Disease treatment mechanism analysis method, system and application based on combination of single cell sequencing, spatial transcriptome and mass spectrometry imaging

Abstract

The application provides a disease treatment mechanism analysis method, system and application based on single-cell sequencing, spatial transcriptome and mass spectrometry imaging, and specifically belongs to the technical field of multi-omics analysis.The disease treatment mechanism analysis method comprises the following steps: preparing a tissue section;respectively performing tissue staining, single-cell sequencing, Spatial-seq 2.0 spatial transcriptome sequencing and mass spectrometry imaging on the tissue section; and performing single-cell spatial multi-omics affine registration and data joint analysis.The disease treatment mechanism analysis method fuses the data of multiple omics, has high spatial resolution, can realize cell subtype positioning, can comprehensively analyze the multidimensional mechanism of drug treatment of complex diseases from gene expression to metabolite, and breaks through the spatial and data integration limitations of traditional technologies.

Description

Technical Field

[0001] This invention belongs to the field of multi-omics analysis technology, specifically involving methods, systems, and applications for elucidating disease treatment mechanisms based on the combined use of single-cell sequencing, spatial transcriptomics, and mass spectrometry imaging. Background Technology

[0002] Elucidating the mechanisms by which drugs treat diseases is beneficial for revealing the complexity of diseases, promoting the development of targeted drugs, improving treatment efficacy, reducing medical costs, and thus driving medical progress. Currently, specific analytical methods include molecular biology methods, cell biology methods, clinical research methods, and bioinformatics methods. Among these, molecular biology methods include metabolomics, gene expression analysis, and proteomics.

[0003] Currently, the most commonly used technologies are single-omics techniques, such as single-cell sequencing, traditional spatial transcriptomics, and mass spectrometry imaging. Single-cell sequencing can only analyze cellular heterogeneity but lacks spatial localization (e.g., it cannot distinguish gene expression differences between the hippocampus and the cortex); traditional spatial transcriptomics (e.g., Visium) has low resolution (>500 μm) and cannot accurately locate cerebellar regions (e.g., the striatum); mass spectrometry imaging (MALDI-MSI) can provide spatial distribution of metabolites but cannot correlate gene expression or cell type. Existing technologies have also attempted to combine various analytical methods, such as combining single-cell and spatial transcriptomics, but currently, this combination method relies on coarse registration (e.g., region segmentation), leading to a mismatch between spatial and cellular resolution (error >30%); the combination of spatial transcriptomics and mass spectrometry imaging lacks single-cell-level cell subtype analysis, making it difficult to reveal the effects of drugs on specific cells (e.g., anti-inflammatory microglia). Currently, there is a lack of a multi-omics approach that comprehensively understands the therapeutic mechanisms of drugs at every level from gene expression to metabolites. Summary of the Invention

[0004] The purpose of this invention is to provide a method, system, and application for elucidating disease treatment mechanisms based on the combined use of single-cell sequencing, spatial transcriptomics, and mass spectrometry imaging. The disease treatment mechanism elucidation method described in this invention integrates data from multiple omics disciplines, achieving high spatial resolution and enabling cell subtype localization. It can comprehensively and systematically analyze the multi-dimensional mechanisms of drug treatment for complex diseases, from gene expression to metabolites, overcoming the spatial and data integration limitations of traditional technologies.

[0005] This invention provides a method for elucidating disease treatment mechanisms based on the combined use of single-cell sequencing, spatial transcriptomics, and mass spectrometry imaging, comprising the following steps: Tissue sections were prepared; tissue sections were stained, single-cell sequencing, Spatial-seq 2.0 spatial transcriptome sequencing, and mass spectrometry imaging were performed on the tissue sections; single-cell spatial multi-omics affine registration and data analysis were performed.

[0006] Preferably, the method of tissue staining includes HE staining.

[0007] Preferably, the Spatial-seq 2.0 spatial transcriptome sequencing is a high-throughput spatial transcriptome sequencing method based on a laser microdissection system and DNA barcode labeling.

[0008] Preferably, the single-cell sequencing includes single-cell transcriptome sequencing; the platform for the single-cell transcriptome sequencing includes the 10x Genomics-Chromium platform, the BD Rhapsody platform, or the BGI Genomics-DNBelab C4 platform.

[0009] Preferably, the method for single-cell transcriptome sequencing includes: homogenizing tissue sections, enzymatically digesting them to obtain single cells, constructing a transcriptome sequencing library, and performing sequencing on a single-cell transcriptome sequencing platform.

[0010] Preferably, the mass spectrometry imaging includes MALDI-MSI mass spectrometry imaging.

[0011] Preferably, the single-cell spatial multi-omics affine registration and data joint analysis includes: Construct a spatial reference coordinate system, map single-cell to spatial transcriptome data, and perform affine registration of multimodal data; The construction of the spatial reference coordinate system includes: using QuickNII software to extract histological features from HE-stained sections, establishing an initial spatial coordinate system based on ventricular structure, hippocampal dentate gyrus morphology and cortical layering features, and registering it to the Allen Brain Atlas standard reference map to complete millimeter-level spatial positioning; The single-cell-spatial transcriptome data mapping includes: joint analysis of single-cell transcriptome sequencing data and Spatial-seq 2.0 data; multimodal data fusion is performed using the Tangram algorithm: using single-cell sequencing data as input, cell types are identified based on the Leiden clustering algorithm, and a cell type-specific gene expression feature matrix is ​​constructed. The multimodal data affine registration includes: morphological erosion processing (3×3 structural nuclei) of HE-stained sections, Spatial-seq 2.0 spatial transcriptome raw images and mass spectrometry imaging sections, extraction of tissue contour feature point clouds, and feature point matching. After registration, multi-omics data of five independent regions, namely the hippocampus, striatum, motor cortex, sensory cortex and olfactory cortex, can be analyzed independently.

[0012] Preferably, the tissue sections are serial sections; the thickness of the tissue sections for tissue staining is 5-18 μm; the thickness of the tissue sections for single-cell sequencing is 20-50 μm; the thickness of the tissue sections for Spatial-seq 2.0 spatial transcriptome sequencing is 10-20 μm; and the thickness of the tissue sections for mass spectrometry imaging is 10-20 μm.

[0013] The present invention also provides a disease treatment mechanism analysis system based on the combined use of single-cell sequencing, spatial transcriptomics and mass spectrometry imaging, including a single-cell sequencing unit for obtaining single-cell sequencing data; Spatial-seq 2.0 spatial transcriptome sequencing unit, used to obtain Spatial-seq 2.0 spatial transcriptome sequencing data; Mass spectrometry imaging unit, used to acquire space metabolomics data; The unit includes a single-cell spatial multi-omics affine registration and data analysis unit, used to perform affine registration and analysis on data acquired by the single-cell sequencing unit, the Spatial-seq 2.0 spatial transcriptome sequencing unit, and the mass spectrometry imaging unit.

[0014] This invention also provides the application of the disease treatment mechanism analysis method or the disease treatment mechanism analysis system described in the above technical solutions in any one of the following studies: ① disease mechanism analysis research; ② drug development research; ③ biomarker discovery research; ④ developmental and regenerative medicine research; ⑤ tumor immunotherapy development research; the application is for non-clinical treatment purposes.

[0015] This invention provides a method for elucidating disease treatment mechanisms based on the combined use of single-cell sequencing, spatial transcriptomics, and mass spectrometry imaging. This invention combines single-cell sequencing, Spatial-seq 2.0, and mass spectrometry imaging; it is not a simple superposition of individual technologies, but rather a systematic analytical framework integrating "gene-cell-metabolism-space" through data complementarity and algorithm integration, enabling in-depth exploration of complex disease mechanisms.

[0016] The experimental results show that the method for analyzing the disease treatment mechanism of the present invention has the following beneficial effects: Synergy and complementarity: Single-cell sequencing resolves cellular heterogeneity but lacks spatial information; Spatial-seq 2.0 provides high-precision spatial localization of gene expression (50 μm) but cannot distinguish subtype functions; Mass spectrometry imaging provides dynamic distribution of metabolites but cannot correlate genes or cell types. Combining these three methods, single-cell data is mapped to spatial coordinates using the Tangram algorithm, precisely correlating cell subtypes (e.g., single-cell sequencing and mass spectrometry). Socs3 hiSpatial colocalization of microglia and metabolites (such as LPC) overcomes the data fragmentation inherent in single technologies.

[0017] Dynamic network construction: Traditional methods can only provide isolated level analysis (such as genes or metabolism), while this invention reveals the cascade mechanism by which drugs regulate lipid metabolism through anti-inflammatory microglia by integrating: gene expression (Spatial-seq 2.0) → cell function (single-cell sequencing) → metabolic regulation (mass spectrometry imaging).

[0018] Precise mechanism positioning: The "space-cell-function" correlation problem that cannot be solved by a single technology is addressed in this invention through cross-omics validation: for example, mass spectrometry imaging reveals abnormal PC metabolism in the hippocampus, and spatial-seq locates this region. Socs3 High gene expression and single-cell sequencing further identified the anti-inflammatory microglia subtype as the key cell type, forming a closed-loop chain of evidence.

[0019] In a stroke model, single-cell sequencing alone could identify anti-inflammatory microglia, but could not pinpoint their concentration in the hippocampus; Spatial-seq 2.0 alone could detect abnormal gene expression in the hippocampus, but could not distinguish cell subtypes; and mass spectrometry imaging alone could detect LPC accumulation, but could not explain the mechanism. The combined use of these three methods revealed for the first time that NGR1 upregulates the hippocampus... Socs3 hi Cellular (single-cell data), inhibition of local LPC generation (mass spectrometry imaging), thereby reducing inflammatory damage (spatial transcriptomics validation), demonstrates the irreplaceable nature of the combined technologies.

[0020] This invention transforms multi-dimensional data into a dynamically interacting biological network through technological collaboration and algorithm integration, solving the problem of systematic mechanism analysis that cannot be achieved by traditional single or pairwise technology combinations, and providing a brand-new research paradigm for the treatment of complex diseases. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 A flowchart of the method for elucidating disease treatment mechanisms based on the combined use of single-cell sequencing, spatial transcriptomics, and mass spectrometry imaging provided by this invention; Figure 2The overall efficacy evaluation results of NGR1 on MCAO mice provided by the present invention are shown in the figure; (A) is a schematic diagram of the 24-hour efficacy experiment protocol; (B) is a figure showing the effect of different doses of NGR1 on the neurobehavioral scores of MCAO mice. Figure 3 This is a representative HE staining image of the mouse brain tissue as a whole and the cortical region provided by the present invention; Figure 4 The single-cell atlas of NGR1-treated MCAO mice provided for this invention includes: (A) a schematic diagram of the single-cell sequencing process; (B) a t-SNE diagram of 37,212 cells with 24 cell clusters (left) and 14 cell types (right); and (C) a map showing the expression levels of marker genes for the 14 cell types, displayed in blue. Figure 5 The following diagrams provide a comprehensive analysis of microglia subtypes for this invention: (A) Diagram showing the identification of 11,791 microglia into 6 subtypes; (B) Diagram showing the expression results of the marker genes for the six subtypes; (C) Diagram showing the number of each microglia subtype in the three samples; (D) Diagram showing the proportion of each microglia subtype in the three samples; (E) Diagram showing the proportion of each microglia subtype in the three samples. Socs3 hi (F) GO analysis results of high-expression genes in microglia subsets; (G) Prediction results of differentiation degree of different cell subtypes; (H) Pseudo-time trajectory results of different microglia subtypes; (I) High-expression marker genes of anti-inflammatory subtypes. Socs3 Expression level violin plot; (J) ELISA results of changes in Socs3 content in brain tissue; Figure 6 Representative images of brain lipids obtained by MALDI-MSI imaging provided by the present invention; wherein, (A) spatial metabolic changes of various lipids; (B) spatial metabolic changes of lipid-related metabolites; Figure 7 The following diagrams show the spatial metabolite analysis results provided for this invention: (A) a schematic diagram of LCM and mass spectrometry imaging data registration; (B) diagrams showing the recovery regulation rates of PC, PE, LPC, and LPE in different spatial regions; (C) diagrams showing the relative abundance changes of representative lipids in the hippocampus; and (D) diagrams showing the relative abundance changes of representative lipids in the striatum. Figure 8 The spatial heterogeneity of microglia regulated by NGR1 in the present invention is illustrated in the diagram; (A) a schematic diagram of single-cell spatial mapping; (B) a diagram showing the spatial distribution of microglia in different groups; (C) Socs3 Map showing the spatial distribution differences of genes; (D) Adam8 Map showing the spatial distribution differences of genes; Figure 9 This is a schematic diagram illustrating the panoramic mechanism of action of Panax notoginseng saponin R1 against acute ischemic stroke at the spatial level, as provided by the present invention. Detailed Implementation

[0023] This invention provides a method for elucidating disease treatment mechanisms based on the combined use of single-cell sequencing, spatial transcriptomics, and mass spectrometry imaging, comprising the following steps: Tissue sections were prepared; tissue sections were stained, single-cell sequencing, Spatial-seq 2.0 spatial transcriptome sequencing, and mass spectrometry imaging were performed on the tissue sections; single-cell spatial multi-omics affine registration and data analysis were performed.

[0024] Single-cell sequencing reveals differences in gene expression at the individual cell level, spatial transcriptomics data provides the spatial distribution of these differences in tissues, and spatial metabolomics data reveals the spatial distribution of metabolites. This invention integrates multiple omics data to comprehensively understand the biological basis of stroke at every level, from gene expression to metabolites. The disease treatment mechanism analysis method described in this invention has high spatial resolution, overcoming the limitations of traditional methods in achieving high-precision (<100 μm) brain region-specific analysis; it can construct a dynamic "gene-cell-metabolism" network, overcoming the data isolation problem inherent in single omics or pairwise combinations; it can accurately analyze the spatial regulatory effects of drugs on specific cell subtypes, addressing the deficiency of cell subtype localization in existing technologies. This invention, by integrating single-cell sequencing (cell heterogeneity), Spatial-seq 2.0 (high-resolution spatial transcriptomics), and mass spectrometry imaging (metabolite spatial dynamics), systematically analyzes the multi-dimensional mechanisms of drug treatment for complex diseases, breaking through the spatial and data integration limitations of traditional technologies.

[0025] The flowchart of the disease treatment mechanism analysis method based on the combined use of single-cell sequencing, spatial transcriptomics, and mass spectrometry imaging of this invention is as follows: Figure 1As shown. After preparing tissue sections, the tissue sections are stained. The stained tissue is used as a reference for registration. Specifically, after tissue staining, images are collected for pathological analysis and registration analysis. In a specific embodiment, the tissue includes animal tissue. In a specific embodiment, the preparation of animal tissue includes taking animal organs, wiping away surface blood, quick-freezing with liquid nitrogen, and storing in a refrigerator. In a specific embodiment, the quick-freezing time with liquid nitrogen can be 1 minute; the storage temperature in the refrigerator can be -80℃. The mechanism of action of Panax notoginseng saponin R1 (NGR1) against acute stroke, as analyzed in this embodiment, was used to construct a mouse animal model. In a specific embodiment, the animal organs include mouse brain tissue. This invention uses a cryostat for sectioning. In a specific embodiment, the tissue staining method includes HE staining. In a specific embodiment, the tissue sections are continuous sections; the thickness of the tissue sections for tissue staining is 5-18 μm; the thickness of the tissue sections for single-cell sequencing is 20-50 μm; the thickness of the tissue sections for Spatial-seq 2.0 spatial transcriptome sequencing is 10-20 μm; and the thickness of the tissue sections for mass spectrometry imaging is 10-20 μm.

[0026] This invention performs Spatial-seq 2.0 spatial transcriptome sequencing on tissue sections. In a specific embodiment, the Spatial-seq 2.0 spatial transcriptome sequencing is a high-throughput spatial transcriptome sequencing method based on a laser microdissection system and DNA barcode labeling, as detailed in patent application number "202510221432.3". In a specific embodiment, the high-throughput spatial transcriptome sequencing method based on a laser microdissection system and DNA barcode labeling includes the following steps: mixing carboxyl magnetic beads with EDC for activation to obtain activated carboxyl magnetic beads; mixing the activated carboxyl magnetic beads with a first spatial barcode primer for dehydration condensation to obtain a first magnetic bead; and mixing the first magnetic beads with a second and a third spatial barcode primer and Phanta Super-Fidelity DNA. Polymerase was mixed and ligated into DNA spatial barcodes for PCR reaction in well plates. Animal tissue was frozen and sectioned, and the surface of the sections was destaticated using a destatic device to obtain destaticated sections. The destaticated sections were microdissected using a laser capture microdissection instrument, and single cells were collected using well plates. The DNA spatial barcodes and cell lysis buffer were added to the well plates containing single cells, and the cells in the well plates were lysed using the DNA spatial barcodes to capture mRNA. The raw materials used to prepare the cell lysis buffer included RNase inhibitors. The captured mRNA was reverse transcribed and washed to obtain washed magnetic beads. The washed magnetic beads were resuspended in a Pre-Amp PCR system and subjected to the first PCR amplification. The beads were placed on a magnetic rack, and the supernatant was collected to obtain the first PCR product. The first PCR product was purified using VAHTS DNA Clean Beads. The purified product was mixed with 2× Kapa HiFi HotStartReadymix and TSO-PCR primers for the second PCR amplification to obtain the second PCR amplification product. The second PCR amplification product was obtained by re-purifying the first PCR product using 0.7× VAHTS DNA Clean Beads. Beads purified the second PCR amplification product to obtain cDNA; a cDNA library was constructed using the TruePrep Flexible DNA Library Prep Kit for Illumina; paired-end sequencing was performed using the Salus Pro gene sequencer; the sequencing results were processed and spatial transcriptome data were analyzed.

[0027] This invention involves mixing carboxyl magnetic beads with EDC (electrode disposable) and performing an activation reaction to obtain activated carboxyl magnetic beads. Before mixing the carboxyl magnetic beads with EDC, they are washed with MES buffer. In this invention, the washing can be performed 1-2 times. In a specific embodiment, 0.1M MES buffer is used for washing. In a specific embodiment, the activation reaction is carried out in MES buffer. In a specific embodiment, the activation reaction is performed at a temperature of 15-30°C for 20 minutes.

[0028] Activated carboxyl magnetic beads were mixed with first spatial barcode primers and subjected to a dehydration condensation reaction to obtain the first magnetic beads. In a specific embodiment, the first spatial barcode primers comprised A1-A4, with nucleotide sequences as shown in SEQ ID NO. 1-4, respectively. In a specific embodiment, 400 μL of activated carboxyl magnetic beads were mixed with 80 μL of 0.2 M MES buffer and 80 μL of the first spatial barcode primers before the dehydration condensation reaction was performed. In a specific embodiment, the concentration of the first spatial barcode primers was 400 μM. In a specific embodiment, the rotation speed of the dehydration condensation reaction (low-speed rotation) was 10-20 rpm. In this invention, the dehydration condensation reaction time could be 20 min. After low-speed rotation, 32.36 μL (0.45 mg) of EDC solution was added, and the low-speed rotation was repeated, including one 20 min and one 80 min low-speed rotation, at a temperature of 15-30°C. Repeated addition of EDC can improve the efficiency of cross-linking, avoid premature deactivation of reaction intermediates, and ensure the quality and yield of the final product. After rotation, the product is placed on a magnetic rack, the supernatant is aspirated, washed, and resuspended to obtain the first magnetic beads. In a specific embodiment, the washing and resuspension are sequentially performed with 100 μL of 0.1M PBS containing 0.02% Tween-20, enzyme-free water, and TE buffer at pH 8.0.

[0029] After obtaining the first magnetic bead, the present invention mixes the first magnetic bead with the second and third spatial barcoding primers and Phanta Super-Fidelity DNA Polymerase, and performs PCR ligation to obtain DNA spatial barcodes. In a specific embodiment, the second spatial barcoding primers include B1-B12, with nucleotide sequences as shown in SEQ ID NO. 5-16, respectively; the third spatial barcoding primers include C1-C8, with nucleotide sequences as shown in SEQ ID NO. 17-24, respectively. In a specific embodiment, the well plate includes a 384-well plate; each well of the PCR reaction system includes 13.5 μL of Phanta Super-Fidelity DNA Polymerase, 4 μL of the first magnetic bead, 1 μL of the second spatial barcoding primer (50 μM), and 1.5 μL of the third spatial barcoding primer (50 μM). In a specific embodiment, the concentrations of the second and third spatial barcoding primers are both 50 μM. In a specific embodiment, the PCR reaction procedure is as follows: 94℃ for 5 min; 95℃ for 15 s, 48.8℃ for 4 min, 72℃ for 4 min, 5 cycles; 94℃ for 5 min, 48.8℃ for 20 min, 72℃ for 20 min. After the PCR reaction, the present invention also includes post-processing to ensure the accuracy and usability of the barcode. In a specific embodiment, the post-processing includes: first, placing the well plate for the PCR reaction on a magnetic rack, operating on ice, aspirating the supernatant, and washing with enzyme-free water to remove residual PCR reaction mixture. Then, incubation with ExoI exonuclease Mix to digest the unamplified single-stranded DNA. To ensure the reaction proceeds fully, the magnetic beads need to be suspended periodically, approximately every 6 min, to promote sufficient contact between the enzyme and DNA. Afterward, washing and resuspending are performed, including continuous resuspending of the magnetic beads with 10 μL of TE-SDS, TE-TW, and 20 μL of enzyme-free water to thoroughly remove residual substances that may affect subsequent experiments. Then, the plate was sealed with a sealing film and placed in a 95 °C metal bath for 6 min. The plate was then quickly removed and placed on a magnetic rack, and the supernatant was aspirated. 20 μL of enzyme-free water was added to each well, and this step was repeated twice. Finally, 15 μL of TE-TW resuspended magnetic beads were added to each well and stored at 4 °C to obtain the Beads preservation plate. These meticulous post-processing steps not only ensured the quality of the DNA barcodes but also provided a stable and reliable foundation for subsequent applications. Traditional methods involve first ligating the second-space barcode primers to the magnetic beads and then ligating them to the third-space barcode primers. This process is cumbersome, time-consuming, and inefficient, especially under the condition of pairing two different barcode primers.This invention employs a one-step method, adding the second and third spatial barcode primers together to the PCR reaction system. Leveraging the highly efficient amplification capability of the Phanta enzyme, both ligation and amplification steps are completed simultaneously within the same reaction system. This improves efficiency (synchronized primer amplification and ligation reduce experimental steps and time, increasing overall reaction efficiency), simplifies operation (reducing multiple reaction steps, reagent addition, and washing processes, thus lowering operational complexity and the probability of errors), and enhances reaction consistency (simultaneous processing within the same reaction system ensures that the second and third barcode primers react under identical conditions, reducing variability caused by differences in different reaction systems). Specifically, by selecting Phanta Super-Fidelity DNA Polymerase and performing a one-step PCR reaction in a novel reaction system, this invention improves the accuracy of barcode ligation and the reliability of amplification; significantly simplifies experimental steps, reducing time and reagent consumption; ensures high fidelity, reduces error introduction, and guarantees barcode accuracy. Therefore, compared to the traditional two-step reaction, this invention's new system setup generates target DNA spatial barcodes more efficiently and accurately, providing a more stable foundation for subsequent experiments.

[0030] This invention, after synthesizing the DNA spatial barcode, also includes quality control of the DNA spatial barcode. In a specific embodiment, the storage plate containing the DNA spatial barcode is removed, placed on a magnetic rack, the supernatant is discarded, enzyme-free water is added, and the magnetic beads are diluted with enzyme-free water to obtain a diluted magnetic bead liquid. Then, the diluted magnetic bead liquid is mixed with qPCR reaction solution and universal and tail-specific primers for qPCR reaction. After the reaction, the mixture is resuspended using TE-EW. In this invention, each 20 μL of the qPCR reaction system includes 10 μL of 2×SYBR Green premix, 1 μL of DNA spatial barcode, 1 μL each of universal and tail-specific primers, and 7 μL of enzyme-free water. In this invention, the qPCR reaction program is: 95℃ for 1 min; 95℃ for 5 s, 55℃ for 15 s, for 35 cycles. The core principle of quality control in this invention is to detect whether the specific sequence of the DNA spatial barcode has been successfully ligated by qPCR, and to ensure that the primers are correctly paired and amplified in the target region. qPCR (quantitative polymerase chain reaction) can monitor the accumulation of products during the PCR amplification process in real time and determine the success of the amplification based on the Ct value (Cycle threshold).

[0031] Tissue was frozen sectioned, and the surface of the sections was treated with an antistatic device to obtain antistatic sections. The antistatic sections were then micro-dissected using a laser capture microdissection instrument, and single cells were collected using a well plate. Animal tissues were also rapidly frozen and stored at -80°C before frozen sectioning. After embedding the frozen animal tissues, they were frozen sectioned. In a specific embodiment, the thickness of the frozen sections was 10 μm. In a specific embodiment, the antistatic treatment time was 3 minutes. In a specific embodiment, the antistatic device included a benchtop ion blower LA-211. Static electricity on LCM instruments and sections is very severe, affecting the capture rate. This invention, by adding an antistatic device to blow out an airflow of positive and negative charges, neutralizes the charges, thereby increasing the capture rate of Spots from 50% to 95%. In a specific embodiment, the micro-cutting conditions are as follows: the cutting area is automatically identified using a laser capture micro-cutting instrument, with parameters set to 10× objective lens, Final Pulse mode, power = 17, aperture = 1, speed = 20, bridging size = 5, and final pulse = 16. After cutting, the material is collected using a well plate and immediately stored at -80°C.

[0032] DNA spatial barcodes and cell lysis buffer are added to well plates containing single cells. The DNA spatial barcodes are used to lyse the cells in the well plates and capture mRNA. The raw materials used to prepare the cell lysis buffer include an RNase inhibitor. In a specific embodiment, before adding the DNA spatial barcodes, the process includes aspirating the supernatant (TE-TW), washing and resuspending with enzyme-free water, all performed on ice. In a specific embodiment, the cell lysis buffer includes Tris-HCl, LiCl, SDS, EDTA, DTT solution, RNase inhibitor, and nuclease-free water; each 1200 μL of cell lysis buffer contains 120 μL of pH 7.5 Tris-HCl, 80 μL of LiCl, 120 μL of 10% SDS solution, 16 μL of EDTA, 12 μL of 500 mM DTT solution, 12 μL of 40 U / μL RNase inhibitor, and the remainder is nuclease-free water. The cell lysis buffer described in this invention can inhibit the effect of exogenous RNases on RNA degradation and improve RNA quality. In this invention, the volume ratio of DNA spatial barcodes to cell lysis buffer is 1:19. Preferably, the cell lysis buffer and DNA spatial barcodes are mixed first, and then the cell lysis buffer containing the DNA spatial barcodes is added to the well plate containing collected single cells. In a specific embodiment, the lysis conditions include: lysis at room temperature for 5 min, incubation on ice for 12 min, and resuspending the magnetic beads every 2 min during this period to promote effective mRNA capture. In this invention, room temperature refers to 15-30°C. After lysis, the magnetic beads are washed. In a specific embodiment, 6 × SSC is used for washing the magnetic beads. Afterwards, centrifugation is performed, and the supernatant is discarded.

[0033] The captured mRNA was reverse transcribed, washed, and the resulting magnetic beads were cleaned. The cleaned magnetic beads were resuspended in a Pre-Amp PCR system for the first PCR amplification. The beads were placed on a magnetic rack, and the supernatant was collected to obtain the first PCR product. In a specific embodiment, it is preferred to resuspend the magnetic beads in 50 mM Tris pH 8.0 solution. After discarding the supernatant, 20 μL of RT Mix (prepared on ice) was immediately added for reverse transcription of the mRNA. In this invention, the reverse transcription conditions were 42°C for 90 min. In a specific embodiment, resuscitation was performed every 15 min. After the reaction, the beads were centrifuged, placed on a magnetic rack, and the supernatant was discarded. In a specific embodiment, the washing could be performed using TE-SDS, TE-TW, and 10 mM Tris-HCl (pH = 8.0), respectively. After washing, the magnetic beads were resuspended in an exonuclease system and incubated at 37°C for 60 min, with resuscitation every 10 min. After the reaction, the beads were placed on a magnetic rack, the supernatant was discarded, and the beads were then washed with TE-SDS, TE-TW, and 10 mM Tris-HCl (pH = 8.0). Subsequently, the beads were resuspended in the Pre-AmpPCR system for the first PCR amplification. In this invention, the reaction program for the first PCR amplification was: 98℃ for 3 min; 98℃ for 20 s, 65℃ for 45 s, 72℃ for 6 min, 6 cycles; 72℃ for 10 min.

[0034] The first PCR product was purified using VAHTS DNA Clean Beads. The purified product was then mixed with 2× KapaHiFi HotStart Readymix and TSO-PCR primers for a second PCR amplification to obtain the second PCR amplification product. In a specific embodiment, 0.8× VAHTS DNA Clean Beads were used to purify the PCR product. Preferably, the purification process included the following steps: mixing 0.8× VAHTS DNA Clean Beads with the first PCR product, incubating at room temperature for 15 min, discarding the supernatant, washing twice with 80% ethanol aqueous solution, incubating at room temperature for 30 s, removing the supernatant, eluting cDNA with nuclease-free water, and incubating at room temperature for 10 min to obtain the purified product. The purified product was then mixed with 2× Kapa HiFi HotStart Readymix and TSO-PCR primers for a second PCR amplification. In this invention, the reaction program for the second PCR amplification was: 98℃ for 3 min; 98℃ for 20 s, 72℃ for 6 min, 10 cycles; 72℃ for 10 min. Preferably, the purification conditions were the same as above. In a specific embodiment, the TSO-PCR primer includes TSO LNA and TSO-PCR; the nucleotide sequence of the TSO LNA is shown in SEQ ID NO.27, and the TSO LNA has two trans-modified guanine / rG / rG and one LNA (locked nucleus) modified guanine / iXNA_G at its end to enhance its stability in binding to the template and reduce non-specific binding; the nucleotide sequence of the TSO-PCR is shown in SEQ ID NO.28.

[0035] After obtaining the second PCR amplification product, the present invention further purifies the second PCR amplification product using 0.7 × VAHTS DNA Clean Beads to obtain cDNA. In a specific embodiment, the length of the cDNA is greater than 650 bp.

[0036] This invention uses the TruePrep Flexible DNA Library Prep Kit for Illumina to construct cDNA libraries. PE150 paired-end sequencing is performed using a Salus Pro gene sequencer.

[0037] This invention processes sequencing results. In a specific embodiment, processing the sequencing results includes: splitting the sequencing data into independent files using a custom Python script based on the different spatial barcode sequences, and processing them according to the Drop-seq core analysis workflow (specifically, using Drop-seq_tools v2.5.1 for quality control and preliminary processing according to Drop-seq Core Computational Protocol v2.0.0), followed by gene alignment using STAR v2.7.8a; the quality control and preliminary processing are performed using the FastQC tool.

[0038] This invention performs single-cell sequencing on tissue sections. In a specific embodiment, the single-cell sequencing includes single-cell transcriptome sequencing; the platform for single-cell transcriptome sequencing includes the 10x Genomics-Chromium platform, the BDRhapsody platform, or the BGI Genomics-DNBelab C4 platform. In a specific embodiment, the method for single-cell transcriptome sequencing includes: homogenizing the tissue sections, enzymatically digesting them to obtain single cells, constructing a transcriptome sequencing library, and performing sequencing on the single-cell transcriptome sequencing platform. In a specific embodiment, mechanical homogenization is performed under sterile conditions, followed by enzymatic digestion using a digestion solution containing trypsin to isolate single cells. When using the 10x Genomics-Chromium platform for single-cell sequencing, this invention utilizes the Chromium single-cell 3' library preparation kit of the 10x Genomics platform, following the manufacturer's instructions, to capture single cells and construct a transcriptome sequencing library using a microfluidic chip system. During the microfluidic chip sorting and cDNA synthesis process, the mRNA of each cell is reverse transcribed into cDNA, and cell-specific barcodes and UMIs (Unique Molecular Identifiers) are added via transposase reaction during the template conversion stage. After the transcriptome sequencing library passes quality testing, it is placed on the Illumina NovaSeq 6000 sequencing platform for paired-end sequencing.

[0039] After single-cell sequencing, the Cell Ranger software package provided by 10x Genomics was used for data quality control, read alignment, UMI counting, and cell identification. The latest genome version corresponding to the tissue was selected as the reference genome for alignment. Cells expressing more than 200 but less than 7500 unique genes and whose read mapping was less than 10% of mitochondrial genes in tissue sections were screened using the Seurat package 5.0.1 of R Studio 4.2.1. Then, the doubletFinder function of DoubletFinder 2.0.4 was used to remove bimodal peaks. Subsequent analysis was performed using the merge function, including principal component analysis using the RunPCA function and removing inter-group effects using harmony 1.2.0. In a specific embodiment, the different groups may include a blank control group, a model group, and a treatment group. The top 20 principal components were selected for unsupervised clustering and t-distributed stochastic neighbor embedding (t-SNE) dimensionality reduction. Cells were clustered using the FindClusters function, and each cluster was labeled and annotated. Downstream analysis was then performed. The CellChat package 1.6.1 was used to analyze the communication signaling pathways between different cell types in single-cell sequencing data, including signal transmission, reception, and potential biological effects. GO and KEGG pathway enrichment analysis of differentially expressed genes was performed using the Metascape online platform. After the enrichment analysis, the enrichment results provided by Metascape were exported and visualized using the R package ggplot2 to obtain single-cell transcriptome data information.

[0040] This invention relates to mass spectrometry imaging of tissue sections. In a specific embodiment, the mass spectrometry imaging includes MALDI-MSI mass spectrometry imaging. This invention utilizes MALDI-MSI mass spectrometry imaging to explore spatial metabolic changes in tissues. In a specific embodiment, the MALDI-MSI mass spectrometry imaging method includes the following steps: drying the tissue sections, capturing optical images, applying a matrix to the sections, and performing mass spectrometry imaging analysis using an AP-SMALDI 10 instrument system. In a specific embodiment, the matrix includes a DHB matrix solution. In a specific embodiment, the application conditions include a DHB matrix solution flow rate of 20 μL / min and a nitrogen flow rate of 5 L / min applied to the sections, with a nozzle movement speed of 3 mm / s, to ensure that the entire brain tissue section is uniformly covered by the matrix. In this specific embodiment, the mass spectrometry parameters were set in Tune software (Thermo Fisher Scientific), including a spray voltage of 4 kV, an S-Lens of 100 eV, and a capillary temperature of 250 °C. A positive ion full scan mode was used, with a mass range of m / z 80-1200 and a mass spectrometry resolution of R = 70000 @ m / z 200. Internal calibration was performed using the instrument's lock-in mass function. In this specific embodiment, the sample stage step size was set to 50 μm, and the laser energy was set to Filter 20% + 20°. In this specific embodiment, the acquired data was processed using Mirion software and the open-source MSiReader software for imaging data processing.

[0041] In a specific embodiment, the single-cell spatial multi-omics affine registration and joint data analysis includes: Construct a spatial reference coordinate system, map single-cell to spatial transcriptome data, and perform affine registration of multimodal data; The construction of the spatial reference coordinate system includes: using QuickNII software to extract histological features from HE-stained sections, establishing an initial spatial coordinate system based on ventricular structure, hippocampal dentate gyrus morphology and cortical layering features, and registering it to the Allen Brain Atlas standard reference map to complete millimeter-level spatial positioning; The single-cell-spatial transcriptome data mapping includes: joint analysis of single-cell transcriptome sequencing data and Spatial-seq 2.0 data; multimodal data fusion is performed using the Tangram algorithm: using single-cell sequencing data as input, cell types are identified based on the Leiden clustering algorithm, and a cell type-specific gene expression feature matrix is ​​constructed. The multimodal data affine registration includes: morphological erosion processing (3×3 structural nuclei) of HE-stained sections, Spatial-seq 2.0 spatial transcriptome raw images and mass spectrometry imaging sections, extraction of tissue contour feature point clouds, and feature point matching. After registration, multi-omics data of five independent regions, namely the hippocampus, striatum, motor cortex, sensory cortex and olfactory cortex, can be analyzed independently.

[0042] The present invention also provides a disease treatment mechanism analysis system based on the combined use of single-cell sequencing, spatial transcriptomics and mass spectrometry imaging, including a single-cell sequencing unit for obtaining single-cell sequencing data; Spatial-seq 2.0 spatial transcriptome sequencing unit, used to obtain Spatial-seq 2.0 spatial transcriptome sequencing data; Mass spectrometry imaging unit, used to acquire space metabolomics data; The unit includes a single-cell spatial multi-omics affine registration and data analysis unit, used to perform affine registration and analysis on data acquired by the single-cell sequencing unit, the Spatial-seq 2.0 spatial transcriptome sequencing unit, and the mass spectrometry imaging unit.

[0043] This invention also provides applications of the disease treatment mechanism analysis method or system described in the above-mentioned technical solutions, such as in any one of the studies described in ① to ⑤: ① Disease mechanism analysis research; ② Drug development research; ③ Biomarker discovery research; ④ Developmental and regenerative medicine research; ⑤ Tumor immunotherapy development research; the applications are for non-clinical treatment purposes. In this invention, the disease mechanism analysis includes: analyzing disease-related cellular heterogeneity, molecular interaction networks, and metabolic dynamics by integrating single-cell transcriptome, spatial transcriptome, and spatial metabolome data. In this invention, the drug development includes: screening disease treatment targets, evaluating drug metabolic distribution and mechanisms of action. In this invention, the biomarker discovery includes: identifying multi-omics biomarkers related to disease subtypes or prognosis. In this invention, the developmental and regenerative medicine research includes: analyzing stem cell differentiation trajectories and spatiotemporal molecular changes during tissue regeneration. In this invention, the tumor immunotherapy development includes: studying the interaction mechanisms between immune cells and metabolites in the tumor microenvironment. The applications of this invention are for non-clinical treatment purposes, covering basic research, disease model construction, and medical technology development.

[0044] To further illustrate the present invention, the following detailed description, in conjunction with embodiments, of the disease treatment mechanism analysis method, system, and application based on the combined use of single-cell sequencing, spatial transcriptomics, and mass spectrometry imaging provided by the present invention, should not be construed as limiting the scope of protection of the present invention.

[0045] Experimental materials and equipment: 20 μm carboxyl magnetic beads, Suzhou Beaver Biomedical Engineering Co., Ltd., item number 40200; Phanta Super-Fidelity DNA Polymerase: Nanjing Novizan Biotechnology Co., Ltd., Product No. P501; PrimeScript™ II Reverse Transcriptase: Takara Corporation, Japan, Product No. 2690A; Recombinant RNasin® Ribonuclease Inhibitor: Promega Inc., USA, Product No. N2515; SUPERase In RNase inhibitor: ABI Pharmaceuticals, USA, catalog number AM2694; KAPA HiFi High-Fidelity Hot-Start DNA Polymerase Premix: Kapa Biosystems, USA, Catalog No. KK2602; Rapid Real-Time PCR Kit (SYBR Green): Biosharp, China, Catalog No. BL705A; Qubit® dsDNA HS Assay Kits: Thermo Scientific, USA, Product No. Q32851; Leica Cryo-Gel cryosection embedding medium: Leica GmbH, Germany, product number 14020108926; VAHTS DNA Clean Beads: Nanjing Novizan Biotechnology Co., Ltd., Product No. N411-01; TruePrep Flexible DNA Library Prep Kit for Illumina: Nanjing Novizan Biotechnology Co., Ltd., Catalog No. TD504; The PCR primer names and sequences were synthesized by Shanghai Sangon Biotech Co., Ltd. A1: TTTAGGGATAACAGGGTAATAAGCAGTGGTATCAACGCAGAGTACGTTTTAGGCGACTCACTACAGGG (SEQ ID NO. 1).

[0046] A2: TTTAGGGATAACAGGGTAATAAGCAGTGGTATCAACGCAGAGTACGTATTCCACGACTCACTACAGGG (SEQ ID NO. 2).

[0047] A3: TTTAGGGATAACAGGGTAATAAGCAGTGGTATCAACGCAGAGTACGTGCTCAACGACTCACTACAGGG(SEQ ID NO.3)。

[0048] A4: TTTAGGGATAACAGGGTAATAAGCAGTGGTATCAACGCAGAGTACGTCATCCCCGACTCACTACAGGG(SEQ ID NO.4)。

[0049] B1:CGATCGTGTCACCGACCTAAACCCTGTAGTGAGTCG(SEQ ID NO.5)。

[0050] B2:CGATCGTGTCACCGATGGAATCCCTGTAGTGAGTCG(SEQ ID NO.6)。

[0051] B3:CGATCGTGTCACCGATTGAGCCCCTGTAGTGAGTCG(SEQ ID NO.7)。

[0052] B4:CGATCGTGTCACCGAGGGATGCCCTGTAGTGAGTCG(SEQ ID NO.8)。

[0053] B5:CGATCGTGTCACCGAGTCCAACCCTGTAGTGAGTCG(SEQ ID NO.9)。

[0054] B6:CGATCGTGTCACCGAACACAGCCCTGTAGTGAGTCG(SEQ ID NO.10)。

[0055] B7:CGATCGTGTCACCGAATGTCCCCCTGTAGTGAGTCG(SEQ ID NO.11)。

[0056] B8:CGATCGTGTCACCGAACTTTGCCCTGTAGTGAGTCG(SEQ ID NO.12)。

[0057] B9:CGATCGTGTCACCGACCGCTTCCCTGTAGTGAGTCG(SEQ ID NO.13)。

[0058] B10:CGATCGTGTCACCGATTTATTCCCTGTAGTGAGTCG(SEQ ID NO.14)。

[0059] B11:CGATCGTGTCACCGACTCCTCCCCTGTAGTGAGTCG(SEQ ID NO.15)。

[0060] B12:CGATCGTGTCACCGATGTACCCCCTGTAGTGAGTCG(SEQ ID NO.16)。

[0061] C1: AAAAAAAAAAAAAAAAAAAAAAAAAAAAAANNNNNNCCTAAACGATCGTGTCACCGA(SEQ IDNO.17)。

[0062] C2: AAAAAAAAAAAAAAAAAAAAAAAAAAAAANNNNNNTGGAATCGATCGTGTCACCGA(SEQ IDNO.18)。

[0063] C3: AAAAAAAAAAAAAAAAAAAAAAAAAAAAAANNNNNNTTGAGCCGATCGTGTCACCGA(SEQ IDNO.19)。

[0064] C4: AAAAAAAAAAAAAAAAAAAAAAAAAAAAAANNNNNNGGGATGCGATCGTGTCACCGA(SEQ IDNO.20)。

[0065] C5: AAAAAAAAAAAAAAAAAAAAAAAAAAAAAANNNNNNGTCCAACGATCGTGTCACCGA(SEQ IDNO.21)。

[0066] C6: AAAAAAAAAAAAAAAAAAAAAAAAAAAAAANNNNNNACACAGCGATCGTGTCACCGA(SEQ IDNO.22)。

[0067] C7: AAAAAAAAAAAAAAAAAAAAAAAAAAANNNNNNATGTCCCGATCGTGTCACCGA (SEQ ID NO. 23).

[0068] C8: AAAAAAAAAAAAAAAAAAAAAAAAAAANNNNNNACTTTGCGATCGTGTCACCGA (SEQ ID NO. 24).

[0069] Universal: GGTATCAACGCAGAGTA (SEQ ID NO. 25).

[0070] Tail: GTCGGGAAAAAAAAAAAAAAAAAG (SEQ ID NO. 26).

[0071] TSO LNA: AAGCGTGGTATCAACGCAGAGTGAAT / rG / rG / iXNA_G (The nucleotide sequence before modification is shown in SEQ ID NO.27).

[0072] TSO-PCR: AAGCAGTGGTATCAACGCAGAGT (SEQ ID NO. 28).

[0073] Cell lysis buffer: 120 μL of Tris-HCl (pH 7.5), 80 μL of LiCl, 120 μL of 10% SDS solution, 16 μL of EDTA, 12 μL of 500 mM DTT solution and 12 μL of RNase inhibitor (40 U / μL) were added to sufficient nuclease-free water to prepare a specific cell lysis buffer; RT mix: 16.16 μL nuclease-free water, 8 μL 5× RT buffer, 8 μL betaine solution, 4 μL dNTP (10mM), 0.2 μL DTT, 0.24 μL MgCl2, 0.4 μL TSO primer, 2 μL RimeScript II reverse transcriptase (200U / μL), 1 μL RNase inhibitor, total 40 μL.

[0074] Gradient PCR amplification instrument: Thermo Fisher Scientific, USA, model Applied biosystem ProFlexBase; Real-time quantitative PCR system: Bio-Rad Laboratories, USA, model CFX96 Touch; Laser capture microdissection system: Leica GmbH, Germany, including an upright microscope (model DM68), a laser cutter (model LMD6), and a single-cell capture and recovery device (model LMT350). Agilent Nucleic Acid Microfluidic Electrophoresis Analysis System: Agilent Technologies, USA, Model 4200 TapeStation; Gene sequencer: Salus Medical Co., Ltd. (China), model Salus Pro; Static electricity removal equipment: LA-211 desktop ion fan, LAOGE (Shanghai) Co., Ltd.

[0075] Panax notoginsenoside R1 standard: purity ≥98%, specification 20 mg / vial, provided by Chengdu Mansite Biotechnology Co., Ltd., product number A0273; Animal experiments with Panax notoginseng saponin R1 solution: Accurately weigh Panax notoginseng saponin R1 standard and dissolve it in physiological saline. Prepare solutions with concentrations of 1.5 mg / mL, 3 mg / mL, and 6 mg / mL in the dark. These solutions are used as the administration solutions for the low-dose (15 mg / kg), medium-dose (30 mg / kg), and high-dose (60 mg / kg) groups of Panax notoginseng saponin R1, respectively. The administration volume for mice is 0.1 mL / 10 g / d. The solution should be prepared and used immediately.

[0076] C57BL / 6 mice: SPF grade, male, 8-10 weeks old, 22-25 g, purchased from Shanghai Slack Laboratory Animal Co., Ltd. (Animal Quality Certificate No.: 20220004003234). The mice were housed in the SPF grade animal room of the Experimental Animal Center of Zhejiang University, placed in a constant temperature room (25±1 ℃), and had free access to food and water in light and dark environments for 12 h each.

[0077] Edaravone: Nanjing Sinopharm Pharmaceutical Co., Ltd., specification 5 mL: 10 mg, batch number 80-151203; Physiological saline: Shandong Qidu Pharmaceutical Co., Ltd., product number 1250850; Silicone-coated sutures: Guangzhou Jialing Biotechnology Co., Ltd., L1800 type, item number 1800AAA; 4% Paraformaldehyde: Beijing Solarbio Technology Co., Ltd., Product No. P1110; Chromium Next GEM Single Cell 3ʹ Reagent Kits v3.1: 10x Genomics, USA, catalog number PN-1000121; 2,5-Dihydroxybenzoic acid (DHB): Shanghai Maclean Biochemical Technology Co., Ltd., product number D806755; DHB matrix solution: Weigh a certain mass of DHB powder and dissolve it in a certain volume of 50% acetonitrile aqueous solution (containing 0.1% TFA) to obtain a 30 mg / mL DHB matrix solution, which is then stored in a refrigerator at 4 ℃ for later use.

[0078] Single-cell library preparation system: 10x Genomics, Inc., USA; Model: 10x Genomics Chromium Controller. Novaseq 6000 sequencer: Illumina Corporation, USA, model NovaSeq 6000; Matrix sprayer: TransMIT GmbH, Germany, model SMALDIPrep Matrix Sprayer; Mass spectrometry imaging system: TransMIT, Germany, model AP-SMALDI 10 (equipped with a Thermo Scientific Q Exactive™ mass spectrometer).

[0079] Example 1 Spatial-seq 2.0 spatial transcriptome sequencing method Step 1: Synthesis of DNA spatial barcodes The first step involves synthesizing a positioning magnetic bead to connect to the first segment of the spatial barcode. This is done by first taking four 1.5 mL centrifuge tubes and adding approximately 1.67 × 10⁻⁶ mg of the solution to each tube. 6232 μL of carboxyl magnetic beads were collected. The supernatant was then aspirated, and the tubes were washed twice with 0.1 M MES buffer. After washing, approximately 358 μL of 0.1 M MES buffer and 2.06 mg (34.4 μL) of EDC were added to each tube, bringing the final volume to approximately 424 μL. After mixing, 400 μL of the mixture was removed from each tube. Then, 80 μL of 0.2 M MES and 80 μL of the corresponding first-space barcode primers (primers A1–A4) were added to each tube, and the mixture was thoroughly mixed. The tubes were fixed on a centrifuge and rotated at low speed for 20 min at room temperature. Then, 32.36 μL (0.45 mg) of EDC solution was added to each tube, and the low-speed rotation was repeated, including one 20-min and one 80-min rotation. After rotation, the centrifuge tube was placed on a magnetic rack, the supernatant was aspirated, and then the magnetic beads were resuspended and washed sequentially with 100 μL of 0.1M PBS containing 0.02% Tween-20, enzyme-free water, and TE buffer at pH 8.0. Finally, the magnetic beads were resuspended in 280 μL of enzyme-free water to obtain the first set of magnetic beads.

[0080] The next step is to synthesize positioning magnetic beads to connect the second and third spatial barcode segments. First, you need to prepare the primers for the second spatial barcode (primer B, B1~B12), the primers for the third spatial barcode (primer C, C1~C8), and Phanta Super-Fidelity DNA Polymerase. Take them out of 4 ℃ in advance, mix them, and then centrifuge them quickly to prepare a concentration of 50 μM.

[0081] During the PCR ligation stage, each well of the 384-well PCR plate contained 13.5 μL of Phanta Super-Fidelity DNA Polymerase, 4 μL of the first magnetic bead, 1 μL of the second spatial barcode primer (50 μM), and 1.5 μL of the third spatial barcode primer (50 μM).

[0082] After mixing and rapidly centrifuging the 384-well plate, PCR amplification was performed using the following program: suspend beads at 94 ℃ for 5 min; suspend beads at 95 ℃ for 15 s, 48.8 ℃ for 4 min, 72 ℃ for 4 min, for a total of 5 cycles; suspend beads at 94 ℃ for 5 min; 48.8 ℃ for 20 min; 72 ℃ for 20 min; hold at 4 ℃.

[0083] After completing the PCR reaction for DNA spatial barcoding, the next step is to post-process the 384-well plate to ensure the accuracy and usability of the barcodes. First, the 384-well plate is placed on a magnetic rack and operated on ice to aspirate the supernatant. Then, each well is washed once with enzyme-free water to remove residual PCR reaction mixture. Next, 10 μL of ExoI exonuclease Mix is ​​added to each well and incubated at 37 °C for 15 min to digest unamplified single-stranded DNA. To ensure the reaction proceeds fully, the magnetic beads are suspended periodically, approximately every 6 min, to promote adequate contact between the enzyme and DNA. Following this, a series of washing and resuspending steps are performed, including continuous resuspending of the magnetic beads with 10 μL of TE-SDS, TE-TW, and 20 μL of enzyme-free water to thoroughly remove any residual substances that may affect subsequent experiments. Next, the sealing film was applied, and the 384-well plate was placed in a metal bath at 95 °C for approximately 6 min. The plate was then quickly removed and placed on a magnetic rack, and the supernatant was aspirated. 20 μL of enzyme-free water was added to each well, and this step was repeated twice. Finally, 15 μL of TE-TW resuspended magnetic beads were added to each well, labeled, and stored at 4 °C to obtain the Beads preservation plate. These meticulous post-processing steps not only ensured the quality of the DNA barcodes but also provided a stable and reliable foundation for subsequent applications.

[0084] Step 2: Micro-cutting and Barcode Capture Marking Frozen sections of animal tissue were gently blown onto the surface for 3 minutes using an antistatic device (LA-211 benchtop ionizer), and then placed on a laser microscopy (LCM) instrument for high-resolution microscopic scanning. Each section can be divided into multiple spot regions according to user needs, with the radius of each spot approximately customizable, down to a minimum of about 10 μm, achieving single-cell level. The LCM instrument automatically identified the cutting regions, with parameters set as follows: objective lens 10 ×, Final Pulse mode, power = 17, aperture = 1, speed = 20, bridging size = 5, final pulse = 16. Enzyme-free plates were used for collection, and the collected plates were immediately stored at -80℃.

[0085] The magnetic bead storage plate synthesized in step 1 and stored at 4 °C was rapidly centrifuged to remove residual liquid. The plate was then placed on a magnetic rack for aspiration of the supernatant (TE-TW). Next, the magnetic beads were washed 1 to 2 times with 20 μL of enzyme-free water each time, followed by rapid centrifugation and resuspending to remove the supernatant. After washing, the magnetic beads in each well were resuspended with 20 μL of enzyme-free aqueous solution. The entire process was performed on ice.

[0086] Next, a 384-well plate was used as the container plate, with 19 μL of Lysis Buffer added to each well, and kept on ice throughout the process. 1 μL of magnetic bead suspension was transferred from each well of the original magnetic bead storage plate to the corresponding well of the container plate, also performed on ice, to obtain cell lysis buffer containing magnetic beads. The collection plate was removed from storage at -80 °C and briefly centrifuged to remove any possible ice crystals. 20 μL of cell lysis buffer containing magnetic beads was added to each well of the collection plate, and lysis was performed at room temperature for 5 min, followed by incubation on ice for 12 min, resuspending the magnetic beads every 2 min to promote efficient mRNA capture. After incubation, all magnetic beads were collected into 1.5 mL centrifuge tubes.

[0087] Next, the magnetic beads were washed twice with 500 μL of 6 × SSC, and the supernatant was discarded after a few seconds of rapid centrifugation on the last wash. The magnetic beads were then resuspended in 300 μL of 50 mM Tris pH 8.0 solution, and the supernatant was discarded. Immediately afterward, 20 μL of RT Mix (prepared on ice) was added to perform reverse transcription of mRNA.

[0088] Step 3: Reverse transcription and cDNA amplification The reverse transcription reaction was performed at 42°C in a metal bath for 90 min, with resuscitation every 15 min. After the reaction, the beads were rapidly centrifuged for a few seconds, and the centrifuge tubes were placed on a magnetic rack. The supernatant was carefully discarded. The magnetic beads were washed with 200 μL of TE-SDS, TE-TW, and 10 mM Tris-HCl (pH = 8.0). The beads were resuspended in 200 μL of exonuclease and incubated at 37°C for 60 min, with resuscitation every 10 min. After the reaction, the beads were placed on a magnetic rack, and the supernatant was discarded. The beads were washed with 200 μL of TE-SDS, TE-TW, and 10 mM Tris-HCl (pH = 8.0).

[0089] Subsequently, PCR amplification was performed after resuspending the magnetic beads in the Pre-Amp PCR system. The PCR program was as follows: 98 ℃ for 3 min; 98 ℃ for 20 s, 65 ℃ for 45 s, 72 ℃ for 6 min, for a total of 6 cycles; 72 ℃ for 10 min; and held at 4 ℃.

[0090] Remove VAHTS DNA Clean Beads 30 minutes before the scheduled time and bring to room temperature, then vortex to mix. Place the PCR product on a magnetic rack and transfer the supernatant to a new centrifuge tube. Purify the PCR product using 0.8× VAHTS DNA Clean Beads as follows: Add VAHTS DNA Clean Beads to the centrifuge tube, mix well, and incubate at room temperature for 15 minutes. Discard the supernatant, wash twice with 200 μL of freshly prepared 80% ethanol, incubate at room temperature for 30 seconds, remove the supernatant, remove the centrifuge tube from the magnetic rack, wash away cDNA with 13 μL of nuclease-free water, incubate at room temperature for 10 minutes, transfer 12 μL of supernatant to a new centrifuge tube, add 12.5 μL of 2×Kapa HiFi HotStart Readymix and 0.5 μL of 10 μM TSO-PCR primers. A second PCR amplification was performed using the following reaction program: 98 ℃ for 3 min; 98 ℃ for 20 s, 72 ℃ for 6 min, for a total of 10 cycles; 72 ℃ for 10 min; and then held at 4 ℃.

[0091] The amplification product was purified again using 0.7 × VAHTS DNA Clean Beads, following the same procedure as above. The cDNA concentration was measured using a Qubit 4.0 fluorometer, and the cDNA bands were detected by Agilent 4200 nucleic acid microfluidic electrophoresis.

[0092] Step 4: cDNA library construction and sequencing cDNA libraries were constructed according to the TruePrep Flexible DNA Library Prep Kit for Illumina. Paired-end sequencing was performed using a Salus Pro gene sequencer (model PE150) from Salus Medical Systems, China.

[0093] Step 5: Sequencing Result Processing Spatial-seq high-throughput pooled sequencing generates massive amounts of data. To achieve effective gene alignment, the sequencing data was first separated into individual files using a custom Python script based on the different spatial barcode sequences (Barcode 1-384). Following the Drop-seq Core Computational Protocol v2.0.0, Drop-seq_tools v2.5.1 was used for quality control and preliminary processing to ensure data quality and format met the requirements for subsequent analysis. After this processing, the ultrafast RNA sequence alignment tool STAR v2.7.8a was used, combined with the reference genome GRCm39 (Release-109 version) genome sequence file (FASTA format) and gene annotation file (GTF format) provided by the Ensembl database, for precise sequence alignment. Preliminary quality control of the sequencing data was performed using the FastQC tool, a Java-based tool that rapidly assesses the quality of "fq" format sequencing files to ensure data quality meets the requirements for subsequent analysis. Using a custom Python script, the raw sequencing data "fq" files were meticulously split and organized according to the differences in Barcode 1-384, preparing for further molecular identification.

[0094] Step 6: Spatial-seq 2.0 spatial transcriptome data analysis The generated numerical expression matrix was standardized using the `calcNormFactors` function in the `edgeR` package v3.40.2 to eliminate the influence of differences in sequencing depth and RNA composition on expression level estimation. The likelihood ratio test (LRT) was used to assess the variability in gene expression across different groups. LRT determines the significance of effects by comparing the goodness of fit between the full model including specific effects and the simplified model excluding those effects. P The value was corrected through multiple tests, and the FDR method was used to control the false positive rate. The final threshold was set (…). P Genes with differential expression <0.05 and |log2 fold change|>1 were identified. GO and KEGG pathway enrichment analyses were performed on the differentially expressed genes using the Metascape online platform. After the enrichment analysis, the enrichment results provided by Metascape were exported and visualized using the ggplot2 package in R.

[0095] Example 2 The development of ischemic stroke involves multiple targets and pathways, including oxidative stress, glutamate excitotoxicity, neuroinflammation, autophagy, and apoptosis. However, most current research methods rely on traditional bulk analysis. For example, transcriptomics studies typically involve homogenizing brain tissue samples and extracting total RNA for expression profiling microarrays or bulk RNA-seq analysis. This only provides the average value of the total cell count and fails to reveal functional heterogeneity between different spatial locations and cell types, making it difficult to precisely elucidate the mechanisms of ischemic stroke at the molecular level. Spatial heterogeneity reflects the different regions and degrees to which stroke affects brain tissue, including changes in the affected brain regions, the degree of damage, and differences in recovery potential. This heterogeneity exists not only between the infarct core and the ischemic penumbra but also across different brain regions. Therefore, a deep understanding and study of the spatial and cellular heterogeneity of stroke is crucial for revealing the complexity of the disease, advancing the development of targeted drugs, and guiding clinical treatment decisions. This embodiment selects notoginsenoside R1 (NGR1), an active ingredient of Panax notoginseng, and uses multi-omics technologies such as single-cell sequencing, spatial-seq, and mass spectrometry imaging to comprehensively reveal the mechanism of action of NGR1 in the fight against acute stroke, providing a scientific basis for further interpreting the modern scientific connotation of the efficacy and clinical application of Panax notoginseng.

[0096] A method for studying the mechanism of action of Panax notoginseng saponin R1 (NGR1) against acute stroke based on the disease treatment mechanism analysis method of this invention: Step 1: Establishment of a mouse model of middle cerebral artery occlusion (MCAO). Male C57BL / 6 mice were acclimatized for one week, and a modified suture occlusion method was used to establish a mouse model of permanent middle cerebral artery occlusion (pMCAO). The specific procedure was as follows: Before surgery, mice were anesthetized with 1.0% sodium pentobarbital. The skull of the anesthetized mice was exposed, and a fiber optic probe of a laser Doppler flowmeter was fixed to the surface of the skull (central area: 2 mm posterior to the anterior fontanelle, 6 mm to the right of the midline; peripheral area: 2 mm posterior to the anterior fontanelle, 3 mm to the right of the midline) to monitor changes in cortical blood flow corresponding to the middle cerebral artery in real time. After the fiber optic was fixed, the right middle cerebral artery of the mouse was blocked using the suture occlusion method to create a permanent focal cerebral ischemia model.

[0097] Step 2: Animal grouping and administration. Using a random number table, all mice were divided into the following 6 groups, with 5 mice in each group: (1) Sham group: Mice underwent MCAO surgery but without suture insertion; (2) Model group: Mice were placed in the MCAO model after successful establishment; (3) Low-dose Panax notoginseng saponin R1 (NGR1-L): Mice were injected intraperitoneally with 15 mg / kg Panax notoginseng saponin R1 solution after establishing the MCAO model; (4) Medium-dose Panax notoginseng saponin R1 (NGR1-M): Mice were injected intraperitoneally with 30 mg / kg Panax notoginseng saponin R1 solution after establishing the MCAO model; (5) High-dose Panax notoginseng saponin R1 (NGR1-H): Mice were injected intraperitoneally with 60 mg / kg Panax notoginseng saponin R1 solution after establishing the MCAO model; (6) Edaravone positive control group: Mice were injected intraperitoneally with 4 mg / kg edaravone solution after establishing the MCAO model. The sham group and the model group received an equal volume of physiological saline solution intraperitoneally.

[0098] Step 3: Mouse neurobehavioral scoring. Mouse neurobehavioral scoring was performed 24 hours after ischemia. The Longa neurobehavioral score was based on the reference [Longa EZ, Weinstein PR, Carlson S, et al. Reversible middle cerebral occlusion without craniectomy in rats[J]. Stroke, 1989,20(1):84-91.]. The specific evaluation criteria are as follows: ① No symptoms: 0 points; ② Inability to fully extend the forepaw on the paralyzed side: 1 point; ③ Circling towards the paralyzed side while walking: 2 points; ④ Leaning towards the paralyzed side while walking: 3 points; ⑤ Inability to walk automatically, with loss of consciousness: 4 points.

[0099] Step 4: Animal brain tissue collection. 24 hours after mouse modeling, the mice were quickly euthanized by cervical dislocation. The brain tissue was carefully removed, the surface blood was wiped off, and the tissue was flash-frozen in liquid nitrogen for 1 minute and stored at -80°C. Step 5: Serial sectioning. Brain tissue from mice in the sham-operated group, model group, and NGR1-M medium-dose treatment group was extracted. Using the anterior fontanelle (Bregma) as the anatomical reference, coronal sections containing the hippocampus, cerebral cortex, and striatum were selected in a cryostatus microtome (corresponding to the mouse brain stereotactic coordinates Bregma range of -1.5 mm to -2.5 mm). Four consecutive coronal sections were prepared, designated as sections ① to ④. Note that section ① is 10 μm thick, section ② is 50 μm thick, section ③ is 15 μm thick, and section ④ is 16 μm thick.

[0100] Step 6: HE staining of sections. Section ① was fixed in 4% paraformaldehyde solution for approximately 15 minutes. Then, the sections were rinsed with PBS to remove the fixative, rinsing 3-3 times for 5 minutes each time. After rinsing, the sections were stained with hematoxylin for 5 minutes. The sections were rinsed with water to remove excess dye. The sections were then stained with eosin for 3 minutes. After staining, the sections were sequentially dehydrated in different concentration gradients of ethanol (70%, 85%, 95%, and 100%), 1 minute for each stage. The sections were then cleared by immersion in xylene for 2 minutes. Finally, the sections were mounted with neutral resin. The stained sections were observed under a microscope, and images were collected using an image acquisition system for pathological analysis and registration analysis.

[0101] Step 7: Single-cell transcriptome sequencing. The corresponding slides ② from 3 mice in each group were combined into one sample, for a total of 3 samples. Single-cell transcriptome sequencing technical services were provided by Shanghai Ouyi Biomedical Technology Co., Ltd., and all experiments were conducted strictly according to the 10x Genomic Single Cell Operation Manual (USER GUIDE: Chromium Next GEM Single Cell 3'Reagent Kits v3.1, CG000204 RevD). First, single-cell suspensions were prepared using tissue homogenization and enzymatic digestion. The obtained brain tissue was mechanically homogenized under sterile conditions, followed by enzymatic digestion using a trypsin-containing digestion solution to isolate single cells. Then, using the Chromium Single-Cell 3' Library Preparation Kit on the 10x Genomics platform, following the manufacturer's instructions, single cells were captured using a microfluidic chip system to construct a transcriptome sequencing library. During the microfluidic chip sorting and cDNA synthesis process, the mRNA of each cell is reverse transcribed into cDNA, and cell-specific barcodes and UMIs (Unique Molecular Identifiers) are added via transposase reaction during the template conversion stage. After the transcriptome sequencing library passes quality testing, it is placed on the Illumina NovaSeq6000 sequencing platform for paired-end sequencing.

[0102] After single-cell sequencing, the Cell Ranger package provided by 10x Genomics was used for data quality control, read alignment, UMI counting, and cell identification. The latest mouse genome version was selected as the reference genome for alignment. Cells expressing more than 200 but less than 7500 unique genes and with read mapping less than 10% of mitochondrial genes in tissue sections were screened using the Seurat package 5.0.1 in R Studio 4.2.1. Then, the doubletFinder function of DoubletsFinder 2.0.4 was used to remove bimodalities. The merge function was then used to combine the remaining cells from the three datasets for subsequent analysis, including principal component analysis using the RunPCA function and removing inter-group effects using harmony 1.2.0. The top 20 principal components (PCs) were then selected for unsupervised clustering and t-distributed stochastic neighbor embedding (t-SNE) dimensionality reduction. The FindClusters function was used to cluster the cells, and each cluster was labeled and annotated. Finally, the remaining cells were analyzed downstream. The CellChat package 1.6.1 was used to analyze communication signaling pathways between different cell types in single-cell transcriptome data, including signal transmission, reception, and potential biological effects. The Metascape online platform was used to perform GO and KEGG pathway enrichment analysis on differentially expressed genes. After the enrichment analysis, the enrichment results provided by Metascape were exported and visualized using the R package ggplot2. Single-cell transcriptome data information was obtained.

[0103] Step 8: Spatial-seq 2.0 spatial transcriptome sequencing. Stroke disease exhibits spatial heterogeneity, with the dominant regions concentrated in the cortex, hippocampus, and striatum. Therefore, Spatial-seq 2.0 spatial transcriptome technology primarily analyzes the cortex, hippocampus, and striatum. Each slice ③ underwent high-resolution microscopic scanning to obtain true LCM scan slices. The Spot regions selected for each slice included the right motor cortex, sensory cortex, olfactory cortex, hippocampus, striatum, and the contralateral cortex, totaling 20 Spots. A total of 180 Spot regions were segmented from all mice. To match the resolution of mass spectrometry imaging, the radius of each Spot was set to approximately 50 μm, and the area to approximately 8,000 μm. 2 There are approximately 100 cells. The experimental procedure is as shown in Example 1: 1. Synthesis of DNA spatial barcodes; 2. Microcutting and barcode capture tagging; 3. Reverse transcription and cDNA amplification; 4. cDNA library construction and sequencing; 5. Sequencing result processing; 6. Spatial-seq 2.0 spatial transcriptome data analysis. Obtain spatial transcriptome data information.

[0104] Step 9: Mass Spectrometry Imaging. Matrix-assisted laser desorption / ionization-mass spectrometry (MALDI-MSI) was used to explore spatial metabolic changes in brain tissue. Tissue section ④ was attached to a glass slide, dried in a vacuum desiccator for 1 h, and optical images were captured using an optical microscope. Next, an automated matrix sprayer was used to spray the section with a prepared DHB matrix solution at a flow rate of 20 μL / min and a nitrogen flow rate of 5 L / min. The spraying process lasted 50 min, with a nozzle movement speed of 3 mm / s to ensure uniform matrix coverage of the entire brain tissue section. Subsequently, in the MALDI-MSI experiment, mass spectrometry imaging analysis was performed using an AP-SMALDI 10 instrument system. The laser focus was manually optimized to a focused beam diameter of 5 μm using the source camera. Spectral data for each pixel were acquired through 50 cumulative laser irradiations at a frequency of 100 Hz. Mass spectrometry parameters were set in Tune software (Thermo Fisher Scientific), including a spray voltage of 4 kV, S-Lens of 100 eV, and capillary temperature of 250 °C. A positive ion full scan mode was used, with a mass range of m / z 80–1200 and a mass spectrometry resolution of R = 70000 @ m / z 200. Internal calibration was performed using the instrument's lock-in mass function. The sample stage step size was set to 50 μm, and the laser energy was set to Filter 20% + 20°. Finally, the acquired data were processed using Mirion software and the open-source MSiReader software to complete the entire MALDI-MSI analysis process and obtain spatial metabolomics data.

[0105] Step 10: Single-cell spatial multi-omics affine registration and joint data analysis.

[0106] (1) Construction of spatial reference coordinate system: The histological features of the HE stained sections obtained in step 6 were extracted using QuickNII software. Based on the ventricular structure, hippocampal dentate gyrus morphology and cortical layering features, an initial spatial coordinate system was established and registered to the Allen Brain Atlas standard reference map to complete millimeter-level spatial positioning.

[0107] (2) Single-cell-spatial transcriptome data mapping: Single-cell transcriptome sequencing data and Spatial-seq 2.0 data were jointly analyzed. Multimodal data fusion was performed using the Tangram algorithm: Using single-cell sequencing data as input, cell types were identified based on the Leiden clustering algorithm, and a cell type-specific gene expression feature matrix was constructed.

[0108] (3) Affine registration of multimodal data: HE stained sections, LCM raw scan images (Spatial-seq 2.0 spatial transcriptome raw images) and mass spectrometry imaging sections were subjected to morphological erosion processing (3×3 structural nuclei), tissue contour feature point clouds were extracted, feature point matching was performed, and after registration, multi-omics data of five independent regions, namely hippocampus, striatum, motor cortex, sensory cortex and olfactory cortex, could be analyzed independently.

[0109] result: Result 1: Evaluation of the overall pharmacodynamics of NGR1 in MCAO mice.

[0110] Figure 2 The diagram shows the overall efficacy evaluation results of NGR1 in MCAO mice; (A) is a schematic diagram of the 24-hour efficacy experiment protocol; (B) is a diagram showing the effect of different doses of NGR1 on the neurological and behavioral scores of MCAO mice. Twenty-four hours after MCAO modeling, neurological and behavioral scores were assessed in each group of mice to evaluate the efficacy of different doses of NGR1 on the MCAO group mice. Figure 2 (B) In the neurological deficit assessment, NGR1-L administration showed comparable efficacy to edaravone in improving neurological function in MCAO mice. NGR1-M and NGR1-H exhibited superior neuroprotective effects compared to the positive control drug edaravone. NGR1-M (30 mg / kg) showed the best efficacy, and further multi-omics mechanism studies will be conducted using the medium-dose NGR1 group.

[0111] Result 2: Effect of NGR1 on pathological morphological changes in brain tissue of MCAO mice.

[0112] When observing the effects of NGR1 on the pathological morphological changes of brain tissue in mice with midbrain artery occlusion (MCAO), the results of HE staining were as follows: Figure 3The images show representative HE staining of the entire mouse brain tissue and cortical areas (n = 3); scale bars are 2000 μm and 100 μm. In the sham-operated group, the left and right hemispheres of the brain showed uniform staining, regular cell morphology, and no edema or vacuolation was observed. The cell nuclei appeared plump, and the nucleoli were clearly visible. In contrast, the infarcted areas of the brain tissue in the model group mice showed lighter staining, significant edema, and numerous vacuolations. The cell nucleoli were shrunken and darkened. However, compared with the model group, mice treated with NGR1 showed a significant reduction in brain edema and nucleolar shrinkage, indicating that NGR1 can effectively improve the pathological changes in the brain tissue of MCAO mice.

[0113] Result 3: Effect of NGR1 on the heterogeneity of brain tissue cells in MCAO mice.

[0114] To investigate the dynamic changes of NGR1 in mouse brain cell subtypes after MCAO model, this invention selected brain tissue from the sham-operated group (Sham), the model group (Model), and the medium-dose group of Panax notoginseng saponin R1 (NGR1-M) after 24 hours of treatment. Brain cells were extracted from slide ②, and single-cell suspensions were prepared. scRNA-seq analysis was performed using the 10x Genomics platform, etc. Figure 4 (A in the text). After obtaining the data, it was filtered by Seurat quality control and double cells and red blood cells were removed. The RNA sequencing data of 37,212 brain tissue cells from single-cell suspensions of 9 samples (sham-operated group, model group, and 3 samples from the medium-dose group of Panax notoginseng saponin R1, each sample containing a mixture of hemispheres of 3 mice in the group) were collected into one dataset. The number of cells and genes after filtering for each group is shown in Table 1.

[0115] Table 1. Data from brain cell scRNA-seq groups

[0116] After filtering out low-quality cells, t-SNE was used to perform dimensionality reduction, clustering, and visualization analysis on the single-cell transcriptome data. Figure 4 Results B showed that brain cells were separated into 24 cell clusters based on similar gene expression patterns. By cross-referencing the unique gene characteristics of each cluster with literature information and known characteristics of mouse brain cell types in the CellMarker database, the 24 cell clusters were partially merged into 15 major cell classes. Figure 4 As shown in C, the 15 major cell classes and characteristic marker genes are as follows: endothelial cells ( Cldn5 ), microglia ( Cx3cr1 ), B cells ( Cd79a ), T cells ( Itk ), neurons Sox11 ), smooth muscle cells ( Acta2 ), macrophages ( Cybb Oligodendrocytes ( Mog ), astrocytes ( Slc1a2 ), neutrophils ( Hdc ), pericytes ( Vtn ), choroid plexus cells ( Igfbp2 ), ependymal cells ( Tmem212 Oligodendrocyte precursor cells ( Fabp7 ) and fibroblasts ( Timp1 ).

[0117] Figure 4 Single-cell atlas of NGR1-treated MCAO mice; (A) Schematic diagram of single-cell sequencing process; (B) t-SNE plot of 37,212 cells with 24 cell clusters (left) and 15 cell types (right); (C) Characteristic map of expression levels of marker genes in 15 cell types, shown in blue.

[0118] Result 4: NGR1 on Socs3 hi Regulatory role of anti-inflammatory subtypes of microglia.

[0119] Further cluster analysis was performed on a total of 11,791 microglia from the three groups mentioned above, identifying 11 distinct subclusters. These subclusters were then identified into 6 subtypes based on the expression of specific marker genes: P2ry12 hi , Socs3 hi , Adam8 hi , Apod hi , Irf7 hi , Prkce hi Microglia ( Figure 5 A and Figure 5 (B in the text). The MCAO and NGR1 treatment groups showed significant differences between their subtypes ( Figure 5 C and Figure 5 (D in the text). All six microglia subtypes showed a regression state after NGR1 administration. GO enrichment analysis was performed on the highly expressed genes of the six cell subtypes to determine... Socs3 hi Microglia belong to the anti-inflammatory subtype. Adam8 hi Microglia belong to the pro-inflammatory subtype.

[0120] Socs3 The expression of this substance can suppress excessive inflammatory responses and help maintain the balance of the immune system. Figure 5(E in the original text). Therefore, Socs3 Often regarded as an anti-inflammatory marker gene, it plays a crucial role in various anti-inflammatory processes.

[0121] Further analysis of NGR1 regulation of microglia differentiation trajectory was conducted, and the impact of NGR1 treatment on microglia fate transition was further determined by constructing a pseudo-time trajectory. First, CytoTRACE was used to predict the differentiation status of six microglia subtypes (…). Figure 5 (F in) Socs3 hi Microglia had the highest CytoTRACE score, indicating Socs3 hi Microglia subtypes appear more "mature" and are closer to the endpoint of the differentiation pathway. Similarly, Figure 5 The same result was obtained for G in the equation. Socs3 hi Microglia subtypes are located at the end of the trajectory, indicating that they are more mature cell types with specific functional roles. Observation of the pseudo-time trajectories of three groups of microglia ( Figure 5 In the study of H), it was found that, compared with the control group, the NGR1 group showed higher rates of development in terminal cells (H). Socs3 hi The continuous increase in microglia indicates that NGR1 releases anti-inflammatory factors, gradually restoring homeostasis. To verify the anti-inflammatory effect of NGR1, we identified a marker gene highly expressed in the anti-inflammatory subtype of microglia. Socs3 The expression of. For example. Figure 5 As shown in Figure I, compared to the model group, the medium-dose group of Panax notoginseng saponin R1 showed... Socs3 The expression abundance of [a substance] was significantly increased, a result further validated by ELISA experiments (see [link]). Figure 5 J in the middle.

[0122] Figure 5 This is a comprehensive analysis of microglia subtypes; (A) results showing the identification of 11,791 microglia into 6 subtypes; (B) expression results of marker genes for the six subtypes; (C) the number of each microglia subtype in the three samples; (D) the proportion of each microglia subtype in the three samples; (E) Socs3 hi (F) GO analysis results of high-expression genes in microglia subsets; (G) Prediction results of differentiation degree of different cell subtypes; (H) Pseudo-time trajectory results of different microglia subtypes; (I) High-expression marker genes of anti-inflammatory subtypes. Socs3 Expression level violin plot; (J) ELISA results of changes in Socs3 content in brain tissue (n = 5).

[0123] Result 5: Mass spectrometry imaging verified that NGR1 reduced inflammation in the brain of MCAO mice.

[0124] This study comprehensively evaluated the impact of NGR1 on the anti-inflammatory response using mass spectrometry imaging experiments, and simultaneously verified how these molecular events affect the entire brain tissue. Figure 6 A and Figure 6 As shown in Figure B, several lipid metabolic pathways related to phospholipids, sphingolipids, and neutral lipids showed significant changes in different groups (sham-operated group, model group, and medium-dose Panax notoginseng saponin R1 group) and spatially (ipsilateral / contralateral brain region). Twenty-four hours after cerebral ischemia-reperfusion injury, compared to the sham-operated group, the model group showed a decrease in phosphatidylcholine (PC) components (e.g., PC (34:1) and PC (32:0)) and an increase in their metabolized lysophosphatidylcholine (e.g., LPC (16:0) and LPC (18:0)) on the ipsilateral side of the brain, a situation improved after NGR1 treatment. Furthermore, phosphatidylethanolamine (PE) components and their metabolites (LPE) showed similar trends. This invention suggests that the reduction in PC and PE after brain injury may be due to the activation of the lipoprotein-associated phospholipase A2 catalytic mechanism induced by cerebral ischemia and intracellular calcium overload. Reports indicate that the hydrolysis of PC and PE produces various free fatty acids, including LPC and LPE, which can generate free radicals and exacerbate cell damage. Mass spectrometry imaging experiments revealed the spatial distribution changes of major lipid components and their metabolites during cerebral ischemia-reperfusion injury from a holistic perspective, demonstrating a close relationship between cerebral lipids and the evolution of tissue infarction. NGR1 influences the degradation or transformation of phospholipids in the brain, providing more n-3 or n-6 unsaturated fatty acids to ischemic brain tissue, producing more anti-inflammatory mediators or modulating the inflammatory response, thereby helping to alleviate ischemic-induced intracranial inflammation.

[0125] Figure 6 These are representative images of brain lipids obtained by MALDI-MSI imaging; (A) spatial metabolic changes of various lipids; (B) spatial metabolic changes of corresponding lipid metabolites.

[0126] Result 6: NGR1 regulates the spatial heterogeneity of lipid distribution in the brain of MCAO mice.

[0127] As mentioned earlier, spatial heterogeneity plays a crucial role in the development and progression of stroke. Single-cell sequencing reveals differences in gene expression at the individual cell level, spatial transcriptomic data provides the spatial distribution of these differences within tissues, and spatial metabolomic data reveals the spatial distribution of metabolites. Integrating multiple omics data allows for a comprehensive understanding of the biological basis of stroke at every level, from gene expression to metabolites. Similarly, to accurately determine spatial differences, registration of mass spectrometry imaging data with actual LCM scan sections and the Allen Brain Atlas standard reference atlas revealed different variations in local metabolites in different regions of the hippocampus, striatum, and cortex (divided into sensory cortex, motor cortex, and olfactory cortex). Figure 7 A in the middle. Figure 6 The expression abundance of all phosphatidylcholine components (PC), phosphatidylethanolamine components (PE), lysophosphatidylcholine (LPC), and lysophosphatidylethanolamine (LPE) in different brain regions was quantitatively analyzed. The effect of NGR1 on the restorative balance of the metabolic network in MCAO mice was further evaluated using the efficiency of recovery regulation (EoR). This method mainly involves comparing the changes in the expression of endogenous metabolites in normal states, disease states, and after NGR1 administration. Figure 7 As shown in Figure B, after NGR1 administration, the phosphatidylcholine component (PC) showed retrograde activity in the hippocampus, striatum, sensory cortex, and olfactory cortex; the phosphatidylethanolamine component (PE) showed retrograde activity in the hippocampus, striatum, and motor cortex; and lysophosphatidylcholine (LPC) and lysophosphatidylethanolamine (LPE) showed retrograde activity in all brain regions. Taking the common intersection of the retrograde regions of the four components revealed that the hippocampus and striatum may be key brain regions for NGR1 to spatially regulate lipid metabolism. Figure 7 C and Figure 7 As shown in D, five types of lipid metabolism (PC), lipoprotein (LPC), polymorphic acid (PE), polymorphic acid (LPE), and styrax (SM) were randomly selected and their changing trends in the hippocampus and striatum were quantitatively analyzed. Compared with the sham-operated group, the present invention observed a significant decrease in the levels of PC, PE, and SM in the hippocampus and striatum of the model group. This result suggests that specific lipid metabolism imbalances in brain regions after stroke may be related to cell membrane damage and brain cell dysfunction. Furthermore, after NGR1 treatment, compared with the model group, the levels of PC, PE, and SM in the hippocampus and striatum significantly increased, while the levels of LPC and LPE significantly decreased, gradually approaching those of the sham-operated group. This indicates that NGR1 has a potential protective effect against brain injury caused by stroke by regulating lipid metabolism and reducing intracranial inflammation in the hippocampus and striatum.

[0128] Figure 7The results of spatial metabolite analysis are shown in the figure. Among them, (A) is a schematic diagram of LCM and mass spectrometry imaging data registration; (B) is a diagram of the recovery regulation rate of PC, PE, LPC and LPE in different spatial regions; (C) is a diagram of the relative abundance change of representative lipids in the hippocampus; and (D) is a diagram of the relative abundance change of representative lipids in the striatum.

[0129] Result 7: NGR1 regulates the spatial heterogeneity of pro-inflammatory / anti-inflammatory microglia in the brain tissue of MCAO mice.

[0130] Spatial-seq transcriptome sequencing data provides spatial distribution information of gene expression, but its spatial resolution is relatively low. Single-cell sequencing data provides high-resolution cell type information, but lacks spatial information. Using a Tangram data integration strategy to map single-cell sequencing data onto spatial-seq transcriptome sequencing data, combining the two can obtain a more fine-grained cell type distribution map while preserving spatial information, thereby enhancing the understanding of cellular heterogeneity at the spatial level. Figure 8 A). Extract spatial distribution expression information of microglia, such as Figure 8 As shown in Figure B, compared with the sham-operated group, the spatial distribution of microglia in various spatial regions of the model group was significantly reduced, and microglia in most spatial regions showed regression under the action of NGR1. Further evaluation of the rebalancing effect of NGR1 on the transcriptional network of MCAO mice using EoR showed that, except for the olfactory cortex, other brain regions showed some regression, with the hippocampus showing the most significant regression. The hippocampus is a key area in the brain closely related to learning and memory, and is particularly sensitive to hypoxia and inflammation. After stroke, the hippocampus exhibits more pronounced damage and inflammatory responses due to its unique structure and function. Therefore, the therapeutic effect of NGR1 is more significant in the hippocampus, possibly reflecting the highly active repair process after injury in this region. Spatially evaluating pro-inflammatory subtypes (… Adam8 hi ) and anti-inflammatory subtypes ( Socs3 hi The distribution and abundance changes of microglia. Results are as follows: Figure 8 As shown in C, after NGR1 administration Socs3 There was some degree of regression in the hippocampus, striatum, motor cortex, and sensory cortex, with the most significant regression in the hippocampus. For example... Figure 8 As shown in D, Adam8 There was no reversion in any brain region, and compared to the sham surgery group, the model group... Adam8 NGR1 expression was lowest in the hippocampus. In summary, compared to other brain regions, NGR1 has a more pronounced regulatory effect on microglia in the hippocampus, and NGR1 promotes... Socs3 hiThe transformation of anti-inflammatory microglia subtypes and the mechanisms of reducing inflammatory responses have a more significant effect in the hippocampus region.

[0131] Figure 8 The diagram shows the spatial heterogeneity of NGR1 regulation of pro-inflammatory / anti-inflammatory microglia; (A) is a schematic diagram of single-cell spatial mapping; (B) is a diagram showing the spatial distribution changes of microglia in different groups; (C) Socs3 Map showing the spatial distribution differences of genes; (D) Adam8 A graph showing the spatial distribution differences of genes.

[0132] Result 8: Single-cell sequencing, Spatial-seq 2.0 spatial transcriptome sequencing, and mass spectrometry imaging multi-omics technologies jointly revealed the key role of Panax notoginseng saponin R1 in regulating intracranial inflammation and promoting cerebral blood flow recovery.

[0133] NGR1 can specifically act on Socs3 High expression of microglia subpopulations promotes their transformation into anti-inflammatory subtypes and increases the production of anti-inflammatory factors. Mass spectrometry imaging further demonstrates that NGR1 regulates lipid metabolism in the brain, particularly reducing the levels of phosphatidylcholine (PC) and phosphatidylethanolamine (PE) while increasing the levels of phosphatidylcholine (LPC) and phosphatidylethanolamine (LPE), thus promoting the formation of an anti-inflammatory phenotype. Multi-omics analysis reveals that microglia mainly exert their immune and anti-inflammatory effects in the hippocampus. NGR1 also specifically regulates pericytes and smooth muscle cell subtypes, exerting vasodilatory effects and regulating muscle contraction and relaxation, thereby exerting anti-inflammatory effects and promoting blood flow restoration in vivo. This provides a new strategy for the treatment of cerebral ischemia-reperfusion injury and also reveals the panoramic mechanism of action of NGR1 at the spatial level. Figure 9 This is a schematic diagram illustrating the panoramic mechanism of action of Panax notoginseng saponin R1 in combating acute ischemic stroke at the spatial level.

[0134] Results analysis: Result 4, a single-cell sequencing result, could not pinpoint the distribution of anti-inflammatory cell subtypes in the brain. Spatial-seq 2.0 could not distinguish microglia subtypes (such as...). Socs3 hi and Adam8 hi Result 5 is a mass spectrometry imaging result image, and the changes in metabolites obtained could not be correlated with genes and spatial locations.

[0135] Result 6 is the correlation between mass spectrometry imaging and Spatial-seq 2.0 spatial information: it can be found that there are different changes in local metabolites in different regions of the hippocampus, striatum, and cortex (divided into sensory cortex, motor cortex, and olfactory cortex).

[0136] Result 7 is the association between Spatial-seq 2.0 spatial transcriptome sequencing and single-cell data: localizing gene expression, distinguishing cell subtypes through single-cell sequencing, and achieving single-cell spatial mapping using the Tangram algorithm. In the stroke model, it was found that NGR1 has a more significant regulatory effect on microglia in the hippocampus, and NGR1 promotes... Socs3 hi The transformation of anti-inflammatory microglia subtypes and the mechanisms of reducing inflammatory responses have a more significant effect in the hippocampus region.

[0137] Results 6 and 7 provide multi-dimensional data synergistic validation: mass spectrometry imaging validated regions of abnormal accumulation of metabolites (such as LPC), which highly overlapped with the expression of inflammatory genes in the spatial transcriptome, demonstrating the synergistic effect of anti-inflammatory and metabolic regulation.

[0138] Result 8 demonstrates the advantages and importance of the combined tri-omics approach. By integrating single-cell transcriptomics, spatial transcriptomics, and mass spectrometry imaging data, it overcomes the limitations of traditional single-omics or dual-omics technologies, achieving a complete chain mechanism analysis encompassing "cell type-spatial location-metabolic microenvironment." Compared to the limitations of single-cell sequencing technology, which can only identify cell subpopulations, and the lack of metabolic dynamic verification in single-cell and spatial omics dual-modal analysis, the combined tri-omics approach not only provides a more comprehensive explanation of the multi-dimensional regulation of the brain microenvironment by Panax notoginseng saponin R1, promoting cerebral blood flow recovery and reducing inflammation, but also fully reveals the spatiotemporal dynamic mechanism by which Panax notoginseng saponin R1 synergistically regulates microglial anti-inflammatory polarization and pericyte vascular repair function through lipid metabolism reprogramming, from three dimensions: molecular interaction networks, cellular spatial polarization, and lipid metabolism gradients. The combined tri-omics analysis provides experimental evidence for the modern scientific explanation of the "blood-activating and stasis-removing" effect of Panax notoginseng in traditional medicine, and also provides an innovative methodological paradigm for the modern scientific interpretation of the multi-target synergistic effects of traditional Chinese medicine.

[0139] It is important to note that the combined analysis of three omics is not a simple superposition of technologies, but rather faces numerous technical challenges. Conventional omics integration methods have significant limitations: First, different omics technologies have inherent differences in resolution, sensitivity, and dynamic range. Single-cell transcriptomics provides gene expression information at the cellular level, spatial transcriptomics presents gene localization at the tissue slice level, while mass spectrometry imaging reflects metabolite distribution at the molecular level. These three technologies are difficult to directly match in terms of data dimensions and scales. Second, existing dual-omics integration methods typically employ simple coordinate mapping or correlation analysis, which cannot effectively address the nonlinear relationships and spatiotemporal heterogeneity between different omics data, leading to information loss or false positive associations. Third, traditional sample processing workflows are often optimized for a single omics. When attempting to combine multiple omics, differences in sample fixation, staining, or processing conditions can lead to irreversible loss of biological information. For example, RNA degradation affects transcriptomics data quality, or the mass spectrometry ionization process interferes with spatial localization accuracy.

[0140] This application innovatively develops a method for single-cell spatial multi-omics affine registration and joint data analysis, establishing a multimodal mapping system of spatial reference coordinates, single cells, spatial transcriptomics, and spatial metabolomics, thus solving the problems of technical deviations and spatial positioning mismatches between different omics platforms. This invention optimizes the continuous tissue section preparation process for multi-omics detection, designs a standardized processing scheme compatible with single-cell sequencing, Spatial-seq 2.0 spatial transcriptomics sequencing, mass spectrometry imaging, and corresponding tissue staining, and sets differentiated section thicknesses for different detection needs, maximizing the preservation of the original biological information of tissue samples. Simultaneously, this invention constructs a multimodal data fusion analysis system based on the Tangram algorithm, combined with the Leiden clustering algorithm and multimodal affine registration strategy, which can effectively capture the complex correlations between gene expression, cell spatial distribution, and metabolite spatial distribution, rather than a simple linear correspondence. These technological breakthroughs enable a true synergistic effect of "1+1+1>3" for the three omics data, rather than a simple data aggregation. It is precisely these key technological innovations that enable this invention to accurately analyze the dynamic regulatory network of "gene expression regulation - cell subtype polarization - spatiotemporal distribution of metabolites" of Panax notoginseng saponin R1 in the process of cerebral ischemia injury repair, providing an unprecedented multi-dimensional and complete chain of evidence for the analysis of the synergistic mechanism of multi-target effects of traditional Chinese medicine.

[0141] This invention constructs a "gene (Socs3)-cell (microglia)-metabolism (PC / LPC)" network by integrating data, revealing the global mechanism by which drugs regulate lipid metabolism through anti-inflammatory subtype cells. In the examples, after NGR1 treatment, the activity of the anti-inflammatory pathway in the hippocampus was enhanced, and the content of the metabolite PC returned to normal levels.

[0142] Although the above embodiments have provided a detailed description of the present invention, they are only some embodiments of the present invention, and not all embodiments. People can obtain other embodiments based on these embodiments without creative effort, and these embodiments all fall within the protection scope of the present invention.

Claims

1. A method for elucidating disease treatment mechanisms based on the combined use of single-cell sequencing, spatial transcriptomics, and mass spectrometry imaging, comprising the following steps: Tissue sections were prepared; tissue sections were stained, single-cell sequencing, Spatial-seq 2.0 spatial transcriptome sequencing, and mass spectrometry imaging were performed on the tissue sections; single-cell spatial multi-omics affine registration and data joint analysis were performed. The diseases mentioned include stroke; When the disease is stroke, the tissue is brain tissue; The tissue sections are continuous sections; the thickness of the tissue sections for tissue staining is 5~18μm, and the thickness of the tissue sections for single-cell sequencing is 20~50μm. The thickness of tissue sections for Spatial-seq 2.0 spatial transcriptome sequencing is 10–20 μm; the thickness of tissue sections for mass spectrometry imaging is 10–20 μm. When the disease is stroke, the single-cell spatial multi-omics affine registration and data joint analysis includes: constructing a spatial reference coordinate system, single-cell-spatial transcriptome data mapping, and multimodal data affine registration; The construction of the spatial reference coordinate system includes: using QuickNII software to extract histological features from HE-stained sections, and based on the ventricular structure, hippocampal dentate gyrus morphology and cortical layering features, establishing an initial spatial coordinate system and registering it to the AllenBrain Atlas standard reference map to complete millimeter-level spatial positioning. The single-cell-spatial transcriptome data mapping includes: joint analysis of single-cell transcriptome sequencing data and Spatial-seq 2.0 data; and multimodal data fusion using the Tangram algorithm: using single-cell sequencing data as input, identifying cell types based on the Leiden clustering algorithm, and constructing a cell type-specific gene expression feature matrix. The multimodal data affine registration includes: morphological erosion processing (3×3 structural nuclei) of HE-stained sections, Spatial-seq 2.0 spatial transcriptome raw images and mass spectrometry imaging sections, extraction of tissue contour feature point clouds, and feature point matching. After registration, multi-omics data of five independent regions, namely the hippocampus, striatum, motor cortex, sensory cortex and olfactory cortex, can be analyzed independently.

2. The method for analyzing the disease treatment mechanism according to claim 1, characterized in that, The tissue staining method includes HE staining.

3. The method for analyzing disease treatment mechanisms according to claim 1, characterized in that, The Spatial-seq 2.0 spatial transcriptome sequencing is a high-throughput spatial transcriptome sequencing method based on a laser microdissection system and DNA barcode labeling.

4. The method for analyzing disease treatment mechanisms according to claim 1, characterized in that, The single-cell sequencing includes single-cell transcriptome sequencing; the single-cell transcriptome sequencing platform includes the 10x Genomics-Chromium platform, the BDRhapsody platform, or the BGI Genomics-DNBelab C4 platform.

5. The method for analyzing disease treatment mechanisms according to claim 4, characterized in that, The method for single-cell transcriptome sequencing includes: homogenizing tissue sections, enzymatically digesting them to obtain single cells, constructing a transcriptome sequencing library, and performing sequencing on a single-cell transcriptome sequencing platform.

6. The method for analyzing disease treatment mechanisms according to claim 1, characterized in that, The mass spectrometry imaging includes MALDI-MSI mass spectrometry imaging.

7. A disease treatment mechanism analysis system based on the combined use of single-cell sequencing, spatial transcriptomics, and mass spectrometry imaging, comprising a single-cell sequencing unit for obtaining single-cell sequencing data; Spatial-seq 2.0 spatial transcriptome sequencing unit, used to obtain Spatial-seq 2.0 spatial transcriptome sequencing data; Mass spectrometry imaging unit, used to acquire space metabolomics data; The unit includes a single-cell spatial multi-omics affine registration and data analysis unit, used to perform affine registration and analysis on data acquired by the single-cell sequencing unit, the Spatial-seq2.0 spatial transcriptome sequencing unit, and the mass spectrometry imaging unit.

8. The application of the disease treatment mechanism analysis method according to any one of claims 1 to 6 or the disease treatment mechanism analysis system according to claim 7 in any one of the above-mentioned studies as described in ① to ⑤: ① disease mechanism analysis research; ② drug development research; ③ biomarker discovery research; ④ development and regenerative medicine research; ⑤ tumor immunotherapy development research; the application is for non-clinical treatment purposes.

Images