Additional file 1 of CHSY1 promotes CD8+ T cell exhaustion through activation of succinate metabolism pathway leading to colorectal cancer liver metastasis based on CRISPR/Cas9 screening

Additional file 1: Table S1. Primer sequences, shRNAs used in this study. Figure S1. (A) Detailed analysis process for bioinformatic analysis after obtaining raw data of sufficient sequencing depth. (B-C) Average matrix values and average sequencing error rates of metastatic (B) and primary (C) foci at different sequencing sites were obtained. (D-E) The separation of AT and GC in metastatic foci (D)and primary foci (E) was detected by statistical analysis of alkali content distribution. Figure S2. (A-B) Detection of primary CRC, adjacent normal bowel tissue, CRC liver metastases, adjacent normal live, and preoperative blood samples using single-cell RNA sequencing. Taxonomic definition of specific gene markers using UMAP plots identified 12 cell clusters. (C-D) The bar chart and violin figure showed that CHSY1 was relatively highly expressed in the total sample analysis of the cancer cell population. (E)The dot plot shows the expression of CHSY in colorectal cancer primary tumor(CT), adjacent tissue(CP), liver metastases(LT), and adjacent metastatic tissue(LP). Figure S3. (A-B) Gene overexpression efficiency of pcDNA3-CHSY1 assessed by qRT-PCR and Western blotting. (C-D) pcDNA3-CHSY1 promoted the proliferation of HCT116 and LOVO cells according to the results of CCK-8 and EdU assays. (E-F) Transwell and wound healing assays show that pcDNA3-CHSY1 significantly promoted the invasive and migratory functions of HCT116 and LOVO cells.Scale bar, 100μm. **, P < 0.01; ***, P < 0.001;****,P < 0.0001. Figure S4. The TISIDB database was used to predict the correlation between CHSY1 expression and immune factors(for example,PD-L1, PD1, LAG3, IDO1 and CTLA4). Figure S5. Single, viable and intact CD45+ 21 immune cells were selected and circulated in liver metastases. There are a total of 33 cell clusters, each of which was defined based on markers specific to the respective. Figure S6. Negative patterns of metabonomics analysis of down-regulating CHSY1 in CRC cells. (A)Volcano maps classify metabolites by HMDB. (B) circos plot showing correlations between multiple differential metabolites. (C) According to the structure and function of the metabolites, the different metabolites in each control group were classified and counted, and the classification results of the substances in the KEGG and HMDB databases were provided respectively. (D) Top 20 up-regulated metabolites and top 20 down-regulated metabolites. (E-F) KEGG pathway analysis indicated the main concentrated pathways of these differential metabolites. (G) Sankey diagram visualization of data flow trends between down-regulated metabolites and various pathways. Figure S7. Positive patterns of metabonomics analysis of down-regulating CHSY1 in CRC cells. (A)Volcano maps classify metabolites by HMDB. (B) circos plot showing correlations between multiple differential metabolites. (C) According to the structure and function of the metabolites, the different metabolites in each control group were classified and counted, and the classification results of the substances in the KEGG and HMDB databases were provided respectively. (D) Top 20 up-regulated metabolites and top 20 down-regulated metabolites. (E-F) KEGG pathway analysis indicated the main concentrated pathways of these differential metabolites. (G) Sankey diagram visualization of data flow trends between down-regulated metabolites and various pathways. Figure S8. (A-B) Inhibitory effect of artemisinin on CHSY1 assessed by qRT-PCR and Western blotting. (C-D) Artemisinin inhibited the proliferation of HCT116 and LOVO cells according to the results of CCK-8 and EdU assays. (E-F) Transwell and wound healing assays show that Artemisinin significantly inhibited the invasive and migratory functions of HCT116 and LOVO cells.**, P < 0.01; ***, P < 0.001;****, P < 0.0001. Figure S9. (A-C) Three-dimensional diagram and two-dimensional diagram show artemisinin can bind to the 429 PHE amino acid residue, 410 MET amino acid residue, 295 SER amino acid residue and 409 VAL amino acid residue of the receptor protein CHSY1 through hydrophobic forces. (D-E) Immunohistochemistry results of CD8, CD4, Ki67, PD-L1 and PD1 expression in the respective groups.*, P < 0.05; **, P < 0.01; ***, P < 0.001;****, P < 0.000.

附加文件1：表S1。本研究中使用的引物序列及短发夹RNA（short hairpin RNA, shRNAs）。图S1。(A) 获得足够测序深度的原始数据后，生物信息学分析的详细流程。(B-C) 获取不同测序位点的转移灶(B)与原发灶(C)的平均矩阵值及平均测序错误率。(D-E) 通过统计碱基含量分布，分析转移灶(D)与原发灶(E)中AT与GC的分离情况。图S2。(A-B) 采用单细胞RNA测序检测结直肠癌原发组织、癌旁正常肠组织、结直肠癌肝转移灶、转移灶旁正常肝组织及术前血液样本。通过统一流形近似与投影（Uniform Manifold Approximation and Projection, UMAP）图对特异性基因标志物进行分类定义，共鉴定出12个细胞簇。(C-D) 柱状图与小提琴图显示，在癌细胞群体的整体样本分析中，软骨素合成酶1（CHSY1）呈相对高表达。(E) 点状图展示了结直肠癌原发肿瘤(CT)、癌旁组织(CP)、肝转移灶(LT)及转移灶旁组织(LP)中CHSY的表达情况。图S3。(A-B) 通过实时荧光定量聚合酶链反应（quantitative Real-Time Polymerase Chain Reaction, qRT-PCR）与蛋白质印迹法（Western blotting）验证pcDNA3-CHSY1的基因过表达效率。(C-D) 细胞计数试剂盒-8（Cell Counting Kit-8, CCK-8）与5-乙炔基-2'-脱氧尿苷（5-Ethynyl-2'-deoxyuridine, EdU）实验结果显示，pcDNA3-CHSY1可促进HCT116与LOVO细胞的增殖。(E-F) Transwell小室实验与划痕愈合实验结果表明，pcDNA3-CHSY1可显著增强HCT116与LOVO细胞的侵袭与迁移能力。比例尺：100μm。**, P < 0.01；***, P < 0.001；****, P < 0.0001。图S4。采用肿瘤免疫互作数据库（TISIDB）预测CHSY1表达与免疫因子（如程序性死亡受体配体1(PD-L1)、程序性死亡受体1(PD1)、淋巴细胞活化基因3(LAG3)、吲哚胺2,3-双加氧酶1(IDO1)及细胞毒性T淋巴细胞相关蛋白4(CTLA4)）的相关性。图S5。选取单个、存活且完整的CD45+免疫细胞，共21个，这些细胞在肝转移灶中循环分布。本研究共鉴定出33个细胞簇，每个细胞簇均依据其对应的特异性标志物完成分类定义。图S6。结直肠癌细胞中下调CHSY1的代谢组学分析负向模式。(A) 火山图依据人类代谢组数据库（Human Metabolome Database, HMDB）对代谢物进行分类。(B) 环状图展示多种差异代谢物间的相关性。(C) 根据代谢物的结构与功能，对各对照组中的差异代谢物进行分类统计，并分别给出京都基因与基因组百科全书（Kyoto Encyclopedia of Genes and Genomes, KEGG）及HMDB数据库的物质分类结果。(D) 排名前20的上调代谢物与排名前20的下调代谢物。(E-F) KEGG通路富集分析显示，这些差异代谢物主要富集于相应通路。(G) 桑基图可视化展示下调代谢物与各类通路间的数据流向趋势。图S7。结直肠癌细胞中下调CHSY1的代谢组学分析正向模式。(A) 火山图依据HMDB对代谢物进行分类。(B) 环状图展示多种差异代谢物间的相关性。(C) 根据代谢物的结构与功能，对各对照组中的差异代谢物进行分类统计，并分别给出KEGG及HMDB数据库的物质分类结果。(D) 排名前20的上调代谢物与排名前20的下调代谢物。(E-F) KEGG通路富集分析显示，这些差异代谢物主要富集于相应通路。(G) 桑基图可视化展示下调代谢物与各类通路间的数据流向趋势。图S8。(A-B) 通过qRT-PCR与Western blotting评估青蒿素对CHSY1的抑制效果。(C-D) CCK-8与EdU实验结果显示，青蒿素可抑制HCT116与LOVO细胞的增殖。(E-F) Transwell小室实验与划痕愈合实验结果表明，青蒿素可显著抑制HCT116与LOVO细胞的侵袭与迁移能力。**, P < 0.01；***, P < 0.001；****, P < 0.0001。图S9。(A-C) 三维结构图与二维结构图展示，青蒿素可通过疏水作用力结合至受体蛋白CHSY1的429位苯丙氨酸（PHE）残基、410位甲硫氨酸（MET）残基、295位丝氨酸（SER）残基及409位缬氨酸（VAL）残基。(D-E) 各组样本中CD8、CD4、Ki67、PD-L1及PD1表达的免疫组化结果。*, P < 0.05；**, P < 0.01；***, P < 0.001；****, P < 0.000。