Droplet microfluidics for high-sensitivity and high-throughput detection and screening of disease biomarkers

Abstract

Biomarkers are nucleic acids, proteins, single cells, or small molecules in human tissues or biological fluids whose reliable detection can be used to confirm or predict disease and disease states. Sensitive detection of biomarkers is therefore critical in a variety of applications including disease diagnostics, therapeutics, and drug screening. Unfortunately for many diseases, low abundance of biomarkers in human samples and low sample volumes render standard benchtop platforms like 96-well plates ineffective for reliable detection and screening. Discretization of bulk samples into a large number of small volumes (fL-nL) via droplet microfluidic technology offers a promising solution for high-sensitivity and high-throughput detection and screening of biomarkers. Several microfluidic strategies exist for high-throughput biomarker digitization into droplets, and these strategies have been utilized by numerous droplet platforms for nucleic acid, protein, and single-cell detection and screening. While the potential of droplet-based platforms has led to burgeoning interest in droplets, seamless integration of sample preparation technologies and automation of platforms from biological sample to answer remain critical components that can render these platforms useful in the clinical setting in the near future. This article is categorized under: Diagnostic Tools > Biosensing Diagnostic Tools > Diagnostic Nanodevices Therapeutic Approaches and Drug Discovery > Emerging Technologies Therapeutic Approaches and Drug Discovery > Nanomedicine for Infectious Disease.

Keywords: biomarker; clinical; diagnostic; droplet; microfluidics.

Figure 1. Discretization facilitates high-throughput, high-sensitivity, rapid, and quantitative analysis for rare biomarkers in a sample

For effective detection of a biomarker of interest, conventional bulk analysis (left) is restricted to a few replicate reactions between microliters and milliliters each, wherein signal can be drowned out by high concentration of sample background, limiting overall sensitivity and speed. In contrast, digitization of sample (right) into femtoliter to nanoliter volume droplets facilitates background reduction and subsequently greater sensitivity and speed. Furthermore, encapsulation of single targets into these droplets allow for absolute quantification of rare targets.

Figure 2. Methods for high-throughput droplet generation

(A) Valve-actuation involves pneumatically pulsing PDMS “Quake” valves to generate droplets. They offer greater control of droplet size, content, and motion, but are limited in generation speed and throughput (Reprinted from (Guo et al., 2010) with the permission of AIP Publishing). (B) Cross-flow devices feature a “T-junction” where an aqueous stream meets a flowing continuous phase to generate droplets. (Reprinted with permission from (Zagnoni, Anderson, & Cooper, 2010). Copyright 2010 American Chemical Society) (C) Flow-focusing devices are most commonly used for droplet generation and feature a junction where a flowing aqueous stream is sheared by two perpendicularly intersecting streams of the continuous phase (Hindson et al., 2011). (D) Step-emulsification devices feature a 3-D step where an aqueous stream enters into a much larger oil reservoir, creating a droplet. (Reprinted from (R. Dangla et al., 2013). Copyright 2012 National Academy of Sciences). Step-emulsifiers may be parallelized to generate droplets at very high speeds, but are more difficult to control in content and movement.

Figure 3. Fragmented and integrated droplet platforms

(A) Most droplet platforms utilize a separate device for (i) droplet generation, (ii) a thermocycler or incubator for droplet incubation, and (iii) another device for droplet detection. (Adapted from (Pekin et al., 2011) with permission of The Royal Society of Chemistry) (B) Imaging-based integrated platforms use (i) a single device for droplet generation and incubation. Droplets are incubated in an on-chip droplet reservoir that rests on a heat block, and (ii) real-time imaging or microscopy may be used to observe these droplets. (Reproduced from (Hatch, Fisher, Tovar, et al., 2011) with permission of The Royal Society of Chemistry). (C) Integrated platforms that facilitate continuous-flow detection of droplets typically contain (i) delay lines for on-chip droplet incubation. Often, these delay lines may contain constrictions to reduce variability in droplet speed through the incubation region. (Reproduced from (Frenz et al., 2009) with permission of The Royal Society of Chemistry). These constrictions also serve as points for droplet detection where individual droplets can be sequentially measured. (ii) Integrated continuous flow platforms may feature a Peltier heater to heat the incubation line and a laser excitation source and APD detector for continuous measurements of droplet fluorescence. (Reprinted from Biosensors and Bioelectronics, 97, Kaushik and Hsieh et al., Accelerating bacterial growth detection and antimicrobial susceptibility assessment in integrated picoliter droplet platform, 260-266, Copyright 2017, with permission from Elsevier.) (Kaushik et al., 2017).

Figure 4. Droplet-based detection of nucleic acid biomarkers

(A) Droplet-based digital PCR was used to screen for mutations in the KRAS oncogene. (Reproduced from (Pekin et al., 2011) with permission of The Royal Society of Chemistry). TaqMan probes specific for the wild-type and mutant genes were encapsulated in droplets that contain at most one haploid genome. The emulsion was then thermocycled off-chip, and reinjected for fluorescence detection. By optically coding droplet groups, parallel analysis of six mutations in KRAS were detected simultaneously using this platform. (B) An integrated platform for viral RNA detection using RT-PCR was developed, where 70 pL droplets containing MS2 virions and RT-PCR reagents were immobilized in microfluidic channels. An integrated thermal cycler provided temperatures necessary for reverse transcription and PCR, and a CCD camera was used for fluorescence detection. (Reprinted with permission from (Beer et al., 2008). Copyright 2007 American Chemical Society.) (C) Loop-mediated isothermal amplification was demonstrated for detection and quantification of N. gonorrhoeae gDNA down to 600 copies per μL. Critically, the assay was conducted in continuous-flow in an integrated platform that facilitated digitization of targets, on-chip incubation, and detection. (Reproduced from (Rane, TD; Chen, L; Zec, HC; Wang, 2014) with permission of The Royal Society of Chemistry).

Figure 5. Amplification-free quantification of RNA using droplet digital enzyme-linked oligonucleotide hybridization assay (ELOHA) (Reprinted from (Guan et al., 2015) under the terms of the Creative Commons Attribution 4.0 License)

(A) A sandwiched complex consisting of capture oligo-coated magnetic beads hybridized to a single molecule of target RNA which is then hybridized to an enzyme-linked detection oligo is co-flowed into droplets along with corresponding enzyme substrate. Following in-line incubation, droplets containing single RNA exhibit strong fluorescence, whereas droplets without the sandwiched complex exhibit weak fluorescence. (B) By interrogating ~10⁶ droplets, the percentage of highly fluorescent droplets can be used to quantify input concentration of 16S rRNA from N. gonorrhoeae down to 600 molecule copies in 100 μL of sample.

Figure 6. Droplet based detection of protein biomarkers

(A) A bead based sandwich assay, similar to ELISA, was developed for quantifying prostate specific antigen, a prostate cancer biomarker, in buffer (Shim et al., 2013). The target protein is tagged with a reporter enzyme and hybridized to the bead. Beads and substrate are co-flown to generate femtoliter droplets. After 10 min. of incubation, this platform is able to detect protein biomarker concentrations as low as 46 fM. Reprinted with permission from (Rane, TD; Chen, L; Zec, HC; Wang, 2014). Copyright 2013 American Chemical Society.) (B) Matrix metalloproteinase (MMP) biomarkers secreted into the peritoneal fluid of clinical endometriosis tissue sample was combined with a droplet library of 4 FRET based protease substrates with and without specific inhibitors and incubated on chip. Following at least 3 h of fluorescence monitoring, proteolytic activity matrix analysis (PrAMA) was conducted to assay protease activity within droplets. (Reprinted with permission from (Chen et al., 2013). Copyright 2013 American Chemical Society.) (C) In order to scale up PrAMA analysis to a larger set of MMPs and substrates, a high-throughput combinatorial valve-based platform was developed (Tushar D. Rane et al., 2015). 650 unique combinations of MMPs and substrates were screened in continuous flow following 12 min. incubation using this approach.

Figure 7. Droplet based detection of single-cell biomarkers

(A) An integrated amplification-free platform for genetic detection of pathogens was developed. Single cells of E. coli were co-encapsulated with peptide nucleic acid beacons, complimentary to a specific region within the 16S rRNA of the pathogen. Following on-chip thermal lysis, rRNA release, and probe hybridization, droplet fluorescence was detected and used to quantify pathogen load within a sample. (Reproduced from (Tushar D. Rane et al., 2012) with permission of The Royal Society of Chemistry). (B) An enzyme-tagged antibody was used to detect the typically low-abundance cell-surface protein biomarker CCR5 (a co-receptor in HIV-1 infection) in U937 cells. (Reprinted from (Joensson et al., 2009) with permission from John Wiley and Sons).

Figure 8. Droplet based bacterial growth quantification and antibiotic susceptibility testing (AST)

(A) A ddCFU platform was developed to rapidly count viable bacteria within a sample. Critically, this platform followed a fragmented workflow that necessitated separate modules for (i) droplet generation and encapsulation of single bacteria, (ii) off-chip bacterial incubation, and (iii) droplet detection. (Adapted from (O. Scheler et al., 2017) with permission of The Royal Society of Chemistry.). (B) A similar workflow was used to develop a droplet based AST assay. After co-encapsulation of single cells of S. aureus with blood plasma and ampicillin into 4 nL droplets, the pathogen’s susceptibility/resistance to the drug was determined in 7.5 h. (Reproduced from (Boedicker et al., 2008) with permission of The Royal Society of Chemistry.). (C) An integrated one-step platform was developed for bacterial growth analysis and antibiotic susceptibility testing (Kaushik et al., 2017). Critically, the use of much smaller 20 pL droplets in this platform enabled detection of single-cell E. coli growth and its susceptibility/resistance to gentamicin in as little as one hour, equivalent to 2-3 replications of the bacterial cell. (Reprinted from Biosensors and Bioelectronics, 97, Kaushik and Hsieh et al., Accelerating bacterial growth detection and antimicrobial susceptibility assessment in integrated picoliter droplet platform, 260-266, Copyright 2017, with permission from Elsevier).

Figure 9. Increasing multiplexing and screening capacity of droplet platforms

(A) A droplet-on-demand system was developed to interface droplet chips with off-chip reservoirs in order to construct libraries of droplets with controllable input concentrations of sample/reagents and controllable volumes at 30 Hz. (Reproduced from (Churski et al., 2010) with permission of The Royal Society of Chemistry.) (B) A device was developed to facilitate the transition of nanoliter plugs that were constructed with controlled combinations of sample/reagents, into picoliter droplets without compromising droplet stability and uniformity. (Copyright 2017 IEEE. Reprinted, with permission, from (P. Zhang et al., 2017)) (C) A serial sample loading (SSL) system was developed to interface droplet platforms with conventional 96-well plates to facilitate automated sample and reagent loading. (Reproduced from (T. D. Rane et al., 2012) with permission of Sage Publishing.)