For a technical summary of the droplet generator designs scroll down to the bottom of the page
You may purchase any of these chips at a reasonable price without paying for any mold or setup fees.
Cells, by nature, are immensely heterogeneous. Tissues can contain many different types of cells and cells of each type can be at different stages. This drastic level of diversity among cells, narrows our knowledge about biological systems. Studying biological systems in single cell resolution, however, can be a great leap towards better understanding the functionality of different cell types at various stages in tissues. A major indicator of cell functionality is its transcriptional program. Although there are methods for sequencing m-RNA, they have not been much advantageous since they are limited by time and cost of analysis. In 2015, McCaroll’s lab introduced drop-seq, a droplet based microfluidic device, for analyzing RNA expression of individual cells. Since then, hundreds of labs have employed this technique for mRNA sequencing and profiling of thousands of single cells.
Their method hinges on encapsulating single cells, cell lysis reagent, and barcoded micro particles in nanoliter droplets. The barcoded micro particles were used as a molecular memory to identify the cell of origin of each mRNA. Upon droplet formation, the cells lyse and their mRNA content hybridize to the primers on the surface of the micro particle and form STAMPS (Single-cell Transcriptomes Attached to Micro Particles). Next, the droplets were demulsified and the mRNA-bound micro particles were collected. The STAMPS were then amplified by PCR reaction followed by reverse transcription of the mRNA pool. Finally, the resulting molecules were sequenced using a high-capacity parallel sequencing method.
Single-cell barcoding and sequencing
Single-cell RNA-sequencing is reported by researchgate as the top trend in biotechnology in 2017. This could stem from the advent of a new generation of methods for genome sequencing. Conventionally, the scientists used to investigate a few thousand cells to find the average value of a characteristic. However, novel technologies such as droplet microfluidics have enabled researchers to take one step further and capture cell heterogeneity in single-cell resolution.
In this paper, Zilionis et al. provided a platform using inDrop chips for barcoding mRNA by a reverse transcription (RT) reaction. Their method consists of following steps:
First, they used a flow focusing chip (shown below) to generate hydrogel beads. Here, Acrylamide primer is the dispersed phase (Qd= 900 µL/hr) while Oil+TEMED serves as the continuous phase (Qd= 1800 µL/hr). The beads were then collected and incubated at 65oC to polymerize the beads. The beads were then attached to a photocleavable spacer, and subsequently barcodes using a split and pool method. The resultant then would be a solid hydrogel conjugated to a barcode chain via a photocleavable spacer (BHB–> Barcoded Hydrogel Bead).
In the next step, they run the BHB solution along with cell solution and lysis/RT reagents through the pictured device. As soon as the droplets are formed in this device, the lysis reagent starts lysing the cells. Afterward, the droplets are collected and exposed to UV light to break the connection between the BHBs and the barcode. The lysate then reacts with the barcode inside the droplets. Next, the droplets are broken to analyze generated cDNAs using a method similar to Hashimony technique.
In the past, studying expression from cells of complex tissues, such as neurons in the brain, at single-cell level has been challenging for researchers. Isolating the cells in such tissues may result in altering the RNA content. In addition, clinical samples and archived tissues could not be processed. This led scientists to develop methods, such as sNuc-seq and Div-seq, for single nucleus RNA sequencing instead of single cells. However, these techniques relied on 96-484 well plates for operating thus were limited in throughput. Here, Habib et al., inspired by Drop-seq device, introduced a microfluidic chip called DroNc-seq to address this issue. They combined DroNc-seq with a single nucleus sequencing method (sNuc-seq) to profile nuclei at high throughput.
Nucleus accommodates lower RNA amount compared to cells. Therefore, they decreased the depth of the Drop-seq device to make smaller droplets. DroNc-seq was benchmarked against Drop-seq and was shown that the number of transcripts vs. sorted nuclei were considerably higher for DroNc-seq. Their device. Their technique was further applied to the nuclei from a mouse cell line, frozen mouse brain tissue, frozen human brain tissue. They showed that they could identify expressions that are particular to neurons and were able to capture closely related cell types.
Thermoresponsive materials such as Agarose have provided researchers a facile tool for replicating single DNA molecules in nanoliter compartments. The Agarose droplets can serve as an inert bioreactor for DNA to perform PCR and. They can also be gelated after the reaction to preserve the amplicons.
In this article, Lent et al. have used a flow-focusing device to generate Agarose droplets that contained DNA templates for subsequent PCR analysis. In their device, an ultra-low gelling Agarose (Melting point= 56oC, Gelling point= 16oC) mixed with PCR reagent serves as the dispersed phase (Qd= 0.4 ml/hr). An ultra-low gelling Agarose ensures easy on-chip droplet production under room temperature as well as remaining in the liquid phase during PCR. Coupling a primer with Agarose matrix avoids diffusion of PCR from the droplets. DNA template molecules are introduced into the PCR mix in a statistically diluted concentration so that on average there will be no more than one template in one droplet. This mixture then encounters oil as the continuous phase (Qc= 2.5 ml/hr) and breaks into droplets. This resulted in generation of highly uniform Agarose droplets with 180±7 µm (3 nL ±0.3 nL) at >500 Hz frequency. Additionally, they studied the effect of the continuous phase flow rate on droplet size and reported the formation of droplets with 110-200 µm diameters.
The droplets are then collected and experienced 25 rounds of thermocycling. The droplets were thermostable since not much size variation was observed before and after thermocycling. The PCR was followed by cooling at 4oC to cure the Agarose droplets. The cured droplets (beads) were then washed to remove the oil phase. The agarose bead showed strength in stirring and centrifugation (necessary for removing the oil phase and replacing it with water) after solidification. Next, the beads were stained and fluorescence images of the droplets with and without DNA templates were taken. It was seen that droplets containing DNA templates were very bright while droplets without DNA templates did not have much fluorescence intensity.
Finally, Q-PCR was employed to determine the PCR efficiency which yielded a 95% efficiency and was claimed to be way higher than 40% ePCR microbead efficiency. The fact that the PCR is carried out in the liquid phase is said to be the reason behind the significantly higher efficiency. Also, various Agarose concentrations were used to investigate the effect of Agarose concentration on the PCR process. PCR efficiency dropped to below 80% when Agarose concentration was >4%. However, 3% and lower did not have any observable effect on PCR efficiency.
Quantitative detection of rare mutations
Oncogenes are a group of genes that have the potential to change normal cells into cancerous cells if they encounter mutation. One of the common oncogenes involved in human cancer is KRAS. KRAS belongs to a family of genes called RAS that have important roles in cell division and growth. Therefore, hyperactivity (overactivity) of these genes can result in cancer.
In this paper, Pekin et al. have used droplet microfluidics to capture rare mutations in KRAS oncogene inside a large population of wild-type cells. They employed digital PCR for quantitative detection of mutants by using the following procedure:
1- An aqueous phase containing gDNA, TaqMan probes, and PCR reagent forming the dispersed phase and oil (HFE-7500, 3M, 2% w/w EA-surfactant) as the continuous phase (each in a 1 mL syringe) ran through the device at dispersed phase and continuous phase flow rates of Qd= 100 µL/hr and Qc= 200 µL/hr, respectively. This resulted in production of 10 pL droplets with 30 kHz frequency.
*The TaqMan probes are specific to mutant and wild-type DNA in a way that a red fluorescent signal is emitted in the presence of wild-type DNA while a green signal is emitted in the case of a mutant DNA.
*The DNA concentration is adjusted so not more than one DNA gets trapped in a droplet.
2- The droplets are collected in a PCR tube (0.5 mL) sealed by a PDMS plug in which two holes (0.75 mm) were made to encompass the collection and reinjection tubes. The collected droplets then experienced a thermocycling process.
3- Subsequently, the droplets are reloaded into the device with oil (HFE-7500 with no surfactant) (Qdroplet= 50 µL/hr, Qoil= 100 µL/hr). These flow rates guarantee a 20 µm distance between successive droplets. The fluorescent signal of the droplets is then analyzed using a data acquisition system comprising a laser with the acquisition rate of 1000 kHz.
4- The bandwidth of the fluorescent signal is measured to remove the droplets with unexpected size and the peek is taken for each droplet to represent the signal. Next, the ratio of green to red is calculated to arrive at the sensitivity of the device.
5- In another experiment, they looked for detection of 6 common mutations in codon 12 of the KRAS oncogene. For this, they fabricated a device comprising 7 droplet generators to simultaneously capture the 6 common type mutants along with the wild type. As expected, each in each inlet a different TaqMan probe was used. The produced droplets were then injected into a fusion device were they were merged with other droplets that contained PCR reagents and gDNA. The rest of the process is similar to the above.
Directed enzyme evolution assays
The new field of protein engineering has provided us with the opportunity of producing proteins with desired functionality for health and industrial applications. Directed evolution is a technique in protein engineering to evolve proteins or nucleic acids with customized functionality. Direct evolution method is very similar to natural selection process and involves iterative rounds of subjecting a gene to mutagenesis, screening the functionality of the template, and amplifying the desired template for the next round. Integration of multiple operations into a robust workflow of an evolution cycle has proven to be challenging.
In this work, Kintses et al. proposed a microfluidic platform for a frequently employed screening process that consists of three steps:
– Cytoplasmic Protein Expression
– Cell Lysis
– Assessment of reaction process by fluorescence measurement
The aqueous phase was introduced using two inlets; One carrying the cell solution and the other one the substrate and cell reagents. The aqueous phase was run at 10 µL/hr while the oil phase (HFE-7500 with EA surfactant) was run at 120 µL/hr to produce particles with the diameters of 30-40 µm. In order to measure the evolution of the reaction, they used the following device and measured the fluorescence intensity of the droplets at 10 control points using an online data acquisition system.
Next, the droplets were stored off-chip for incubation and were reloaded into a droplet sorting device (Qd=200 µL/hr ) along with the oil phase. The oil phase was run at Qc= 1300 µL/hr to ensure the appropriate distance between successive droplets for droplet sorting. The droplets then passed a spot where they were captured by a laser system to analyze their fluorescence intensity. This system generated an electric pulse to activate two electrodes located downstream of the channel to move the droplet with intensities higher than the designated threshold to the top channel while the droplets with lower intensity were kept in the bottom channel.
Single-cell analysis and sorting
In this paper, Mazutis et al. proposed a droplet-based microfluidic platform to compartmentalize cells in single droplets for analysis of antibody secretion by the cells. Their method consists of the following steps.
First, they co-encapsulate the antibody-producing cells and non-producers along with streptavidin beads covered with capture antibody, and detection antibody using the following device. Cell encapsulation is carried out such that no more than one cell gets encompassed in each droplet. Cell and growth medium do not have equal densities (1.1 g ml-1 for cells and 1.0 g ml-1 for growth medium) and the cell encapsulation experiment can take up to 1hr resulting in sedimentation of the cells. Therefore, OptiPrep is added to the solution to match the density with that of the cell.
The droplets are then collected and incubated off-chip for 15 minutes at 37oC and 5% CO2. During the incubation, the secreted antibody attaches to the bead from one side and the detection antibody (green fluorescent–labeled) from the other side resulting in making the beads highly fluorescent.
Next, the droplets are reloaded into a droplet sorting device where a data acquisition system hooked up with a laser beam is used to separate droplets containing fluorescent beads from the rest. The laser beam detects the fluorescent droplets and triggers two electrodes downstream of the channel causing dielectrophoretic displacement of the desired droplets. The sorted droplets transfer to the top channel while the unsorted ones go to the bottom one (waste channel). The flow resistance of the bottom channel is lower than the top channel to ensure that the droplets go to the waste channel when electrodes are off. The sorting throughput depends on the droplet size. The throughput is higher for smaller droplets and for applications in which large droplets are not needed, the throughput can increase to 1kHz and above. The sorted droplets can then be collected and used for determining the sequences.
Chromatin immunoprecipitation and sequencing
Chromatin that forms chromosomes in Eukaryotic cells consists of DNA, RNA, and proteins such as histone. Albeit its importance in protecting DNA and controlling gene expression, its structure is not still well understood. Therefore, Chromatin immunoprecipitation (ChIP) assays were developed. Conventional ChIP techniques often require a large population of cells, and are time-consuming. However, the development of droplet microfluidics has enabled researchers to study the Chromatin’s structure with higher throughput but at a much shorter time (hours instead of days) and by consuming lower reagents.
Here, Rotem et al. provided a microfluidic platform to overcome the deficiency of current assays in capturing cell heterogeneity and profile chromatin in single-cell resolution. Their technique consists of the following steps:
1- A drop-maker was used in which ES cells, a cell lysing detergent, and micrococcal nuclease (MNase) were co-encapsulated. The concentrations were adjusted such that the droplets did not contain more than one cell.
2- A barcode library was engineered, 1152 oligonucleotide adaptors were made and a drop-maker was used to elicit the oligonucleotides from a 3840well plate. This was performed such that each droplet encompassed copies of the same barcode.
3- A 3-point droplet merger was used to merge the droplets from steps 1 and 2 and a small aliquot of an enzymatic buffer. The droplet fusion was triggered by an electric field.
4- The samples were read using the barcodes.
Protein crystallization is highly sensitive to the environmental conditions such as pH and temperature as well as the concentration of the reagents. Therefore, producing a single-protein crystal is often challenging and laborious using conventional methods. However, it is highly desired to have a single-protein crystal to obtain the structure of the protein using x-ray crystallography. In this study, Maeki, et al. studied the effect of droplet size on protein crystallization in a microfluidic device. They could calculate the minimum droplet volume required for crystallization based on the maximum distance that protein can travel by diffusion. This can be used for designing microchannels with proper size that can produce the droplets with the calculated size for crystallization of desired proteins.
Four types of protein with different molecular weights, namely lysozyme, thaumatin, glucose isomerase, and ferritin, were co-flown through the device along with precipitant. They broke into droplets upon meeting a fluorinated oil. The droplets were then collected and the number of crystals was counted every hour. The results revealed the possibility of forming single-protein crystals in droplets and showed the approximate agreement of the estimated critical droplet volume with the experimental values.
3D cell culturing in double emulsions
It is known that 3D cultured cells could be a better representative of in vivo environments. One of the useful 3D models of cells used for creating 3D tissues is spheroids that are a spherical aggregation of cells. Here, a microfluidic device has been introduced for encapsulating single spheroids in hydrogels with tunable compositions. The technique appears to be applicable for various cell lines.
In this work, Chan et al. have employed 2 microfluidic devices for creating an environment for culturing hepatocytes. Hepatocytes are the cells that form a large portion of the liver. Engineered tissues containing hepatocytes are thus of high importance in liver research. They first suspended rat hepatocyte cells in an aqueous solution containing alginate. Next, the cells’ solution and an oil phase were introduced into the first device where the droplets containing cells were formed inside a carrier oil phase. These droplets were then transferred to the second device to generate water in oil in water double emulsions with an inner diameter of approximately 200 µm. The droplet size was shown to be adjustable by changing the flow rate or the device design. Double emulsions were then incubated for 4h to form spheroids followed by removal of the outer shell using cell strainers. Next, the core droplets were polymerized using calcium chloride that resulted in hydrogel droplets containing single spheroids of approximately 80 µm.
Janus microparticles from functional precursor polymers
Janus particles are compartmentalized bifacial particles that are named after the two face Roman God, Janus. Usually, the two sides of the Janus particles, have different surface properties, polarity or chemistry. This anisotropic characteristics, makes Janus particles a good option for a variety of applications. For example, Amphiphilic Janus particles can be used as powerful surfactant. They have also been used as microsensors, actuators, and models for self-assembly study.
In this work, Seiffert and coworkers have used the following device to make Janus particles. The novelty of their work lies in separating the droplet synthesis and particle gelation steps. However, the delay between droplet generation and gelation raises the chance of distortion of the phases inside the droplet. One major concern in Janus particle generation has always been retaining the structure of the particle upon droplet formation. The contents of the droplet are prone to intermixing due to the diffusion or secondary induced flows inside the droplets. In this paper, however, it has been shown that by proper selection of the channel size and reagents, the initial structure could be retained even after through a serpentine channel. Consequently, after generation, the droplets travel a 2cm distance to reach the delay line (a serpentine channel) where they get exposed to a strong UV light and solidified. This approach proved to be promising for making three types of Janus particles, namely, Janus particles, Janus shells using double emulsions, Janus microgels that are loaded with ferromagnetic additives. All these products could be made with the same junction and by a slight change in flow rates or reagents.
Fluorescent-magnetic bifunctional microparticles
The advent of microfluidic techniques, has enabled production of multifunctional particles for applications in drug delivery, biochemical analysis, etc. The size of the particles is usually dominated by size of the droplets that is often in 10-100 µm range. Few devices have been reported to make particles in 1-10 µm size span.
Here, Yang et al. developed a droplet splitter device that enabled manufacturing fluorescent-magnetic droplets with the diameters of 2.8-19.2 µm in a single step. In their device, the aqueous solution contained 10 wt%poly(ethylene glycol) diacrylate, 11.8 g L_1 PEI–Fe3O4, 0.2 mg mL_1 acryloyl-RhB and 0.5 wt% Irgacure 2959 while mineral oil with 2 wt% span 80 and 0.5 wt% 2,2-dimethoxy-2-phenylacetophenone was used as the oil phase. The Aqueous phase and oil phase were pumped to a flow focusing device where the mother droplets were generated with an average monodispersity of 3.7%. The droplets then encounter an orifice that causes them to split and go to different outlets. The droplets are then collected for off-chip photopolymerization. The size of the droplets and accordingly the particles can be adjusted by changing the flow rate ratio. Plots of the droplet size for various flow rate ratios are provided.
Yang, Yu-Jun, et al. “Generation of sub-femtoliter droplet by T-junction splitting on microfluidic chips.” Applied Physics Letters 102.12 (2013): 123502.
Yang, Chun-Guang, Ru-Yi Pan, and Zhang-Run Xu. “A single-cell encapsulation method based on a microfluidic multi-step droplet splitting system.” Chinese Chemical Letters 26.12 (2015): 1450-1454.
Generation of composite biopolymer microgels
Here, Chau et al. produced composite biopolymers of agarose-gelatin using a microfluidic approach. In their work, they introduced agarose solution, gelatin-ph solution, and hydrogel peroxide through their device to form the aqueous phase inside the fluorinated oil HFE 7500 as the oil phase.
Upon generation, the droplets passed through a serpentine channel that allowed the preoxidase-catalyzed gelation of the gelatin modified with phenolic hydroxil group. The partially gelled droplets were then collected and kept in ice bath to gel the agarose component of the droplets. Subsequently, the composite microgels (105-175 diameter-3%CV) were analyzed to investigate the effect of flow rate on composition, morphology, structure, and stiffness. This method was shown to be a promising tool for making customized artificial cellular matrices.
Janus and ternary particles
In this study, Nie et al. reported a technique for producing Janus microparticles in 40-100 micron diameter range. They then extended their work and employed their device for fabrication of ternary (3 phase) particles. Briefly, two aqueous phase M1 (methacryloxypropyl dimethylsiloxane) and M2 (mixture of pentaerythritol triacrylate, poly(ethylene glycol diacrylate, and acrylic acid) were co-flown through the device. In the case of ternary particles, M2 was sandwiched between two streams of M1.
In this stage, proper selection of M1 and M2 was very important to make sure a sharp interface exists between the phases. Diffusion of one phase to another thus making a gradient in the particle can limit its application for many cases. An aqueous phase consisting of 2% wt sodium dodecylsulfate was used as the carrier phase. The droplets were then exposed to UV light to form solid particles with amphilicphilic properties. By changing the inlet flow rates, Nie et al. could successfully generate monodisperse Janus and ternary particles (2.8% < CV < 4.7%) with 40-100 micron diameters. Furthermore, they showed that by changing the M1 and M2 flow rates, various particles with asymmetric properties can be generated.
Hydrogels such as Poly(ethylene glycol) family (PEGs) are widely used polymers with applications in drug delivery, biomedicine, tissue engineering, etc. Conventional methods for hydrogel production lack monodispersity and are mostly time consuming. However, micofluidics has appeared as a powerful tool for producing hydrogels with customized properties.
In this regard, Dang et al. devised a method to generate monodisperse PEGDA and PEGMEA hydrogels. They first performed their experiments with three flow focusing devices to investigate the effect of device geometry, flow rate and reagent concentration on particle size. Next, the results was used to make particles functionalized with 6 micron beads. Briefly, their method is as follows:
Mineral oil and PEG solutions are introduced through the device as the continuous phase and dispersed phase, respectively. The droplets then get formed due to hydrodynamic breaking of the dispersed phase into droplets. Next, the droplets travel through a serpentine channel where they get exposed to UV through a pre-embedded window open window on top of the device and solidify.
Biomimetic material synthesis
In their novel study, Bawazer et al. combined combinatorial techniques with an exquisite platform for high-throughput droplet generation and screening to determine the best combinations of oil/surfactant that produce droplets with the highest endurance. This technique has the potential to be used for rapid identification of the superior oil/surfactant combination for a given application.
Here, they interfaced a 96 well microplate with droplet generator PDMS devices. Each device was connected to 4 well. Two wells were connected to the inlets (for oil and water phase) and one well to the outlet. The fourth well was used as an observation window. The plate was connected to a standard vacuum manifold (Supleco) to suck the reagents.
A broad range of oils and surfactants including small-molecules, polymeric, ionic, nonionic, and non-particulate surfactants were sampled and the stability of the droplets was then analyzed using a genetic algorithm to come up with the best combination that produces droplets with the highest stability.
Generation of inorganic-organic particles
In the study conducted by Prasad et al., hybrid Janus microparticle with organic and inorganic sides were generated. At first, an organic solution (photocurable PFPE) and an inorganic phase (AHPCS) were introduced to a Y-channel to form a narrow laminar stream. This stream then faced a continuous oil phase and broke into dumbbell-like Janus particles with 162 µm diameter and 3.5% polydispersity.
Following the generation, the droplets were cured via a UV light. Thereafter, the microparticles underwent pyrolysis treatment to remove the organic phase. Removal of the organic side, transformed the dumbbell-like particle into a ceramic hemisphere in two steps. First, the organic phase shrunk into a small hemisphere attached to the inorganic phase thus making a mushroom shape. Next, the organic phase got removed completely with further heating resulting in a concave hemisphere remnant. Furthermore, magnetic particles were added to the particle and the generated particles were tested under magnetic field for forced assembly.
Synthesis of cell-laden alginate particles
Conventional methods for producing alginate microgels are not capable of producing particles with high monodispersity. However, having monodisperse particles with the possibility of functionalization is important for applications such as serving as carriers for cells, proteins, and drugs.
Here Akbari and Pirbodaghi proposed a microfluidic approach for encapsulating mammalian cells inside alginate droplets. They selected an off-chip method for polymerization since, depending on channel dimensions, on-chip gelation can cause device clogging. For an external gelation method, a divalent ion such as Ca2+ can be added to the droplets. However, exposure of cells to these ions can reduce the cell viability significantly. Consequently, they devised a droplet generator in which hybridoma cells and mouse breast cancer cells immersed in alginate made a coaxial stream with alginate mixed with CaCO3. These streams then met the oil in the second junction and broke into droplets. CaCO3 was used as a means for producing Ca2+ ions required for polymerization. Next, the droplets were collected and acetic acid was added to the droplets to trigger the polymerization. The exact stoichiometric amount of acetic acid required for initiating the gelation was calculated based on the following equation:
2CH3COOH+CaCO3–>Ca(CH3COO)2+ H2O+ CO2
The particles were then immediately washed and moved to the cell growth medium to minimize the exposure of cells to low pH. The cells were then incubated and 84% cell viability was shown.
Double emulsion droplets for flow cytometry analysis
Here, Yan et al. utilized a double flow-focusing device to generate w/o/w double emulsions for further flow cytometry analysis. Four solutions were used as the inner core of the double emulsion:
1- PBS (Phosphate-buffered saline)
2- 0.025 μM FITC (fluorescein isothiocyanate) in PBS
3- 0.25 μM FITC in PBS
4- 2.5 μM FITC in PBS
The oil phase and outer aqueous phase consisted of HFE-7500 and DI water, respectively. Upon generation, the droplets were stored in storage channels and their fluorescent intensities were analyzed. It was found that proper choice of the oil phase inhibits the diffusion of the inner core content into the surrounding. It was proven by showing that no fluorescent intensity was emitted from the outer aqueous phase. The channel height was 25 micron and the size analysis revealed the inner core and outer shell diameters to be 71.4 µm and 84.2 µm, respectively.
Enzyme expression study in E. coli
Directed evolution is a strong means for engineering new proteins based on natural selection-like process. Briefly, it involves adding a mutation to create new variants and selecting the one with desired properties followed by its amplification. This process requires fast and high-throughput devices to meet the ever-increasing demand of the protein engineering market. The fastest conventional enzymatic screening method employs robotic for microtitre plates. This method, however, has reached its limit since not more than 1 microL of reagent could be processed due to evaporation. This opens up the space for droplet-based microfluidics that is capable of compartmentalizing the desired reagents in nL-pL droplets and enables faster (10 folds) assays with less reagent consumption (1000 folds).
Here, Beneyton et al. proposed a device for screening cotA laccase activity in E. coli. Their technology relied on following steps:
1- Encapsulating E. coli in droplets
E. coli cotA and E. coli ∆cotA (frameshifted inactive variant) were encapsulated in 14pL aqueous droplets immersed in HFE7500 oil. A green fluorescent barcode was co-encapsulated with E. coli to tag the bacteria.
2- Incubation to allow cell growth and protein expression
3- Injection of a fluorogenic substrate
Droplets were reloaded into an injection device where 1pL droplets containing fluorogenic substrates were merged with the incubated droplets using an electric field. 99.5% of the picoinjection was reported to be successful. The droplets were then collected for the next step. It should be noted that due to the proper choice of surfactant, the droplets were highly stable such that only 0.6% of droplets coalesced while being reloaded.
4- Incubation to allow enzymatic activity
5- Sorting droplets using a fluorescent activated cell sorter (FACS) based on fluorescence intensity gained in step 4.
High-throughput screening of filamentous fungi
Filamentous fungi are of utmost importance in industrial scale because of its capacity for secreting proteins. However, current techniques for enzymatic screening of filamentous fungi in 96-well plates are expensive and slow (100 fungi/hr) thus limiting the possibility of producing new strains. This necessitates the need for developing novel methods for high throughput, low-cost screening of enzymatic activity of filamentous fungi.
Here, Beneyton et al. combined droplet microfluidics with microtiter assays to come up with a rapid, high throughput method ro screen enzymatic activities in Aspergillus niger. First, they co-encapsulated single pores along with the growth media and fluorogenic α-amylase in 18 nL (<5% polydispersity) droplets in an oil carrier phase (HFE 7500). Next, the droplets were collected and incubated overnight to allow. At this stage, the droplets with enzymatic activity begin to emit green fluorescent signals.
The bottom line for employing droplet microfluidics for filamentous fungi assay is using large droplets. This is because detectable enzymatic activity can require at least 24 hr of incubation. On the other hand, the rapid growth rate of the fungi causes the hyphal tips to exit the droplets even as large as 250 pL in 15 hr. However, using nL droplets enables longer incubation without the risk of unwanted coalescence due to the hyphal growth.
The incubated droplets were then introduced to a fluorescent activated droplet sorter (FADS) device where the droplets with α-amylase activity where separated from inactive droplets. Finally, the sorted samples were analyzed with a 96-well micro for α-amylase activity screening. The results showed that 98.9% of the sorted droplets were active revealing the high success rate of the microfluidic device for enzymatic screening. Although the device was used for screening α-amylase activity, it has the potential for screening other enzymes and metabolites.
Encapsulating bacteria in agarose
Minimum inhibitory concentration (MIC), defined as the minimum antibiotic concentration that inhibits the visual growth of bacteria, is an important factor in prescribing the correct drug to patients. Conventional methods for determining MIC rely on the preparation of various concentration of the antibiotic and incubation with the bacterial followed by serial culturing and inspection to determine the MIC. Although these methods are vital for microbiology, they are not satisfactory for slow growing bacteria. Also, they require a considerable amount of reagents that make these assays expensive. Here, Eun et al. showed the capability of droplet microfluidics for determining the minimum inhibitory concentration (MIC) of E. coli with respect to rifampicin.
They also used their technique to identify the spontaneous mutants that caused the resistance to antibiotic. In order to investigate the effect of rifampicin concentration on growth of E. coli was investigated by encapsulating E. coli (MG1655-ptetEGFP) in agarose beads using a flow-focusing device. The device made monodisperse droplets ( 30µm diameter). Upon formation, the droplets were incubated and analyzed using flow cytometry. The results was then compared with that of induced and uninduced samples to arrive at the minimum inhibitory concentration.
Similarly, the resistant mutants were selected by encapsulating E. coli along with rifampicin in agarose beads. However, this time the rifampicin concentration was selected to be higher than its MIC followed by incubation. Next, the droplets were analyzed with a FACS device to find the GPF+ samples. This subpopulation was then collected to isolate and sequence mutants. This approach decreases the amount of required reagent and assay time subsequently.
All in all, this approach was proven as a powerful tool for encapsulating, incubating and analyzing bacteria with single-cell resolution and was successful in determining the MIC and spontaneous mutants. This system is envisioned to be used for bacteria isolation based on genotype and phenotype.
A programmable microenvironment for cellular studies
The advent of microfluidics has revolutionized the biological research in cellular level by increasing the throughput and enabling simultaneous screening of cellular activities. As the field grows, more complex systems and compartments are required to facilitate intricate cellular activities. For example, use of W/O emulsions can be unfavorable when droplet generation should be followed by a flow cytometry assay that requires aqueous based compartments. Another example can be for cell/bacteria culturing where continuous supply of nutrients is needed for long incubation times. Zhang and coworkers proposed double emulsion W/O/W to solve this shortcoming.
Here, the oil phase acts as a selective membrane that enables the small nutrients to enter the core phase and prohibits large molecules from intrusion. They used this technique to encapsulate bacteria and a reporter gene in a double emulsion to show that the core environment inside the double emulsion can easily be controlled by chemical diffusion from the surrounding environment. The double emulsion generation was conducted in two steps. First, W/O droplets were generated in a hydrophobic device and then the droplets were transferred to another device with a hydrophilic surface where they formed the W/O/W double emulsions.
To show chemical diffusibility of the oil layer, Rhodamine B (a small fluorescent molecule) and Rhodamine B labeled BSA (Bovine Serum Albumin) were co-encapsulated in the droplets. Results showed that Rhodamine B diffused to the surrounding while Rhodamine B labeled BSA could not protrude.
To prove the capability of a double emulsion for controlling the microenvironment inside the double emulsion, E. coli (MC4100Z1) along with a GFP gene, that gets controlled by a protein(Tetr), were co-encapsulated in a double emulsion. MC4100ZA strain is capable of expressing Tetr. However, a small molecule called anhydrotetracycline (aTC) is required for GFP expression. Next, aTC was introduced to the aqueous phase surrounding the double emulsion. After 1hr, fluorescent signals were emitted proving the diffusion of aTC through the oil layer. This was further validated by encapsulating a bacteria that constantly emits GFP. In this case, the signal intensity is dependent on the cell population. Consequently, the double emulsions were moved to PBS and nutrients were added to the surrounding. Fluorescence intensity was then measured and significant increase of the intensity was observed. This confirms transport of the nutrients to the double emulsion resulting in multiplication of the bacteria inside the core thus increasing the fluorescence signal.