Quantitative SRS Imaging of Cancer Metabolism at Single Cell Level

项目概要： 虽然改变细胞代谢正在成为癌症的一个既定标志， 在单个活细胞水平上定量代谢物的新工具。特别是，由于缺乏特定的标签， 对于代谢物，对于能够绘制代谢物的高分辨率成像工具存在未满足的需求， 小分子（脂肪酸、碳水化合物、氨基酸）在癌症发病机制中起重要作用。 在IMAT项目的R21资助的支持下，我们的团队通过开发一个 多重受激拉曼散射（SRS）显微镜，使振动成像的代谢物， 活的肿瘤细胞和完整的活组织检查，速度为每个光谱5微秒。此R33应用程序旨在 通过（i）技术简化，将高光谱受激拉曼成像技术推向下一个水平 和验证，（ii）开发一个强大的高光谱图像分割框架，以及（iii）整合 SRS模态与商业自发拉曼显微镜走向广泛使用的非专家。我们有 组建了一个跨学科的团队进行拟议的开发。三个具体目标是：（1）发展 一个易于操作的，高灵敏度的逐线高光谱SRS显微镜，并验证其能力， 单细胞水平的癌症代谢成像。(2)建立一个用于分割的特征分析框架， 高光谱SRS图像使用非参数贝叶斯模型。(3)集成和验证 自发拉曼显微镜上受激拉曼成像模态。通过完成拟议的 开发和验证活动，我们将产生一个非常新颖的光谱成像系统， 广泛适用于分析细胞和组织的化学成分，用于发现驱动的研究和标记， 基于精确诊断。

Project summary: Although altered cell metabolism is becoming an established hallmark of cancer, there remains a crucial need of new tools for quantitation of metabolites at single living cell level. In particular, due to lack of specific labels for metabolites, there is an unmet need for high-resolution imaging tools capable of mapping metabolites and small molecules (fatty acids, carbohydrates, amino acids) that play essential roles in pathogenesis of cancer. Supported by a R21 grant through the IMAT program, our team partially addressed this need via developing a multiplex stimulated Raman scattering (SRS) microscope, which enabled vibrational imaging of metabolites in live tumor cells and intact biopsies at the speed of 5 microseconds per spectrum. This R33 application aims to push the hyperspectral stimulated Raman imaging technology to the next level through (i) technical simplification and validation, (ii) developing a robust hyperspectral image segmentation framework, and (iii) integrating the SRS modality with a commercial spontaneous Raman microscope towards broad use by non-experts. We have assembled an inter-disciplinary team for the proposed development. The three specific aims are: (1) Developing an easy-to-operate, highly sensitive line-by-line hyperspectral SRS microscope and validate its capacity for cancer metabolic imaging at single cell level. (2) Establishing a feature analysis framework for segmentation of hyperspectral SRS images using the non-parametric Bayesian model. (3) Integrating and validating the stimulated Raman imaging modality on a spontaneous Raman microscope. By completing the proposed development and validation activities, we will have generated a highly novel spectroscopic imaging system broadly applicable to analysis of chemical contents of cells and tissues for discovery-driven research and marker- based precision diagnosis.