Label-free high-throughput live-cell sorting of genome-wide random mutagenesis libraries for metabolic traits by Raman flow cytometry

作者：Xixian Wang, Sen Wang, Zhidian Diao, Xibao Hou, Yanhai Gong, Qing Sun, Jiaping Zhang, Lihui Ren, Yuandong Li, Yuetong Ji, Wei Shen, Yifeng Yin, Shi Huang, Xiaojin Song, Qiu Cui, Yingang Feng, Jian Xu, Bo Ma · 发表于：Proceedings of the National Academy of Sciences · 年份：2025 · DOI：10.1073/pnas.2503641122 · 被引用次数：16 · 研究领域：Microfluidic and Bio-sensing Technologies、Microfluidic and Capillary Electrophoresis Applications、Innovative Microfluidic and Catalytic Techniques Innovation

A full spontaneous single-cell Raman spectrum captures the metabolic phenome in a label-free and noninvasive manner. However, Raman-activated cell sorting (RACS) of rare target cells from highly heterogeneous systems has remained largely conceptual. Here, we present a positive dielectrophoresis-induced deterministic lateral displacement (pDEP-DLD)-based RACS (pDEP-DLD-RACS), in which a modulated pDEP-DLD force is applied to focus, trap, and functionally sort fast-moving single cells in a wide channel. For pigment- and oil-producing yeasts, pDEP-DLD-RACS shows high sorting accuracy (>90%), high throughput (~600 events min −1 ), high yield (>85%), and long stable running time (~10 h), and can sort rare cells while preserving full cellular vitality. Moreover, label-free sorting directly from a genome-wide random mutagenesis library with >10 5 Aurantiochytrium sp. Mutants, based on intracellular docosahexaenoic acid (DHA) content, produces mutant cells with 58% higher DHA productivity in just two RACS runs over two days, representing two-orders-of-magnitude higher time- and cost-efficiency than conventional approaches. This superior trait arises from global remolding of transcriptomes, including enhanced carbon metabolism, reduced intracellular NADPH synthesis rates, and increased triacylglycerol (TAG) synthesis. By enabling direct screening of metabolic traits from genome-wide mutagenesis libraries, pDEP-DLD-RACS is a powerful platform for synthetic biology.