2024 年 47 巻 1 号 p. 104-111
White matter lesions induced by chronic cerebral hypoperfusion can cause vascular dementia; however, no appropriate treatments are currently available for these diseases. In this study, we investigated lipid peroxidation, which has recently been pointed out to be associated with cerebrovascular disease and vascular dementia, as a therapeutic target for chronic cerebral hypoperfusion. We used ethoxyquin, a lipid-soluble antioxidant, in a neuronal cell line and mouse model of the disease. The cytoprotective effect of ethoxyquin on glutamate-stimulated HT-22 cells, a mouse hippocampal cell line, was comparable to that of a ferroptosis inhibitor. In addition, the administration of ethoxyquin to bilateral common carotid artery stenosis model mice suppressed white matter lesions, blood–brain barrier disruption, and glial cell activation. Taken together, we propose that the inhibition of lipid peroxidation may be a useful therapeutic approach for chronic cerebrovascular disease and the resulting white matter lesions.
Chronic cerebral hypoperfusion leads to white matter lesions. It has been suggested that cerebral white matter lesions caused by circulatory deterioration are the core pathology of subcortical vascular dementia (VaD) and also contribute to the pathogenesis of Alzheimer’s disease.1,2) In white matter lesion areas, demyelination, axonal loss, and gliosis have been observed.3) To date, mouse models of bilateral common carotid artery stenosis (BCAS) that can recapitulate human white matter lesions have been developed and used to investigate mechanisms related to this pathology. It has been revealed that inflammatory responses and the blood–brain barrier (BBB) disruption contribute to the progression of white matter lesions.4,5) However, it is currently unclear how cerebral hypoperfusion causes these phenomena.
The brain is a highly oxygen-consuming organ.6) During cerebrovascular disease, reactive oxygen species (ROS) are released and oxidative stress level increases in the brain.1) Brain tissue is rich in polyunsaturated fatty acids,7,8) which are readily oxidized by ROS, triggering a chain reaction called lipid peroxidation (LPO).9) High levels of terminal products generated by LPO are found in the plasma of VaD patients.10) Moreover, LPO-dependent cell death, ferroptosis, was suggested to be induced in a cerebral ischemia–reperfusion mouse model, in which a ferroptosis inhibitor improved neurological deficits.11) In addition, oxidized lipids act as damage-associated molecular patterns that activate the immune and inflammatory responses.12) Thus, we hypothesized that LPO inhibition might effectively treat white matter lesions caused by chronic cerebrovascular disease.
Ethoxyquin (6-ethoxy-2,2,4-trimethyl-1,2-dihydroquinoline) is a quinoline-based lipophilic antioxidant (Fig. 1) mainly used to preserve feed quality.13) Recently, ethoxyquin derivatives were reported to have protective effects in a rat cerebral ischemia–reperfusion model.14,15) Furthermore, ethoxyquin appeared to protect against nerve injury during chemotherapy in a diabetic model mouse.16,17) However, its effects on chronic cerebral hypoperfusion remain unclear.
In this study, we used ethoxyquin to investigate whether the suppression of LPO could protect against the pathophysiology of cerebrovascular disease. Using a neuronal cell line and a mouse model of chronic cerebral hypoperfusion, we investigated whether ethoxyquin treatment inhibits LPO and pathological progression.
HT-22 mouse hippocampal cells (Merck, Darmstadt, Germany) were maintained in Dulbecco’s modified eagle medium (DMEM) (Nacalai Tesque, Kyoto, Japan) supplemented with 10% fetal bovine serum (Nichirei Bioscience Inc., Tokyo, Japan), 100 U/mL penicillin, and 100 µg/mL streptomycin (Nacalai Tesque) at 37 °C in a humidified atmosphere with 5% CO2. Erastin (Cayman Chemical, Ann Arbor, MI, U.S.A.), ethoxyquin (FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan), and Ferrostatin-1 (Fer-1; Focus Biomolecules, Plymouth Meeting, PA, U.S.A.) were dissolved in dimethyl sulfoxide (DMSO; Nacalai Tesque) to a concentration of 400-fold from the final concentration, diluted in the medium, and added. Glutamate (FUJIFILM Wako Pure Chemical Corporation) was then dissolved in the medium. If no compound was added during the stimulation, an equal volume of DMSO was added instead.
Cell Viability AssayCell viability was assessed using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. HT-22 cells were seeded in a 96-well culture transparent plate at 100 µL/well to achieve a density of 1 × 104 cells/well. After 6 or 24 h of incubation, 100 µL/well of medium containing double concentration of glutamate or DMSO-dissolved erastin, and each compound was added simultaneously. After 24 h, MTT solution (15 µL/well; Nacalai Tesque) was added and aspirated after 2 h, followed by the addition of DMSO (100 µL/well), and absorbance was measured using the Enspire multimode plate reader (PerkinElmer, Inc., Waltham, MA, U.S.A.) at 570 and 630 nm. The cell viability was determined by comparing it with be the control group (DMSO dissolved in the medium) as 100%.
LPO AssayHT-22 cells (100 µL/well) were seeded in a 96-well black plate with a clear bottom to achieve a density of 1 × 104 cells/well. After 24 h of incubation, 100 µL/well of medium containing double the concentration of glutamate or DMSO-dissolved erastin was added, and each compound was added simultaneously. After 6 h, the medium was aspirated and 100 µL/well of medium containing 2 µM NBD-Pen was added and incubated for 15 min.18) Each well was washed with 100 µL/well Hank’s balanced salt solution (HBSS), and 100 µL/well FluoroBrite DMEM (Gibco, Grand Island, NY, U.S.A.) was added for detection. Fluorescence was measured using the IN Cell Analyzer 2000 (Cytiva, Tokyo, Japan) at wavelengths of λex/em = 495/520 nm. The fluorescence intensity per cell was determined by analyzing the imaging data and LPO levels were presented as fluorescence intensities relative to the control.
Animal ExperimentAll experiments and animal care were approved by the Committee on Ethics of Animal Experiments, Graduate School of Pharmaceutical Sciences, Kyushu University, and performed according to the Guidelines for Animal Experiments of the Graduate School of Pharmaceutical Sciences, Kyushu University.
Male C57BL/6J mice (10–12 weeks old; weighing 24–29 g) were purchased from Jackson Laboratory Japan, Inc. (Kanagawa, Japan). All mice were kept under a 12-h light–dark cycle (lights on from 7:00 to 19:00) with free access to a standard diet (CLEA Japan, Tokyo, Japan) and water.
BCASMice were subjected to the BCAS procedure by attaching microcoils (specifications: piano wire diameter 0.08 mm, coiling pitch 0.5 mm, total length 2.5 mm, inner diameter 0.16 mm and 0.18 mm; purchased from SAMINI Co., Ltd., Shizuoka, Japan) with a modified method reported in a previous study.4) The mice were anesthetized with 2–3% isoflurane and maintained on 1.5% isoflurane using a face mask. After making a midline skin incision, the common carotid artery (CCA) was exposed and isolated bilaterally. Then, 0.16 and 0.18 mm microcoils were applied to the right and left CCAs, respectively, and the incised region was sutured. The treated mice were observed until awakened and then, allowed to feed and water ad libitum. In sham mice, the same procedure was performed without CCA stenosis.
BCAS mice were divided into several groups: (1) vehicle, (2) ethoxyquin (100 µmol/kg), and (3) Tempol 5 mM. The solvent for ethoxyquin was 50% PEG-300 (FUJIFILM Wako Pure Chemical Corporation) in phosphate-buffered saline (PBS) and for Tempol was water. Vehicle or ethoxyquin was administered orally 30 min before BCAS operation and three times weekly after BCAS. Tempol (Tokyo Chemical Industry, Tokyo, Japan) was administered ad libitum via drinking water 3 d prior to surgery.
RNA Extraction and Quantitative RT-PCRTotal RNA was extracted from the frozen brains (corpus callosum) using ISOGEN II (Nippon Gene, Tokyo, Japan). The expression of mRNA level was measured using ReverTra Ace® qPCR RT Master Mix (TOYOBO Co., Ltd., Osaka, Japan) and THUNDERBIRD® SYBR® qPCR Mix (TOYOBO Co., Ltd.) with CFX Connect Real-Time PCR Detection System (Bio-Rad Laboratories, Inc., Hercules, CA, U.S.A.) using the thermal cycle condition as follows: one cycle of 95 °C for 60 s, followed by 40 cycles of 95 °C for 15 s and 55–60 °C for 60 s, with a final stage of melting curve analysis. Gapdh was used for normalization, and relative gene expression was calculated using the 2−ΔΔCt method. The primers used in this study were as follows: Gapdh-F 5′-AATGTGTCCGTCGTGGATCTGA-3′, Gapdh-R 5′-GATGCCTGCTTCACCACCTTCT-3′, Hmox1-F 5′-GTCAAGCACAGGGTGACAGA-3′, Hmox1-R 5′-ATCACCTGCAGCTCCTCAAA-3′, Iba1-F 5′-GAAGCGAATGCTGGAGAAA-3′, Iba1-R 5′-GACCAGTTGGCCTCTTGTGT-3′, Gfap-F 5′- AGGCAGAAGCTCCAAGATGA-3′, Gfap-R 5′-TGTGAGGTCTGCAAACTTGG-3′, and Il1b-F 5′-GCAACTGTTCCTGAACTCAACT-3′, Il1b-R 5′-ATCTTTTGGGGTCCGTCAACT-3′
Protein Extraction and Western BlottingHippocampal and white matter (corpus callosum and striatum) tissues were homogenized in a lysis buffer (50 mM Tris–HCl, pH 7.5; 150 mM NaCl; 0.1% sodium dodecyl sulfate (SDS); 1% Triton X-100; and 1% sodium deoxycholate) containing phenylmethylsulfonyl fluoride (PMSF), NaF, sodium orthovanadate, and a protease inhibitor cocktail. The homogenates were sonicated on ice, incubated for 30 min, and centrifuged (16000 × g, 4 °C, 10 min). The supernatants were used to assay protein contents using the BCA method using the Pierce®BCA™ Protein Assay kit (Thermo Fisher Scientific, Waltham, MA, U.S.A.). Briefly, protein samples (20 µg) were mixed with loading buffer (55.0% glycerol, 0.05% bromophenol blue, 158.9 mM Tris–HCl (pH 6.8), and 4.76% SDS), and were electrophoresed using SDS-polyacrylamide gel electrophoresis (PAGE). Then, protein samples were transferred to polyvinylidene difluoride (PVDF) membranes (0.45 or 0.20 µm, Millipore, Burlington, MA, U.S.A.), and blots were blocked with Blocking one (Nacalai Tesque). The membranes were incubated overnight at 4 °C with the following primary antibodies: anti-Ionized calcium-binding adapter molecule 1 (IBA-1) (Wako, [1/2000]), anti-glial fibrillary acidic protein (GFAP) (GeneTex, Irvine, CA, U.S.A. [1/2000]), anti-matrix metalloproteinase-9 (MMP-9) (Abcam, Cambridge, U.K. [1/1000]), anti-4-hydroxy-2-nonenal (HNE) (Abcam [1/2000]), or anti-glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (MBL, Tokyo, Japan [1/4000]). Thereafter, the membranes were incubated with appropriate secondary antibodies for 1 h at room temperature. Each antibody was diluted in Can Get Signal solution (TOYOBO Co., Ltd.). EzWest Lumi Plus (ATTO, Tokyo, Japan) was used as the detection reagent. Luminescence intensity was calculated using ImageLab (Bio-Rad Laboratories), normalized to GAPDH levels, and evaluated as a ratio to the sham group.
Statistical AnalysesData are expressed as the mean ± standard error of the mean for each group. Statistical significance was evaluated using one-way ANOVA, followed by Dunnett’s multiple comparison test. p < 0.05 was considered statistically significant. Analyses were performed using GraphPad Prism version 9.0 (GraphPad Software, San Diego, CA, U.S.A.).
We first examined whether ethoxyquin protected against cytotoxicity induced by stimuli during cerebrovascular damage in a mouse hippocampal cell line HT-22. During cerebral vascular injury, glutamate is released from damaged neurons.19) Excess glutamate induces excitotoxicity via N-methyl-D-aspartate (NMDA) receptors or cytotoxicity upon glutathione depletion resulting from cystine efflux via cystine/glutamate antiporter (xCT).20,21) HT-22 cells lack NMDA receptors but express xCT, resulting in the glutamate-stimulated enhancement of intracellular LPO and subsequent cell death.22) Fer-1, a lipophilic radical-trapping antioxidant, was added for the positive control group and evaluated as previously described.23) First, we evaluated the cytoprotective potential of each compound using the xCT inhibitor erastin. Erastin-stimulated HT-22 cell death was entirely inhibited by ethoxyquin, as measured using the MTT assay (Fig. 2a). Furthermore, 24 h after glutamate stimulation ethoxyquin completely inhibited HT-22 cell death (Fig. 2b). In addition, the increase in LPO levels measured by NBD-Pen, at 6 h after treatment, when glutamate-induced cell death had not yet occurred, was reduced by treatment with ethoxyquin (Fig. 2c, Supplementary Fig. 1). These inhibitory effects were comparable to those of Fer-1 (Fig. 2).
(a, b) MTT assay results when cells were stimulated with erastin or glutamate; HT-22 cells were stimulated with 0.25 µM of erastin or 5 mM of glutamate, with 2 µM of each compound added. Cell viability was expressed as a percentage of the absorbance of the control as 100%. (c) LPO level during glutamate treatment. HT-22 cells were treated with NBD-Pen for 6 h after addition of glutamate and each compound, and then, the fluorescence intensity per cell was calculated. LPO levels were represented as a ratio of control to 1. (n = 3–6, ** p < 0.01, *** p < 0.001 compared with glutamate treatment group analyzed using one-way ANOVA with Dunnett’s multiple comparison test). EQ, ethoxyquin.
The BCAS model, a chronic cerebral hypoperfusion model, recapitulates clinical white matter injury.4) We orally administered ethoxyquin to mice with BCAS to evaluate its protective effects on pathology. Tempol, a free radical scavenger previously reported as pathoprotective in this model, was administered via drinking water as a positive control24) (Fig. 3). The decrease in myelin-associated glycoprotein (MAG) level, a marker of white matter injury that occurs 4 weeks after BCAS,25) was prevented by treatment with ethoxyquin (Fig. 4a). As oxidative stress level increased during chronic cerebral hypoperfusion, we evaluated changes in heme oxygenase-1 (Hmox1) mRNA expression. Increased Hmox1 expression level 1 week after BCAS was suppressed by ethoxyquin (Fig. 4b). The post-BCAS increase in levels of 4-HNE-modified proteins, the end products of LPO, was assayed using Western blot. The results showed that the levels of 4-HNE-modified proteins increased in the hippocampus 1 week after cerebral hypoperfusion and were significantly decreased upon the administration of ethoxyquin (Fig. 4c).
Ethoxyquin was administered orally at 100 µmol/kg 30 min before and 3 times a week after the BCAS surgery. Tempol was dissolved in water (5 mM) and provided ad libitum for 3 d before surgery. Tempol water was prepared every 2 d.
(a) Western blot of myelin component protein MAG in white matter areas 4 weeks after BCAS surgery. (n = 5–9, * p < 0.05 compared with BCAS vehicle group). (b) mRNA expression of oxidative stress-related factor Hmox1 in white matter regions at 1 week after BCAS surgery. mRNA expression was normalized to Gapdh expression in each sample. (n = 8–12, * p < 0.05, * p < 0.01 compared with BCAS vehicle group). (c) Western blot of 4-HNE modified protein (50 kDa band) in the hippocampus, 1 week after BCAS surgery. (n = 3, * p < 0.05, * p < 0.01 compared with BCAS vehicle group) Each significance test was performed using one-way ANOVA with Dunnett’s multiple comparison test.
Chronic hypoperfusion increases inflammatory response levels through the activation of glial cells, such as astrocytes and microglia, along with the disruption of the BBB.4,26) Oxidized lipids have also been reported to trigger glial cell activation.12,27) Therefore, we investigated whether LPO suppression by ethoxyquin inhibited glial cell activation in a BCAS mouse model. The protein levels of MMP-9, which is believed to contribute to BBB disruption during cerebrovascular pathology, were estimated. MMP-9 protein expression, which increased in white matter regions one week after BCAS surgery, was suppressed in mice treated with ethoxyquin (Fig. 5). We then assessed changes in the expression of glial cell activation markers and inflammatory cytokines in the BCAS model mice. Evaluation of mRNA and protein expression levels showed increased activation markers for astrocyte and microglia after 1 week of BCAS and were significantly suppressed by ethoxyquin (Figs. 6a, b, d, e). Furthermore, the mRNA expression of interleukin (IL)-1β (Il1b), which was increased in the BCAS model mice, was also decreased in the ethoxyquin-treated group (Fig. 6c). Even when half dose of ethoxyquin (50 µmol/kg) was administered, the increased protein expression of the glial cell markers GFAP and IBA1 was attenuated in the BCAS model mice (Figs. 6d, e).
Western blot analysis of a matrix-degrading enzyme MMP-9 in white matter regions 1 week after BCAS surgery. (n = 7–8, * p < 0.05, compared with the BCAS vehicle group analyzed using one-way ANOVA with Dunnett’s multiple comparison test).
(a–c) Quantitative RT-PCR analysis of the expression of astrocytic marker Gfap (a), microglial marker Iba1 (b), and inflammatory cytokine Il1b (c) in white matter regions 1 week after BCAS surgery. mRNA expression was normalized to Gapdh. (n = 8–12, * p < 0.05, ** p < 0.01, *** p < 0.001 compared with BCAS vehicle group). (d, e) Western blot analysis of GFAP (d) and IBA1 (e) in white matter regions 1 week after BCAS surgery. (n = 6–7, * p < 0.05, * p < 0.01, *** p < 0.001 compared with BCAS vehicle group) Each significance test was performed using one-way ANOVA with Dunnett’s multiple comparison test.
In this study, we treated glutamate-stimulated neuronal cell line and mouse model of BCAS with the lipid-soluble antioxidant ethoxyquin. The results showed that ethoxyquin inhibits LPO, protecting against cell death, inflammatory responses, and white matter lesions. These suggest that LPO inhibition is a promising therapeutic strategy for the treatment of cerebrovascular disorders, particularly chronic cerebral hypoperfusion.
Although the involvement of oxidative stress in cerebrovascular diseases has been previously reported, most studies have focused on ROS.28) The brain contains lipids that are highly susceptible to oxidation, and ROS generation may result in the formation of oxidized lipids. Additionally, the biological activities of lipid oxidation products and oxidized lipid-dependent cell death have been reported in recent years.29,30) Ethoxyquin is highly lipophilic; therefore, even when administered orally, it easily crosses the BBB and prevents the formation of oxidized lipids in the brain of pathological mouse models.31)
Ethoxyquin protects against peripheral neuropathy in type II diabetes and neurotoxicity during chemotherapy.16,17) Moreover, it exerts a protective effect by inhibiting ferroptosis in a doxorubicin-induced cardiomyopathy model.32) Neurons are susceptible to oxidative stress, and lipids are easily oxidized by ROS.33) In the brain, inhibition of ferroptosis resulted in a neuroprotective effect in ischemia–reperfusion model mice.11) Fer-1, commonly used to inhibit ferroptosis, has low metabolic stability and is unsuitable for clinical application.34) In this study, we treated a group of BCAS mice with Tempol, a free radical scavenger, which showed a similar protective effect. However, Tempol exhibits a short half-life in vivo.35) In contrast, ethoxyquin demonstrated a protective effect when administered 3 times weekly.
In this study, 4-HNE-modified proteins were used as indicators of LPO generation in BCAS mice. 4-HNE is mainly produced by the oxidation of arachidonic acid,36) and arachidonic acid-containing phospholipids are concentrated in the hippocampus.37) 4-HNE is believed to modify various proteins and alter their functions.9) It has been reported that 4-HNE activates toll-like receptor 4, which is expressed on immune cells, and induces proinflammatory cytokine release.38) Further investigations on 4-HNE are required with respect to whether it is directly or indirectly involved in cellular responses and whether it is generated in white matter regions and its pathological mechanism in the hippocampus.
Previous studies have shown that inhibiting microglial and astrocyte activation during chronic hypoperfusion can suppress white matter lesions and subsequent cognitive impairment.39,40) Additionally, oxidized lipids affect immune and glial cells, causing inflammatory response.12,27) In this study, glial cell activation was inhibited by the administration of ethoxyquin, an LPO inhibitor. However, the molecular structure of the oxidized lipids that activate glial cells remains unclear. It is necessary to elucidate the relationship between oxidized lipids generated during pathogenesis and glial cell responses using structural analysis and other methods.
A further limitation of this study is about the dosage of ethoxyquin. The dose of ethoxyquin used in the mice in this study relative to that used in humans (1.76 mg/kg; converted from 100 µmol/kg [21.7 mg/kg], in accordance with U.S. Food and Drug Administration (FDA) guidance41)) was higher than the published acceptable daily intake (0–0.005 mg/kg) or acute reference dose (0.5 mg/kg), as reported by the WHO (https://apps.who.int/pesticide-residue-jmpr-database). In this study, these compounds were used at high doses to identify the therapeutic targets by inhibiting LPO. However, in our laboratory, a reduction in glial cell activation in the white matter was observed even at half the dose (Figs. 6d, e). Furthermore, optimizing the number of doses and route of administration and the development of potent derivatives of ethoxyquin may provide protection at lower doses.
In conclusion, our results suggested that ethoxyquin protects against the progression of chronic cerebrovascular disease by inhibiting LPO. Collectively, LPO may be responsible for the progression of this disease in multiple targets. Cellular damage and an increased inflammatory response caused by chronic cerebral hypoperfusion may lead to long-term deterioration of brain function and cognitive decline. Based on the protective effects of ethoxyquin against various disease states, we propose LPO as a therapeutic target for chronic cerebrovascular disorders and associated VaD.
This study was supported in part by an AMED-CREST Grant (JP22gm0910013 to KY) and JSPS KAKENHI Grants (23H05481, 22H05572, and 20H00493 to KY). This study was also supported by the Platform Project for Supporting Drug Discovery and Life Science Research of AMED. We appreciate the technical support provided by the Research Support Center of the Graduate School of Medical Sciences at Kyushu University.
The authors declare no conflict of interest.
This article contains supplementary materials.
