A high-throughput method for simultaneous profiling of mRNA and protein levels in

描述(由申请人提供)：对单个细胞的分析有望揭示在大量细胞群体研究中隐藏的组织生理学和疾病的洞察力。通过分析单个细胞，有可能搜索稀有细胞亚群 例如抗药性癌细胞，这可能在健康和疾病中发挥深远的作用。单细胞分析还可以通过测试细胞信号、新陈代谢、细胞周期和细胞之间的相关性来阐明细胞行为是如何受到控制的 组织内的分化。要实现单细胞分析的承诺，至少需要满足四个技术要求。分析必须在大量的细胞上进行；它应该捕获多个测量以构建“细胞轮廓”或信使核糖核酸水平、蛋白质水平等；它应该足够灵敏，以检测不同细胞之间轮廓的变化；它应该在组织中进行，或与立即从组织中取出的细胞一起进行，以确保测量直接反映临床情况。扩大到大量细胞的问题尤其紧迫。例如，如果耐药癌细胞只占肿瘤的1%，那么即使找到10个这样的细胞，也需要分析总共1000个细胞。然而，今天，在分析许多细胞和生成每个细胞的广泛细胞轮廓之间存在权衡。分析许多细胞的方法，例如荧光激活细胞分选(FACS)，仅限于有限数量的成分，因此可能无法识别感兴趣的稀有细胞。能够提供更全面的轮廓的方法成本高且劳动密集型，因此不能用于大量细胞。理想的方法应该是为每个电池生成一个1美元的手机配置文件。我们提出了一种方法，使用广泛可用的测序技术，每次运行同时分析超过1,000个细胞。该方法结合并适应了许多现有的分子技术，可以同时测量单个细胞中数十种蛋白质和100-200个mRNA的水平。这种测量的广度，特别是蛋白质水平的测量，是任何现有的单细胞水平的系统都不可能实现的。这种方法也应该比当今最接近的可比技术更准确，因为它需要的信号放大要少得多。该方法的长期潜力主要受到高通量测序技术能力的限制，随着测序技术在提供1,000美元人类基因组的竞赛中继续大幅改进，该方法的成本效率、速度和细胞吞吐量也有望提高。我们预计，这种方法最终的成本将大大低于每个细胞配置文件1美元。因此，如果成功，这种方法将为单细胞分析向前迈出深刻的一步。 公共卫生相关性：单细胞分析面临的一个关键限制是目前在分析多个细胞和许多细胞成分之间的权衡。我们提出了一种克服这一限制的方法，该方法以合理的成本(约1美元/细胞)和较少的步骤，在1000个或更多细胞中分析100-200个mRNA转录本和数十个蛋白质的水平。该方法结合了几种已建立的分析方法，并引入了一种新的“组合条形码”策略，使用下一代测序同时读出所有细胞的轮廓。

DESCRIPTION (provided by applicant): The analysis of individual cells promises to reveal insights into tissue physiology and disease that remain hidden in the study of bulk cell populations. By profiling single cells it is possible to search for rare cell sub- populations with distinct characteristics, such as drug-resistant cancer cells, which may have profound roles in health and disease. Single cell analysis can also shed light on how cell behavior is controlled, by testing for correlations between the activity of cell signaling, metabolism, cell cycle and cell differentiation within a tissue. To deliver on the promise of single cell analysis, it is necessaryto satisfy at least four technical requirements. The analysis must be carried out on a large number of cells; it should capture multiple measurements to construct a "cellular profile" or mRNA levels, protein levels, and so on; it should be sufficiently sensitive to detect changes in the profile between different cells; and it should be performed in tissues, or with cells immediately removed from tissues, to ensure that the measurements directly reflect the clinical situation. The problem of scale-up to large numbers of cells is particularly urgent. For example, if drug-resistant cancer cells constitute only 1% of a tumor, then to find even ten such cells requires analyzing 1,000 cells in total. Yet today there is a trade-off between analyzing many cells, and generating a wide cellular profile per cell. Methods that analyze many cells, for example Fluorescent Activated Cell Sorting (FACS), are restricted to a limited number of components and may therefore fail to identify rare cells of interest. Methods that can provide a more comprehensive profile are costly and labor-intensive, and therefore cannot be used on large numbers of cells. An ideal method should generate a cellular profile for <$1 per cell. We propose a method to simultaneously profile over 1,000 cells per run, using widely available sequencing technology. The method combines and adapts a number of existing molecular techniques to measure tens of proteins and 100-200 mRNA levels simultaneously in single cells. This breadth of measurement, particularly of protein levels, is not possible to achieve by any existing system at the single cell level. The method should also be more accurate than the closest comparable technology today, as it requires significantly less signal amplification. The long-term potential of the method is limited primarily by the capabilities of high-throughput sequencing technology, and as sequencing technology continues to improve dramatically in the race to deliver a $1,000 human genome, the cost-efficiency, speed and cell throughput of the method are also expected to improve. We project that this method will eventually cost significantly less than $1 per cell profile. Thus, if successful, this method would provide a profound step forward for single cell analysis. PUBLIC HEALTH RELEVANCE: A critical limitation facing single cell analysis is the current trade-off between analyzing many cells and many cellular components. We propose a method that overcomes this limitation by profiling the levels of 100-200 mRNA transcripts and tens of proteins in 1,000 or more cells at reasonable cost (~$1/cell) and with few steps. The method combines several established assays, and introduces a novel "combinatorial bar-coding" strategy to simultaneously read out the profile of all cells using next-generation sequencing.