$$\rightleftharpoonup{xx}$$
$$\longleftharp{xx}$$,
$$\longrightharp{xx}$$,
The fabricated polymeric droplet-based microfluidic platform consists of two PDMS layers (Figure 1a). Three kinds of microfluidic channel networks are used for generating microspheres: 1) Flow-focusing geometry as shown in Figure 1b, 2) a serpentine channel for mixing solution I and solution II, and 3) a polymerization channel for microsphere solidification. The height of all channels was 60 μm. The channel length for mixing and polymerization were 74.35 mm and 94.45 mm, respectively. The widths of the microchannel for two immiscible fluid flows and the one for mineral oil flow were 100 μm and 200 μm, respectively. The orifice structure used for microsphere formation was of 50 μm length and 25 μm width. The angle of the diffuser structure was 37°. A lab-based pneumatic control system for continuous flow of the mineral oil and two syringe-pumps for the solution flow (Figure 1c) were used for generating microspheres in the microfluidic platform (Figure 1d). The production speed of microspheres was about 30 microspheres per second when the flow rate of solutions and applied pressure were at 0.6 mL/h and 108 kPa, respectively (Table 2). Its diameter in the micro channel is 78.7 ± 2.5 μm. The on-flow synthesis of microspheres can be successfully manipulated using the microfluidic device. Bead sizes were measured after swelling. The average diameter of the resulting microspheres was 150.4 ± 12.8 μm. The size variations were about 8.5%. The microspheres undergo swelling in water, resulting in an enormous size increase.
The copolymerizable property of microsphere can be varied. We incorporated a 5'-acrydite-DNA probe into the microsphere solution (solution I, see Table 1). Following co-polymerization, a complementary DNA probe is labeled with fluorescent dye, Cy3, at room temperature for 1 h. Confocal microscopy can be used to prove the synthesis of copolymerizable oligomicrospheres as well as functional hybridization on the surface of microspheres. Fluorescent images of the microsphere are shown in Figure 2. If microspheres are correctly functionalized with the 5'-acrydite-DNA probe, they should result in coverage of fluorescent activity on the surface during the hybridization experiment. If copolymerization did not occur correctly, the optical microsphere image would exhibit internal-contamination inside the microspheres. As shown in Figure 2, there is no interference of random-interior orientation. This result allowed us to carry out the DNA probe presentation in a 3-dimensional (3-D) arrangement and microsphere-PCR. It should be noted that an identical DNA probe with a 5'-NH2-group instead of a 5'-acrytide modification did not copolymerize during on-flow synthesis of microsphere8.
When working with microspheres, manipulating the 3-D surface with a DNA oligo-probe is a much faster process14. Therefore, an on-flow microsphere synthesis platform can provide a tool for ssDNA amplification and purification, as outlined in Figure 3. The anti-sense DNA template (-, complementary template) can be extended by adding a random DNA template (+, template). In this case, the initially copolymerized 5'-Ap provides a 3'-OH end for DNA polymerization after annealing to its complementary template (76 nt). Microsphere-PCR can be performed in a single PCR microtube using functionalized microspheres, a random DNA template, and a Tag-forward primer. In order to distinguish the resultant ssDNA amplicons from random DNA template (+ template, 76 nt), the forward primer has an additional 24 nucleotide tag. If the forward primer does not have an additional tag sequence, it is very hard to recognize between ssDNA amplicons and the initial random DNA template.
An asymmetric PCR experiment was performed. We were able to observe dsDNA contaminants as shown in Figure 4. In most cases, it is necessary to isolate ssDNA using streptavidin-coated magnetic beads and exonuclease digestion. However, microsphere-PCR makes efficient use of a single primer (Tag-forward primer) to accumulate ssDNA in aqueous phase. It is expected that dsDNA contaminants will attach to the surface of microspheres. Therefore, ssDNA amplicons can be obtained through pipetting without the need for centrifugation. The resultant ssDNA was demonstrated by comparing it to the synthetic size markers (76 nt ssDNA and 100 nt ssDNA) through gel electrophoresis analysis.

Figure 1: The microfluidic platform. (a) Fabricated microfluidic platform, (b) enlarged view of the flow-focusing geometry, (c) experimental set-up, and (d) Captured images showing continuous generation of microspheres. 1: Micro channel structure for flow focusing geometry, 2: for mixing solutions, 3: for microsphere solidification. Please click here to view a larger version of this figure.

Figure 2: Fluorescent readout of DNA hybridization on the surface of microspheres. 5'-Acrydite-modified DNA probes (Ap) were capable of hybridizing with complementary Cy3 labeled DNA probes (cAp). Please click here to view a larger version of this figure.

Figure 3: Illustration of the microsphere-PCR protocol. Please click here to view a larger version of this figure.

Figure 4: Comparison between conventional asymmetric PCR and microsphere-PCR. M; ssDNA marker (76mer and 100mer), Lane 1; Asymmetric PCR, Lane 2; Microbeads-PCR. Reprinted with permission from previous work8. Please click here to view a larger version of this figure.
| Reagent | Volume in mix (μL) | Final Concentration |
| Solution I | 40% Acrylamide:bis solution (19:1) | 25 | 10% |
| 100 μM Acrydite probe (Ap, 5’-Acrydite-(TTTTTTT, linker sequence) AGA TTG CAC TTA CTA TCT-3’) | 10 | 10 μM |
| 5X TBE buffer (Tris-base-EDTA) | 10 | 0.5X |
| Water | 5 | - |
| Solution II | 20% Ammonium persulfate | 50 | 10% |
| Solution III | TEMED (N,N,N′,N′-Tetramethylethylenediamine) | 8 | 0.40% |
| Mineral oil | 1000 | - |
Table 1. Reagent mix components for on-flow polyacrylamide microsphere synthesis.
| Operational condition | Microsphere
production speed |
| Solution flowrate | Oil pressure |
| (mL/h) | (kPa) | ( / sec) |
| 0.4 | 82 | 7.1 |
| 0.5 | 94 | 13.3 |
| 0.6 | 108 | 28.9 |
| 0.7 | 116 | 36.5 |
Table 2. Summary of operational conditions and produced microsphere.
| DNA probe | Sequences |
| Cy3 labeled complementary oligonucleotide probe (cAp) | 5’-Cy3-AGATAGTAAGTGCAATCT-3’ |
| Tag (first 24 nt)-forward primer | 5’- GGT AAT ACG ACT CAC TAT AGG GAG ATA CCA GCT TAT TCA ATT-3’ |
Table 3. Sequence information of DNA probes.
| Reagent | Volume for 1x reaction (μL) | Final concentration |
| Random DNA template | 1 | 1 ng/μL |
| Tag-forward primer | 1 | 0.4 μM |
| Reverse primer | 1 | 0.02 μM |
| 10X Taq buffer | 5 | 1X |
| 10 mM of dNTP | 4 | 2.5 mM |
| Ex taq (1000U) | 0.2 | 1 U |
| Water | 37.8 | - |
| Total | 50 | - |
Table 4. Asymmetric PCR reagent mix components in 20 μL reaction.
| Template | Sequence | Length (nt) |
| Random DNA template | 5’- ATA CCA GCT TAT TCA ATT (Random sequence, 40 mer) AGA TAG TAA GTG CAA TCT-3’ | 76 |
| Tag-Forward primer (Tag-F) | 5’- GGT AAT ACG ACT CAC TAT AGG GAG ATA CCA GCT TAT TCA ATT-3’ | 42 |
| Reverse primer | 5’- AGA TTG CAC TTA CTA TCT-3’ | 18 |
Table 5. Sequence information for Microsphere-PCR.
| Reagent | Volume for 1x reaction (μL) | Final concentration |
| Ap-Microspheres | ~25 microspheres | - |
| Random DNA template | 1 | 1 ng/μL |
| Tag-Forward primer | 1 | 0.4 μM |
| 10X Taq buffer | 5 | 1X |
| 10 mM dNTP | 4 | 2.5 mM |
| Ex taq (1000U) | 0.2 | 1 U |
| Water | 37.8 | - |
| Total | 50 | - |
Table 6. Microsphere-PCR reagent mix components in 50 μL reaction.
| Scan mode | Plane |
| Scaling | X: 2.49 μm, Y: 2.49 μm |
| Stack size | X: 1272.79 μm, Y: 1272.79 μm |
| Scan zoom | 0.7 |
| objective | EC Plan-Neofluar 10 x / 0.3 M27 |
| Average | 1 |
| Pinhole | Ch2: 104 um |
| Filters | Ch3: LP 420 |
| Beam splitters | MBS: HFT 488 / 543, DBS1: mirror, DBS2: NFT 515, FW1: None |
| Wavelength | 543 nm 100.0 % |
Table 7. Settings for the confocal microscope.