Use of a ZBED1 expression inhibitor in the preparation of a drug for treating cytarabine-resistant leukemia

Abstract

The present invention relates to the field of bioengineering technology, and particularly relates to the application of a ZBED1 expression inhibitor in the preparation of drugs for treating cytarabine-resistant leukemia. The present invention provides the application of a ZBED1 expression inhibitor in the preparation of drugs for treating cytarabine-resistant leukemia, and the drugs for treating cytarabine-resistant leukemia include a ZBED1 expression inhibitor. The present invention first establishes the key role of ZBED1 in leukemia drug resistance, and discovers the active ingredient of traditional Chinese medicine diosgenin that selectively inhibits the function of ZBED1. This ingredient can not only act synergistically with cytarabine to reverse the cytarabine resistance of leukemia, but also provide a new target site and molecular marker for clinically predicting and improving the treatment effect of patients. This innovative strategy brings new hope for leukemia treatment, especially in the problem of cytarabine resistance, providing a more effective solution.

Description

Technical Field

[0001] The present invention relates to the field of bioengineering technology, and particularly to the application of a ZBED1 expression inhibitor in the preparation of a drug for treating cytarabine-resistant leukemia. Background Art

[0002] Acute myeloid leukemia (AML) is a hematological malignancy mainly characterized by the clonal proliferation of myeloid primitive cells, with high heterogeneity. In the past few decades, its incidence has been increasing continuously. Among all acute leukemia patients, AML patients have the shortest survival period, and the five-year survival rate is less than 30%. Acute myeloid leukemia is a highly heterogeneous disease with a complex pathogenesis and limited treatment effects. Therefore, it is crucial to search for and discover new AML molecular targets, which is of great significance for improving the clinical prognosis of patients and increasing the survival rate.

[0003] The tumor immune microenvironment of AML plays an important role in tumor progression. Recent studies have shown that in the treatment of AML, the immune balance of the bone marrow microenvironment plays a key role. Tumor cells can hijack and reshape the bone marrow microenvironment, making it an environment that supports tumor growth, thereby helping AML bypass the attack of the immune system and resistance to treatment. In this altered bone marrow microenvironment, the stem cells and initiating cells of AML can maintain their regenerative ability, and minimal residual disease can be effectively incubated, ultimately leading to the recurrence of AML. In recent years, studies have shown that improving the understanding of the immune microenvironment of the bone marrow microenvironment will inspire the formulation of new treatment strategies, thereby more effectively controlling the development of AML.

[0004] The recurrence of AML is a major challenge in the treatment process, often leading to disease progression and deterioration of the patient's prognosis. Recurrence is usually closely related to drug resistance. Cytarabine is currently widely used in AML and can be used as an effective single therapy and the backbone of combination chemotherapy regimens. The formation of cytarabine resistance reduces the effectiveness of traditional treatment methods, limits treatment options, and increases the difficulty of treatment. Nowadays, more and more studies have shown that traditional herbal medicines (THMs) cover a variety of natural products and are considered an important new source of candidate small molecule drugs due to their diverse biological activities, and they play an important role in reversing cancer chemotherapy resistance. Summary of the Invention

[0005] In order to solve the above technical problems, the present invention provides the application of a ZBED1 expression inhibitor in the preparation of a drug for treating cytarabine-resistant leukemia. The present invention discovers that diosgenin, a traditional Chinese medicine active ingredient with selective inhibition of ZBED1 function, can not only act synergistically with cytarabine to reverse the cytarabine resistance of leukemia, but also provide new target sites and molecular markers for clinically predicting and improving the treatment effect of patients.

[0006] To this end, the present invention provides the following technical solutions:

[0007] In a first aspect, in an alternative embodiment, the present invention provides an application of a ZBED1 expression inhibitor in the preparation of a drug for treating cytarabine-resistant leukemia, and the drug for treating cytarabine-resistant leukemia includes a ZBED1 expression inhibitor.

[0008] Furthermore, the ZBED1 expression inhibitor is diosgenin.

[0009] In a second aspect, in an alternative embodiment, the present invention provides a pharmaceutical composition for inhibiting leukemia, which includes the above-mentioned ZBED1 expression inhibitor and cytarabine.

[0010] Furthermore, the pharmaceutical composition for inhibiting leukemia includes diosgenin and cytarabine.

[0011] Preferably, in the pharmaceutical composition for inhibiting leukemia, the concentration of cytarabine is 0.01 - 2 μM, and the concentration of diosgenin is 2 - 64 μM. Further preferably, in the pharmaceutical composition for inhibiting leukemia, the concentration of cytarabine is 0.1 μM, and the concentration of diosgenin is 16 μM.

[0012] Preferably, the leukemia inhibitor further includes excipients.

[0013] Furthermore, the leukemia inhibitor is an oral preparation or an injection preparation. The oral preparation includes capsules, tablets or granules.

[0014] In the present invention, the leukemia inhibitor can be any conventional oral preparation or injection preparation that is pharmaceutically acceptable and is made by a conventional preparation process. Oral preparations such as capsules, tablets, granules or liquids. To enable the above oral preparations to be realized, pharmaceutically acceptable excipients need to be added during the preparation process. The pharmaceutically acceptable excipients are fillers, disintegrants, lubricants, suspending agents, binders or sweeteners, etc.

[0015] The fillers include at least one of starch, lactose, microcrystalline cellulose or sucrose; the disintegrants include at least one of starch, sodium carboxymethyl starch, pregelatinized starch, low-substituted hydroxypropyl cellulose or cross-linked sodium carboxymethyl cellulose; the lubricants include at least one of magnesium stearate or silica, sodium dodecyl sulfate; the suspending agents include at least one of polyvinylpyrrolidone, sucrose or hydroxypropyl methylcellulose; the binders include at least one of hydroxypropyl methylcellulose, starch paste or polyvinylpyrrolidone; the sweeteners are at least one of sodium saccharin, glycyrrhetinic acid, sucrose, aspartame or sodium cyclamate.

[0016] To enable the above injection preparation to be realized, pharmaceutically acceptable excipients need to be added during the preparation process, including water for injection and sodium chloride solution for injection.

[0017] Compared with the prior art, the present invention has one of the following beneficial effects:

[0018] 1. The present invention for the first time established the key role of ZBED1 in leukemia drug resistance and discovered the active ingredient of traditional Chinese medicine diosgenin that selectively inhibits the function of ZBED1. This ingredient can not only synergistically act with cytarabine to reverse the cytarabine resistance of leukemia, but also provide a new target site and molecular marker for clinically predicting and improving the treatment effect of patients. This innovative strategy brings new hope for leukemia treatment, especially in the problem of cytarabine resistance, providing a more effective solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is the single-cell analysis result in Example 1 of the present invention, revealing the characteristics of the AML microenvironment, where Figure 1 A shows the cell clustering results at different resolutions. Each node represents a cluster, and the connection lines between the nodes represent the clustering relationship when the resolution changes, with the resolution changing from low to high in sequence, Figure 1 B shows 11 different cell types identified using specific markers. The cell types include: AML progenitor, B cell, Dendritic cell, Erythroid cell, GMP (Granulocyte-Monocyte Progenitor), HSC (Hematopoietic Stem Cell), Mono / Mac (Monocyte / Macrophage), Neutrophil, NK cell (Natural Killer cell), Plasma and T cell, Figure 1 C shows the sources of different cells. Each point represents a single cell, and different colors represent different source samples, reflecting the diversity of the AML microenvironment, Figure 1 D shows the gene expression levels of specific markers for each cell type. Each column represents a cell type, and each row represents a specific marker gene. The depth of the color indicates the level of gene expression, Figure 1 E shows the distribution ratios of different cell types in each sample. Each bar represents a sample, and different colors represent the ratios of different cell types, providing an overview of the cell composition in the AML microenvironment;

[0020] Figure 2Schematic diagram of the results of studying the heterogeneity of AML cells in Example 1 of the present invention, where Figure 2 A is a marker gene feature map of different types of acute myeloid leukemia progenitors, showing the expression density of multiple key genes (such as DEFA3, MALAT1, CST3, RPS 12, S100A8, H3F3A, LGALS1, RPL9) in the UMAP space. Figure 2 B shows the identification of 4 major subsets of acute myeloid leukemia progenitors, namely AML.progenitor.1, AML.progenitor.2, AML.progenitor.3, and AML.progenitor.4. Figure 2 C is the expression heat map and GO analysis of DEGs in 4 subsets of acute myeloid leukemia progenitors. The heat map shows the specific gene expression of each subset, and the right side shows the GO functional annotations related to these genes. Figure 2 D shows the differential genes between AML progenitors that relapse after chemotherapy and AML progenitors in remission. Figure 2 E is the distribution ratio of different cell types in each sample, further showing the proportion of various subsets of acute myeloid leukemia progenitors in different samples. Different colors represent different cell subsets.

[0021] Figure 3 Characteristics of T lymphocyte subsets in the AML microenvironment in Example 1 of the present invention, where Figure 3 A shows the clustering results of all cells at different resolutions. Figure 3 B shows the t-SNE and UMAP maps of different cell types and different clinical types. Figure 3 C shows the expression levels of marker genes of each T cell subset. The depth of color indicates the high or low gene expression. Figure 3 D shows the distribution ratio of different cell types in each sample. Figure 3 E shows the differential genes between T cells in remission after chemotherapy and T cells not in remission after chemotherapy.

[0022] Figure 4 Detailed explanation of the communication network between cell subsets in Example 1 of the present invention, where Figure 4 A - Figure 4 B shows the communication frequency and degree between different cell subsets. Figure 4 A shows the interaction frequency between different cell types. Figure 4B shows the interaction strength between cell types. Each node represents a cell type, and the connections between nodes represent the communication relationships between cells. The thickness and color of the lines indicate the frequency or strength of communication. Figure 4 C is a graphical illustration of the main signal senders (sources) and receivers (targets), showing the signal-sending and signal-receiving capabilities of each cell type. The horizontal axis represents the strength of the transmitted signal, the vertical axis represents the strength of the received signal, and the size of the points represents the number of cells. Figure 4 D reveals the main signal-sending and signal-receiving trends, showing the signal-sending and signal-receiving patterns of each cell type. The left side shows the signal-sending pattern, and the right side shows the signal-receiving pattern. The color of each matrix block represents the relative signal strength, and the darker the color, the stronger the signal.

[0023] Figure 5 This is a schematic diagram for comprehensively examining the cell communication network in Example 1 of the present invention. Among them, Figure 5 A and Figure 5 E are used to determine the number of output ( Figure 5 A) or input ( Figure 5 E) signal patterns using cophenetic and silhouette metrics. Figure 5 B and Figure 5 F are used to identify various cell subsets and their respective output ( Figure 5 B) or input ( Figure 5 F) signal patterns. Figure 5 C and Figure 5 G show the distribution of output ( Figure 5 C) or input ( Figure 5 G) signal patterns emitted from signal-sending cells. On the left side of the figure is the cell population, in the middle is the identified signal pattern, and on the right side is the specific signal pathway, further showing the overall flow from the cell population to the signal pattern and then to the signal pathway. Figure 5 D and Figure 5 H show the changes in output ( Figure 5 D) or input ( Figure 5 H) signals in the key path, further showing the output signal patterns of different cell types in each key signal pathway. Different colors represent different cell types, and the size of the points represents the strength of signal contribution.

[0024] Figure 6 This is a schematic diagram of the analysis results of the differences in cell - cell communication between relapsed and remitted AML after chemotherapy in Example 1 of the present invention. Among them, Figure 6 A shows the cell - cell communication network in relapsed and remitted AML tissues after chemotherapy. Figure 6To compare the changes in the quantity and intensity of cell - cell interactions in relapsed AML versus remission AML after chemotherapy, in the figure, enhanced relative interactions in relapsed AML are shown in red, and weakened relative interactions in relapsed AML are shown in blue. Figure 6 C is to compare the differences in signal intensity and quantity among different clinical types. Figure 6 D shows the differences in tissue - tissue interaction frequency and intensity between relapsed AML and remission AML after chemotherapy. The color depth of each cell in the figure represents the frequency and intensity of interactions between different cell types. Dark colors indicate high interaction intensity, and light colors indicate low interaction intensity, providing a detailed comparison of the cell - cell communication patterns in the relapse and remission states. Figure 6 E is to compare the dominant signal emitters and receivers in the tissues of relapsed AML versus remission AML after chemotherapy. Figure 6 F is to compare the expression of key signal molecules and pathways between relapsed AML and remission AML tissues after chemotherapy. The figure shows the differences in the expression levels of important signal molecules and signal pathways in the relapse and remission states, and the relative expression levels of each molecule or pathway in the two states are shown by color and height. Figure 6 G shows the analysis of outgoing signals among different cell subsets, further demonstrating the patterns and changes in the outgoing signals of different cell subsets in the relapse and remission states. Figure 6 H shows the analysis of received signals among different cell subsets. Figure 6 I shows the overall signal transmission analysis of different cell subsets;

[0025] Figure 7 This is a schematic diagram of the important regulatory role of ZBED1 in AML in Example 1 of the present invention. Among them, Figure 7 A shows the intersection genes of SUMOylation - related genes, AML relapse genes, and AML prognosis genes. Figure 7 B - Figure 7 C shows the Kaplan - Meier survival curves of ZBED1 in TCGA survival data ( Figure 7 B) and GSE71014 survival data ( Figure 7 C). Figure 7 D is a UMAP plot of ZBED1 expression, showing the expression levels of ZBED1 in different cell clusters. Each point represents a single cell, and different colors represent different expression levels. Figure 7 E is a volcano plot of differentially expressed genes relative to the expression level of ZBED1. Figure 7 F - Figure 7 H is a schematic diagram of the results of analyzing differentially expressed genes (DEGs) using GSEA, GO, and KEGG;

[0026] Figure 8Schematic diagram of the results of ZBED1 remodeling the tumor microenvironment (TME) in Example 1 of the present invention, where Figure 8 A shows the immune infiltration of each sample in the TCGA-AML database, Figure 8 B shows the distribution of various cell types in all AML samples, Figure 8 C shows the correlation analysis of ZBED1 with cells in the immune microenvironment, further showing the correlation coefficient between ZBED1 expression and the abundance of various cell types in the bone marrow microenvironment, Figure 8 D shows the Kaplan-Meier survival curve presenting the prognosis analysis results of patients with high expression of CD8+ T cells and low expression of CD8+ T cells;

[0027] Figure 9 Schematic diagram of the association results between the increased expression of ZBED1 and enhanced drug resistance in AML in Example 1 of the present invention, where Figure 9 A and Figure 9 B are used to study the relationship between ZBED1 expression and the half-maximal inhibitory concentration (IC 50 ) of various drugs using the GSE71014 database (A) and the TCGA-AML (B) database, Figure 9 C shows the study of the interaction between ZBED1 protein and several commonly used AML treatment drugs (such as Cytarabine, Doxorubicin, Mitoxantrone, and Fludarabine) through molecular docking technology. The molecular docking results show the possible binding sites of ZBED1;

[0028] Figure 10 Schematic diagram of ZBED1 promoting CD8+ T cell exhaustion in Example 1 of the present invention, where Figure 10 A- Figure 10 C show the Celltype ( Figure 10 A), pseudotime plot ( Figure 10 B), and ZBED1 expression level ( Figure 10 C) of CD8+ T cells, Figure 10 D shows the distribution of Marker genes of different types of T cells, Figure 10 E- Figure 10 F show the CD8+ T cell subsets ( Figure 10 E), pseudotime analysis ( Figure 10 F), Figure 10 G- Figure 10 H show the dynamic changes of cell type ( Figure 10 G) and ZBED1 expression level ( Figure 10 H), Figure 10 I shows the Marker genes related to developmental time and cell subsets;

[0029] Figure 11 Schematic diagram of the results showing the important role of ZBED1 in drug-resistant AML cells in Example 2 and Example 3 of the present invention, wherein Figure 11 A shows the results of comparing the expression of ZBED1 in AML cell lines and normal control HS-5 cells using qRT-PCR; Figure 11 B- Figure 11 D is for qRT-PCR ( Figure 11 B) and Western blot ( Figure 11 C- Figure 11 D) showing the results of detecting the expression levels of ZBED1 in MOLM13, MV4-11, MOLM13 / R, and MV4-11 / R cells respectively, Figure 11 E shows the localization of ZBED1 in drug-resistant cells and corresponding parental cell lines by immunofluorescence, with the nucleus stained with DAPI, scale bar 20 μm, Figure 11 F shows the results of verifying the knockdown efficiency of ZBED1 using qRT-PCR in MOLM13 / R and MV4-11 / R cells, Figure 11 G- Figure 11 I is for using CCK-8 ( Figure 11 G) and EDU ( Figure 11 H- Figure 11 I) to detect the proliferation ability of cells after knocking out ZBED1, Figure 11 J- Figure 11 M is for using WB ( Figure 11 J- Figure 11 K), flow cytometry ( Figure 11 L), Caspase3 activity assay ( Figure 11 M) to detect the apoptosis of cells, Figure 11 N- Figure 11 O is for using flow cytometry to detect the cell cycle of cells, Figure 11 P shows the tumor entity of the dissected mouse, Figure 11 Q- Figure 11 R shows the tumor volume and weight of each group of mice, Figure 11 Q represents the tumor growth kinetics of NSG mice in the sh-ZBED1 group and shRNA-NC group shown by the growth curve, Figure 11 R represents the comparative analysis of the tumor masses of nude mice in the sh-ZBED1 group and shRNA-NC group;

[0030] Figure 12 Schematic diagram of the results showing that ZBED1 induces osteolysis in AML in Example 1 of the present invention, wherein, Figure 12 A- Figure 12 E are the anteroposterior and lateral three-dimensional reconstructions of the femur in the sh-NC group ( Figure 12 A), the coronal cross-sectional view of the whole femur (Figure 12 B), Cross-sectional view and three-dimensional reconstruction of the distal femur near the metaphysis ( Figure 12 C), Three-dimensional reconstruction of the cortical bone in the middle of the femur ( Figure 12 D), Coronal view of the proximal tibia ( Figure 12 E), Figure 12 F- Figure 12 J is the anteroposterior and lateral three-dimensional reconstruction views of the femur in the sh-ZBED1 group ( Figure 12 F), Coronal cross-sectional view of the whole femur ( Figure 12 G), Cross-sectional view and three-dimensional reconstruction of the distal femur near the metaphysis ( Figure 12 H), Three-dimensional reconstruction of the cortical bone in the middle of the femur ( Figure 12 I), Coronal view of the proximal tibia ( Figure 12 J), Figure 12 K- Figure 12 P shows the changes in trabecular bone microstructure parameters, and BMD represents bone mineral density ( Figure 12 K); BS represents bone surface area ( Figure 12 L), BV / TV represents trabecular bone volume fraction ( Figure 12 M), Tb.Th represents trabecular thickness ( Figure 12 N), Tb.N represents trabecular number ( Figure 12 O), Tb.Sp represents trabecular spacing ( Figure 12 P), Figure 12 Q- Figure 12 T shows the changes in cortical bone microstructure parameters, and Tt.Ar represents the total cross-sectional area within the periosteal envelope ( Figure 12 Q), Ct.Ar represents cortical bone area ( Figure 12 R), Ct.Ar / Tt.Ar represents cortical bone area fraction ( Figure 12 S), Ct.Th represents average cortical bone thickness ( Figure 12 T), *P < 0.05, **P < 0.01, ***P < 0.001;

[0031] Figure 13 This is a schematic diagram of the results of the inhibition of the malignant phenotype of cytarabine-resistant AML cells by diosgenin and the increase in their sensitivity to cytarabine in Examples 1-3 of the present invention. Among them, Figure 13 A shows the results of WB detection of the effects of ursolic acid (Mairin), astilbin, diosgenin, obacunone, and isoimperatorin on ZBED1 expression, Figure 13 B shows the molecular docking display of diosgenin and ZBED1, Figure 13 C- Figure 13 D shows the effects of diosgenin at different concentrations on the apoptosis rate detected by flow cytometry,Figure 13 E is a photo of the tumor entity of the dissected mouse, Figure 13 F shows the tumor growth kinetic curves of different groups of NSG mice, Figure 13 G shows the comparative analysis results of the tumor masses of different groups of NSG nude mice, Figure 13 H is the confidence interval graph of the synergistic drug combination (dioscin and cytarabine), Figure 13 I- Figure 13 L is the ZIP ( Figure 13 I), Bliss ( Figure 13 J), HSA ( Figure 13 K), and Loewe ( Figure 13 L) synergy graphs of the combined treatment of cytarabine and dioscin on MOLM13 / R cells. The synergy scores were calculated using the SynergyFinder R package. Positive or negative values indicate synergistic or antagonistic effects respectively, Figure 13 M is the measurement result of the synergistic effect with the dose combination of 0.1 μM cytarabine and 16 μM dioscin, Figure 13 N- Figure 13 O is the result of the flow cytometry experiment based on the recommended concentration of the metering indicator in Figure M;

[0032] Figure 14 This is the schematic diagram of the mechanism of the synergistic treatment of leukemia by the traditional Chinese medicine monomer (diosgenin) targeting and inhibiting ZBED1 and cytarabine in the present invention. Detailed implementation mode

[0033] In order to make the objectives, technical solutions and advantages of the present invention clearer, the following further elaborates on the present invention in combination with specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0034] The experimental methods used in the following embodiments are all conventional methods unless otherwise specified.

[0035] The materials, reagents, etc. used in the following embodiments can be obtained from commercial channels unless otherwise specified.

[0036] For the schematic diagram of the mechanism of the synergistic treatment of leukemia by the traditional Chinese medicine monomer (diosgenin) targeting and inhibiting ZBED1 and cytarabine in the present invention, see Figure 14 .

[0037] Example 1

[0038] 1. Data source and processing

[0039] Three datasets (GSE235063, GSE116256, and GSE130756) were screened from the GEO database. The RNA-seq data of TCGA-AML and GSE71014 and their corresponding clinical records were integrated. Samples without complete survival or clinical information were excluded. The scRNA-seq data were processed in detail using the Seurat R package (version 4.3.0). Potential doublets were excluded using DoubletFinder, and cells with too high rRNA and mitochondrial proportions were filtered. All gene data were normalized and dimensionally reduced by PCA. Then, the "Harmony" toolkit was used to adjust the differences between batches. Highly variable genes were screened using the FindVariableFeatures function. On the corrected data, further dimensional reduction was performed using the UMAP and tSNE methods. Marker genes were identified through the Cell Marker database. The results are shown in Figure 1 , where Figure 1 A shows the cell clustering results at different resolutions. Each node represents a cluster, and the connections between nodes represent the clustering relationships when the resolution changes, with the resolution changing from low to high in sequence, Figure 1 B shows 11 different cell types identified using specific markers. The cell types include: AML progenitor, B cell, Dendritic cell, Erythroid cell, GMP (Granulocyte-Monocyte Progenitor), HSC (Hematopoietic Stem Cell), Mono / Mac (Monocyte / Macrophage), Neutrophil, NK cell (Natural Killer cell), Plasma, and T cell, Figure 1 C shows the origin of different cells. Each point represents a single cell, and different colors represent different source samples, reflecting the diversity of the AML microenvironment, Figure 1 D shows the gene expression levels of specific marker genes for each cell type. Each column represents a cell type, and each row represents a specific marker gene. The color intensity indicates the level of gene expression, Figure 1 E shows the distribution proportions of different cell types in each sample. Each bar represents a sample, and different colors represent the proportions of different cell types, providing an overview of the cell composition in the AML microenvironment.

[0040] Then, Monocle2 and Monocle3 were used to visualize the pseudotemporal developmental trajectories of cells, and the plotpseudotime heatmap function was used to display the heatmap of the dynamic changes in gene expression. The results are shown inFigure 3 and Figure 10 , where Figure 3 A shows the clustering results of all cells at different resolutions, Figure 3 B shows the t-SNE and UMAP plots of different cell types and different clinical types, Figure 3 C shows the expression levels of marker genes of each T cell subset, and the color depth indicates the high or low gene expression, Figure 3 D shows the distribution proportions of different cell types in each sample, Figure 3 E shows the differential genes between T cells that are in remission after chemotherapy and those that are not in remission after chemotherapy; Figure 10 A- Figure 10 C shows the Celltype of CD8+ T cells ( Figure 10 A), the pseudotime plot ( Figure 10 B), the expression level of ZBED1 ( Figure 10 C), Figure 10 D shows the distribution of Marker genes of different types of T cells, Figure 10 E- Figure 10 F shows the CD8+ T cell subset ( Figure 10 E), the pseudotime analysis ( Figure 10 F), Figure 10 G- Figure 10 H shows the Celltype ( Figure 10 G) and the dynamic changes of the expression level of ZBED1 ( Figure 10 H), Figure 10 I shows the Marker genes related to the developmental time and cell subsets.

[0041] 2. Intercellular communication analysis

[0042] Using the RNA expression data and cell annotations, a CellChat object was initialized through the "create Cell Chat" function. Then, the interaction database containing "Secreted Signaling" was selected for signal network analysis. Through the "compute Commun Prob" function, the communication probability between cells was estimated. When using the "selectK" function, the global communication mode was used, and the nPatterns parameter was set to 2 for both the input and output of the information flow. The results are shown in Figures 4 - 6 , where Figure 4 A- Figure 4 B shows the communication frequency and degree between different cell subsets, Figure 4 A shows the interaction frequency between each cell type, Figure 4B shows the interaction strength between cell types. Each node represents a cell type, the connections between nodes represent the communication relationships between cells, and the thickness and color of the lines indicate the frequency or strength of communication. Figure 4 C is a graphical illustration of the main signal senders (sources) and receivers (targets), showing the signal-sending and signal-receiving capabilities of each cell type. The horizontal axis represents the strength of the sent signal, the vertical axis represents the strength of the received signal, and the size of the dots represents the number of cells. Figure 4 D reveals the main signal-sending and signal-receiving trends, showing the signal-sending and signal-receiving patterns of each cell type. The left side shows the signal-sending pattern, and the right side shows the signal-receiving pattern. The color of each matrix block represents the relative signal strength, and the darker the color, the stronger the signal. Figure 5 A and Figure 5 E is to determine the number of output ( Figure 5 A) or input ( Figure 5 E) signal patterns using cophenetic and silhouette metrics. Figure 5 B and Figure 5 F is to identify various cell subpopulations and their respective output ( Figure 5 B) or input ( Figure 5 F) signal patterns. Figure 5 C and Figure 5 G shows the distribution of output ( Figure 5 C) or input ( Figure 5 G) signal patterns from signal-sending cells. On the left side of the figure is the cell population, in the middle is the identified signal pattern, and on the right side is the specific signal pathway, further showing the overall flow from the cell population to the signal pattern and then to the signal pathway. Figure 5 D and Figure 5 H shows the changes in output ( Figure 5 D) or input ( Figure 5 H) signals in the key path, further showing the output signal patterns of different cell types in each key signal pathway. Different colors represent different cell types, and the size of the dots represents the strength of the signal contribution. Figure 6 A shows the intercellular communication network in relapsed and remitted AML tissues after chemotherapy. Figure 6 B is to compare the changes in the number and strength of intercellular interactions in relapsed AML relative to remitted AML after chemotherapy. In the figure, red indicates an increase in relative interactions in relapsed AML, and blue indicates a decrease in relative interactions in relapsed AML. Figure 6 C is to compare the differences in signal strength and quantity among different clinical types. Figure 6D shows the differences in the frequency and intensity of tissue - to - tissue interactions between relapsed AML and remission AML after chemotherapy. The color depth of each cell in the figure represents the frequency and intensity of interactions between different cell types. Dark colors indicate high interaction intensity, and light colors indicate low interaction intensity, providing a detailed comparison of the inter - cellular communication patterns in the relapse and remission states. Figure 6 E shows the dominant signal emitters and receivers in the tissues of relapsed AML compared to remission AML after chemotherapy. Figure 6 F compares the expression of key signaling molecules and pathways between the tissues of relapsed AML and remission AML after chemotherapy. The figure shows the differences in the expression levels of important signaling molecules and signaling pathways in the relapse and remission states, and the relative expression levels of each molecule or pathway in the two states are shown by color and height. Figure 6 G shows the analysis of outgoing signals between different cell subsets, further demonstrating the patterns and changes in the outgoing signals of different cell subsets in the relapse and remission states. Figure 6 H shows the analysis of received signals between different cell subsets. Figure 6 I shows the overall signal transduction analysis of different cell subsets.

[0043] 3. Identification and survival analysis of SUMOylation - related genes

[0044] Several candidate genes were selected from the intersection of SUMO - related genes and differentially expressed genes (DEGs). By analyzing the TCGA - AML and GSE71014 datasets, survival analysis of individual genes was performed using the R package "survival". According to the cutoff value of each gene, patients were divided into high - expression and low - expression groups. The Log - Rank test determined the statistical significance of the survival curves, and curves with P < 0.05 were considered different. The results are shown in Figure 7 A - Figure 7 E, where Figure 7 A shows the intersection genes of SUMOylation - related genes, AML relapse genes, and AML prognosis genes. Figure 7 B - Figure 7 C shows the Kaplan - Meier survival curves of ZBED1 in the TCGA survival data ( Figure 7 B) and GSE71014 survival data ( Figure 7 C). Figure 7 D is a UMAP plot of ZBED1 expression, showing the expression levels of ZBED1 in different cell clusters. Each point represents a single cell, and different colors represent different expression levels. Figure 7 E is a volcano plot of differentially expressed genes relative to the expression level of ZBED1.

[0045] 4. Enrichment analysis

[0046] Differentially expressed genes (DEGs) were identified by the FindMarkers function, with the thresholds of Padj.P < 0.05 and log 2 FC > 0.25. GO and KEGG analyses were performed using the cluster Profiler package, and GSEA analysis was performed using their respective packages. The results are shown in Figure 2 、 Figure 7 F- Figure 7 H, where Figure 2 A is a marker gene feature map of different types of acute myeloid leukemia progenitor cells (AMLprogenitors), showing the expression density of multiple key genes (such as DEFA3, MALAT1, CST3, RPS12, S100A8, H3F3A, LGALS1, RPL9) in the UMAP space. Figure 2 B shows the identification of 4 major subsets of acute myeloid leukemia progenitor cells (AML progenitors), namely AML.progenitor.1, AML.progenitor.2, AML.progenitor.3, and AML.progenitor.4. Figure 2 C is the expression heatmap and GO analysis of DEGs in 4 subsets of acute myeloid leukemia progenitor cells (AML progenitors). The heatmap shows the specific gene expression of each subset, and the right side shows the GO functional annotations related to these genes. Figure 2 D shows the differential genes between AMLprogenitors with relapse after chemotherapy and AML progenitors in remission. Figure 2 E is the distribution ratio of different cell types in each sample, further showing the proportion of various subsets of acute myeloid leukemia progenitor cells (AML progenitors) in different samples. Different colors represent different cell subsets. Figure 7 F- Figure 7 H is a schematic diagram of the results of analyzing differentially expressed genes (DEGs) using GSEA, GO, and KEGG.

[0047] 5. Evaluation of the distribution of immune cell subtypes

[0048] The infiltration of cell subtypes in the TCGA-AML clinical cohort was analyzed using CibersortX and the average expression values of signature genes. The association between ZBED1 and different cell populations was revealed by Spearman correlation analysis. The results are shown in Figure 8 ,where Figure 8 A shows the immune infiltration of each sample in the TCGA-AML database. Figure 8 B shows the distribution of various cell types in all AML samples.Figure 8 C shows the correlation analysis of ZBED1 with cells in the immune microenvironment, and further shows the correlation coefficients between ZBED1 expression and the abundances of various cell types in the bone marrow microenvironment. Figure 8 D shows the Kaplan-Meier survival curves presenting the prognostic analysis results of patients with high and low expression of CD8+ T cells.

[0049] 6. Chemotherapy drug sensitivity analysis

[0050] Using the R package oncoPredict, the IC 50 values of 198 drugs were obtained from the Genomics of Drug Sensitivity in Cancer (GDSC; https: / / www.cancerrxgene.org / ) and the Cancer Cell Line Encyclopedia (CCLE, https: / / sites.broadinstitute.org / ccle / ). The correlation between the drug IC 50 values and the risk score was explored by Spearman analysis to identify relevant drugs. Subsequently, the IC 50 differences between the high and low expression groups were compared, with a focus on analyzing drugs with an absolute correlation value exceeding 0.2. The correlation analysis dot plots and lollipop plots in R language's ggplot2 were used to visualize the results, and the results are shown in Figure 9 A- Figure 9 B.

[0051] 7. Molecular docking analysis

[0052] The three-dimensional structures of the ZBED1 protein and various drug molecules were obtained from the PubChem database. The docking analysis preparation of the protein and drug structures included removing water molecules and adding non-polar hydrogen atoms using PyMOL and Auto Dock4. Subsequently, appropriate docking parameters and grid box sizes were established. Molecular docking simulations were performed using AutoDock4, and the results were analyzed and visualized using PyMOL. The results are shown in Figure 9 C and 13B. The docking results showed that a binding energy lower than -2.0 kcal / mol indicated significant molecular interactions.

[0053] 8. Screening of the best traditional Chinese medicine active ingredients

[0054] The Coremine Medical database was used to collect gene-related information. The Coremine Medical database is an open retrieval platform that integrates comprehensive medical information such as traditional Chinese herbs, gene ontology, protein expression, and anatomy. The selected key genes were respectively mapped into Coremine Medical (http: / / www.coremine.com / ), and traditional Chinese medicines related to the key genes were screened out. A P value < 0.05 was considered statistically significant. Traditional Chinese medicines (TCMs) significantly related to the key gene were determined, including Evodia rutaecarpa, Jujube Fruit, Rheum palmatum, and Smilax glabra, etc. (P < 0.05). The traditional Chinese medicine systems pharmacology database and analysis platform (TCMSP) was used to screen the main active ingredients of high-frequency TCMs, requiring an oral bioavailability (OB) ≥ 30% and a drug likeness (DL) ≥ 0.18. Subsequently, molecular docking was performed to verify the binding energy of all TCM monomers to ZBED1, and 20 different docking simulations were carried out for each monomer. The five monomers with the lowest binding energy were identified as Ursolic acid (Mairin), Astilbin, Diosgenin, Obacunone, and Isoimperatorin. Then, further exploration was carried out to determine whether the above-mentioned traditional Chinese medicines had an effect on ZBED1 when treating MV4-11 / R and MOLM13 / R cells for 48 h at the IC 50 value concentration. The results are shown in Figure 13 A- Figure 13 B. It was found that among the above 5 drugs, only Astilbin, Diosgenin, Diosgenin, and Isoimperatorin had an inhibitory effect on ZBED1, and Diosgenin had the strongest inhibitory effect (see Figure 13 A). The results of molecular docking showed that Diosgenin might be able to directly bind to ZBED1 and form a hydrogen bond connection with the R33 site on the ZBED1 protein chain (see Figure 13 B).

[0055] Example 2

[0056] In vitro experiment

[0057] 9. Quantitative real-time PCR analysis

[0058] Total RNA was isolated using TRIzol reagent (Invitrogen), and then reverse transcription was performed using the Revert Aid First Strand cDNA Synthesis Kit (Thermo Fisher Scientific, USA). SYBR-Green assay mixture was prepared, and quantitative PCR analysis was carried out using a Roche Light Cycler 480 II instrument. To ensure the reproducibility of the experiment, the number of replicate wells for each sample was three. The gene expression levels were determined by comparing 2 -ΔΔCt methods, and the results are shown in Figure 11 A, Figure 11 B and Figure 11 F.

[0059] 10. Small interfering RNA (siRNA), short hairpin RNA (shRNA), and plasmid transfection

[0060] Three siRNAs targeting ZBED1 and one shRNA sequence were designed using the online siRNA and shRNA design tools on the Genepharma website. In the shRNA experiment, the designed shRNA sequence was cloned into the expression vector pLKO.1. Subsequently, the pLKO.1 shRNA plasmid, packaging plasmid, and envelope plasmid were co-transfected into HEK293T cells to produce and collect lentiviral particles. Leukemia cell lines in the logarithmic growth phase were seeded into 6-well plates until they reached 70%-80% confluence, at which time the collected viral particles were used for cell transfection, and Polybrene was added to improve the transfection efficiency. Twelve hours after transfection, the cells were maintained in normal medium for 24 hours and then harvested. Transfection success was verified by qRT-PCR. For the siRNA experiment, when the cell density reached 70%-80%, cells in the logarithmic growth phase were seeded into 6-well plates and transfected using Lipofectamine 3000 transfection reagent. Six hours after transfection, the medium was changed to normal medium, and the cells were cultured for another 24 hours before being collected for subsequent experiments. Cells were transfected with the designated plasmid using Lipofectamine 2000 (Invitrogen, Life Technologies) according to the manufacturer's protocol. The transfection efficiency was optimized by adjusting the ratio of plasmid to Lipofectamine and the incubation time, and the results are shown in Figure 11 F.

[0061] 11. EdU staining assay

[0062] The cells were incubated with 10 μM EdU staining solution (provided by Apexbio). After incubation, the cells were treated with 500 μl of Click reaction mixture containing Cy3 azide for 30 minutes, and the reaction was carried out at room temperature in the dark. After the reaction, the cells were rinsed with PBS and then stained with 1 ml of Hoechst 33342 dye at a final concentration of 5 μg / ml, and incubated at room temperature for another 30 minutes in the dark. Subsequently, the stained specimens were observed using a fluorescence microscope. See Figure 11 H- Figure 11 I。

[0063] 12. CCK-8 Assay

[0064] Cell viability was evaluated using the Cell Counting Kit-8 (CCK-8, Sigma-Aldrich, St. Louis, MO). According to the kit's instructions, the cell lines were seeded into 96-well plates. When the cells adhered, they were exposed to the specified drugs for the specified time. Subsequently, 10 μl of CCK-8 solution was added to each well and incubated for one hour in the dark. Cell survival rate was evaluated by reading the optical density at a wavelength of 450 nm. The results are shown in Figure 11 G。

[0065] 13. Western Blotting

[0066] Cells were lysed using radioimmunoprecipitation assay buffer (RIPA, Beyotime), and protein levels were quantified using the Beyotime BCA Protein Assay Kit (Beyotime). Equal amounts of cell extracts were separated by 10% SDS-PAGE and then transferred to PVDF membranes. To block non-specific binding, the membranes were incubated in 5% non-fat milk and then incubated with the primary antibody overnight on a shaker at 4°C. Subsequently, the membranes were incubated with the HRP-conjugated secondary antibody (1:2000, Proteintech) for one hour at room temperature. After thorough washing, the fluorescence signals were detected using a UVP ChemStudio system (Ultraviolet Products, USA), and semi-quantitative analysis was performed using VisionWorks software (Analytik Jena, Germany). The results are shown in Figure 11 C、 Figure 11 D、 Figure 11 K、 Figure 11 J and Figure 13 A。

[0067] The primary antibodies used included: anti-ZBED1 (1:1000, Proteintech), anti-BCL2 (1:200, Abcam), anti-BAX (1:1000, Abcam), and anti-Beta-Tubulin (1:1000, Santa Cruz Biotechnology).

[0068] 14. Immunofluorescence

[0069] For immunofluorescence staining of paraffin-embedded sections, the sections were deparaffinized and rehydrated, and then antigen retrieval was performed in citrate buffer at pH 6.0. Subsequently, antibody staining was carried out. First, the sections were treated with antigen blocking solution at room temperature for 20 minutes to prevent non-specific binding. Then they were incubated overnight at 4°C in a 1:1000 diluted ZBED1 antibody solution. After thorough washing with PBST, the sections were incubated with Alexa Fluor 555-labeled IgG antibody at room temperature for 20 minutes in the dark, at a dilution ratio of 1:600. Finally, nuclear staining was performed with a medium containing DAPI, and images were taken using a fluorescence microscope. The results are shown in Figure 11 E.

[0070] 15 Flow cytometry

[0071] Flow cytometry apoptosis: The FITC Annexin V Apoptosis Detection Kit (BD, UK) was used to detect cell apoptosis. After washing and centrifuging the cells with PBS, the cells were resuspended with buffer, counted, and 1×10^5 cells were collected and placed in a 1.5 ml EP tube. 5 μl of PI and 5 μl of FITC Annexin V were added and incubated for 20 minutes. The results are shown in Figure 11 L, Figure 13 C, Figure 13 D, Figure 13 N and Figure 13 O.

[0072] Flow cytometry cell cycle: The Cell Cycle Staining Kit (MultiSciences, China) was used to detect the cell cycle. After washing and centrifuging the cells with PBS, the upper layer of PBS was discarded. 10 μl of Permeabilization solution and 1 ml of DNA Staining solution were added and incubated for 25 minutes. The results are shown in Figure 11 N and Figure 11 O.

[0073] The processed cells above were all detected on a flow cytometer (CytoFlex SRT, Beckman, USA), and the collected data was imported into FlowJo software (version 10.8.1) for analysis.

[0074] 16 Detection of Caspase 3 activity

[0075] The activity of Caspase 3 was detected using the Caspase 3 Activity Assay Kit (Beyotime, China). Cells were first seeded into 6-well plates. After specific treatment, cells were collected by centrifugation with PBS, 150 μl of lysis buffer was added, and after sufficient lysis, the supernatant was collected by centrifugation. An appropriate amount of Ac-DEVD-pNA (2 mM) was added according to the instructions to prepare the reaction system. After incubation at 37 °C for 70 minutes, detection was performed using a microplate reader, and the results are shown in Figure 11 M.

[0076] 17 Visualization of drug synergy

[0077] The R package synergyfinder was used to evaluate the synergy of anti-AML drugs. To obtain as reliable results as possible, four different models were adopted: ZIP, LOEWE, BLISS, and HSA. BLISS model: Assuming independent drug action, the effect is predicted by calculating the probability of independent action of each drug in the combination. HSA model: Defines that if the response of the combination is greater than the response of any single drug, it is considered a synergistic effect. LOEWE model: Considers the dose-response relationship of individual drugs, calculates the expected additive response, and a response higher than expected is considered a synergistic effect. ZIP model: Assumes that each drug does not affect the potency of the other drug and calculates the combined effect. Each model analyzes the effect of drug combinations based on different assumptions and methods, thus providing a comprehensive understanding of drug interactions. Mairin, Astilbin, Diosgenin, Diosgenin, Isoimperatorin were all purchased from Sigma-Aldrich.

[0078] For the specific evaluation results, see Figure 13 H- Figure 13 M. Among them, Figure 13 H is the confidence interval plot of the synergistic drug combination (diosgenin and cytarabine), Figure 13 I- Figure 13 L is the ZIP ( Figure 13 I), Bliss ( Figure 13 J), HSA ( Figure 13 K), and Loewe ( Figure 13L) Synergy diagram. The synergy score was calculated using the SynergyFinder R package. Positive or negative values indicate synergy or antagonism, respectively. Figure 13 M represents the result of synergy measurement with a dose combination of 0.1 μM cytarabine and 16 μM dioscin.

[0079] Example 3

[0080] In vivo experiments

[0081] 18. Animal experiments

[0082] In vivo animal experiments were conducted using the previously constructed female humanized NSG mice. Six 4-week-old humanized NSG mice were randomly divided into the sh-NC group (n = 3) and the sh-ZBED1 group (n = 3). The mice were respectively inoculated with MOLM13 / R cells transfected with sh-ZBED1 or sh-NC (1×10^7 cells, 0.1 mL PBS). Each group of mice was intraperitoneally injected with cytarabine (at a dose of 250 mg / kg, Sigma-Aldrich, USA) for 7 consecutive days, and then the tumor volume was measured every 3 days (the formula for calculating the tumor volume is: length × width × width / 2). After 21 days, the excised tumor tissues of the mice were collected and weighed. See Figure 11 P- Figure 11 R, where Figure 11 P is the tumor entity of the mouse after dissection, Figure 11 Q- Figure R is the tumor volume and weight of each group of mice, ​ Q represents the tumor growth kinetics of NSG mice in the sh-ZBED1 group and the shRNA-NC group presented by the growth curve, ​ R represents the comparative analysis of the tumor mass of nude mice in the sh-ZBED1 group and the shRNA-NC group.

[0083] Six 4-week-old humanized NSG mice were randomly divided into the control group (n = 3) and the dioscin group (n = 3). The mice were all inoculated with MOLM13 / R cells (1×10^7 cells, 0.1 mL PBS), and each group of mice was intraperitoneally injected with cytarabine (at a dose of 250 mg / kg, Sigma-Aldrich, USA) for 7 consecutive days. The mice in the dioscin group were orally administered dioscin at a concentration of 40 mg / kg / day. The control group was orally administered the same volume of normal saline. Then the tumor volume was measured once a week (the formula for calculating the tumor volume is: length × width × width / 2). After 4 weeks, the excised tumor tissues of the mice were collected and weighed. See ​ E- ​ G, ​ E is the photo of the tumor entity of the mouse after dissection,​ F shows the growth kinetic curves of tumor growth in different groups of NSG mice. ​ G shows the results of comparative analysis of tumor masses in different groups of NSG mice.

[0084] 19. micro-CT analysis

[0085] When constructing the orthotopic model, 3-week-old female humanized NSG mice were randomly assigned to two groups: sh-ZBED1 group and shRNA-NC group. Approximately 2×10 6 cells transfected with luciferase plasmid were transplanted into the distal femur. When the average tumor volume reached approximately 100 mm 3 , the mice in each group were intraperitoneally injected with cytarabine (at a dose of 250 mg / kg, Sigma-Aldrich, USA) for 7 consecutive days. Micro CT scan analysis was performed on the dissected femur and tibia samples. The samples were first fixed with 4% paraformaldehyde and then stored in PBS solution at 4°C. High-resolution μCT scans were performed using Skyscan 1276 (from Skyc an, Aartselaar, Belgium), with the scan resolution set to 20.376 μm per pixel, the voltage at 100 kV, and the current at 200 μA. When selecting the region of interest (ROI), it started 0.45 mm below the distal growth plate and extended 0.45 mm proximally to analyze the relevant parameters of trabecular bone, such as bone mineral density (BMD), bone surface (BS), bone volume ratio (BV / TV), trabecular thickness (Tb.Th), trabecular number (Tb.N), and trabecular separation (Tb.Sp). In addition, the cortical bone parameters within a 0.2-mm range in the middle of the femur were also analyzed, such as total cross-sectional area (Tt.Ar), cortical bone area (Ct.Ar), cortical area ratio (Ct.Ar / Tt.Ar), and cortical thickness (Ct.Th), etc. The results are shown in ​ , among which, ​ A - ​ E are the anteroposterior and lateral three-dimensional reconstruction diagrams of the femur in the sh-NC group ( ​ A), the coronal cross-sectional diagram of the whole femur ( ​ B), the cross-sectional diagram and three-dimensional reconstruction diagram of the distal femur near the metaphysis ( ​ C), the three-dimensional reconstruction diagram of the cortical bone in the middle of the femur ( ​ D), the coronal diagram of the proximal tibia ( ​ E), ​ F - ​ J are the anteroposterior and lateral three-dimensional reconstruction diagrams of the femur in the sh-ZBED1 group ( ​ F), the coronal cross-sectional diagram of the whole femur ( ​ G), the cross-sectional diagram and three-dimensional reconstruction diagram of the distal femur near the metaphysis (​ H), Three-dimensional reconstruction diagram of the mid-femur cortical bone( ​ I), Coronal view of the proximal tibia( ​ J) ​ K- ​ P is the change in trabecular bone microstructure parameters, and BMD represents bone mineral density( ​ K); BS represents bone surface area ​ L), BV / TV represents trabecular bone volume fraction ​ M), Tb.Th represents trabecular thickness ​ N), Tb.N represents trabecular number ​ O), Tb.Sp represents trabecular spacing ​ P) ​ Q- ​ T is the change in cortical bone microstructure parameters, and Tt.Ar represents the total cross-sectional area within the periosteal envelope ​ Q), Ct.Ar represents cortical bone area ​ R), Ct.Ar / Tt.Ar represents cortical bone area fraction ​ S), Ct.Th represents average cortical bone thickness ​ T), *P < 0.05, **P < 0.01, ***P < 0.001

[0086] Although the principles of the present invention have been described in detail above in connection with the preferred embodiments of the present invention, those skilled in the art should understand that the above embodiments are merely explanations of the illustrative implementation manners of the present invention and do not limit the scope of the present invention. The details in the embodiments do not constitute a limitation on the scope of the present invention. Without departing from the spirit and scope of the present invention, any obvious changes such as equivalent transformations and simple substitutions based on the technical solutions of the present invention all fall within the protection scope of the present invention

Claims

1. A pharmaceutical composition for inhibiting cytarabine-resistant leukemia MOLM13 / R cells, characterized in that: The invention comprises diosgenin and cytarabine, wherein the concentration of the cytarabine is 0.1 μM, and the concentration of the diosgenin is 16 μM.

2. The pharmaceutical composition according to claim 1, characterized in that Also includes auxiliary materials.

3. The pharmaceutical composition according to claim 2, characterized in that The auxiliary materials include fillers, disintegrants, lubricants, suspending agents, binders or sweeteners.

4. The pharmaceutical composition according to any one of claims 2 to 3, characterized in that The pharmaceutical composition is an oral preparation or an injection preparation.

5. The pharmaceutical composition according to claim 4, characterized in that The oral preparations include capsules, tablets or granules.

Images