壳层隔绝增强拉曼/定量相位成像用于干细胞分化状态表征研究

Monitoring the differentiation of stem cell into specific cell lineage and the quality of differentiated cells is a difficult problem to be solved urgently in the research of the stem cell differentiation mechanism and its application. And it is needed to achieve the differentiation state characterization such as cell morphology and biochemical index. In this study, we hope to achieve non-intervention differentiation state characterization during mesenchymal stem cell into islet like cell through using shell isolated nanoparticles enhanced Raman spectroscopy (SHINERS) and quantitative phase imaging (QPI). Study on the evolution process and rule of biomolecule during stem cell differentiation by using Raman spectroscopy. Study on the differentiation state characterization method with high sensitivity, high stability and high reliability differentiation by using SHINERS and nuclear targeting SHINERS. It is hoped to achieve the characterization method of phase variation rule of stem cell differentiation through using polarization phase-shifting dual-channel phase measurement method with the advantages of non-contact, non-intervention, full-field, rapid and nano-scale high-precision imaging measurement. By solving the scientific and technological problems of shell isolated nanoparticles preparation, SHINERS measurement, data processing and analysis, spectral identification and imaging, as well as the QPI algorithm with high-precision, high-sensitivity, the fusion and evaluation algorithm of Raman data and phase data, it is possible to provide a new effective method for stem cell differentiation and related research.

干细胞向特定细胞谱系分化及所分化细胞品质监测，是干细胞分化调控和分化机理研究中急需解决的难点问题，需要研究对分化中细胞形态和生化参量等分化状态无干预的动态测量和表征方法。项目以间充质干细胞向胰岛样细胞分化为研究对象，用壳层隔绝增强拉曼光谱和光学相位测量方法对分化状态非干预表征进行研究。通过拉曼光谱，研究分化过程中生物分子的演化过程及规律；通过非靶向和细胞核靶向壳层隔绝增强拉曼光谱探测，研究高灵敏和高稳定的细胞和细胞核生化参量拉曼光谱表征方法。通过偏振相移双同通道相位测量方法对分化过程细胞相位的无接触、非干预、全场、快速和纳米级高精度成像，研究干细胞分化的形态变化特点和规律。研究解决壳层隔绝增强拉曼光谱基底制备、拉曼光谱测量、数据测量处理及分析、光谱指认、高精度高灵敏定量相位成像方法及算法中的科学技术问题，研究拉曼与相位数据融合与评价问题。为干细胞分化状态及相关研究提供新的有效手段。

诱导多能干细胞iPSCs 的获取和医学应用避免伦理道德和免疫排斥问题，为再生医学发展提供了全新的方案，但是其分化过程中形成的畸胎瘤极大限制其发展和应用，急需研究对分化过程中细胞形态和生化参量的无干预动态测量和表征方法。本项目针对多能干细胞iPSCs分化为神经细胞NPC过程中细胞状态的精确识别和表征展开研究。通过拉曼光谱，获得了多能干细胞iPSCs分化为神经细胞NPC过程中生物分子的演化过程及规律。同时，为满足细胞分子拉曼探测安全、普适和可靠的需求，针对不同尺度如纳米、亚微米和微米等待测物，利用微纳光学结构近场增强与远场辐射特性，我们设计并制备了壳层隔绝、间隙、双共振和多面体核壳等多种微纳结构，解决拉曼光谱探测需要低热损伤、宽频谱探测范围和高信噪比的需求。另外，为满足高灵敏和高稳定的细胞分子、细胞核和线粒体等靶向探测，我们设计并制备了对巯基苯硼酸MPBA和混合膜仿生普鲁士蓝等SERS纳米探针，对细胞内活性氧、PI3K/Akt 信号通路和PH值等生化参量进行探测，实现了干细胞分化状态的多生化参量动态表征；利用课题组专有基于液晶相移的数字全息/微分干涉双通道相移相位显微成像系统，研究出一系列提高数字全息积分相位精度和微分相位剪切量测量方法，一系列定量相位显微成像质量如分辨率、干涉对比度、稳定性和系统集成度方法；对多能干细胞iPSC向NPC分化过程中细胞积分相位和微分相位进行无接触、非干预、全场、快速和纳米级高精度成像，获得了一些多能干细胞分化过程中形态变化如细胞体积、表面积、干质量等参量的特点和规律。在此基础上，基于深度学习，提出了基于自动聚焦技术实现拉曼光谱信号增强；基于卷积注意力机制的细胞分化和凋亡状态表征方法；基于残差网络的拉曼光谱分析获得了一些干细胞分化的光谱变化特点和规律；对多能干细胞分化过程中的拉曼光谱、积分相位、微分相位、相衬图像等多模态信号特征进行杂化融合，获得了一些iPSC向NPC分化过程中细胞精确分类表征和快速智能识别的方法。