Introduction
Neonatal hypoxic-ischemic (HI) encephalopathy (HIE) caused by perinatal asphyxia induces neurosensory deficits, cognitive impairment, lifelong cerebral palsy, and death in infants and young children (Zou et al., 2018). The incidence of HIE is 3 per 1000 live births in developed countries and 26 per 1000 live births in underdeveloped countries (Wachtel et al., 2019). Approximately 24% of children with HIE die from the disease, and approximately 25% of survivors are left with permanent neurophysiological disorders, including visual and auditory impairment, epilepsy and cognitive dysfunction, and sensory deficits in motor and behavior (Douglas-Escobar and Weiss, 2015; Patel et al., 2015; Tan et al., 2021; Xiong et al., 2021). At present, mild hypothermia treatment is the only clinically recognized treatment of moderate to severe hypoxic-ischemic brain injury (HIBD). Although mild hypothermia treatment reduces the mortality and morbidity of HI, it does not show an obvious effect on improving related diseases caused by HI (Mike and Ferriero, 2021). Therefore, identifying novel and effective treatments for HIE is a critical issue.
The pathogenesis of HIBD is complex. In the initial stage of an injury (within 1 hour after injury), partial healing is initiated. A continuously activated apoptotic cascade, oxidative stress, and inflammation occur during the injury incubation period. The incubation period is a major therapeutic target of current research on neuroprotective agents, as even relatively severe damage is potentially related to the short-term recovery of oxidative metabolism in the incubation period after reperfusion (Bennet et al., 2006, 2012). Several studies have confirmed that oxidative stress is the major pathogenesis factor of HIE (Devos et al., 2014; Barbariga et al., 2015; Zhao et al., 2016).
The high metabolic activity of brain tissue makes it extremely sensitive to oxidative stress. When oxidative stress occurs, phospholipids and cholesterol contained in polyunsaturated fatty acids in cell membranes and lipoproteins are easily oxidized via free radical-induced lipid peroxidation to form oxidation products (Sies, 2015), ultimately causing nerve damage. Therefore, reducing oxidative stress damage, balancing the intracellular redox system, and reducing lipid oxidation levels are potential strategies for the treatment of HI during the incubation period.
Chlorogenic acid (CGA) is a natural compound that is present in various plants and a main phenolic compound in coffee. CGA has been shown to exhibit pharmacological effects including anti-inflammatory, antioxidant, anti-cancer, and neuroprotection activities (Miao and Xiang, 2020). Studies indicate that coffee intake may be a protective factor against Alzheimer’s disease, and increased coffee consumption may reduce cognitive decline (Gardener et al., 2021). In a rat model of global cerebral ischemia/reperfusion, CGA significantly reduced the level of glutamate in the cortex, hippocampus, cerebellum, and cerebrospinal fluid (Kumar et al., 2019). CGA was also shown to inhibit the transcriptional activity of hypoxia-inducible factor-1α under hypoxic conditions and reduced the polymerization extent of ferritin induced by Fe3+ and Fe2+ (Ho et al., 2012; Park et al., 2015; Yang et al., 2021). CGA treatment increases the survival rate of dopaminergic neurons, reduces anxiety, and improves exercise, spatial learning, and memory capabilities in mouse models of anxiety and a rabbit model of infarct ischemic strokes (Bouayed et al., 2007; Lapchak, 2007; Shen et al., 2012). These studies show that CGA has a therapeutic effect in these diseases. Nevertheless, the effects of CGA in HIBD have not been examined.
Ferroptosis is a form of programmed cell death that is mediated by lipid peroxides and dependent on iron ions (Yang and Stockwell, 2016; Zhu et al., 2021). Glutathione peroxidase 4 (GPX4) is the key factor in ferroptosis and also a lipid oxidation-related marker. Increased expression of GPX4 is an important biochemical feature of ferroptosis. Under normal conditions, GPX4 regulates the prevention of ferroptosis (Yagoda et al., 2007; Yang et al., 2014; Yang and Stockwell, 2016). In contrast, when GPX4 is inactivated, lipid oxides cannot be reduced. The divalent iron ions will thus oxidize lipids and generate a large number of reactive oxygen species, thus enabling redox imbalance in the cell and inducing ferroptosis (Lu et al., 2017). The synthesis of GPX4 is regulated by system Xc– (the glutamate/cystine antiporter), and the system Xc–/GPX4 axis is a key pathway for regulating ferroptosis (Bannai and Kitamura, 1980; Bridges et al., 2012). The function of system Xc– in the plasma membrane is crucial to inhibiting ferroptosis. Suppression of ferroptosis has been shown to be effective in treating diseases including organ damage caused by ischemia/reperfusion in mouse models (Friedmann Angeli et al., 2014; Linkermann et al., 2014) and is highly related to neurodegenerative diseases (including Parkinson’s disease, Alzheimer’s disease, and Huntington’s disease) (Bogdan et al., 2016; Stockwell et al., 2017).
Therefore, we hypothesized that CGA may play an important role in HIBD, and ferroptosis regulated by the system Xc–/GPX4 axis may be implicated in the pathogenesis of neonatal HIBD. Therefore, we established HIBD mice and the oxygen-glucose deprivation (OGD) models to explore the effects of CGA on HIBD.
Methods
Establishment of the HIBD mouse model and drug administration
All animal care and experiments were conducted following the “Guidelines for the Care and Use of Laboratory Animals” of the National Institute of Health (8th ed, 2011) and were approved by the Experimental Animal Ethics Committee of Wenzhou Medical University (ethics No. Wydw2019-0723 on May 20, 2019). Female (including pregnant) and male C57BL/6J mice aged 6–8 weeks and weighing 18–25 g were purchased from the Wenzhou Medical University Animal Experiment Center, Wenzhou, China (license No. SCXK (Zhe) 2018-228) and raised in the Animal Experiment Center (specific-pathogen-free level) of Wenzhou Medical University in an environment maintained at 23 ± 2°C, 60 ± 10% humidity, with a 12-hour light/dark cycle. Adult mice had free access to standard food and drinking water and were allowed to freely mate to produce offspring for subsequent experiments.
To identify the optimal dosage of CGA, a series of dose-dependent trials were conducted. The mice were randomly divided into Sham (n = 45), Sham + erastin (n = 10), HI (n = 45), HI + CGA (n = 49), and HI + CGA + erastin groups (n = 10). Mice in the CGA groups were intraperitoneally injected with different concentrations of CGA (12.5, 25, 50 mg/kg) (HY-N0055, MedChemExpress, Monmouth Junction, NJ, USA). Mice in the erastin groups were simultaneously intraperitoneally injected with CGA (25 mg/kg) and erastin (20 mg/kg, HY-15763, MedChemExpress). In the sham operation and HI groups, the same amount (25 mg/kg of bodyweight) of normal saline and dimethyl sulfoxide was injected, based on previous explorations.
An improved version of the Rice-Vannucci model was used to establish a neonatal mouse HIBD model (Shah et al., 2017): neonates at 10 days after birth were used in the experiment (pathological changes in 7–10-day-old neonatal mice and human neonates are similar, but 10-day-old mice had a higher survival rate after surgery, so we chose P-10-day mice). Mice were anesthetized by inhalation of anhydrous ether (2–4%) (Changshu Hongsheng Fine Chemical Co., Ltd., Changshu, China) and then were cut along the middle of the neck. After separation and ligation, the left common carotid artery was cut off. The procedure was conducted within approximately 5 minutes. After the operation, the mouse was placed in a 37°C incubator until animals regained consciousness and returned to the cage to stay with their mothers for 2 hours. Mice were then placed into an anoxic box filled with 92% nitrogen and 8% oxygen that was ventilated at a flow rate of 3 L/min for 1.5 hours; the temperature in the box was maintained at approximately 37°C (Yager and Ashwal, 2009). In the Sham group, only the neck skin was cut and the left carotid artery was lifted without ligation and other treatments. After hypoxia exposure, all mice were returned to cages for subsequent experiments.
In this study, we chose a moderate injury model on the basis of clinical symptoms (Chalak et al., 2014); samples with severe conditions such as convulsions or apnea were excluded. A total of 200 mice were used in this study, and the survival rate was 93.5%. Twenty-eight mice were excluded because of severe injury. The flow chart of the experimental procedures is shown in Figure 1. A laser Doppler imager (MoorLDI-2, Moor Instruments Limited, Devon, UK) with a laser wavelength of 633 nm, scan of 55 cm and scan duration of 5 minutes was used to verify the success of the model, and MoorLDI Review V6.1 software was used to quantify the results.
;)
Figure 1: The flow chart of animal experiments.CGA: Chlorogenic acid; HI or HIBD: hypoxic-ischemic brain injury; qPCR: quantitative real-time polymerase chain reaction; TTC: 2,3,5-triphenyl tetrazolium chloride.
Determination of cerebral infarction volume
The mice were euthanized 48 hours after the induction of HIBD. The brain tissue was immediately frozen for 5 minutes at –20°C, cut into 2 mm coronal sections, and immersed in 1% 2,3,5-triphenyl tetrazolium chloride (T8877-10G, Sigma-Aldrich, St. Louis, MO, USA) in the dark at 37°C for 30 minutes. Tissues were then soaked in 4% paraformaldehyde overnight. ImageJ software (1.8.0, National Institutes of Health, Bethesda, MD, USA) (Schneider et al., 2012) was used to measure and calculate the volume of the cerebral infarction area.
Brain water content measurement
We measured the edema state of the damaged cerebral hemisphere (left hemisphere) after HIBD and administration, as previously described (Zhang et al., 2016). At 24 hours after HIBD, one mouse from each group was anesthetized by inhalation with anhydrous ether and sacrificed. Pictures were taken immediately after obtaining fresh tissue, and the left hemibrain was weighed to obtain a wet weight. Brain tissues were placed in a dry oven (100°C) for 48 hours to obtain a dry weight (accurate to 0.1 mg). The dry-wet ratio of the left hemisphere of the brain was calculated, which represented the percentage of brain water content: ([wet weight – dry weight]/wet weight) × 100%. ImageJ software was used for the determination of residual brain volume (ratio of the injured hemisphere to the contralateral hemisphere).
Real-time reverse transcription polymerase chain reaction
An RNA extraction kit (RC101-01, Vazyme, Nanjing, China) was used to extract total RNA from the hippocampus following the manufacturer’s instructions. The optical density 260/280 ratio and RNA concentration were measured using an ultra-micro spectrophotometer (NanoDrop 2000, Thermo Scientific, Wilmington, DE, USA). The optical density ratio was within the range of 1.8–2.0. RNA was reverse transcribed into complementary DNA using a reverse transcriptase system (122-01, Vazyme) and a PCR thermal cycler (T100 thermal cycler, Bio-Rad, Singapore). Quantitative PCR was performed using the CFX96 Optics Module (Bio-Rad) using the following cycling conditions: 95°C for 3 minutes, 95°C for 15 seconds, and 60°C for 30 seconds (40 cycles) and decreased to 65°C at a gradient of 0.5°C (41 cycles; melting curve step). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA was used as an internal control, and gene expression was determined using the 2–ΔΔCT method (Pfaffl, 2001). The forward and reverse primer sequences are shown in Table 1
Table 1: Primers for quantitative polymerase chain reaction
Oxidative stress evaluation
Oxidative stress in brain tissue was evaluated by the level of malondialdehyde (MDA). Brain tissue samples frozen at –80°C were lysed and homogenized, and protein concentration was determined using a BCA Protein Assay Kit (P0012, Beyotime) following the manufacturer’s instructions. MDA detection was performed using an MDA kit (S0131M, Beyotime, Shanghai, China). Briefly, brain tissue samples were mixed with lysate (P0013B, Beyotime) and homogenized, and the concentration was calculated according to the standard protein curve. Next, 100 µL brain tissue samples with known protein concentration were mixed with 200 µL MDA working solution, heated at 100°C for 15 minutes, cooled to room temperature in a water bath, and centrifuged at 1000 × g for 10 minutes at room temperature. The supernatant (200 µL) was immediately transferred to a 96-well plate. A spectrophotometer (Thermo Scientific) was used to measure the absorbance at 532 nm. MDA concentration was calculated using the standard curve (MDA µmol/mg).
Histopathological analysis and tissue immunofluorescence staining
Tissue samples were collected on the 7th day after HI injury (P17). After anesthesia of mice by anhydrous ether inhalation, the heart was perfused with 10 mL of phosphate-buffered saline (PBS; pH 7.4, P1020, Solarbio) and then with 10 mL of paraformaldehyde (pH 7.4). Fresh brain tissue was immediately transferred to 4% paraformaldehyde, embedded in paraffin at 4°C overnight, and cut into 5 µm coronal sections for subsequent histological evaluation. The paraffin-embedded coronal brain sections were deparaffinized, hydrated, and stained with a hematoxylin and eosin staining kit (G1120, Solarbio, Beijing, China). Briefly, mouse brain sections were fixed in 4% paraformaldehyde for at least 24 hours, dehydrated and embedded in paraffin. Serial 5-µm brain sections were immersed in xylene I and II for 10 minutes for deparaffinization, followed by immersion in 100% ethanol for 2 minutes, 95% ethanol for 1 minute, 80% ethanol for 1 minute, 75% ethanol for 1 minute, and distilled water for 2 minutes. Brain sections were stained with hematoxylin for 5–7 minutes, washed with tap water, washed with 1% hydrochloric alcohol, washed with tap water, and counterstained with eosin for 3 minutes prior to routine procedures, such as dehydration, hyalinization and sealing with neutral resin. The hippocampal CA1 area was observed using an optical microscope (LEICA DFC7000T, Wetzlar, Germany). The procedure for Nissl staining was similar to those described above; after deparaffinization and hydration, brain slices were processed using a Nissl staining kit (G1432, Solarbio, Beijing, China). Briefly, brain slices were incubated with methyl violet staining solution for 15 minutes and rinsed with distilled water, and Nissl’s differentiation solution was used until most of the staining was eliminated. Routine dehydration and other procedures were then performed.
The mouse brains were perfused with PBS, fixed with paraformaldehyde (P1110, Solarbio), and dehydrated with sucrose; samples were frozen and sectioned into 24 µm sections by a computer microtome (KD-3390, KEDEE, Jinhua, Zhejiang, China). For immunofluorescence staining of tissue sections, the frozen section was immersed in 4% paraformaldehyde for 24 hours and dehydrated using 10%, 20%, and 30% sucrose gradients. The slices were permeabilized with Triton X-100 (0.2%) for 15 minutes, blocked in PBS containing 10% goat serum (SL038, Solarbio) for one hour, and incubated with primary antibody overnight at 4°C. The sections were then incubated with secondary antibodies at 25°C for 2 hours. Next, 4′,6-diamidino-2-phenylindole (DAPI) anti-fluorescence attenuation mounting tablets (Solarbio, S2110) were used to mount the tissues and images were obtained using a fluorescence microscope (Nikon TI-DH, Tokyo, Japan). The antibodies used were as follows: mouse anti-myelin basic protein (MBP; 1:200, Biolegend, San Diego, CA, USA, Cat# 808401, RRID: AB_2564741), rabbit anti-glial fibrillary acidic protein (GFAP; 1:200, Abcam, Cat# ab7260, RRID: AB_305808), rabbit anti-GPX4 (1:100, Affinity Biosciences, Liyang, China, Cat# DF6701, RRID: AB_2838663), mouse anti-microtubule associated protein 2 (MAP2; 1:200, ZEN-BIOSCIENCE, Chengdu, China, Cat# 250035, RRID: AB_2909468), rabbit anti-neuronal nuclei antigen (NeuN; 1:200, Cell Signaling Technology, Danvers, MA, USA, Cat# 12943s, RRID:AB_2630395), goat anti-rabbit IgG H&L (Alexa Fluor® 488, 1:200, Abcam, Cat# ab150077, RRID: AB_2630356), goat anti-mouse IgG H&L (Alexa Fluor® 488, 1:200, Abcam, Cat# ab150113, RRID: AB_2576208), and goat anti-mouse IgG H&L (Alexa Fluor® 594, 1:200, Abcam, Cat# ab150116, RRID: AB_2650601). The cell fluorescence treatment steps were similar to those described above.
Isolation and culture of primary cortical neurons
Pregnant mice at approximately 18 days of gestation were anesthetized with isoflurane (R510-22-10, RWD, Shenzhen, China) and the fetuses were removed and placed on ice. Brain tissue was removed and quickly transferred to Hanks’ balanced salt solution (11575032, Gibco, Grand Island, NY, USA) after disinfecting with 75% alcohol. The cortex was separated under a dissecting microscope to remove the meninges and cut the tissue. A papain separation system kit (A003176, A003182, Sangon Biotech, Shanghai, China) was used to digest and separate cells. After digestion at 37°C for 1 hour, the reaction was terminated using Dulbecco’s modified Eagle medium containing 10% fetal bovine serum, which was then used to dilute and resuspend the cells. Cells (2.5 × 105/mL) were inoculated on a culture dish or glass coverslip coated with poly-L-ornithine solution (P4957, Sigma), and the dishes were placed in a CO2 cell incubator for 4 hours or overnight. The culture medium was replaced with neurobasal medium (containing fibroblast growth factor and 1% B27) and cells were returned to culture.
Cell viability assay
Cell viability was determined using the Cell Counting Kit-8 (C0038, Beyotime) following the manufacturer’s protocol. First, the primary cortical neurons were cultured in 96-well plates at a density of 2 × 104/mL. After 3 days, the cells were incubated with various concentrations of CGA (0, 5, 10, 20, 40, 80, 100, 120, 160, 200 µM) for 24 hours; the OGD group was treated as described above after drug pretreatment. Finally, 10 µL of CCK-8 solution were added to each well and cells were incubated at 37°C for 2 hours. Absorbance was detected using a spectrophotometer at 450 nm wavelength.
OGD model and drug treatment
Cells were cultured for 4 days and then divided into five groups: the control, erastin control (10 µM erastin) (Huo et al., 2016), OGD, OGD + CGA (100 µM CGA), and OGD + CGA + erastin (100 µM CGA and 10 µM erastin) groups. Because of the irreversibility of neuronal damage, OGD + CGA and OGD + CGA + erastin groups were pretreated with CGA for 12 hours. After pretreatment, the culture medium was replaced with sugar-free medium and cells were cultured for 3 hours under 99% nitrogen and 1% oxygen to simulate HI conditions. Cells were then transferred to a regular CO2 incubator at 37°C for 24 hours.
Measurement of reactive oxygen species
Oxidative stress levels were analyzed using a reactive oxygen species assay kit (S0033M, Beyotime). Briefly, the treated primary neurons were incubated with 10 µM 2,7-dichlorodi-hydrofluorescein diacetate (DCFH-DA; S0033M-1, Shanghai, China, Beyotime) at 37°C in the dark for 20 minutes and washed three times with serum-free Dulbecco’s modified Eagle medium. The neurons were then observed under a fluorescence microscope.
TdT-mediated dUTP nick-end labeling (TUNEL) staining
An In Situ Cell Death Detection Kit (Roche, 11684817910, Basel, Switzerland) was used to detect apoptotic DNA fragments. Paraffin sections (5 µm) of the brain were obtained 24 hours after the above HIBD. After routine deparaffinization and hydration, the brain tissue was incubated with 20 µg/mL proteinase K working solution (ST533, Beyotime) at 37°C for 30 minutes and washed with PBS, and the area around the sample was dried. The sample and TdT-mediated dUTP nick-end labeling (TUNEL) reaction mixture were incubated in a dark and humid environment at 37°C for 1 hour and then washed three times with PBS (pH 7.4); the plate was then mounted with an anti-fluorescence quencher containing DAPI. A fluorescence microscope was used to examine apoptosis. ImageJ was used to measure the number of TUNEL-positive cells and calculate the apoptotic index (the ratio of the number of apoptotic cells to the total number of cells in a field of view).
Western blot analysis
Tissues were lysed in radioimmunoprecipitation buffer (P0013B, Beyotime) with a phenylmethylsulfonyl fluoride protease inhibitor (ST506, Beyotime), homogenized, and centrifuged at 13,201 × g and 4°C for 30 minutes. To prepare neuron lysates, neurons were lysed on ice, sonicated for 10 minutes, and centrifuged under similar conditions. Protein concentration was determined using a BCA protein assay kit (P0012, Beyotime). Equivalent amounts of protein (50 mg) were separated using sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred to a polyvinylidene difluoride membrane (PR05509, Millipore, Cork, Ireland). After blocking with 5% skimmed milk for 2 hours, the membranes were incubated with primary antibodies overnight at 4°C. The primary antibodies used included rabbit anti-4-hydroxynonenal (4-HNE; 1:2000, Abcam, Cat#ab46545, RRID: AB_722490), rabbit anti-glutathione peroxidase 4 (GPX4; 1:100, ABclonal, Wuhan, China, Cat# a1933, RRID:AB_2763960), rabbit anti-solute carrier family 3 member 2 (SLC3A2; 1:200, ABclonal, Cat# a5702, RRID:AB_2766461), rabbit anti-solute carrier family 7 member 11 (SLC7A11; 1:1000, ABclonal, Cat# a15604, RRID:AB_2763010), rabbit anti-glutaminase 2 (GLS2; 1:1000, ABclonal, Cat# a16029, RRID:AB_2763466), rabbit anti-glutathione synthase (GSS; 1:2000, ABclonal, Cat# a14535, RRID: AB_2861700), rabbit anti-ferritin light chain (FLC; 1:2000, Cat# a11241, ABclonal, RRID:AB_2861532), and rabbit anti-ferritin heavy chain (FHC; 1:2000, ABclonal, Cat# a1144, RRID: AB_2758562). Membranes were then incubated with goat anti-rabbit IgG (1:5000, Abcam, Cat# ab97051, RRID:AB_10679369) or goat anti-mouse IgG (1:5000, Abcam, Cat# ab97023, RRID: AB_10679675) secondary antibody at 25°C for 2 hours. Membranes were washed three times with Tris-buffered saline with Tween 20 (TBST; T1085, Solarbio) and ChemiDox XRS (Bio-Rad) was used for detection. ImageJ software was used to analyze the optical density of the bands. All experiments were repeated at least three times.
Statistical analysis
No statistical methods were used to predetermine sample sizes. Animals with severe injury (as mentioned above) were euthanized and not included in the statistical sample data. The evaluator was blind to the animal groupings. All data were obtained from at least three independent experiments. Data are presented as mean ± standard error of mean (SEM). Statistical analyses were performed using GraphPad Prism 8.0 (GraphPad Software, San Diego, CA, USA, www.graphpad.com). Inter-group comparisons were performed using a one-way analysis of variance, followed by Tukey’s post hoc test when analyzing data from more than two groups. The Student’s t-test was used for the comparison of two groups. P < 0.05 was considered statistically significant.
Results
CGA attenuates HIBD in newborn mice
We established the HIBD model as described in the Methods. A laser Doppler imager was used to verify that the mouse left carotid artery was successfully ligated and quantify cerebrovascular flow (Figure 2A and B). We observed a significant reduction in blood flow on the left side of the model mice compared with the sham surgery mice (P = 0.0034). Infarct volume was quantified in mice from the Sham + normal saline (NS), HI + NS, and HI + different concentrations of CGA (12.5, 25, 50 mg/mL) groups. CGA at various concentrations significantly reduced the volume of cerebral infarction. Analysis using 2,3,5-triphenyl tetrazolium chloride revealed that 25 mg/kg was the most effective dose that significantly reduces the volume of cerebral infarction (P < 0.0001; Figure 2C and D). Compared with HI + CGA (25 mg/kg), 50 mg/kg CGA did not further reduce the volume of cerebral infarction (P = 0.7474). Thus, 25 mg/kg CGA was used in subsequent experiments.
Figure 2: Chlorogenic acid attenuates the hypoxic-ischemic brain damage in newborn mice.(A) Representative image of laser Doppler scan photographs of the cerebrovascular region in the Sham and HI groups. Cerebral blood flow was decreased significantly in the HI group compared with other groups. (B) Quantification of the blood flow volume (n = 4). (C) Quantification of infarct volume based on TTC staining (n = 4). (D) Representative images of TTC-stained coronary brain 48 hours after HIBD; significant reduction in cerebral infarct size (arrows) was observed after 25 mg/kg CGA treatment. (E) Quantification of the dry-wet ratio (n = 6). (F) Representative images of the brain 7 days after HIBD. The area of brain liquefaction was significantly reduced after CGA treatment. Scale bar: 1 mm. (G) Quantification of residual brain volume (n = 4). Data are presented as mean ± SEM and were analyzed by Student’s t-test (B) or one-way analysis of variance followed by Tukey’s post hoc test (C, E, and G). **P < 0.01, ***P < 0.001, vs. Sham (+ NS) group; ##P < 0.01, ###P < 0.001, vs. HI + NS group. CGA: Chlorogenic acid; HI or HIBD: hypoxic-ischemic brain injury; NS: normal saline; TTC: 2,3,5-triphenyl tetrazolium chloride.
We then measured the dry-wet ratio of the injured cerebral hemisphere and found that the HI + CGA group showed significantly reduced cerebral edema after HIBD compared with the Sham + NS group (P = 0.0035; Figure 2E). We also examined the brain anatomy seven days after HIBD and the remaining hemisphere volume was quantified (Figure 2F and G). The injured cerebral hemisphere in the HI + NS group had severe brain atrophy or even liquefaction; in contrast, the degree of atrophy and liquefaction in the HI + CGA group was significantly reduced (P < 0.0001).
CGA induces neuroprotective effects in brain tissue after HIBD in newborn mice
To examine neuronal arrangement after treatment, hematoxylin and eosin staining on brain slices of mice was performed seven days after HIBD (Figure 3A); these slices directly revealed the integrity of the cerebral hemispheres. Brain cells in hippocampal CA3, CA1, and dentate gyrus areas of the Sham + NS group were arranged in a regular pattern. In the HI + NS group, the hippocampal area was atrophied, with apparent cell arrangement disorder, rupture of nucleus, and partial neuron loss. In the HI + CGA group, these changes were alleviated; the atrophy of the hippocampus was relieved, and many areas were orderly arranged with an increased number of neurons.
Figure 3: CGA induces neuroprotective effects on the brain tissue after HIBD in newborn mice.(A) Representative image of hematoxylin and eosin staining in hippocampal CA3, CA1 and DG region 7 days after HIBD. Atrophy in the hippocampus was relieved in the HI + CGA group, and many areas remained orderly arranged with an increased number of neurons. (B) Representative immunofluorescence staining images of MBP (red) and DAPI (blue) on the corpus callosum. MBP staining was increased in the HI + CGA group. (C) Representative immunofluorescence staining images of GFAP (green) and DAPI (blue) on the corpus callosum. GFAP staining was decreased in the HI + CGA group. (D, E) Quantification data of mean fluorescence intensity of MBP (n = 3) and GFAP (n = 4) in the brain. (F) Representative immunofluorescence staining images of TUNEL (green) and DAPI (blue) on the cerebral cortex. The number of TUNEL-positive cells was decreased in the HI + CGA group. Scale bars: 100 µm. (G) Quantification of TUNEL staining in the brain (n = 3). Data are presented as mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, vs. Sham + NS group; #P < 0.05, ##P < 0.01, vs. HI + NS group (one-way analysis of variance followed by Tukey’s post hoc test). CGA: Chlorogenic acid; DAPI: 4’,6-diamidino-2-phenylindole; DG: dentate gyrus; GFAP: glial fibrillary acidic protein; HI or HIBD: hypoxic-ischemic brain injury; MBP: myelin basic protein; NS: normal saline; TUNEL: TdT-mediated dUTP nick-end labeling.
To investigate whether CGA accelerates axon repair in neonatal HI-injured mice, immunofluorescence staining was used to detect the expression of MBP and GFAP. MBP staining only detects the corpus callosum part in the brain, because this region highly expresses MBP. MBP immunopositivity on the corpus callosum of the HI + NS group was significantly lower than that of the Sham + NS group (P = 0.0076); in the group treated with CGA, upregulated MBP expression was observed. GFAP expression was significantly upregulated in the HI + NS group, whereas the group treated with CGA showed inhibited GFAP expression (P = 0.0189; Figure 3B–E). Furthermore, a significant increase in the number of TUNEL-positive cells after HI injury was observed, and this increase was reduced by CGA intervention (P = 0.0034; Figure 3F–G). Together, these results indicate that CGA promotes axon repair and reduces the number of TUNEL-positive cells with neuroprotective effects on newborn mice after HIBD.
CGA improves HIBD by reducing oxidative stress
To explore the impact of CGA on oxidative stress after HIBD (Lu et al., 2015), the level of MDA, a byproduct of lipid oxidation (Tsikas, 2017), in the cerebral cortex was measured. The level of MDA in the HI + NS group was significantly higher than that in the Sham group, and CGA significantly reduced the level of MDA (P < 0.0001; Figure 4A). We also measured and analyzed the lipid oxidation-related markers 4-HNE and GPX4 by western blotting (Figure 4B–E). The expression level of 4-HNE in the HI + NS group was significantly upregulated (P = 0.0341), whereas GPX4 expression was significantly downregulated (P = 0.013); these changes were partially reversed in the HI + CGA group. Together, these results imply that CGA may reduce HIBD by inhibiting oxidative stress.
Figure 4: CGA improves hypoxic-ischemic brain damage by reducing oxidative stress.(A) The MDA level in brain tissues. (B, C) Western blot of 4-HNE and GPX4 in brain tissues. β-Actin was used as an internal reference. (D, E) Quantitative analyses of 4-HNE and GPX4 (normalized to β-actin). Data are presented as mean ± SEM (n = 4). **P < 0.01, vs. Sham + NS group; #P < 0.05, ##P < 0.01, vs. HI + NS group (one-way analysis of variance followed by Tukey’s post hoc test). 4-HNE: 4-Hydroxynonenal; CGA: chlorogenic acid; GPX4: glutathione peroxidase 4; HI: hypoxic-ischemic brain injury; MDA: malondialdehyde; NS: normal saline.
CGA regulates the levels of ferroptosis-associated proteins after HIBD in newborn mic
Lipid peroxidation and iron metabolism disorders are the primary causes of ferroptosis (Yang and Stockwell, 2016). Our findings revealed changes in MDA, 4-HNE, and GPX4 expression in the animal groups treated with CGA. Notably, these changes are consistent with the biochemical characteristics of ferroptosis, and GPX4 is a key factor in the regulation of ferroptosis.
To explore whether CGA acts on HIBD through influencing ferroptosis, western blot was used to analyze the expression of ferroptosis-related proteins in the brain of mice (Figure 5A–F). Ferritin heavy chain and ferritin light chain are related to iron homeostasis (Li et al., 2010). Both were significantly reduced after HIBD but recovered after CGA intervention (P = 0.0025; P = 0.0366). This shows that a large amount of Fe2+ was inactivated after HIBD, and additional Fe2+ needed to be obtained from the iron pool to maintain balance. Glutaminase 2 regulates the conversion of glutamine to glutamate and increases damage after HIBD. This also indirectly confirms that CGA reduced ferroptosis after HIBD. Quantitative PCR was performed to evaluate ferroptosis-related mRNA levels in the hippocampus (Figure 5G). The mRNA levels of nuclear factor erythroid-2 (a factor that regulates GPX4 (Deng et al., 2020)) (P = 0.0001), glutathione (P < 0.0001) and GPX4 (P < 0.0001) in the HI + NS group were significantly lower than those in the Sham + NS group, while cyclooxygenase-2 (P = 0.0049) was higher. CGA partially reversed the mRNA levels of these ferroptosis-related factors.
Figure 5: CGA regulates the relative expression level of ferroptosis proteins after HIBD in newborn mice.(A–F) Western blot of SLC7A11, FHC, FLC, SLC3A2, GLS2, and GSS expression in the cortex (n = 4). β-Actin was used as an internal reference. (G) Cox-2, Nrf2, GSH and GPX4 mRNA levels in hippocampus tissue 24 hours after HIBD; GAPDH mRNA was used as a normalization control (n = 3). Data are presented as mean ± SEM. *P < 0.05, **P < 0.01, vs. Sham + NS group; #P < 0.05, ##P < 0.01, vs. HI + NS group (one-way analysis of variance followed by Tukey’s post hoc test). CGA: Chlorogenic acid; COX2: cyclooxygenase-2; FHC: ferritin heavy chain; FLC: ferritin light chain; GAPDH: glyceraldehyde-3-phosphate dehydrogenase; GLS2: glutaminase 2; GSH: L-glutathione; GSS: glutathione synthase; HI or HIBD: hypoxic-ischemic brain injury; Nrf2: nuclear factor (erythroid-derived 2)-like 2; NS: normal saline; SLC3A2: solute carrier family 3 member 2; SLC7A11: solute carrier family 7 member 11.
CGA protects primary cortical neurons from OGD-induced damage
We next examined the effects of various concentrations CGA on the viability of primary cortical neurons using CCK-8 assay. Treatment of neurons with CGA at concentrations up to 100 µM for 24 hours did not induce any toxic effects; decreased neuronal viability was observed at a dose of 120 µM (P = 0.0401; Figure 6A). Next, neurons were pretreated with different concentrations of CGA (0–120 µM) and then subjected to OGD injury. The therapeutic effect of CGA on primary cortical neurons treated with OGD was dose-dependent (P < 0.0001; Figure 6B). From these results, we selected the concentration of 100 µM CGA for subsequent experiments.
Figure 6: CGA protects primary cortical neurons from OGD-induced damage.(A) Neuronal viability of primary cortical neurons treated with various doses of CGA for 24 hours was determined by CCK8. (B) Neuron viability after reoxygenation for 24 hours after OGD for 3 hours. (C) Representative image of primary cortical neurons with OGD treatment for 4 hours. CGA protected primary neurons from OGD injury. (D) ROS generation of each group was detected by DCFH-DA (green) staining. ROS production was significantly reduced in the OGD + CGA groups compared with the OGD group. (E) Representative immunofluorescence staining images of GPX4 (green, FITC) and DAPI (blue). The immunopositivity of GPX4 was significantly increased in the OGD + CGA groups compared with the OGD group. (F) Representative immunofluorescence staining images of TUNEL (green) and DAPI (blue) on the primary neurons. The number of TUNE-positive cells was significantly increased in the OGD group compared with the control and OGA + CGA groups. (G) Representative immunofluorescence staining images of NeuN (green, FITC) and DAPI (blue) on the primary neurons. The number of NeuN positive cells was significantly reduced in the OGD group compared with the control and OGA + CGA groups. (H) Representative immunofluorescence staining images of MAP2 (green, FITC) and DAPI (blue) on the primary neurons. MAP2 staining was significantly reduced in the OGD group compared with the control and OGA + CGA groups. Scale bars: 50 µm in C, G, H; 100 µm in D, F; 20 µm in E. (I–K) Quantification data of mean fluorescence intensities of GPX4 (I), NeuN (J) and MAP2 (K) in the primary neurons. (L) Quantification of TUNEL staining in the primary neurons 24 hours after OGD. (M) Cox-2, GSH and GPX4 mRNA expressions in the primary neurons 24 hours after OGD; GAPDH mRNA was used for normalization. Data are presented as mean ± SEM (n = 3). *P < 0.05, **P < 0.01, vs. control group; #P < 0.05, ##P < 0.01, ###P < 0.001, vs. OGD group (one-way analysis of variance followed by Tukey’s post hoc test). CGA: Chlorogenic acid; DAPI: 4′,6-diamidino-2-phenylindole; DCFH-DA: 2,7-dichlorodi-hydrofluorescein diacetate; GPX4: glutathione peroxidase 4; MAP2: microtubule association protein-2; NeuN: neuronal nuclei; OGD: oxygen and glucose deprivation; ROS: reactive oxygen species; TUNEL: TdT-mediated dUTP nick-end labeling.
After subjecting primary cortical neurons to OGD for 24 hours (Xu et al., 2020) to simulate the hypoxic-ischemic environment, analysis of cells using an optical microscope showed that most axons were broken following OGD exposure. Notably, treatment with CGA alleviated the effects of OGD. CGA also reduced the production of reactive oxygen species (Figure 6C and D). Results of immunofluorescence staining of GPX4 (P = 0.0103), the neuronal markers NeuN (P = 0.0436) and MAP2 (P = 0.0481), and TUNEL-positive cells (P = 0.001) on primary cortical neurons 24 hours after exposure to OGD (Figure 6E–L) were consistent with in vivo findings. In addition, the mRNA levels of glutathione (P < 0.0001) and GPX4 (P < 0.0001) were significantly lower in the OGD group than those in the control group, while the mRNA level of cyclooxygenase-2 was higher than that in the control group (P < 0.0001). Notably, CGA partially reversed these mRNA levels (Figure 6M). Together these findings indicate that CGA treatment suppressed OGD-induced damage and reduced the number of TUNEL-positive cells on primary neurons.
Erastin partially reverses the neuroprotective effect of CGA following HIBD in newborn mice
To determine whether the effects of CGA against HIBD were mediated by the system Xc–/GPX4, erastin (a system Xc– inhibitor) was applied. To exclude the effects of erastin on mice, a control + erastin group was established. Changes in brain tissue structure 7 days after HIBD treatment were examined by Nissl staining (Figure 7A–D). In both HI and HI + CGA + erastin groups, the damage was decreased. The number of Nissl bodies in the cortex, hippocampal CA1, CA3 and dentate gyrus areas was markedly decreased following HIBD treatment (P < 0.05). In the sham and sham + erastin groups, the Nissl bodies showed neat and dense arrangement. Immunofluorescence staining of MBP and GFAP is shown in Figure 7E–H. In the sham group, the axons around the corpus callosum appeared dense and branched, while GFAP immunopositivity was low. CGA treatment restored the growth of axons, suppressed the expression of GFAP (P = 0.0047), and increased the number of Nissl bodies. Erastin treatment partially alleviated the changes induced by CGA treatment and partially reversed the neuroprotective effect of CGA treatment.
Figure 7: Erastin partially reverses the neuroprotective effect of CGA following HIBD in newborn mice.(A) Representative image of Nissl staining in the cortex, hippocampal CA1, CA3 and DG region 7 days after HIBD. The numbers of Nissl bodies were significantly reduced in the HI and HI + CGA + Erastin groups compared with other groups. Scale bars: 100 µm. (B–D) The number of Nissl bodies in the cortex (B), hippocampal CA1 (C), and CA3 regions (D). (E) Representative immunofluorescence staining images of MBP (red, Cy3) and DAPI (blue) on the corpus callosum. MBP staining was significantly decreased in the HI and HI + CGA + erastin groups compared with other groups. Scale bars: 50 µm. (F) Representative immunofluorescence staining images of GFAP (green, FITC) and DAPI (blue) on the corpus callosum. GFAP expression was significantly decreased in the HI and HI + CGA + erastin groups compared with other groups. Scale bars: 50 µm. (G, H) Quantification data of mean fluorescence intensity of MBP (G) and GFAP (H) in the brain. Data are presented as mean ± SEM (n = 3). *P < 0.05, **P < 0.01 (one-way analysis of variance followed by Tukey’s post hoc test). AU: Arbitrary unit; CGA: chlorogenic acid; DAPI: 4’,6-diamidino-2-phenylindole; DG: dentate gyrus; GFAP: glial fibrillary acidic protein; HI or HIBD: hypoxic-ischemic brain injury; MBP: myelin basic protein.
CGA exerts neuroprotection via the system Xc–/GPX4 axis
Activation of the system Xc–/GPX4 axis inhibits lipid peroxidation, thereby preventing ferroptosis (Yang and Stockwell, 2016). Western blotting results showed that the expression levels of SLC7A11 and SLC3A2, the main components of system Xc– (Bridges et al., 2012), were significantly reduced after HIBD (P = 0.006, P = 0.007; Figure 8A–H), and treatment with CGA significantly increased the expression of the proteins (P = 0.003, P = 0.004; Figure 8F–I). However, simultaneous co-treatment of erastin with CGA partially reversed the therapeutic effect of CGA (P = 0.002, P = 0.03). GSS (a key factor involved in glutathione biosynthesis (Wang et al., 2021)) and its downstream factor GPX4 have a similar expression pattern. Erastin partially reversed the antioxidant effect of CGA (P < 0.0001; Figure 8I). In summary, there results suggest that CGA activated the system Xc– and increased the expression of GPX4, thereby exerting neuroprotection.
Figure 8: CGA exerts neuroprotection via the system Xc–/GPX4 axis.(A–D) Western blot analysis of SLC3A2, SLC7A11, GSS, and GPX4 in the cortex. β-Actin was used as an internal reference. (E–H) Quantitative analyses of SLC3A2, SLC7A11, GSS, and GPX4 expression normalized with β-actin in the cortex. (I) The MDA level in the cortex. Data are presented as mean ± SEM (n = 4). *P < 0.05, **P < 0.01 (one-way analysis of variance followed by Tukey’s post hoc test). CGA: Chlorogenic acid; GPX4: glutathione peroxidase 4; GSS: glutathione synthase; HI or HIBD: hypoxic-ischemic brain injury; MDA: malondialdehyde; SLC3A2: solute carrier family 3 member 2; SLC7A11: solute carrier family 7 member 11.
Discussion
Neonatal HIE is the primary cause of chronic neurological diseases and neonatal death. However, effective treatment methods for neonatal HIE are still lacking. Oxidative stress is one of the pathophysiological mechanisms of neonatal HIE (Lu et al., 2015; Huang et al., 2019).
Cell death is crucial for development, maintaining homeostasis, and preventing and treating diseases (Fuchs and Steller, 2011). Ferroptosis, a newly discovered form of programmed cell death, has been shown to be highly associated with HIE (Lin et al., 2022). Ferroptosis is prevented by GPX4, an enzyme that relies on the most abundant antioxidant glutathione in the cell, which reduces phospholipid hydroperoxide and inhibits lipoxygenase-mediated lipid peroxidation (Yagoda et al., 2007; Yang et al., 2014; Yang and Stockwell, 2016). The synthesis of GPX4 is regulated by cysteine and the cysteine transport system Xc–; inhibition of system Xc– on the plasma membrane results in decreased GPX4 expression, triggers ferroptosis induced by lipid peroxidation, and ultimately causes damage to the nervous system (Yan et al., 2021).
CGA is a polyphenol that is present in many types of plants as well as coffee and tea. Although some clinical studies have suggested that high consumption of coffee increases the risk of neurodegenerative disease in humans (Pham et al., 2021), drinking coffee and tea separately or in combination was associated with a lower risk of stroke and dementia (Zhang et al., 2021), suggesting that increasing coffee or tea consumption may reduce the risk of neurodegenerative diseases in humans. CGA contains five active hydroxyl groups and one carboxyl group. The phenolic hydroxyl structure easily reacts with free radicals to form hydrogen free radicals with antioxidant effects, eliminating the activity of hydroxyl free radicals and superoxide anions; thus, its antioxidant effect is significant (Tošović et al., 2017; Miao and Xiang, 2020). The role of CGA in cardioprotection and neuroprotection has been extensively studied. The neuroprotective effect is primarily attributed to its high antioxidant activity in the brain (Han et al., 2010; Kwon et al., 2010; Heitman and Ingram, 2017). Nevertheless, the role and mechanism of CGA in brain injury are still unclear. Therefore, here we established in vivo and in vitro mouse models to analyze the potential protective effect of CGA on HIBD and explored the possible mechanisms. We also used erastin, a drug that directly inhibits system Xc– influencing downstream GPX4 synthesis (Xie et al., 2016), to establish whether ferroptosis regulated by the system Xc–/GPX4 axis is involved in this process.
Our results revealed that intraperitoneal injection of 25 mg/kg CGA into mice significantly reduced the cerebral infarction volume of mice, state of cerebral edema in a short period, and inhibited oxidative stress. CGA also improved the morphology of brain tissue, promoted axon growth recovery, and inhibited cell apoptosis and GFAP formation in the long term. Numerous studies have identified GFAP as a potential biomarker of HIE to assess the severity of neonatal HIBD (Douglas-Escobar et al., 2010; Chalak et al., 2014). The in vitro OGD model established in our study showed that CGA promoted the survival of primary cortical neurons, and the therapeutic effect was dose dependent. These findings suggest that CGA improves brain damage caused by hypoxia-ischemia. Western blotting revealed that while CGA was working, the changes in GPX4 were also increased. GPX4 is an enzyme that reduces lipid peroxidation in biological membranes, and mitochondrial-specific overexpression of GPX4 alleviates ischemia/reperfusion-related cardiac dysfunction (Dabkowski et al., 2008; Brigelius-Flohé and Maiorino, 2013). This suggests that ferroptosis may be involved in the damage caused by HI. Our findings show that CGA attenuated this phenomenon and improved the brain damage caused by hypoxia and ischemia.
To investigate whether GPX4-mediated ferroptosis is implicated in brain damage caused by HI, we analyzed the key upstream regulatory factors of GPX4 synthesis, i.e., the cysteine and cysteine transport system Xc–. System Xc– comprises the catalytic subunit SLC7A11 and the partner subunit SLC3A2. Subunit SLC7A11 takes in extracellular cysteine, which is converted to cysteine in the cytoplasm through the reduction reaction of the reduced form of nicotinamide-adenine dinucleotide phosphate, while subunit SLC3A2 is involved in the synthesis glutathione, a key substance for GPX4 activation (Yang et al., 2014; Xie et al., 2016; Koppula et al., 2021). Our data revealed that CGA may play a neuroprotective effect after HIBD in newborn mice by activating system Xc– and upregulating GPX4 expression. Erastin partially reversed the neuroprotective effect of CGA, and the expression of related proteins on the system Xc–/GPX4 axis after CGA treatment was restored. Thus, these findings suggest that CGA may promote the recovery of neonatal HIBD by activating the system Xc–/GPX4 axis (Figure 9).
Figure 9: Schematic model for the mechanism of CGA in HIE.CGA inhibits ferroptosis by activating the system Xc–/GPX4 axis after HI. CGA: Chlorogenic acid; GCL: glutamate cysteine ligase; GLS: glutaminase; GPX4: glutathione peroxidase 4; GSH: L-glutathione; GSS: glutathione synthase; GSSG: L-glutathione oxidized; HIE: hypoxic-ischemic encephalopathy; ROS: reactive oxygen species; SLC3A2: solute carrier family 3 member 2; SLC7A11: solute carrier family 7 member 11; Xc–: the glutamate/cystine antiporter.
Many studies have indicated CGA, a polyphenol with significant antioxidant effects, as a key therapeutic drug, including an agent for the treatment of ischemic infarction and neurodegenerative diseases (Loader et al., 2017; Bao et al., 2018; Williamson, 2020; Bobadilla et al., 2021). Although we explored the therapeutic effects of CGA in the acute phase and within 7 days after HIBD, we did not investigate whether CGA provides long-term treatment and protection after HIBD. We hypothesize that by increasing CGA expression in the neonatal brain, the activation of system Xc– to inhibit ferroptosis would substantially improve cognition, memory, and exercise capabilities after HIBD. Furthermore, we confirmed that CGA alleviates the harm caused by HIBD by activating the regulation of the system Xc–/GPX4 axis, reducing the occurrence of ferroptosis. However, the impact of other pathways cannot excluded. Nevertheless, as an effective treatment plan for HIE, our findings suggest that CGA may be an inexpensive and effective agent to alleviate the harm caused by HIBD. Further exploration of the relationship between HIE and ferroptosis may provide novel ideas for effective treatments for HIE.
In conclusion, this work confirms the neuroprotective impact of CGA on HIE and its potential mechanism of action on HIE through ferroptosis. Although we explored the therapeutic effects of CGA in the acute phase and within seven days after HIBD, we did not investigate whether CGA provides long-term treatment and protection after HIBD. Furthermore, the specific target of CGA still needs further study. Nevertheless, without an effective treatment plan for HIE, CGA may be a current strategy to alleviate the harm caused by HIBD. Exploring the relationship between ferroptosis and HIE may provide insights into new treatment options.
Author contributions:Study conception and design: LYL, QW, XQF; main experiment implementation and manuscript draft: LYL, QW; result explain and manuscript modifying: PJL, ZLL; Figure 1 experiment implementation: JJL, XYW; Figure 2 experiment implementation: LD, ZLL; data analysis: YHZ, TYS, WL; study supervision and manuscript revision: XQF. All authors joined in discussing the results, and reading the manuscript.
Conflicts of interest:The authors declare that they have no conflict of interest.
Availability of data and materials:All data generated or analyzed during this study are included in this published article and its supplementary information files.
Open peer reviewers:Yiran Zhu, Zhengzhou University, China; Junfan Chen, The Chinese University of Hong Kong, China.
Additional file:Open peer review reports 1 and 2.
P-Reviewers: Zhu Y, Chen J; C-Editor: Zhao M; S-Editors: Yu J, Li CH; L-Editors: Yu J, Song LP; T-Editor: Jia Y
References
1. Bannai S, Kitamura E 1980 Transport interaction of L-cystine and L-glutamate in human diploid fibroblasts in culture J Biol Chem 255 2372 2376
2. Bao L, Li J, Zha D, Zhang L, Gao P, Yao T, Wu X 2018 Chlorogenic acid prevents diabetic nephropathy by inhibiting oxidative stress and inflammation through modulation of the Nrf2/HO-1 and NF-ĸB pathways Int Immunopharmacol 54 245 253
3. Barbariga M, Curnis F, Andolfo A, Zanardi A, Lazzaro M, Conti A, Magnani G, Volontè MA, Ferrari L, Comi G, Corti A, Alessio M 2015 Ceruloplasmin functional changes in Parkinson's disease-cerebrospinal fluid Mol Neurodegener 10 59
4. Bennet L, Roelfsema V, Pathipati P, Quaedackers JS, Gunn AJ 2006 Relationship between evolving epileptiform activity and delayed loss of mitochondrial activity after asphyxia measured by near-infrared spectroscopy in preterm fetal sheep J Physiol 572 141 154
5. Bennet L, Tan S, Van den Heuij L, Derrick M, Groenendaal F, van Bel F, Juul S, Back SA, Northington F, Robertson NJ, Mallard C, Gunn AJ 2012 Cell therapy for neonatal hypoxia-ischemia and cerebral palsy Ann Neurol 71 589 600
6. Bobadilla M, García-Sanmartín J, Martínez A 2021 Natural food supplements reduce oxidative stress in primary neurons and in the mouse brain, suggesting applications in the prevention of neurodegenerative diseases Antioxidants (Basel) 10 46
7. Bogdan AR, Miyazawa M, Hashimoto K, Tsuji Y 2016 Regulators of iron homeostasis:new players in metabolism, cell death, and disease Trends Biochem Sci 41 274 286
8. Bouayed J, Rammal H, Dicko A, Younos C, Soulimani R 2007 Chlorogenic acid, a polyphenol from Prunus domestica (Mirabelle), with coupled anxiolytic and antioxidant effects J Neurol Sci 262 77 84
9. Bridges RJ, Natale NR, Patel SA 2012 System xc? cystine/glutamate antiporter:an update on molecular pharmacology and roles within the CNS Br J Pharmacol 165 20 34
10. Brigelius-Flohé R, Maiorino M 2013 Glutathione peroxidases Biochim Biophys Acta 1830 3289 3303
11. Chalak LF, Sánchez PJ, Adams-Huet B, Laptook AR, Heyne RJ, Rosenfeld CR 2014 Biomarkers for severity of neonatal hypoxic-ischemic encephalopathy and outcomes in newborns receiving hypothermia therapy J Pediatr 164 468 474.e1
12. Dabkowski ER, Williamson CL, Hollander JM 2008 Mitochondria-specific transgenic overexpression of phospholipid hydroperoxide glutathione peroxidase (GPx4) attenuates ischemia/reperfusion-associated cardiac dysfunction Free Radic Biol Med 45 855 865
13. Deng HF, Yue LX, Wang NN, Zhou YQ, Zhou W, Liu X, Ni YH, Huang CS, Qiu LZ, Liu H, Tan HL, Tang XL, Wang YG, Ma ZC, Gao Y 2020 Mitochondrial iron overload-mediated inhibition of Nrf2-HO-1/GPX4 assisted ALI-induced nephrotoxicity Front Pharmacol 11 624529
14. Devos D, Moreau C, Devedjian JC, Kluza J, Petrault M, Laloux C, Jonneaux A, Ryckewaert G, Garçon G, Rouaix N, Duhamel A, Jissendi P, Dujardin K, Auger F, Ravasi L, Hopes L, Grolez G, Firdaus W, Sablonnière B, Strubi-Vuillaume I, et al. 2014 Targeting chelatable iron as a therapeutic modality in Parkinson's disease Antioxid Redox Signal 21 195 210
15. Douglas-Escobar M, Weiss MD 2015 Hypoxic-ischemic encephalopathy:a review for the clinician JAMA Pediatr 169 397 403
16. Douglas-Escobar M, Yang C, Bennett J, Shuster J, Theriaque D, Leibovici A, Kays D, Zheng T, Rossignol C, Shaw G, Weiss MD 2010 A pilot study of novel biomarkers in neonates with hypoxic-ischemic encephalopathy Pediatr Res 68 531 536
17. Friedmann Angeli JP, Schneider M, Proneth B, Tyurina YY, Tyurin VA, Hammond VJ, Herbach N, Aichler M, Walch A, Eggenhofer E, Basavarajappa D, Rådmark O, Kobayashi S, Seibt T, Beck H, Neff F, Esposito I, Wanke R, Förster H, Yefremova O, et al. 2014 Inactivation of the ferroptosis regulator Gpx4 triggers acute renal failure in mice Nat Cell Biol 16 1180 1191
18. Fuchs Y, Steller H 2011 Programmed cell death in animal development and disease Cell 147 742 758
19. Gardener SL, Rainey-Smith SR, Villemagne VL, Fripp J, Doré V, Bourgeat P, Taddei K, Fowler C, Masters CL, Maruff P, Rowe CC, Ames D, Martins RN 2021 Higher coffee consumption is associated with slower cognitive decline and less cerebral Aβ-amyloid accumulation over 126 months:data from the Australian Imaging, Biomarkers, and Lifestyle Study Front Aging Neurosci 13 744872
20. Han J, Miyamae Y, Shigemori H, Isoda H 2010 Neuroprotective effect of 3,5-di-O-caffeoylquinic acid on SH-SY5Y cells and senescence-accelerated-prone mice 8 through the up-regulation of phosphoglycerate kinase-1 Neuroscience 169 1039 1045
21. Heitman E, Ingram DK 2017 Cognitive and neuroprotective effects of chlorogenic acid Nutr Neurosci 20 32 39
22. Ho L, Varghese M, Wang J, Zhao W, Chen F, Knable LA, Ferruzzi M, Pasinetti GM 2012 Dietary supplementation with decaffeinated green coffee improves diet-induced insulin resistance and brain energy metabolism in mice Nutr Neurosci 15 37 45
23. Huang J, Liu W, Doycheva DM, Gamdzyk M, Lu W, Tang J, Zhang JH 2019 Ghrelin attenuates oxidative stress and neuronal apoptosis via GHSR-1α/AMPK/Sirt1/PGC-1α/UCP2 pathway in a rat model of neonatal HIE Free Radic Biol Med 141 322 337
24. Huo H, Zhou Z, Qin J, Liu W, Wang B, Gu Y 2016 Erastin disrupts mitochondrial permeability transition pore (mPTP) and induces apoptotic death of colorectal cancer cells PLoS One 11 e0154605
25. Koppula P, Zhuang L, Gan B 2021 Cystine transporter SLC7A11/xCT in cancer:ferroptosis, nutrient dependency, and cancer therapy Protein Cell 12 599 620
26. Kumar G, Mukherjee S, Paliwal P, Singh SS, Birla H, Singh SP, Krishnamurthy S, Patnaik R 2019 Neuroprotective effect of chlorogenic acid in global cerebral ischemia-reperfusion rat model Naunyn Schmiedebergs Arch Pharmacol 392 1293 1309
27. Kwon SH, Lee HK, Kim JA, Hong SI, Kim HC, Jo TH, Park YI, Lee CK, Kim YB, Lee SY, Jang CG 2010 Neuroprotective effects of chlorogenic acid on scopolamine-induced amnesia via anti-acetylcholinesterase and anti-oxidative activities in mice Eur J Pharmacol 649 210 217
28. Lapchak PA 2007 The phenylpropanoid micronutrient chlorogenic acid improves clinical rating scores in rabbits following multiple infarct ischemic strokes:synergism with tissue plasminogen activator Exp Neurol 205 407 413
29. Li L, Fang CJ, Ryan JC, Niemi EC, Lebrón JA, Björkman PJ, Arase H, Torti FM, Torti SV, Nakamura MC, Seaman WE 2010 Binding and uptake of H-ferritin are mediated by human transferrin receptor-1 Proc Natl Acad Sci U S A 107 3505 3510
30. Lin W, Zhang T, Zheng J, Zhou Y, Lin Z, Fu X 2022 Ferroptosis is involved in hypoxic-ischemic brain damage in neonatal rats Neuroscience 487 131 142
31. Linkermann A, Skouta R, Himmerkus N, Mulay SR, Dewitz C, De Zen F, Prokai A, Zuchtriegel G, Krombach F, Welz PS, Weinlich R, Vanden Berghe T, Vandenabeele P, Pasparakis M, Bleich M, Weinberg JM, Reichel CA, Bräsen JH, Kunzendorf U, Anders HJ, et al. 2014 Synchronized renal tubular cell death involves ferroptosis Proc Natl Acad Sci U S A 111 16836 16841
32. Loader TB, Taylor CG, Zahradka P, Jones PJ 2017 Chlorogenic acid from coffee beans:evaluating the evidence for a blood pressure-regulating health claim Nutr Rev 75 114 133
33. Lu B, Chen XB, Ying MD, He QJ, Cao J, Yang B 2017 The role of ferroptosis in cancer development and treatment response Front Pharmacol 8 992
34. Lu Q, Harris VA, Kumar S, Mansour HM, Black SM 2015 Autophagy in neonatal hypoxia ischemic brain is associated with oxidative stress Redox Biol 6 516 523
35. Miao M, Xiang L 2020 Pharmacological action and potential targets of chlorogenic acid Adv Pharmacol 87 71 88
36. Mike JK, Ferriero DM 2021 Efferocytosis mediated modulation of injury after neonatal brain hypoxia-ischemia Cells 10 1025
37. Park JJ, Hwang SJ, Park JH, Lee HJ 2015 Chlorogenic acid inhibits hypoxia-induced angiogenesis via down-regulation of the HIF-1α/AKT pathway Cell Oncol (Dordr) 38 111 118
38. Patel SD, Pierce L, Ciardiello A, Hutton A, Paskewitz S, Aronowitz E, Voss HU, Moore H, Vannucci SJ 2015 Therapeutic hypothermia and hypoxia-ischemia in the term-equivalent neonatal rat:characterization of a translational preclinical model Pediatr Res 78 264 271
39. Pfaffl MW 2001 A new mathematical model for relative quantification in real-time RT-PCR Nucleic Acids Res 29 e45
40. Pham K, Mulugeta A, Zhou A, O'Brien JT, Llewellyn DJ, Hyppönen E 2021 High coffee consumption, brain volume and risk of dementia and stroke Nutr Neurosci doi:10.1080/1028415X.2021.1945858
41. Schneider CA, Rasband WS, Eliceiri KW 2012 NIH Image to ImageJ:25 years of image analysis Nat Methods 9 671 675
42. Shah TA, Nejad JE, Pallera HK, Lattanzio FA, Farhat R, Kumar PS, Hair PS, Bass WT, Krishna NK 2017 Therapeutic hypothermia modulates complement factor C3a and C5a levels in a rat model of hypoxic ischemic encephalopathy Pediatr Res 81 654 662
43. Shen W, Qi R, Zhang J, Wang Z, Wang H, Hu C, Zhao Y, Bie M, Wang Y, Fu Y, Chen M, Lu D 2012 Chlorogenic acid inhibits LPS-induced microglial activation and improves survival of dopaminergic neurons Brain Res Bull 88 487 494
44. Sies H 2015 Oxidative stress:a concept in redox biology and medicine Redox Biol 4 180 183
45. Stockwell BR, Friedmann Angeli JP, Bayir H, Bush AI, Conrad M, Dixon SJ, Fulda S, Gascón S, Hatzios SK, Kagan VE, Noel K, Jiang X, Linkermann A, Murphy ME, Overholtzer M, Oyagi A, Pagnussat GC, Park J, Ran Q, Rosenfeld CS, et al. 2017 Ferroptosis:a regulated cell death nexus linking metabolism, redox biology, and disease Cell 171 273 285
46. Tan LL, Jiang XL, Xu LX, Li G, Feng CX, Ding X, Sun B, Qin ZH, Zhang ZB, Feng X, Li M 2021 TP53-induced glycolysis and apoptosis regulator alleviates hypoxia/ischemia-induced microglial pyroptosis and ischemic brain damage Neural Regen Res 16 1037 1043
47. Tošović J, Marković S, DimitrićMarković JM, Mojović M, Milenković D 2017 Antioxidative mechanisms in chlorogenic acid Food Chem 237 390 398
48. Tsikas D 2017 Assessment of lipid peroxidation by measuring malondialdehyde (MDA) and relatives in biological samples:Analytical and biological challenges Anal Biochem 524 13 30
49. Wachtel EV, Verma S, Mally PV 2019 Update on the current management of newborns with neonatal encephalopathy Curr Probl Pediatr Adolesc Health Care 49 100636
50. Wang K, Zhang Z, Tsai HI, Liu Y, Gao J, Wang M, Song L, Cao X, Xu Z, Chen H, Gong A, Wang D, Cheng F, Zhu H 2021 Branched-chain amino acid aminotransferase 2 regulates ferroptotic cell death in cancer cells Cell Death Differ 28 1222 1236
51. Williamson G 2020 Protection against developing type 2 diabetes by coffee consumption:assessment of the role of chlorogenic acid and metabolites on glycaemic responses Food Funct 11 4826 4833
52. Xie Y, Hou W, Song X, Yu Y, Huang J, Sun X, Kang R, Tang D 2016 Ferroptosis:process and function Cell Death Differ 23 369 379
53. Xiong LL, Chen J, Du RL, Liu J, Chen YJ, Hawwas MA, Zhou XF, Wang TH, Yang SJ, Bai X 2021 Brain-derived neurotrophic factor and its related enzymes and receptors play important roles after hypoxic-ischemic brain damage Neural Regen Res 16 1453 1459
54. Xu B, Qin Y, Li D, Cai N, Wu J, Jiang L, Jie L, Zhou Z, Xu J, Wang H 2020 Inhibition of PDE4 protects neurons against oxygen-glucose deprivation-induced endoplasmic reticulum stress through activation of the Nrf-2/HO-1 pathway Redox Biol 28 101342
55. Yager JY, Ashwal S 2009 Animal models of perinatal hypoxic-ischemic brain damage Pediatr Neurol 40 156 167
56. Yagoda N, von Rechenberg M, Zaganjor E, Bauer AJ, Yang WS, Fridman DJ, Wolpaw AJ, Smukste I, Peltier JM, Boniface JJ, Smith R, Lessnick SL, Sahasrabudhe S, Stockwell BR 2007 RAS-RAF-MEK-dependent oxidative cell death involving voltage-dependent anion channels Nature 447 864 868
57. Yan HF, Zou T, Tuo QZ, Xu S, Li H, Belaidi AA, Lei P 2021 Ferroptosis:mechanisms and links with diseases Signal Transduct Target Ther 6 49
58. Yang R, Tian J, Liu Y, Zhu L, Sun J, Meng D, Wang Z, Wang C, Zhou Z, Chen L 2021 Interaction mechanism of ferritin protein with chlorogenic acid and iron ion:The structure, iron redox, and polymerization evaluation Food Chem 349 129144
59. Yang WS, Stockwell BR 2016 Ferroptosis:death by lipid peroxidation Trends Cell Biol 26 165 176
60. Yang WS, SriRamaratnam R, Welsch ME, Shimada K, Skouta R, Viswanathan VS, Cheah JH, Clemons PA, Shamji AF, Clish CB, Brown LM, Girotti AW, Cornish VW, Schreiber SL, Stockwell BR 2014 Regulation of ferroptotic cancer cell death by GPX4 Cell 156 317 331
61. Zhang W, Zhang H, Mu H, Zhu W, Jiang X, Hu X, Shi Y, Leak RK, Dong Q, Chen J, Gao Y 2016 Omega-3 polyunsaturated fatty acids mitigate blood-brain barrier disruption after hypoxic-ischemic brain injury Neurobiol Dis 91 37 46
62. Zhang Y, Yang H, Li S, Li WD, Wang Y 2021 Consumption of coffee and tea and risk of developing stroke, dementia, and poststroke dementia:A cohort study in the UK Biobank PLoS Med 18 e1003830
63. Zhao M, Zhu P, Fujino M, Zhuang J, Guo H, Sheikh I, Zhao L, Li XK 2016 Oxidative stress in hypoxic-ischemic encephalopathy:molecular mechanisms and therapeutic strategies Int J Mol Sci 17 2078
64. Zhu R, Zeng Q, Huang GZ, Zhong Z, Xu YG, Huang ZZ 2021 Ferroptosis and stroke Zhongguo Zuzhi Gongcheng Yanjiu 25 3734 3739
65. Zou R, Xiong T, Zhang L, Li S, Zhao F, Tong Y, Qu Y, Mu D 2018 Proton magnetic resonance spectroscopy biomarkers in neonates with hypoxic-ischemic encephalopathy:a systematic review and meta-analysis Front Neurol 9 732
