基于磷酸化蛋白质组的肝细胞癌激酶组活性分析

Substrate phosphorylation mediated by protein kinases is one of the most important post-translational modifications, and involved in regulating almost all of biological processes and pathways. Protein kinases have been identified as key drug targets for cancer therapy, and the abnormality of kinase activity is highly associated with tumorigenesis, cancer progression and metastasis. The protein kinase activity is composed by enzymatic activity and the specificity of substrate recognition, whereas related researches such as kinome activity profiling in cancers have been emerged to be great challenges, and can provide potentially more efficient drug targets and treatments for personalized medicine of cancers. In this project, we will combine multiple techniques, such as non-label quantitative phosphoproteomic identification, bioinformatic prediction, and various experimental approaches of molecular biology, biochemistry, cell biology and mouse model, to characterize and quantify phosphorylation sites with modification levels in a variety of human normal liver and hepatocellular carcinoma (HCC) cell lines. We will design novel computational algorithms for profiling kinome activity in liver cancers, develop an integrative platform for both computation and experiments, and accurately predict significantly activated or inhibited kinases in HCCs, together with following experimental validations. Based on the results, we will experimentally identify new protein kinases that participate in the regulation of HCC metastasis and drug-resistance. Using small-molecule compounds of newly identified kinases, we will design potentially useful treatments by individual or combinatorial drugs, and probe the efficacy on HCC inhibition in both cell and mouse levels.

蛋白激酶介导的底物磷酸化是重要的翻译后修饰之一，参与调控几乎所有的生物学过程和通路。蛋白激酶是癌症治疗的重要药物靶标，其活性异常与肿瘤的发生、发展和转移密切相关。激酶活性由酶反应活性和底物识别特异性两部分构成，其相关研究即癌症激酶组活性分析已成为重要挑战，有望为癌症的个性化医疗提供潜在的、更有效的药靶和治疗方案。本项目中，我们拟结合非标记定量磷酸化蛋白质组鉴定技术、生物信息学预测方法以及分子、生化、细胞和小鼠实验等研究手段，分别鉴定和定量多种人类正常肝细胞和肝癌细胞系在索拉非尼处理前后的磷酸化位点及修饰水平，设计新的肝癌激酶组活性分析的计算方法，建立整合的计算分析与实验验证平台，准确预测并证实肝癌中活性显著差异的激酶，发现参与新的、调控肝癌转移和耐药性的激酶，利用相应的小分子化合物探索单独或联合用药的潜在治疗方案，并在细胞和小鼠中验证其对肝癌抑制的有效性。

蛋白激酶是癌症治疗的重要药物靶标，其活性异常与肿瘤的发生、发展及转移密切相关。因此本项目的关键科学问题是，发展新的技术手段，建立新的癌症激酶组活性分析方法，高覆盖率、全面检测癌症样本中激酶活性的变化情况，发现新的、参与调控肿瘤的激酶和信号通路。本项目的主要研究成果包括四个部分：1）围绕多种翻译后修饰类型和重要生物学过程及人类疾病，构建了节律基因及修饰信息数据库、赖氨酸修饰数据库、自噬蛋白质及修饰数据库、泛素和类泛素偶联调控因子数据库、人类疾病相关修饰数据库、磷酸化调控因子数据库、液-液相分离数据库、原核和真核磷酸化位点数据库；2）基于前沿人工智能技术，设计了组蛋白乙酰转移酶特异性位点预测工具、赖氨酸丙二酰化位点预测工具、磷酸化蛋白质结合结构域特异性结合位点预测工具、琥珀酰化修饰位点预测工具、真核激酶特异性磷酸化位点预测工具、棕榈酰化位点预测工具和基于词云的富集分析可视化软件；3）开展多组学整合和图像处理等研究，包括自噬活性定量工具的开发、PD-L1棕榈酰化对T细胞肿瘤免疫应答的影响、多组学整合策略揭示果蝇节律激酶组分析、基于多组学数据整合的调控自噬关键激酶的预测方法设计和基于磷酸化组数据整合的肝癌关键激酶预测和候选精准药靶分析的算法开发；4）利用本项目研发的技术，设计了新冠肺炎患者临床结局的人工智能预测系统和基于新冠血浆蛋白质组学数据的分子标志物预测算法。计划发表SCI 论文5~6 篇，获颁计算机软件著作权登记证书2~3 项。实际发表标注资助的SCI论文26篇，其中以通讯或共同通讯作者身份发表Immunity 1篇，Nature Biomedical Engineering 1篇和Nature Communications 1篇。在国际会议上作特邀报告11次，申请10项发明专利（2项已授权），获得9项软件著作权登记证书。2017年入选教育部“长江学者奖励计划”青年学者。项目执行期间共培养毕业博士生8人，毕业硕士生1人。