Neuroglial cell marker set for early warning critical point of brain aging and application of neuroglial cell marker set

Abstract

The invention relates to the field of senescence and systematic dynamics research, in particular to a glial cell marker set of a brain senescence critical point and application of the glial cell marker set in senescence mechanism research. 45 healthy human brain samples of 29-94 years old are analyzed through a single cell transcriptome, it is found for the first time that 56-60 years old are brain aging critical points, and a dynamic network marker (DNB) set composed of microglial cells, astrocytes and oligodendrocytes is identified and comprises key molecules mTOR, PADI2, JAK1, ARHGEF11 and KAZN. The marker set realizes early warning of an aging critical point by detecting the following characteristics: (1) PASP-GPR37L1 / PASP-GPR37 communication between glial cells is obviously weakened; (2) the activity of the neuron-microglial cell CX3CL1-CX3CR1 / IL34-CSF1R pathway is reduced, and the activity of the neuron-microglial cell pathway is reduced; and (3) DNB molecular expression fluctuation peak. The marker set disclosed by the invention can be used for developing a brain aging early warning kit and screening intervention targets.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of aging and systems dynamics, and in particular to a set of glial cell markers of brain aging critical point and its application in the study of aging mechanism. BACKGROUND

[0002] Brain aging is a complex process that affects everything from the subcellular to the organ level, starting in early life and accelerating with age. Morphologically, there are two major manifestations: one is the gradual reduction in brain volume, which is a slight and continuous reduction that occurs in the cortex and mainly affects functions such as executive movement and psychological activity; the other is the slow shrinkage of the white matter of the hippocampus, which is responsible for long-term memory and complex psychological activity. Pathophysiology, brain aging is associated with neuronal cell atrophy, dendritic degeneration, demyelination, small vessel disease, metabolic slowing, microglial cell activation, and white matter lesion formation. Aging is also a major risk factor for neurodegenerative diseases, and common neurodegenerative diseases such as Alzheimer's disease (AD), Parkinson's disease (PD), and Huntington's disease (HD) often occur in the elderly and affect hundreds of millions of people.

[0003] The dynamic and nonlinear characteristics of brain aging have been gradually revealed, mainly manifested as fluctuation peaks of molecular expression at specific ages during the aging process. Lehallier et al. found that protein clusters changed in different patterns at 40, 60, and 80 years of age in the aging plasma proteome. Shen et al. found significant peaks at around 40 and 60 years of age through multi-omics data, marking the dysregulation of molecules. The latest research evaluated the changes in brain aging through machine learning models and brain tissue data based on multi-modal magnetic resonance imaging, and found three aging inflection points at 57, 70, and 78 years of age, as well as specific signaling pathways at different inflection points. These results suggest that brain aging is not just the result of damage accumulation, but more likely a manifestation of a series of dynamic networks from homeostasis to collapse. Therefore, it becomes more meaningful to determine which factors and stages lead to the nonlinear characteristics of aging or the critical state of aging.

[0004] DNB theory believes that the change of the system in the development process is not gradual, but the system gradually changes internally but always maintains the original state. When it reaches a "critical state" at a certain moment, once it passes through the "critical state", the system will be sharply directed to a disastrous state and it is difficult to recover to the initial state. DNB theory provides a quantitative tool for identifying the critical point in the aging process. DNB can capture the signal that the system is close to the critical point in the early stage of aging, such as increased volatility, enhanced network coupling and abnormal changes in the covariance matrix. This perspective can help us explore potential anti-aging strategies and biological regulation targets from the dynamic behavior of the whole system, thereby promoting the further development of healthy aging research. SUMMARY

[0005] In order to solve the above technical problems existing in the prior art, the present application provides a glial cell marker set for early warning of brain aging critical point and its application. Specifically, it is realized by the following technical solutions: A glial cell marker set for early warning of brain aging critical point, the marker is derived from a glial cell dynamic molecular network and contains gene products expressed in microglia, astrocytes and oligodendrocytes.

[0006] Further, the gene products include products of mTOR, PADI2, JAK1, ARHGEF11 and KAZN genes.

[0007] The above-mentioned glial cell marker set for early warning of brain aging critical point is applied to early warning of brain aging critical point, especially in the preparation of early warning reagent kit for early warning of brain aging.

[0008] A kit for early warning of brain aging critical point contains the above-mentioned glial cell marker set for early warning of brain aging critical point or its detection reagent.

[0009] Further, the kit for early warning of brain aging critical point is an early warning kit for early warning of brain aging. The kit can be applied to early warning of brain aging critical point.

[0010] A method for early warning of brain aging critical point, which detects the expression level of the identified aging critical state marker gene in microglia, astrocytes and oligodendrocytes to early warning of critical point.

[0011] Further, the expression level of the identified senescence critical point gene in microglia cells, astrocytes and oligodendrocytes is detected, and if the gene expression of mTOR, PADI2, JAK1, ARHGEF11 and KAZN is significantly up-regulated in the neural glial cells of the brain tissue, it is in the neural glial cell-mediated senescence critical point.

[0012] Further, the method further comprises detecting whether the following biological characteristics exist, and if the following biological characteristics exist, it is in the neural glial cell-mediated senescence critical point: (a) The strength of PSAP-GPR37L1 and PSAP-GPR37 ligand-receptor communication between neural glial cells decreases; (b) The activity of CX3CL1-CX3CR1 and IL34-CSF1R signaling pathways between neurons and microglia cells decreases.

[0013] (c) The gene expression of mTOR, PADI2, JAK1, ARHGEF11 and KAZN is significantly up-regulated in the neural glial cells of the brain tissue.

[0014] The present application proposes a set of neural glial cell markers for early warning of brain senescence critical point and its application, based on 45 cases of 29-94 year-old human brain single cell nucleus transcriptome data, a single cell transcriptome atlas of healthy aging brain is constructed. At the cell type level, we observed that the proportion of neurons in the elderly group decreased, indicating that there was damage or death of neurons during the aging process; the proportion of glial cells including microglia cells, astrocytes and oligodendrocytes increased significantly, suggesting their potential role in aging. In addition, we also identified the key features of brain tissue aging, such as decreased neuronal differentiation, synaptic assembly and plasticity regulation disorder, and pathway enrichment related to inhibition of neuronal migration, immune phagocytosis and apoptosis.

[0015] Further explore the relationship between cell type and aging, we used SASP gene set score, coefficient of variation calculation and transcriptional noise evaluation, the results showed that microglia cells, astrocytes and oligodendrocytes showed higher senescence score and greater transcriptional instability, while neurons were relatively stable. Our findings highlight the important association between glial cells and age-related changes that occur in the healthy aging human brain, indicating that the involvement of glial cells may lead to impaired neural development and synaptic regulation disorders during the aging process.

[0016] Subtype subdivision and aging-related analysis of three types of glial cells were performed to find their respective subtypes with higher correlation with aging and their functions. The DNB calculation method was introduced to identify the cell population of each cell type in the critical state of aging, and the key critical state molecules were identified through the DNB molecular network. The results show that the key DNB molecules mTOR, PADI2, JAK1, ARHGEF11, KAZN can be used as early warning markers.

[0017] According to the results of the DNB calculation method, we determine 56-60 years old as the critical point of healthy aging of the human brain, at which the disorder and dysfunction of the system will lead the system to an irreversible aging state. More importantly, glial cells are the cell type that first reaches the peak of DNB at this age and is presumed to mediate the critical state. In contrast, neurons show a later critical point age, indicating that they are more slowly and passively affected by the aging critical state. We also found the critical point corresponding to the age of 56-60 years old in the brain tissue of 18-month-old mice and identified several key DNB molecules as candidate genes for aging markers.

[0018] The potential biological characteristics and mechanisms of the aging critical state at 56-60 years old are further described. For example, the PSAP-GPR37L1 and PSAP-GPR37 signaling axes between glial cells are significantly weakened, which may weaken their nerve repair function and anti-apoptotic ability. At the same time, the activity of the CX3CL1-CX3CR1 and IL34-CSF1R signaling pathways between excitatory neurons and microglia cells decreases, or leads to abnormal activation of microglia cells, which in turn induces chronic neuroinflammation and destroys the steady-state environment of neurons. These changes collectively constitute the risk of functional imbalance that the brain tissue faces in the aging critical state, which may lay the foundation for the occurrence of aging-related pathological states.

[0019] Compared with the prior art, the technical effects of the present application are embodied in: 1. The present application constructs a single-cell nucleus transcriptome atlas of a healthy aging human brain, identifies 10 cell types, and describes the cell, molecular and functional differences in the aging process from the overall and cell type levels.

[0020] 2. Among all cell types, the proportion of neurons and glial cells changes significantly after aging. The proportion of neuronal cells decreases after aging but is relatively stable, with low transcriptional instability. In contrast, the proportion of glial cells significantly increases after aging, and shows high dynamic instability and enhanced transcriptional noise, suggesting that glial cells may play a more critical regulatory role in the nonlinear process of brain aging.

[0021] 3. The application focuses on the subtyping and aging-related differential analysis of three types of glial cells, and introduces the theory of dynamic network marker (DNB) to identify specific cells and key aging tipping point markers at the tipping point. A dynamic network marker (DNB) set consisting of microglia, astrocytes and oligodendrocytes is identified, including key molecules mTOR, PADI2, JAK1, ARHGEF11 and KAZN.

[0022] 4. The application finds that there is a significant tipping point at the age of 56-60, and the tipping point of the three types of glial cells appears at the same time at this age, while neurons and other cell types show a relatively slow and passive regulation of the tipping point, so it is believed that glial cells mainly mediate the regulation of the tipping point.

[0023] 5. The application explores the potential characteristics and mechanisms that lead to the tipping point at the age of 56-60, and finds that the significant weakening of the number and intensity of communication related to the function of glial cells is a significant feature of the tipping point at this time. Among them, the PSAP-GPR37L1 and PSAP-GPR37 communication between glial cells is significantly weakened, leading to impaired nerve repair and anti-apoptotic ability. The weakening of CX3CL1-CX3CR1 and IL34-CSF1R communication between excitatory neurons and microglia may lead to excessive activation of microglia, resulting in neuroinflammation and destruction of neuronal homeostasis. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 Sample composition and data processing instructions. The single-cell data of 8 published studies were filtered and batch effect processed.

[0025] Figure 1 B. Integrating the UMAP dimensionality reduction map of human brain, different colors represent different cell types (astrocytes n = 172571; endothelial cells n = 15972; ependymal cells n = 2185; excitatory neurons n = 219154; fibroblasts n = 22444; inhibitory neurons n = 130127; microglia n = 115559; oligodendrocytes n = 249254; oligodendrocyte precursor cells n = 71106; T cells n = 2716).

[0026] Figure C. Distribution of samples of different ages and data sources in UMAP.

[0027] Figure 1 D. The proportion of data sources, age, cell type and tissue type.

[0028] Figure 2 Expression of SASP genes in each cell type. Black represents high expression, light color represents low expression. The size of the circle represents the proportion of cells expressing the gene in this cell type.

[0029] Figure 2 B SASP activity score of young vs old group (Wilcoxon rank-sum test, *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001).

[0030] Figure 2 C UMAP of SASP score of each cell type.

[0031] Figure 2 D UMAP of CSP activity score of each cell type.

[0032] Figure 3 A Coefficient of variation (cv) of each cell type in young vs old group. Pink represents young group, dark color represents old group.

[0033] Figure 3 B Variance transcription noise calculation of each cell type in young vs old group (Wilcoxon rank-sum test, *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001).

[0034] Figure 4 A Key DNB molecular network of microglia at Tipping point period.

[0035] Figure 4 B Differential expression of PADI2 gene in young and old groups (****p<0.0001).

[0036] Figure 4 C hdWGCNA analysis clusters molecules with similar expression patterns.

[0037] Figure 4 D Correlation calculation of each clustered module with age-related traits (Student's t-test, *q<0.05, **q<0.01, ***q<0.001, ****q<0.0001).

[0038] Figure 5 A Key DNB molecular network of astrocytes at Tipping point period.

[0039] Figure 5B Differential expression of JAK1 and ARHGEF11 genes between young and old groups (****p<0.0001).

[0040] Figure 5 C hdWGCNA analysis clusters molecules with similar expression patterns.

[0041] Figure 5 D Correlation of each clustered module with age-related traits computed (Student's t-test, *q<0.05, **q<0.01, ***q<0.001, ****q<0.0001).

[0042] Figure 6 A Key DNB molecular networks at the tipping point of oligodendrocytes.

[0043] Figure 6 B Differential expression of KAZN genes between young and old groups (****p<0.0001).

[0044] Figure 6 C hdWGCNA analysis clusters molecules with similar expression patterns.

[0045] Figure 6 D Correlation of each clustered module with age-related traits computed (Student's t-test, *q<0.05, **q<0.01, ***q<0.001, ****q<0.0001).

[0046] Figure 7 DNB computes tipping point predictions for each cell type at each age Figure 8 A PCA results showing the distribution of age samples in published mouse data.

[0047] Figure 8 B Age-related genes in mouse hippocampus.

[0048] Figure 8 C Tipping point DNB computations for mouse at each age.

[0049] Figure 8 D KEGG pathway enrichment of key DNB molecules at the tipping point in mouse.

[0050] Figure 9 A Differential analysis of cell communication in 56-60 year old neuroglia cells compared to adjacent 50-55 and 65-70 year old age groups.

[0051] Figure 9B identified conserved and specific signaling pathways in three age groups, with the vertical dashed line representing 50% of the total information flow.

[0052] Figure 9 Differences in the intensity and quantity of cell communication across the three age groups, and receptor-ligand analysis of cell communication in each age group. Detailed Implementation

[0053] The technical solution of the present invention will be further defined below with reference to specific embodiments, but the scope of protection is not limited to the description made.

[0054] Example 1: Construction of a single-cell atlas of brain aging We collected snRNA-seq data related to human brain health and aging from public databases. For datasets from different sources, we first conducted independent analyses, then selected and filtered samples based on phenotypic and tissue factors to compile the data suitable for this study. The final data comprised cerebral cortex and hippocampal samples from 45 healthy individuals, ranging in age from 29 to 94 years, and included 29 age data points. Figure 1 A). After filtering and quality control, we obtained 17,145 genes and 1,001,088 high-quality cell nuclei for analysis in this study. Figure 1 B).

[0055] Given the complexity of sample sources, we used Harmony for batch processing, and the results showed no specific aggregation among cells from different sources, platforms, and tissues. Figure 1 C). We also statistically analyzed the proportions of each sample source, age, cell type, and tissue type, demonstrating the scientific validity and rationality of our data collection. Specifically, at the cell type level, the three types of glial cells accounted for approximately 53.7%, and neurons accounted for approximately 34.9% (…). Figure 1 D).

[0056] Example 2: Calculation of SASP gene set activity for cell types To further clarify which cell types are more closely related to or more significantly affected by brain aging, we quantified their relationship with aging by assigning aging scores to each cell type. SASPs, including pro-inflammatory cytokines and chemokines, growth regulators, angiogenic factors, and matrix metalloproteinases, are a typical marker of senescent cells and mediate various physiological and pathological mechanisms of senescent cells. We used the AUCell tool to analyze SASPs (… Figure 2 A) Senescence scores were assigned at both the overall cell count and for each cell type. As expected, the older group had significantly higher SASP scores at the overall level (…). Figure 2B). UMAP results showed SASP scores and classical aging pathway (CSP) scores for each cell type, with the three types of glial cells, fibroblasts, and endothelial cells showing higher aging scores (B). Figure 2 (C, D) indicates that their molecular and functional changes during aging are more closely related to aging.

[0057] Example 3: Calculation of transcriptional instability of cell types in senescent tissues With increasing age, cellular transcriptional regulation becomes unstable, and the production of "junk" mRNA inevitably impacts cellular function. Our analysis of the coefficient of variation (CV) revealed that the CV increases with age in almost all cell types, suggesting the widespread presence of gene expression variations during aging. Figure 3 A). Furthermore, we used transcriptional noise computation to quantify transcriptional instability in different cell types. By dividing the cells into multiple age groups for computation and comparison, we found that compared to other cell types, transcriptional noise in microglia, astrocytes, and oligodendrocytes increased significantly with age. Figure 3 (B), which is consistent with previous reports.

[0058] Example 4: Identification of key genes in microglia senescence using dynamic molecular networks snRNA-seq data of microglia were extracted, DNB calculations were performed, the tipping points of microglia were predicted, and the molecular network of key regulatory roles in the tipping point state was plotted. Among them, mTOR showed the most and most important connections. Figure 4 A), and previous studies have shown that inhibiting mTOR can prolong the lifespan of model organisms. Furthermore, we discovered the PADI2 gene, which is one of the Top 10 genes upregulated in microglia during aging. Figure 4 B), consistent with previous results in primates. Further clustering of molecules using hdWGCNA analysis and association with age-related traits revealed a high correlation between PADI2 and module 4, and the age phenotype. Figure 4 (C, D). The above results suggest that mTOR and PADI2 may serve as aging markers indicating the critical state of microglia aging.

[0059] Example 5: Identification of key genes for astrocyte senescence using dynamic molecular networks SnRNA-seq data of astrocyte populations were extracted, and a dynamic molecular network of astrocyte tipping pont was constructed. Figure 5 A) found that JAK1 and ARHGEF11 may be potential key genes in a critical state (Figure 5 B). hdWGCNA analysis showed that JAK1 was located in module 4, which was negatively correlated with age-related phenotypes. Inhibition of JAK1 has been shown to reduce the secretion of SASP in senescent cells, effectively delaying aging. In contrast, ARHGEF11, as a guanine exchange factor for RhoA, regulates cell proliferation, migration, and epithelial-mesenchymal transition. It is significantly up-regulated during aging. It is a member of module 11 in the results of hdWGCNA, which is positively correlated with age (Fig. 4B). Figure 5 C, D), further emphasizing the potential importance of these genes in the senescence tipping point.

[0060] Example 6: Dynamic molecular network identifies key genes of oligodendrocyte senescence snRNA-seq data of oligodendrocyte population was extracted, and it was found that the DNB molecular network of oligodendrocytes was more complex than the previous two (Fig. 5A). Figure 6 A). Among them, KAZN was identified, which encodes kazrin, a protein involved in hemidesmosome assembly and adherens junction, and was significantly up-regulated in the old group (Fig. 5B). Figure 6 B), and clustered into module 2 in the results of hdWGCNA, which is positively correlated with age (Fig. 5C). Figure 6 C, D).

[0061] Example 7: DNB tipping point prediction for each cell type Next, we avoided senescence-related screening and directly applied DNB scores to all cells of all ages to predict specific Tipping point peaks for all cell types (Fig. 6A). Figure 7 ). The results found that the tipping point of microglia, astrocytes and oligodendrocytes appeared earliest and simultaneously in the 56-60 age group. Combined with the previous analysis results of glial cells, we believe that the critical tipping point that leads to irreversible aging of brain tissue during the aging process of human brain health occurs at the age of 56-60, and this may be driven by the functional changes of glial cells.

[0062] Example 8: Verification of the senescence tipping point age in mouse brain tissue To further verify our results and explore the senescence tipping point in other species, we performed the same senescence-related analysis in mouse aging brain tissue samples. The data came from publicly available RNA-seq data of 59 mice aged 3 to 28 months (Fig. 7A). Figure 8 A), and the hippocampus and cortex were also extracted. The results found that the samples of different age groups showed significant distribution differences consistent with the direction of age growth. Age correlation calculation on all genes found that the Padi2 gene was positively correlated with age (Fig. 7B). Figure 8B), which is consistent with the previous results. Applying DNB to each age sample of mice for critical state prediction calculation, it is predicted that there is a significant aging critical state in 18-month-old mice, which is approximately 56 years old in humans ( Figure 8 C), which is consistent with our critical point prediction results in human brain aging samples. Functional analysis of the key DNB molecules of the mouse critical state is also enriched in functions such as aging and immune inflammation pathways ( Figure 8 D).

[0063] Example 9: Dynamic molecular network identifies oligodendrocyte aging key genes To determine the pathways and functions that lead to the critical state of irreversible aging at the age of 56-60 years old, we selected the 56-60 year old sample for cellchat analysis. And set 50-55 years old and 65-70 years old as the critical age before and after the state stage for comparison. The results found that the cell communication of the three types of glial cells at the age of 56-60 years old had significant differences compared with 50-55 years old and 65-70 years old, mainly manifested as a significant decrease in the number and intensity of intercellular communication, especially in microglial cells ( Figure 9 A, C). A total of 25 pathways were identified to be involved in intercellular communication in the three age groups, including 13 conserved pathways (PSAP, SPP1, ANGPT, SEMA3, VEGF, BMP, VISFATIN, PTN, TGFb, EGF, FGF, PDGF and NRG), 6 pathways specific to 50-55 years old (GAS, WNT, CXC3, IL16, EDN and CXCL), 5 pathways existing in both 50-55 years old and 65-70 years old (PARs, NT, GSF, PROS and ANGPL) and 1 pathway existing in both 50-55 years old and 56-60 years old (IGF) ( Figure 9 B). Specifically, the PSAP-GPR37L1 and PSAP-GPR37 communication between the three types of glial cells was significantly weakened, representing the impairment of functions related to nerve protection and nerve repair, which may be an important reason for the functional disorder of the critical state. As a receptor, microglial cells also down-regulate pathways related to cell proliferation, migration, nerve repair and immune regulation (such as CX3C, CSF, IL16, GAS and TGFb). Among them, the weakening of CX3CL1-CX3CR1 and IL34-CSF1R communication between microglial cells and excitatory neurons may lead to interference with the immune environment during the aging process, exacerbating tissue damage and cell apoptosis ( Figure 9 C).

[0064] Finally, it should be pointed out that the above embodiments are only representative examples of the present application. Obviously, the technical solutions of the present application are not limited to the above embodiments, and there can be many variations. All variations that can be directly derived or inferred by those of ordinary skill in the art from the content disclosed herein should be considered as falling within the scope of protection of the present application.

Claims

1. A set of glial cell markers for early warning of a brain senescence critical point, characterized in that, The marker is derived from a dynamic molecular network of glial cells, including gene products expressed in microglia, astrocytes and oligodendrocytes.

2. The glial cell marker panel for early warning of critical points of brain aging according to claim 1, characterized in that, The gene products include products of mTOR, PADI2, JAK1, ARHGEF11 and KAZN genes.

3. Use of the glial cell marker set for early warning of brain aging critical point according to claim 1 in early warning of brain aging critical point.

4. A kit for early warning of a critical point of brain aging, characterized in that, A kit containing the glial cell marker set for early warning of brain aging critical point according to claim 1 or its detection reagent.

5. The kit for early warning of brain aging critical point according to claim 4, characterized in that, The kit for early warning of brain aging critical point is an early warning kit for early stage of brain aging.

6. Use of the kit according to claim 4 in early warning of brain aging critical point.

7. A method of prewarning a critical point of brain aging, characterized by, Early warning of the critical point is achieved by detecting the expression level of the identified aging critical point genes in microglia, astrocytes and oligodendrocytes.

8. The method of claim 7, wherein, The detection of the expression level of the identified aging critical point genes in microglia, astrocytes and oligodendrocytes is the detection of the products of mTOR, PADI2, JAK1, ARHGEF11 and KAZN genes, such as mTOR, PADI2, JAK1, ARHGEF11 and KAZN, in the brain tissue of glial cells. If the gene expression of mTOR, PADI2, JAK1, ARHGEF11 and KAZN is significantly up-regulated, it is in the glial cell-mediated aging critical point.

9. The method of claim 7, wherein, It also includes detection to determine whether the following biological characteristics exist, such as the following biological characteristics exist, it is in the glial cell-mediated aging critical point: (a) The strength of PSAP-GPR37L1 and PSAP-GPR37 ligand-receptor communication between glial cells decreases; (b) The activity of CX3CL1-CX3CR1 and IL34-CSF1R signaling pathways between neurons and microglia decreases.

Images