大数据与精准医学

大数据与精准医学

1. ⼤大数据与精准医学 Big Data and Precision Medicine 个性化药物,花椒素的发现与抗衰老作用 Personalized Medicine, a Drug Candidate From Chinese Pepper and Its Anti-aging Role 魏冬青 上海交通大学生命科学技术学院生物信息与生物统计系 微生物代谢国家重点实验室 Wei, Dongqing State Key Lab. Of Microbial Metabolism College of Life Science and Biotechnology and Shaanghai Jiaotong University dqwei@sjtu.edu.cn
2. 2016大数据与精准医学-浪潮与交大-梅宏
3. HPC, 1986-2016 IBM 3090-A Vector Machine at U. British Columbia NEC SGI Pittsburg Super-Computer Center Centre de Recherche en Calcul Applique(CERCA) McGill, Concordia, U. de Montreal, U. Quebec A Nature Comm. Paper just published by Yukun Wang et al. 博士生顾若虚获得上海市优博
4. Acknowledgements   Team Members/Collaborators   Professor K.C. Chou, Shanghai Jiaotong U. and Gordon Life Sciences Institute, US–               Rank No1. in the world on hot papers(17) Professor Weizhu Zhong- Pfizer Pharma., USA Professor Yixue Li, Zhongdong Qiao, Yilei Zhao, Qishi Du, Haifeng Chen, Yimin Liu, Liu Qi, Wei Chaochun, Juan Huo, Shanghai Jiaotong University Prof./Dr. Suzanne Sirois, Univesite de Quebec A Montreal(UQAM) Prof. Hong Guo, Univ. of Tennessee, Dept. of Biochem., Cell and Mol. Biol., Knoxville Yiru Gan, Tianjin University Researchers: Wang Jingfang, Wu Maoying, Wang Zhuo, Li Na Students: Ruoxu Gu, Tao Zhang, Peng Lian, Ying Wang, Yukun Wang, Mingzhu Zhao, Qiang Zhou, Huaimeng Fan, Li Li, Ying Wang, Peisi He, Yufang Wang, Hao Dai, Shigao Chen, Jing Chang, Kai Xu, Shouxin Cui, Yanzhi Bai, Laiyu Lu, Xiaoli Yuan, Yi, Xiong, Lin Gao, Shuo Liu, Jinghe Hu, Yingjie Wang, Liwei Yan, Jiao Zhang, Jian Yang, Xiaolei Cui, Yu Yao, Lin Huang, Zhaobin, Xu, Hui Gu, Bei Tang, Suqing Wang, Rui Zhang Rui, Yun Li, Hui Gao, Weina Gao, Huiqin, Zheng, Huachun Wei, Yuqing Zhong, Qi Chen, Zhiyuan Xie, Jue Li, Ke Gong, Jing He, Congying Dai, Chengcheng Zhang, Jingyi Yan
5. The Shanghai Jiaotong University A public school founded in 1896, the second oldest in China, picture shows students of the Nanyang Public School during the reign of Emperor Guangxu
6. Used to be an engineering school Instrumental for railway and ship building, and many other engineering applications
7. One of the most prestigious universities in China with full academic programs 7 campuses, the Minhang campus is one of the largest among Chinese universities with 2822903m2 (200 hectare)
8. Beautiful Campus
9. Campus Location Campus Name Xuhui Campus Address and Zip Code Size of Area 1954 Huashan Road, Shanghai, 200030 236492m2 800 Dongchuan Road, Min Hang, Min Hang Campus Shanghai, 200240 2822903m2 Fa Hua Zhen Road 535 Fahuazhen Road, Shanghai, Campus 200052 33721m2 Shang Zhong Road 100 Shangzhong Road,Shanghai, Campus 200232 31785m2 Qibao Campus 2678 Qixin Road, Shanghai, 201101 218360m2 South Chongqing Road Campus 227 South Chongqing Road 200025 Tel:021-63846590 78667 m 2 other 113786m2 Total 3535714m 2
10. General Information 1.  full-time students amounts to 38,000, there are 18,000 undergraduates and 18,100 graduates 2.  20 academic schools, 60 undergraduate programs, 152 masters-degree programs, 93 Ph.D. programs, 16 post-doctorate programs, 16 state key doctorate programs and 14 State key laboratories and National engineering centers and 13 affiliated hospitals(major ones in Shanghai, No. 1 in clinical medicine) 3.  22 academicians of the Academy of Sciences and Academy of Engineering, 31 "Changjiang Chair Professors" and more than 3,000 faculty members
11. Well Known Alumni JIANG Zemin(ex-president), LU Dingyi, DING Guangen, WANG Daohan, QIAN Xuesen(icon of Chinese science), WU Wenjun, ZOU Taofen, MAO Yisheng, CAI Er, HUANG Yanpei, SHAO Lizi, WANG An(US computer Entrepreneur) Of all the academicians of China's Academy of Sciences and Academy of Engineering, more than 200 are the alumni of Jiao Tong University—close to 10% Among 23 national celebrated scientists who made primary contributions to Chinese nuclear program, 6 are Alumni of Jiaotong University Among 14 recipients of the National Supreme Science Award, 2 are Jiaoda Alumni(Wu Wenjun, Xu Guangxian
12. Contributions from alumni of SJTU         Libraries by Bao’s Family And many others 爱国荣校饮水思源 Bear in mind where water comes from, so be a patriotic student makes your university proud Mulan Shipping and Architecture Building Donated by Mother of E. Chao
13. Interdisciplinary Sciences-- Computational Life Sciences Editor-in-Chief: Dong-Qing Wei ISSN: 1913-2751 (print version) Springer Journal no. 12539 International Association of Scientists in the Interdisciplinary Areas Email:dqwei@sjtu.edu.cn Currently Cited by PubMed/ Medline, Chem. Abstract, and many others SCI soon! http://www.springer.com/life+sci/ bioinformatics/journal/12539
14. 2008 Nobel Prize on Physiology or Medicine Chairman of our editorial board Prof. Luc Montagnier for his discovery of human immunodeficiency virus(HIV) Recently Joint our university to establish the Luc Montagnier BioMedical Research Institute(LMBMRI), to work closely with us trying to find new ways to tread AIDS and other chronicle diseases. 首位中组部“顶尖千人” 上海交大大学讲座教授
15. Jan. 20, 2015 ⼤大数据与精准医学 Big Data and Precision Medicine “I want the country that eliminated polio(小儿麻痹症)and mapped the human genome to lead a new era of medicine —  one that delivers the right treatment at the right time. In some patients with cystic fibrosis(囊胞性纤维症), this approach has reversed a disease once thought unstoppable. Tonight, I’m launching a new Precision Medicine Initiative to bring us closer to curing diseases like cancer and diabetes — and to give all of us access to the personalized information we need to keep ourselves and our families healthier.” “Let’s Call it Precision Medicine!”
16. 健康科学的发展: 转化医学 个体化医学 精准医学 精准诊断和精准治疗 本质性转变: 从诊断治疗到健康保障 精准预防 基于精准医学理念的个体化治疗市场规模日益扩大,2018 年前全 球市场规模将达到2238 亿美元
17. 世界卫生组织(WHO)《2014 年非 传染性疾病国家概况》统计数据显示,我国 在2000-2012 年肿瘤、2 型糖尿病、心血管疾 病这三种慢病死亡总数达近700 万,占所有 死亡人数的70%,远高于全球平均(45%)和 英美等发达国家(57%-61%);到2030 年我 国阿尔茨海默病患者将达1200万。我国心血 管疾病患者约有2.9 亿人,每年约有350 万人 死于心血管疾病,占总死亡原因的41%,高 居死因榜首。
18. 精准医学研究已成为新一轮国家科技竞争和 引领国际发展潮流的战略制高点。 美国积极推动精准医学的发展,将其上升为国家战略;欧盟以 精准医学理念指导其创新药物二期计划;英国开展十万人测序 计划并成立精确癌症研究所;日本将精准医学相关内容列入科 技创新计划中。这标志着国际上在基因资源利用、新药靶点发 现、新的诊断治疗方法开发、生物医药新产品研发等的竞争进 入新的阶段,对我国生物医药与健康产业的发展形成严峻挑战。
19. 美国积极推动精准医学的发展 2011 年,美国国家科学院研究理事会(NRC)发布了题为“ 迈向精准医学:构建生物医学研究知识网络和新的疾病分类 体系”的报告,首次提出精准医学概念。2015 年,美国开始启动 精准医学研究计划,标志着精准医学上升为国家战略。为了促进 精准医学的发展,美国国立卫生研究院(NIH)在2015 财年、2016 财年预算中将精准医学作为重点领域进行资助,计划在2015年10 月开始投入2.15 亿美元启动精准医学计划,首先进行100 万人基 因组测序,与美国生物库中的数据信息联合形成大型研发资源库 ,作为全面加速生物医学研发计划的一部分,助力开发新一代 药物;启动了肿瘤基因组图谱二期计划(TCGA2),进一步加大 肿瘤机制研究和肿瘤治疗个体化药物研发的“精准”性。在药物 开发方面,2014 年美国NIH 和生物制药公司联合启动重大研究项 目——加速建立医学合作AMP 计划,旨在发展用于新的诊断和药 物开发的疾病靶标模型,专注于阿尔茨海默病、2 型糖尿病、自 身免疫性疾病类中的风湿性关节炎和系统性红斑狼疮等三类疾病 的研究。
20. 欧盟及成员国力推精准医学研究 2014 年3 月,欧盟发布创新药物2 期计划战略 研究议程(IMI2),其主题是实现精准医疗,即正 确的时机向正确的病人提供正确的预防治疗措施 。IMI2 将带来新的工具、方法及预防和治疗方案 ,(直接或间接)促进个体化医疗的发展。2012 年 ,英国宣布对患有癌症及罕见疾病的十万英国人进 行全基因组测序,旨在根据基因组学和临床数据为 患者制定个性化疗法。英国技术战略委员会(TSB )还在2014 年建立了“精准医学孵化器”,帮助英 国在该领域加快创新步伐;牛津大学已投入约1.5 亿英镑,成立精准癌症医学研究所。法国早在2012 年就在“投资未来计划”国家计划中,出资1 亿欧 元资助个体化医疗项目。
21. 日本将精准医学相关内容列入科技创新计划中 日本在2011 年实施的FANTOM计划第五阶段中 投入1 亿美元开展功能基因组研究。在2014 科技创 新计划中将“定制医学/基因组医学”列为重点关注 领域之一。计划在2015 年建立疾病的全基因组数 据库,识别日本人的标准基因序列及有助于疾病预 后的基因,并利用基因数据对抗癫痫剂的副作用进 行预测性诊断。至2020-2030 年大幅度改善终身性疾 病(糖尿病、中风、心脏病)的干预效果;建立对 癌变可能性及抗癌药物的治疗效果或副作用的预测 性诊断方法;开展针对抑郁症和痴呆症的临床研究 ;开发诊断和治疗神经肌肉疾病的诊断和治疗方法。
22. 国家发改委批复建设基因检测应用示范中心 在通知中,基因检测技术列入“新型健康技术惠民工程”,其总体目标是“支 持拥有核心技术、创新能力和相关资质的机构,采取网络化布局,率先建设30 个基因检测技术应用示范中心,以开展遗传病和出生缺陷基因筛查为重点,推 动基因检测等先进健康技术普及惠民,引领重大创新成果的产业化”。 2015年-2017年年度实施重点和工作要求如下: 2015年重点:启动第一批10个左右省市试点建设基因检测技术应用示范中心。 以目前相对成熟的遗传性耳聋和唐氏综合征等遗传性疾病墓因筛查为重点,优 先支持地方政府已确定相关政策措施和运行模式、建设条件具备的示范中心 建设,推进基因检测技术在遗传性疾病、肿瘤、心脑血管疾病和感染性疾病等 重大疾病防治上的应用,促进健康惠民。 2016年重点:实施第二批墓因检测技术应用推广示范中心建设。继续优选启动1 0个左右省市示范中心建设,快速推进基因检测技术在遗传性疾病大规模筛查 上的应用。依托具有个体化医学检测等资质的墓因检测技术应用示范中心,开 发新的疾病基因检测技术,探索基因检测技术在个人墓因组检测、墓因身份证 等新领域的产业化应用。 2017年重点:继续实施第三批墓因检测技术应用推广示范中心建设,推动自主 研制基因检测仪器、试剂的成果应用。引导墓因检测仪器设备及试剂的骨干 企业,通过上下游联动、协同推进具有我国自主知识产权的基因检测仪器设备 及试剂的产业化应用,提升我国基因检测产业的竞争力。
23. 产业化前景 精准医学的发展将带动相关产业的快速发展,孕育巨大市场 空间 首先,生物样本和数据本身就具有市场价值。据统计,2015 年生物 样本库市值将超过22.5 亿美元。至2018 年生物大数据的市场总额将增长 至76 亿美元,年复合增长率达到71.6%。生物数据的商业价值已经初步 体现,2015 年1 月,罗氏制药子公司Genentech 向美国基因测序公 司23andMe 注资 6000 万美金,用于共享23andMe 收集的帕金森病患者 的基因数据,并基于数据信息研发帕金森病的治疗方案。 国家后续投资万亿用于精准医学领域
24. 基因测序是精准医疗产业的重要组成部分。来自BBC research的数据显示,全球基因测序市场总量从2007 年的794.1 万美元增长至2013 年的45 亿美元,预计未来几年全球市场仍将 继续保持快速增长,2018 年达到117 亿美元。 分子诊断是精准医疗的另一重要子行业,已经成为生 物医药行业新热点,据Markets and Markets 公司估测, 2018 年 的全球市场市值将达到79 亿美元,2013-2018 年间的复合年增 长率为9.7%。 基于精准医学理念的个体化治疗市场规模日益扩大,2018 年前全球市场规模将达到2238 亿美元。美国十大商业保险公司 已将50 余项疾病个体化诊疗分子检测项目列入医疗保险。巨大 市场空间吸引众多医药公司开展研发,目前已有多种个体化诊 疗产品上市。 截至2013 年,美国FDA 已批准了100 多种个体 化药物,重点关注癌症和慢性疾病的个性化治疗。
25. 我国精准医学的发展目标 以我国常见高发重大疾病及若干罕见病为切入点, 构建百万人级自然人群国家大型健康队列和特定疾病队列、 多层次精准医疗知识库体系和生物医学大数据共享平台,突 破新一代生命组学技术和大数据分析技术,建立创新性的大 规模研发疾病预警、诊断、治疗与疗效评价的生物标志物、 靶标、制剂的实验和分析技术体系。 以临床应用为导向,形成重大疾病的风险评估、预 测预警、早期筛查、分型分类、个体化治疗、疗效和安全性 预测及监控等精准防诊治方案和临床决策系统,建设中国人 群典型疾病精准医疗临床方案的示范、应用和推广体系,推 动一批精准治疗药物和分子检测技术产品进入国家医保目录 ,为显著提升人口健康水平、减少无效和过度医疗、避免有 害医疗、遏制医疗费用支出快速增长提供科技支撑,使精准 医疗成为经济社会发展新的增长点。
26. 转变的基础: 1、 组学: 基因组, 转录组,蛋白质组,代谢组……. 大数据:人群、队列 是当前国际两大科学前沿的交汇 2、 基因型与表型的关联 生物信息学 生物网络 系统生物学 在此基础上融合临床检验、影像学等指标
27. Department of Bioinformatics and Biostatistics, SJTU   Established in 2002, The first undergraduate bioinformatics program in China; The first Ph.D. Program; Ranked No. 1 nationally   Faculties: Honorary professor: Tim Springer(Havard), Rudy Marcus(Caltec), Martin Karplus(Harvard) Guest Professors, Heping Zhang(Yale), ZJ Tian(Johnson & Johnson), Li Yixue(Shanghai Center of Bioinformatics Technology) Professors: Wei Dongqing, Zhao Yilei Associate Professors: Liu Qi, Chen Haifeng, Xiao Xianglin, Xie Lu, Liu Limin, Wei Chaochun   Facilites: Lenov Clusters-350 processors, and future Blue Gene   Grants: National 863(3), 973(2) and NSFC(5) Currently hiring, write to me:dqwei@sjtu.edu.cn 15 faculties by 2010
28. Computational Biology/Chemistry 2013 Nobel Prize in Chemistry   Chemists used to create models of molecules using plastic balls and sticks.   Today, the modelling is carried out in computers. In the 1970s, Martin Karplus, Michael Levitt and Arieh Warshel laid the foundation for the powerful programs that are used to understand and predict chemical processes. Computer models mirroring real life have become crucial for most advances made in chemistry today. Chemical reactions occur at lightning speed. In a fraction of a millisecond, electrons jump from one atomic nucleus to the other. Classical chemistry has a hard time keeping up; it is virtually impossible to experimentally map every little step in a chemical process. Aided by the methods now awarded with the Nobel Prize in Chemistry, scientists let computers unveil chemical processes, such as a catalyst’s purification of exhaust fumes or the photosynthesis in green leaves.
29. 科学研究-结构生物信息学与生物统计研究室 研究方向:结构生物信息学,生物统计学,生物与统计物 理学, 计算生物与化学 M2通道,药物结合位点的 自由能计算研究,与基于 机构的药物设计 人类基因多态性(SNP),分子代谢机理与个 性化药物设计 生物膜系统稀有事件自由能的计算 以及方法学研究 Aβ淀粉蛋白 gx50 100ns MD simulation 药物分子与烟碱乙酰胆碱受体 及Aβ淀粉蛋白的相互作用 科研人员:徐沁副研究员,霍娟助理研究员,赵林静,杨风雷,王斌讲师(后三位为访问学者) 博士生: 顾若虚 硕士生: 黄琳 连鹏 范怀蒙 陈琦 李莉 金超慧 吕建平 马丽娜 赵明珠 孙一凡 王昱焜 赵唐祯 戴昊 覃艳艳 崔守鑫 白燕枝 李全一 张文 微生物代谢国家重点实验室 徐凯 张会圆
30. 课题意义, Purpose of Our Studies CYP 450 and Personalized Drug Design   In clinical treatment, people have quite different response to the same drug, 在临床治疗中,不不同病⼈人对相同药物的反 应不不完全相同,甚⾄至⼤大相径庭。   It has been shown that it has to do with the genetic Polymorphism(SNPs) CYP 450, 研究表明,这些差异是由 于药物代谢相关基因变异,即单核苷酸多态性(SNP)引起的。   The proteins code the genes would bind and metabolize drugs in different fashion, 由于SNP改变了了个体对药物的 代谢及药物结合的⼿手性蛋⽩白的特性,因⽽而决定了了某种药物是 否对个体有效,是否会引起不不良反应等。   Studying interaction of proteins of the SNPs with the drug molecules would gain some insight of the personalized drug design, 研究SNP与药物相互作⽤用关系,以及针对SNP 的个性化药物初步研究,对于临床医⽣生的个性化⽤用药有着重 要的指导意义。
32. ShAnghai Molecular Modeling(SAMM) Invited Review Suzanne Sirois, Rui Zhang, Weina Gao, Hui Gao, Yun Li and Dong-Qing Wei*, “Discovery of Potent antiSARS-CoV MPro inhibitors”, Current Computer Aided Drug Design, 3, 341-352(2007). William Kem, Ferenc Soti, Susan LeFrancois, Kristin Wildeboer, Kelly MacDougall, Dong-Qing Wei, KuoChen Chou and Hugo R. Arias, “The Nemertine Toxin Anabaseine and its Derivative DMXBA (GTS-21): Chemical and Pharmacological Properties”, Marine Drugs 4, 55-273(2006). Kuo-Chen Chou, Dong-Qing Wei, Qi-Shi Du, Suzanne Sirois, Wei-Zhu Zhong, Progress in Drug Development against SARS, Current Medicinal Chemistry, 13, 3263-3270(2006).
33. Sequence to 3D Structure
34. Ramachandran plot for the model 3D structure Ramachandran plot statistics Regions Residue number Percent Most favorable 353 88.7% Additional allowed 41 10.3 % Generously allowed 2 0.5 % Disallowed 2 0.5 % Total residue number 398 100.0 % (non-Gly and Pro)
35. The total energies of system, root mean square deviations (RMSD) for both backbone atoms of CYP2E1 and acetaminophen as well as the hydrogen bond numbers formed by CYP2E1 and acetaminophen during our MD simulations are shown in Supplementary information Figure 10. Although having some fluctuations in the first 1 ns, acetaminophen shows relatively stabilized, forming one or two hydrogen bonds with CYP2E1. In most of time, acetaminophen can form hydrogen bonding interactions with Leu368 and/or Ala438, which seem in charge of the substrate recognition in the binding pocket of CYP2E1.
36. Molecular Docking and Simulation
37. Docking-Preliminary Studies
38. Library of existing drug molecules metabolized by the CYP 450
39. Residue-drug molecule complex
40. Analysis
41. 3D structure modeling of cytochrome P450 2C19 and its implication for personalized drug design JF Wang, DQ Wei et al, Biochemical and Biophysical Research Communications 355 (2007) 513 –519 How could a limited number of proteins could metabolize so many kinds of drugs with different shape, hydrophobicity Hydrogen bonding etc Enough Information for an ADME Model Template :CYP2C9 (PDB code 1R90)
42. 四种药物分⼦子 Name Formula Chemical Name CEC C11H9NO3 3-Cyano-7-Ethoxycoumarin Fluvoxamine C15H21F3N2O2 2-[(5-methoxy-1-[4 (trifluoromethyl)phynyl]pentylidene) amino]oxyethanamine Lescol C24H26FNO4 7-[3-(4-fluorophynyl)-1 -(methylethyl)-1H-indol-2-yl-]-3,5 -dihydroxy-hept-6-enoic acid Ticlopidine C14H14ClNS 3-[(2-chlorophenyl)methyl]-7-thia-3 -azabicyclo[4.3.0]nona-8,10-diene
43. Molecular structures of ligands F F F N H2N OH O N (E) (Z) OH (R) O (S) O H F O Lescol Fluvoxamine Cl N (Z) S N O Ticlopidine O CEC O
44. Results of Molecular Docking Ligands E_ele E_vdw E_binding CEC -4.23 -14.82 -19.05 Fluvoxamine -3.25 -14.84 -18.09 Lescol -7.14 -13.45 -20.59 Ticlopidine -5.53 -13.63 -19.17
45. Drug Molecules and Molecular Surfaces
46. Two SNPs of CYP2C19
47. W120R的分子对接结果 Ligands E_ele E_vdw E_binding A B A B A B CEC -5.00 -3.10 -10.10 -15.12 -15.10 -18.22 Fluvoxamine -3.72 -4.78 -15.75 -14.39 -19.97 -19.18 Lescol -5.87 -4.26 -14.44 -15.79 -20.91 -20.05 Ticlopidine -0.20 -1.20 -14.26 -16.15 -14.46 -17.35
48. W120R分子表面亲水图
49. I331V的分子对接结果 Ligands E_ele E_vdw E_binding A B A B A B CEC -1.74 -4.80 -13.25 -14.30 -14.99 -19.10 Fluvoxamine -3.84 -4.78 -12.02 -17.77 -15.87 -22.55 Lescol -5.29 -9.73 -16.04 -15.22 -21.33 -24.95 Ticlopidine -0.78 -1.27 -14.34 -15.47 -15.12 -16.74
50. I331V分子表面亲水图
51. 2.CYP相关的药物小分子数据库 及其它数据库   CYP-nsSNP 数据库 基于实验数据及文献搜索我们构建出一个关于单核苷酸 多态性(SNPs)主要是错义性突变导致细胞色素酶P450 酶活性发生改变的数据库CYP-nsSNP(missense mutation SNP) (http://cypdatabase.sjtu.edu.cn:8000/.)。
52.   CYP-Meta细胞色素酶代谢数据 该数据库中主要收录了以每一家族为单位,化合 物分成为代谢底物、抑制剂、诱导剂等三类。信息包括化 合物的名称、分子式、分子量及一些物理化学性质如脂水 分配系数logP、解离常数、酶动力学活反应实验数据。
53. CYP-Meta目前包括药物分子查询、CYP代谢底物查询两 种查询功能。用户可以通过输入药物分子名称中的关键字, 并设定CYP的代谢类型进行药物分子的信息查询,并可以通 过点击 每种药物分子后面所 对应的连接查看所关 心的药物分子的详细 信息;用户也可以通 过选择自己所关心的 CYP亚家族,来查看 能被这种CYP亚型所 代谢的药物分子信息。
54. Accurate prediction of SNPs in human cytochrome P450 Work flow
55. Results: performance of 16 models Hybrid features Only Sequences features and Target site features Incorporating with PSSM Incorporating with phylogenetic entropy Incorporating with both PSSM and phylogenetic entropy S1+S2 S1+S2+T1 S1+S2+T2 S1+S2+T1+T2 S1+S2+E1 S1+S2+T1+E1 S1+S2+T2+E1 S1+S2+T1+T2+E1 S1+S2+E2 S1+S2+T1+E2 S1+S2+T2+E2 S1+S2+T1+T2+E2 S1+S2+E1+E2 S1+S2+T1+E1+E2 S1+S2+T2+E1+E2 S1+S2+T1+T2+E1+E2 Combinations C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12 C13 C14 C15 C16 Specificity 0.675 0.675 0.938 0.913 0.750 0.95 0.875 0.938 0.75 0.800 0.838 0.888 0.713 0.713 0.900 0.913 Performance Sensitivity Accuracy 0.825 0.750 0.925 0.800 0.813 0.875 0.850 0.881 0.875 0.813 0.888 0.919 0.875 0.875 0.913 0.925 0.800 0.775 0.875 0.838 0.800 0.819 0.938 0.913 0.913 0.813 0.875 0.794 0.900 0.900 0.850 0.881 AUC 0.830 0.925 0.920 0.885 0.885 0.956 0.912 0.957 0.853 0.865 0.858 0.971 0.899 0.891 0.948 0.976
56. The ROC curves of top 3 models
57. Compare with other prediction methods Approaches Accuracy Align ACE 50.4% ANN 50.2% Motif Sampler 50.1% YMF 51.0% Weeder 50.0% Projection 49.7% Random Forests 57.0% KNNs 54.0% Basic sequence features + DNA sequence based Protein sequence based Sequence features + CG feature + SVM 75.6% SCYPPred 66.7% Our model
58. 模型预测准确性的评价 利用PLS方法构建模 型(SAMM软件)
59. 同时,我 们将构建出 的几种不 同CYP蛋白 酶代谢药物 分子的SVM 预测模型整 合入网络平 台上。
60. 预测ADME网络平台的具体网页形式如下:
61. MD Studies of Cooperativity of Molecule Binding 利用分子动力学方法对细胞色素酶CYP2E1协同位点机理研究(J. Li, D.Q. Wei et al, J. Chem. Info. Modeling) 实验证明CYP2E1可以显示出非典型的动力学作用,而这种非典型的 动力学作用和协同结合(CB)两个小分子在一个酶上有关。我们基于X射 线解析出来的CYP2E1晶体结构,使用分子对接和分子动力学模拟,清晰 的阐释CYP2E1协同的位点的作用机理。
62. 通过比较不同状态下的体系的RMSD及B因子,不同 位点小分子结合的残基变化和各二级结构之间的相关性 ,我们发现协同作用位点位于F116,F207, L210, L368, F498 和F478残基间。
63.   协同位点的氨基酸作用协同小分子在L210和F478左右 活动,这使得两个关键残基和T303的正相关协同性 降低,而这种协同性是催化效率高不可少的。然后, 由于协同小分子的存在,使得活性位点的残基和催化 小分子的波动性增加,稳定性降低,结合自由能升高 ,从而降低了催化效率。
64. Detailed Study for CYP 1A2   The crystal structure of CYP1A2 has been determined in 2007, the first 3D structure of CYP 1A family. 260 Å3 345 Å3 CYP2A6 PDB: 1Z10 CYP1A2 PDB: 2HI4 1385 Å3 The volumes of active sites were calculated using a 1.4Å Probe with the program VOIDOO CYP3A4 PDB: 1TQN
65. Introduction
66. SNPs and Metabolism, MD Studies Experimental observations of CYP1A2 F186L mutation 1. Protein expression level 2. Protein function level 7-ethoxyresorufin Phenacetin
67. Introduction E-helix F-helix Phe186 D-helix 26Å G-helix I-helix Several mechanisms supported by experimental evidences and computational studies (1)  “Coupling mechanism” of surface residues was found in the human purine nucleoside phosphorylase (hPNP) Saen-Oon et al. Biophysical Journal 2008,94:4078
68. (2) “Environment mechanism” was proposed in the site-directed mutagenesis of Thermolysin. 2007,141:835 (3) Tatsumi et al. J Biochem. “Recognition mechanism” was confirmed for surface residue T192 of P450cam based on binding energy. Behera et al. Biophysical chemistry 2008,135:1 Questions:   How does the surface residue F186L affect the enzymatic activity of CYP1A2 ?   Is there a common mechanism for changes in enzymatic activity of CYP1A2 due to the mutation of surface residues ?
69. Computational methods 20 ns 1 1.  2.  3.  Structural comparison To remove ligand from the of substrate-free between crystal structure; Wild-type and mutant To change the side chain from 2 Phe to Leu; To perform 20 ns simulation Molecular docking using AMBER8.0 5 with multiple receptor conformations Multiple-sequence Alignment among CYP1A family 3 The substrate access channel analysis 4 Interactive essential dynamics
70. Results 1. The structural comparison between substrate-free CYP1A2 and F186L mutant A. Secondary structure estimated by DSSP program based on average structure
71. Results B. The differences of active site between CYP1A2 wild-type and F186L mutant (1) 1D-Distance (2) 2D-Surface area (3) 3D- Active site volume * *The active site volume is solvent-accessible volume calculated by Pocket program
72. Results It is apparent, based on the analysis of ligand-free structures that the substitution of Phe186 with Leu did not cause the secondary structure disruption rather than result in change in the geometry of active site.
73.   Two relevant studies of active site of CYPs (1)  It is hard to accurately delimit the active site in P450 structures due to existence of the channels around the cavity. Redberg et al. J.Phys.Chem.B 2007,111:5445 (2) The changes in volume of active site will influence the substrate-binding within active site, and lead to the decrease of enzymatic activity. Sano et al. J. Comp. Chem. 2010,31:2746
74. Results 2. The differences of ligand-bound structures between wild type and F186L mutant obtained by Molecular docking method (1)  The validation of docking method is performed based on crystal structure of CYP1A2 ANF Crystal structure Strucutre-1 Strucutre-2 Bauer et al. Chem Res Toxicol 1995
75. Results (2) Docking results 7-Ethoxyresorufin
76. Results Phenacetin
77. Results (3) Summary
78. Results The catalytic cycle of P450 http://elearn.pharmacy.ac.uk/flash/view/Cytochrome_P450.html How do the reactants enter into the active site or products exit from the active site of P450 ?
79. Results 3 The substrate accessible channel analysis Cytochrome P450 cam (PDB 1AKD) RC Wade et al. BBA-Biomembranes 2007
80. Results (1) The investigation of substrate accessible channels was performed on the 21 structures extracted from the trajectory of 20 ns using the software CAVER. Wild-type
81. Results Mutant
82. Results Wild-type F186L
83. Results (2) The comparison of position between channels and binding poses out of active site
84. Results 4 What’s happened to the structure of F186L mutant resulted in different distribution of channels.
85. Results The interactive essential dynamics based on PCA
86. Results A. The swing motion of D-helix D-helix F-helix E-helix
87. Results B. The movement between up and down of C-D loop D-helix Wild-type Mutant C-helix
88. Results C. The shrinkage and stretch of F-G loop F-G loop
89. Results 5 Multiple-sequence alignment of CYP1A family (contained 22 protein sequences from 15 species) is performed to identify the important residues for the function of CYP1A2
90. Results   The distribution of conserved residues in A. secondary structures of CYP1A2 B. (1). 60/80=75% (2). 29/56=52% D. C.
91. Conclusions   1. Eight substrate accessible channels of CYP1A2 are identified based on the trajectory of simulation.   2. The conserved residues with aromatic ring surrounding the substrate channels may serve as the relay site during the process of substrate accession to active site.   3. The surface residues may affect the function of CYP1A2 by influencing the adjacent channels   4. The substitution of F186L can change the structure of CYP1A2 from open state to close state via protein motions.
92. Car-Parrinello Molecular Dynamics / Molecular Mechanics (CPMD/MM) Simulation study of coupling and uncoupling mechanisms of Cytochrome P450cam Peng Lian, Jue Li, Dong-Qi Wang, Dong-Qing Wei
93. Background The Cytochrome P450 catalytic circle Coupling? Or Uncoupling? Highly exothermic ? Or Thermoneutral ? Altarsha 2009 J. AM. Chem. SOC.
94. Protonation mechanism Three possible proton translocation channels in P450 1st the Asp251-Thr252 channel 2nd the Glu366-Thr252 channel 3rd the Arg299 channel
95. Asp251-Thr252   Asp251-Wat901-Thr252-FeOOH   Asp251-Wat901- - - - - - FeOOH   Asp251-Wat901-WatS-FeOOH Altarsha 2009 J. AM. Chem. SOC.
96. Glu366-Thr252   Glu366-Wat-Wat-Wat-Wat-Thr252 Altarsha 2010 J. Biol. Inorg. Chem.
97. Arg299   Active stie – Arg299(rotation) – protein surface Oprea Proc. Natl. Acad. Sci. USA 1997
98. Some Important Exp. Facts   D251N ~   E366M ~   T252A/G ~ a very large kinetic solvent isotope effects little change of activity Cpd 0 is still formed but no camphor hydroxylation occurs The Coupling and Uncoupling Rate Under the T252X Mutation O2 consuming rate (uM/min) Coupling Uncoupling T252T 1370 100% - T252S 830 ~85% ~15% T252V 420 ~22% ~45% D251N 6 90% - D251N+T252A 5 ~52% ~64% T252A/G 1100 <6% >83% Hishiki 2000 J. Biochem. Altarsha 2009 J. AM. Chem. SOC.
99. Details of the Exp. Findings P450cam Connolly Solvent Excluded Volume of sidechain(A3) Hydropathy Index of amino acid O2 consuming rate (uM/Min) 5-exohy-droxycamphor Coupling (%) H2O2 Uncoupling (%) D251G 21 99 2 D251A 3 89 12 D251N 6 90 - D251N+T252A 5 52 64 T252T(WT)* 48.6 -0.7 1370 100 2 T252T(WT) 48.6 -0.7 1330 96 5 T252T(WT) 48.6 -0.7 1350 97 3 T252T(WT)** 48.6 -0.7 1340 100 - T252T(WT) 48.6 -0.7 1350 100 - T252S 31.4 -0.8 1100 81 15 T252S 31.4 -0.8 830 85 15 T252T-OMe** 64.9 - 410 100 - T252V 58.5 4.2 420 22 45 T252I 75.9 4.5 277 44 40 T252A 24.3 1.8 1100 6 83 T252A 24.3 1.8 1150 5 89 T252G 8.9 -0.4 1090 3 88
100. Possible channels for proton translocation in T252X mutaions.
101. Materials and Methods QM region & H3O+ probes Sys: 24639 atoms 6156 H2O QM: 49 atoms CPMD VDB pp, BLYP MM: Wat, Pr, Ion Gromacs OPLS aa
102. Results and Discussion Coupling (Group B) Uncoupling (Group E)
103. Energy Profile Coupling Uncoupling Position A B C D E △EQM -208 -235 -131 -111 -106 (kcal/mol) In former QM-only studies, coupling is thermodynamically more favorable than uncoupling by 6 kcal/mol. However, in our QM/MM study, coupling is much more favorable than uncoupling by ~100 kcal/mol. That’s mainly owing to the presence of protein environment.
104. The time-dependent bond length and bond order evolution in simulation. The reaction is a dynamic process. And O-O cleavage and proton transfer are coupled. Time (ps)
105. Electrostatic structure during each reaction Coupling Uncoupling
106. Conclusion   Both coupling and uncoupling are highly exothermic.   The enzyme environment makes coupling much more favorable       than uncoupling. O-O cleavage and proton transfer are associated for coupling. The Fe-O cleavage associated with protonation as well during uncoupling. The PCET (proton-coupled electron transfer) is confirmed by our simulations. Pathway through which the second proton is delivered play a key role in the determination of coupling and uncoupling.
107. Vast Growth in (Structural) Data... but number of Fundementally New (Fold) Parts Not Increasing that Fast Structural Prediction– Structural Bioinformatics is in high demand Total in Databank New Submissions New Folds
108. Bioinformatics schematic diagram for rational drug design
109. Major Application I: Designing Drugs  Understanding How Structures Bind Other Molecules (Function)  Designing Inhibitors  Docking, Structure Modeling (From left to right, figures adapted from Olsen Group Docking Page at Scripps, Dyson NMR Group Web page at Scripps, and from Computational Chemistry Page at Cornell Theory Center).
110. 药物发展现状/Current Status of Drug Discoveries:   1500-1600万化合物/15-16 million compounds   500多个靶点/ 500 targets   7000多个药物/ 7000 drugs 后基因组研究计划/The Post Genome Age   1500-1600万化合物/15-16 million compunds   2000个靶点/2000 targets   2万~3万个药物/ 20,000-30,000 drugs ⻄西药市场(chemical drugs):98.5% 中药市场(Chinese Drugs):1.5%
111. An Integrated Biochemoinformatics System for Drug Discovery Reference: Frontiers in Biochip Technology, Springer, 2006, L. Shi, Z. Su et al
112. 屠呦呦,中国人的骄傲 中国中医研究院终身研究员兼首席研究员 ,多年从事中药和中西药结合 研究,突出贡献是创制新型抗疟药青蒿素和双氢青蒿素。1972年成功提取 到了一种分子式为C15H22O5的无色结晶体,命名为青蒿素。2011年9月 ,因为发现青蒿素——一种用于治疗疟疾的药物,挽救了全球特别是发展 中国家的数百万人的生命获得拉斯克奖和葛兰素史克中国研发中心“生命 科学杰出成就奖”。2015年10月,屠呦呦获得诺贝尔生理学或医学奖,理 由是她发现了青蒿素,这种药品可以有效降低疟疾患者的死亡率。她成为 首获科学类诺贝尔奖的中国人。
113. Drug Design with Database of the Effective Components of Chinese Medicine, System One Disease Biology Approach TCM One to Many Therapy Correlation Target 1 Target 2 Target 3 Many to Many Compound 1 Compound 2 Compound 3 Many to Many TMC 1 TMC 2 TMC 3 TMC- Traditional Chinese Medicine Many to Many Dose 1 Dose 2 Dose Doses of Traditional Chinese medicine Graphing by Sun Liya 中 医 疗 法
114. Traditional Chinese Medicine Database(TCMD) Informatics Platform of Effective Compounds Web浏览 Web Viewer (表现层) Web服务器 (逻辑层) Web Server Oracle数据库 Oracle Database (数据层)
115. image 查看分子信息 Molecule Viewer 10000 molecules so far
116. Image 对比查看 Molecule Comparison Correlation with Target Databases and Doses Virtual Screening, Mining
117. 数据库表结构-Database Structure CHMED MOLDATA MOL2CHMED ID MOLECULARID CHMEDID MOLECULARID CASID RXLINK CHMEDID LATINNAME CHINESENAME DISTRIBUTING USEFULPART TASTE CHARACTER TOXICITY FUNCTION TARGET APPEND DRUGTARGET TARGET_SNP ID TargetID SNPSequence TARGETS TargetID TargetName Sequence NumberofResidues GenbankID SwissProtID pdbFile GeneSequence SNP_DRUG TargetID DrugID MOLECULAR ARTICLE MOLCHR MOLECULARID CHEMFORMULA IUPACNAME MOLWEIGHT MELTING H2OSOLUBILITY STATE LOGP PKA SDF INCHI LASTUPDATE JOB PROC MOLECULARID SMILESSTRING CREATEDATE LASTUPDATE SLCS ID TYPE SRCID OBJID LCSSTR SCORE LASTUPDATE ARTICLEID MOLECULARID TITLE AUTHOR ABSTRACT INDEXID PUBDATE JOBID MINCPU MAXCPU STATUS PRIORITY PTIME TYPE PROCID NODEID JOBSTATUS STARTTIME JOBGID UPLOAD PRIORITY Allele DrugName ReactionType DrugEffect ActivityOf Enzyme ExperimentalApproaches EnzymeExpression Vm Kcat Km ratioOfVmAndKm RatioOfKcatAndKm TheUrinaryMolarRatio Reference SNP NODE NODEID CPU MEMORY IPNPORT FREE MAXFREE DRUG DrugName DrugDetail MolecularFormula Structure Remark Allele Family Ethnic ChangeOfDNA GeneRef ProteinName ChangeOfProtein ProteinRef
118. 课题技术方案:多靶点预测的SEA方法 SEA (Similarity Ensemble Approach)– 分子相似性系宗收索方法---基于候 选化合物与靶标配体相似性的统计预测模型, 多靶标分子收索的理论基础:通 过其已知配体与候选化合物匹配的评分机制 目前尚处在起始阶段, 还有许多尚待解决的生物信息学问题, 如分子指纹的确 定,匹配技术,打分函数以及非线性统计模型的建立等 Keiser, M. J. et al. Relating protein pharmacology by ligand chemistry. Nature Biotechnology 25, 197( 2007);and Nature, 462, 175(2009).
119. 课题技术方案:多靶点预测的SEA方法 函数µσ需要优化,统计模型需要进一步研究   BLAST algorithms •  Tanimoto coefficients (Tc) •  Raw similarity score •  Z-score •  E-value
120. 课题技术方案:基于“配体-靶点”数据库的数据挖掘—Network Pharmacology 构建“配体-靶点”相互作用网 络 基于“配体-靶点”相互作用网 络开发新的药物靶点预测方法
121. 课题技术方案:多靶点抗阿尔兹海默症药物设计 多条调控通路; 多个药物靶点:烟碱乙酰胆碱受体,Aβ淀 粉蛋白,α、β、γ分泌酶; 分子对接虚拟筛选,靶点预测,筛选法
122. 自有知识产权:软件著作权 Software copyrights                               基于Maccskey 分子指纹搜索软件V1.0 2009SR030823 2009.08.05 天然中药小分子数据库软件V1.0 2009SR042815 2009.09.27 基于神经网络的药物代谢预测软件V1.0 2009SR056211 2009.12.02 基于SVM 的细胞色素P450 酶SNP 预测软件V1.0 2010SR019961 2010.05.01 基于药物分子数据库的分子指纹搜索软件V1.0 2010SR042160 2010.08.18 细胞色素酶P450 氨基酸突变酶活性改变数据库软件V1.0 2010SR042161 2010.08.18 药物、药物靶标数据库及其网络搜索平台软件V1.0 2010SR042163 2010.08.18 基于蛋白质信息的DNA 结合位点预测工具软件 2010R11L055771 2010.08.09 基于支持向量机的药代预测软件 2010R11L055833 2010.08.09 蛋白质活性位点抽取软件 2010R11L055755 2010.08.09 基于Convex Hull 的蛋白质活性位点搜索软件 2010R11L055766 2010.08.09 SNP 预测的序列数据处理软件 2010R11L055742 2010.08.09
123. 魏冬青, 教授,上海交大   Springer期刊“Interdisciplinary Sciences”主编,“J. Atomic and Molecular Physics” managing editor, “Molecular Simulation”等9家杂志编委   主持并完成国家863项目:“药物代谢酶SNPs与药物的类药性一体 化预测软件的研究与开发”(2010年底结题), 主持一项,完成4 项国家自然科学基金项目。   共发表150篇SCI科研论文章,11篇邀请评述文章,5本专著,1最 高单篇他引204次,130次以上4篇,70次以上14篇,50次以上20 过 4000次,H factor 40. 1.  IF: 3.23 Hugo R. Arias, Ruo-Xu Gu, Dominik Feuerbach, Bao-Bao Guo, Yong Ye, and Dong-Qing Wei*, Biochemistry, 50, 5263–5278 (2011), H.R.A. 2.  IF: 3.23 Hugo R. Arias*, Ruo-Xu Gu, Dominik Feuerbach,and Dong-Qing Wei,Biochemistry, 49,4169-4180 (2010), H.R.A. 顾若虚共同第一作者.. 3.  IF: 8.58 Ruo-Xu Gu, Limin Angela Liu, Dong-Qing Wei*, Lei-Liu, Hong-Liu and Jian-Guo Du, JACS, 133 (28), 10817–10825(2011). 4.  7.33-Jing Chang, Peng Lian, Dong-Qing Wei*, Xiang-Rong Chen, Zizheng Gong and Qingming Zhang, Phys. Rev. Lett. 105, 188302 -188305(2010). 5.  IF:4.351 - Peng Lian, Dong-Qing Wei*, Jing-Fang Wang*, Kuo-Chen Chou, PLoS ONE , 6, e18587(2011). 6.  Cited 66 times, IF: 2.59 - Jing-Fang Wang, Dong-Qing Wei*, Lin Li, Si-Yuan Zheng, Yi-Xue Li, Kuo-Chen Chou, BBRC, 355, 513–519 (2007).
124. 抗⽼老老年年痴呆的药物设计 Drug Candidates From Chinese Medicine Database Against Alzheimer's Disease(AD)
125. Background of The Receptor Cys-loop Ligand   Alpha7 烟碱⼄乙酰胆碱受体(Nicotinic acetylcholine Gated Ion Channel receptor(nAChR ))是⽼老老年年痴呆病症的药物靶点之⼀一   It’s a membrane of Cys-loop ligand gated ion channel superfamily A /   组成nAChR的亚基⽬目前已经发现了了⼗十七种(α1-10,GABA β1-4, γεδ), Glutamate/Glycine/ nAChR nAChR⼀一般由五个亚基组成同源或者是异源多聚体 5-HT3/P2X   根据192-193两个位点上的氨基酸是否为连续的两个半胱氨酸,把 这些亚基分成α型与⾮非α型,不不同的亚基之间有⾮非常⾼高的相似程度 (40%-70%) Neuro nAChR Neuromuscal nAChR Alpha7 Receptor
126. Background Info. of the receptor
127. The role of α7 receptor   胆碱能神经元的概念   实验结果:在⽼老老年年痴呆患者体内,⼄乙酰胆碱的量量明显降低   α7 受体是β样淀粉蛋⽩白的受体,但β样淀粉蛋⽩白的毒性可以被⼀一些⼩小 分⼦子消除,⽐比如GTS-21   能够和nAChR受体相互作⽤用的⼩小分⼦子有很多种,但是同⼀一种⼩小分⼦子 对不不同受体的作⽤用差异很⼤大,并且⼀一般没有选择性,   我们所需要的是具有⽐比较⾼高的特异性的agonist OMe OMe H N N
128. The screening method DMXBA(GTS-21) is used as a template molecule The homology model of about alpha7 dimer is used in 10000 the screening Similarity search and flexible alignment are conducted to exclude the molecules 590 don’t match the template Lipinski’s rule of five and the molecule volume to exclude those too large 100 or too polarity 43 dock Drug 7 candidate Agonist? Analysis of hydrogen bond and hydrophobic and hydrophilic 21 interaction and binding energy 9 MD
129. U vdw U ligand U binding Numbers of H bond -1.55 -20.49 66.72 -22.04 1 47.36 -1.91 -16.09 65.36 -18.01 3 gx-50 72.03 3.06 -9.33 78.31 -6.28 3 cana-36993 82.91 1.63 -5.26 86.54 -3.63 1 gx-51 365.32 -1.44 4.34 362.42 2.90 2 open3d-41249 87.35 -0.70 7.72 80.33 7.02 1 gx-52 75.27 -0.71 9.97 66.02 9.25 4 133573 129.80 0.49 16.79 112.52 17.28 1 aido5682 104.66 -0.45 22.26 82.85 21.81 2 198860 87.38 3.79 18.06 65.53 21.86 1 aido5758 112.78 4.40 17.90 90.48 22.30 1 80835 137.42 -1.10 33.31 105.21 32.21 1 gx-180 86.87 -3.60 39.34 51.14 35.74 4 open3d-19047 133.22 -2.82 40.50 95.54 37.68 1 open3d-99008 179.72 1.36 61.34 117.02 62.70 3 open3d-99662 172.55 -3.74 82.13 94.16 78.39 1 aug00-2d-14717 205.76 -0.84 84.10 122.50 83.26 1 aido42661 228.06 1.39 97.44 129.22 98.83 1 open3d-51265 280.58 1.83 101.66 177.10 103.49 3 open3d-60247 651.67 1.13 111.14 539.40 112.27 2 DMXBA 222.64 0.49 120.35 101.80 120.84 2 Molecule code U total aido1698 44.68 open3d-77464 U ele
130. OMe OMe H N N
131. OMe H H3CO OMe N MeO O H3CO N MeO N N Gl n117 OMe 3. 19N H2N OMe N O N N OMe Leu119 3. 33 3. 49 3. 24 OMe Trp149
132. wgx-50 – The best molecule found in TCM Database N-(2-(3,4-DIMETHOXYPHENYL)ETHYL)-3-PHENYLACRYLAMIDE A molecule from Zanthoxylum Bungeanum (Sichuan pepper)
133. Zanthoxylum Bungeanum (Sichuan pepper)
134. 麻辣的感觉Spicy , Hot and Numb 花椒(学名:Zanthoxylum bungeanum Maxim.),别名:檓、大椒、秦椒、蜀椒、川椒,大红袍 或山椒。为芸香科、花椒属落叶灌木或小乔木,可孤植又可作防护刺篱。《神农本草经·注》中 记载:“始产于秦。其果皮可作为调味料,并可提取芳香油,又可入药,种子可食用,也可加工制 作肥皂。”原属野生。花椒可除各种肉类的腥气;促进唾液分泌,增加食欲;使血管扩张,从而起 到降低血压的作用。花椒形状球形,椒皮外表红楬色,晒干后呈黑色。适宜人群一般人群均能食用 ,孕妇,阴虚火旺者忌食。有龟裂纹,顶端开裂。果实呈圆形,绿豆大小,其外皮是一种常用香料 。内含种子一粒,圆形,有光泽。果实成熟时红色或紫红色果皮叫椒红,种子叫椒目,都是中药材 ,家庭常用调味品,多见于海拔2500米的坡地,也有栽种。花椒含有柠檬烯、香叶醇、异茴香醚 、花椒油烯、水芹香烯、香草醇等等挥发性物质。具有独特浓烈香气。花椒按大小分为大椒,(大 椒又称大红袍、狮子头),其果粒大,色艳红或紫红,内皮呈淡黄色。),按采收季节又分为秋椒 和伏椒。 【花椒的功效与作用】:温中止痛;除湿止泻;杀虫止痒。主脾胃虚寒之脘腹冷痛;蛔虫腹痛;呕 吐泄泻;肺寒咳喘;肺寒咳喘;龋齿牙痛;阴痒带下;湿疹皮肤瘙痒。用于脘腹冷痛,呕吐泄泻, 虫积腹痛,蛔虫症;外治湿疹瘙痒。 Anti-inflamation, infection!!
136. O R1 MeO N MeO Original R1: -CH2-CH2Modified R1: (1) -CH2- ; (2) -(CH2)3- ; (3) -(CH2)4- ; (4) -CH(CH3)CH2- ; (5) -CH2CH(CH3)- ; (6) -C(CH3)=CH- ; (7) -CH=C(CH3)- N MeO R3 R2 MeO Original: R2=O, and R3= (8) R2=O, R3= N Modified: ; (9) R2=O, R3= (11) R2=O, R3= NH2 S ;(10) R2=O, R3= ;(12) R2=S, R3= N ;
137. No. U_ele U_vdw U_binding U_ligand H bond gx-50 3.06 -9.33 -6.28 78.31 3 1 0.07 -7.93 -7.86 78.83 1 2 -0.88 18.17 17.29 87.73 1 3 -0.17 115.03 114.86 107.84 1 4 1.58 1.63 3.21 75.57 4 5 0.58 35.59 36.17 110.40 1 6 -1.86 12.32 10.46 74.09 0 7 0.08 511.83 511.91 99.88 1 8 -0.98 -13.50 -14.47 81.44 1 9 3.54 23.57 27.11 130.15 2 10 1.07 -4.79 -3.73 67.44 0 11 -2.44 63.91 61.47 78.56 4 12 2.55 -14.77 -12.22 78.83 2
138. 小鼠巨噬细胞(Mouse Macrophages)预实验 Ligand of Alpha7 Inhibits The Release of Inflammatory factors Induced by LPS LPS (Lipopolysaccharides)
139. 小鼠巨噬细胞预实验 受体(α7)抑制剂 α7 Inhibitor LPS LPS gx-50 炎性因子 Inflammatory factor gx-50 炎性因子 inflammatory factor
140. Fig 1 900 800 * I L - 1β r el eas e( pg/ ml ) 700 600 * ** 500 ** 400 300 200 100 0 BLANK DMSO LPS 0.001 concent r at i on 0.1 10 1000 nM Fig1. 随着浓度增加对LPS诱导的细胞因子表达抑制作用增强
141. 700 600 LP S+ Me LP y+ S gx -5 0( LP 10 S+ nM BT ) X+ gx -5 0( 10 nm LP ) S+ gx -5 0( 10 nM ) BL AN K I L - 1β r el eas e( pg/ ml ) Fig 2 900 800 * * 500 400 300 200 100 0 Fig2. 受体抑制剂能够阻断gx-50对细胞因子表达的抑制作用
142. 小鼠巨噬细胞预实验-MTT Pre-experiment in macrophage (MTT) Cell viability (%) 120 100 80 60 40 20 0 0M 10-8M 10-7M 10-6M 10-5M gx-50对巨噬细胞活性影响 10-4M
143. 结论- Conclusion   gx-50能抑制LPS诱导的巨噬细胞炎性因⼦子分泌泌, 起到抗炎作⽤用 It has been shown that gx-50 could inhibit the release of inflammatory factor induced by LPS   gx-50的抗炎作⽤用是通过α7nAChR发挥作⽤用的 The action is made through α7nAChR gx-50 a possible agonist( a leading compound) !
144. 新生大鼠皮质神经元细胞分离 NSE抗体免疫 组化鉴定纯度 不同浓度gx-50处理 MTT法检测细胞 活性 gx-50+Aβ 不同浓度Aβ处理 FCM检测细胞凋亡 α银环蛇毒素预处理 gx-50+Aβ FCM,TUNEL,DNA LADDER检测凋亡 RT-PCR,WB检测凋亡相关基因Bcl-2和Bax
145. Aβ解聚实验-原子力显微镜 Aβ depolymerization- Atomic Force Microscope(AFM) 通过原子力显微镜观察发 现gx-50在体外可以直接促 进Aβ的解聚 AFM detection demonstrated that gx50 could induce Aβ depolymerization.
146. The Inhibitory Effect of wgx-50 on brain amyloid plaques The relative plaque area showed that wgx-50 administration could reduce A plaques deposition by approximately 60% (**p < 0.01). amyloid- protein precursor transgenic (APP-Tg) mice
148. Neuron protection by wgx-50 from Aβ -induced apoptosis
149. To further understand the anti-apoptotic effect of wgx-50, we examined two principal regulators of apoptosis, anti-apoptotic gene bcl-2 and pro-apoptotic gene bax. These apoptotic proteins were measured in neurons treated with 10M A42 in the absence or presence of 10−6M wgx-50 Real Time PCR GAPDH:glycerald ehyde-3 phosphate dehyd rogenase Western Blotting
150. The abnormal rise of intercellular Ca2+ content is one of the mechanisms of A-induced neuronal toxicity. Therefore, we also examined the effect of gx-50 on neuronal [Ca2+]i by confocal microscopy.
151. 2NNT_AD43不同时刻构象 wgx-50可结合于β淀粉样蛋白原 丝多肽五聚体的三个位点A、B 、C。由于结合不同位点后会导 致五聚体不同的构象变化,这里 使用结合前的五聚体结构以作示意 B I31 M35 I32 L34 C V18 A F20 A L34 B I32 L34 K28 wgx-50 D23 Fibril axis wgx-50深入结合 于位点C导致五聚 体单元β片层结构的 严重形变,使多肽 链间相互作用能下 降,促进其解聚
152. 最显著的成果 Initial Conformations GX50经AutoDock对接与2BEG结合
153. 最显著的成果 10 Neno seconds 2BEG_GX50_1C在平衡 后的10ns内的平均结构
154. gx50对东莨菪碱诱导的小鼠急性记忆获得性障碍的干扰作用,左图为穿越平台的次数(空间记忆),右图为到 达第一象限的时间(方向位置) 为进一步评价gx50对老年痴呆症的作用,我们采用了ppr转基因小鼠作为老年痴呆症的模型小鼠。选择了鼠 龄为10周的同窝的ppr+雄鼠及ppr-雄鼠,随机各分为2组,双盲法注射gx50或生理盐水10天后,Morris水迷宫 法评价gx50对老年痴呆的影响。结果如图6所示,gx50无论是对小鼠穿越平台的次数(空间记忆),还是寻找 有效路径的时间(方向位置),都可以有效地改善ppr+转基因小鼠的记忆能力。 图6、gx50对ppr+转基因小鼠记忆力的影响,左图为穿越平台的次数(空间记忆),右图为到达第一象限的时间(方向位置), Group1 ppr-/water 生理盐水处理的转基因阴性小鼠,Group2 ppr-/gx50 gx50处理转基因阴性小鼠,Group3 ppr+/water 生理盐水处理转基因阳性小鼠,Group4 ppr+/gx50 gx50处理转基因阳性小鼠,上述实验证明,gx50 可以通过激动α7乙酰胆碱能受体及解聚β-amyloid等途径,有效地治疗由β-amyloid诱导的老年痴呆症。
156. 水迷宫实验-Morris Water Maze 通过对东莨宕碱模型小鼠和APP转基 因小鼠进行水迷宫实验,结果表明 gx-50可以明显的改进小鼠的记忆力 Morris Water Maze test demonstrated that wgx50 could improve the memory ability of dementia mice
157. 皮肤衰老机制Skin Aging Mechanism Ø 内源性生理衰老 endogenous physiological senescence ① 自由基引发 ② 线粒体损伤 ③ 端粒缩短 ④ 基因表达控制 ⑤ 器官水平的衰老 ⑥ 外源性环境衰老 ⑦ 紫外线引起肌肤光老化 Exogenous ROS:Reactive Oxygen Species 16/12/9 157
158. 自由基理论 Free Radicals ①对核酸的损伤:碱基的双键,戊糖的氢 ②对蛋白质的损伤:胶原、弹性蛋白和表皮 生长因子受体蛋白交联变性 皮肤变薄 、起皱,弹性降低,细胞生长变缓. ③对糖的损伤:黏多糖、透明质酸、糖醛 与DNA、RNA、蛋白质发生进一步交联变性 ④对脂质的损伤:膜上的不饱和脂肪酸 脂质过氧化物及其代谢物丙二醛 与蛋 白质核酸交联形成脂褐质(老年斑) 16/12/9 158
159. 线粒体理论 Mitochondrial Ø 线粒体的氧化损伤导致基因突变是人体 衰老与退行性疾病的主要原因。 线粒体的氧化损伤 自由基聚集 16/12/9 氧化磷酸化被抑制 氧化压力升高 能量产生减少,细胞凋亡 159
160. 端粒理论 Telomere Theory Ø 随着细胞分裂端粒逐渐缩短,短至一定 程度则启动停止分裂信号,正常的体细 胞即开始衰老死亡。 16/12/9 160
161. 器官水平的衰老理论 Ø 衰老时神经元及相关激素的功能减退导 致全身功能的退行性变化。 ①激素降解率降低,通过反馈作用引起激 素分泌减少。 ②内分泌腺分泌的激素原发性减少。 ③激素受体数量减少且敏感性降低。 ④内分泌系统在调节酶合成方面功能衰退 16/12/9 161
162. 紫外线引发的光老化 UV Ø 抑制Bcl2和Brn3a基因的表达,使自 发永生化角质形成细胞 (HaCaT细胞) 凋亡 Ø 透明质酸量下降及成纤维细胞损伤 Ø 影响Ⅰ型胶原形成,Ⅲ型胶原相对增加 Ø 氧化表皮中的还原黑色素而直接晒黑 16/12/9 皮肤 162
163. 基因表达调控 Ø 生物成年以后,基因组内特定的遗传衰老 程序启动,按时激活褪变过程,逐渐展开 ,导致衰老。 ①随着年龄增长,修饰基因逐渐丧失 ②DNA自我修复功能出现障碍 ③基因调控的改变 16/12/9 163
164. NF-kB途径Pathways 16/12/9 2007. Dec .15th. Genes & NF-kB基因的活跃程 度与年龄增长成正比 ,将NF-kB阻断两 周后,基因表达谱和 组织特征都显示,老 鼠皮肤已恢复到一只 164 年轻老鼠的状态。
165. WGX50 NFκB途径 进入细胞核 调控基因表达 16/12/9 165
166. 二、WGX50抗皮肤衰 老的应用 Ø 内源性生理衰老 抗氧化作用? ① 自由基引发 ② 线粒体损伤 WGX50 ③ 端粒缩短? NF-KB途径! ④ 基因表达调控 皮肤抗衰老 增加胶原蛋白! ⑤ 器官水平的衰老 Ø 外源性环境衰老 抵抗光辐射? 紫外线引起肌肤光老化 16/12/9 166
167. 三、WGX50发现的机缘 与巧合 Discovery of its function on skin anti-aging You find more beauties in Sichuan since ancient days with flower face and jade-like skin, warmness and smoothness 16/12/9 天现数 然它据 存在库 在花中 椒比 中对 竟时 然发 congealed fat 有的它教经 显保对授 著护于等 作和皮研 用抗肤究 衰组发 老织现 Zhongdong Qiao 的由据魏 抗计生冬 衰算物青 老机信教 药设息授 物计学团 产原队 生理根 167 自 古 川 蜀 多 佳 丽 面 若 桃 花 , 肌 如 凝 脂
168. 四、研究进展 Progresses 组织水平 Tissue Le vel Made H H3CO H3CO N O gx50 WGX50的化学结构 16/12/9 细胞水平 Cell Level 168
169. 组织水平 Tissue Level 预实验 Pre-Tests 扩大实验 Expanded 16/12/9 169
170. 组织水平 Tissue Level Apply WGX-50 For Two Weeks Frozen(Cyostat) Section Hematoxylin-eosin Staining Extract Proteins, Perform EMSA, Western Blot 16/12/9 170
171. 组织水平 Tissue Level •  20个月龄小鼠背部皮肤一 侧涂抹溶于DMSO和甘油中 的WGX50,另一侧只涂DMSO 和甘油做为对照,涂药两 周后,剪取皮肤制作冰冻 切片,进行HE染色 •  Apply WGX-50 desolved in DMSO and Glycerol to one side of 20 month mice, various tests are made after two weeks 16/12/9 171
172. HE染色结果 Frozen(Cyostat) Section 图一 皮肤组织冰冻切片HE染色4X 实验组 Exp. 16/12/9 对照组 Control 172 16
173. 动物组织水平 预实验 图二 皮 肤 组 织 冰 冻 切 片HE 染色 10X Left 左 侧涂 药皮肤 With WGS-50 右侧 对照 皮肤 Witho ut WGX-50 16/12/9 173 17
174. 组织水平扩大实验 Extended Tests with 7 Mices 对7只1年3个月龄的小鼠进行了涂药实验,再 进行冰冻切片的制作。 图三 皮肤组织冰冻切片HE染色10X 实验组 16/12/9 174 对照组 19
175. 动物组织水平 HE染色结果: 表一 HE染色结果皮肤形态对比表 表皮 真皮 16/12/9 涂有GX50实验组 未涂GX50对照组 角质层 较平整 凹凸不平 基底层 致密,数目较多 稀疏,数目较少 黑色素细 胞 数目较多,沉积较少 数目较少,黑色较 深 真皮乳头 数目较多,突起较多 数目较少,较平坦 胞外基质 排列规整,分布致密 排列无序,分布松 散 毛囊 较多 较少 175 18
176. 免疫组化染色 Immunohistochemical staining of collagen Collagen Protein 胶原蛋白是一种天然蛋白质,它在皮 肤中的内含比例和分布,会随着不同生理环境而变化,特 别是与年龄相关的生理变化。使用了胶原蛋白的抗体来对 涂药处理实验组和对照组的皮肤进行免疫组化染色 图四 皮肤组织冰冻切片胶原蛋白免疫组化染色10X 实验组Experimental 16/12/9 对照组Control 176 20
177. 动物组织水平 Collagen 表一 皮肤真皮层胶原蛋白密度图像分析结果 使用GX50之后,皮肤的真皮层中,胶原蛋 白颗粒的密度对比同期未用WGX50的对照 组皮肤,增长16.68% Density of Collagen Protein increase 16/12/9 177 21
178. 动物组织水平NF-κB Path Way •  对NF-κB(核转录因子)通路的作用 •  EMSA检验皮肤组织中NF-κB入核量即调节活性变化 对 照组 空白 NF-κB complex 迟滞带 实 验组 NF-κB 探针条带 WGX50对皮肤中NF-κB活性的影响的EMSA图 Spectrum 16/12/9 178 22
179. NF-κB-down regulated 24.4% •  对NF-κB通路的作用 •  通过Image J软件对NF-κB与探针的复合物滞后 带的灰度值分析后,得知实验组采用WGX50处 理后,NF-κB信号通路的活化被抑制了24.4% ,继而使得入核后调节下游基因表达的NF-κB 减少 NF-κB 16/12/9 对照组 Control 实验组 Exp. 179 23
180. Myogen •  对NF-κB通路的作用 (2)为了检测GX50是作用在该通路的哪个步骤, 对肌 浆蛋白(Myogen)进行Western Blot分析,检测IκB-α 的磷酸化状态,以看家基因(house-keeping gene) GADPH作为内参 对照组 实验组 p- IκB-α GADPH 16/12/9 图七 Western Blot 分析GX50对皮肤组织中NF-κB通路的影响图 180
181. 细胞水平 1. 表皮细胞的原代培养及鉴定 2. 加药处理 3. 提取核蛋白、浆蛋白和膜蛋白 4. EMSA检验NF-κB的活性 5. Western Blot检验I-κB 磷酸化的程度 6. 检验GX50与表皮细胞α7乙酰胆碱受体的结 合能力 16/12/9 181
182. Cellular Level with epidermal cel l表皮细胞的鉴定 使用单克隆抗体一抗PCK(Mouse Anti-Pan Cytokeratin)广谱角蛋白 一抗标记角质形成细胞表面标志,通过免疫荧光染色检测表面标记 情况,鉴定KC的纯度。 图八 角质形成细胞细胞免疫荧光染色鉴定图 16/12/9 A 普通光 182 B 荧光 26
183. 对六孔板中的细胞分组并加药处理2h 16/12/9 GX50 1µmol /ml Bl ank LPS 100ng 183 /ml
184. EMSA检验表皮细胞NF-κ B的活性 未处理 GX50 16/12/9 LPS Blank 184
185. EMSA检验表皮细 胞NF-κB的活性 WGX-50 Blank LPS(炎性因子 16/12/9 185
186. Western Blot检验I-κB 磷酸 化水平Phosphorylation GAPDH I-ΚB p- I-κB 16/12/9 GX50 Blank LPS 186
187. Western Blot检验I-κB的磷酸化水平 16/12/9 187
188. Western Blot检验p-I-κB的磷酸 化水平 16/12/9 188
189. WGX50与α7乙酰胆碱受体 的结合实验 Binding Exp. 16/12/9 189
190. 实验证明, experiment show Ø  WGX50通过抑制IκB的磷酸化,使NF-κB不易从其 抑制因子IκB上解离下来,从而减少了NF-κB的入 核量,降低了NF-κB的活性。 Ø  WGX50 inhibits Phosphorylation of IκB, so that NF -κB does not dissociate from its inhibitory factor IκB, reduces its activity Ø  其结合受体可能为a-7乙酰胆碱受体,a-7 could be a receptor of wgx50 16/12/9 190
191. β-galactosidase(半乳糖苷酶)staining identification of cell senescence (10 ×) Hydrogen Peroxide H2O2 WGX50 Control group A, WGX50 alone treatment groups B, hydrogen peroxide alone treatment group C, adding WGX50 after hydrogen peroxide treatment group D, adding hydrogen peroxide after GX50 treatment group E, the hydrogen peroxide alone treatment group showed obvious blue, which is β-galactosidase, cells were oxidized and aging
192. Cellular Metabolic Rate Marker MDA Determination of malondialdehyde (MDA) secretion分泌in WGX50 treated cells
193. WGX50’s effective regulation of collagen-related cytokine secretion分泌 WGX50’s regulation of cell secretion of transforming growth factor (TGF-β), which increase transforming growth factor (TGF-β) secretion in a certain period of time, but will not promote its long-term over-expression
194. 抗紫外线-标记物染色
195. WGX-50,紫外光对胶原蛋白,活性 氧(ROC),MDA的影响 ROC refers to Reactive Oxygen Species, 荧光强度与其正相关
196. WGX-50 non-irritating to the rabbit eye test Take 3 rabbit weighed 2.0 - 2.5kg, ensure that each rabbit’s cornea, iris, conjunctiva is normal before introducing drug. We introduce 0.1mL of 10-6mol / L solution GX50 to the left eye conjunctival sac of rabbits, and 0.1mL phosphate buffer to the right eye as a control. After introducing drug, rabbit eye is passively closed for 8-10s, we introduce drug four times a day for 7 days and observe the eye condition before each dose, and observe the responses of cornea, iris, conjunctiva at 4,12,24,48,72 h after the last time of drug introduction. During introducing the drug, we observe two times a day, take the highest two points as score of that day, after the teat stopped, we continue to observe the eye reaction for 3 days. The results show that, GX50 had no irritating effect on rabbit cornea, iris and conjunctiva.
197. 结论 Ø GX50可通过a-7乙酰胆碱受体来调节 与皮肤衰老相关的NF-κB通路,使得 皮肤保持年轻状态,对于已发生衰老 的皮肤则有一定的年轻化作用。 16/12/9 197
198. 今后的工作计划 Ø 我们现在的研究都是在小鼠身上进 行的,这些结果并不能完全代 表GX-50对人体的影响。换句话说 ,GX-50对人皮肤的影响是什么, 还需要我们继续深入研究。 16/12/9 198
199. 16/12/9 199
200. Gx-50可以与神经细 胞膜表面的α7胆碱能 受体结合
201. 肖申克的救赎 Gx-50可以使小鼠的皮肤 变年轻GX-50 makes skin younger(20月龄57BL6/j 雌鼠)
202. NF-κB途径 皮肤组织水平 空 NF-κB complex迟滞 带 白 对 照 组 实 验 组 探针条带 control 表皮细胞水平 Blank GX50 LPS 阴性 Wgx-50
203. 心电图-electrocardiogram 2.5min gx-50处理组 gx-50 treatment group 麻醉剂对照组 Anesthetic control group 12min 起 止 标记 最大值 最小值 平均值 心率 (s) (s) 起 止 标记 最大值 最小值 平均值 心率 (s) (s) 平均值 标准差 0.093667 0.055199 -0.511 0.002828 -0.36633 0.004028 312.341 37.88454 平均值 标准差 0.030667 0.040549 -0.476 0.012961 -0.37233 0.007318 206.924 74.03907 20min 148.609 150.031 711.87 1200.023 713.456 1201.885 0.024 -0.513 -0.364 361.446 0.098 -0.513 -0.363 306.346 148.446 150.242 712.155 1205.261 715.065 1208.801 0 -0.525 -0.375 241.845 0.043 -0.482 -0.377 172.697 0.159 -0.507 -0.372 269.231 0.073 -0.458 -0.378 138.34
204. PK profiles of GX-50 after IV and PO administrations to male SD rats Plasma concentration (ng/mL) Individual plasma concentration-time profiles of GX-50 after an IV dose of 1 mg/kg in SD rats (N=3) Rat#1 10000 Rat#2 1000 Rat#3 100 10 1 0 4 8 12 Time (hr) 16 20 24
208. Summary WGX50 can be used for delaying even reversing skin aging WGX50 molecule can be used as potential anti-aging skin care products, solely, or as an active additive to the existing cosmetics, and as medicine for skin anti-aging.  
209. The Future We believe we have found a magic molecule, your help is Needed to have it change the world!
210. Our team!
211. Thank you for your attention 谢谢您的关注! Grants From: National 863, Bioinformatics, 973 Program And National Science Foundation of China(NSFC)
212. References 1.  2.  3.  4.  5.  6.  R.-X. Gu, H. Gu, Z.-Y. Xie, J.-F. Wang, H.R. Arias, D.-Q. Wei* and K. Ch. Chou, “Possible Drug Candidates for Alzheimer’s Disease Deduced from Studying their Binding Interactions with α7 Nicotinic Acetylcholine Receptor”, Medicinal Chemistry, 5, 250-262(2009). Shigao Chen, Ruoxu Gu* and Dong-Qing Wei*, “Virtual Screening with Respect to alpha7 Nicotinic Acetylcholine Receptor for Possible Leads against Alzheimer's disease”, Journal of Molecular Graphics and Modeling, 39, 98-107(2013). H. R. Arias, Ruo-Xu Gu, Dominik Feuerbach, Bao-Bao Guo, Yong Ye, and D.Q. Wei*, “Novel Positive Allosteric Modulators of the Human α7 Nicotinic Acetylcholine Receptor”, Biochemistry, 50, 5263–5278(2011). Ruo-Xu Gu, Limin Angela Liu*, and Dong-Qing Wei*, “ Equilibrium of four binding states of anti-viral drug rimantadine in M2-lipid bilayer system”, J. Am. Chem. Soc., 133 (28), 10817–10825(2011). H. R. Arias, Ruo-Xu Gu, Dominik Feuerbach, Bao-Bao Guo, Yong Ye, and D.Q. Wei*, “Novel Positive Allosteric Modulators of the Human α7 Nicotinic Acetylcholine Receptor”, Biochemistry, 50, 5263–5278(2011). Maoping Tang, Zhaoxia Wang, Ying Zhou, Wangjie Xu, Shengtian Li,Lianyun Wang, Dong-Qing Wei*, Zhongdong Qiao*, “A novel drug candidate for Alzheimer disease treatment - gx-50 derived from Zanthoxylum Bungeanum”, J. Alzheimer’s Disease, 34, 203–213 (2013).