1. Introduction
The term
biomolecules generally refers to the molecules that naturally exist in living organisms and serve as the fundamental units of life. Biomolecules can provide information about the body of an organism and reflect its physiological state. They primarily include small molecules (sterols, vitamins, hormones, and carbohydrates); monomers (amino acids, nucleotides, and phosphoric acid); and polymers (proteins, nucleic acids, and polysaccharides). Currently, phenotypes, as the physical manifestations of gene function, have attracted considerable interest and have emerged as a focal point of medical research. There is also great interest in intracellular or membrane proteins, which are involved in transcription, translation, metabolism, growth, adhesion, signaling, and other functions, forming the basis of cell structure in single cells. Single-cell secreted proteins primarily include cytokines, growth factors, and hormones, which can mediate cell communication and govern cell, tissue, and organ activity; and secreted proteins that inhibit tumor development or encourage tumor spread. Hence, the monitoring of biomolecules and single cells is crucial in clinical diagnosis, molecular biology, and the early diagnosis and treatment of diseases
[1].
Various assays based on macromolecules and single cells have been developed in previous studies. Among them, traditional detection methods have great value, such as enzyme-linked immunosorbent assay (ELISA)
[2], polymerase chain reaction (PCR)
[3], mass spectrometry
[4], western blot
[5], and so forth. Although these methods are accurate, their complex procedures and the need for specialized personnel to carry them out
[6],
[7] limit their applications. To address these limitations and meet the requirements for speed, high sensitivity, and low detection limits, electrochemical
[8], fluorescence
[9], Raman
[10], microfluidic
[11], and other technologies have been developed and widely applied. However, many technologies are constrained to the study of a single biomolecule, which limits their effectiveness in comprehensive disease diagnosis, detection, and treatment. Assessing multiple potential targets through such methods not only increases the expense of the experiment but also escalates the demand for samples while reducing efficiency. In contrast, high-throughput analysis enables the concurrent detection of multiple targets across various samples within a single test. Compared with traditional methodologies, this approach offers several unique advantages, such as miniaturization, low consumption, expedited processing time, high-throughput capabilities, and the elimination of costly analytical equipment.
In recent years, microfluidic technology, with its advantages of being high-throughput and rapid, with low cost and low reagent consumption
[12], has had the capability to precisely manage the volume and execute a large number of tests in a short period of time, particularly for the high-throughput analysis of complicated biological materials. These technologies include droplet microfluidics
[13], lattice microfluidics
[14], and microfluidic barcode technology
[15],
[16]. Among them, droplet-based microfluidics is capable of precisely controlling discrete droplets ranging from the picoliter to nanoliter scale
[17]. This technology is now widely applied in various fields, including DNA identification and sequencing, immunoassays, clinical diagnosis, and drug screening
[18]. Xiao et al.
[19] presented an innovative droplet-based DNA-walker platform technology that uses two-dimensional (2D) barcodes to encode droplets, enabling rapid, high-throughput, one-step, and synchronous bacterial identification (
Fig. 1(a)). Although significant advancements have been achieved in droplet-based microfluidics, this technology still presents challenges in practical applications, such as limited stability and high cost, and is typically confined to specific microfluidic laboratory environments.
Microfluidic lattice technology is realized by depositing microspots of DNA or proteins on a functionalized glass slide
[20]. This technique has been widely used in drug screening, protein–protein interactions, and proteomics, due to its excellent sensitivity and specificity
[21]. Chi et al.
[22] described a method for the quantitative detection of biomarkers by means of a microarray and reported the high sensitivity and high stability storage of this platform (
Fig. 1(b)). However, its low detection repeatability, complex preparation, low detection sensitivity, limited dynamic detection range, trailing effect, and uneven spot arrays contribute to analyte cross-contamination and inaccurate signal readings
[20],
[23]. Microfluidic barcode detection technology is based on the formation of antibody/probe barcodes via fixing on a functional glass substrate. To date, this technique has been extensively applied in biomarker screening (
Fig. 1(c))
[24] and single-cell analysis (single-cell secreted protein heterogeneity, single-cell exosome heterogeneity, etc.)
[15],
[16],
[25],
[26],
[27],
[28].
With innovations in nanotechnology, complex analytical functions have been downsized and combined onto a single detection platform. This has led to the widespread application of microfluidic systems in tandem with diverse technologies for the study of biomolecules and single-cell analysis
[16],
[25],
[26]. The integration of microfluidic barcode technology with nanomaterials and polymeric materials notably addresses the limitations often encountered in other technologies. These include issues such as low flux, poor reproducibility, difficulty in preparation, low sensitivity, a restricted dynamic detection range, trailing effects, and uneven speck arrays, all of which contribute to the risks of analyte cross-contamination and inaccurate signal readings. At the same time, the design of a multichannel configuration serves to partition the test samples into numerous isolated reaction units. This ensures that the samples do not interfere with each other, ultimately enabling high throughput, high sensitivity, and rapid detection. The benefits of microfluidic barcode technology include the following aspects: First, the nanomaterials or polymer materials uniformly self-assemble onto a microfluidic barcode substrate. Second, nanomaterials offer the advantages of catalytic properties, optical characteristics, a large surface-area ratio, Förster resonance energy transfer (FRET), and functional simplicity. These features improve capture rates, boost the signal-to-noise ratio, reduce detection limits, and improve sensitivity
[29],
[30]. Overall, microfluidic barcode detection technology enables the simultaneous detection of multiple biomarkers, making such detection faster, more efficient, accurate, and sensitive. This innovation lays a robust foundation for future medical research and opens up new prospects.
With the development of microfluidic technology, several reviews have been published to summarize and evaluate the potential applications of this technology over the past few years. Chen et al.
[31] offered a comprehensive overview of the advancements in microfluidic technologies based on nanomaterials, focusing on optical, electrical, magnetic, and acoustic aspects. Li et al.
[32] reviewed recent advancements in nucleic acid detection with microfluidic chips in molecular diagnostics. Zhang et al.
[33] reviewed the application of droplet microfluidic technology in biomolecules. Qin et al.
[34] concentrated their research on paper-based microfluidic chips for rapid molecular detection. However, a comprehensive review focusing specifically on the application of microfluidic barcode technology to biomolecules and single cells has not yet been conducted.
As microfluidic barcode technology has become widely accepted, achieved unprecedented development, and realized high-throughput detection capabilities, it possesses bright prospects in the fields of biomolecule and single-cell analysis
[26],
[27]. In this work, we conducted an online search for articles related to microfluidic high throughput and microfluidic barcodes published in the last 20 years, as illustrated in
Figs. 2(a) and
(b). The findings suggest that microfluidic barcode technology is garnering increasing interest, but there is still much room for advancement. In this review, we summarize the main advantages and recent advances in microfluidic barcode biochips for the high-throughput detection of biomolecules and single cells. This review aims to ① discuss common high-throughput analysis techniques based on microfluidics, such as droplet microfluidics, paper-based microfluidics, and microfluidic arrays, and their advantages and disadvantages; ② emphasize the advantages and disadvantages of microfluidic lattice technology and microfluidic barcode technology; ③ summarize the materials used as self-assembled microfluidic barcode biochip substrates and compare their advantages and disadvantages; ④ summarize the application of high-throughput detection based on microfluidic barcode technology in the fields of biomolecule and single-cell analysis, as illustrated in
Fig. 2(c); and ⑤ present the latest challenges and future perspectives in these fields. This work reviews the future development of microfluidic barcode biochips for the high-throughput detection of biomolecules and single cells, offering valuable insights for future endeavors in biological diagnostics and research.
2. Microfluidic barcode biochip self-assembly substrate materials
A microfluidic barcode biochip consists of three components: a microfluidic chip self-assembly substrate, a polydimethylsiloxane (PDMS) antibody loading layer chip, and a sample loading layer chip
[16],
[35]. The ability of the microfluidic self-assembly substrate to support patterned antibody barcodes in order to identify target molecules is critical in the entire process of microfluidic barcode biochip detection. We investigated some of the currently available substrates for microfluidic barcode biochip self-assembly, including poly-
L-lysine (PLL), nanomaterials, and nanocomposites. In general, these materials possess excellent physical and chemical characteristics, such as biocompatibility, stability, optical qualities, and ease of functionalization
[36],
[37],
[38]. Herein, we discuss representative materials utilized in microfluidic barcode biochips, highlighting their significant advantages.
2.1. Poly-L-lysine
PLL is an edible, water-soluble, biodegradable, and nontoxic cationic polymer that can be employed in medicine delivery, biodegradation, absorbent gel, food preservation, drug- and gene-delivery vectors, and other applications in biomedicine and other fields
[39],
[40],
[41],
[42]. Because of its positive charge, PLL can bind a DNA probe with a negative charge to identify the target
[43]. Moreover, thanks to its exceptional characteristics, PLL can serve as an appropriate scaffold for protein loading
[44]. Based on this, the Fan group
[25],
[26],
[45] and the Lu group
[46],
[47] used PLL self-assembled chips to immobilize antibody barcodes in order to identify single-cell secreted proteins and single-cell exosomes (
Figs. 3(a) and
(b)). Our group
[48] developed a novel PLL-integrated graphene field-effect transistor (PGFET) biosensor for RNA detection, thereby laying a solid foundation for cancer diagnosis and disease screening (
Fig. 3(c)). We
[49],
[50] also reported a graphene oxide (GO) and PLL self-assembled microfluidic dual-layer chip for high-throughput microRNA (miRNA) detection (
Fig. 3(d)).
2.2. Nanomaterials
Due to their large surface area, high conductivity, chemical stability, facile functionalization, biocompatibility, FRET, and biodegradability
[51],
[52], nanomaterials are widely employed in disease monitoring, environmental pollution, and food safety. In microfluidic barcode sensing platforms, the localized self-assembly of nanomaterials on microfluidic substrates significantly improves the high-throughput detection performance of microfluidic barcode biochips for biomolecules and single cells
[15],
[28],
[35], especially in the following aspects: ① With their large surface area and active functional groups, nanomaterials facilitate more efficient target enrichment; ② the unique optical, electrical, and plasmonic properties of nanomaterials contribute to enhancing the detection sensitivity and reducing the detection limit; and ③ FRET can reduce nonspecific adsorption, thereby improving the signal-to-noise ratio and sensitivity.
Research indicates that a limited range of materials have been integrated into the antibody barcode patterns of microfluidic barcode biochip systems. Herein, we focus on GO, graphene oxide quantum dots (GOQDs), quantum dots (QDs), metal nanoparticles, and ZnO, along with their applications in microfluidic barcode biochips. GO, which has exceptional fluorescence-quenching capabilities, good biocompatibility, extensive contact area, excellent water dispersibility, and ease of surface modification, is considered a highly promising 2D nanomaterial
[53],
[54]. The 2D surfaces of GO exhibit strong noncovalent interactions with identified biomolecules through π–π interactions, electrostatic adsorption, or hydrogen bonding, providing valuable information for biomolecular detection
[54],
[55]. In a microfluidic barcode biochip, GO is self-assembled onto a functionalized glass substrate through electrostatic attraction. Subsequently, antibody or DNA probes are loaded through the PDMS channel layer to construct a microfluidic barcode array for antibody profiling or DNA profiling. Our research
[49],
[56],
[57] was used to create a self-assembled microfluidic barcode biochip based on GO to detect nucleic acids and other markers. Although various studies have widely revealed that GO can be self-assembled in microfluidic sensing platforms, there are limited applications at present in microfluidic barcode biochip platforms (
Fig. 4(a) [49]).
The advantages shared by GOQDs and GO include nontoxicity, solubility, biocompatibility, and environmental friendliness. In particular, the size of GOQDs is less than 10 nm, such that their diminutive dimensions result in low steric hindrance and enhanced stability, facilitating easier penetration through the cell membrane. This unique property makes GOQDs a candidate for drug delivery and
in vivo and
in vitro bioimaging
[9],
[58]. Our group
[24],
[27],
[35],
[59] developed a microfluidic barcode chip based on GOQD self-assembly, which takes advantage of its FRET, large specific surface area, small particle size, biocompatibility, and other advantages to detect and analyze biomolecules, the heterogeneity of single-cell secreted proteins, and single-cell exosomes (
Figs. 4(b) and
(c)). This microfluidic chip is capable of simultaneously detecting 20 different biomarkers and 60 samples, significantly enhancing work efficiency.
Compared with other fluorescent-labeled sensors, QDs demonstrate distinct advantages. QDs of different sizes emit light of varying wavelengths when excited at a single wavelength; this property allows them to be easily attached to biological groups or biomolecules, which can then be identified through multicolor fluorescence reactions. Hu et al.
[60] developed a microfluidic barcode protein chip that integrates the high-throughput detection capability of microfluidics with the highly sensitive, multicolor imaging capability of QDs for the ultimate detection of biomarkers in serum (
Fig. 5(a)). This detection system opens new avenues for the practical clinical application and proteomics research of protein chips.
Furthermore, the combination of ZnO and microfluidics provides a promising analytical method
[61]. Liu et al.
[62] developed ZnO microfluidic arrays for proteins. This combination provides valuable information for the development of high sensitivity, low-cost, and high-throughput protein detection platforms. Based on previous work, Hu et al.
[63] also reported a ZnO nanorod-enhanced fluorescence microarray chip for the first time. A layer of organic matter on the ZnO surface was used to enhance the loading of capture proteins and reject the adsorption of nonspecific proteins (
Fig. 5(b) [63]). Guo et al.
[64] directly grew ZnO nanowires in a microfluidic channel by means of a hydrothermal method for carcinoembryonic antigen (CEA) and alpha-fetoprotein (AFP) detection, as shown in
Fig. 5(c).
2.3. Nanocomposites
Nanocomposites exhibit various structural forms, such as nanoflowers, nanostars, and nanoprisms, all comprising three-dimensional (3D) structures assembled from nanomaterials of varying sizes
[65]. In microfluidic platforms, nanomaterials have been extensively applied for biomarker detection, biomarker imaging, and environmental protection
[16],
[29],
[66]. To date, very few nanocomposites have been used in microfluidic barcode biochips.
Hogan et al.
[67] achieved precise
in situ alignment monitoring for the first practical realization of 3D photonic metastructure shaping based on 2D-fluid composites and a complementary metal-oxide-semiconductor (CMOS) photonics platform (
Fig. 6(a)). To improve the efficiency of the experiment by addressing the limitations of low experimental flux and long reaction time, our group
[16] developed a self-assembled microfluidic barcode biochip based on L-Cysteine (Lys)-Au nanoparticles (NPs)@MoS
2, which contains ten parallel channels and can simultaneously detect ten proteins with high throughput and sensitivity. In this microfluidic platform, Lys-AuNPs@MoS
2 has a large contact area, excellent stability, and multiple binding sites, which increases the antibody loading density and provides a promising method for clinical diagnostic applications (
Fig. 6(b) [16]). Wang et al.
[68] prepared a surface-enhanced Raman scattering (SERS) integrated microfluidic chip for the high-throughput detection of melamine in milk (
Fig. 6(c)). Due to the limitations of the microfluidic SERS chip in clinical detection, a microfluidic SERS chip integrated with Ag/AuNCs SERS substrates and automatic sampling was designed and prepared for the detection of clinical samples, providing reasonable chip design and manufacturing and reduced interference (
Fig. 6(d))
[69].
At present, the variety of substrate materials available for microfluidic technology is somewhat limited and costly. Among various products, the PLL substrate (model: P4981-001) manufactured by Thermo Fisher Scientific (USA) is the representative substrate. However, organic substrate materials such as PLL are susceptible to environmental influences and are not conducive to long-term preservation—characteristics that compromise the uniformity of the substrate to some extent. In contrast, inorganic materials, which are characterized by their nanoscale size, more uniform self-assembly, and longer shelf life, are considered to be the main trend for future substrate development. Consequently, the development of substrate materials that are stable, uniform, cost-effective, and suitable for large-area applications remains a significant challenge and is crucial for the development of high-performance sensing chips.
3. Microfluidic barcode biochips for the high-throughput detection of biomarkers
Biomarkers are generally biochemical molecules, such as proteins, genomes, and metabolites
[70],
[71],
[72]. They are commonly employed in medical research as biochemical indicators that can be utilized to characterize human systems, organs, tissues, cells, and subcellular structures. At present, the biological processes of organisms and the early detection and diagnosis, prevention, and prognosis of diseases are studied by measuring biomarkers
[73],
[74]. Therefore, in this chapter, we classify three important groups of molecular biomarkers namely, proteins, nucleic acids, and small molecules and briefly introduce their characteristics, importance, advantages, and detection methods, with a focus on the high-throughput detection of biomarkers using microfluidic barcode biochip technology.
3.1. Proteins
Proteins are organic macromolecules that serve as the foundation of biological matter. They are key biomarkers that engage in a range of biological functions, including information transmission, viral defense, cell structure and support, and catalysis. Abnormal protein expression has been well established to be connected with disorders such as cardiovascular disease, tumors, and cancer
[22],
[75]. As a result, protein biomarkers are crucial for early disease diagnosis, therapy, and prognosis. The detection of protein markers plays an important role in clinical practice. Current traditional methods for protein detection include electrochemistry, fluorescence, Raman spectroscopy, colorimetry, and ELISA
[76],
[77],
[78]. However, these methods usually take a long time, have a high detection limit, and are expensive
[79]. To further reduce disease risk, novel sensing technologies for the rapid and convenient screening of protein markers must be developed
[80]. With technological development and innovation, complex analysis techniques can be miniaturized and integrated into a platform. Microfluidic technology has developed into a potential technology for high-throughput analysis. It can perform high-throughput analysis in a short time and has a number of advantages, particularly in complicated multi-biological samples, such as high throughput, a small sample size, a high degree of integration, ease of transport, and low cost
[12]. Below, we discuss the application of an integrated microfluidic platform in the detection of protein markers.
Microfluidic paper-based analysis devices have the advantages of small size, portability, biodegradability, and low cost, and have been widely used in the detection of protein markers
[81]. Sanjayan et al.
[82] developed an integrated perovskite QD-based paper microfluidic device for the simultaneous detection of CEA and neuron-specific enolase (NSE) with high throughput, high sensitivity, and multicolor imaging capabilities. Chang et al.
[83] reported a low-cost 3D microfluidic chip that detects three proteins simultaneously by recognizing the color of photonic crystal beads (PCBs). Zhao et al.
[78] proposed a novel fluorescence enhancement based on the structure of ZnO nanorod arrays to achieve highly sensitive CEA detection.
Due to its advantages of high sensitivity and limit of detection (LOD) microfluidic technology is currently being extensively used in the field of protein analysis. Therefore, the rapid development of microfluidic barcode biochip technology is attributed mostly to its benefits of multiple channels, sample conservation, and the potential to combine nanomaterials
[16],
[35],
[60].
High-throughput, high-sensitivity, and high-specificity protein marker detection can be achieved by integrating nanomaterials with a small size, high energy resonance transmission, and high capture efficiency onto microfluidic chips. Hu et al.
[60] proposed integrating the high-throughput features of microfluidic chips with the fluorescence properties of QDs to create a microfluidic protein chip, thereby offering a valuable tool for large-scale clinical applications. Recently, Qiu et al.
[16] proposed a Lys-AuNPs@MoS
2 self-assembled microfluidic biochip with a digital signal output and applied it to the simultaneous detection of multiple biomarkers. The biochip can achieve the detection of ten biomarkers within a short time frame (
Fig. 7(a) [16]). At the same time, our group
[24],
[35] developed a microfluidic barcode platform based on GOQDs, which can detect 60 samples and 20 kinds of biomarkers based on previous works (
Fig. 7(b)). We also investigated the effects of mechanical force on the cytokine release of early astrocytes differentiated from human neural stem cells
[59].
3.2. miRNA
Nucleic acids, which include DNA and RNA, are a type of biological macromolecule
[84]. Nucleic acid sequences convey a significant amount of genetic information, are vital in cell activity, and are utilized as biomarkers for cancer and other disorders
[85],
[86]. Rapid, efficient, low-cost, and noninvasive nucleic acid testing technologies are critical in clinical applications such as disease detection, assessment, and therapy
[87],
[88],
[89]. The nucleic acid testing procedure includes sample pretreatment, nucleic acid extraction amplification, and signal detection
[90],
[91]. The sample pretreatment technique extracts nucleic acid from the sample while avoiding interference from interfering chemicals in the sample. Because the concentration of nucleic acid in a sample is too low to meet the detection requirements, amplification is a necessary condition to raise the detection limit and sensitivity. Amplification technologies used in medical diagnostic research include rolling circle amplification (RCA)
[92], PCR
[93], chain replacement amplification
[94], loop-mediated isothermal amplification (LAMP)
[95], and nucleic acid sequence-based amplification (NASBA)
[36],
[96]. Finally, the signal-detection output mode is based on classic laboratory methods, such as electrochemical luminescence and chemiluminescence. Among these methods, PCR amplification is the established gold standard for nucleic acid detection. Although these approaches are highly accurate, they have drawbacks such as expensive equipment, long reaction times, limited sensitivity, significant background noise, and complex operation. Therefore, a speedy, accurate, and simple nucleic acid detection technique is urgently needed. Many new nucleic acid detection techniques, including electrochemistry
[97], fluorescence
[98], Raman spectroscopy
[99], isothermal amplification
[95], and microfluidics
[91], have been developed in recent years. Among these, microfluidics technology, with its high throughput, fast reaction time, low cost, high sensitivity, and direct detection has received widespread attention and is widely employed in clinical diagnostics, food safety, and other fields.
Droplet-based microfluidics technology accomplishes detection by encapsulating and separating chemicals in a small volume of droplets and reacting in a very short time
[13],
[100]. Azizi et al.
[101] developed droplet microfluidic technology with LAMP to detect pathogens. This detection method is straightforward, sensitive, focused, quick, and adaptable to a range of biological measurements. Centrifugal microfluidic chips are a novel nucleic acid detection technique. The chip avoids the typical complex design of switches and valves by allowing liquid flow and mixing samples via centrifugal force. In clinical research, centrifugal microfluidic chips can be used to extract markers in blood and for DNA purification
[102]. Xiong et al.
[102] created a portable centrifugal microfluidic technology for the rapid, accurate, and simultaneous detection of seven coronaviruses. The centrifugal microfluidic system eliminates manual involvement in collection, purification, separation, amplification, and detection, decreases interference, and provides high sensitivity and specificity. Tian et al.
[103] reported an automatic centrifugal microfluidic device with good sensitivity and specificity, quick detection, and 21 reaction units. This closed automatic microfluidic system eliminates the requirement for qualified individuals to perform multiple sample detection, reduces detection complexity, and enhances diagnostic efficiency. Li et al.
[104] developed a stretch-driven microfluidic chip that performs nucleic acid detection and determines the results of samples by detecting the color of indoor liquids.
Despite its popularity, traditional microfluidic detection technology still has some problems, such as long preparation and analysis times and relatively complex operation, which limit its application. Therefore, microfluidic barcode biochips are becoming widely used in molecular diagnosis, biosensing, bioimaging, environmental monitoring, and other fields because of their high throughput, short time, reagent savings, and low cost
[49],
[56],
[105]. Yang et al.
[105] proposed a simple, fast, cost-effective, stable, and multiplex barcode detection platform compatible with mobile devices (
Fig. 8(a)). Du et al.
[106] developed a microfluidic sample preparation multimeter and test program capable of running 80 samples on a single chip, which lays a good foundation for immediate detection. Gao et al.
[56] created a microfluidic barcode chip platform for PLL local self-assembly with 60 samples loaded and 50 independent-channel fixed probes for high-throughput miRNA detection. The platform’s data analysis uses the cross-fluorescence values between two chip channels to provide a quick detection time (
Fig. 8(b) [56]). Our research group also developed a unique nanomaterial GO local-self-assembly three-layer microfluidic barcode biochip that is ultrasensitive, rapid, inexpensive, dependable, low cost, and high throughput. This biochip detects miRNA in breast cancer samples using GO to adsorb a single-chain probe, a parsing double-chain probe, and PLL to collect signals
[49],
[57], based on previous work (
Figs. 8(c) and
(d)).
3.3. Circulating tumor DNA
ctDNA refers to the tumor-derived free DNA fragments found in the blood of cancer patients. These biomarkers are released when cancer cells and tumor growth are replaced by new cells. As ctDNA is released from multiple tumor regions (both primary and clonal), it provides extensive information about tumor heterogeneity and clonal evolution, aiding in the early detection and diagnosis of cancer
[107]. Compared with traditional physical and chemical methods, blood-serum-based ctDNA analysis offers a noninvasive approach for effective cancer diagnosis, prognosis, and precision treatment
[108]. However, due to the laboratory-bound nature of ctDNA analysis, which is characterized by low throughput and reliance on large-scale equipment and specialized personnel, there is still a need for highly sensitive and selective equipment to fully realize the potential of ctDNA. Microfluidic platforms, with their advantages of high throughput, compactness, time efficiency, rapid processing, and minimal sample requirements, can perform simultaneous or continuous ctDNA testing as needed to characterize the evolving trends of ctDNA in cancer patients
[109].
To address limitations in the previous ctDNA detection methods, the latest research has identified a high-throughput plasma ctDNA analysis method based on PCR sequencing technology, significantly improving the accuracy, sensitivity, and specificity of ctDNA analysis
[109]. The study introduced superparamagnetic microbeads (SPMs) for the separation of ctDNA on a microfluidic platform. The platform is used for early cancer detection and classification, and has been validated using computer simulation techniques. The process of ctDNA separation involves the following steps: separating ctDNA from mixed samples based on their size; attaching these ctDNA fragments to SPMs (
Fig. 9(a) [109]); and, finally, collecting the attached ctDNA fragments at the microfluidic channel outlet by applying a magnetic field. Although this method is highly effective in providing high specificity and sensitivity, it is relatively complex to perform. Therefore, a more straightforward method has been proposed that using an array of magnetoresistive a sensors integrated in a portable biochip platform for magnetic labeling and detection of cell-free DNA (cfDNA) fragments. This method not only simplifies the operation but also reduces nonspecific effects, raising the detection limit to the picomolar (pM) level (
Fig. 9(b))
[110]. The advancement of ctDNA detection technology has led to the discovery of applications for microfluidic SERS chips. In particular, pumpless multichannel microfluidic chips based on SERS utilize signal-amplification strategies that not only enable the measurement of multiple biomarkers in real samples but also offer a novel strategy for detecting other types of tumor markers (
Fig. 9(c))
[111].
Although ctDNA presents numerous challenges in clinical diagnostic applications, it exhibits greater accuracy, specificity, and informativeness than protein biomarkers
[112]. More importantly, this technology has propelled the advancement of whole-genome sequencing, facilitating the conversion of sequence data into ctDNA for cancer detection
[113]. Detection based on ctDNA in serum not only offers personalized treatment options for cancer patients but also plays a crucial role in diagnosis, prognosis, and treatment decision-making
[114]. Nevertheless, there is still significant room for improvement in the clinical application of this technology.
In summary, while a number of detection methods excel in accuracy and sensitivity, they are still incapable of meeting the requirements for multiple performance metrics simultaneously. In contrast, microfluidic barcode biochips are capable of balancing multiple performance attributes, including high throughput, high sensitivity, high specificity, and high accuracy, thereby facilitating effective biomarker screening. For example, even with a low volume of samples, microfluidic barcode biochips can achieve high-throughput detection across multiple samples. This detection system significantly reduces costs while substantially improving work efficiency and cost-effectiveness. That said, to ensure the accuracy of the results, it is necessary to validate a substantial number of experimental samples. In addition, scaling up the production of these biochips poses a significant challenge for the future.
4. Microfluidic barcode biochips for single-cell analysis
Biological functions such as cell proliferation, immune responses, and tumor metastasis play a crucial role in regulating physiological processes, with interactions and communications between cells leading to collective responses
[115],
[116]. Current research has found that these intercellular interactions offer new avenues for disease diagnosis and therapeutic targets
[117],
[118]. However, the study of cell populations often obscures the unique characteristics of individual cells. Consequently, research on single cells and their subpopulations has garnered widespread attention. In particular, proteins within or associated with the cell membrane form the structural basis of the cell, facilitating functions such as transcription, translation, metabolism, growth, adhesion, and signal transduction
[119],
[120]. However, proteins secreted by immune or stromal cells—including cytokines, growth factors, and hormones—play a crucial role in cell communication, in tumor development and progression, and in regulating the functions of cells, tissues, or organ groups
[121],
[122]. Extracellular vesicles (EVs) are essential in the regulation of intercellular communication and the transmission of biological information molecules, including cytosolic proteins, lipids, and nucleic acids
[123],
[124]. Traditional methods of EV surface labeling maps based on population measurements mask the intercellular heterogeneity of EV secretion and phenotype. Thus, in order to assess cell-to-cell connections and biological functions, researchers must simultaneously and directly investigate a variety of signaling behaviors at the single-cell level.
Traditionally, microscopes, flow cytometry, and mass spectrometry have been frequently utilized in single-cell research
[125]. Because these methods often have limitations such as low throughput, strong background signals, the inability to measure cellular metabolites, and the need for bulky and expensive instruments
[26], the field of single-cell omics requires innovative techniques to meet current needs. Therefore, emerging single-cell omics analysis methods based on microfluidic chips have received extensive attention, as they provide high-throughput, high-sensitivity, and low-cost approaches to evaluate cell omics
[26],
[59]—including single-cell analyses of the genome, epigenome, transcriptome, proteome, and metabolome, as well as their interactions
[126],
[127],
[128].
4.1. Detection of real-time secreted proteins from single cells
Proteomic research plays a vital role in assessing biological mechanisms, disease-related biomarkers, therapeutic targets, and other protein complexes
[119]. In particular, microfluidic barcode biochip technology has been widely used in single-cell proteomics. Current techniques suffer from limitations such as low throughput and low sensitivity, restricting the detection to certain proteins
[129],
[130]. Thus, microfluidic barcode biochip platforms—which can overcome these drawbacks while achieving highly sensitive and high-throughput single-cell protein detection—have garnered widespread attention
[46],
[59].
Depending on the specific objectives of the experiment, microfluidic barcode biochips can be designed with various antibody strips for immunological quantification studies of single-cell proteins
[15],
[27],
[28],
[59],
[131],
[132]. Ma et al.
[132] proposed a single-cell barcode biochip specifically for measuring a range of protein factors secreted by lymphocytes, thereby uncovering the heterogeneity and functional diversity of proteins (
Fig. 10(a)). As shown in
Fig. 10(b), Elitas et al.
[131] designed a biochip capable of the simultaneous determination of 16 secreted proteins of a single cell. This single-cell chip features 5000+ microchambers that randomly load single cells. The changes in the functional phenotypes of tumor cells were revealed by the determination of 16 proteins. Moreover, the original technology has been improved. Lu et al.
[25] reported a microfluidic barcode biochip for single-cell analysis. By integrating spectral encoding with the spatial encoding of various fluorescent colors, these researchers achieved the detection of 42 proteins the highest throughput currently available in single-cell protein analysis, as depicted in
Fig. 10(c) [25]. The platform has been used in cell signaling, immune response, cell therapy, and other aspects
[25],
[26],
[131]. Given the low single-cell utilization and reagent waste associated with the aforementioned methods, Khajvand et al.
[133] introduced an integrated droplet microfluidic device (
Fig. 10(d)). This device, linked with antibody barcoding, facilitates the efficient multiplex analysis of proteins secreted by individual cells and has confirmed the heterogeneity of protein secretion at the single-cell level.
The majority of the preceding work is based on PLL functionalized glass plates for antibody mapping, and the format is relatively simple. Based on this, as depicted in
Fig. 10(e), our research group
[27] developed a spatially coded antibody barcode chip based on self-assembled GOQDs for single-cell secreted protein analysis. The material not only provides a suitable environment for cell growth but also allows for flexible assembly on the chip, enabling the precise measurement of proteins secreted by individual cells.
The integration of machine learning has further facilitated the classification and sub-group identification of single cells. Building on the foundation of single-cell cytokine measurements, our group delved into the impact of mechanical forces on the release of cytokines by human neural stem cells in the early stages of differentiation into astrocytes, as shown in
Fig. 10(f) [59]. Single-cell analysis confirmed that mechanical stress leads to the downregulation of certain cytokines, offering new approaches and unique insights into the immune-regulatory mechanisms of the central nervous system
[59].
4.2. Detection of exosomes from single cells
Fig. 11(a) [134] demonstrates a novel microfluidic barcode biochip platform with multipleparallel microchannels for the high-throughput visualization analysis of single-cell EVs. The proposed platform effectively addresses urgent challenges in the technology of tumor-cell communication mediated by EVs. The results indicated that the decrease in EV phenotypes (e.g., CD63
+ EVs) is related to the invasive potential of both the cell lines and primary cells. Furthermore, as shown in
Fig. 11(b) [135], the researchers reported the use of a home-use scanner to realize high-throughput single-cell EV secretion analysis without cell counting. The above methods are based on research involving PLL substrate barcoding, while our team analyzed the single-cell exosomes of ovarian cancer tumor cells and epithelial cells based on a GOQD self-assembled substrate, demonstrating cell–cell functional heterogeneity, as shown in
Fig. 11(c) [28]. The proposed analytical platform can simultaneously analyze 20 different types of secretion profile biomarkers. It also allows for customization of the number and size of microchannels within the single-cell capture platform, thereby increasing the throughput of single-cell analysis. This strategy holds promise for application in other types of cancers as well, enabling the in-depth analysis of highly complex tumor microenvironments and potentially unraveling the multidimensional spectrum of heterogeneity within tumors at a higher resolution.
4.3. Analysis of cell interactions
Single-cell responses and cellular heterogeneity are outcomes regulated by both intrinsic and environmental factors. Single-cell analysis primarily encompasses aspects such as the detection of single-cell characteristics, the assessment of genetic information, and cell phenotype correlations. At present, with the use of microfluidic technology, the rapid and high-throughput screening of single cells can be achieved, enabling in-depth research in fields such as single-cell genetics, proteomics, genomics, transcriptomics, cellular heterogeneity, and metabolic profiling
[136],
[137]. Among these areas, studies have revealed that paracrine signaling transduction between glioblastoma cells and macrophages can lead to changes in the functional phenotypes of tumor cells. Moreover, correlations among proteins have unveiled critical signaling transduction nodes in the alteration of tumor–macrophage communication (
Fig. 12(a))
[131]. As shown in
Fig. 12(b) [128], the use of single-cell barcode chips has been instrumental in studying the role of paracrine signal transduction in the toll-like receptor 4 (TLR4) response to lipopolysaccharide (LPS) stimulation. The results indicate that, at the population level, paracrine signaling amplifies the secretion of cytokine subgroups, including IL-6 and IL-8, upon LPS stimulation, while inhibiting the secretion of TNF-α and other cytokines in macrophages
[138].
Despite the ongoing development of single-cell analysis technology, limitations remain in the engineering design and integration of microenvironment platforms for single-cell analysis. To address these limitations, a single-cell cytokine-secretion analysis platform has been proposed. This platform was employed to explore the immune response of macrophages stimulated by TLR ligands in the tumor microenvironment and revealed the differential regulation of cytokine secretion, including a decrease in the antitumor cytokine TNF-α and an increase in IL-6, as shown in
Fig. 12(c) [139]. To characterize a substantial volume of single-cell dynamics and secretion information, Chen et al.
[140] developed a method to simultaneously measure ten secreted proteins in 5000 single cells, as depicted in
Fig. 12(d). Their results demonstrated that macrophage responses to TLR4 stimulation primarily exist in two distinct states, which correlate with the baseline functional states of single cells. This method is primarily used for translation and validation procedures, revealing heterogeneity among cells within phenotypically homogeneous cell populations. Dual-cell secretion assays have also been found to be valuable for understanding how intercellular communication regulates cell motility behaviors and for identifying key proteins that maintain the free energy gradient driving intercellular movement
[141].
Regarding traditional flow cytometry analysis methods, it is challenging to conduct tests of multiple parameters while maintaining cell viability. However, under conditions in which cells are well-maintained, microfluidic barcode chips in conjunction with single-cell chips can perform the multi-parametric detection of secreted proteins/exosomes from a single cell. This capability can enhance our understanding of the heterogeneity among cells and the underlying mechanisms in physiological and pathological processes.
In the research process, microfluidic barcode biochips still present significant challenges in terms of sensitivity, cell throughput, multifunctionality, and repeatability. Nevertheless, combining microfluidic barcode biochips with single-cell analysis in proteomics, genomics, and metabolomics offers significant opportunities for a comprehensive understanding of single-cell characteristics, which is important for advancing biological research and precision medicine.
5. Conclusions and future perspectives
Microfluidic barcode biochip technology has become a popular tool for the high-throughput detection of biomolecules and single cells analysis in medical and bioanalytical science research. This article provided a comprehensive review of microfluidic barcode technology. First, it introduced a brief classification of microfluidic technology; then, it presented the advantages and disadvantages of microfluidic barcode technology and summarized other microfluidic approaches. Next, the article summarized the substrate materials used for microfluidic barcode chips. Finally, it introduced the application of microfluidic barcode biochips for the analysis of biomolecules (proteins and nucleic acids) and single cells (secreted proteins and exosomes), laying a solid foundation for the application of bioanalysis. Compared with microfluidic lattice technology, microfluidic barcode technology overcomes the shortcomings of low flux, low detection reproducibility, difficult preparation, low detection sensitivity, a limited range of dynamic detection, the trailing effect, and uneven dot arrays, which lead to the cross-contamination of analytes and incorrect signal reading. In addition, it has the advantages of parallel multi-channels, high throughput, low sample consumption, speed, and low cost. Furthermore, the integration of nanomaterials or polymer materials with microfluidic technology is accelerating the development and application of microfluidic barcode technology. These materials have the advantages of a large surface area, FRET, small size, and easy functionalization; they reduce nonspecific adsorption, improve the signal-to-noise ratio, and lower the detection limits. Microfluidic barcode biochip technology benefits from these advantages and can use the smallest number of samples for the high-throughput detection of biomarkers with higher sensitivity, high specificity, a LOD, a short detection time, and so forth. Finally, although microfluidic barcode technology has found extensive applications in biomedical research and other fields, problems remain to be solved, and some directions require further study.
There is still a long way to go before the innovation of microfluidic barcode technology can be commercialized and popularized in clinical diagnosis. First, this study revealed that the available materials are rather simple; each material has its own scope of use, and optimal capture efficiency may be achieved under particular conditions. Therefore, the urgent need to develop new and reproducible materials with good properties, stability, and clinical applications remains a major challenge. Second, microfluidic barcode technology can be combined with other technologies to form a variety of analysis methods to achieve high-throughput analysis. The different methods support each other, markedly enhancing the efficiency, sensitivity, and specificity of the analysis. The mechanisms of disease development in organisms can be studied through biomolecules and single cells, and some biological information can be screened. Finally, microfluidic barcode biochip technology has been improved to realize the analysis of biological samples in a more complex environment and through a fully automated detection process. Great efforts are underway to push this technology from the laboratory to the market for large-scale applications.
In short, the development of microfluidic barcode biochip technology for material self-assembly, chip manufacturing, automation, portability, and versatility is far from complete. Nevertheless, the rapid and high-quality development of this technology is accelerating its industrialization; moreover, its integration with other technologies is greatly promoting the development of real-time detection and high-efficiency detection technology, as well as its large-scale use in the clinical diagnosis of diseases.
Acknowledgments
This work was supported by the National Key Research and Development Plan of China (2023YFB3210400), the Natural Science Innovation Group Foundation of China (T2321004), the National Natural Science Foundation of China (62174101), Shandong University Integrated Research and Cultivation Project (2022JC001), Key Research and Development Plan of Shandong Province (Major Science and Technology Innovation Project; 2022CXGC020501).
Compliance with ethics guidelines
Jiaoyan Qiu, Yanbo Liang, Chao Wang, Yang Yu, Yu Zhang, Hong Liu, and Lin Han declare that they have no conflict of interest or financial conflicts to disclose.
PDF
(7211KB)
