Fubp1 knockdown in BCR-ABL1 induced chronic myeloid leukemia (CML) like myeloproliferative neoplasm mouse model

We examine the oncogenic function of the Far Upstream Element Binding Protein 1 (FUBP1) in leukemia-initiating cells. For this perpose we performed knockdown of Fubp1 in CML cells in our mouse model. Here, we transduce donor bone marrow cell with the oncogene BCR-ABL1 and Fubp1 or scrambled shRNA and transplanted in recipient mice. Lineage- BCR-ABL1+ shRNA+ cells were then sorted from diseased mice and analyzed by RNA sequencing. We compare the gene expression of the Fubp1 shRNA expressing CML cells versus the control (scrambled shRNA expressing) cells in order to identify the relevant Fubp1 target genes. Mouse donor BM cells that were harvested from 5-fluorouracil-treated mice, cotransduced twice with cryopreserved and thawed retrovirus expressing MSCV-IRES-RFP BCR-ABL1 and pLKO.1-Venus Fubp1- or scrambled shRNA lentivirus. 2.5-3x10^5 transduced cells were transplanted intravenously into sublethally irradiated (900 cGy) recipient mice. Lineage- BCR-ABL1-RFP+ shRNA-Venus+ DAPI- cells from bone marrow of CML mice (4 mice in the Fubp1 and 4 mice in the scrambled shRNA group) were sorted directly into TRIzol. RNA isolation, library preparation and next generation sequencing were performed by Transcriptome and Genome Analysis Laboratory (TAL, Göttingen, Germany), as described previously.24, 25 RNA-Seq data will be deposited in the GEO database upon acceptance of this manuscript. Read counts per gene were statistically analysed in R (version 3.4.3) using ‘RStudio’ (version 1.1.423). Differential analysis was performed using the ‘DESeq2’ tool (version 1.18.1). For annotation of genes with mouse gene names and human orthologue HGNC symbols, the ‘biomaRt’ package (version 2.34.2) was used. Candidate genes were filtered by a padj value < 0.05 and a log2 fold change < -1 or >1 and visualised in a MA-plot. Principal component analysis was computed with the ‘prcomp’ function and results were visualized using the ‘pca3d package’(version 0.10). Gene set enrichment analysis was performed using the ‘GSEAPreranked’ tool from the GSEA software (version 2.2.3; Broad Institute).

本研究旨在探究远端上游元件结合蛋白1（Far Upstream Element Binding Protein 1, FUBP1）在白血病起始细胞中的致癌功能。为此，我们在小鼠模型的慢性髓系白血病（CML）细胞中开展了Fubp1基因敲低实验。具体操作为：将供体骨髓细胞分别转导携带致癌基因BCR-ABL1与Fubp1或乱序短发卡RNA（scrambled shRNA）的载体，随后将其移植至受体小鼠体内。从患病小鼠中分选得到谱系阴性（Lineage-）的BCR-ABL1+ shRNA+细胞，通过RNA测序（RNA sequencing, RNA-seq）进行分析。为鉴定Fubp1的核心靶基因，我们比较了表达Fubp1 shRNA的CML细胞与对照组（表达乱序shRNA的细胞）的基因表达谱。我们从经5-氟尿嘧啶（5-fluorouracil, 5-FU）处理的小鼠中分离供体骨髓细胞，先后两次转导经冻存复苏的逆转录病毒（该病毒携带MSCV-IRES-RFP-BCR-ABL1）与pLKO.1-Venus-Fubp1或乱序shRNA慢病毒。将2.5×10^5至3×10^5个转导后的细胞经静脉移植至接受亚致死剂量辐射（900 cGy）的受体小鼠体内。从慢性髓系白血病小鼠的骨髓中分选得到谱系阴性、BCR-ABL1-RFP+、shRNA-Venus+且DAPI-的细胞（Fubp1敲低组与乱序shRNA对照组各4只小鼠），直接将其加入TRIzol试剂中。RNA提取、文库构建与下一代测序（next generation sequencing, NGS）均由德国哥廷根的转录组与基因组分析实验室（Transcriptome and Genome Analysis Laboratory, TAL）完成，操作流程参照此前发表的文献24、25。本研究的RNA-seq数据将在论文被接收后提交至基因表达综合数据库（Gene Expression Omnibus, GEO）。利用R软件（版本3.4.3）结合RStudio（版本1.1.423）对每个基因的读段计数进行统计学分析。采用'DESeq2'工具（版本1.18.1）进行差异表达分析。为对基因进行注释（获取小鼠基因名称与人类同源基因的HGNC符号），我们使用了'biomaRt'软件包（版本2.34.2）。候选基因的筛选阈值为校正后P值（padj）<0.05且log2倍变化（log2 fold change）<-1或>1，并通过MA图进行可视化展示。采用'prcomp'函数进行主成分分析（principal component analysis, PCA），并利用'pca3d'软件包（版本0.10）对结果进行可视化。基因集富集分析（gene set enrichment analysis, GSEA）采用Broad研究所开发的GSEA软件（版本2.2.3）中的'GSEAPreranked'工具完成。