Acoustofluidic Separation of Placental Nanovesicle Subpopulations in Obstetrical Diseases

胎盘对于胎儿的发育和生长、母体的动态平衡乃至妊娠健康都是必不可少的。然而，由于胎盘位于子宫内的深度和高度的血管成分，我们在怀孕期间非侵入性探测胎盘健康的能力受到了阻碍，使得胎盘在很大程度上无法进行安全和动态的研究。虽然胎盘研究通过细胞培养、体外系统、动物模型和产后分析得到了发展，但这些间接方法提供了关于胎盘健康的事后信息。胎盘成像使胎盘医学领域发生了革命性的变化，但在分子、细胞或代谢水平上的分辨率仍然有限。为了应对这些挑战，我们和其他人专注于从胎盘滋养层细胞释放细胞外小泡(EV)，在人类中，胎盘滋养层细胞直接沐浴在母体血液中。我们专注于外体(现在称为小EV或SEV)、微泡和凋亡小泡，它们从滋养层持续而丰富地释放到母体循环中，并可通过外周血液测试在整个怀孕期间获得。在这些EV中，我们主要关注胎盘SEV，它们含有其他任何细胞类型很少表达的信息，并执行独特的胎盘生物学功能，如抗病毒反应。虽然信息量很大，但最近的数据表明，SEV不是一个统一的囊泡群体，而是包括几个亚群，定义为大SEV、小SEV和外显子。除了它们的大小，这些SEV亚型的特点是独特的货物。尽管最近SEV亚群的发现使研究人员感到兴奋，因为它们有可能给非侵入性诊断领域带来革命性的变化，但SEV亚群仍未用于临床环境。这在很大程度上是由于与分离和隔离纳米尺寸的SEV亚群有关的困难。我们的团队现在已经开发出先进的声流控技术，旨在有效、可重复地和快速地从血液中分离SEV。我们证明，我们可以将胎盘SEV分成它们特定的亚群，这是以前没有完成的。因此，我们建议的研究重点是人类胎盘SEV亚群的产生，以及它们的RNA和蛋白质组货物。我们假设，通过分析SEV亚群中的这些分析物，我们可以阐明与胎盘健康相关的生物活性分子的独特景观。为了降低数据复杂性，我们提出了一种机器学习管道，将用于探测正常和病理怀孕期间的亚SEV频谱。此外，我们将提高我们从脂蛋白中提纯SEV亚群的能力，并产生一种单一的、集成的设备，可以可靠地在人类怀孕期间实时分离囊泡。我们相信，我们的自动化声流方法以高产量、生物相容的方式分离SEV亚群，对于释放SEV的临床应用至关重要。从我们的研究中获得的见解将改进怀孕期间的非侵入性诊断，并可能发现个性化胎盘治疗的新靶点。

The placenta is essential for fetal development and growth, maternal homeostasis, and broadly, pregnancy health. Yet, our ability to non-invasively probe placental health during human pregnancy is hampered by its deep intrauterine location and its highly vascular composition, rendering the placenta largely inaccessibly for safe and dynamic investigation. Whereas placental research has been advanced by cell culture, ex vivo systems, animal models, and postpartum analyses, these indirect approaches provide ex post facto information about placental health. Placental imaging has revolutionized the field of placental medicine, but resolution at the molecular, cellular, or metabolic level remains limited. To address these challenges, we and others have focused on the release of extracellular vesicles (EVs) from placental trophoblasts, which, in humans, are directly bathed in maternal blood. We focused on exosomes (now termed small EVs or sEVs), microvesicles, and apoptotic blebs, which are continuously and abundantly released from trophoblasts into the maternal circulation and are accessible throughout pregnancy by peripheral blood tests. Among these EVs, we focus mainly on placental sEVs, which harbor messages that are seldom expressed by any other cell types and execute unique placental biological functions, such as an antiviral response. While informative, recent data indicate that sEVs are not a uniform population of vesicles, but comprise several subgroups, defined as large sEVs, small sEVs, and exomeres. In addition to their size, these sEV subtypes are characterized by distinctive cargo. Although the recent discovery of sEV subpopulations has excited researchers due to their potential to revolutionize the field of non-invasive diagnostics, sEV subpopulations have yet to be utilized in clinical settings. This is largely due to the difficulties associated with separation and isolation the nano-sized sEV subpopulations. Our group has now developed advanced acoustofluidic technologies designed to effectively, reproducibly, and rapidly isolate sEVs from blood. We show that we can separate placental sEVs into their specific subpopulations, which has not been previously accomplished. Our proposed investigation therefore focuses on the production of human placental sEV subpopulations, along with their RNA and proteome cargo. We posit that, by profiling these analytes from sEV subpopulations, we can illuminate a unique landscape of bioactive molecules that are relevant to placental health. To reduce data complexity, we propose a machine learning pipeline that will be used to probe the sub-sEV spectra during normal and pathological pregnancies. Further, we will improve our ability to purify sEV subpopulations from lipoproteins, and generate a single, integrated device that can reliably separate vesicles in real time across human gestation. We believe that our automated acoustofluidic approach to separating sEV subpopulations in a high-yield, biocompatible manner is critical to unlocking the clinical utility of sEVs. Insights gained from our investigation will improve non-invasive diagnostics during pregnancy and may uncover new targets for personalized placental therapeutics.