Stephen Quake, a bioengineer and applied physicist at Stanford University, has always been interested in the interface between physics and biology. During his work at the California Institute of Technology in the late 1990s, he focused on DNA research. However, he was frustrated by the time-consuming processes at the benchtop. Interested in new automated technologies, he wondered how he could escape the tyranny of pipetting.
He wondered if the answer lay in soft lithography, a method developed by George Whitesides’ team at Harvard University in the late 1990s that used polydimethylsiloxane (PDMS).1 This material had quickly found its way into microfluidic devices, tiny chips that were gaining popularity in life sciences, due to its mechanical flexibility, low cost, and ease of mass production. Most importantly, the high transparency of PDMS meant that it was suitable for optical microscopy, making these microfluidic devices particularly attractive for biological applications.
Due to the benefits of PDMS, Quake sought to create valves and pumps in the same material to enable researchers to conduct high degrees of automation within a chip. As an extension of Whitesides’ technique, Quake and his colleagues devised microfabricated mechanical valves made of PDMS and dubbed them Quake valves.2 This design integrated these microvalves onto a bilayer PDMS microfluidic chip. Liquid flowed through the bottom channel, while the upper channel sat perpendicular to it. When activated under pressure, the upper channel pinched off the flow of the channel below it (much like stepping on a garden hose). Relieving the pressure caused the lower channel to reopen.
You have good technology when the applications emerge as we learn more about the world around us. It's not a one-trick pony, so to speak.
–Aaron Streets, University of California, Berkeley
These microvalves offered an improvement for conventional methods: They provided a powerful platform for automation. For instance, Quake integrated these pneumatic microvalves onto microfluidic chips to easily separate and sort DNA molecules instead of relying on DNA gel electrophoresis. With the integration of this technology, Quake and his team also streamlined cell sorting and performed polymerase chain reaction on microfluidic chips.3
Microfluidic systems greatly aided researchers in exploring applications from structural biology, such as protein crystallization, lab-on-a-chip, and beyond.4 While this technology powered many scientific discoveries, it completely transformed the field of single-cell biology.
Microfluidics, Single Cells, and DamID
With single-cell genomics techniques introduced in 2009, researchers could amplify transcriptomes from single cells but lacked the automation to capture and analyze many cells. “The big roadblock was how to take these techniques beyond bespoke experiments with a handful of cells and do them at scale. Microfluidic plumbing enabled us to do just that,” explained Quake.
The Human Genome Project benefitted from these powerful microfluidic platforms. Ongoing during the late 1990s and early 2000s, the project presented researchers with a daunting challenge: sequence millions of cells. Microfluidics technology offered a solution to this labor intensive and time-consuming task. The automation of microfluidics served as a critical factor in helping researchers collect genomic data and made DNA sequencing faster and simpler to perform for a fraction of the price.
Subfields such as microvalve, microwell, and microdroplet platforms quickly emerged and greatly increased the scalability and throughput of single-cell sequencing techniques.5-7 Now, these different technologies aid in mapping the Human Cell Atlas, which aims to identify and catalogue all human cell types and cell states in health and disease.
“There are trillions of cells, and the more we explore cellular diversity at the single cell level, the more we learn about this continuum and complexity of cell-state,” said Aaron Streets, a bioengineer at the University of California, Berkeley. “It would never be possible to imagine capturing enough single cells to create a Human Cell Atlas without microfluidic technology.”
Streets’ foray into using microfluidics for single-cell genomic analysis began in the mid-2010s. While sequencing technology provided valuable information, Streets wanted to obtain a more complete picture while characterizing cells. So, he coupled optical and genomic measurements to learn how genomic information relates to what a cell looks like.
Streets developed a new technique called μDamID.8 Adapted from DNA adenine methyltransferase identification (DamID), this method enabled single cell sorting on a microfluidic chip and collection of genomic material for sequencing.
μDamID paired two methods, one for sequencing data and the other for spatial imaging data. First, the researchers tagged cells with m6A-Tracer, a fusion protein containing green fluorescent protein and a domain bound to these methylated sites. Then they individually isolated, imaged, and sorted the cells on the microfluidic platform. To see the spatial location of these interactions in the cell nucleus, Streets used fluorescence microscopy.
Following imaging, the researchers performed DamID, which relied on identifying a chemical recording of protein-DNA interactions on the DNA. It deposited methyl groups along the DNA and functioned as a chemical signal trail. They selectively amplified this signal for analysis to map out the sequence positions of protein-DNA interactions. DNA collection occurred on the microfluidic chip and the researchers performed subsequent sequencing off-chip. This technology improved researchers’ abilities to create these protein-DNA binding maps for studying how DNA is regulated in the nucleus, with the major advantage of linking imaging and sequencing data for individual cells.
Since μDamID focused on short-read sequencing, Streets developed another method called DiMeLo-seq to assess long-read sequencing data.9 His team continues to develop tools for making new multimodal measurements that will give researchers insights into a biological system that wasn’t previously possible with traditional methods.
“You have good technology when the applications emerge as we learn more about the world around us. It's not a one-trick pony, so to speak,” remarked Streets. “That’s why [microfluidics] has been such an exciting field.”
Microfluidic Lab-on-a-Chip Devices aid Diagnostics
While microfluidic chips offered improvements to cell culture practices at the lab bench, the medical field wasn’t far behind in their adoption. Microfluidic devices soon found their way in pathophysiology, drug discovery and development, and point-of-care diagnostics.10-12
Joel Voldman, an electrical engineer at the Massachusetts Institute of Technology, was fascinated with the electrical properties of cells. “These properties are informative because if you’re trying to discern a pathological state versus a nonpathological state, there may be a difference in electrical properties,” explained Voldman.
With a goal of applying his knowledge in diagnostics, he set off to design a way to measure those differences. In collaboration with physician researchers who wanted to shrink the components of the lab onto a chip to use less blood and receive faster results, they developed a microfluidic lab-on-a-chip that quantified the number of circulating activated leukocytes for rapidly monitoring sepsis.13
To characterize the intrinsic electrical properties of leukocytes, the team passed a suspension of cells obtained from a sepsis model in mice across a microfluidic chip. These cells traveled through a channel with electrodes placed above and below the cells, guiding them across a conductivity gradient. The electrodes applied a dielectrophoretic force that kept cells away from walls as they traversed across the channel. Based on their electrical properties, cells moved to different isodielectric positions (IDP). These IDP resulted in a distribution of cells and with imaging software, researchers visualized the cellular distribution of activated and nonactivated leukocytes. This work correlated with traditional flow cytometry analysis, and Voldman later developed an assessment platform as an extension of this work that analyzed healthy human volunteers and volunteers with sepsis.14
Confined Chambers for Wiggly Worms
The influence of microfluidic technology extended beyond cells and molecules when worm researchers realized that they had found the perfect chambers for live worm studies.
It's a great tool, and we're going to see some good biology coming out of it.”
–Hang Lu, Georgia Institute of Technology
Hang Lu, a chemical and biomolecular engineer at the Georgia Institute of Technology, was interested in studying the sensory motor behavior of Caenorhabditis elegans. She often tested how different stimulus conditions such as food concentrations and pheromone exposures affected the worms.
“The tricky part is that it’s often difficult to try to get precision stimulus with a small organism because it wiggles a lot,” explained Lu. Because of the worms’ small size and tendency to escape, Lu designed a microfluidic system called worm-on-a-chip (or worm chip) for their worm experiments.
In one study, Lu investigated how the microenvironment affected the long-term development of C. elegans.15 Under normal conditions, C. elegans larvae develop into reproductive adults. However, under unfavorable conditions, larvae develop into an alternative growth stage called dauer. These dauer exhibit stagnated growth and slower movement.
By utilizing four growth chambers, the researchers could easily accommodate up to 50 animals from the larval to reproductive adult stages. This platform allowed them to precisely control multiple stimuli such as food delivery, pheromone delivery, and temperature. “The microfluidic device is basically agnostic to whatever observation that you're trying to make—or at least the way that we designed it—and we try to make it compatible with all sorts of microscopy modalities,” commented Lu.
By pumping the chambers with varying concentrations of food and pheromones, the researchers observed the stages of development over a period of 150 hours using a micrograph. Micrograph images of the growth chambers revealed that C. elegans larvae developed into either dauer or reproductive adults depending on their food and pheromone exposure. Low food concentrations and prolonged pheromone exposures promoted developing larvae to enter the dauer stage.
Overall, the ease of integrating different modalities with samples of drastically different sizes contributed to the progress of microfluidic technology. “It's a great tool, and we're going to see some good biology coming out of it,” commented Lu.
3D Printed Microfluidic Devices
As new functionalities of microfluidics expanded, so did the need for new materials. When Whitesides developed PDMS, he inspired others to jump onto these techniques. Albert Folch, a bioengineer at the University of Washington, recalled how its development and his own use of PDMS spurred his interest in the properties of microfluidic materials for applications in cell culture and cancer diagnostics.
Each material presented various properties to consider: biocompatibility, transparency, manufacturability, and elasticity. In 2013, Folch serendipitously received a brochure advertising a three-dimensional (3D) printed microfluidic device. It opened his eyes to the digital manufacturing of microfluidic devices. “People have been digitally manufacturing a lot of other things in the meantime in these last two decades, from cars to houses to everything, but they have not digitally manufactured microfluidic devices consistently in a systematic way," explained Folch.
3D printing microfluidic components appealed to Folch for its potential for rapid commercialization. Another draw was its ability to easily develop microfluidic chips with complex designs. Folch’s team set out to 3D print microfluidic devices using various resins that boast similar advantages to PDMS.16-17 They wanted these materials to have properties such as transparency and biocompatibility. However, 3D printing resins are not as mature in terms of biocompatibility as PDMS technology. It will take time to fully optimize these materials, but Folch is optimistic about 3D printing’s growth in the field of microfluidics.
“Everything tends towards using the most convenient and most cost-effective solution, which is always 3D printing. It’s just a matter of time before people find a solution with a 3D printable material that is biocompatible,” said Folch.
For Folch, 3D printing allows for more complex designs and eases the transition of prototypes to commercialization. Like the devices themselves, he envisions a more streamlined way of manufacturing and is excited to see the efforts of the upcoming 3D printing revolution in microfluidics.
“It's been one of the great pleasures of my career to see this work go from technology in the academic literature to work that's being used in thousands of labs around the world and [to see] robust commercial development changing the course of science. That's very rewarding for us in the field,” remarked Quake.
- Xia Y, Whitesides GM. Soft lithography. Annu Rev Mater Sci. 1998;28:153-184.
- Unger MA, et al. Monolithic microfabricated valves and pumps by multilayer soft lithography. Science. 2000;288(5463):113-116.
- Liu J, et al. A nanoliter rotary device for polymerase chain reaction. Electrophoresis. 2002;23(10):1531-1536.
- Hansen CL, et al. A microfluidic device for kinetic optimization of protein crystallization and in situ structure determination. J. Am. Chem. Soc. 2006;128(10):3142-3143.
- Oh KW, Anh CH. A review of microvalves. J. Micromech. Microeng. 2006;16:R13-R39.
- Yuan J, Sims P. An automated microwell platform for large-scale single cell RNA-Seq. Sci Rep. 2016;6:33883.
- Macosko EZ, et al. Highly parallel genome-wide expression profiling of individual cells using nanoliter droplets. Cell. 2015;161(5):1202-1214.
- Altemose N, et al. μDamID: A microfluidic approach for joint imaging and sequencing of protein-DNA interactions in single cells. Cell Syst. 2020;11(4):354-366.
- Altemose N, et al.DiMeLo-seq: a long-read, single-molecule method for mapping protein–DNA interactions genome wide. Nat Methods. 2022;19:711-723.
- Huh D, et al. Reconstituting organ-level lung functions on a chip. Science. 2010;328(5986):1662-1668.
- Dittrich P, Manz A. Lab-on-a-chip: microfluidics in drug discovery. Nat Rev Drug Discov. 2006;5:210-218.
- Martinez AW, et al. Diagnostics for the developing world: microfluidic paper-based analytical devices. Anal. Chem. 2010;82(1):3-10.
- Prieto JL, et al. Monitoring sepsis using electrical cell profiling. Lab Chip. 2016;16(22): 4333-4340.
- Jeon H, et al. Fully automated, sample-to-answer leukocyte functional assessment platform for continuous sepsis monitoring via microliters of blood. ACS Sens. 2021;6(7):2747-2756.
- Zhuo W, et al. Microfluidic platform with spatiotemporally controlled micro-environment for studying long-term C. elegans developmental arrests. Lab Chip. 2017;17(10):1826-1833.
- Urrios A, et al. 3D-printing of transparent bio-microfluidic devices in PEG-DA.Lab Chip. 2016;16(12):2287-2294.
- Bhattacharjee N, et al. Desktop-stereolithography 3D-printing of a poly(dimethylsiloxane)-based material with sylgard-184 properties. Adv Mater. 2018;30(22):e1800001.