Language: Chinese | English

南方科技大学理学院化学系, 广东 深圳 518055

Department of Chemistry, School of Science, Southern University of Science and Technology, Shenzhen 518055, China

南方科技大学理学院化学系, 广东 深圳 518055

Department of Chemistry, School of Science, Southern University of Science and Technology, Shenzhen 518055, China

南方科技大学理学院化学系, 广东 深圳 518055

Department of Chemistry, School of Science, Southern University of Science and Technology, Shenzhen 518055, China

基于质谱的蛋白质组学技术已经日趋成熟,可以对细胞和组织中的成千上万种蛋白质进行全面的定性和定量分析,逐步实现“深度覆盖”。随着生物医学日益增长的大队列蛋白质组学分析需求,如何在保持较为理想的覆盖深度下实现短时间、快速的“高通量”蛋白质组学分析已成为当前亟需解决的关键问题之一。常规的蛋白质组学分析流程通常包括样品前处理、色谱分离、质谱检测和数据分析。该文从以上4个方面展开介绍近10年以来高通量蛋白质组学分析技术取得的一系列研究进展,主要包括:(1)基于高通量、自动化移液工作站的蛋白质组样品前处理方法;(2)基于微升流速液相色谱与质谱联用的高通量蛋白质组检测方法;(3)利用灵敏度高、扫描速度快的质谱仪实现短色谱梯度分离下蛋白质组深度覆盖的分析方法;(4)基于人工智能、深度神经网络、机器学习等的蛋白质组学大数据分析方法。此外,对高通量蛋白质组学面临的挑战及其发展进行展望。总而言之,预期在不久的将来高通量蛋白质组学技术将会逐步“落地转化”,成为大队列蛋白质组学分析的利器。

Proteomic analysis aims at characterizing proteins on a large scale, including their relative abundance, post-translational modifications, protein-protein interactions and so on. Proteomic profiling helps to elucidate the mechanisms of disease occurrence and to discover new diagnostic markers and therapeutic targets. Mass spectrometry (MS)-based proteomic technologies have advanced to allow comprehensive qualitative and quantitative proteome profiling across a myriad proteins in cells and tissues. High-throughput proteomics is the core technique for large-scale protein characterization. With the increased demand for large cohort proteomic analysis in the biomedical research field, high-throughput proteomic analysis has become a critical issue that needs to be urgently addressed. The standard shotgun proteomic workflow comprises four steps, including sample preparation, peptide separation, MS acquisition, and data analysis. Advances in these four steps have contributed to the development of high-throughput proteomics. In this review, we aimed at summarizing the current information on the state-of-the-art development of high-throughput proteomic analysis, mainly including the following topics: (1) High-throughput, automatic proteomic sample preparation methods based on liquid-handling workstations. The automation of the proteomic sample preparation steps is essential for high-throughput proteomic analysis, which will significantly reduce variation of manual operation and sample loss by multistep sample processing. The commercial liquid handling workstations, including King FisherTM Flex, Agilent Bravo, AssayMAP Bravo, and Biomek® NXP, perform the handling steps of 96- or 384-channel microplate formats using a mechanical arm that increases the throughput and robustness of sample preparation. (2) High-throughput proteomic detection methods based on microliter-flow-rate liquid chromatography coupled with mass spectrometry (micro-flow LC-MS/MS). Nanoliter-flow-rate liquid chromatography coupled with mass spectrometry (Nano-flow LC-MS/MS) is widely used in classic proteomic research due to its excellent sensitivity, which often comes at the expense of robustness. Owing to the improved robustness and decreased injection-to-injection overheads, micro-flow LC-MS/MS has become increasingly popular in high-throughput proteomic analysis. (3) Using MS instrumentation with high sensitivity and fast scanning speed to realize in-depth proteomic analysis coupled with short chromatographic gradient separation. In recent years, new MS instrumentation continues to exhibit speed of analysis and sensitivity enables the large-scale profiling of hundreds of samples. In particular, ion mobility-based MS, such as timsTOF Pro and Exploris 480 equipped with a front-end high field asymmetric waveform ion mobility spectrometry (FAIMS), which provides fast, sensitive, and robust proteome profiling, thus shifting proteomics to the high-throughput era. (4) Artificial intelligence-, deep neural network-, and machine learning-based proteome data analysis methods. These approaches have improved comprehensive proteomic analysis efficiency. Specifically, the emergence of new algorithms and the up gradation of search engines accelerate the process of high-throughput data analysis. Additionally, the challenges and future development of high-throughput proteomics are prospected. In conclusion, high-throughput proteomic technologies are expected to gradually “transform” and become powerful tools for large cohort proteomic analysis in the near future.

蛋白质组学是指大规模地对蛋白质的表达水平、翻译后修饰、蛋白质相互作用等进行研究。蛋白质组研究不仅可以全景式地揭示生命活动的分子本质,还能阐明生命在生理或病理条件下的变化机制[1]。近年来,色谱和质谱技术的进步驱动了蛋白质组学的快速增长,逐步实现了“深度覆盖”,科学家们先后完成包括人类在内的多个物种的蛋白质组图谱解析[2,3]。2020年,Mann等[4]系统性地绘制了100个跨物种的蛋白质组图谱,获得了200万条肽段和34万个蛋白质的鉴定信息,为整个进化范围内生物的功能组织研究提供重要依据。随着生物医学研究日益增长的大队列蛋白质组学分析需求,如何实现“高通量”的蛋白质组学分析已成为当前亟需解决的关键问题之一。常规的蛋白质组学分析流程通常包括样品前处理、色谱分离、质谱检测和数据分析。该文将从以上4个方面介绍近10年以来高通量蛋白质组学分析技术取得的相关研究进展。

样品前处理是整个蛋白质组学分析流程的关键步骤,传统样品前处理流程依赖于多步的手工操作,不仅耗时且不可避免地存在人为误差,成为限制蛋白质组学发展的瓶颈之一。集成化的样品前处理技术有利于提升蛋白质组学分析的整体性能和灵敏度,如大连化物所张丽华课题组[5]结合在线蛋白质酶解,同位素二甲基化标记和多维肽段分离发展了蛋白质组定量集成化平台,具有灵敏度高、定量重现好等优点。本课题组前期开发了一种基于离心力的集成化蛋白质组学样品前处理技术(simple and integrated spintip-based proteomics technology, SISPROT)[6],通过将强阳离子交换填料和C18膜填充在枪头中以集成化方式完成样品前处理和多肽分级的全过程,显著提升了蛋白质组学分析的整体性能和灵敏度,并得到进一步推广[7,8,9,10,11]。高通量蛋白质组学则对样品前处理提出了更高要求,发展分析通量高、快速而稳定的样品前处理方法成为这一领域亟需解决的首要问题。2014年,Yu等[12]在超滤膜辅助的样品前处理方法(filter-aided sample preparation, FASP)基础上发展了96通道的FASP(96FASP),该方法借助于96孔滤膜板进行离心操作,从10 μg临床尿液中可鉴定700~900个蛋白质。Mann课题组[13]则发展了in-StageTip(iST)技术,在内置C18膜的封闭Tip头内可完成样品前处理,并开发了PreOmics试剂盒(https://preomics.com/products)。通过进一步定制96通道的iST设备(96-well iST)并借助于多通道移液器可完成96个样品前处理操作。然而,以上的样本前处理方法虽然提高了分析通量,但仍然依赖于手工移液和转移操作。

近十几年来,自动化移液工作站相继问世,如赛默飞的自动化磁珠提取系统King FisherTM Flex、安捷伦高精度的Bravo和精确流速控制的AssayMAPBravo、贝尔曼库尔特的Biomek® NXP自动化工作站等,使样品前处理逐渐实现自动化。自动化移液工作站借助于机械臂完成96甚至384通道的移液操作,解放手工操作,大大促进了高通量的样品前处理技术发展。Mann等[14]将iST技术与Bravo自动化平台联用,2 h内可完成96个血浆样品的前处理。最近,他们还将这项前处理工作流程应用于100个不同类别生物体的蛋白质图谱解析[4]。Krijgsveld等开发了基于磁珠的样品前处理技术(single-pot solid-phase-enhanced sample preparation, SP3)[15,16],进一步将SP3技术与Bravo平台相结合开发了autoSP3技术,3.5 h可完成96个样品的前处理,并具有优异的稳定性[17]。Leutert等[18]发展了快速,自动化的磷酸化蛋白质组学分析流程(rapid-robotic phosphoproteomics, R2-P2),在King FisherTM Flex自动化工作站上整合了SP3技术和基于Fe-IMAC磁珠的磷酸化富集技术,实现5 h的磷酸化蛋白质组前处理。自动化移液工作站保证高通量分析的同时,还能极大缩短前处理时间,以及减少大队列样品处理造成的人为误差。然而,以上的这些样品前处理工作流程还未能实现“零人工”操作的全自动化。随着高通量样品前处理的不断智能化,开发“零人工”操作的全自动化样品前处理方法成为必然的发展趋势。

色谱分离是液相色谱-质谱联用蛋白质组学分析中的关键环节[19,20,21],但目前的液相色谱-质谱联用技术存在液相分离速度难以匹配质谱采集速度、稳定性不够好等问题[22],限制了高通量蛋白质组学的进一步发展。因此,在保证较为理想的蛋白质组覆盖度前提下,缩短液相色谱分离时间并提高分离稳定性以满足高通量数据采集的需求显得尤为重要。传统的“鸟枪法”蛋白质组学研究一般选择纳升级液相来获得更高的检测灵敏度,与此同时却牺牲了分析通量与液相色谱分离的稳定性。近年来,微升流速液相色谱(micro-flow LC)或高流速液相色谱(high-flow LC)因其色谱稳定性好、周转时间短等优点广泛应用在高通量蛋白质组学。通过配合使用先进的高灵敏度质谱仪,可以有效解决由于稀释效应导致的灵敏度下降问题。本课题组在前期开发的Photo-pTyr-scaffold[23]与基于96孔板的Photo-pTyr-scaffold正向蛋白阵列平台[24]的基础上,开发了高通量、快速的酪氨酸磷酸化(pTyr)信号复合物的蛋白质组学分析工作流程[25]。利用150 mm×150 μm分离柱在分离流速800 nL/min和20 min有效色谱梯度条件下完成高通量蛋白质组学分析;通过与Q Exactive HF-X质谱仪联用,在不增加样品上样量和不影响鉴定覆盖度前提下,能够快速、可重复地分析动态pTyr信号蛋白质复合物[25]。Bruderer等[26]利用150 mm×300 μm分离柱(流速5 μL/min,梯度40 min)完成了1508例血浆样本的分析,并且超过2000次的进样没有发现堵塞(背压只增加14%),表明micro-flow LC具有良好的稳定性。Bian等[27]利用商品化150 mm×1 mm的C18色谱分离柱,以50 μL/min微升级流速联合Q Exactive HF-X质谱仪进行分析。在2000个蛋白质组的分析中具有色谱保留时间(相对标准偏差