Single-cell assay on microfluidic devices

Qiushi Huang , Sifeng Mao , Mashooq Khan and Jin-Ming Lin *
Department of Chemistry, Beijing Key Laboratory of Microanalytical Methods and Instrumentation, MOE Key Laboratory of Bioorganic Phosphorus Chemistry & Chemical Biology, Tsinghua University, Beijing 100084, China. E-mail: jmlin@mail.tsinghua.edu.cn

Received 12th June 2018 , Accepted 22nd August 2018

First published on 22nd August 2018


Advances in microfluidic techniques have prompted researchers to study the inherent heterogeneity of single cells in cell populations. This would be helpful in the identification of major diseases and the design of personalized medicine. Different microfluidic approaches provide a variety of functions in the process of single-cell analysis. In this review, we take a broad overview of various microfluidic-based approaches for single-cell isolation, single-cell lysis, and single-cell analysis. Up-to-date flagship techniques and the pros and cons of these methods are discussed in detail.


image file: c8an01079j-p1.tif

Qiushi Huang

Qiushi Huang is a PhD student in the research group of Prof. J.-M. Lin in the Department of Chemistry, Tsinghua University, China. He got his BS from the Department of Chemistry, Tsinghua University, China. His research focus is microfluidic-based single-cell analysis and mass spectrometry.

image file: c8an01079j-p2.tif

Sifeng Mao

Dr Sifeng Mao is currently a postdoctoral research fellow in the group of Prof. J.-M. Lin in the Department of Chemistry, Tsinghua University, China. He got his PhD in 2016 from Tokyo Metropolitan University, Japan. He obtained his BS and MS in 2009 and 2012, respectively, from the Department of Chemistry, Tsinghua University, China. His current research focuses on the design of microfluidic devices for label-free isolation and analysis of single cells and microchemical pen technology.

image file: c8an01079j-p3.tif

Mashooq Khan

Dr Mashooq Khan is currently working as a postdoctoral research fellow in the group of Prof. J.-M. Lin in the Department of Chemistry, Tsinghua University, China. He got his integrated MS-PhD from Kyungpook National University (KNU), South Korea in 2016 and his BS in Chemistry (with distinction) in 2009 from the University of Malakand, Pakistan. He has worked as a visiting research scientist (after completion of his PhD) and research associate (after completion of his BS) in the Department of Chemistry, University of Malakand, Pakistan. His current research interest is microfluidic devices for biodiagnostics and biodefence.

image file: c8an01079j-p4.tif

Jin-Ming Lin

Professor Jin-Ming Lin graduated from Fuzhou University in 1984 and received a PhD at Tokyo Metropolitan University in 1997. He studied and worked at Showa University and Tokyo Metropolitan University during 1992–2002. He was selected as part of the “100 Talented Researcher Program” for the Chinese Academy of Sciences and obtained the Fund for Distinguished Young Scholars of the National Natural Science Foundation of China in 2001. He was a full professor in the Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences during 2002–2004. He has been a full professor in the Department of Chemistry, Tsinghua University since 2004 and was selected as Cheung Kung Scholars Professor of the Ministry of Education, China in 2008. He is a Fellow of the Royal Chemical Society, a vice-president of the Chinese Society of Mass Spectrometry, General Secretary and Deputy Director of the Committee of Mass Spectrometry in the Chinese Chemical Society, and has served as an associate editor or member of the editorial board of ten international journals. He is the author or co-author of 418 original research papers published in international journals, 45 reviews, 4 books and 51 patents. His current research focuses on cell analysis on microfluidic devices with mass spectrometry (Chip-MS) and chemiluminescence.


1 Introduction

Single-cell assays reveal heterogeneities in the morphology, functions, composition, and genetic performance of seemingly identical cells. Therefore, advances in single-cell analysis can overcome the difficulties that arise from the heterogeneity of cells in diagnostics to enable a targeted model of disease. A single cell contains a minute amount of biomolecules such as proteins, DNA, RNA, or other cell surface molecules and hence requires an effective, sensitive, and quantitative technique for detection. Hitherto, the development of modern analytical chemistry has promoted numerous approaches that have the sensitivity and resolution to analyse the chemical substances in a single cell.1 However, these approaches always encounter the difficulty of separating and working on a single cell in the process of sample preparation. Microfluidic devices that are fitted approximately to the size of an individual cell have opened new avenues for point-of-care diagnostics and provided fascinating solutions to many issues in single-cell analysis. Microfluidic platforms have the advantages of portability, parallel processing, automation, and a large surface-to-volume ratio. This enables the integration of multiple liquid handling processes, such as pumping, metering, sampling, dispensing, sequential loading, and washing. These advantages lead to enhanced compatibility in the use of microfluidics and substantially reduce the labor involved with respect to conventional laboratory techniques.2,3 This review describes the various microfluidic-based approaches used for single-cell isolation, single-cell lysis, and single-cell analysis. Up-to-date flagship techniques and the pros and cons of these methods are discussed in detail. This article will be of great interest for researchers working in the same field and an informative tool for researchers from other fields and beginners.

2 Single-cell isolation

Single-cell isolation has been one of the most important steps in single-cell analysis. The first process in single-cell analysis comprises the collection of single cells from their native environment such as a cell suspension or culture dish. Commonly, there are several kinds of method of single-cell isolation, including mechanical, hydrodynamic, dielectrophoretic, optical and droplet techniques. These methods depend on various microfluidic devices for their operation. These devices have several advantages for single-cell isolation. A compartment for single cells can be designed and miniaturized to fit the size and volume of each unit, which is significant for the analysis of biological molecules on the microscale. The volume of reagents consumed by the microfluidic system is also reduced, which will decrease the cost. Most importantly, microfluidic systems are fully automated and closed, which reduces the risk of sample contamination. In this section, we mainly review the methods of single-cell isolation on microfluidic devices listed above.

2.1 Mechanical methods

In a cell suspension, single cells can be separated by the borders of a microfluidic chip and finally fixed in traps. These traps are so-called mechanical traps, which utilize specific structures on a chip to capture single cells and separate them from a cell suspension. Among these structures, microdams and microwells have been widely integrated into microfluidic devices. Both of these depend on the design of a cell chamber to trap single cells from a moving suspension. Besides, valves on a chip provide a convenient system for single-cell isolation. Owing to the characteristics of valves, a fluid can be restricted to a certain region. The flow and flow direction can also be controlled manually. Thus, single cells can be isolated mechanically together with a suspension.
Microwells. The mechanism of a microwell on a chip is directly derived from the concept of a multiwell plate. The difference between microwells and conventional multiwell plates lies in their size and volume. Microwells are generated on microchips using microfabrication techniques such as soft lithography, as shown in Fig. 1A. After a cell suspension passes above the microwells, single cells will be seeded into the microwells as targets. This method enables high-throughput single-cell analysis that is able to capture hundreds to millions of single cells at the same time.4,5 Chin et al.6 demonstrated the fabrication of approximately 10[thin space (1/6-em)]000 microwells on a glass coverslip and their application for long-term and quantitative analysis of the proliferation of single stem cells. In order to improve the efficiency of single-cell isolation, surface modification has been utilized for cell trapping. Revzin et al.7 reported a polyethylene glycol (PEG) microwell array modified with a T-cell-specific anti-CD5 antibody for immobilizing single T-lymphocytes, which achieved a capture rate of 95%. Similarly, Chen et al.8 developed a microfluidic microwell array coated with a cell-specific aptamer that was capable of isolating and capturing single cells from an artificial complex of a biological sample.
image file: c8an01079j-f1.tif
Fig. 1 Mechanical methods of single-cell isolation. (A) Microwells. Reprinted with permission from ref. 7. Copyright 2005 The Royal Society of Chemistry. (B) Microdams. Reprinted with permission from ref. 11. Copyright 2006 American Chemical Society. (C) Valve on chip. Reprinted with permission from ref. 17. Copyright 2000 The American Association for the Advancement of Science. (D) Integrated valves on chip. Reprinted with permission from ref. 17. Copyright 2000 The American Association for the Advancement of Science.

Recently, microwells have undergone much development and found many applications. Han et al. developed a microwell-based high-throughput single-cell RNA sequencing system to analyze more than 400[thin space (1/6-em)]000 single cells covering all the major organs in mice and constructed a basic scheme for an atlas of mouse cells. This study demonstrated the widespread applicability of microwell-based sequencing technology.9 Wu et al.10 reported the design and fabrication of a planar chip for high-throughput trapping (more than 2400 single cells at a time) and pairing of cells in an area of 1 × 1.5 cm by positive dielectrophoresis (p-DEP) within only a few minutes. The combination of the microwell structure and dielectrophoresis constitutes an effective means of high-throughput single cell–cell pairing that could provide a facile platform for cell communication and enable a precise cell pairing step in cell fusion. Although microwells enable a high-throughput method of single-cell isolation, some drawbacks still remain, of which one is that, owing to the risk of contamination from an open environment, microwell approaches are more suitable for microscopic single-cell studies instead of molecular analysis. Another drawback is that the degree of operability of this method is limited in terms of the analysis of single cells.

Microdams. Microdams comprise a second approach for the mechanical isolation of single cells that is similar to the use of microwells. These devices both utilize physical boundaries to trap single cells from a suspension, whereas the difference lies in the manufacture of the hole or embossment. The physical boundaries can be U-shaped cups that may have cutaways that allow fluids to flow through an unoccupied trap, which thereby increases the frequency of trapping. The classical U-shaped microdam (Fig. 1B) was developed by Di Carlo et al.11 In his system, quantities of U-shaped structures were demonstrated to capture about 100 individual HeLa cells to observe their growth and adherence within 24 h.12 Furthermore, U-shaped traps can be modified with drainage channels, which could help reduce the pressure on cells to maintain cell integrity.13 In order to increase the occupancy of each microdam and the loading speed, an optimized structure was utilized to sort cells into individual channels, which was able to quickly trap single cells using a gravity-driven flow and achieve high trapping efficiency.14,15 Moreover, U-shaped microdams can be employed for the high-throughput trapping of two cells in physical contact.16 The drawbacks of microdams are the same as those of microwells, namely, that an open environment is not suitable for molecular analysis and they have limited operability.
Valves. The development of valves on a chip can overcome the drawbacks of microwells and microdams to a certain extent. Mechanical actuators provided by valve-based microfluidic devices exhibit high programmability and a relatively high throughput. Valve-based microfluidic devices were first developed by Quake et al.17 in 2000 (Fig. 1C and D). With the inspiration of the concept of transistors, two intersecting channels in multiple layers were fabricated as a unit. In each unit, the upper channel acts as a switch to control the other channel. One chip may contain thousands of units that form hundreds of individually controlled chambers. A valves-on-a-chip system can be used in single-cell isolation steps for single-cell analysis.18 Using a valves-on-a-chip system, Streets et al.19 described the heterogeneity of gene expression in single mouse embryonic cells and mouse embryonic fibroblasts by RNA sequencing. In this study, pressure-actuated valves were used to capture single cells in a chamber with a volume of only 0.86 nL. The capture rate in this study was made to approach 100% by ensuring that no cells were lost between injection and trapping, which would be particularly valuable for applications that require transcriptome analysis of rare cells. Fan et al.20 utilized valves to capture single cells and partition lysates of a chromosome suspension into 48 different chambers for amplification, which resulted in whole-genome haplotype analysis. The mechanism of the valves was also based on pressure actuation. Similarly, White et al.21 presented an integrated microfluidic device for the high-throughput digital polymerase chain reaction (dPCR) analysis of single cells. The device was used to analyse 200 single cells on a 10 cm2 chip and was operated by individually controlling 12 pneumatic valves. Although valve-based systems can overcome the drawbacks of microwells and microdams, such as limited operability and molecular analysis, the complex multilayer structure and typical employment of computer-controlled pneumatics would complicate the fabrication and operation of devices and increase the cost to a certain extent. Moreover, the relatively limited throughput is an issue that needs more work to improve.

2.2 Hydrodynamic methods

A hydrodynamic microfluidic system utilizes a specific fluid flow to capture single cells or particles, which is different from the mechanical methods that use borders of a microfluidic chip. This approach does not require complicated experimental systems and is commonly used for single-cell isolation. There are several common forms of hydrodynamic microfluidic systems. Induced by a recirculating fluid flow, vortices22 were demonstrated to trap single cells, and this effect was termed “hydrodynamic tweezers”. Lutz et al.23 reported an audible frequency oscillation around a cylinder used to generate four recirculating vortices, which were able to capture single cells at the centers of the vortices. Using this method, cell lysates are usually not isolated. As a result, hydrodynamic tweezers are commonly used for microscopic analysis.

Dean flow, which is induced by inertial effects, results in the formation of a vortex that is perpendicular to the original flow direction. Owing to this phenomenon, cells with different sizes, densities, or shapes in a cell suspension respond differently to inertial effects and are concentrated at different locations in a channel, which will be helpful for single-cell encapsulation. Using Dean flow, Warkiani et al.24 reported the label-free capture of CTCs from blood with high throughput (1.7 mL min−1) and conducted fluorescence in situ hybridization detection for DNA analysis. Kemna et al.25 introduced a high-yield/speed method of single-cell droplet encapsulation using a Dean-coupled inertial ordering system in a curved continuous PDMS-based microchannel. The yield of this approach can reach 77% at a speed of 2700 cells per s. Similarly, Huang et al.26 used a capillary to generate Dean flow to isolate single cells for mass spectrometry. Inertia-based microfluidics enables the passive and label-free isolation of single cells. This will greatly simplify the operation and fabrication of devices and thus increase their applicability. Furthermore, the fluid flow in the channel is continuous, which enables high-throughput experiments. A disadvantage is that most devices require diluted samples for better performance. Other forms of hydrodynamic methods, such as cross-flow27 and hydrophoretic28 methods and pinched flow fractionation,29 have also been investigated. Recently, Mao et al.30 reported a live single-cell extractor (LSCE) based on laminar flow for capturing adhered single cells on a cell culture dish to study the cell heterogeneity (Fig. 2). Based on soft lithography techniques, the tip of the LSCE was fabricated from PDMS and was immersed perpendicularly into a Petri dish. The extraction of adhered single cells was performed in automated mode. At a sufficiently high ratio of aspiration to injection, a stable microjet of laminar flow formed underneath the tip of the LSCE, which enabled the extraction of adhered cells. Hydrodynamic single-cell isolation is one of the most cost-efficient ways to achieve high-throughput cell capture in microfluidic chips. However, there are several limitations, such as the reliability of device fabrication and complex optimization of the fluidic device. Moreover, the user interface needs to be addressed for advanced applications.


image file: c8an01079j-f2.tif
Fig. 2 Hydrodynamic method of single-cell isolation. Reprinted with permission from ref. 30. Copyright 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim.

2.3 Dielectrophoresis

Cell capture by electrical methods can be an attractive option because the physical pressure on the membrane is low in comparison with the above methods. Among many such methods, dielectrophoresis (DEP) has been a popular method of cell capture owing to its strong driving force and convenient integration with microfluidic systems. Owing to the advantages of DEP, such as label-free manipulation and low-cost fabrication, it has become a common method for the accurate and efficient manipulation of single cells. The usual DEP system includes a disposable cartridge and an analysis platform. The cartridge is an array composed of a series of individual electrodes that enable single-cell isolation31 (Fig. 3A). After being trapped, single cells are identified by a microscope and sorted into other traps or isolated off chip. By utilizing DEP technology, different types of cells can be captured selectively and controlled in both space and time.32 Zhang et al.33 demonstrated the prototype of a fast Raman-activated cell sorting (RACS) system, which combined a positive dielectrophoresis (pDEP) unit with a solenoid valve suction-based switch for trapping and sorting single cells. The system enabled label-free cell sorting with an average accuracy of 73%, which demonstrated rapid RACS at the sub-second level. In contrast to pDEP, several novel designs that use negative dielectrophoresis (nDEP) to trap cells in physiological media have been presented.34,35 Jaeger et al.36 employed nDEP to trap a single yeast cell without contact in an electric field cage for a study of single-cell proliferation. Similarly, Thomas et al.37 developed concentric-ring nDEP traps to hold the selected cell in place while repelling other cells in a flow system. This study illustrates the potential of dielectrophoretic devices for the capture of cells from heterogeneous populations. Overall, DEP-based methods of single-cell isolation have relatively low throughput, which limits their applicability to smaller samples, and higher labor costs. Furthermore, when using DEP approaches, it is important to distinguish signals due to single-cell heterogeneity and experimental noise in a series of single-cell analyses.
image file: c8an01079j-f3.tif
Fig. 3 Other methods of single-cell isolation. (A) Dielectrophoresis traps. Reprinted with permission from ref. 31. Copyright 2013 Elsevier Ireland Ltd. (B) Droplet-based single-cell isolation. Reprinted with permission from ref. 52. Copyright 2008 The Royal Society of Chemistry. (C) Optical single-cell isolation. Reprinted with permission from ref. 42. Copyright 2010 Optical Society of America. (D) Droplet-based single-cell transcriptome barcoding. Reprinted with permission from ref. 57. Copyright 2016 Macmillan Publishers Limited, part of Springer Nature.

2.4 Optical methods

When optical instruments are integrated into microfluidic platforms, optofluidic systems are created, which are promising for the manipulation and measurement of single cells. Optical tweezers are important tools in single-cell analysis and have been widely employed for manipulating cells and the measurement of forces.38,39 This approach relies on a tightly focused laser beam to manipulate single cells with little damage to the cells themselves. In an optical process, single cells are easily trapped at the focal point of laser beams, which enables the isolation of single cells with great convenience. The mechanism of this technique has been reported previously.40,41 When the refractive index of a cell is higher than that of the surrounding fluid, the cell will tend to migrate to the region with the highest light intensity, and vice versa. In both cases, the cell will move towards the direction of light propagation. Under these conditions, single-cell isolation can be achieved in several cases, as shown in Fig. 3C.42 Bragheri et al.43 presented a device with a fluorescence-triggered optical mechanism for analysis and sorting at the single-cell level, which enabled the isolation of subpopulations with high selectivity from heterogeneous samples. However, the device could only sort approximately 100 cells per minute owing to a lag in software communications. Huang et al. reported a novel microfluidically integrated optoelectronic tweezers platform for the preparation and analysis of single-cell samples. This system can optically select a specific cell from a population and move it to a separate channel. Once all the channels are loaded with single cells, the cells can be flushed off the chip for further analysis.44 Chen et al.45 utilized a tightly focused laser to induce a cavitation bubble in a fluid flow, which provided a mechanical force for cell manipulation. Two disadvantages of this approach were dilution of the sample due to the sheath flow and the complexity of building 3D microfluidic channels. In the following year, they improved their initial design by using an inertia-based focusing step,46 which therefore removed the need for a 3D structure. The disadvantage of optical tweezers is obvious in that the low throughput capability remains a considerable limitation.

2.5 Droplet methods

Droplet-based microfluidic systems provide an efficient means of single-cell isolation. They enable high-throughput encapsulation of single cells and are convenient tools for controlling the environment of individual cells, which is useful for the molecular analysis of single-cell lysates. Monodisperse droplets are generated by the confluence of two immiscible fluids, and no complex fabrication of the microfluidic system is required. High throughput is achieved by working in a continuous flow with no pauses. This field is now quite popular, and a number of extraordinary works have been published.47–51 Edd et al.52 demonstrated a droplet-based microfluidic system (Fig. 3B), which provides a purely passive method of controllably loading single cells into drops, which overcomes the intrinsic limitations imposed by Poisson statistics and ensures that virtually every drop contains exactly one cell. Kemna et al.25 utilized Dean flow in a spiral microchannel for prefocusing cells and reported a single-cell encapsulation efficiency of 77%. Recently, Chen et al.53 described a method of single-cell lipid profiling that combined drop-on-demand inkjet cell printing and probe electrospray ionization mass spectrometry. Using this technique, high-throughput single-cell analysis was achieved, as well as cell classification and the detection of cell markers. Guo et al.54 combined contact printing and droplet-based microfluidic techniques and established a solid pin-based droplet system for performing automated and flexible miniaturized liquid–liquid reactions and assays in the range from femtoliters to nanoliters. This system has been utilized in the measurement of kinetic parameters for matrix metalloproteinase-9 in picoliter droplets and assays of enzymatic activity in single cells. Lake et al.55 reported high-throughput methods for single-nucleus droplet-based sequencing and hypersensitive site sequencing of single-cell transposomes. This method was used to acquire maps of the nuclear transcriptome and DNA accessibility for >60[thin space (1/6-em)]000 single cells from the adult human visual cortex, frontal cortex, and cerebellum.

Droplet methods have inherent advantages over single-cell RNA sequencing (Fig. 3D), and several related works have been published recently.56,57 Stephenson et al.58 developed a 3D-printed low-cost instrument for controlling droplet microfluidics and employed it in a clinical environment to perform single-cell transcriptome profiling of disaggregated synovial tissue from five patients with rheumatoid arthritis. Using this technique, 20[thin space (1/6-em)]387 single cells were sequenced, which revealed 13 transcriptomically distinct clusters. These constitute an unsupervised draft atlas of autoimmune infiltrates that contribute to disease biology. Because the quantities and quality involved in single-cell droplet microfluidics have increased significantly over the last 5 years, we can expect considerable growth in this research field. Future work might focus on the preservation/storage of single-cell-containing droplets, standardized frameworks for barcoding and packaging or increasing the sensitivity of barcoding.

Although many microfluidic platforms for single-cell isolation have been reviewed in this section, each of these platforms has advantages and limitations in operation. Researchers could follow suitable approaches to improve performance, taking into consideration throughput, vulnerability, and the risk of contamination, etc.

3 Single-cell lysis

Single-cell lysis is a significant step in the molecular analysis of single cells. It plays an important role in subsequent studies such as genome analysis, transcriptome analysis, and proteome analysis. Microfluidic devices provide ideal platforms for this process59 owing to their characteristic controllable dimensions, laminar liquid flow, transparency for visualization, tunable mechanical or chemical perturbation, and reduced extraneous contamination. In comparison with conventional methods, microfluidic-based lysis minimizes dilution of the lysate, which mostly increases the sensitivity of assays.60 Here, several microfluidic-based methods of single-cell lysis, namely, mechanical, thermal, chemical, and electrical methods, are discussed (Fig. 4). Each method has its own merits and limitations.
image file: c8an01079j-f4.tif
Fig. 4 Methods of single-cell lysis. (A) Mechanical. Reprinted with permission from ref. 61. Copyright 2003 The Royal Society of Chemistry. (B) Thermal. Reprinted with permission from ref. 64. Copyright 2010 The Royal Society of Chemistry. (C) Chemical. Reprinted with permission from A. Sarkar et al., Nat. Commun., 2014, 5, 3421. Copyright 2014 Macmillan Publishers Limited. (D) Electrical. Reprinted with permission from ref. 82.

3.1 Mechanical methods

Mechanical cell lysis crushes cell membranes by a mechanical force, which can comprise shearing, compression, collision with a sharp structure, or other methods. Di Carlo et al.61 demonstrated mechanical lysis using nanostructured filters, which can result in the accessibility of biomolecules for bioassays (Fig. 4A). HL-60 cells and samples of whole blood were tested in an experiment. When the cells passed through arrays of sharp blades, the friction forces and shearing stresses easily penetrated the cell membranes, which resulted in lysis. Hoefemann et al.62 achieved single-cell lysis using a vapor bubble generation technique. Cells passed through an integrated heater and were lysed by bubbles, which compressed the cells against the border of a channel. Single-cell lysis was demonstrated with an efficiency of 100% within 20 ms. However, the throughput was relatively low because there was a single bubble generator in the device, and multiple generators may be considered as an improvement in future work. Recently, Cheng et al.63 presented a pump-on-a-chip microfluidic platform for mechanical cell lysis. Using this platform, they achieved cell lysis rates of 80.6% and 90.5% for samples of HEK293 and human natural killer cells, respectively, after 30 cycles of circulating lysis steps. It is believed that an increase in the number of repeated cycles of circulating lysis could further improve the cell lysis rate. The advantages of microfluidic-based mechanical cell lysis in comparison with other methods are the more effective procedure, more facile and simpler operation, absence of a requirement for chemical agents and negligible influence on intracellular compounds because no new reagents are introduced.

3.2 Thermal methods

Thermal cell lysis relies on thermally induced denaturation of cell membrane proteins, which thereby disrupts the cell membrane. This method is most frequently used for PCR analysis that requires additional temperature cycling rather than protein analysis. Gong et al.64 developed a microwell-based approach for detecting copies of mRNA transcripts obtained directly from cells by a one-step single-cell reverse transcription polymerase chain reaction (RT-PCR) (Fig. 4B). In experiments, they loaded more than 6000 single cells into on-chip wells and performed single-cell lysis by heating at 50 °C for 40 min, which was followed by temperature cycling to amplify the transcribed cDNA products. The heating method does not incur the risk of contamination with chemical detergents, in contrast to chemical lysis.65 However, the applicability of this method is limited because many intracellular molecules, such as proteins, are very sensitive to temperature. As for single-cell molecular analysis, prolonged lysis times could represent a considerable limitation in the measurement of signals that occur within a short time in cells. This could explain why thermal lysis is commonly only used in studies of gene expression, which last for several hours.66

3.3 Chemical methods

Owing to the long history of chemical cell lysis in the bulk analysis of cells, it has become a popular technique in the field of single-cell lysis. The introduction of a lysis buffer enables lysis with high efficiency and throughput, which has been a common tool for the analysis of single-cell genomes, transcriptomes and proteomes. Sarkar et al.67 presented a microfluidic probe that was able to lyse single adherent cells from a standard tissue culture. The probe was able to hydrodynamically confine the lysis buffer from the tip to an area on the scale of a single cell via a balanced surrounding inflow (Fig. 4C). The buffer was a complex mixture containing 1% commercial Triton X-100 detergent, which is essential for single-cell lysis. Thus far, many commercial systems have been developed for cell lysis. Treutlein et al.,68 by utilizing a commercially available Fluidigm C1 system, captured single cells for RNA sequencing and qPCR. The system used a lysis buffer to disrupt the cells. Other commercial detergents also have wide applicability, such as Thermo Scientific's NP-40 nonionic detergent. Streets et al.19 fabricated their own microfluidic chip for single-cell RNA sequencing. The cell lysis process was completed by the addition of a buffer containing 10% NP-40. The selection of a suitable protocol for the lysis of mammalian69 and bacterial cells70 and tissue71 can be made by consulting manufacturers’ guides for bulk cell applications or in accordance with the above references.

The velocity of the lysis process depends on several factors, such as the detergent or enzyme being utilized, the concentration, and the efficiency of contact. An increase in the efficiency of contact can greatly enhance single-cell lysis. Shi et al.72 brought cells into strict contact with lysis buffer via diffusion for 20 min and incubation for a further 20 min. However, prolonged contact limits the analysis of faster intracellular events, as has already been discussed. DeKosky et al.73 captured single lymphocytes together with lithium dodecyl sulfate in microdroplets and reported a lysis efficiency of 100%. Similar works have been published on single-cell lysis in droplets.74–77 Chemical single-cell lysis is a well-established approach with the characteristics of easy implementation and limited requirements of the lysis process. However, the lysis buffer may denature the intracellular proteins, and the additional separation steps used for removing the lysis reagent increase the complexity of system integration.

3.4 Electrical methods

Electrical cell lysis relies on molecular reorientation induced by an electric field to disrupt cells. The cell plasma membrane is composed of a lipid bilayer, which has fluid qualities. Upon exposure to an electric field, the lipid bilayer will undergo molecular reorientation, which forms pores in the membrane. Once the electric field is strong enough or the cell is treated for a sufficient time in the field, pore formation will be irreversible, which causes cell lysis.78 Electrical cell lysis has an advantage in that the electric field can be optimized for rapid cell lysis with little influence on target biomolecules and even some organelles with membranes.79 Mellors et al.80 combined capillary electrophoresis (CE) and electrospray ionization mass spectrometry (ESI-MS) on a microfluidic chip for the automated real-time analysis of single cells. In experiments, single cells traveled to a T-intersection where a strong electric field (4 kV) lysed the cells at a rate of 0.2 cells per s. Young et al.81 reported the on-chip electrical lysis of single lymphoma cells using a voltage of 15 V and a frequency of 1 MHz. The success rate of cell driving in this experimental architecture was 80%, with an average driving speed of 17.74 μm s−1. However, this method processes a single cell at one time. Recently, de Lange et al.82 developed a cell lysis method that comprised applying an electric field and using lysozyme for lysis in droplets (Fig. 4D). This approach is simple to integrate into microfluidic devices and compatible with high-throughput single-cell screening assays. The lysis efficiency was >90%, which is a great improvement on that of lysis using lysozyme alone. Electrical approaches provide rapid cell lysis in comparison with thermal approaches and are also able to avoid extraneous contamination in comparison with chemical lysis methods. However, the integration of on-chip electrodes will make the fabrication of the chip difficult and also complicate the operation, which will limit the applicability of this approach.

Besides the approaches listed above, other methods are employed in single-cell lysis, such as optical83 and acoustic methods.84 Readers with more interest can read the related references.

4 Single-cell analysis

Single-cell analysis on microfluidic devices has found a significant number of applications in biological and chemical research owing to its remarkable capacity for the manipulation of single cells. After a single cell is brought into an ideal condition for analysis, it is convenient to study both on a phenotypic level (gene/protein expression) and a genome sequencing level (DNA analysis). Most of the information will finally be obtained by molecular analysis, and we divide the relevant methods into three categories, namely, optical, electrochemical, and mass spectrometric analysis.

4.1 Optical analysis

Microscopic techniques have been widely used in single-cell analysis for a long time. The ability to directly visualize the morphology of cells makes this a popular approach. The development of fluorescence methods reveals the spatial distribution of molecules, cellular constituents, or dynamic states of biological processes. In recent years, the fast development of techniques based on fluorescence spectroscopy has enabled the rapid tracking of biomolecules, and the creation of super-resolution microscopy has permitted imaging with resolutions down to several nanometers.85 In this section, we briefly introduce recent advances in fluorescence techniques, including labeling and label-free approaches, and finally introduce some live-cell imaging techniques.
Fluorescence techniques. Advanced fluorescence techniques are now able to analyze target molecules and image single cells with sub-micrometer resolution. Alternative approaches such as fluorescence resonance energy transfer (FRET), fluorescence in situ hybridization (FISH) and fluorescence super-resolution microscopy (SRM) are introduced as follows.

The FRET technique is extensively utilized for biochemical reactions and studies of cell biology that involve energy transfer from an excited donor fluorophore to an acceptor molecule.86 The advantage is that fluorescence is emitted upon binding to the target with no necessity for more washing steps. Son et al.87 developed a reconfigurable microfluidic platform integrated with a micropatterned photodegradable hydrogel array, which enabled single-cell secretion analysis. FRET peptides were trapped inside microfabricated compartments to monitor the activity of protease molecules secreted by single cells. Ng et al.88 developed multicolored FRET-based enzymatic substrates integrated into a microfluidic platform to simultaneously measure the activities of multiple specific proteases from single cells in droplets, as shown in Fig. 5A.


image file: c8an01079j-f5.tif
Fig. 5 Optical methods of single-cell analysis. (A) FRET-based method. Reprinted with permission from ref. 88. Copyright 2016 Elsevier B.V. (B) FISH-based microfluidic system. Reprinted with permission from ref. 89. Copyright 2015 The Royal Society of Chemistry. (C) Super-resolution imaging of fixed single cells. Reprinted with permission from ref. 91. Copyright 2014 Macmillan Publishers Limited. (D) Laser tweezers Raman spectroscopy set-up. Reprinted with permission from ref. 98. Copyright 2016 The Royal Society of Chemistry. (E) iSCAT-based methods. Reprinted with permission from ref. 104. Copyright 2015 American Chemical Society.

The FISH technique is used to detect and localize specific sequences in DNA or RNA molecules in single cells. Although the throughput is sacrificed, it is particularly useful for identifying heterogeneities in gene expression within single cells or complex tissues. Perez-Toralla et al.89 developed a protocol that enabled the immobilization and fixation of cells in the liquid phase and their quantitative characterization by fluorescence in situ hybridization (Fig. 5B). Moffitt et al.90 introduced a new approach for multiplexed error-robust fluorescence in situ hybridization (MERFISH). This method has higher throughput than traditional methods.

SRM has recently enabled optical imaging with resolutions of only 10 nanometers. This can be useful in single-cell studies for localizing target molecules in the cell. Jungmann et al.91 reported a novel method that suffers from high exposure of the sample to light but has multiplexing capabilities (Fig. 5C). They employed a sequential labelling/image acquisition protocol (Exchange-PAINT) and were able to achieve resolutions of below 10 nm. Niehörster et al.92 utilized spectrally resolved fluorescence lifetime imaging microscopy (sFLIM) to visualize up to nine different target molecules simultaneously in C2C12 mouse cells. The results demonstrate that sFLIM is suitable for super-resolution multi-target imaging by stimulated emission depletion (STED). For other related studies, readers may be interested in references that will not be introduced.93–95

Label-free optical methods. There are several optical microscopy techniques for single-cell analysis without the need for fluorescent labels or staining. Here, we introduce three widely used methods, namely, Raman spectroscopy, surface plasmon resonance, and interferometric scattering microscopy.

Raman spectroscopy is a spectroscopic technique that is widely used to observe vibrational, rotational, and other low-frequency modes in a system and has recently been used in single-cell analysis.96,97 Casabella et al.98 reported an automated microfluidic platform for single-cell Raman spectroscopy (Fig. 5D). Results were presented for the discrimination of live prostate epithelial cells and lymphocytes, together with a consideration of the consequences of traditional ‘batch analysis’, which is typically used for LTRS of live cells. Kang et al.99 utilized Raman spectroscopy in combination with fluorescence microscopy to probe the dynamics of drug delivery in single cells. Besides, Raman spectroscopy can be employed for single-cell sorting. Readers interested in this field could focus on the review below.100

Surface plasmon resonance (SPR) imaging is widely used for determining the binding/dissociation constant of two molecules, which can be applied in single-cell analysis. Stojanović et al.101 utilized SPR to detect and quantify secreted antibodies from single hybridoma cells. Epithelial cell adhesion molecule (EpCAM) was pre-immobilized on the SPR sensor to induce the secretion of monoclonal antibodies by hybridoma cells. The result showed that the production of antibodies by cells could be measured quantitatively at a single-cell level.

Interferometric scattering microscopy (iSCAT) is also an optical microscopy technique with no requirement for labels. iSCAT is based on scattered light and interference with a reference light field and is mainly utilized for visualizing cell membranes or detecting the movement of biomolecules.102,103 Agnarsson et al.104 invented a light scattering microscopy technique for measuring the kinetics of binding of single cells to a given surface on the basis of evanescent fields (Fig. 5E). A theoretical analysis of the results showed that light scattering signals from lipid vesicles on a single surface can be characterized in terms of the vesicle size without the use of fluorescent labels.

Live-cell imaging. Live-cell imaging commonly focuses on the prolonged analysis of cells with regard to their proliferation and metabolic processes. This method can be carried out either with or without a fluorescent label.105,106 Its remarkable compatibility with super-resolution microscopy is also highly convenient.107 Moffitt et al.108 achieved live-cell monitoring of single bacteria and bacterial communities via time-lapse studies over 30 to 40 generations. Hoffman et al.109 utilized live-cell imaging of bacteria transfected with GFP to distinguish reversible and irreversible adhesion. Under common conditions, normal cells do not express fluorescent proteins or molecules. Hence, the introduction of additional dyes into cells may influence the vitality of the cells, which will limit the study of quantitative dynamic long-term imaging of cell vitality. To address this problem, Krämer et al.110 established a system for noninvasive perfusion with propidium iodide (PI) and counterstain for the analysis of single-cell vitality. In their study, a nontoxic method for the continuous monitoring of cell vitality was established by changing the conventional concentrations of the PI stain and testing the effect of PI on cell vitality.

4.2 Electrochemical analysis

In recent years, advances in electrochemical techniques have enabled progress in applications in the field of single-cell analysis.111,112 Electrochemistry is an ideal technique for miniaturization and integration into microfluidic platforms. Owing to the absence of a requirement for optical transparency, microelectrodes can be easily fabricated on a series of substrates such as polymers, silicon and glass. Because of their sensitivity to a wide range of biomolecules, electrochemical techniques are especially useful for the study of neuronal cells by measuring the release of neurotransmitters. Liu et al.113 reported an approach based on Au nanoelectrodes with high spatial resolution for the detection of the release of dopamine from rat pheochromocytoma cells (Fig. 6A). By cyclic voltammetric scanning, Au nanoelectrodes with a radius from 970 nm down to 3 nm can be controllably fabricated. Carbon fiber microelectrodes also have excellent electrical properties for the detection of the release of transmitters from single vesicles in single cells. Robinson et al.114 utilized nanowire electrode arrays to track the signals of multiple interconnected neurons. In their experiment, the vertical nanowire electrode arrays that were utilized could stimulate and record neuronal activity intracellularly in rat cortical neurons, which can also be used to detect multiple single-synapse connections. The scalability of this platform with silicon nanofabrication techniques provides an efficient route for simultaneous high-fidelity interfacing with hundreds of single neurons. Li et al.115 reported a method termed intracellular vesicle electrochemical cytometry. In this method, a carbon fiber microelectrode was fabricated into a conical nanotip to lyse individual nanoscale vesicles inside single PC12 cells and determine the total content of electroactive neurotransmitters (Fig. 6B). The results showed that during exocytosis only a fraction of the quanta of neurotransmitters were released. Wigström et al.116 introduced the lithographic microfabrication of a flexible thin-film microelectrode array (MEA) probe with 16 platinum band electrodes (Fig. 6C). Using this device, exocytosis from single bovine chromaffin cells was detected, which indicated the possibility of the two-dimensional localization of neurotransmitter release.
image file: c8an01079j-f6.tif
Fig. 6 Electrochemical methods of single-cell analysis. (A) Detection of exocytosis from single PC12 cells by an Au nanoelectrode. Reprinted with permission from ref. 113. Copyright 2015 American Chemical Society. (B) Intracellular vesicle electrochemical cytometry using a carbon fiber microelectrode. Reprinted with permission from ref. 115. Copyright 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim. (C) Flexible thin-film microelectrode array probe for the detection of exocytosis from single bovine chromaffin cells. Reprinted with permission from ref. 116. Copyright 2016 American Chemical Society. (D) Surfactant-assisted electrochemical detection of single cells. Reprinted with permission from ref. 118. Copyright 2016 The Royal Society of Chemistry.

Another important parameter for electrochemical approaches in single-cell analysis is oxidative stress. Piskounova et al.117 reported an extraordinary study that revealed in detail the inhibition of distant metastases of cancer cells by oxidative stress in a study conducted on human melanoma cells. Besides, a platinum electrode can also detect ions released as a result of oxidative stress such as free radicals and peroxides. Jeffrey E. Dick118 reported the electrochemical detection of reactive oxygen species (ROS) in single cells (Fig. 6D), which was achieved by the surfactant-assisted monitoring of the consumption of the contents of single cells owing to collisions with a microelectrode. The results showed a thousand-fold difference between acute lymphoblastic lymphoma T-cells and healthy thymocytes. Other advanced techniques can be derived from electrochemical analysis, such as electrochemical enzymatic assays,119–121 impedance spectroscopy,122,123 and scanning electrochemical microscopy.124–126 Readers interested in this field could consult the above related references. A future challenge comprises utilizing the high sensitivity of this technique to investigate infinitesimal intracellular and extracellular contents for an in-depth study of single-cell biology.

4.3 Mass spectrometric analysis

Mass spectrometry (MS) has the advantages of high detection limits, high sensitivity, label-free detection, and the ability to analyse hundreds of biomolecules from a single specimen.127,128 MS techniques are attractive in single-cell analysis owing to their ability to generate high-dimensional data within one single cell. We briefly classify the different single-cell MS techniques that are utilized by the form of ionization, namely, electrospray ionization MS (ESI-MS), matrix-assisted laser desorption/ionization MS (MALDI-MS), secondary-ion mass spectroscopy (SIMS), and inductively coupled plasma MS (ICPMS). Different ionization techniques have different focuses in terms of the biomolecules that are detected.

Electrospray ionization produces ions via a high voltage, which is applied to a liquid to create an aerosol. It maintains the integrity of macromolecules with little fragmentation and is thus a so-called soft ionization technique. This is attractive for the analysis of single-cell contents. Single-cell analysis utilizing ESI-MS requires the lysate of a single cell, which needs steps of sample pretreatment for better performance.129 Fujii et al.130 utilized microcapillaries to extract single plant cells and analyzed these by ESI-MS (Fig. 7A). This method enables the analysis of metabolites from a single cell with rapid sampling and high sensitivity. Gong et al.131 utilized a capillary to extract single live plant cells and detected metabolites at the cellular and subcellular levels.


image file: c8an01079j-f7.tif
Fig. 7 MS methods of single-cell analysis. (A) ESI approach for MS of single live plant cells. Reprinted with permission from ref. 130. Copyright 2015 Nature America, Inc. (B) High-throughput microscale sample preparation for single-cell MALDI-MS. Reprinted with permission from ref. 133. Copyright 2015, American Society for Microbiology. (C) TOF-SIMS detection of nanoparticle-induced changes in lipids in single cells. Reprinted with permission from ref. 135. Copyright 2018 American Chemical Society. (D) ICP-MS-based single-cell imaging. Reprinted with permission from ref. 137. Copyright 2016 American Chemical Society.

The MALDI technique is a relatively mild method of obtaining ions of macromolecules in the gas phase with little fragmentation, which is similar to the ESI method. The difference is that MALDI uses a matrix to absorb laser energy to help to create ions. MALDI-MS has been applied in single-cell analysis over the last few years. Ong et al.132 developed a single-cell MALDI-MS method that enabled the label-free classification of cells in the endocrine system on the basis of peptide profiles. In their study, a variety of cell types from the endocrine and nervous systems were analyzed, including cells from different organs in rats. Krismer et al.133 utilized MALDI to screen different populations of Chlamydomonas reinhardtii at a single-cell resolution. A stainless-steel plate with integrated microwells was utilized to capture single cells, and mass spectra of single cells were obtained by fast freezing and matrix deposition (Fig. 7B).

The SIMS technique sputters the surface of the sample with a focused primary ion beam and collects the ejected secondary ions. This approach is utilized to analyze the composition of solid surfaces and thin films and is the most sensitive technique for surface detection. It has the advantages of a low detection limit and high spatial resolution. The requirement for an ultrahigh vacuum for analysis limits the study of living organisms. Graham et al.134 utilized 3D TOF-SIMS to localize nanoparticles in single HeLa cells. A comparison of the TOF-SIMS results and optical images of cells incubated with fluorescent-labeled polymer nanoparticles demonstrated that the nanoparticles were clustered and correlated this with endosomal encapsulation. Hua et al.135 utilized TOF-SIMS to analyse the cytotoxicity of Ag NPs in combination with PCA at the molecular level (Fig. 7C). The results showed that levels of DAG and cholesterol increased after treatment with Ag NPs and most lipids migrated to the area surrounding the cells, which indicated the death of the cells. This approach has shown potential for the study of the metabolism of nanoparticles in single cells.

Inductively coupled plasma MS ionizes the sample with an inductively coupled plasma and is able to detect metals and several non-metals at extremely low concentrations of non-interfered low-background isotopes. The analysis of unlabeled single cells by ICP-MS is a recently developed field. Wang et al.136 developed a time-resolved ICP-MS approach for the detection of essential mineral elements in single cells. In experiments, the contents and distribution of several elements in different kinds of single cells were analyzed, and the results showed obvious differences. This approach provides a method for the analysis of mineral elements at a single-cell level. Van Malderen et al.137 were the first to carry out a cross-evaluation of LA-ICP-MS-based intracellular imaging and intracellular elemental distributions based on SR-XRF elemental analysis (Fig. 7D). Using both techniques, a center of accumulation was observed in single cells, which demonstrated the potential applicability of LA-ICP-MS in bioimaging at the subcellular level.

Single-cell MS is able to profile and image the compounds in a single cell. The individual profiling of a series of single cells is able to reveal important information on the heterogeneities of cells. However, this technique has a limited spatial resolution, and one single cell can only give rise to one spectrum. Further research may focus on the investigation of subcellular structures for an exhaustive understanding of these structures.

5 Conclusion and outlook

Currently, microfluidic devices for single-cell analysis have become more integrated, high-throughput, and sensitive. Developments will tend towards high-throughput, fully automated and completely integrated systems with advanced instruments. In the meantime, with the development of detection techniques such as optical microscopy, electrochemistry and mass spectrometry, the target of single-cell analysis will gradually tend towards subcellular and even single-molecule analysis in the near future. It is believed that modern single-cell analysis will lead us to a new insight into tumor cells and cancer, which will promote the field of personalized and precision medicine.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

We acknowledge the financial support from the National Key R&D Program of China (2017YFC0906800) and National Natural Science Foundation of China (No. 21435002, 21727814 and 21621003).

Notes and references

  1. L. Armbrecht and P. S. Dittrich, Anal. Chem., 2017, 89, 2–21 CrossRef CAS PubMed .
  2. G. M. Whitesides, Nature, 2006, 442, 368–373 CrossRef CAS PubMed .
  3. J. El-Ali, P. K. Sorger and K. F. Jensen, Nature, 2006, 442, 403–411 CrossRef CAS PubMed .
  4. S. Lindström and H. Andersson-Svahn, Biochim. Biophys. Acta, Gen. Subj., 2011, 1810, 308–316 CrossRef PubMed .
  5. S. H. Kim, G. H. Lee and J. Y. Park, Biomed. Eng. Lett., 2013, 3, 131–137 CrossRef .
  6. V. I. Chin, P. Taupin, S. Sanga, J. Scheel, F. H. Gage and S. N. Bhatia, Biotechnol. Bioeng., 2004, 88, 399–415 CrossRef CAS PubMed .
  7. A. Revzin, K. Sekine, A. Sin, R. G. Tompkins and M. Toner, Lab Chip, 2005, 5, 30–37 RSC .
  8. Q. Chen, J. Wu, Y. Zhang, Z. Lin and J. M. Lin, Lab Chip, 2012, 12, 5180–5185 RSC .
  9. X. Han, R. Wang, Y. Zhou, L. Fei, H. Sun, S. Lai, A. Saadatpour, Z. Zhou, H. Chen and F. Ye, Cell, 2018, 172, 1091–1107 CrossRef CAS PubMed .
  10. C. Wu, R. Chen, Y. Liu, Z. Yu, Y. Jiang and X. Cheng, Lab Chip, 2017, 17, 4008–4014 RSC .
  11. D. Di Carlo, N. Aghdam and L. P. Lee, Anal. Chem., 2006, 78, 4925–4930 CrossRef CAS PubMed .
  12. D. Di Carlo, L. Y. Wu and L. P. Lee, Lab Chip, 2006, 6, 1445–1449 RSC .
  13. H. Chen, J. Sun, E. Wolvetang and J. Cooperwhite, Lab Chip, 2015, 15, 1072–1083 RSC .
  14. J. Chung, Y. J. Kim and E. Yoon, Appl. Phys. Lett., 2011, 98, 123701 CrossRef PubMed .
  15. F. Yesilkoy, R. Ueno, B. X. Desbiolles, M. Grisi, Y. Sakai, B. J. Kim and J. Brugger, Biomicrofluidics, 2016, 10, 014120 CrossRef CAS PubMed .
  16. A. Skelley, O. Kirak, H. Suh, R. Jaenisch and J. Voldman, Nat. Methods, 2009, 6, 147–152 CrossRef CAS PubMed .
  17. M. A. Unger, H. P. Chou, T. Thorsen, A. Scherer and S. R. Quake, Science, 2000, 288, 113–116 CrossRef CAS PubMed .
  18. E. A. Ottesen, J. W. Hong, S. R. Quake and J. R. Leadbetter, Science, 2006, 314, 1464–1467 CrossRef CAS PubMed .
  19. A. M. Streets, X. Zhang, C. Cao, Y. Pang, X. Wu, L. Xiong, L. Yang, Y. Fu, L. Zhao and F. Tang, Proc. Natl. Acad. Sci. U. S. A., 2014, 111, 7048–7053 CrossRef CAS PubMed .
  20. H. C. Fan, J. Wang, A. Potanina and S. R. Quake, Nat. Biotechnol., 2011, 29, 51–57 CrossRef CAS PubMed .
  21. A. K. White, K. A. Heyries, C. Doolin, M. Vaninsberghe and C. L. Hansen, Anal. Chem., 2013, 85, 7182–7190 CrossRef CAS PubMed .
  22. A. Karimi, S. Yazdi and A. M. Ardekani, Biomicrofluidics, 2013, 7, 21501 CrossRef CAS PubMed .
  23. B. R. Lutz, J. Chen and D. T. Schwartz, Anal. Chem., 2006, 78, 5429–5435 CrossRef CAS PubMed .
  24. M. E. Warkiani, G. Guan, K. B. Luan, W. C. Lee, A. A. Bhagat, P. K. Chaudhuri, D. S. Tan, W. T. Lim, S. C. Lee and P. C. Chen, Lab Chip, 2014, 14, 128–137 RSC .
  25. E. W. Kemna, R. M. Schoeman, F. Wolbers, I. Vermes, D. A. Weitz and D. B. A. Van, Lab Chip, 2012, 12, 2881–2887 RSC .
  26. Q. Huang, S. Mao, M. Khan, L. Zhou and J.-M. Lin, Chem. Commun., 2018, 54, 2595–2598 RSC .
  27. C. M. Earhart, C. E. Hughes, R. S. Gaster, C. C. Ooi, R. J. Wilson, L. Y. Zhou, E. W. Humke, L. Xu, D. J. Wong and S. B. Willingham, Lab Chip, 2013, 14, 78–88 RSC .
  28. S. Choi, S. Song, C. Choi and J. K. Park, Lab Chip, 2007, 7, 1532–1538 RSC .
  29. X. Lu and X. Xuan, Anal. Chem., 2015, 87, 6389–6396 CrossRef CAS PubMed .
  30. S. Mao, W. Zhang, Q. Huang, M. Khan, H. Li, K. Uchiyama and J. M. Lin, Angew. Chem., Int. Ed., 2017, 57, 236–240 CrossRef PubMed .
  31. F. Fabbri, S. Carloni, W. Zoli, P. Ulivi, G. Gallerani, P. Fici, E. Chiadini, A. Passardi, G. L. Frassineti and A. Ragazzini, Cancer Lett., 2013, 335, 225–231 CrossRef CAS PubMed .
  32. S. K. Ameri, P. K. Singh, M. R. Dokmeci, A. Khademhosseini, Q. Xu and S. R. Sonkusale, Biosens. Bioelectron., 2014, 54, 462–467 CrossRef CAS PubMed .
  33. P. Zhang, L. Ren, X. Zhang, Y. Shan, Y. Wang, Y. Ji, H. Yin, W. E. Huang, J. Xu and B. Ma, Anal. Chem., 2015, 87, 2282–2289 CrossRef CAS PubMed .
  34. N. Mittal, A. Rosenthal and J. Voldman, Lab Chip, 2007, 7, 1146–1153 RSC .
  35. R. S. Thomas, H. Morgan and N. G. Green, Lab Chip, 2009, 9, 1534–1540 RSC .
  36. M. S. Jaeger, K. Uhlig, T. Schnelle and T. Mueller, J. Phys. D, 2008, 41, 2329–2342 CrossRef .
  37. R. S. W. Thomas, P. D. Mitchell, R. O. C. Oreffo and H. Morgan, Biomicrofluidics, 2010, 4, 1884–1888 CrossRef PubMed .
  38. A. Ashkin, J. M. Dziedzic and T. Yamane, Nature, 1987, 330, 769–771 CrossRef PubMed .
  39. C. Piggee, Anal. Chem., 2009, 81, 16–19 CrossRef CAS .
  40. A. Jonas and Z. Pavel, Electrophoresis, 2008, 29, 4813–4851 CrossRef CAS PubMed .
  41. J. R. Moffitt, Y. R. Chemla, S. B. Smith and C. Bustamante, Annu. Rev. Biochem., 2015, 77, 205–228 CrossRef PubMed .
  42. F. Bragheri, I. Cristiani, K. C. Vishnubhatla, L. Ferrara, N. Bellini, P. Minzioni, R. Osellame and R. Ramponi, Opt. Express, 2010, 18, 4679–4688 CrossRef PubMed .
  43. F. Bragheri, P. Minzioni, V. R. Martinez, N. Bellini, P. Paiè, C. Mondello, R. Ramponi, I. Cristiani and R. Osellame, Lab Chip, 2012, 12, 3779–3784 RSC .
  44. K. W. Huang, Y. C. Wu, J. A. Lee and P. Y. Chiou, Lab Chip, 2013, 13, 3721–3727 RSC .
  45. Y. Chen, T. H. Wu, Y. C. Kung, M. A. Teitell and P. Y. Chiou, Analyst, 2013, 138, 7308–7315 RSC .
  46. C. Yue, A. J. Chung, T. H. Wu, M. A. Teitell, D. Di Carlo and C. Pei-Yu, Small, 2014, 10, 1746–1751 CrossRef PubMed .
  47. A. Kang, J. Park, J. Ju, G. S. Jeong and S. H. Lee, Biomaterials, 2014, 35, 2651–2663 CrossRef CAS PubMed .
  48. E. Brouzes, M. Medkova, N. Savenelli, D. Marran, M. Twardowski, J. B. Hutchison, J. M. Rothberg, D. R. Link, N. Perrimon and M. L. Samuels, Proc. Natl. Acad. Sci. U. S. A., 2009, 106, 14195–14200 CrossRef CAS PubMed .
  49. D. J. Collins, A. Neild, A. Demello, A. Q. Liu and Y. Ai, Lab Chip, 2015, 15, 3439–3459 RSC .
  50. S. Koster, F. E. Angilè, H. Duan, J. J. Agresti, A. Wintner, C. Schmitz, A. C. Rowat, C. A. Merten, D. Pisignano and A. D. Griffiths, Lab Chip, 2008, 8, 1110–1115 RSC .
  51. N. Wen, Z. Zhao, B. Fan, D. Chen, D. Men, J. Wang and J. Chen, Molecules, 2016, 21, 881 CrossRef PubMed .
  52. J. F. Edd, D. Di Carlo, K. J. Humphry, S. Köster, D. Irimia, D. A. Weitz and M. Toner, Lab Chip, 2008, 8, 1262–1264 RSC .
  53. F. Chen, L. Lin, J. Zhang, Z. He, K. Uchiyama and J. M. Lin, Anal. Chem., 2016, 88, 4354–4360 CrossRef CAS PubMed .
  54. X. L. Guo, Y. Wei, Q. Lou, Y. Zhu and Q. Fang, Anal. Chem., 2018, 90, 5810–5817 CrossRef CAS PubMed .
  55. B. B. Lake, S. Chen, B. C. Sos, J. Fan, G. E. Kaeser, C. Y. Yun, T. E. Duong, D. Gao, J. Chun and P. V. Kharchenko, Nat. Biotechnol., 2017, 36, 70–80 CrossRef PubMed .
  56. H. S. Moon, K. Je, J. W. Min, D. Park, K. Y. Han, S. H. Shin, W. Y. Park, C. E. Yoo and S. H. Kim, Lab Chip, 2018, 18, 775–784 RSC .
  57. R. Zilionis, J. Nainys, A. Veres, V. Savova, D. Zemmour, A. M. Klein and L. Mazutis, Nat. Protoc., 2016, 12, 44–73 CrossRef PubMed .
  58. W. Stephenson, L. T. Donlin, A. Butler, C. Rozo, A. Rashidfarrokhi, S. M. Goodman, L. B. Ivashkiv, V. P. Bykerk, D. E. Orange and R. B. Darnell, Nat. Commun., 2018, 9, 791 CrossRef PubMed .
  59. L. Nan, Z. Jiang and X. Wei, Lab Chip, 2014, 14, 1060–1073 RSC .
  60. R. B. Brown and J. Audet, J. R. Soc., Interface, 2008, 5, S131–S138 CrossRef CAS PubMed .
  61. D. Di Carlo, K. H. Jeong and L. P. Lee, Lab Chip, 2003, 3, 287–291 RSC .
  62. H. Hoefemann, S. Wadle, N. Bakhtina, V. Kondrashov, N. Wangler and R. Zengerle, Sens. Actuators, B, 2012, 168, 442–445 CrossRef CAS .
  63. Y. Cheng, Y. Wang, Z. Wang, L. Huang, M. Bi, W. Xu, W. Wang and X. Ye, Biomicrofluidics, 2017, 11, 024112 CrossRef PubMed .
  64. Y. Gong, A. O. Ogunniyi and J. C. Love, Lab Chip, 2010, 10, 2334–2337 RSC .
  65. F. Stumpf, J. Schoendube, A. Gross, C. Rath, S. Niekrawietz, P. Koltay and G. Roth, Biosens. Bioelectron., 2015, 69, 301–306 CrossRef CAS PubMed .
  66. K. Leung, H. Zahn, T. Leaver, K. M. Konwar, N. W. Hanson, A. P. Pagé, C.-C. Lo, P. S. Chain, S. J. Hallam and C. L. Hansen, Proc. Natl. Acad. Sci. U. S. A., 2012, 109, 7665–7670 CrossRef CAS PubMed .
  67. A. Sarkar, S. Kolitz, D. A. Lauffenburger and J. Han, Nat. Commun., 2014, 5, 3421–3421 CrossRef PubMed .
  68. B. Treutlein, D. G. Brownfield, A. R. Wu, N. F. Neff, G. L. Mantalas, F. H. Espinoza, T. J. Desai, M. A. Krasnow and S. R. Quake, Nature, 2014, 509, 371–375 CrossRef CAS PubMed .
  69. C. Noelker, L. Morel, A. Osterloh, D. Alvarez-Fischer, T. Lescot, M. Breloer, M. Gold, W. H. Oertel, C. Henze and P. P. Michel, J. Neuroinflammation, 2014, 11, 86 CrossRef PubMed .
  70. D. Anjan, P. Derek, R. M. Andrade, H. Chen, E. R. Cobo, H. Ken, S. Chen, G. R. Guillermina, O. Esther and M. B. Martínez, Nat. Med., 2012, 18, 956–960 CrossRef PubMed .
  71. Z. Yu, Y. Q. Gao, H. Feng, Y. Y. Lee, M. S. Li, Y. Tian, M. Y. Y. Go, D. Y. Yu, Y. S. Cheung and P. B. S. Lai, Gut, 2014, 63, 1793–1804 CrossRef CAS PubMed .
  72. Q. Shi, L. Qin, W. Wei, F. Geng, R. Fan, Y. S. Shin, D. Guo, L. Hood, P. S. Mischel and J. R. Heath, Proc. Natl. Acad. Sci. U. S. A., 2012, 109, 419–424 CrossRef CAS PubMed .
  73. B. J. DeKosky, T. Kojima, A. Rodin, W. Charab, G. C. Ippolito, A. D. Ellington and G. Georgiou, Nat. Med., 2015, 21, 86–91 CrossRef CAS PubMed .
  74. E. Z. Macosko, A. Basu, R. Satija, J. Nemesh, K. Shekhar, M. Goldman, I. Tirosh, A. R. Bialas, N. Kamitaki and E. M. Martersteck, Cell, 2015, 161, 1202–1214 CrossRef CAS PubMed .
  75. A. Zinchenko, S. R. A. Devenish, B. Kintses, P. Y. Colin, M. Fischlechner and F. Hollfelder, Anal. Chem., 2014, 86, 2526–2533 CrossRef CAS PubMed .
  76. A. M. Klein, L. Mazutis, I. Akartuna, N. Tallapragada, A. Veres, V. Li, L. Peshkin, D. A. Weitz and M. W. Kirschner, Cell, 2015, 161, 1187–1201 CrossRef CAS PubMed .
  77. A. Rival, D. Jary, C. Delattre, Y. Fouillet, G. Castellan, A. Bellemincomte and X. Gidrol, Lab Chip, 2014, 14, 3739–3749 RSC .
  78. S. W. Lee and Y. C. Tai, Sens. Actuators, A, 1999, 73, 74–79 CrossRef CAS .
  79. H. Lu, M. A. Schmidt and K. F. Jensen, Lab Chip, 2005, 5, 23–29 RSC .
  80. J. S. Mellors, K. Jorabchi, L. M. Smith and J. M. Ramsey, Anal. Chem., 2010, 82, 967–973 CrossRef CAS PubMed .
  81. C.-W. Young, J.-L. Hsieh and C. Ay, Sensors, 2012, 12, 2400–2413 CrossRef CAS PubMed .
  82. N. de Lange, T. M. Tran and A. R. Abate, Biomicrofluidics, 2016, 10, 024114 CrossRef CAS PubMed .
  83. H.-H. Lai, P. A. Quinto-Su, C. E. Sims, M. Bachman, G. Li, V. Venugopalan, N. L. Allbritton and J. R. Soc, Interface, 2008, 5, S113–S121 CAS .
  84. T. C. Marentis, B. Kusler, G. G. Yaralioglu, S. Liu, E. O. Haggstrom and B. Khuri-Yakub, Ultrasound Med. Biol., 2005, 31, 1265–1277 CrossRef PubMed .
  85. A. M. Sydor, K. J. Czymmek, E. M. Puchner and V. Mennella, Trends Cell Biol., 2015, 25, 730–748 CrossRef CAS PubMed .
  86. S. Hohng, S. Lee, J. Lee and M. H. Jo, Chem. Soc. Rev., 2014, 43, 1007–1013 RSC .
  87. K. J. Son, D. S. Shin, T. Kwa, J. You, Y. Gao and A. Revzin, Lab Chip, 2015, 15, 637–641 RSC .
  88. E. X. Ng, M. A. Miller, T. Jing and C. H. Chen, Biosens. Bioelectron., 2016, 81, 408–414 CrossRef CAS PubMed .
  89. K. Perez-Toralla, G. Mottet, E. T. Guneri, J. Champ, F. C. Bidard, J. Y. Pierga, J. Klijanienko, I. Draskovic, L. Malaquin and J. L. Viovy, Lab Chip, 2015, 15, 811–822 RSC .
  90. J. R. Moffitt, J. Hao, G. Wang, K. H. Chen, H. P. Babcock and X. Zhuang, Proc. Natl. Acad. Sci. U. S. A., 2016, 113, 11046–11051 CrossRef CAS PubMed .
  91. R. Jungmann, M. S. Avendaño, J. B. Woehrstein, M. Dai, W. M. Shih and P. Yin, Nat. Methods, 2014, 11, 313–318 CrossRef CAS PubMed .
  92. T. Niehörster, A. Löschberger, I. Gregor, B. Krämer, H. J. Rahn, M. Patting, F. Koberling, J. Enderlein and M. Sauer, Nat. Methods, 2016, 13, 257–262 CrossRef PubMed .
  93. N. M. Mishina, A. S. Mishin, Y. Belyaev, E. A. Bogdanova, S. Lukyanov, C. Schultz and V. V. Belousov, Nano Lett., 2015, 15, 2928–2932 CrossRef CAS PubMed .
  94. D. M. Andrade, M. P. Clausen, J. Keller, V. Mueller, C. Wu, J. E. Bear, S. W. Hell, B. C. Lagerholm and C. Eggeling, Sci. Rep., 2015, 5, 11454 CrossRef CAS PubMed .
  95. B. Dudok, L. Barna, M. Ledri, S. I. Szabó, E. Szabadits, B. Pintér, S. G. Woodhams, C. M. Henstridge, G. Y. Balla and R. Nyilas, Nat. Neurosci., 2015, 18, 75–86 CrossRef CAS PubMed .
  96. Y. Wang, W. E. Huang, L. Cui and M. Wagner, Curr. Opin. Biotechnol., 2016, 41, 34–42 CrossRef CAS PubMed .
  97. B. Kann, H. L. Offerhaus, M. Windbergs and C. Otto, Adv. Drug Delivery Rev., 2015, 89, 71–90 CrossRef CAS PubMed .
  98. S. Casabella, P. Scully, N. Goddard and P. Gardner, Analyst, 2016, 141, 689–696 RSC .
  99. B. Kang, M. M. Afifi, L. A. Austin and M. A. El-Sayed, ACS Nano, 2013, 7, 7420–7427 CrossRef CAS PubMed .
  100. Y. Song, H. Yin and W. E. Huang, Curr. Opin. Chem. Biol., 2016, 33, 1–8 CrossRef CAS PubMed .
  101. I. Stojanović, V. D. V. Tj, H. W. Mulder, R. B. Schasfoort and L. W. Terstappen, Anal. Biochem., 2015, 485, 112–118 CrossRef PubMed .
  102. W. G. De, J. S. Danial, P. Kukura and M. I. Wallace, Proc. Natl. Acad. Sci. U. S. A., 2015, 112, 12299–12303 CrossRef PubMed .
  103. J. A. Ortega, J. Andrecka, K. M. Spillane, N. Billington, Y. Takagi, J. R. Sellers and P. Kukura, Nano Lett., 2014, 14, 2065–2070 CrossRef PubMed .
  104. B. Agnarsson, A. Lundgren, A. Gunnarsson, M. Rabe, A. Kunze, M. Mapar, L. Simonsson, M. Bally, V. P. Zhdanov and F. Höök, ACS Nano, 2015, 9, 11849–11862 CrossRef CAS PubMed .
  105. K. Nienhaus and G. U. Nienhaus, Chem. Soc. Rev., 2014, 43, 1088–1106 RSC .
  106. D. L. S. Van, M. Heilemann and M. Sauer, Annu. Rev. Phys. Chem., 2012, 63, 519–540 CrossRef PubMed .
  107. K. R. Lee, K. Kim, J. Jung, J. H. Heo, S. Cho, S. Lee, G. Chang, Y. J. Jo, H. Park and Y. K. Park, Sensors, 2013, 13, 4170–4191 CrossRef PubMed .
  108. J. R. Moffitt, J. B. Lee and P. Cluzel, Lab Chip, 2012, 12, 1487–1494 RSC .
  109. M. D. Hoffman, L. I. Zucker, P. J. B. Brown, D. T. Kysela, Y. V. Brun and S. C. Jacobson, Anal. Chem., 2015, 87, 12032–12039 CrossRef CAS PubMed .
  110. C. E. M. Krämer, W. Wolfgang and K. Dietrich, Sci. Rep., 2016, 6, 32104 CrossRef PubMed .
  111. E. S. Bucher and R. M. Wightman, Annu. Rev. Anal. Chem., 2015, 8, 239–261 CrossRef CAS PubMed .
  112. A. E. Vasdekis and G. Stephanopoulos, Metab. Eng., 2015, 27, 115–135 CrossRef CAS PubMed .
  113. Y. Liu, M. Li, F. Zhang, A. Zhu and G. Shi, Anal. Chem., 2015, 87, 5531–5538 CrossRef CAS PubMed .
  114. J. T. Robinson, M. Jorgolli, A. K. Shalek, M. H. Yoon, R. S. Gertner and H. Park, Nat. Nanotechnol., 2012, 7, 180–184 CrossRef CAS PubMed .
  115. X. Li, S. Majdi, J. Dunevall, H. Fathali and A. G. Ewing, Angew. Chem., Int. Ed., 2015, 54, 11978–11982 CrossRef CAS PubMed .
  116. J. Wigström, J. Dunevall, N. Najafinobar, J. Lovric, J. Wang, A. G. Ewing and A. S. Cans, Anal. Chem., 2016, 88, 2080–2087 CrossRef PubMed .
  117. E. Piskounova, M. Agathocleous, M. M. Murphy, Z. Hu, S. E. Huddlestun, Z. Zhao, A. M. Leitch, T. M. Johnson, R. J. Deberardinis and S. J. Morrison, Nature, 2015, 527, 186–191 CrossRef CAS PubMed .
  118. J. E. Dick, Chem. Commun., 2016, 52, 10906–10909 RSC .
  119. T. S. Safaei, R. M. Mohamadi, E. H. Sargent and S. O. Kelley, ACS Appl. Mater. Interfaces, 2015, 7, 14165–14169 CrossRef CAS PubMed .
  120. X. Chen, Y. Wang, Y. Zhang, Z. Chen, Y. Liu, Z. Li and J. Li, Anal. Chem., 2014, 86, 4278–4286 CrossRef CAS PubMed .
  121. J. Chen, C. Xue, Y. Zhao, D. Chen, M. H. Wu and J. Wang, Int. J. Mol. Sci., 2015, 16, 9804–9830 CrossRef CAS PubMed .
  122. J. Kim, H. Cho, S. I. Han and K. H. Han, Anal. Chem., 2016, 88, 4857–4863 CrossRef CAS PubMed .
  123. N. Haandbæk, O. With, S. C. Bürgel, F. Heer and A. Hierlemann, Lab Chip, 2014, 14, 3313–3324 RSC .
  124. J. A. Koch, M. B. Baur, E. L. Woodall and J. E. Baur, Anal. Chem., 2012, 84, 9537–9543 CrossRef CAS PubMed .
  125. M. N. Zhang, Z. Ding and Y. T. Long, Analyst, 2015, 140, 6054–6060 RSC .
  126. F. P. Filice, M. S. M. Li, J. D. Henderson and Z. Ding, Anal. Chim. Acta, 2016, 908, 85–94 CrossRef CAS PubMed .
  127. T. J. Comi, T. D. Do, S. S. Rubakhin and J. V. Sweedler, J. Am. Chem. Soc., 2017, 139, 3920–3929 CrossRef CAS PubMed .
  128. Y. Yang, Y. Huang, J. Wu, N. Liu, J. Deng and T. Luan, TrAC, Trends Anal. Chem., 2017, 90, 14–26 CrossRef CAS .
  129. L. Zhang, D. P. Foreman, P. A. Grant, B. Shrestha, S. A. Moody, F. Villiers, J. M. Kwak and A. Vertes, Analyst, 2014, 139, 5079–5085 RSC .
  130. T. Fujii, S. Matsuda, M. L. Tejedor, T. Esaki, I. Sakane, H. Mizuno, N. Tsuyama and T. Masujima, Nat. Protoc., 2015, 10, 1445–1456 CrossRef CAS PubMed .
  131. X. Gong, Y. Zhao, S. Cai, S. Fu, C. Yang, S. Zhang and X. Zhang, Anal. Chem., 2014, 86, 3809–3816 CrossRef CAS PubMed .
  132. T. H. Ong, D. J. Kissick, E. T. Jansson, T. J. Comi, E. V. Romanova, S. S. Rubakhin and J. V. Sweedler, Anal. Chem., 2015, 87, 7036–7042 CrossRef CAS PubMed .
  133. J. Krismer, J. Sobek, R. F. Steinhoff, S. R. Fagerer, M. Pabst and R. Zenobi, Appl. Environ. Microbiol., 2015, 81, 5546–5551 CrossRef CAS PubMed .
  134. D. J. Graham, J. T. Wilson, J. J. Lai, P. S. Stayton and D. G. Castner, Biointerphases, 2015, 11, 02A304 CrossRef PubMed .
  135. X. Hua, H. W. Li and Y. T. Long, Anal. Chem., 2018, 90, 1072–1076 CrossRef CAS PubMed .
  136. H. Wang, B. Wang, M. Wang, L. Zheng, H. Chen, Z. Chai, Y. Zhao and W. Feng, Analyst, 2015, 140, 523–531 RSC .
  137. S. J. M. Van Malderen, E. Vergucht, M. De Rijcke, C. R. Janssen, L. Vincze and F. Vanhaecke, Anal. Chem., 2016, 88, 5783–5789 CrossRef CAS PubMed .

This journal is © The Royal Society of Chemistry 2019