Recently, Sun Jiashu, a researcher at the National Nanoscience Center of the Chinese Academy of Sciences, and Zhang Shaohua, a professor at the Fifth Medical Center of the Chinese people’s Liberation Army General Hospital, have developed a new microfluidic pulse filtration technology and a highly sensitive thermal swimming detection platform to achieve efficient separation and analysis of extracellular tumor vesicles (Extracellular vesicle,EV) from whole blood samples for early diagnosis of breast cancer. The related research results are published in Science Advances under the title of Cascaded microfluidic circuits for pulsatile filtration of extracellular vesicles from whole blood for early cancer diagnosis.
Extracellular vesicles (EV) are lipid bilayer sealed biological particles with a diameter of 30 to 250nm, which are actively secreted into the surrounding environment by most mammalian cells (especially tumor cells) and play an important role in intercellular communication (1-53). Tumor-derived EV (tEV) carries the payload of proteins and nucleic acids that reflect the molecular characteristics of cell origin, and is found in large quantities in blood and other body fluids. Therefore, electric vehicles are considered to be a promising source of biomarkers for liquid biopsies for early cancer diagnosis and real-time monitoring of tumor development (4-9). However, a major obstacle to electric vehicle analysis is the lack of standardized, automated and repeatable methods to separate and purify electric vehicles. In order to isolate EV from whole blood samples, sequential centrifugation, including ultracentrifugation (UC), is now the gold standard, involving multiple centrifugation steps to remove blood cells, cell fragments, and other interfering substances (6, 10, 11). The whole workflow is time-consuming and laborious, and requires the use of a variety of special instruments. In addition, a longer UC cycle may lead to a low yield of EV with protein aggregate coprecipitation (122.13). Several other separation methods, such as polyethylene glycol (PEG)-based precipitation, volume exclusion chromatography, immunoaffinity capture and microfluidic separation, have been used to isolate EV with different results in terms of morphology, purity and yield (14-21). Therefore, there is an urgent need to develop a high-performance, fast, integrated and cost-effective isolation technology between electric vehicles and biological fluids.
Membrane filtration is an alternative strategy for size selective separation of EV by using micro / nano porous membranes as size exclusion filters (size exclusion filters). It is reported that the exocrine total separation chip integrated with several filter modules with different membrane apertures can separate EV (24) with sizes from 30 to 200nm from plasma and urine. A centrifugal microfluidic device containing two nanofilters allows automatic enrichment of EV (25) smaller than 600nm from urine within 30 minutes. For these dead corner filtration equipment, membrane fouling is a major problem, which will lead to the decline of performance and the recovery rate of electric vehicles. In addition, the high pressure and shear force in the dead-end filtration may cause damage to the EV (26pcm27). In order to reduce filter clogging, a double membrane filtration method combined with harmonic oscillation was designed to extract EV (14) from urine samples by periodically re-suspending EV around the membrane surface. However, the application of high-frequency oscillations (3 to 7kHz) for dozens of minutes may change the EV (size and completeness of 28pc29). In addition, the demand for complex external control systems may limit the widespread use of membrane filtration in clinical applications.
Here, we have designed a cascade microfluidic loop for pulse filtering of EV directly from whole blood samples (figure 1A). The platform consists of a cell removal circuit with an external polycarbonate (PC) membrane filter with an aperture of 600nm and an EV isolation circuit with an anodic alumina (AAO) membrane filter with an aperture of 20nm. On the basis of circuit analogy, the designed microfluidic loop can generate pulsating flow through the porous membrane, lifting particles away from the surface, thereby inhibiting filter scaling and particle aggregation (figure 1A). Microfluidic pulsating filtration can separate EV from non-metastatic breast cancer (NMBC) blood samples quickly, with high yield and high purity within 30 minutes. The protein profile of isolated EV was measured by one-step PEG enhanced thermal swimming aptamer sensor (pTAS) in less than 15 minutes to diagnose NMBC.
Tumor EV is a nano-scale phospholipid vesicle (30-250nm) secreted by tumor cells into humoral environment such as peripheral blood, which plays an important role in the occurrence and development of tumor. The development of a simple, rapid, efficient and standardized method for EV separation and analysis has become a key technical issue in the field of tumor fluid biopsy.
In this work, a microfluidic pulse filtration system integrated with microchannel (resistor), elastic film (capacitor) and one-way valve (diode) is fabricated by using flow path-circuit analogy design strategy. The system only needs to be driven by unidirectional pulse pressure to produce a micro-upgraded pulse flow that can reciprocate through the porous membrane, so as to restrain the blockage of the membrane and the accumulation of particles, and effectively improve the filtration effect. By cascading 600nm pore filter membrane (for blood cell removal) and 20nm aperture filter membrane (for free protein, free nucleic acid, lipoprotein and other interfering substances), this method can efficiently and completely separate and purify EV from whole blood samples within 30 minutes, and the recovery rate is as high as 80%. The researchers further used the nucleic acid aptamer sensing platform of PEG enhanced thermoswimming to detect tumor-associated proteins carried by EV in a highly sensitive one-step method within 15 minutes, achieving accurate diagnosis of early breast cancer patients with an accuracy of 91%.
