北京生物医学工程

基于差异代谢网络的泛癌代谢分析

Pan-cancer analysis based on differential metabolic network

作者：熊宇峰郑浩然

单位：中国科学技术大学计算机科学与技术学院(合肥 230027)

分类号：R318.04

出版年·卷·期（页码）：2020·39·5（506-512）

摘要：

目的基因组规模代谢网络模型（genome-scale metabolic model, GEM）是当前癌症代谢研究常用的一种分析工具，高质量的癌症代谢网络有助于人们了解癌症内部的代谢特征。泛癌代谢分析需要为多种癌症构造代谢网络模型，然而传统的直接基于癌症基因表达或蛋白质数据进行的网络重构方法，无法避免引入癌症所在组织的代谢特征，从而影响探寻疾病本身的代谢特点。为了削弱组织代谢特征对研究的影响，更加准确地提取癌症相对正常组织细胞在代谢层面上发生的变化，提出一种基于差异代谢网络的泛癌代谢分析方法。方法以癌症基因组图谱（The Cancer Genome Atlas, TCGA）数据库中17种癌症作为分析对象，通过分析癌症及癌旁组织的基因表达数据，采用基因芯片线性模型（linear models for microarray analysis, Limma）工具包分别提取这17种癌症的差异表达基因，并将人类全基因组规模代谢网络Recon3D作为背景网络，结合代谢网络重构算法FASTCORE为每种癌症构建差异代谢网络。最后联合多种癌症网络利用流量平衡分析（flux balance analysis, FBA）方法进行基因敲除以及毒性测试，分析癌症代谢共性，发现潜在药物治疗靶点。结果得到了每种癌症的差异表达基因、差异代谢网络模型，以及潜在的药物靶点，通过联合所有癌症进行分析，最后得到了7个潜在的泛癌药物靶点，其中包括CRLS1、CTH、PTDSS1、SLC6A14、SGMS1、PKM以及PLD2。7个药物靶标中已经有5个用于不同的癌症治疗策略中，而其余结果代表了潜在的药物靶点。结论本文提出的差异代谢网络模型可以有效削弱组织代谢特征对癌症本身特征的干扰，有助于泛癌代谢分析。

Objective Genome-scale metabolic model (GEM) is currently a commonly used tools for cancer metabolism research. High-quality cancer metabolism models can help us understand the metabolic characteristics of cancer. In pan-cancer analysis we need to reconstructed connected genome-scale metabolic models for each cancer. However, the traditional way based on gene expression or protein data to construct metabolic network lead to the metabolic characteristics of cancer tissue, which influence understanding the metabolic characteristics of disease. In order to eliminate the influence of tissue metabolism on the research and accurately extract the metabolic changes of cancer relative to normal tissues, we propose a pan-cancer metabolic analysis method based on differential metabolic networks. Methods Taking 17 cancers in The Cancer Genome Atlas (TCGA) as the analysis objects, we used the linear models for microarray analysis (Limma) toolkit to determine genes that are differentially expressed between cancer and benign tissue. Then we combined differential genes, reference generic human GEM named Recon3D and metabolic network reconstruction algorithm FASTCORE to build differential metabolic networks for each cancer. Flux balance analysis (FBA) was used for metabolic flux analysis and gene knockout on a variety of cancer networks to study the common features of cancer metabolism and potential drug targets. Results In the experiments, we obtained differentially expressed genes, differential metabolic network models, and potential drug targets for each type of cancer. Through the analysis of all cancers, seven potential pan-cancer drug targets were obtained, including CRLS1, CTH, PTDSS1, SLC6A14, SGMS1, PKM, and PLD2. Five of the seven drug targets had already been used in different cancer treatment strategies, while the remaining drug targets represent new potential drugs. Conclusion The differential metabolic network model proposed in this paper can effectively eliminate the differences between different tissues, which is helpful to studying the metabolic characteristics of pan-cancer.

参考文献：

[1] DeBerardinis R J, Chandel N S. Fundamentals of cancer metabolism[J]. Science advances, 2016, 2(5): e1600200.

[2] Luengo A, Gui D Y, Vander Heiden M G. Targeting metabolism for cancer therapy[J]. Cell chemical biology, 2017, 24(9): 1161-1180.

[3] Pavlova N N, Thompson C B. The emerging hallmarks of cancer metabolism[J]. Cell metabolism, 2016, 23(1): 27-47.

[4] Zhang C, Hua Q. Applications of genome-scale metabolic models in biotechnology and systems medicine[J]. Frontiers in physiology, 2016, 6: 413.

[5] Agren R, Mardinoglu A, Asplund A, et al. Identification of anticancer drugs for hepatocellular carcinoma through personalized genome‐scale metabolic modeling[J]. Molecular systems biology, 2014, 10(3).

[6] Gatto F, Miess H, Schulze A, et al. Flux balance analysis predicts essential genes in clear cell renal cell carcinoma metabolism[J]. Scientific reports, 2015, 5: 10738.

[7] Gatto F, Ferreira R, Nielsen J. Pan-cancer analysis of the metabolic reaction network[J]. Metabolic engineering, 2020, 57: 51-62.

[8] Brunk E, Sahoo S, Zielinski D C, et al. Recon3D enables a three-dimensional view of gene variation in human metabolism[J]. Nature biotechnology, 2018, 36(3): 272.

[9] Vlassis N, Pacheco M P, Sauter T. Fast reconstruction of compact context-specific metabolic network models[J]. PLoS computational biology, 2014, 10(1): e1003424.

[10] Ritchie M E, Phipson B, Wu D, et al. limma powers differential expression analyses for RNA-sequencing and microarray studies[J]. Nucleic acids research, 2015, 43(7): e47-e47.

[11] Mardinoglu A, Agren R, Kampf C, et al. Genome-scale metabolic modelling of hepatocytes reveals serine deficiency in patients with non-alcoholic fatty liver disease[J]. Nature communications, 2014, 5: 3083.

[12] Thiele I, Swainston N, Fleming R M T, et al. A community-driven global reconstruction of human metabolism[J]. Nature Biotechnology, 2013, 31(5): 419-425.

[13] Sharma B, Kanwar S S. Phosphatidylserine: A cancer cell targeting biomarker[C]//Seminars in cancer biology. Academic Press, 2018, 52: 17-25.

[14] Heirendt L, Arreckx S, Pfau T, et al. Creation and analysis of biochemical constraint-based models using the COBRA Toolbox v. 3.0[J]. Nature protocols, 2019, 14(3): 639.

[15] Wang H, Marci?auskas S, Sánchez B J, et al. RAVEN 2.0: A versatile toolbox for metabolic network reconstruction and a case study on Streptomyces coelicolor[J]. PLoS computational biology, 2018, 14(10): e1006541.

[16] Agren R, Mardinoglu A, Asplund A, et al. Identification of anticancer drugs for hepatocellular carcinoma through personalized genome‐scale metabolic modeling[J]. Molecular systems biology, 2014, 10(3).

[17] Bidkhori G, Benfeitas R, Elmas E, et al. Metabolic network-based identification and prioritization of anti-cancer targets based on expression data in hepatocellular carcinoma[J]. Frontiers in physiology, 2018, 9: 916.

[18] Vallabhapurapu S D, Blanco V M, Sulaiman M K, et al. Variation in human cancer cell external phosphatidylserine is regulated by flippase activity and intracellular calcium[J]. Oncotarget, 2015, 6(33): 34375.

[19] Sikder M O F, Yang S, Ganapathy V, et al. The Na+/Cl?-Coupled, Broad-Specific, Amino Acid Transporter SLC6A14 (ATB 0,+): Emerging Roles in Multiple Diseases and Therapeutic Potential for Treatment and Diagnosis[J]. The AAPS journal, 2018, 20(1): 12.

[20] Rozhkova A V, Zinovyeva M V, Sass A V, et al. Expression of sphingomyelin synthase 1 (SGMS1) gene varies in human lung and oesophagus cancer[J]. Molekuliarnaia biologiia, 2014, 48(3): 395-402.

[21] Gomez-Cambronero J. Phospholipase D in cell signaling: from a myriad of cell functions to cancer growth and metastasis[J]. Journal of Biological Chemistry, 2014, 289(33): 22557-22566.

服务与反馈：

【文章下载】【加入收藏】

提示：您还未登录，请登录！点此登录