细胞是生命活动的基本单位。近来的研究表明即使是相同基因的细胞也会由于微环境的不同和随机过程而呈现出差异，这就是单细胞的异质性。若在群体细胞层面研究，取得的是平均化的结果，掩盖了细胞之间的差异。为了更好地理解生命活动的现象和进程，我们需要在单细胞层面进行研究。相对于单细胞的基因组学、蛋白组学，单细胞的代谢组学可以提供细胞最实时和动态的信息，也有助于发现潜在的生物标志物。然而单细胞的体积很小、基质干扰严重、代谢物在环境中很快就发生变化以及获取到的代谢物信息有限，这都使单细胞代谢组学的发展受到限制。为了解决以上问题，本研究提出了一种新的分析方法，主要内容如下： 1. 建立了多溶剂液滴萃取与电喷雾质谱联用的方法。单液滴微萃取技术萃取到的代谢物信息有限。因此，为了既能去除基质干扰问题，也能获得高覆盖度的单细胞代谢物信息，我们建立了多溶剂液滴微萃取高覆盖检测方法，并将之用于了乳腺癌细胞亚型的分型研究上。在单细胞水平上实现了高覆盖度的代谢物检测，可以区分亚型之间的微小区别，实现了乳腺癌细胞亚型的区分。我们也探究了不同亚型之间有显著性差异的组分，为寻找特定细胞亚型的潜在标志物提供了参考。 2. 基于液滴微萃取与皮升电喷雾质谱联用对不同迁移能力的乳腺癌亚型进行单细胞代谢物分析。基于小室迁移实验可以反映细胞迁移能力，我们选用迁移能力较强的 HER2 型 SKBR-E 为模型，对其迁移前后的单细胞进行萃取，进而对两种情况下的单细胞代谢物进行分析。结合多变量数据分析，发现不同迁移能力的 SK-BR-3 的磷脂分布的确存在差异。此外，我们还发现了单不饱和磷脂与乳腺癌的迁移能力相关，这可能是 HER2 型乳腺癌的潜在生物学标志物，为 HER2 型乳腺癌的早期诊断和后续治疗提供了基础，也为其他乳腺癌亚型的研究提供了参照。

Cell is the basic unit of life. The recent studies revealed that the cells with the same gene still had the different metabolic profiles due to the microenvironment and stochastic process. The phenomenon is called single cell heterogeneity. To meet the sensitivity of traditional methods, population cells were used to get sufficient amounts of compounds, thus the results are average, which masked the difference between cells. Therefore, we need to study at the single-cell level to understand the phenomena and process of life. Compared with the single-cell genomics and proteomics, the single-cell metabolomics can reveal the real-time and dynamic information, which was helpful to find potential biomarkers. However, the small volume of single cell, the severe interference, the quick change of metabolites and the limited coverage of metabolites make it hard for single-cell metabolomics. We proposed a new method to solve the problem. 1. we have developed multiple microextractions coupled to ESI-MS as a method for the discrimination of four breast cancer subtypes at the single-cell level. traditional one-stepextraction method was powerless for the complicated system due to the limited coverage of metabolited. In order to remove the interference and obtain the high coverage single cell metabolite, our method was proposed. Discrimination among four breast cancer subtypes implies that our method is essential to capture the differences among cancer cell subtypes due to its high degree of coverage. We also identified the characteristic molecules that show signifi-cantly different levels of abundance in different cell lines, which may potentially serve as biomarkers of a specific cell type. 2. Single-cell metabolic analysis of breast cancer subtypes with different migration abilities was performed by droplet microextraction- ESI-MS. Based on the transwell assay which can reflects the migration abilities, we selected the more migrate subtypes, HER2 enriched subtypes SK-BR3 as the model. The cells before and after migration were extracted to analyze the single-cell metabolites in both cases at the single cell level. Combined with multivariate data analysis, it was found that the phospholipid distribution of SK-BR-3 with different migration abilities was different. In addition, we have also found that monounsaturated phospholipids are associated with breast cancer migration, which may be a potential biomarker for HER2 breast cancer, providing a basis for early diagnosis, treatments for HER2 breast cancer and a reference for other breast cancer subtypes.