Application of overexpressed PSAP in preparation of medicine for improving secondary brain injury after subarachnoid hemorrhage

Abstract

The invention belongs to the field of medicine, and relates to application of overexpressed PSAP in preparation of medicine for improving secondary brain injury after subarachnoid hemorrhage. HuPSAP is adopted for overexpressing PSAP, the HuPSAP is combined with a Gpr37 receptor on the surface of an oligodendrocyte cell membrane, downstream mTOR and ERK signal channels are activated, myelin sheath related gene expression is enhanced, mitochondrial dependent cell apoptosis is inhibited, and the effect of treating secondary brain injury after subarachnoid hemorrhage is achieved. Therefore, the conduction function of the CL-mPFC loop after SAH is protected, and disturbance of consciousness is reduced. Experimental basis and theoretical basis are provided for treatment of secondary brain injury after SAH, especially disturbance of consciousness, by the HuPSAP.

Description

Technical Field

[0001] The invention belongs to the field of medicine and relates to the application of overexpressed PSAP in the preparation of a medicine for improving secondary brain damage after subarachnoid hemorrhage. Background Art

[0002] Subarachnoid hemorrhage (SAH) is a cerebrovascular emergency with a high mortality and morbidity rate, with a 30-day mortality rate of 40%-50%. Approximately 60% of SAH patients experience impaired consciousness during the acute phase of the disease, and the severity of impaired consciousness is significantly correlated with a poor prognosis. Current SAH treatments are primarily aimed at preventing rebleeding and cerebral vasospasm, but even with successful control of these complications, approximately 40% of patients still develop a poor prognosis, suggesting the existence of an underlying pathological mechanism that has not been effectively addressed.

[0003] Recent studies have shown that structural disconnection of white matter fiber tracts contributes significantly more to brain dysfunction than local gray matter lesions. In the pathological setting of SAH, blood metabolites diffuse through the cerebrospinal fluid circulation, attacking oligodendrocytes in the white matter, triggering axonal swelling and myelin degradation. Imaging studies have confirmed a strong positive correlation between the integrity of the thalamocortical white matter pathway and the level of consciousness (r = 0.73), suggesting that white matter damage disrupts the connectivity of long-range projection fibers, leading to global network functional decoupling, a key mechanism of impaired consciousness.

[0004] The neural circuit formed by the central lateral thalamic nucleus (CL) and the medial prefrontal cortex (mPFC) plays a key role in maintaining wakefulness. CL connects to the mPFC via glutamatergic projections, forming a closed "thalamus-cortex-thalamus" loop that regulates the synchronized discharge of cortical neurons. Clinical studies have found that the degree of consciousness recovery in patients with traumatic brain injury is positively correlated with the strength of functional connectivity in the CL-mPFC loop, and deep brain stimulation targeting the CL nucleus can significantly increase the duration of wakefulness in patients in a vegetative state. However, the mechanism of CL-mPFC circuit damage after SAH has not been fully elucidated.

[0005] Oligodendrocytes are the sole executors of myelination and maintenance in the central nervous system, and their functional status directly determines the efficiency of signal transmission within the white matter network. Mitochondrial dysfunction and endoplasmic reticulum stress are seen in oligodendrocytes early after SAH. Red blood cell lysis products activate inflammatory signaling pathways, triggering an oxidative burst in oligodendrocytes (ROS increase by 3.2-fold) and a collapse in energy metabolism (ATP decrease by 60%).

[0006] Prosaposin (Psap) is a multifunctional secretory protein that, as a precursor of the Saposin family of proteins, is directly involved in sphingolipid metabolism and provides essential substrates for the lipid components of the myelin membrane. Psap binds to the G protein-coupled receptor Gpr37 on the surface of oligodendrocytes, activating the downstream mTOR and ERK signaling pathways, regulating the expression of myelin-related genes and inhibiting cell apoptosis. Studies have shown that Psap plays a key role in maintaining myelin integrity and protecting oligodendrocytes. Reduced expression of Psap is associated with white matter damage, while increased levels can alleviate myelin and axonal damage. In spinal cord injury models, activation of the Psap / Gpr37 signaling axis is positively correlated with myelin regeneration efficiency, suggesting its universal applicability in neurological repair.

[0007] Current treatments for white matter damage after SAH primarily include antioxidant intervention, inflammatory regulation, and growth factor supplementation. However, antioxidants have poor targeting and are difficult to accumulate in oligodendrocytes; anti-inflammatory drugs may inhibit beneficial reparative inflammation; and growth factors are unstable and have difficulty penetrating the blood-brain barrier. These approaches often target a single pathological link and are unable to comprehensively regulate the complex cascade of SAH-induced white matter damage.

[0008] Through in-depth analysis of single-cell nuclear RNA sequencing data, we found that SAH may lead to myelin destruction and impaired CL-mPFC circuit function by inhibiting Psap expression in oligodendrocytes and its interaction with Gpr37, thereby causing impaired consciousness. Experiments confirmed that Psap transcription levels in oligodendrocytes were significantly downregulated after SAH, and the protein interaction between Psap and Gpr37 was weakened. Concomitantly, myelin basic protein expression decreased in white matter regions such as the corpus callosum, and T2 magnetic resonance imaging signal increased, indicating extensive white matter damage. These findings provide an important theoretical basis for the development of therapeutic drugs targeting the Psap-Gpr37 signaling axis, which may improve white matter damage and impaired consciousness after SAH by restoring Psap function.

[0009] Therefore, there is an urgent need to develop new therapeutic drugs that can target and regulate the Psap-Gpr37 signaling axis and simultaneously affect multiple key links of white matter damage, so as to provide new strategies and directions for the treatment of secondary brain injury after SAH, especially impaired consciousness. Summary of the Invention

[0010] This study aimed to explore the mechanism of action of recombinant human prosaposin (HuPSAP) in the treatment of SAH and its secondary injuries, and to evaluate the efficacy of HuPSAP for secondary brain injury (SBI) after SAH. Experiments have shown that HuPSAP restores the function of the Psap-Gpr37 signaling axis, protects oligodendrocytes, and maintains myelin integrity, thereby alleviating functional damage to the CL-mPFC neural circuit and impaired consciousness after SAH.

[0011] The technical solution of the present invention is:

[0012] Application of overexpression of PSAP in improving secondary brain injury after subarachnoid hemorrhage.

[0013] Furthermore, the application includes overexpressing PSAP using HuPSAP.

[0014] Furthermore, the HuPSAP is used to reduce myelin damage after subarachnoid hemorrhage, alleviate axonal pathological damage, and improve subarachnoid hemorrhage-induced consciousness disorders and neurological function prognosis.

[0015] Furthermore, the secondary brain injury after subarachnoid hemorrhage includes one or more of the following:

[0016] disrupts consciousness and damages the central-lateral thalamus-medial prefrontal cortex neural circuit;

[0017] Disruption of central-lateral thalamic-medial prefrontal cortex circuit connectivity;

[0018] Myelin damage and axonal injury in the central-lateral thalamo-medial prefrontal cortex circuit are accompanied by oligodendrocyte dysfunction;

[0019] Disruption of Psap-Gpr37 interaction in oligodendrocytes.

[0020] Furthermore, the concentration of HuPSAP is 0.25-1 ng / μl.

[0021] Furthermore, the HuPSAP is an injection.

[0022] Furthermore, the HuPSAP is used to protect the conduction function of the CL-mPFC circuit after subarachnoid hemorrhage and alleviate impaired consciousness.

[0023] Furthermore, the HuPSAP is used to alleviate cognitive impairment after subarachnoid hemorrhage.

[0024] Furthermore, the HuPSAP is used to alleviate subarachnoid hemorrhage-induced white matter damage in the CL-mPFC pathway.

[0025] Furthermore, the HuPSAP is used to rescue axonal transport defects by alleviating myelin breakdown and oligodendrocyte dysfunction.

[0026] Beneficial effects

[0027] HuPSAP binds to the Gpr37 receptor on the surface of oligodendrocytes, activating the downstream mTOR and ERK signaling pathways, enhancing the expression of myelin-related genes and inhibiting mitochondria-dependent apoptosis. This, in turn, protects the conduction function of the CL-mPFC circuit after SAH and alleviates impaired consciousness. This provides an experimental and theoretical basis for HuPSAP treatment of secondary brain injury after SAH, particularly impaired consciousness. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 Subarachnoid hemorrhage disrupts consciousness and impairs central-lateral thalamic-medial prefrontal cortex neural circuitry in mice. (A) Schematic diagram of the EEG / EMG electrode implantation and recording apparatus. Representative EEG / EMG recordings and power spectra of wakefulness, non-rapid eye movement (NREM), and rapid eye movement (REM) sleep states. (B) Representative EEG spectrograms, EEG / EMG recordings, and sleep state distribution in sham and SAH mice over a 24-hour period. (C) Quantification of time spent in wakefulness, NREM, and REM states (n = 6 per group, ****P < 0.0001, *P < 0.05, two-way ANOVA). (D) Schematic diagram of HSV-tdTomato virus injection into the CL for anterograde tracing. (E) Representative fluorescence images of tdTomato-labeled projections from the CL to the mPFC in sham and SAH mice. (F) Schematic diagram of retrograde viral tracing using AAV2-retro-hSyn-EGFP injected into the mPFC. (G) Representative fluorescence images of EGFP-labeled CL neurons in sham and SAH mice. (H) Schematic diagram of optogenetic modulation and fiber photometry recording using AAV-Syn-GCaMP6s (mPFC) and AAV-CaMKIIa-ChrimsonR (CL). (I) Representative fluorescence images of GCaMP6s and ChrimsonR expression in sham and SAH mice. (J, K) GCaMP6s fluorescence tracks and corresponding heat maps showing optogenetic stimulation-induced calcium responses in sham and SAH groups.

[0029] (L) Quantification of the normalized area under the curve (AUC) of optogenetic responses in sham and SAH mice (****P < 0.0001, unpaired t-test). Scale bar, 200 μm.

[0030] Figure 2 Subarachnoid hemorrhage disrupts central-lateral thalamus-medial prefrontal cortex circuit connectivity. (A) Schematic diagram of sample collection.

[0031] (B) Uniform Manifold Approximation and Projection (UMAP) dimensionality reduction visualization of cell type clusters identified by single nucleus RNA sequencing (snRNA-seq), including excitatory neurons (EXN), inhibitory neurons (IN), oligodendrocytes (OL), oligodendrocyte precursor cells (OPC), astrocytes (Astro), vascular and leptomeningeal cells (VLMC), and endothelial cells (EC); (C) UMAP feature maps of excitatory neuron clusters (clusters 1-6), showing transcriptome heterogeneity. (D) Dot plot showing the percentage of cells expressing key marker genes in each excitatory neuron subtype (EXN1-EXN6), with expression intensity indicated by color scale. (E) Allen Brain Atlas image showing that Lef1 expression is primarily localized in the thalamus. (F) Volcano plot of differentially expressed genes in EXN1. (G) GO functional enrichment analysis of differentially expressed genes upregulated in EXN1. (H) Reactome pathway enrichment analysis of differentially expressed genes upregulated in EXN1. (I) Schematic diagram of the fiber photometry recording experimental setup, with AAV-CaMKIIa-GCaMP6s injected into the CL. (J) Representative fluorescence images of GCaMP6s expression in the CL. (K, L) Example calcium fluorescence traces from sham and SAH mice, showing increased spontaneous calcium transients in the SAH group. (M) Quantification of calcium transient duration, showing a significant increase in SAH mice (*P < 0.05, unpaired t-test). (N, R) Schematic diagram of patch clamp recording sites in the CL and mPFC, respectively. (O,S) Representative action potential traces of sham and SAH neurons in the CL and mPFC. (P,T) Quantification of action potential frequency in the CL (**p < 0.01, unpaired t-test) and mPFC (ns, not significant). (Q,U) Resting membrane potential (RMP) measurements revealed significant depolarization in CL neurons after SAH, but no significant changes in the mPFC.

[0032] Figure 3Subarachnoid hemorrhage-induced myelin damage and axonal injury are associated with oligodendrocyte dysfunction. (A) Representative T2-weighted magnetic resonance imaging (MRI) images showing areas of T2 hyperintensity in the corpus callosum (CC), external capsule, fornix, and internal capsule / cerebral peduncles / optic tract (IC / CP / OT) of SAH mice. (B) Quantification of T2 hyperintensity in regions of interest (****P < 0.0001, unpaired t-test). (C) Representative images showing AAV9 transfection, with mCherry (red) as a viral marker and DAPI (blue) labeling cell nuclei. Scale bar, 200 μm. (D) Immunofluorescence staining of myelin markers (AAV2 / 9, dMBP, MBP) in the CC of sham and SAH mice 48 hours after SAH. Scale bar, 50 μm. (E) Quantification of myelin markers in the CC, showing increased APP and dMBP and decreased MBP in SAH mice (*P < 0.05, **P < 0.01, unpaired t-test). (F) UMAP visualization of oligodendrocytes, showing clear separation between Sham and SAH groups. (G) Volcano plot showing differential expression of genes in oligodendrocytes after SAH. (H) Pathway enrichment analysis of downregulated genes in oligodendrocytes, showing significant enrichment in lipid metabolism. (I) GO analysis of five gene clusters from pseudo-chronic analysis, identifying enriched pathways related to myelination, lipid transport, and ion homeostasis. (J) Downregulation of key myelination-related genes in pseudo-chronic analysis, highlighting the significant downregulation of Psap, Abca2, Mal, and Mbp after SAH.

[0033] Figure 4Downregulation of Psap-Gpr37 interaction in oligodendrocytes following subarachnoid hemorrhage. (A) Bubble plot showing changes in Psap-Gpr37 interaction after SAH. (B) Colocalization analysis of Psap (green) and Gpr37 (red) immunofluorescence staining under a confocal microscope (Pearson correlation coefficient = 0.84); Scale bar, 20 μm. (C, D) Representative images of in situ PLA analysis showing Psap and Gpr37 interaction in the corpus callosum of brain tissue sections from sham and SAH mice (green represents the interaction complex, blue represents the nucleus); Scale bar, 20 μm. (E, F) ELISA bar graphs showing Psap levels in human and mouse plasma (SAH / Sham, n = 6, **P < 0.01, ****P < 0.0001). (G) UMAP and violin plots showing Psap gene expression in sham and SAH groups, **P < 0.01. (H) UMAP and violin plots showing Gpr37 gene expression in sham and SAH groups, ns = no significant difference. (I) Representative immunofluorescence images of Psap (green) and CC1 (red) co-stained in the corpus callosum of sham and SAH mice. DAPI (blue) labels the nuclei, and the dotted outline indicates the corpus callosum. Scale bar, 100 μm. (J) Quantification of Psap expression in CC1-positive cells in the corpus callosum of sham and SAH mice (n = 6, **P < 0.01). (K) Representative immunofluorescence images of Gpr37 (green) and CC1 (red) co-stained in the corpus callosum of sham and SAH mice. DAPI (blue) labels the nuclei, and the dotted outline indicates the corpus callosum. Scale bar, 100 μm. (L) Quantification of Gpr37 expression in CC1-positive cells in the corpus callosum of sham and SAH mice (n = 6, ns = no significant difference). (M) Representative immunofluorescence images of Psap (green) and CC1 (red) co-stained in brain tissue sections from SAH patients and non-SAH controls. DAPI (blue) labels cell nuclei. Scale bar, 20 μm. (N) Quantification of Psap expression in CC1-positive cells in brain tissue sections from human non-SAH controls and SAH patients (n = 6, ****P < 0.0001). (O) Representative immunofluorescence images of Gpr37 (green) and CC1 (red) co-stained in brain tissue sections from human controls and SAH patients. DAPI (blue) labels cell nuclei. Scale bar, 20 μm. (P) Quantification of Gpr37 expression in CC1-positive cells in brain tissue sections from SAH patients and non-SAH controls (n = 6, ns = no significant difference).

[0034] Figure 5Effects of Psap overexpression on the integrity of the corpus callosum white matter after subarachnoid hemorrhage. (A) Schematic diagram of stereotaxic injection of Psap-overexpressing virus into the corpus callosum. (B) Representative images showing Psap-overexpressing virus transfection, with mCherry (red) used as a viral marker and DAPI (blue) used to label cell nuclei. Scale bar, 200 μm. (C) Immunofluorescence co-staining of Psap (green) and Gpr37 (red) in the corpus callosum of brain tissue sections from Psap-overexpressing (Psap-OE) and control (Psap-Vec) groups. In situ PLA analysis reveals the interaction between Psap and Gpr37 (green indicates the interaction complex; blue indicates the nucleus). Scale bar, 20 μm.

[0035] (D) Quantification of PLA experiments, showing the number of positive cells per high-power field (n = 6, ****P < 0.0001). (E) Representative immunofluorescence images of Psap (green) and the oligodendrocyte marker CC1 (red) in the corpus callosum of the Psap OE and Psap-Vec groups. The dotted outline indicates the corpus callosum. Scale bar, 20 μm. (F) Representative T2-weighted MRI images of the corpus callosum of the Psap OE and Psap-Vec groups after SAH. (G) Quantification of normalized T2 signal intensity in the corpus callosum of the Psap OE and Psap-Vec groups after SAH (n = 6, ****P < 0.0001). (H) Immunofluorescence staining of myelin markers (AAV2 / 9, dMBP, MBP) in the CC of the Psap OE and Psap-Vec groups. Scale bar, 50 μm. (I) Quantification of the relative fluorescence intensity of MBP and dMBP in the corpus callosum (OE / Vec, n = 6, **P < 0.01). (J) Representative electron microscopy images of myelin sheaths in the corpus callosum from different groups. Scale bar, 5 μm. (K) G-ratio quantification of the myelin ultrastructure shown in (J).

[0036] Figure 6 Psap overexpression alleviates circuit dysfunction and improves consciousness and cognitive impairment after subarachnoid hemorrhage. (A) Schematic diagram of the optogenetic experimental setup for Psap overexpression and circuit conduction assessment. (B) Representative images showing successful transfection of the light-sensitive protein ChrimsonR and the calcium indicator GCaMP6s into the corpus callosum and medial prefrontal cortex, respectively. Scale bar, 100 μm. (C) Average histograms of post-stimulation ΔF / F calcium signals for Psap-overexpressing (Psap-OE) and control (Psap-Vec) groups. The black line indicates the mean, and the shaded area indicates the standard error. (D) Heatmaps of ΔF / F calcium signals for Psap OE and Psap-Vec groups. The color bars indicate individual mice and groups.

[0037] (E) Quantification of the area under the curve (AUC) of normalized calcium signals in the Psap OE and Psap-Vec groups (n = 6, **P < 0.01). (F) Representative electroencephalogram (EEG) and electromyogram (EMG) recordings, as well as 24-hour sleep-wake stage analysis, recorded 48 hours after surgery in the control and Psap OE groups. (G) Proportion of time spent in each sleep-wake stage in the Psap OE and Psap-Vec groups (n = 6, *P < 0.05, ****P < 0.0001). (H, K) Representative images of the Psap OE and Psap-Vec groups in the Morris water maze and probe tests. (I, J, L) Quantitative analysis of escape latency, swimming speed, and time spent in the target quadrant in the Psap OE and Psap-Vec groups in the Morris water maze and probe tests (n = 12, ****P < 0.0001, ns = not significant difference, *P > 0.05).

[0038] Figure 7 Human prosaposin activator protein (HuPSAP) alleviates circuit dysfunction and improves consciousness and cognitive deficits after subarachnoid hemorrhage. (A) Schematic diagram of stereotaxic injection of viruses and HuPSAP for optogenetic experiments. (B) Representative images showing transfection of the light-sensitive protein ChrimsonR and the calcium indicator GCaMP6s. The white dashed line marks the boundary. Scale bar, 100 μm. (C) Average histogram of post-stimulation ΔF / F calcium signals for different HuPSAP concentrations. The thick black line indicates the mean, and the shaded area indicates the standard error.

[0039] (D) Heatmap of ΔF / F calcium signals for different HuPSAP concentration groups. (E) Quantification of the area under the curve (AUC) of normalized calcium signals for different HuPSAP concentration groups (n = 6, **P < 0.01, ***P < 0.001). (F) Representative EEG and EMG recordings and sleep-wake stages over 24 hours for HuPSAP and saline control groups. (G) Proportion of time spent in each sleep-wake stage for HuPSAP and saline control groups (n = 6, *P < 0.05, **P < 0.01). (H, K) Representative images of HuPSAP and saline control groups in the Morris water maze and probe tests. (I, J, L) Quantitative analysis of escape latency, swimming speed, and time spent in the target quadrant in the Morris water maze and probe tests (n = 12, **P < 0.01, ***P < 0.001, ns = not significant).

[0040] Figure 8Overexpression of HuPSAP and Psap attenuates subarachnoid hemorrhage-induced white matter damage in the central-lateral thalamus-medial prefrontal cortex pathway. (A) Immunofluorescence staining of myelin markers (AAV2 / 9, dMBP, and Psap) in the corpus callosum (CC) of the sham, SAH, SAH-saline, and SAH-HuPsap groups. Scale bar: 50 μm. (B) Quantification of relative dMBP fluorescence intensity in the corpus callosum (n = 6, **P < 0.01). (C) Schematic diagram of the mechanism of action. DETAILED DESCRIPTION

[0041] 1. Methods

[0042] 1.1 Establishment of mouse subarachnoid hemorrhage model

[0043] (1) Preoperative preparation: 8-week-old mice were used, and the weight of the experimental mice was within the range of 22 g ± 2 g. Microscope, microsurgery instruments, 5-0 sutures, alcohol cotton balls, erythromycin eye ointment, normal saline, nylon sutures with microscissors at a 45-degree angle to create the tip, and sterilized modeling instruments were prepared.

[0044] (2) Mice were placed in an anesthesia induction machine for isoflurane-induced anesthesia (3% concentration). After the mice were fully anesthetized, they were fixed in a supine position on a stereotaxic apparatus (Reward Life Science Co., Ltd., China) while maintaining isoflurane anesthesia (1.5% concentration). A constant temperature blanket was used under the stereotaxic apparatus to maintain the body temperature of the mice at 37°C.

[0045] (3) Remove the hair on the mouse's neck and thoroughly disinfect the skin with an alcohol cotton ball, and apply erythromycin eye ointment to the eyes.

[0046] (4) Perform a midline neck incision and use blunt dissection techniques to separate the digastric aponeurosis to fully display the common carotid artery and its accompanying vagus nerve bundle. Perform a fine separation operation on the artery while strictly avoiding mechanical stimulation of the vagus nerve. In parallel, free fixation of the common carotid artery is performed using two lines.

[0047] (5) Follow the long axis of the artery to the bifurcation anatomical structure, free the internal carotid artery and external carotid artery segments respectively, perform double ligation of the external carotid artery 1 cm distal to the bifurcation and leave a traction line, and then complete the arterial transection operation distal to the ligation point.

[0048] (6) Perform electrocoagulation or ligation on the occipital artery branches, adjust the angle between the external carotid and internal carotid arteries to 180° through the pre-set suspension wire system, and use a micro-arterial clamp to temporarily block the blood flow of the common carotid artery. Use microscissors to open a window in the wall of the external carotid artery and introduce the embolization catheter. When the embolization wire is advanced to the bifurcation area, tighten the fixed wire immediately to prevent displacement. Moderately loosen the internal carotid artery traction device and continue to advance the catheter. Stop when a typical breakthrough feeling occurs and the 2cm marking line reaches the arterial bifurcation plane. Maintain for 5 seconds and then remove the catheter to restore blood perfusion to the common carotid artery. After confirming that the vascular resuscitation is good, suture the incision layer by layer to complete the postoperative monitoring process.

[0049] 1.2 Mouse EEG-EMG recording and analysis

[0050] (1) Preoperative preparation: 8-week-old mice weighing 22 g ± 2 g. Microscope, microsurgery instruments, 5-0 sutures, alcohol cotton balls, erythromycin eye ointment, normal saline, EEG electrodes, and dental cement.

[0051] (2) Mice were placed in an anesthesia induction machine for isoflurane-induced anesthesia (3% concentration). After the mice were fully anesthetized, they were fixed in a prone position on a stereotaxic apparatus (Reward Life Science Co., Ltd., China) while maintaining isoflurane anesthesia (1.5% concentration). A constant temperature blanket was used under the stereotaxic apparatus to maintain the body temperature of the mice at 37°C.

[0052] (3) Shave the mouse's head and thoroughly disinfect the skin with an alcohol cotton ball, and apply erythromycin ointment to the eyes.

[0053] (4) Cut a fingernail-sized incision on the mouse scalp to expose the mouse's anterior and posterior fontanelles. Clean the remaining periosteum with saline, and then wipe the skull surface dry with a clean cotton ball.

[0054] (5) To monitor EEG activity, two small craniotomies were drilled in the frontal and occipital regions, and then four screws were carefully inserted into the cranial holes so that the screw tips were in contact with the meninges.

[0055] (6) Place the EEG electrode in the middle of the four screws, tightly wrap the EEG electrode silver wire with the two screws on the frontal bone, tightly connect the ground wire silver wire with the two screws on the occipital bone, and insert the EMG silver wire into the mouse neck muscle.

[0056] (7) Use dental cement to fix the electrodes, silver wires, and screws to the mouse skull.

[0057] (8) After the operation, the mice were housed in individual cages and allowed to recover and become familiar with the EEG recording environment for one week before EEG recording (Shanghai Quanlan) was started.

[0058] (9) EEG recordings were started at least one week after electrode implantation. Each mouse was recorded for 24 hours, with ZT 0-12 being daytime and ZT 13-24 being nighttime. The sleep analysis software LunionData was then used to automatically analyze the multi-channel recording signals based on the spectral characteristics of the EEG-EMG waveforms and classify them into continuous non-overlapping 4-s periods. The criteria for determining the wakefulness period were the presence of desynchronized, low-amplitude, high-frequency EEG rhythms accompanied by phasic myoelectric activity enhancement; NREM was characterized by synchronized high-amplitude, low-frequency (0.5-4 Hz delta waves) EEG signals and baseline myoelectric activity; REM showed significant theta oscillations (4-10 Hz) accompanied by suppression of myoelectric activity. The scoring results were visually inspected and manually corrected if necessary. The sleep-wake state classification was performed by researchers who were unaware of the experimental procedures.

[0059] (10) In the EEG analysis results, data containing interference artifacts were excluded. In the EEG data recorded from mice with subarachnoid hemorrhage, data in which the brain tissue was removed after recording to confirm that there was no hemorrhage were excluded.

[0060] 1.3 Morris water maze test

[0061] The Morris water maze system consists of a circular polypropylene navigation pool, visual interference barriers, a concealed escape device, and an intelligent trajectory tracking system (Noldus EthoVision XT 8.0). The experimental setup's geometric parameters are: 120 cm diameter × 80 cm height, with the liquid level maintained at 60 cm. A transparent acrylic platform (R = 5 cm) is submerged 1 cm deep and 20 cm from the edge. A white polyethylene particle suspension is dispersed in the water to enhance visual masking and optimize the subject's spatial recognition ability. The maze is arranged in quadrants (NE / SE / SW / NW), with four sets of differentiated geometric symbols hanging on the pool walls as spatial reference cues. A thermally stabilized water temperature of 22 ± 1°C is maintained throughout the experiment, with the target platform fixed in the NE quadrant. The standardized experimental process consists of three progressive phases: an adaptation training phase, a spatial orientation phase, and a target exploration phase.

[0062] During the adaptation phase, mice are placed in the maze and swim for 60 seconds, and the platform is removed to allow them to adapt to the environment; mice that float for a long time or move abnormally will be eliminated. Positioning and navigation phase: The platform is fixed in the NE area, and mice are placed in the maze in the order of SE, SW, and NW, with training in each area 4 times a day. If the mouse can find the platform within 60 seconds and stay for at least 2 seconds, it is considered a success; if it fails to find it, it is manually assisted to climb onto the platform and stay for 20 seconds. The latency refers to the time from entering the water to successfully finding the platform. During the training, the software is used to record the average latency and swimming speed of the mouse in each area. Spatial exploration phase: The platform is removed, and the mouse is placed in the SW area. The software records the mouse's swimming time, latency, number of entries into the target area, and number of times crossing the platform in the target area within 60 seconds.

[0063] 1.4 Virus injection

[0064] Mice (8 weeks old, 22 g ± 2 g) were placed in an anesthesia induction machine for isoflurane-induced anesthesia (3% concentration). After the mice were fully anesthetized, they were fixed in a prone position on a stereotaxic instrument (Reward Life Science Technology Co., Ltd., China) while maintaining isoflurane anesthesia (1.5% concentration). A constant temperature blanket was used under the instrument to maintain the body temperature of the mice at 37°C. The hair on the head was shaved and disinfected with an alcohol cotton ball, and erythromycin ointment was wiped on both eyes. An incision of about 1 cm was made on the head to expose the anterior and posterior fontanelles. The skull was wiped with normal saline, the connective tissue and periosteum on the surface of the skull were removed, and the skull surface was wiped dry with a clean cotton ball. The anterior and posterior fontanelles were adjusted to the same horizontal position and the left and right sides were adjusted to be parallel. The needle of a microsyringe (10 μL, 0.485 mm, Hamilton, USA) was placed at the position of the anterior fontanelle and the coordinates were zeroed. Under a microscope, a small handheld dental drill (Beijing Laiate Technology Development Co., Ltd., China) was used to drill a small hole approximately 0.6 mm in diameter at the virus injection port for virus injection. A virus microinjection pump (Reward Life Science Co., Ltd., China) was used to inject the virus from the hole into the target nucleus. After 2 minutes of incubation, the virus was injected (injection rate: 0.06 μL / min).HSV-tdTomato (50 nL, H01002, Shumi, China) was injected into the CL (coordinates: Bregma, AP = -1.94 mm, ML = 0.85 mm, DV = -2.75 mm, perpendicular to the midline); AAV-hSyn-EGFP (2 / Re) (50 nL, PT-1990, Shumi, China) was injected into the mPFC (coordinates: Bregma, AP = 2.00 mm, ML = 0.30 mm, DV = -2.75 mm, perpendicular to the midline). V = -2.50 mm, perpendicular to the midline); AAV-CaMKIIa-ChrimsonR (50 nL, PT-4285, Shumi, China) was injected into the CL (coordinates: Bregma, AP = -1.94 mm, ML = 0.85 mm, DV = -2.75 mm, perpendicular to the midline); AAV-hSyn-GCaMP6s (50 nL, PT-0145, Shumi, China) was injected into the mPFC (coordinates: Bregma a, AP = 2.00 mm, ML = 0.30 mm, DV = -2.50 mm, perpendicular to the midline) and CL (coordinates: Bregma, AP = -1.94 mm, ML = 0.85 mm, DV = -2.75 mm, perpendicular to the midline); AAV-MBP-Psap-mCherry (200 nL, PT-1515, Shumi, China) was injected into the bilateral corpus callosum (Bregma, AP = -1.06 mm, ML = ± AAV2 / 9-hsyn-mCherry (50 nL, PT-1008, Shumi, China) was injected into the CL (coordinates: Bregma, AP = -1.94 mm, ML = 0.85 mm, DV = -2.75 mm, perpendicular to the midline). After injection, the needle was allowed to rest for 10 minutes before slowly withdrawing (withdrawal speed 1 mm / min). The skin was closed with surgical sutures after surgery. HSV transduction required 3 days, while other viruses required 21 days. Mice were allowed to recover for at least 3 weeks before electrophysiological or behavioral experiments. After the experiment, histological analysis was performed to verify the location of viral transduction. Data were excluded if viral transduction extended beyond the CL / mPFC / corpus callosum region.

[0065] 1.5 Fiber implantation

[0066] Mice (8 weeks old, 22 g ± 2 g) were placed in an anesthesia induction machine for isoflurane-induced anesthesia (3%). After full anesthesia, they were fixed in a prone position on a stereotaxic apparatus (Reward Life Science Co., Ltd., China) while maintaining isoflurane anesthesia (1.5%). A constant temperature blanket was used beneath the apparatus to maintain the mouse's body temperature at 37°C. The head was shaved and disinfected with an alcohol swab. Both eyes were rubbed with erythromycin ointment. A fingernail-sized section of scalp was removed to expose the anterior and posterior bregma. The skull was wiped with saline, and the connective tissue and periosteum on the skull surface were removed. The skull surface was then wiped dry with a clean cotton ball. Virus injection was performed according to the protocol described in 1.4. After the microinjection needle was removed, a ceramic fiber ferrule (1.25 mm, 2.5 mm, Qian Aoxingke, China) was inserted perpendicularly through the original skull opening through a fiber ferrule holder (Qian Aoxingke, China) at a depth 0.2 mm shallower than the virus injection depth. After insertion, fix the optical fiber with dental cement and pull out the optical fiber holder after the dental cement is completely solidified.

[0067] 1.6 Optogenetics Experiment Recording and Analysis

[0068] The virus-injected fiber optic ferrule was implanted, and after the mice recovered for 2 weeks, the ferrule of the mice was connected to the fiber optic recording system. The mice were acclimated for one week (3 weeks after viral transfection) and recorded. 633nm red light was used to activate the photosensitive protein ChrimsonR in the upstream CL nucleus (10ms, 20Hz, 5mW), with each illumination of 5s, an interval of 25s, and 3 cycles per mouse. At the same time, the calcium signal after GCaMP6s transfection was recorded in the downstream mPFC nucleus (470nm, 20mW). If the mouse showed irritability and alertness during the recording process, the mouse was allowed to adapt to the environment for one hour and the baseline changes were recorded. The photometric data were exported as a MATLAB file for further analysis. Three experiments were performed on each mouse, and the fluorescence change value (ΔF / F) was obtained by calculating (F-F0) / F0. ΔF / F. The values are represented by heat maps or average maps. F0 is the baseline average fluorescence signal in the 1-second period before the start of light stimulation. The normalized area under the curve is calculated as the sum of the fluorescence changes during the light stimulation period divided by the duration of the behavioral period. ΔF / F0 is presented as a heat map or as an average map, and the shaded area represents the standard error of the mean (SEM).

[0069] 1.7 Patch clamp experiment

[0070] (1) Mice (8 weeks old, 22 g ± 2 g) were deeply anesthetized with isoflurane and then perfused with ice-cold oxygenated slicing solution.

[0071] (2) The whole brain was rapidly dissected and placed in pre-cooled oxygenated slicing solution.

[0072] (3) Slices containing the target brain region were coronally sectioned at 300 μm using a vibratome (VTA-1200S; Leica) and incubated in slicing solution at 33°C for 15 min, then transferred to incubation solution.

[0073] (4) After incubation at room temperature (RT, 24-26°C) for at least 1 hour, the cells were transferred to a fluorescence microscope (BX51Wl;

[0074] Recordings were performed using an Olympus CT scanner. All neuronal recordings were performed in artificial cerebrospinal fluid (ACSF). The circulatory system was maintained at ~33°C using a temperature control system. To maintain cell viability, all solutions were continuously perfused with 95% oxygen and 5% carbon dioxide.

[0075] (5) After determining the location of the target brain region under the 10x objective lens of the microscope, switch the objective lens to 40x and perform whole-cell patch clamp recording.

[0076] (7) The borosilicate glass electrodes used for recording (BF150-86-10; Sutter Instrument) were pulled using a P97 puller (Sutter Instrument) to ensure an impedance between 3 and 5 MΩ. The electrodes were then perfused with internal solution and used for cell recording.

[0077] (8) Neuronal electrophysiological recordings were performed using a Multiclamp 700B amplifier (Molecular Devices) and a Digi-data 1440A digital-to-analog converter (Molecular Devices). Signals were low-pass filtered at 10 kHz and digitized at 10 kHz (MICRO3

[0078] 1401, Cambridge Electronic Design). Data were acquired and analyzed using Clampfit 10.6 software (Cambridge Electronic Design). We also strictly controlled the quality of the recorded cells, such as series resistance Ra < 25 MΩ, input impedance > 300 MΩ, and leakage current < 50 pA. Cells outside these values were excluded.

[0079] 1.8 MRI scan of mice

[0080] MRI scans were performed using a 7.0T animal scanner (Bruker Biospin) after anesthesia using a gas mixture (induction: 3% isoflurane and 1 L / min O2; maintenance: 1.5% isoflurane and 1 L / min O2). T2 weighted images were acquired using RARE. White matter signal values were obtained from the original DICOM images using a 3D slicer.

[0081] 1.9 Immunofluorescence staining

[0082] After induction of intraperitoneal anesthesia with 1% sodium pentobarbital, animals underwent transcardial perfusion immediately: first, a 20 mL cardiac flush with pre-chilled PBS (0.01 M, pH 7.4) was performed, followed by perfusion with 20 mL of 4% paraformaldehyde (PFA) fixative for tissue fixation. Following perfusion, the entire brain was removed, the pia mater and arachnoid membranes meticulously dissected, and the intact brain specimen immersed in 50 mL of 4% PFA fixative for 24 hours. After fixation, the specimen was transferred to a 30% sucrose gradient dehydration system (refreshed daily until tissue sedimentation reached equilibrium). After dehydration, the brain tissue was embedded in OCT embedding medium and fixed in a cryostat (Leica Biosystems, USA), frozen at -80°C for 12 hours, and then transferred to a cryostat sectioning system (Leica Biosystems, USA) for temperature equilibration (2 hours at -20°C). Serial coronal sections were prepared at 20 μm thickness, and target nuclei were located according to the Paxinos mouse brain stereotaxic atlas. Target sections were rinsed with PBS (3 × 5 min) and then subjected to antigen heat retrieval (15 min, using rapid retrieval solution). These sections were then permeabilized (15 min, using PBS containing 0.3% Triton X-100) and blocked with rapid blocking solution (1 h, room temperature). The primary antibody reaction system was incubated overnight at 4°C in a humidified chamber. After rinsing with PBST (3 × 10 min), secondary antibody labeling was performed in the dark (37°C, 1 h). Following nuclear counterstaining with DAPI, the sections were mounted with glycerol. Microscopic images were acquired using a confocal microscope within 30 min of mounting. (Mouse anti-MBP, 1:200, abcam, ab11159; Rabbit anti-dMBP, 1:1000, merckmillipore, ab5864; Rabbit anti-Psap, 1:200, abcam, ab300469; Rabbit anti-Gpr37, 1:200, Proteintech Group, 14820-1-AP; Mouse anti-APC[CC-1],1:200,abcam,ab16794; Mouseanti-Gpr37,1:50,Santa Cruz,sc-390110;Rabbit anti-Lef1,1:200,abcam,ab137872).

[0083] 1.10 Single-cell nuclear transcriptome sequencing

[0084] This part is provided by Shenzhen BGI Genomics Co., Ltd. for technical support.

[0085] The subarachnoid hemorrhage model (SAH group, n=3) and sham operation control model (Sham group, n=3) were successfully established in mice. The ipsilateral cerebral hemisphere specimens were quickly collected 48 hours after surgery and immersed in neural tissue preservation medium. Standardized nuclear suspension preparation was performed using a commercial sequencing platform. Nuclear transcriptome libraries were constructed using the Chromium Single-Cell Nucleus Capture System (10x Genomics, v3.0), strictly adhering to the single-cell nuclear sequencing standard operating procedure (SOP-00012). After quantification by Qubit and quality control on an Agilent 2100 Bioanalyzer, paired-end high-throughput sequencing (PE150) was performed using the Illumina NovaSeq 6000 Sequencing System.

[0086] 1.11 Single-cell nuclear transcriptome sequencing analysis process

[0087] (1) Preliminary quality control: After sequencing on the NovaSeq 6000 platform, the BCL-FASTQ format was converted using Illumina's official bcl2fastq (v2.20). The raw sequencing data were initially evaluated for quality using FastQC (v0.11.9). The following filtering criteria were applied to obtain high-quality clean reads: Quality threshold setting: Low-confidence bases were defined based on a Phred quality score < 5. Low-quality bases were removed when the cumulative proportion of low-quality bases in a single read was ≥ 20%; Fuzzy base control: Automatic filtering was initiated when the frequency of unknown bases (N) in single-end reads was > 3.

[0088] (2) Obtaining the original expression matrix: Use CellRanger (version 5.0.0) software provided by 10X to perform further cell quality control, detection, and reference genome (mm10) alignment on the clean reads after preliminary quality control to obtain the original expression matrix file for downstream analysis.

[0089] (3) Downstream analysis: The expression matrix data were read using the Read10X function in the Seurat (version 5.0.1) package. The Seurat object was created using the CreateSeuratObject function and genes expressed in less than 3 cells and cells with less than 200 genes were removed. This study believes that in the damaged model samples, directly removing low-quality cells based on characteristics such as the proportion of mitochondrial genes is likely to overlook many damaged cells, and low-quality cells can be clustered in subsequent clustering, so low-quality cells were not further removed. After standardization and normalization using the SCTransform function, the principal components were found using the RunPCA function. The researchers believed that the cellular heterogeneity of brain tissue was relatively complex. After multiple attempts, npcs = 150 was selected and clustered using UMAP plots and FindNeighbors and FindClusters functions. The default value of the npcs parameter was used for subgroup clustering.

[0090] 1.12 Cell type annotation

[0091] Manual annotation of cell types was performed based on known gene expression signatures and referenced to previous literature reports. The cell marker genes used in this study are listed in the table below.

[0092] Cell marker genes

[0093]

[0094] 1.13 Single-Cell Regulatory Network Inference and Clustering (SCENIC) Analysis

[0095] Some experiments have shown that transcription factors are a type of protein that can directly act on the genome and regulate the DNA transcription process. SCENIC analysis is currently the mainstream software for transcription factor analysis of single-cell sequencing data. This software can also obtain visual results while performing data analysis.

[0096] (1) Based on the single-cell transcriptome expression profile (UMI count matrix), the ensemble tree machine learning algorithm GENIE3 (v1.0.0) was used to construct a gene regulatory network. This framework uses a set of known transcription factors as regulatory feature input, establishes P = dim(G) random forest models (P is the gene matrix dimension), and uses the Gini impurity reduction method to quantify the regulatory weight (ω∈[0,1]) of each transcription factor (TF) on the expression of the target gene. The specific process includes: using the TF expression matrix as the explanatory variable of the supervised learning model, generating a decision tree model after 1000 bootstrap resampling, and calculating the predicted contribution of each TF to the gene transcription level. After verification by Monte Carlo simulation, a double filtering strategy was implemented - eliminating weak correlations with regulatory weight ω<0.05 and excluding low-confidence modules with connectivity <50, and finally obtaining a statistically significant (Pearson's r=0.82, p<1e-5) TF-target gene regulatory network. In this model, the ω value represents the regulatory intensity of the transcription factor and the target gene expression level, and its value is positively correlated with the causal relationship of regulation.

[0097] (2) Enrichment analysis of transcription factor-transcription factor binding motifs and finding corresponding target genes (regulatory groups). This process is implemented by the RcisTarget package. RcisTarget can find the transcription start sites of genes in the gene list and find DNA motifs that appear frequently around the transcription start sites. Subsequently, RcisTarget will search a database containing information on various motifs across the genome of different species, and screen out motifs associated with target transcription factors with a normalized enrichment score (NES) higher than 3.0. Subsequently, for each set of motifs and gene lists, RcisTarget will predict possible target genes. The target genes are the genes with the highest predicted results in the gene list. The target genes shared by all motifs plus the corresponding transcription factors are the regulatory groups.

[0098] (3) SCENIC calculates the activity of a regularity using AUCell. AUCell calculates the area under the curve (AUC) based on the expression level of each gene to calculate the activity of the regularity. Genes are ranked on the x-axis according to their expression level to plot the curve. Therefore, AUC reflects the expression level of a given set of genes relative to other genes in each cell.

[0099] Through AUCell, we can obtain a matrix containing the expression level of each group of regulons relative to other genes in each cell, that is, their activity level. Through this matrix, we can cluster cells and see which regulons are active in different cell types.

[0100] 1.14Monocle2 pseudo-timing analysis

[0101] Monocle2 (v2.22.0), the core algorithmic tool for single-cell transcriptome trajectory inference, employs a semi-supervised learning framework to model cell developmental dynamics. Its core algorithm constructs pseudo-time gradients (pseudotime values ranging from 0 to 100) of cell state transitions by screening the expression profiles of key regulatory genes (DDRTree algorithm, P < 0.01). This process integrates manifold learning techniques, using a reverse graph embedding algorithm to reduce the high-dimensional expression matrix (n × p, where n is the number of cells and p is the number of genes) to a two-dimensional manifold space (k = 2). Cell fate decision trajectories are then constructed based on diffusion graph principles. Biological trajectory reconstruction follows a tree-like topology: a root node is defined in Euclidean distance space, branch points are calculated using a minimum spanning tree algorithm (branch point FDR < 0.05), and cell migration distances along the main-branch pathway are quantified (pseudotime value = geodesic distance to the root node). Algorithm verification shows that compared with methods such as PAGA and Slingshot, Monocle2 has significant advantages in trajectory stability (ARI=0.91vs 0.78) and noise resistance (SNR>8dB).

[0102] 1.15 Cellchat Cell Interaction Analysis

[0103] CellChat (v1.6.0) is a single-cell interactome analysis tool that analyzes cell communication networks using a Bayesian probability model (FDR < 0.05). Its computational framework takes the transcriptome matrix of n cells × m genes as input, integrates 1,234 ligand-receptor interaction pairs (including co-stimulatory factors) from the CellTalkDB database (v4.0), and constructs a probability matrix (p∈

[0104] The core algorithmic process includes: first, identifying differentially expressed ligand / receptor genes (log2FC>1, p.adj<0.01) based on the hypergeometric distribution test, then mapping the expression profiles to the STRING protein interaction network (confidence>0.7) using a random walk algorithm, and finally calculating the statistical significance of the communication probability (AUC>0.85) using Monte Carlo simulation (n=1000 permutation tests). The tool provides multidimensional visualization solutions (t-SNE projection, Chord plots, etc.) to display the topological structure of key signaling pathways (such as Wnt and TGF-β).

[0105] 1.16 Analysis of differentially expressed genes and functional enrichment analysis of single-cell transcriptome data

[0106] Differentially expressed genes (DEGs) were identified using the FindMarkers function, with P < 0.05 and |avg_log2FC| > 0.25 set. Subsequently, Gene Ontology (GO) functional enrichment analysis was performed using the ClusterProfiler package. The nonparametric Wilcoxon rank sum test was used as the default method for differential expression analysis. The data were analyzed using a pseudo-bulk differential expression analysis method. Cells with abnormal expression levels and low gene expression levels were removed, and the snRNA-seq data were integrated into transcriptome-wide expression data. Finally, differential analysis was performed using DESeq2 (P < 0.05, |avg_log2FC| > 0.25).

[0107] 1.17 Proximity Ligation Assay (PLA)

[0108] Using proximity connection technology ( PLA (Sigma-Aldrich) detects protein interactions. The probe system consists of a paired antibody-DNA complex (PLUS / MINUS probe, conjugated to a high-affinity DNA probe and a target protein-specific antibody). When the target protein is spatially adjacent (<40 nm), the probe DNA strand forms a closed loop mediated by a 20-nt linker oligonucleotide. T4 DNA ligase (5 U / μL) catalyzes the formation of a single-stranded DNA template. Rolling circle amplification (RCA) is then performed by polymerase (30°C, 16 h). Finally, the signal is quantified using a fluorescently labeled probe. Animals were anesthetized intraperitoneally with 1% sodium pentobarbital (50 mg / kg) and perfused transcardially with 20 mL of pre-chilled PBS (0.1 M, pH 7.4) followed by 20 mL of 4% PFA for in situ fixation. Whole-brain specimens were carefully dissected from the pia mater and postfixed with 4% PFA for 24 h (4°C). The tissue was then dehydrated in a gradient of 30% sucrose solution until sedimentation occurred. After OCT embedding, the samples were quickly frozen at -80°C for 12 h and then thawed and sectioned using a CM1950 cryostat (Leica) (20 μm coronal section, with reference to the Paxinos atlas for locating the target nucleus). The sections were baked at 37°C for 15 min and fixed with 4% PFA for 15 min. After three 5-min rinses in PBS, the sections were antigen-recovered at 95°C with citrate buffer (pH 6.0) and permeabilized with 0.3% Triton X-100. Endogenous peroxidase was then blocked with 3% H2O2 for 25 min (RT) and blocked with 5% BSA for 1 h (37°C). After incubation in a humidified chamber at 4°C overnight with primary antibodies (rabbit anti-Psap 1:200, ab300469; mouse anti-Gpr37 1:50, sc-390110), the cells were incubated with a PLA probe mixture (1:5 dilution) at 37°C for 1 h. The signals were amplified using ligase (1:40 dilution, 30 min) and polymerase (1:80 dilution, 100 min) in sequence, and the cells were counterstained with DAPI and imaged.

[0109] 1.18 Immunofluorescence colocalization analysis

[0110] Immunofluorescence colocalization analysis usually involves the following steps:

[0111] (1) Use ZEISS analysis software ZEN to open the image for colocalization analysis.

[0112] (2) Separate channels (Image-Color-Split Channels), retaining the two channels with the target protein.

[0113] (3) The colocalization analysis of the two target proteins was performed using the Colocalization (Plugins-Colocalization Finder) function. The default parameters were used and a Pearson's correlation coefficient (PCC) between 0.5 and 1 was considered significant.

[0114] 1.19 Electron Microscope Imaging

[0115] (1) Anesthesia perfusion and sampling: The experimental animals were anesthetized with 1% sodium pentobarbital and immediately underwent transcardial perfusion. 20 mL of glutaraldehyde was perfused to complete sample fixation. After the complete brain tissue was taken out, the corpus callosum sample (1 mm) was microseparated according to the brain stereotaxic atlas under ice bath conditions. 3 ), and fixed in 2.5% glutaraldehyde at 4°C for 2-4 hours.

[0116] (2) Post-fixation: The tissue blocks were rinsed with 0.1 M phosphate buffer (3 × 15 minutes) and then fixed in 1% osmium sulfate at 4°C for 4 hours or overnight.

[0117] (3) Gradient dehydration: The samples were dehydrated in a gradient of 30% → 50% → 70% → 80% → 95% → 100% ethanol (40 min per step), and then replaced with 100% propylene oxide three times (30 min each time).

[0118] (4) Resin infiltration: Gradient infiltration of propylene oxide-embedding solution mixture system (2:1 ratio for 4 hours → 1:2 ratio overnight → pure embedding solution overnight → pure embedding solution for 3-4 hours), and finally embedding in a standard mold for curing.

[0119] (5) Polymerization and curing: The embedded plate is thermally polymerized at 60°C for 48 hours until the resin is completely hardened.

[0120] (6) Ultrathin sectioning: After the resin block is roughly trimmed, it is positioned by semi-thin sectioning using an ultrathin microtome. After fine trimming the target area, 70 nm ultrathin sections are prepared and picked up with a copper mesh.

[0121] (7) Double staining: Uranium staining: stain with 3% uranyl acetate solution in ethanol in the dark for 8 minutes - wash with 70% ethanol 3 times - rinse with ultrapure water 3 times; Lead staining: stain with 2.7% lead citrate solution for 8 minutes - wash with ultrapure water 3 times - blot gently with filter paper.

[0122] (8) Electron microscopy observation: imaging analysis under the accelerating voltage of transmission electron microscope.

[0123] 1.20 Stereotaxic injection of recombinant protein

[0124] Lyophilized recombinant protein powder (abcam, USA, ab167924) was added to an appropriate amount of deionized water (200 μl) to prepare a 0.5 mg / ml working solution. The solution was gently shaken until the protein was completely dissolved. The reconstituted protein solution was aliquoted to avoid multiple freeze-thaw cycles and stored at -80°C for long-term storage. Mice (8 weeks old, 22 g ± 2 g) were placed in an anesthesia induction machine for isoflurane-induced anesthesia (3% concentration). Once fully anesthetized, they were fixed in a prone position on a stereotaxic apparatus (Reward Life Science, China) while maintaining anesthesia with isoflurane (1.5% concentration). A constant temperature blanket was used under the apparatus to maintain the mouse's body temperature at 37°C. The head was shaved and disinfected with an alcohol swab, and both eyes were applied with erythromycin ointment. An approximately 1 cm incision was made on the head to expose the anterior and posterior bregma. The skull was wiped with saline, and the connective tissue and periosteum on the skull surface were removed. The skull surface was then wiped dry with a clean cotton ball. The anterior and posterior bregmas were aligned and aligned, and the left and right sides were aligned. A microsyringe (10 μL, 0.485 mm, Hamilton, USA) was placed at the anterior bregma and the coordinates zeroed. Under a microscope, a small handheld dental drill (Beijing Laiate Technology Development Co., Ltd., China) was used to drill a small hole approximately 0.6 mm in diameter at the injection site (in the bilateral corpus callosum, AP = -1.06 mm, ML = ±0.70 mm, DV = -1.40 mm, perpendicular to the midline) for recombinant protein injection. A microsyringe pump (Reward Life Science Co., Ltd., China) was used to enter the corpus callosum through the hole. After 2 minutes of incubation, the recombinant protein was injected. Subarachnoid hemorrhage models were established after injection.

[0125] 1.21 Data Processing

[0126] In this study, post hoc statistical power analyses were performed using G*Power version 3.1.9.234. Power was calculated based on one-way analysis of variance (ANOVA) by setting the mean difference and standard deviation between groups and using an alpha error of 0.05. A power (1-beta error probability) greater than 0.8 was considered statistically significant. The sample size in this study met statistical requirements. Details of all statistical analyses are detailed in the figure legends, and data are presented as mean ± standard deviation of six independent biological replicates. All data were assessed for normality and homogeneity of variance. Variance correction was performed using the Brown-Forsythe and Welch ANOVA tests when necessary, and the Shapiro-Wilk test was used to assess normality for each dataset. One-way ANOVA was used to compare multiple groups, and P < 0.05 was considered statistically significant. All statistical analyses were performed using GraphPad Prism 9.0.0 software.

[0127] 2. Results

[0128] 2.1 Subarachnoid hemorrhage disrupts the consciousness state of mice and damages the central lateral thalamus-medial prefrontal cortex neural circuit To investigate the effect of SAH on consciousness, researchers performed sleep / wake analysis based on electroencephalogram (EEG) / electromyography (EMG) on adult C57BL / 6J male mice implanted with electrodes. Figure 1 A). Wakefulness is defined as a period characterized by an unsynchronized, low-amplitude EEG rhythm and paroxysmal EMG activity. NREM is characterized by synchronized, high-amplitude, low-frequency (0.5-4 Hz, delta) EEG activity and lower EMG activity compared to wakefulness, without phasic bursts. REM is defined as a period characterized by a prominent theta (4-10 Hz) rhythm and minimal EMG activity. SAH mice had significantly less time awake across the light-dark cycle compared to sham-operated (Sham) mice ( Figure 1 B and C), with corresponding increases in NREM and REM sleep.

[0129] Consciousness depends on large-scale thalamocortical and cortical-cortical connections. Previous studies have identified the central lateral thalamic nucleus (CL) as a key structure for maintaining consciousness. To assess whether SAH disrupts CL-mediated cortical connections, the researchers used anterograde herpes simplex virus (HSV) to perform multisynaptic neural circuit tracing. The virus was injected into the CL region 48 hours after modeling ( Figure 1 D), and brain slices were collected three days later. In sham mice, bright red fluorescent tdTomato signals were detected in mPFC neurons, while in SAH mice, the fluorescent signals in mPFC were significantly weakened, indicating that the CL-mPFC connection was impaired ( Figure 1 E). To further validate these findings, AAV2-retro virus was injected into the mPFC of Sham and SAH mice 48 hours after modeling ( Figure 1 F). 21 days after injection, strong EGFP fluorescence was detected in the CL of sham mice, while the fluorescence intensity was significantly weakened in SAH mice ( Figure 1 G), demonstrating disruption of the reciprocal connectivity between the CL and mPFC.

[0130] Furthermore, the researchers used optogenetics to examine functional impairments in the CL-mPFC circuit. Given that CL neurons are primarily glutamatergic, the researchers used AAV carrying the CaMKIIa-ChrimsonR construct for optogenetic activation, while simultaneously measuring calcium activity in the mPFC using the AAV-Syn-GCaMP6s construct. Figure 1H). Optogenetic stimulation (633 nm, 20 Hz, 10 ms pulses, 5 mW, 5 s on-25 s off cycle) was applied to activate ChrimsonR-expressing CL neurons while recording calcium transients in the mPFC. In sham mice, stimulation of CL induced a significant increase in calcium signals in the mPFC (quantified by the area under the curve AUC). However, in SAH mice, optogenetic activation failed to induce a significant calcium response, indicating a profound disruption of functional connectivity within the CL-mPFC circuit ( Figure 1 IL).

[0131] 2.2 Subarachnoid hemorrhage disrupts the central lateral thalamus-medial prefrontal cortex circuit connection

[0132] To investigate the molecular mechanism of SAH-induced CL-mPFC circuit damage, the researchers performed single-cell RNA sequencing (snRNA-seq) on the hemisphere brain tissue of mice 48 hours after SAH modeling, ranging from the mPFC starting point (Bregma 2.34mm) to the CL end point (Bregma -2.18mm) ( Figure 2 A, n = 3). A total of 26,589 cell nuclei were analyzed, of which 13,734 were from sham-operated (Sham) brain tissue and 12,855 were from SAH brain tissue. Cell types including excitatory neurons (EN), inhibitory neurons (IN), oligodendrocytes, astrocytes, vascular and leptomeningeal cells (VLMC), and endothelial cells (EC) were identified and visualized using UMAP ( Figure 2 B) The accuracy of these classifications was validated by heatmaps of differentially expressed genes across annotated cell types and bubble plots of cell type classifications based on classic cell-specific markers.

[0133] Further snRNA-seq analysis revealed transcriptional heterogeneity within excitatory neurons, classifying them into six distinct clusters (clusters 1-6) based on their unique transcriptome signatures. Based on these transcriptome differences, excitatory neurons were further divided into six subtypes, EXN1-EXN6 ( Figure 2 C and D). Notably, EXN1 is enriched for Lef1, a gene expressed primarily in the thalamus, as confirmed by the Allen Brain Atlas ( Figure 2 E). Lef1 expression was highly specific to EXN1 in our snRNA-seq dataset (Figure S2E-G), indicating that EXN1 represents a specific excitatory neuron subtype within the thalamus.

[0134] To investigate the effects of SAH on EXN1 isoforms, we performed KEGG and Reactome pathway enrichment analyses of genes upregulated in EXN1 neurons of SAH mice compared with sham controls. These analyses revealed that SAH-induced upregulated genes in EXN1 were primarily related to ion channel activity and Wnt signaling pathways, rather than pathways associated with neuronal injury or cell death ( Figure 2 Given the role of ion channels and Wnt signaling in neuronal excitability, we further investigated whether SAH affects calcium dynamics within the CL. The calcium-sensitive fluorescent probe GCaMP6s was injected into the CL, followed by implantation of an optical fiber for fiber photometry recording ( Figure 2 I and J). SAH mice showed significantly elevated spontaneous calcium signals in the CL compared to Sham mice ( Figure 3 KM), indicating increased neuronal activity.

[0135] To further evaluate neuronal excitability, we performed patch clamp electrophysiological recordings in the CL (starting point) and mPFC (termination point) of the CL-mPFC pathway ( Figure 2 In CL, SAH neurons showed significantly higher action potential (AP) frequency and more depolarized resting membrane potential (RMP) compared with sham control ( Figure 2 In contrast, there were no significant differences in AP frequency or RMP in mPFC neurons between SAH and Sham groups ( Figure 2 SU). These findings suggest that SAH does not lead to significant neuronal loss in the CL, but rather to increased neuronal excitability. However, mPFC electrophysiology remains unchanged, suggesting that although CL-mPFC circuit dysfunction occurs after SAH, it is not the result of CL neuronal degeneration and death. Instead, this dysfunction may result from disruption of functional connectivity within the CL-mPFC pathway.

[0136] 2.3 Subarachnoid hemorrhage-induced myelin damage and axonal injury in the central lateral thalamo-medial prefrontal cortex circuit accompanied by oligodendrocyte dysfunction

[0137] To investigate the role of oligodendrocytes in the disruption of the CL-mPFC pathway induced by subarachnoid hemorrhage (SAH), we first performed MRI scans using a 7.0T animal scanner to obtain T2-weighted images. T2-hyperintense areas were observed in key white matter regions of SAH mice, including the CC, external capsule, fornix, and internal capsule / cerebral peduncle / optic tract (IC / CP / OT). Figure 3 To investigate the damage of CL-mPFC circuit conduction after subarachnoid hemorrhage, we first stereotaxically injected AAV9-mCherry into the CL nucleus of mice ( Figure 3C). 21 days after transfection, mice underwent SAH or sham surgery. 48 hours after surgery, brain tissue was collected and frozen and sectioned. Immunofluorescence staining was performed on the CC, a representative white matter region of the CL-mPFC circuit. Notably, we observed significant morphological changes 48 hours after SAH. Compared with the sham group, SAH mice showed increased myelin basic protein degradation (dMBP) and decreased myelin basic protein (MBP) ( Figure 3 CE). These findings suggest that SAH induces myelin damage, leading to impaired conduction in the CL-mPFC pathway.

[0138] Oligodendrocytes play a key role in the central nervous system by forming and maintaining the myelin sheath to support the rapid conduction of neural signals. In addition, oligodendrocytes remain active after myelination, maintaining functional connections with axons through myelin channels and supporting long-term axonal integrity. To explore how SAH affects oligodendrocyte function, we isolated oligodendrocytes from single-cell nuclear sequencing data. Reclustering and dimensionality reduction showed that there was a significant separation of oligodendrocytes between the Sham and SAH groups, indicating a major shift in the state of oligodendrocytes after SAH ( Figure 3 F). Differential expression analysis found that the genes downregulated in oligodendrocytes after SAH were mainly related to lipid metabolism ( Figure 3 G,H). Given that lipids constitute approximately 70% of myelin and are crucial for axonal insulation and rapid conduction of nerve impulses, downregulation of lipid metabolism-related genes suggests that myelin lipid metabolism is disturbed, which may lead to myelin damage and subsequent axonal injury.

[0139] After SAH, oligodendrocytes exhibited significant cell state transitions. To further explore these transitions, we applied Monocle2 to accurately reconstruct the complex mononuclear trajectory. Notably, in the developmental trajectory, the Sham and SAH groups were clearly separated after State 1. Gene expression analysis identified five gene clusters with distinct expression patterns, and GO enrichment analysis revealed that these clusters were enriched in pathways related to cellular ion homeostasis, sterol metabolism, lipid transport, and myelination ( Figure 3 I). Cluster 5, which is associated with myelination, showed a gradual downregulation from Sham to SAH, suggesting its potential role in myelin damage after SAH. Key myelination-related genes in cluster 5, such as Abca2, Sox10, Mal, Mbp, Psap, and Plp1, were significantly downregulated after SAH, with the decrease in Abca2, Mal, Mbp, and Psap being particularly prominent ( Figure 3Abca2 regulates lipid metabolism, particularly sphingolipid metabolism and cholesterol homeostasis in late endosomes / lysosomes. Mal plays a role in stabilizing glycosphingolipid-rich membrane microdomains, while Mbp is a major component of oligodendrocytes and myelin. Psap, which supports myelin maintenance, showed the most significant downregulation in pseudo-time series analysis ( Figure 3 J), suggesting that it may play a key role in myelin damage after SAH.

[0140] Subarachnoid hemorrhage-induced disruption of Psap-Gpr37 interaction in oligodendrocytes

[0141] To investigate the potential mechanisms of myelin damage and oligodendrocyte dysfunction after SAH, we performed CellChat cell interaction analysis. The results showed that the intensity of cell-cell interactions was significantly reduced in the SAH group compared with the Sham group. Notably, the PSAP signaling pathway showed a significant decrease after SAH. The significant downregulation of the PSAP signaling pathway was confirmed by ligand-receptor analysis, in which Psap served as a ligand and Gpr37 served as a receptor, both of which are expressed in oligodendrocytes ( Figure 4 A). These findings are consistent with the observed reduction in oligodendrocyte self-interactions after SAH. To verify these findings at the protein level, colocalization analysis of Psap and Gpr37 in the mouse brain corpus callosum showed a Pearson correlation coefficient as high as 0.84, indicating a significant interaction between Psap and Gpr37 ( Figure 4 B). In addition, Duolink proximity ligation assay (PLA) confirmed that Psap-Gpr37 interaction was significantly reduced in the corpus callosum of the SAH group compared with the sham control group, further supporting the observation that the interaction is decreased after SAH ( Figure 4 C, D). Since PSAP is a secretory protein, to verify its changes in the plasma of SAH mice and patients, we collected plasma samples from SAH patients and healthy controls for ELISA experiments. The results showed that PSAP was downregulated in the plasma of both mice and patients ( Figure 4 E, F). These results confirm that Psap-Gpr37 interaction is downregulated after SAH, indicating disruption of the PSAP signaling pathway.

[0142] 2.5 Subarachnoid hemorrhage induces downregulation of Psap expression in oligodendrocytes, while Gpr37 expression remains unchanged

[0143] To explore the mechanism by which subarachnoid hemorrhage (SAH) induces a decrease in Psap-Gpr37 interaction in oligodendrocytes, we used single-cell nuclear sequencing data to examine the effects of SAH on the expression of Psap and Gpr37 in oligodendrocytes. The analysis showed that Psap expression was significantly decreased in the SAH group, while Gpr37 expression did not change significantly ( Figure 4 G, H). Immunofluorescence staining of the corpus callosum of mouse brain tissue confirmed the downregulation of Psap expression in oligodendrocytes in the SAH group ( Figure 4 I, J). In addition, there was no significant difference in Gpr37 expression between the SAH group and the Sham group, which was consistent with the transcriptome data ( Figure 4 K, L). Similarly, immunofluorescence analysis of human brain tissue from SAH patients and non-SAH controls showed that Psap protein levels were decreased in oligodendrocytes after SAH, while Gpr37 protein levels did not change significantly ( Figure 4 MP). Psap is a secreted protein crucial for neuroprotection and glial support, exerting its effects through interactions with GPR37 and GPR37L1 receptors. These findings suggest that SAH-induced downregulation of Psap expression may lead to reduced Psap-Gpr37 interaction, potentially causing myelin damage and impaired conduction in the CL-mPFC circuit after SAH.

[0144] 2.6 Psap overexpression enhances Psap-Gpr37 interaction and reduces myelin and axonal damage after subarachnoid hemorrhage

[0145] To investigate whether regulating Psap expression could improve myelin / axonal damage after subarachnoid hemorrhage (SAH), we used AAV virus driven by the MBP promoter and carrying mCherry marker to overexpress Psap specifically in oligodendrocytes. The virus was injected into the bilateral corpus callosum ( Figure 5 A). High mCherry signal in the corpus callosum confirms successful viral transfection ( Figure 5 B). Immunofluorescence staining further confirmed that this AAV overexpression strategy significantly upregulated Psap levels in oligodendrocytes after SAH ( Figure 5 E). In addition, proximity ligation assay (PLA) showed that Psap overexpression significantly enhanced Psap-Gpr37 interaction in the corpus callosum after SAH ( Figure 5 C,D).

[0146] After Psap overexpression, we evaluated its effect on myelin and axonal integrity after SAH. T2-weighted MRI scans showed that the SAH+Psap OE group had significantly less white matter damage compared to the SAH+Vec group, as evidenced by a decrease in T2 hyperintensities in the corpus callosum ( Figure 5F, G). Immunofluorescence analysis further confirmed the reduction of white matter damage in the SAH+Psap OE group. The results showed that in AAV9-transfected mice, MBP fluorescence increased in the corpus callosum of the SAH+Psap OE group, while dMBP and APP fluorescence decreased. These findings suggest that Psap overexpression has a protective effect on myelin structure after SAH ( Figure 5 H, I). Electron microscopy further confirmed that Psap overexpression alleviated myelin damage caused by SAH, as evidenced by the more complete myelin sheath structure in the SAH+Psap OE group compared with the SAH+Vec group ( Figure 5 J). In addition, g-ratio analysis showed that the myelin integrity was improved in the SAH+Psap OE group, as indicated by a lower g-ratio, reflecting thicker and better-intact myelin sheaths compared with the SAH+Vec group ( Figure 5 K).

[0147] Together, these findings provide strong evidence that Psap overexpression enhances the interaction between Psap and Gpr37, an effect that preserves myelin integrity and reduces axonal damage after SAH, likely through a neuroprotective effect on oligodendrocytes. This protective effect is associated with the maintenance of corpus callosum structural integrity, suggesting that the Psap-Gpr37 interaction plays a crucial role in maintaining neural network connectivity after SAH.

[0148] 2.7 Psap overexpression improves central lateral thalamus-medial prefrontal cortex circuit conduction and alleviates impaired consciousness after subarachnoid hemorrhage

[0149] To examine the effect of Psap overexpression on CL-mPFC circuit conduction, the researchers conducted optogenetic experiments. They injected the light-sensitive channel protein ChrimsonR into the CL nucleus, GCaMP6s into the mPFC, and injected AAV virus overexpressing Psap into the bilateral corpus callosum of mice. After the injection, optical fibers ( Figure 6 A and B). After 21 days of viral expression and SAH modeling, optogenetic experiments were performed 48 hours after SAH. The results showed that compared with the control group, the calcium signal in the downstream mPFC was significantly enhanced in the Psap overexpression group under 633nm light stimulation of the upstream CL nucleus ( Figure 6 CE), indicating that Psap overexpression alleviated SAH-induced CL-mPFC circuit conduction disturbances.

[0150] To evaluate whether Psap overexpression improves consciousness, the researchers performed electroencephalographic recordings and sleep-wake analysis 24 hours after SAH. Figure 6FG). The Psap overexpression group showed a significant increase in the proportion of wakefulness and a decrease in the proportion of NREM and REM sleep, indicating that the consciousness state of SAH mice was improved. It is known that impaired consciousness caused by brain damage will affect the recovery of cognitive function. The CL-mPFC circuit plays a key role in maintaining high-level cognitive functions. In view of this, we further used the Morris water maze test to evaluate cognitive function. In the spatial learning acquisition stage, the time it took SAH mice to find the hidden platform was significantly prolonged compared with Sham mice. In SAH, the escape latency of the Psap+OE group was significantly shortened compared with the Psap+Vec group, especially at 23-26 days, indicating that Psap overexpression improved the spatial learning deficit induced by SAH; in addition, the Psap+OE group also showed more platform crossings and stayed in the target quadrant for a longer time ( Figure 6 HL), indicating that Psap overexpression alleviated cognitive impairment in SAH mice.

[0151] 2.8 Recombinant human prosaposin activator protein (HuPSAP) restores central lateral thalamus-medial prefrontal cortex circuit conduction and alleviates consciousness and cognitive impairment after subarachnoid hemorrhage

[0152] We next used optogenetics to investigate the therapeutic effect of HuPSAP on CL-mPFC circuit damage after subarachnoid hemorrhage (SAH). The photosensitive channel protein ChrimsonR was stereotactically injected into the CL nucleus, GCaMP6s was injected into the mPFC, and optical fibers were implanted at two sites. After 21 days of viral expression, 0.25ng / μl, 0.5ng / μl, 0.75ng / μl, and 1ng / μl of HuPSAP (recombinant human PSAP protein purchased from Abcam (Shanghai) (ab167924), the sequence is shown in SEQ ID No.1) were injected into the corpus callosum. After a 7-day recovery period, SAH modeling was performed, and optogenetic stimulation was performed 48 hours after SAH ( Figure 7 A, B).

[0153] SEQ ID No.1

[0154] ASGPVLGLKE CTRGSAVWCQ NVKTASDCGAVKHCLQTVWN KPTVKSLPCD ICKDVVTAAGDMLKDNATEE EILVYLEKTC DWLPKPNMSASCKEIVDSYL PVILDIIKGE MSRPGEVCSA LNLCESLQKHLAELNHQKQL ESNKIPELDM TEVVAPFMAN IPLLLYPQDG PRSKPQPKDN GDVCQDCIQM VTDIQTAVRTNSTFVQALVE HVKEECDRLG PGMADICKNYISQYSEIAIQ MMMHMQPKEI CALVGFCDEV KEMPMQTLVPAKVASKNVIP ALELVEPIKKHEVPAKSDVY CEVCEFLVKE VTKLIDNNKT EKEILDAFDK MCSKLPKSLSEECQEVVDTYGSSILSILLE EVSPELVCSM LHLCSGTRLP ALTVHVTQPK DGGFCEVCKKLVGYLDRNLEKNSTKQEILA ALEKGCSFLP DPYQKQCDQF VAEYEPVLIE ILVEVMDPSF VCLKIGACPSAHKPLLGTEK CIWGPSYWCQ NTETAAQCNA VEHCKRHVWN KLHHHHHH.

[0155] After 633 nm light stimulation of the upstream CL region, the calcium signal in the downstream mPFC region was significantly enhanced in the HuPSAP treatment group, especially in the 0.75 ng / μl concentration group, compared with the saline control group, indicating that the CL-mPFC loop conduction was improved after SAH ( Figure 7 However, there was no significant difference between the 0.75 ng / μl and 1 ng / μl HuPSAP treatment groups.

[0156] To evaluate whether HuPSAP injection also alleviated consciousness and cognitive impairment, electroencephalogram recordings and sleep-wake analysis were performed 24 hours after SAH. Compared with the control group, the HuPSAP group showed a significant increase in wakefulness time and a decrease in NREM and REM sleep, indicating that the consciousness state of SAH mice was improved ( Figure 7 FG).

[0157] In addition, cognitive function was assessed using the Morris water maze test. During the spatial learning acquisition phase, there was no significant difference in swimming speed between the groups. However, the escape latency in the HuPSAP group was significantly shortened compared with the control group, especially on days 23-26, indicating improved spatial learning ability ( Figure 7 HJ). During the exploration phase, the HuPSAP group showed significantly more platform crossings and stayed in the target quadrant longer, further indicating that HuPSAP injection alleviated cognitive impairment after SAH ( Figure 7 K, L).

[0158] 2.9 Overexpression of HuPSAP and Psap attenuates subarachnoid hemorrhage-induced white matter damage in the central lateral thalamus-medial prefrontal cortex pathway

[0159] To directly assess subarachnoid hemorrhage (SAH)-induced white matter damage in the CL-mPFC pathway, we stereotaxically delivered AAV2 / 9-hsyn-mCherry to the CL nucleus to anterogradely label axons traversing the CC. Immunofluorescence analysis revealed a significant increase in dMBP accumulation within mCherry-positive regions in the CC of SAH mice ( Figure 8 A, B), indicating demyelination and white matter structural destruction. Notably, SAH mice overexpressing Psap or receiving HuPsap had significantly reduced dMBP levels in the CC compared with controls receiving vehicle or saline ( Figure 8 A, B), indicating that Psap restoration mitigates SAH-induced white matter damage in this pathway. Our findings delineate a pathophysiological cascade linking SAH to impaired consciousness: (1) SAH induces white matter damage, particularly in the CC, disrupting axonal integrity and impairing signaling within the CL-mPFC circuit; (2) Psap deficiency exacerbates this process, while HuPsap administration rescues axonal transport deficits by attenuating myelin sheath breakdown and oligodendrocyte dysfunction; (3) Preserving intact CL-mPFC connectivity is associated with improved behavioral markers of consciousness. This work establishes Psap as a key mediator of white matter repair after SAH and highlights circuit-specific therapeutic strategies for restoring consciousness.

Claims

1. Application of overexpression of PSAP in improving secondary brain injury after subarachnoid hemorrhage.

2. The use according to claim 1, characterized in that The application includes overexpressing PSAP using HuPSAP.

3. The use according to claim 2, characterized in that The HuPSAP is used to reduce myelin damage after subarachnoid hemorrhage, alleviate axonal pathological damage, and improve the consciousness disorder and neurological function prognosis induced by subarachnoid hemorrhage.

4. The use according to claim 1, characterized in that The secondary brain damage after subarachnoid hemorrhage includes one or more of the following: disrupts consciousness and damages the central-lateral thalamus-medial prefrontal cortex neural circuit; Disruption of central-lateral thalamic-medial prefrontal cortex circuit connectivity; Myelin damage and axonal injury in the central-lateral thalamo-medial prefrontal cortex circuit are accompanied by oligodendrocyte dysfunction; Disruption of Psap-Gpr37 interaction in oligodendrocytes.

5. The use according to claim 2, characterized in that The concentration of HuPSAP is 0.25-1 ng / μl.

6. The use according to claim 2, characterized in that The HuPSAP is an injection.

7. The use according to claim 2, characterized in that The HuPSAP is used to protect the conduction function of the CL-mPFC circuit after subarachnoid hemorrhage and alleviate consciousness disorders.

8. The use according to claim 2, characterized in that The HuPSAP is used to alleviate cognitive impairment after subarachnoid hemorrhage.

9. The use according to claim 2, characterized in that The HuPSAP is used to alleviate subarachnoid hemorrhage-induced white matter damage in the CL-mPFC pathway.

10. The use according to claim 2, characterized in that The HuPSAP is used to rescue axonal transport defects by alleviating myelin breakdown and oligodendrocyte dysfunction.

Images