Degradation of LOX-1 via Neuronal Autophagy Alleviates Cerebral Ischemia/Reperfusion Injury in Mice

Abstract

Lectin-like oxidized low-density lipoprotein receptor-1 (LOX-1) is critically involved in atherosclerotic plaque formation. Recent clinical evidence suggest that the LOX-1 polymorphisms are linked to the susceptibility and outcomes of ischemic stroke (IS). However, the mechanisms through which LOX-1 influences cerebral ischemia/reperfusion (I/R) injury are not fully understood. In this study, blood samples were obtained from 295 patients with acute IS (AIS), and parallel experimental models including middle cerebral artery occlusion and reperfusion (MCAO/R) in male C57BL/6 mice and oxygen glucose deprivation (OGD) in primary neurons, were used to simulate cerebral I/R injury. Elevated serum soluble LOX-1 (sLOX-1) levels were detected in patients with AIS and showed positive correlations with both infarct volume and the 3-month modified Rankin scale (mRS) score. In mice, LOX-1 overexpression increased infarct volume, aggravated neurological deficits, and reduced cerebral blood flow (CBF), whereas LOX-1 knockdown produced the opposite effects. Co-immunoprecipitation demonstrated that LOX-1 interacts with key autophagy-lysosome pathway proteins (LC3B, P62, and LAMP2) in primary neurons. Western blotting, immunofluorescence and transmission electron microscopy showed that LOX-1 overexpression suppresses neuronal autophagic flux in peri-infarct brain tissue, while LOX-1 knockdown restores it. Moreover, early activation of autophagy with rapamycin (RAPA) promoted LOX-1 degradation, reduced infarct volume, improved neurological deficits, and restored CBF following cerebral I/R. Collectively, these findings indicated that LOX-1 exacerbates cerebral I/R injury by inhibiting neuronal autophagic flux. Early autophagy activation may therefore represent a promising neuroprotective strategy by facilitating LOX-1 degradation during acute cerebral ischemia.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12035-026-05730-1.

Introduction

Ischemic stroke exhibits high incidence, mortality, and disability, all of which seriously threaten human health and place a significant burden on patients’ lives [1]. However, the currently available vascular recanalization therapy has a limited time window and may cause adverse effects, such as cerebral hemorrhage and reperfusion injury [2, 3]. Therefore, further exploration of adjunct and alternative neuroprotective strategies is required to prevent or alleviate neuronal injury.

Lectin-like oxidized low-density lipoprotein receptor-1 (LOX-1) is a type II membrane glycoprotein belonging to the integrin family, typically expressed at low levels under physiological conditions [4]. However, LOX-1 expression increases rapidly under various pro-oxidative and inflammatory stimulation, thereby contributing to cellular damage [5]. LOX-1 mediates inflammation and oxidative stress in endothelial cells and induces endothelial apoptosis, ultimately impairing the blood brain barrier (BBB) integrity [6]. In addition, LOX-1 activates macrophages by binding to oxidized LDL (ox-LDL), promoting matrix metalloproteinase (MMP) release, facilitating atherosclerotic plaque rupture, and consequently increasing the risk of ischemic stroke [7–9]. Recent clinical studies further demonstrate that LOX-1 expression is closely associated with the occurrence, severity, and prognosis of cerebral ischemia. In 2013, Zhang et al. identified that the rs11053646 polymorphism of LOX-1 could serve as a potential susceptibility and prognostic marker of ischemic stroke in Chinese populations [10]. A subsequent genomics study reported that the rs1050283T allele of LOX-1 is significantly linked to a higher risk of cerebral ischemia [11]. In addition, the hydrolyzed form of LOX-1, soluble LOX-1 (sLOX-1), is upregulated in the blood of stroke patients [12] and shows a positive correlation with the degree of intracranial arterial stenosis [13]. sLOX-1 levels are also positively correlated with multiple inflammatory markers in circulation [14], supporting its potential utility as a predictor of long-term prognosis in ischemic stroke. However, despite these clinical associations, the direct pathogenic roles or mechanistic effects of LOX-1 during cerebral ischemia remain largely unexplored.

In this study, we investigated the expression of LOX-1 in peri-infarct brain tissue during the acute phase of cerebral ischemia, as well as the functional effects and underlying mechanisms by which LOX-1 contributes to cerebral ischemic injury. In addition, we examined whether early autophagy activation influences LOX-1 expression and modulates the neuronal injury response following cerebral ischemia.

Materials and Methods

Study Participants and Clinical Data Collection

This study enrolled 342 consecutive patients from the Department of Emergency at Xuanwu Hospital of Capital Medical University. The study protocol adhered to the principles of the Declaration of Helsinki, and written informed consent was obtained from all participants.

Patients were enrolled according to the following criteria: (1) clinical symptoms meeting the diagnostic criteria for cerebral infarction (main symptoms included cognitive impairment, gaze disturbance, speech impairment, motor disability, or visual field loss/neglect); (2) age ≥ 18 years; (3) acute ischemic stroke (AIS) with admission within 12 hours post-stroke; (4) 4 ≤ NIHSS score < 20; and (5) voluntary participation with provision of written informed consent. Patients with (1) cerebral hemorrhage diagnosed via CT or MRI or cerebral infarction secondary to trauma, tumor, abnormal coagulation mechanism, or aneurysm/arteriovenous malformations, and those with (2) serious internal medical conditions such as respiratory or circulatory failure, renal insufficiency, or malignancy were excluded. After applying these criteria, 295 patients with AIS were included in the study. Among them, patients with a score of ≤ 15 were diagnosed with mild-to-moderate ischemic stroke (MMIS), and those with a score of 16–20 were designated as having severe ischemic stroke (SIS). Twenty age- and sex-matched healthy participants were included as controls. Clinically collected data included blood test indicators (e.g., routine blood tests at admission), imaging data for cerebral infarction (MRI/CT), National Institutes of Health Stroke Scale (NHISS) score at admission, and modified Rankin Scale (mRS) score at 3 months.

Blood Samples and Enzyme-linked Immunosorbent Assay (ELISA) Detection

Blood samples were obtained from patients with ischemic stroke and healthy controls; samples from patients were obtained upon admission within 12 hours of stroke onset. The separated serum was stored at –80 ℃. Serum sLOX-1 levels were measured using a human sLOX-1 ELISA kit (Solarbio Life Science, Beijing, China). All assays were conducted by certified laboratory technicians in strict accordance with the manufacturer’s instructions, with technicians remaining blinded to the clinical information of the subjects throughout the testing process.

Primary Neurons Culture

Cortical neurons were obtained from the cerebral cortex of embryonic day 18 (E18) fetuses. The meninges and microvessels were carefully removed under a microscope, and the cortex was isolated. The cortex was digested with 0.025% trypsin (Gibco, USA) for 5 minutes, followed by filtration, centrifugation, resuspension, and trituration to obtain a single-cell suspension. Following cell counting, neurons were plated onto poly-L-lysine-coated (Sigma, USA) culture dishes at an initial density of approximately 1.0 × 10⁶ cells/mL, using Dulbecco’s modified Eagle medium (DMEM, Gibco, USA) supplemented with 10% fetal bovine serum (FBS, Gibco, USA). Cells were then cultured in neurobasal medium supplemented with 2% B27 (Gibco, USA), 0.25% glucose, and 0.5% penicillin/streptomycin after 6 hours. Cultures were maintained at 37 °C under a humidified atmosphere of 5% CO₂/95% O₂, with the medium refreshed every 3 days.

To initiate oxygen glucose deprivation (OGD), primary neurons were transferred to glucose-free medium and placed in a hypoxia incubator chamber flushed with 95% N₂/5% CO₂ at 37 ℃ for 3 hours. Rapamycin (RAPA; Selleck Chemicals, Houston, TX, USA), an autophagy activator that initiates autophagy and promotes autophagosome nucleation by specifically inhibiting mTOR signaling [15], was dissolved in saline to a concentration of 100 nmol/L, and immediately administered after OGD.

Intracerebroventricular Injection

Adult male C57BL/6 mice (Si Pei Fu Biotechnology Co., Ltd., Beijing) weighing approximately 20–22 g were used to establish the middle cerebral artery (MCA) occlusion (MCAO)/reperfusion (MCAO/R) model. A gas mixture of 5% isoflurane and N₂O/O₂ (70%/30%) was used to induce anesthesia, followed by a gas mixture of 2% isoflurane and N₂O/O₂ (70%/30%) to maintain anesthesia. The mouse head was secured on a stereotaxic apparatus, and a midline scalp incision was made to expose the skull. The bregma served as the reference point, and the injection point was located 1 mm posterior and 1 mm lateral (right) to the bregma. The injection depth was set at 2.5 mm, excluding the needle tip length.

For viral delivery, 5 μL of Lenti-LOX-1-GFP or its control Lenti-GFP was injected into the lateral ventricle 7 days before MCAO to induce LOX-1 overexpression. Similarly, 5 μL of siRNA-NC or its control LOX-1- siRNA was injected to achieve LOX-1 knockdown. RAPA was dissolved in saline to a concentration of 500 nmol/L, and 2 μL RAPA was administered via the lateral ventricle immediately after ischemia/reperfusion (I/R).

Mouse Model of Middle Cerebral Artery Occlusion

Seven days after viral injection via the lateral ventricle, a mouse model of MCAO was created using the intraluminal filament method as previously described [16]. Briefly, cerebral blood flow (CBF) was interrupted by occluding MCA with a suture for 45 minutes, following which the filament was withdrawn to allow reperfusion. The body temperature of mice was maintained at 37 ℃ ± 0.5 ℃ during surgery. Sham-operated animals underwent identical surgical exposure without insertion of the occluding suture.

The mice were randomly assigned to the following groups: (1) sham, Lenti-GFP, and Lenti-LOX-1-GFP; (2) sham, siRNA-NC, and LOX-1- siRNA; (3) sham, MCAO, and MCAO + RAPA; and (4) sham, Lenti-LOX-1-GFP, and Lenti-LOX-1-GFP + RAPA.

Cerebral Blood Flow

Midline scalp incision was performed to fully expose the skull. The mouse head was fixed, and changes in global CBF were monitored using a laser speckle imaging system, with images acquired over a 20-second period. CBF was measured at the following time points: pre-MCAO, during MCAO, during reperfusion and 24 hours post-MCAO/R. Saline was used to keep the skull moist. The bilateral middle cerebral artery supply area was selected as the region of interest (ROI), and the relative changes in CBF were calculated using the following formula:

$$\text{rCBF\%}=\frac{\text{CBF value in ROI of infracted side}}{\text{CBF value in ROI of normal side}}$$

Neurological Function Score

Neurological deficits in the mice were evaluated using the balance beam test, modified neurological severity score (mNSS), and Longa score over seven consecutive days after surgery. The balance beam test was used to evaluate motor coordination and limb integration in mice after cerebral ischemia. Preoperative training was performed for three consecutive days to exclude mice with walking defects. Each mouse was placed at the end of the balance beam (length = 120 cm, width = 1 cm, height = 30 cm), and the number of hind limb slips within 1 meters was recorded. Each mouse was tested three times with a 2-minute interval between trials. The mNSS comprehensively evaluates sensation, movement, reflexes, and balance in mice after cerebral ischemia, with higher scores corresponding to more pronounced neurological impairment [17]. The Longa score was used to evaluate limb movement, with elevated scores reflecting more severe neurological dysfunction [18].

Volume Measurement of Cerebral Infarction

Mice were euthanized at 24 hours and 7 days after cerebral ischemic injury, and brain tissue was collected. Five consecutive slices of brain tissue (1-mm thickness) were obtained and placed in 2% 2,3,5-triphenyl tetrazolium chloride (TTC, T8877, Sigma-Aldrich, USA) solution and incubated at 37 ℃ for 10 minutes. Infarct volume was quantified using Image J software. The percentage of brain infarct volume was calculated using the following equation:

$$(\text{area of the contralateral hemisphere}-\text{non}-\text{infarcted area of the ipsilateral hemisphere})/\text{area of the contralateral hemisphere}\times 100\text{\%}$$

Western Blotting

Radioimmunoprecipitation assay (RIPA) lysis buffer containing phosphatase and protease inhibitors was used to extract proteins from the cerebral peri-infarct region of mice 24 hours after cerebral ischemia. For protein separation, 12% sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (SDS-PAGE) was used, and the separated proteins were subsequently transferred onto nitrocellulose membranes. Nitrocellulose membranes were blocked in 5% skim milk for 2 hours and incubated overnight at 4 ℃ with the following primary antibodies: anti-LC3 (M186‐3, MBL, Japan); anti-P62 (PM045, MBL, Japan); anti-LAMP2 (sc‐18822, Santa Cruz, USA); and anti-β-actin (sc-1616, Santa Cruz, USA). The blots were incubated with horseradish peroxidase-conjugated secondary antibodies (Santa Cruz, Biotechnology) for 2 hours. Protein bands were visualized using an enhanced chemiluminescence detection kit (Millipore, Billerica, MA, USA).

Immunofluorescence Staining

Brain tissue collected from mice 24 hours after cerebral ischemia was fixed in 4% paraformaldehyde for over 48 hours and subsequently embedded in paraffin for sectioning. The brain sections were incubated with the following primary specific antibodies at 4 ℃ for 12 hours: anti-LC3 (PM036, MBL, Japan), anti-P62 (ab56416, Abcam, USA), anti-LAMP2 (sc-18822, Santa Cruz, Biotechnology), LOX-1 (bs-2044R, Bioss), NeuN (MAB377, Millipore), GFAP (ab53554 and ab178680, Abcam), Iba1 (ab5076, Abcam; 16,825–1-AP, Proteintech), and GFP (Rockland, 600–101–215). After washing, the sections were stained with appropriate fluorescent-conjugated secondary antibodies (Santa Cruz Biotechnology).

Transmission Electron Microscopy

Autophagosomes and autolysosomes in neurons of the cerebral peri-infarct region were examined using transmission electron microscopy (TEM). The peri-infarct brain tissue was cut into small pieces measuring 1 mm × 1 mm × 1 mm, which were fixed in 2% glutaraldehyde for 2 hours at room temperature. Ultrathin sections were prepared and imaged by Servicebio Biotechnology Co., Ltd.

Co-immunoprecipitation

Primary neurons were lysed in an appropriate volume of IP lysis buffer at 4 ℃ for 30 minutes, followed by centrifugation at 12000 × g for 30 minutes. A portion of the supernatant was retained for western blotting. To the remaining supernatant, 1 μg of anti-LAMP2 antibody (sc‐18822, Santa Cruz, USA), anti-P62 antibody (PM045, MBL, Japan), or anti-LC3 antibody (M186‐3, MBL, Japan) was added along with 10–50μL of protein A/G beads. The mixture was incubated overnight at 4 ℃. The beads were pelleted by centrifugation at 3000 × g for 5 minutes and the supernatant was carefully removed. The protein A/G beads were washed 3–4 times with 1 mL of lysis buffer. Finally, 5 μL of 5 × SDS loading buffer was added, and the samples were heated at 100 ℃ for 10 minutes before the detection of protein expression.

Statistical Analysis

Statistical analysis was performed using GraphPad Prism 9.5.1 (GraphPad Software, La Jolla, CA, USA). Normality of the data was assessed using the Kolmogorov–Smirnov test. Continuous variables conforming to a normal distribution are expressed as mean ± standard error of the mean (SEM), while non-normally distributed variables are presented as the median with the 25th–75th percentile (Q1–Q3). One-way analysis of variance (ANOVA) was used to compare multiple groups, followed by post hoc Turkey’s test. A p-value of < 0.05 was considered statistically significant.

Results

The Levels of Serum sLOX-1 in Patients with AIS

We analyzed the serum sLOX-1 levels in 295 patients with AIS within 12 hours of stroke onset, of whom 182 had MMIS and 113 had SIS. Compared to healthy controls, the serum sLOX-1 levels in patients with MMIS (p < 0.0001, Fig. 1A) and SIS (p < 0.0001, Fig. 1A) were significantly higher, and the serum sLOX-1 levels in patients with SIS were significantly higher than in those with MMIS (p = 0.020, Fig. 1A). These findings indicated that serum sLOX-1 levels are associated with ischemic stroke severity. We further analyzed the correlation between serum sLOX-1 levels, infarct volume, and neurological function scores in patients with ischemic stroke. Serum sLOX-1 levels were significantly positively correlated with infarct volume (r² = 0.030, p = 0.022, Fig. 1B) and the 3-month mRS score (r² = 0.026, p = 0.039, Fig. 1D), whereas no significant correlation with the NIHSS score at admission was observed (r² = 0.002, p = 0.524, Fig. 1C). In addition, serum sLOX-1 levels were significantly positively correlated with neutrophil (r² = 0.056, p = 0.001, Fig. S1A) and leukocyte (r² = 0.061, p = 0.0009, Fig. S1B) counts. These results suggest that changes in serum sLOX-1 levels may reflect the intensity of LOX-1 expression, which increases rapidly in response to pro-oxidative and pro-inflammatory signals. We further divided patients into with and without reperfusion therapy groups and found no significant correlations between serum sLOX-1 levels and 3-month mRS score in patients with (r² = 0.024, p = 0.136, Fig. S2A) and those without (r² = 0.014, p = 0.287, Fig. S2B) reperfusion therapy.

Fig. 1.

Serum sLOX-1 levels in patients with AIS and healthy controls, and the correlation between serum sLOX-1 levels and infarct volume, NIHSS score at admission and 3-month mRS score in patients with AIS. Serum sLOX-1 levels in healthy controls and patients with mild to moderate ischemic stroke, and those with severe ischemic stroke (A). The correlation between serum sLOX-1 levels and infarct volume in patients with AIS (B). The correlation between serum sLOX-1 levels and NIHSS score at admission in patients with AIS (C). The correlation between serum sLOX-1 levels and 3-month mRS score in patients with AIS (D). sLOX-1, soluble LOX-1; AIS, acute ischemic stroke; mRS, modified Rankin scale; NIHSS, National Institutes Health Stroke Scale

LOX-1 is Involved in Cerebral I/R Injury in Mice

We further investigated the effects of LOX-1 on cerebral ischemic injury in mice. Lenti-LOX-1-GFP was used to induce LOX-1 overexpression. GFP was prominently expressed in neurons (GFP⁺NeuN vs. GFP⁺Iba1, p = 0.155; GFP⁺NeuN vs. GFP⁺GFAP, p = 0.013; GFP⁺Iba1 vs. GFP⁺GFAP, p > 0.999; Fig. S3A,B). LOX-1 showed overexpression (p = 0.014, Fig. S4A, 5 A) and knockdown in neurons (p = 0.0001, Fig. S4B, 5B), with no significant differences observed in Iba1 (p = 0.371, Fig. S4C, 5 C; p > 0.999, Fig. S4D, 5D) and GFAP (p > 0.999, Fig. S4E, 5E; p > 0.999, Fig. S4F, 5 F) after Lenti-LOX-1-GFP and LOX-1-siRNA injection, respectively. The Lenti-LOX-1-GFP group showed a significantly larger infarct volume than did the Lenti-GFP group at 24 hours (p = 0.010, Fig. 2A) and day 7 (p = 0.009, Fig. 2A) following cerebral I/R. Compared to the siRNA-NC group, the LOX-1-siRNA group showed a significant decrease in infarct volume at 24 hours (p = 0.016, Fig. 2B) and day 7 (p = 0.003, Fig. 2B) following cerebral I/R. In addition, the Lenti-LOX-1-GFP group showed a decrease in body weight compared with the Lenti-GFP group, although the difference was not statistically significant (p > 0.05, Fig. 2C) during the 7-day period after cerebral I/R. Compared with the siRNA-NC group, the LOX-1-siRNA group showed an increase in body weight, also without statistical significance (p > 0.05, Fig. 2D). Regarding neurological deficits, the Lenti-LOX-1-GFP group had a higher mNSS score (day 6: p = 0.0204, Fig. 2C) than the Lenti-GFP group. However, the LOX-1-siRNA group had a lower mNSS score than did the siRNA-NC group (day 2: p = 0.0401; day 6: p = 0.0130, Fig. 2D). The balance beam score showed that the Lenti-LOX-1-GFP group had a lower score than did the Lenti-GFP group (day 1: p = 0.0052; day 2: p = 0.0290; day 5: p = 0.0177, Fig. 2C). Conversely, the LOX-1-siRNA group had a higher balance beam score than did the siRNA-NC group (day 3: p = 0.0056, Fig. 2D). The Longa score showed that the Lenti-LOX-1-GFP group had a higher score than did the Lenti-GFP group (day 1: p = 0.030, Fig. 2C), whereas the LOX-1-siRNA group had a lower Longa score than did the siRNA-NC group (day 3: p = 0.0276; day 4: p = 0.043; day 6: p = 0.0192; day 7: p = 0.0350, Fig. 2D). CBF values before MCAO, during MCAO, during reperfusion and 24 hours post-MCAO/R were recorded, with the value at 24 hours post-MCAO/R being significantly lower than that at reperfusion in the Lenti-LOX-1-GFP group (p = 0.025, Fig. 2E). No significant difference was detected in the Lenti-GFP group (p > 0.999, Fig. 2E). In addition, the CBF at 24 hours post-MCAO/R was significantly higher in the Lenti-GFP group than that in the Lenti-LOX-1-GFP group (p = 0.003, Fig. 2E). This indicated that LOX-1 overexpression significantly decreased CBF following cerebral I/R. Moreover, CBF was significantly higher at 24 hours post-MCAO/R than during reperfusion in the LOX-1-siRNA group (p = 0.004, Fig. 2F); this indicated that LOX-1 knockdown ameliorated reductions in CBF.

Fig. 2.

The effects of LOX-1 on infarct volume, neurological deficiency and CBF in mice following cerebral I/R injury. The effect of LOX-1 overexpression on infarct volume in mice at 24 hours and 7 days after cerebral I/R injury (A). The effect of LOX-1 knockdown on infarct volume in mice at 24 hours and 7 days after cerebral I/R injury (B). The effect of LOX-1 overexpression on neurological function deficiency in mice within 7 days after cerebral I/R injury (C). The effect of LOX-1 knockdown on neurological function deficiency in mice within 7 days after cerebral I/R injury (D). The effect of LOX-1 overexpression on CBF in mice within 24 hours after cerebral I/R injury (E). The effect of LOX-1 knockdown on CBF in mice within 24 hours after cerebral I/R injury (F) LOX-1, lectin-like oxidized low-density lipoprotein receptor-1; CBF, cerebral blood flow; I/R, ischemia/reperfusion

The Expression of LOX-1 Affects Neuronal Autophagic Flux in the Peri-infarct Area Following Early Cerebral Ischemia Injury

We investigated the mechanisms by which LOX-1 is involved in ischemic damage in mice. First, we evaluated the distribution of LOX-1 expression in peri-infarct aeras and performed immunofluorescence co-staining of LOX-1 with NeuN, Iba1, and GFAP. We found that LOX-1 exhibited strong co-localization signals with NeuN (Fig. S6A), while only minimal co-localization was observed with Iba1 (Fig. S6B) and GFAP (Fig. S6C). These findings indicated that LOX-1 is specifically highly expressed in peri-infarct neurons and may be associated with neuronal injury.

To further investigate the mechanisms by which LOX-1 is involved in neuronal damage, we utilized a primary neuronal OGD model. In-put results of co-immunoprecipitation (co-IP) showed that OGD-treated neurons showed increased LOX-1 levels, a decreased LC3B levels, and elevated levels of P62 and LAMP2 compared with those shown by untreated primary neurons (Fig. 3A). Subsequent treatment of OGD-exposed neurons with increasing concentrations of the autophagy activator RAPA reduced LOX-1 expression in a concentration-dependent manner. These results indicated that cerebral ischemic injury upregulates LOX-1 expression and suppresses autophagic activity in neurons, whereas autophagy activation inhibits LOX-1 expression. To further investigate the interaction between LOX-1 and autophagy-related proteins, we performed IP-P62 and IP-LAMP2 assays (Fig. 3A). These results demonstrated that LOX-1 interacts with LC3B, P62, and LAMP2 after RAPA treatment. Moreover, they indicated that LOX-1 interacts with proteins related to the autophagy-lysosome pathway, and that the damaging effects of LOX-1 in early cerebral ischemia are associated with neuronal autophagic activity.

Fig. 3.

The association between LOX-1 expression and autophagic flux. Co-immunoprecipitation for detection of LOX-1, LC3B, P62 and LAMP2 expressions and the interactions following OGD and RAPA treatment (A). Western blotting and quantitative analysis of LOX-1, LC3B-Ⅱ, LAMP2, and P62 at 24 hours after cerebral I/R injury with LOX-1 overexpression. n = 3, *p < 0.05, **p < 0.01 vs. sham group; #p < 0.05, ##p < 0.01 vs. Lenti-GFP (one-way ANOVA followed by Tukey’s multiple comparison test) (B). Western blotting and quantitative analysis of LOX-1, LC3B-Ⅱ, LAMP2, and P62 at 24 hours after cerebral I/R injury with LOX-1 knockdown. n = 3, *p < 0.05, **p < 0.01 vs. sham group; #p < 0.05 vs. siRNA-NC (one-way ANOVA followed by Tukey’s multiple comparison test) (C). Representative immunofluorescence images showing co-localization of Neun-stained neurons (green) and LC3-positive cells (red) in the peri-infarct areas from mice at 24 hours after cerebral I/R injury in the LOX-1 overexpression (D) and LOX-1 knockdown (E) groups. DAPI (blue) indicates cell nuclei. The white arrows indicate the co-labeling cells. Scale bar, 50 µm. Representative transmission electron microscopy image showing autophagosomes in the LOX-1 overexpression (F) and LOX-1 knockdown (G) groups. I/R, ischemia/reperfusion; LOX-1, lectin-like oxidized low-density lipoprotein receptor-1; OGD, oxygen–glucose deprivation; RAPA, rapamycin

We further examined the association between LOX-1 and autophagic activity in the peri-infarct region at 24 hours after cerebral ischemia in mice. In line with the results observed in co-IP in in vivo experiments (Fig. 3A), the LOX-1 expression were increased in the Lenti-GFP (Fig. 3B, p = 0.551) and siRNA-NC (Fig. 3C, p = 0.006) groups compared with those in the sham group. Furthermore, expression of LC3B-Ⅱ was decreased in the Lenti-GFP (Fig. 3B, p = 0.041) and siRNA-NC (Fig. 3C, p = 0.002) groups, while the expressions of P62 and LAMP2 were increased in Lenti-GFP (Fig. 3B, P62, p = 0.551; LAMP2, p = 0.642) and siRNA-NC (Fig.  3 C, P62, p = 0.024; LAMP2, p = 0.006) groups compared with those in the sham group. These results indicated that cerebral ischemic injury upregulated LOX-1 expression and suppressed autophagic activity. LC3B-Ⅱ expression (p = 0.041) was significantly reduced in the Lenti-LOX-1-GFP group compared with that in the Lenti-GFP group, while P62 (p = 0.021) and LAMP2 (p = 0.003) expressions were significantly increased; This indicatd that LOX-1 overexpression markedly inhibits neuronal autophagic flux (Fig. 3B). The LOX-1-siRNA group showed significantly decreased P62 (p = 0.042) and LAMP2 (p = 0.006) expressions, and increased LC3B-Ⅱ expression (p = 0.041) compared with those in the siRNA-NC group. Thus, LOX-1 knockdown significantly improved autophagy flux (Fig. 3C). Immunofluorescence staining showed that LC3B and NeuN co-localization was significantly lower in the Lenti-LOX-1-GFP group than in the Lenti-GFP group (p = 0.029, Fig. S7A, Fig. 3D), whereas the LOX-1-siRNA group showed significantly greater co-localization than did the siRNA-NC group (p = 0.0004, Fig. S7B, Fig. 3E). Furthermore, TEM revealed fewer autophagosomes in neurons from the Lenti-LOX-1-GFP group than in those from the Lenti-GFP group (Fig. 3F), although the difference did not reach statistical significance (p = 0.071, Fig. S7C). In contrast, a significantly higher number of autophagosomes was observed in the LOX-1-siRNA group than in the siRNA-NC group (p = 0.002; Fig. S7D; Fig. 3G). Taken together, these results suggested that cerebral ischemic injury induce an increase in LOX-1 expression, the damaging effects of LOX-1 are associated with the inhibition of neuronal autophagic flux. However, LOX-1 upregulation reduced autophagy activity rather than completely supressing it.

Activating Autophagy Following Early Cerebral Ischemia Alleviates Cerebral Ischemic Injury and Inhibits LOX-1 Expression in Mice

Autophagy maintains normal cellular function by clearing abnormal proteins. Autophagy activation has protective effects against cerebral ischemic injury during the early stages of cerebral ischemia [19, 20]. RAPA was immediately injected into the lateral ventricle after cerebral I/R to activate autophagy. The infarct volume was significantly lower in the RAPA group than in the MCAO group at 24 hours (p = 0.009, Fig. 4A) and on day 7 (p = 0.015, Fig. 4B). Body weights were significantly higher in the RAPA group than in the MCAO group during the 7 days following cerebral ischemia (day 6: p = 0.0317; day 7: p = 0.0079, Fig. 4C). In the assessment of neurological deficits, the mNSS was significantly lower in the RAPA group than in the MCAO group (day 1: p = 0.0079; day 2: p = 0.0079; day 4: p = 0.0079; day 5: p = 0.0238; day 6: p = 0.0079; day 7: p = 0.0079, Fig. 4C). The balance beam score was significantly higher in the RAPA group than in the MCAO group (day 1: p = 0.0397; day 3: p = 0.0079; day 4: p = 0.0476; day 5: p = 0.0476, day 6: p = 0.0476; day 7: p = 0.0238, Fig. 4C). The Longa score was significantly lower in the RAPA group than in the MCAO group (day 2: p = 0.0317; day 3: p = 0.0317; day 4: p = 0.0317; day 5: p = 0.0317; day 6: p = 0.0317; day 7: p = 0.0397, Fig. 4C). These results indicated that autophagy activation significantly alleviates neurological deficits. As shown in Fig. 3A, RAPA-induced autophagy activation reduced LOX-1 expression in primary neurons, and LOX-1 interacts with LC3B, P62, and LAMP2 after RAPA treatment. We further validated this finding in a mouse model of cerebral ischemia. Western blotting analysis showed that LOX-1 expression was significantly lower in the RAPA group than in the MCAO group (p = 0.031, Fig. 4D). Immunofluorescence analysis showed significantly lower co-localization of LOX-1 and NeuN in the RAPA group than in the MCAO group (p < 0.0001, Fig. 4E). These results indicated that autophagy activation alleviates cerebral ischemic injury by inhibiting LOX-1 expression.

Fig. 4.

Early autophagy activation in cerebral ischemia alleviates cerebral I/R injury and inhibits LOX-1 expression in mice. Representative image and infarct volume analysis at 24 hours in mice after cerebral I/R injury (A). Representative image and infarct volume analysis on day 7 in mice after cerebral I/R injury (B). Analysis of neurological function deficiency within 7 days after cerebral I/R injury (C). Western blotting and quantitative analysis of LOX-1 and LC3B-Ⅱ at 24 hours after cerebral I/R injury. n = 3, ****p < 0.0001 vs. sham group; # p < 0.05, ###p < 0.001 vs. MCAO. (one-way ANOVA followed by Tukey’s multiple comparison test) (D). Representative immunofluorescence images and quantitative analysis showing co-localization of Neun-stained neurons (green) and LOX-1-positive cells (red) in the peri-infarct regions from mice at 24 hours after cerebral I/R injury (E). DAPI (blue) indicates cell nuclei. The white arrows indicate the co-labeling cells. Scale bar, 50 µm. n = 4, *p < 0.05, ****p < 0.0001 vs. sham group; ####p < 0.0001 vs. MCAO group. I/R, ischemia/reperfusion; MCAO, middle cerebral artery occlusion; LOX-1, lectin-like oxidized low-density lipoprotein receptor-1

Activating Autophagy Following LOX-1 Overexpression Significantly Reduces Cerebral Ischemic Injury and Inhibits LOX-1 Expression in Mice

To further validate the protective effects of LOX-1 degradation via autophagy activation against cerebral ischemic injury, we performed a rescue experiment in which RAPA was administered after LOX-1 overexpression in a mouse model of acute cerebral ischemia. The Lenti-LOX-1-GFP + RAPA group exhibited a significantly reduced infarct volume compared with that in the Lenti-LOX-1-GFP group at 24 hours (p = 0.016, Fig. 5A) and on day 7 (p = 0.0002, Fig. 5B). Body weight and neurological function were evaluated over 7 days following cerebral I/R injury. Compared with the Lenti-LOX-1-GFP group, the Lenti-LOX-1-GFP + RAPA group showed an increase in body weight (Fig. 5C), although the difference was not statistically significant. For neurological assessment, the mNSS in the Lenti-LOX-1-GFP + RAPA group was significantly lower than that in the Lenti-LOX-1-GFP group (day 1: p = 0.0003; day 2: p < 0.0001; day 3: p = 0.001; day 4: p = 0.0033; day 5: p = 0.0027; day 6: p = 0.007; day 7: p = 0.0017, Fig. 5C). The balance beam score was significantly higher in the Lenti-LOX-1-GFP + RAPA group than in the Lenti-LOX-1-GFP group (day 1: p = 0.0001; day 2: p = 0.0245; day 3: p = 0.0026; day 4: p = 0.0023; day 5: p = 0.0035; day 6: p = 0.0006; day 7: p = 0.0023, Fig. 5C). The Longa score was significantly lower in the Lenti-LOX-1-GFP + RAPA group than in the Lenti-LOX-1-GFP group (day 3: p = 0.034, Fig. 5C). These results indicated that autophagy activation significantly alleviates neurological deficits in mice following cerebral ischemia. CBF was recorded at pre-MCAO, MCAO, reperfusion and 24 hours post-MCAO/R. In the Lenti-LOX-1-GFP group, CBF at 24 hours post-MCAO/R was significantly decreased compared with reperfusion (p = 0.0006, Fig. 5D), whereas no significant difference was observed in the Lenti-LOX-1-GFP + RAPA group (p = 0.868, Fig. 5D). Furthermore, the CBF at 24 hours post-MCAO/R was significantly higher in the Lenti-LOX-1-GFP + RAPA group than in the Lenti-LOX-1-GFP group (p = 0.003, Fig. 5D).

Fig. 5.

Autophagy activation following LOX-1 overexpression alleviates cerebral I/R injury in mice. Representative image and infarct volume analysis at 24 hours after cerebral I/R injury in mice (A). Representative image and infarct volume analysis on day 7 after cerebral I/R injury in mice (B). Analysis of neurological function deficiency within 7 days after cerebral I/R injury (C). Representative image and CBF analysis within 24 hours after cerebral I/R injury in mice (D). I/R, ischemia/reperfusion; LOX-1, lectin-like oxidized low-density lipoprotein receptor-1; CBF, cerebral blood flow

We further examined the effect of autophagy activation on LOX-1 expression in mice overexpressing LOX-1. Western blotting showed that LOX-1 expression was significantly lower in the Lenti-LOX-1-GFP + RAPA group than in the Lenti-LOX-1-GFP group (p = 0.031, Fig. 6A). Immunofluorescence analysis showed significantly reduced co-localization of LOX-1 and NeuN in the Lenti-LOX-1-GFP + RAPA group (p = 0.008, Fig. 6B). These results further supported the theory that autophagy activation suppresses LOX-1 expression, suggesting that autophagy enhancement may represent a potential neuroprotective strategy.

Fig. 6.

Autophagy activation following LOX-1 overexpression reduces LOX-1 expression. Western blot detection and quantitative analysis of LOX-1 at 24 hours after cerebral I/R injury. n = 3, **p < 0.01 vs. sham group; #p < 0.05 vs. Lenti-LOX-1-GFP. (one-way ANOVA followed by Tukey’s multiple comparison test) (A). Representative immunofluorescence images and quantitative analysis showing co-localization of Neun-stained neurons (green) and LOX-1-positive cells (red) in the peri-infarct areas from mice at 24 hours after cerebral I/R injury (B). DAPI (blue) indicates cell nuclei. The white arrows indicate the co-labeling cells. Scale bar, 50 µm. n = 4, **p < 0.01 vs. sham group; ##p < 0.01 vs. Lenti-LOX-1-GFP group. I/R, ischemia/reperfusion; LOX-1, lectin-like oxidized low-density lipoprotein receptor-1

Discussion

This study demonstrated that the serum sLOX-1 levels in patients with AIS were associated with the severity of cerebral infarction. In a cerebral ischemia mouse model, LOX-1 was involved in cerebral ischemia injury, and we found for the first time that LOX-1 was highly expressed in peri-infarct neurons and inhibited neuronal autophagic flux. Autophagy activation was able to alleviate cerebral I/R injury and reduce LOX-1 expression, providing a novel neuroprotective strategy against ischemic stroke.

LOX-1 is a transmembrane glycoprotein receptor, and sLOX-1 is released into the extracellular space through proteolytic cleavage after cellular injury. Therefore, serum sLOX-1 levels are believed to reflect the expression of LOX-1 in cells [21, 22]. Clinical studies have shown that sLOX-1 levels are markedly elevated in patients with AIS compared with those in healthy controls [12, 23] and are positively correlated with the risk of recurrent ischemic stroke, which may improve the risk stratification of recurrent stroke [24, 25].Consistent with these studies, we found significantly higher serum sLOX-1 levels in patients with AIS than in healthy controls. Moreover, sLOX-1 was further elevated in patients with severe AIS compared with that in patients with mild AIS: this supported an association between sLOX-1 levels and stroke severity. Furthermore, we found a strong positive correlation between serum sLOX-1 levels and infarct volume in patients with AIS, which implied that greater tissue damage leads to increased sLOX-1 release. However, no significant correlation was observed between sLOX-1 levels and NIHSS scores. Although the NIHSS score is related to the infarct volume, it correlates more closely with the specific location of the infarction. Furthermore, the NIHSS score does not directly reflect the degree of cellular injury, which may be the primary reason for the lack of significant correlation between sLOX-1 levels and NIHSS scores. Additionally, the relatively small sample size may have been a contributing factor. In this study, we found a significant positive correlation between serum sLOX-1 levels and neutrophil and white blood cell counts, which supported an association between sLOX-1 and systemic inflammatory activity. As LOX-1 acts as an acute-phase reactant with rapidly increased expression under pro-oxidative and pro-inflammatory conditions, these results also suggest that changes in serum sLOX-1 levels reflect the intensity of LOX-1 expression. We divided the patients into with and without reperfusion therapy groups to further investigate whether reperfusion therapy could affect the serum sLOX-1 level and patient outcomes. However, no significant correlations were found between serum sLOX-1 levels and 3-month mRS scores in patients with and those without reperfusion therapy. The results indicated that reperfusion therapy does not affect serum sLOX-1 levels. In summary, the results of this study suggest that serum sLOX-1 levels may be an important clinical indicator for the early evaluation of the severity of ischemic stroke, and that the expression intensity of LOX-1 may directly affect serum sLOX-1 levels. However, the hydrolysis of sLOX-1 involves many complex processes. In addition, Alexander et al. reported that LOX-1 expression in peripheral blood monocytes transiently increased at 6 hours and returned to baseline at 24 hours after acute ischemic stroke, regardless of thrombolytic therapy [26]. This suggests that LOX-1 expression undergoes dynamic changes, and detecting the dynamic expression of LOX-1 may provide a more precise timing for future LOX-1-targeted interventions.

LOX-1 serves as a pivotal receptor for the initiation and progression of atherosclerosis, which ultimately leads to plaque rupture, thrombosis, and ischemic stroke and indirectly causes neuronal damage. In a cerebral ischemia model, LOX-1 induced neuroinflammation and exacerbated neuronal damage by regulating toxic microglia [27]. Additionally, LOX-1 expression is significantly increased within the ischemic core in animal studies [28], and endothelial LOX-1 transgenic mice show significantly aggravated cerebral infarction volume and neurological deficits [26]. LOX-1 also plays a critical role in BBB injury [6]. Taken together, these findings underscore the significant involvement of LOX-1 in cerebral ischemic injury. In this study, we found that LOX-1 overexpression exacerbated the infarct volume and neurological deficits in mice, whereas LOX-1 knockdown alleviated these ischemic deficits. Given that there is a close relationship between CBF and infarct volume, we found that the CBF was decreased after LOX-1 overexpression, and increased after LOX-1 knockdown. For the first time, we found that the expression of LOX-1 was increased in primary neurons after OGD induction and in the neurons of per-infarct areas of mice after MCAO/R, indicating that the mechanisms of LOX-1 in cerebral ischemic injury are related to neuronal damage. In addition, we found that the interaction between LOX-1 and proteins related to the autophagy-lysosome pathway was enhanced in primary neurons treated with the autophagy activator RAPA via co-immunoprecipitation, suggesting that LOX-1 may mediate neuronal damage by affecting neuronal autophagic flux. We further demonstrated that LOX-1 exacerbated cerebral ischemic injury by inhibiting autophagic activity in peri-infarct neurons in a mouse model of cerebral ischemia. However, TEM revealed the presence of autophagosomes in neurons, even under LOX-1 overexpression conditions, indicating that the upregulation of LOX-1 does not completely suppress the autophagic process. Even with the upregulation of LOX-1, there were still a few autophagosomes.

Autophagy is a metabolic process in eukaryotic cells that involves the clearance of intracellular components to protect against damage and regulate stress [29]. Autophagosome formation involves the transport of damaged organelles or abnormal proteins to lysosomes for degradation, thereby maintaining normal cell structure and eliminating harmful factors [30]. Studies have shown that proteins related to neurodegenerative diseases, such as Hsp-90, a-synaptophysin and tau can be degraded through the autophagic-lysosome pathway, thereby engaging a self-defense mechanism to slow down the pathological process [31, 32]. In addition, autophagy activation reduces neuronal death in the penumbra [33], clears damaged mitochondria [34], and participates in the neuroprotective effects of ischemic adaptation [35]. However, autophagy is a double-edged sword because its timing and levels have varying effects on cell survival [36]. Several studies have shown that autophagy activity in the early stages of cerebral ischemia reduces neuronal necrosis, decreases infarct volume, and improves neurological function[37–39]. Consistent with these studies, we found that the immediate activation of autophagy after cerebral I/R in mice significantly reduced cerebral infarction volume and neurological deficits. Given that autophagy is a self-defense mechanism and that autophagosomal pathway-related proteins interact with LOX-1, we investigated the role of autophagy activation on LOX-1 in the early stages of cerebral ischemia. We found that activating autophagy reduced LOX-1 expression, suggesting that early activation of autophagy exerts neuroprotective effects by inhibiting LOX-1 expression. Therefore, we performed a rescue experiment to validate the neuroprotective effects of LOX-1 degradation via autophagy activation and found that the infarct volume decreased and neurological deficits reduced. Furthermore, CBF increased because of the decreased infarct volume. These results were associated with a decrease in LOX-1 expression. In summary, our results suggest that, in the early stages of cerebral ischemia, clearance of LOX-1 via activation of autophagy can alleviate cerebral ischemic injury.

This study has some limitations: First, the peri-infarct region identified by TTC and immunofluorescence staining does not strictly conform to the definitive criteria of the ischemic penumbra. Second, lackness of dynamic evaluation of LOX-1 expression in serum sLOX-1 level and stroke brain tissue. Third, incorporation of autophagic flux inhibitors would help delineate the specific phase of autophagy that is modulated.

Conclusion

In conclusion, this study demonstrated that sLOX-1 levels are associated with AIS severity. LOX-1 participates in acute cerebral ischemic injury by inhibiting neuronal autophagic flux in peri-infarct brain tissue. Autophagy activation in early cerebral ischemia alleviates cerebral I/R injury by degrading LOX-1, therefore providing a novel therapeutic strategy for ischemic stroke.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contributions

Y.L. and Y.Z. secured funding and designed the experiments. Y.H. performed the experiments; R.W., and T.S. analyzed the data; Y.H. and Y.Z. wrote the original draft; and F.Y., Z.T., Y.L. and P.L. reviewed and edited the manuscript.

Funding

This work was supported by Natural Science Foundation in China (82371306, 82571513,82001390, 82171301).

Data Availability

The data and materials support the findings of this study are available from the corresponding authors upon reasonable request.

Declarations

Ethics Approval

This study was conducted in accordance with the principles outlined in the Declaration of Helsinki and was approved by the Ethics Committee of the Xuanwu Hospital of Capital Medical University (Approval number: Clinical Research [2008] No. 1).

All animal experiments were approved by the Institutional Animal Care and Use Committee of Xuanwu Hospital of Capital Medical University (XW-20211216–1) and were conducted in accordance with the principles outlined in the National Institutes of Health's Guide for the Care and Use of Laboratory Animals.

Consent to Participate

Informed consent was obtained from all individual participants included in the study.

Competing Interests

The authors declare no competing interests.