Hyperplexed Quantum Dots for Multidimensional Cell Classification in Intact Tissue

摘要 这项生物工程研究基金（BRG）提案的目标是开发单细胞荧光标记， 通过完整三维组织成像进行分类。我们专注于半导体量子 量子点（QD），纳米晶体，表现出明亮的荧光和独特的光学和电子性能。我们 设计的新类量子点，我们建议将允许定量，多光谱分析30或更多 不同的分子，以便高光谱光片显微镜可以用于蛋白质组学分析， 在单个染色步骤之后，在整个完整组织中全面地映射20种或更多种不同的细胞类型。这 该技术解决了三维组织的光学显微镜中的突出瓶颈， 只有3种不同的分子标记可以很容易地区分，限制了精确分类细胞类型的能力。 并共同定位不同的细胞类型。这一建议是在新的光片显微镜有 最近可广泛用于以亚细胞分辨率对光学透明组织进行全厚度成像 这样多路复用的快速发展就能产生快速的影响。作为一个例子，这个项目的研究人员 开发了光学透明、免疫标记和成像来自瘦型和肥胖型的完整脂肪组织的工作流程 啮齿动物模型，除了基于荧光免疫染色剂全面识别细胞的软件外， 深度学习算法来自动化微环境分割。有了这些进步，我们能够 发现脂肪组织中新的免疫微环境类型，这些微环境被认为会促进合并症 肥胖症，如2型糖尿病和心脏病。然而，关于这些性质的关键假设 微环境不能容易地解决，直到我们可以离散分类细胞的基础上，其分子 其上下文微环境中的表达模式。在这个方案中，我们的技术目标是 开发用于完整组织中的高含量多路复用的荧光团。我们的生物学目标是利用这些工具 了解在肥胖状态下调节脂肪组织的免疫细胞微环境。我们的具体 目的是（1）设计新类别的基于QD的标记的物理学，（2）将这些标记缀合到 抗体片段，并验证其作为分子探针的靶特异性，（3）定量评估细胞 在三维脂肪组织中的标记和分类准确性，以及（4）应用探针面板来量化 脂肪免疫微环境在细胞水平上在瘦和肥胖状态。这是一个合作 在量子点和分子探针方面有专长的工程师和科学家之间的提议（安德鲁·史密斯）， 高级光学显微镜（Paul Selvin），生物医学图像计算（Mark Anastasio），细胞免疫学 (Erik纳尔逊）和肥胖动物模型（Kelly Swanson）。

ABSTRACT The goal of this Bioengineering Research Grant (BRG) proposal is to develop fluorescent labels for single-cell classification through imaging of intact three-dimensional tissue. We are focusing on semiconductor quantum dots (QDs), nanocrystals that exhibit bright fluorescence and unique optical and electronic properties. We designed new classes of QDs which we propose will allow quantitative, multispectral analysis of 30 or more distinct molecules so that hyperspectral light sheet microscopy can be used to proteomically profile and comprehensively map 20 or more distinct cell types throughout an intact tissue after a single staining step. This technology addresses an outstanding bottleneck in optical microscopy of three-dimensional tissue, for which only 3 distinct molecular markers can be easily distinguished, limiting the capacity to precisely classify cell types and to co-localize different cell types. This proposal comes at a time when new light sheet microscopes have recently become widely available for full-thickness imaging of optically cleared tissues at sub-cellular resolution such that rapid advances in multiplexing could yield rapid impacts. As an example, the investigators of this project developed workflows to optically clear, immunolabel, and image intact adipose tissues from lean and obese rodent models, in addition to software to comprehensively identify cells based on fluorescent immunostains, and deep learning algorithms to automate microenvironment segmentation. With these advances, we were able to discover new classes of immune microenvironments in adipose tissue that are believed to promote comorbidities of obesity, such as type 2 diabetes and heart disease. However key hypotheses regarding the nature of these microenvironments cannot be readily addressed until we can discretely categorize cells based on their molecular expression patterns within their contextual microenvironments. In this proposal, our technological goal is to develop fluorophores for high-content multiplexing in intact tissues. Our biological goal is to use these tools to understand immune cell microenvironments that regulate adipose tissue in the state of obesity. Our Specific Aims are to (1) engineer the photophysics of new classes of QD-based labels, (2) conjugate these labels to antibody fragments and validate their target specificity as molecular probes, (3) quantitatively evaluate cell labeling and classification accuracy in three-dimensional adipose tissue, and (4) apply probe panels to quantify adipose immune microenvironments at the cellular level in the lean and obese states. This is a collaborative proposal between engineers and scientists with expertise in quantum dots and molecular probes (Andrew Smith), advanced optical microscopy (Paul Selvin), biomedical image computing (Mark Anastasio), cellular immunology (Erik Nelson), and animal models of obesity (Kelly Swanson).