Microfluidic Selective Concentration of Microdroplet Contents by

Mar 11, 2015 - Microfluidic protein crystallisation controlled using spontaneous emulsification. Mao Fukuyama , Aoi Akiyama , Makoto Harada , Tetsuo O...
0 downloads 0 Views 1MB Size
Download PDF
Letter pubs.acs.org/ac

Microfluidic Selective Concentration of Microdroplet Contents by Spontaneous Emulsification Mao Fukuyama and Akihide Hibara* Department of Chemistry, Tokyo Institute of Technology, 2-12-1-W4-20 Ookayama Meguro-ku Tokyo, 152-8551, Japan S Supporting Information *

ABSTRACT: The selective concentration of the contents in a microdroplet using spontaneous emulsification was proposed and demonstrated in a microfluidic channel. Aqueous microdroplets having a 40-μm diameter, in octane containing 100 mM of Span 80, shrank to 10 μm within 10 min with nanodroplet formation at the interface of the microdroplets. The microdroplets’ contents either stayed in the microdroplet or partitioned into the nanodroplets, depending on their properties. The size and the hydrophobicity of the contents are two parameters that determine concentration/separation. In addition, this method was applied to a bound complex and free ligand (B/F) separation to demonstrate its applicability to biochemical analyses. Here we report the separation of water-soluble molecules in microdroplets for the first time. This method is expected to enhance the flexibility of the design of droplet analytical processes and widen their applicability.

I

rather than simple, concentration of water-soluble substances is preferred. Here, we use nanodroplet formation at the interface of the microdroplets, by spontaneous emulsification, for the selective concentration of the microdroplets’ contents. Spontaneous emulsification occurs without external energy supply in specific aqueous-organic-surfactant systems.25 Here, we use the spontaneous emulsification induced by Span 80,26 a nonionic surfactant composed of a sugar ring and an alkyl chain, which was used due to its biocompatibility and its frequent use for microdroplet generation in microdevices.27 The outline of the selective concentration of a microdroplets’ contents is shown in Figure 1. Microdroplets are formed by introducing an aqueous sample solution to an organic phase containing Span 80. After generation, spontaneous emulsification occurs at the interface of the aqueous microdroplets. The microdroplets then shrink with nanodroplet formation because water molecules in the microdroplets partition into the newly formed nanodroplets. Then, the microdroplets’ contents either stay in the microdroplet or partition to the nanodroplet, depending on their properties. Here, the selective concentration of the microdroplets’ contents using spontaneous emulsification is proposed and demonstrated. The concentration factor and the solute selectivity are reported. In addition, this method is applied to a bound complex and free ligand (B/F) separation to demonstrate its feasibility for biochemical analyses.

n the past decade, the applications of microfluidic devices to chemical and biochemical analyses have been studied.1,2 This technology enables the reduction of analysis time, required sample amount, and reagent consumption by miniaturizing the analytical operations. Recently, microdroplets formed in microfluidic devices have been studied and used as picolitersized chemical containers.3−5 By introducing two immiscible flows into the microchannels of a microfluidic device, micrometer-sized monodispersed droplets can be generated.6,7 Substances such as molecules,8 nanoparticles,9 and cells10 can be isolated in each microdroplet. In addition, manipulations of microdroplets, such as fusion,11 mixing,12 and splitting13 can be achieved by controlling the structure and wettability of the microchannel. Microdroplets have been applied to highthroughput microchemical and -biochemical analyses,14 such as protein crystallization,15 high-throughput single cell assays,16 and in vitro protein synthesis17 by using these features. In these microdroplet-based microanalytical systems, the detection of each of the microdroplet’s contents might be a limiting factor because of their short optical path length (∼micrometer), and the small amount of sample inside each microdroplet (∼attomole). To overcome this difficulty, several solutions have been used, such as chemical amplification (e.g., enzyme reaction18 and PCR19), highly sensitive laser spectroscopy,20 and preconcentration of the microdroplet’s contents.21−24 Thus, far, several studies on the preconcentration method for microdroplets, by using the dissolution of the droplet medium to the continuous phase, have been reported.21−24 In these studies water-in-oil (W/O) droplets were formed, then the droplets’ contents were concentrated because of the shrinkage of the microdroplets by dissolution of water into the organic phase. However, for versatile analytical pretreatment a selective, © 2015 American Chemical Society

Received: January 13, 2015 Accepted: March 11, 2015 Published: March 11, 2015 3562

DOI: 10.1021/acs.analchem.5b00155 Anal. Chem. 2015, 87, 3562−3565

Letter

Analytical Chemistry

Figure 1. Selective concentration of a microdroplet’s contents by spontaneous emulsification.

Figure 2. Concentration of sulforhodamine B in microdroplets. (a) Bright field and fluorescent micrographs of microdroplets containing 0.1 mM sulforhodamine at 2 and 15 min after formation. (b) Time course of microdroplet diameter. (c) Amount of sulforhodamine in a microdroplet calculated based on the concentration analysis from the fluorescent images (see the Supporting Information). (d) Concentration factor.

■

EXPERIMENTAL SECTION A glass microfluidic device, fabricated by a two-step photolithographic wet-etching, was used. The channel wall was hydrophobized using octadecyltrichlorosilane (Sigma-Aldrich Japan KK, Japan).28 Microdroplets were generated at the Tjunction of a 10-μm-deep and 70-μm-wide shallow channel and a 100-μm-deep and 250-μm-wide deep channel.29 Octane (Wako Pure Chemicals, Japan) containing Span 80 (SigmaAldrich Japan KK, Japan), and aqueous solutions of fluorescent dyes were used as organic and aqueous phases, respectively. Sulforhodamine B (MP Biomedicals LLC), rhodamine 123 (Wako Pure Chemicals, Japan), rhodamine B (Sigma-Aldrich Japan KK, Japan), rhodamine 6G (Wako Pure Chemicals, Japan), fluorescein sodium salt (Wako Pure Chemicals, Japan), and fluorescein-tagged polyethylene glycols (MPEG 350, 550, and 2000, purchased from Nanocs, Inc.) were used as fluorescent dyes. The 40-μm diameter droplets were formed by introducing the aqueous solutions of fluorescent dyes at a flow rate of 0.01 μL/min from the shallow channel into the deep channel, where the organic phase was flowing at a flow rate of 0.1 μL/min. After droplet formation, the flow of the aqueous phase was stopped and the time course of the diameters and the fluorescent intensities of the microdroplets

were observed using a CCD camera (FASTCAM SA3, Photoron, Japan) under an organic flow condition of 0.02 μL/min. To demonstrate a B/F separation, phosphate buffer saline solutions of avidin (Wako Pure Chemicals, Japan) and of 10 μM biotin-fluorescein (Thermo Scientific) were used to form microdroplets. In this experiment, the phosphate buffer saline (PBS, 137 mM NaCl, 2.68 mM KCl, 10 mM Na2HPO4, and 2 mM NaH2PO4) was diluted to 100 mM.

■

RESULTS AND DISCUSSION As shown in Figure 2a, the microdroplets shrank and sulforhodamine was concentrated in the microdroplets. Concurrently, nanodroplets formed at the interface of the microdroplets, which was caused by spontaneous emulsification. The microdroplets of the 0.1 mM aqueous solution of sulforhodamine, which did not dissolve in octane (see the Supporting Information), in octane containing 100 mM Span 80, shrank from 40 to 10 μm and then to 6 μm, in 10 and 30 min, respectively (Figure 2b). This shrinkage was not observed in octane without surfactant (see the Supporting Information for details). The amount of fluorescent dye was estimated from its concentration and the volume of the microdroplets (see the 3563

DOI: 10.1021/acs.analchem.5b00155 Anal. Chem. 2015, 87, 3562−3565

Letter

Analytical Chemistry Supporting Information for a detailed procedure). As a result, most of the sulforhodamine was concentrated in the microdroplets (Figure 2c). In addition, the concentration factor, C, was calculated as C = c/c0, where c and c0 are the final and initial concentrations, respectively. The concentration factor after 30 min was found to be 500 (Figure 2d). When microdroplets of a higher initial dye concentration were formed, the final diameter of microdroplets was larger (Figure 3). This result indicated that water transport from the

Figure 4. Concentration of fluorescent dyes vs microdroplet diameter. Rhodamine dyes (a) and fluorescein-tagged polyethylene glycols (b) were used as fluorescent dyes. The numbers after the names of fluorescent dyes in part a indicate the logarithm of the water-octanol partition coefficient. The molecular weight of the polyethylene glycol moiety is indicated as MPEG.

Figure 3. Concentration dependence of the final diameter of microdroplets. Sulforhodamine B was used as a solute in the microdroplets. The final concentrations of the sulforhodamine for each initial concentration (ci) were cf = 40 mM (ci = 0.1 mM), 120 (ci = 1 mM), 230 (ci = 5 mM), 250 mM (ci = 10 mM), respectively.

molecular weight (MW = 732.80). On the other hand, the complex of avidin and fluorescein-tagged biotin is large (MW > 60 000), and it was expected to be concentrated in the microdroplets. Thus, the fluorescence intensity was expected to increase with an increase of the initial avidin concentration. Figure 5 shows the dependence of the fluorescence intensity

microdroplets to the nanodroplets was considered to be affected by concentration-dependent parameters, such as hydration and osmotic pressure. That is, the concentration factor is controlled by the concentration of the major solute in the microdroplets. While most sulforhodamine was concentrated in the microdroplets, several other fluorescent dyes partitioned into the nanodroplets formed by spontaneous emulsification, instead of concentrating in the microdroplets. The selectivity of this concentration/partition was investigated, and its dependence on the hydrophobicity and the size of the solute was determined. To examine the effect of hydrophobicity, the behaviors of several kinds of rhodamine dyes were observed in the same manner as the experiments for Figure 2 (see the Supporting Information for details). The concentration factor of the rhodamine dyes was increased with a decrease of the water-octanol partition coefficient30,31 (Figure 4a). This result indicates that hydrophilic molecules are concentrated in the microdroplets rather than partitioned into the nanodroplets. To investigate the effect of molecular size, the behaviors of various fluorescein-tagged PEGs (molecular weight of PEG part (MPEG) = 0, 350, 550, and 2000) were observed. Larger molecules tended to concentrate in the microdroplets. In addition, it should be noted that the partition of the fluoresceintagged PEG from the microdroplets to octane without Span 80 was not observed (see the Supporting Information for details). Thus, far, we postulate that the mechanism is similar to that of the reverse-micellar extraction by using nonionic surfactants.32 Further investigation will be reported elsewhere. A B/F separation was demonstrated using a biotin−avidin system, which is often used for biochemical analyses. When avidin and fluorescein tagged biotin exist together in a microdroplet, the free fluorescein-tagged biotin was expected to partition into the nanodroplets because it has small

Figure 5. Fluorescent intensities of microdroplets vs initial avidin concentration. The dashed blue line is the theoretical line calculated from the assumption that (i) avidin was quantitatively concentrated in the microdroplets and (ii) a certain concentration of free biotin, which was determined from the 0 mM avidin result, stayed in the microdroplets. The initial concentration of fluorescein-tagged biotin was 10 μM.

after the shrinkage process (16 min after droplet formation) on the initial avidin concentration. In this experiment, 43-μm-sized microdroplets were formed which then shrunk to 19 μm. The final droplet size of 100 mM PBS (19 μm, final concentration 1.2 M) was larger than that of 10 mM sulforhodamine (12 μm, final concentration 250 mM). This result agreed with the tendency that the higher initial concentration leaded larger final droplet size as shown in Figure 3. As expected, the fluorescence 3564

DOI: 10.1021/acs.analchem.5b00155 Anal. Chem. 2015, 87, 3562−3565

Analytical Chemistry

■

intensity increased with an increase of the initial avidin concentration. The fluorescence intensity in the absence of avidin indicated that about a half of fluorescein-tagged biotin stayed in the microdroplets. The measured fluorescence intensity was in agreement with the values which were calculated from the assumptions that (i) avidin was concentrated in the microdroplets and did not partition to continuous phase and nanodroplets, (ii) biotin molecule were bond to avidin molecules whenever the binding site of avidin is empty, and (iii) a certain concentration of free biotin, which was determined from the 0 mM avidin result, stayed in the microdroplets. This result successfully demonstrated this selective concentration method, which can be used for preconcentration in biochemical analyses, such as competitive assays. This is the first report on the selective concentration of the water-soluble contents inside a microdroplet. The spontaneous emulsification is reported to occur when the Span 80 concentration is higher than the critical micellar concentration and alkanes are used as a continuous phase.26 In addition, it is known that spontaneous emulsification can occur not only by Span 80 but also other surfactants such as AOT.25 Although we demonstrate only a 500-fold increase in concentration here, in principle, the concentration factor of this method would be higher if a larger microdroplet of a dilute solution was used. In this paper, the solute dependency of the final concentration in the microdroplets is still unclear and will be discussed in the following studies. This method not only enables the increase of signal intensity but also reduces the background signals caused by water-soluble biomolecules, which cannot be achieved by nonselective concentration methods. Therefore, this method is expected to increase the flexibility of the detection methods and the design of droplet-based microanalytical systems.

CONCLUSIONS Here, the selective concentration of a microdroplets’ contents by using spontaneous emulsification was proposed and demonstrated. In addition, to indicate the feasibility of this method for biochemical analyses, a B/F separation was demonstrated. We report the separation of water-soluble molecules in microdroplets for the first time. This method is expected to increase the flexibility of the design of droplet analytical processes and widen their applicability. ASSOCIATED CONTENT

* Supporting Information S

Analysis procedure of fluorescent images and experimental results without Span 80. This material is available free of charge via the Internet at http://pubs.acs.org.

■

REFERENCES

(1) Arora, A.; Simone, G.; Salieb-Beugelaar, G. B.; Kim, J. T.; Manz, A. Anal. Chem. 2010, 82, 4830−4847. (2) Salieb-Beugelaar, G. B.; Simone, G.; Arora, A.; Philippi, A.; Manz, A. Anal. Chem. 2010, 82, 4848−4864. (3) Song, H.; Chen, D. L.; Ismagilov, R. F. Angew. Chem., Int. Ed. 2006, 45, 7336−7356. (4) Huebner, A.; Sharma, S.; Srisa-Art, M.; Hollfelder, F.; Edel, J. B.; de Mello, A. J. Lab Chip 2008, 8, 1244−1254. (5) Kelly, B. T.; Baret, J.-C.; Taly, V.; Griffiths, A. D. Chem. Commun. 2007, 1773−1788. (6) Thorsen, T.; Roberts, R. W.; Arnold, F. H.; Quake, S. R. Phys. Rev. Lett. 2001, 86, 4163−4166. (7) Christopher, G. F.; Anna, S. L. J. Phys. D. Appl. Phys. 2007, 40, R319−R336. (8) Chen, D.; Du, W.; Liu, Y.; Liu, W.; Kuznetsov, A.; Mendez, F. E.; Philipson, L. H.; Ismagilov, R. F. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 16843−16848. (9) Takeuchi, S.; Garstecki, P.; Weibel, D. B.; Whitesides, G. M. Adv. Mater. 2005, 17, 1067−1072. (10) He, M.; Edgar, J. S.; Jeffries, G. D. M.; Lorenz, R. M.; Shelby, J. P.; Chiu, D. T. Anal. Chem. 2005, 77, 1539−1544. (11) Bremond, N.; Thiam, A.; Bibette, J. Phys. Rev. Lett. 2008, 100, 1−4. (12) Song, H.; Ismagilov, R. F. J. Am. Chem. Soc. 2003, 125, 14613− 14619. (13) Link, D.; Anna, S.; Weitz, D.; Stone, H. Phys. Rev. Lett. 2004, 92, 1−4. (14) Joensson, H. N.; Andersson Svahn, H. Angew. Chem., Int. Ed. 2012, 51, 12176−12192. (15) Zheng, B.; Roach, L. S.; Ismagilov, R. F. J. Am. Chem. Soc. 2003, 125, 11170−11171. (16) Brouzes, E.; Medkova, M.; Savenelli, N.; Marran, D.; Twardowski, M.; Hutchison, J. B.; Rothberg, J. M.; Link, D. R.; Perrimon, N.; Samuels, M. L. Proc. Natl. Acad. Sci. U.S.A. 2009, 106, 14195−14200. (17) Fallah-Araghi, A.; Baret, J.-C.; Ryckelynck, M.; Griffiths, A. D. Lab Chip 2012, 12, 882−891. (18) Sakakihara; Araki, S.; Iino, R.; Noji, H. Lab Chip 2010, 10, 3355−3362. (19) Pekin, D.; Skhiri, Y.; Baret, J.-C.; Le Corre, D.; Mazutis, L.; Ben Salem, C.; Millot, F.; El Harrak, A.; Hutchison, J. B.; Larson, J. W.; Link, D. R.; Laurent-Puig, P.; Griffiths, A. D.; Taly, V. Lab Chip 2011, 11, 2156−2166. (20) Cecchini, M. P.; Hong, J.; Lim, C.; Choo, J.; Albrecht, T.; de Mello, A. J.; Edel, J. B. Anal. Chem. 2011, 83, 3076−3081. (21) He, M.; Sun, C.; Chiu, D. T. Anal. Chem. 2004, 76, 1222−1227. (22) Sugaya, S.; Yamada, M.; Hori, A.; Seki, M. Biomicrofluidics 2013, 7, 54120. (23) Wu, T.; Hirata, K.; Suzuki, H.; Xiang, R.; Tang, Z.; Yomo, T. Appl. Phys. Lett. 2012, 101, 074108. (24) Bajpayee, A.; Edd, J. F.; Chang, A.; Toner, M. Anal. Chem. 2010, 82, 1288−1291. (25) López-Montilla, J. C.; Herrera-Morales, P. E.; Pandey, S.; Shah, D. O. J. Dispersion Sci. Technol. 2002, 23, 219−268. (26) Gonzalez-Ochoa, H.; Arauz-Lara, J. L. Langmuir 2007, 23, 5289−5291. (27) Baret, J.-C. Lab Chip 2012, 12, 422−433. (28) Hibara, A.; Iwayama, S.; Matsuoka, S.; Ueno, M.; Kikutani, Y.; Tokeshi, M.; Kitamori, T. Anal. Chem. 2005, 77, 943−947. (29) Sugiura, S.; Nakajima, M.; Iwamoto, S.; Seki, M. Langmuir 2001, 17, 5562−5566. (30) Borisover, M.; Graber, E. R.; Bercovich, F.; Gerstl, Z. Chemosphere 2001, 44, 1033−1040. (31) Yu, B.; Dong, C. Y.; So, P. T.; Blankschtein, D.; Langer, R. J. Invest. Dermatol. 2001, 117, 16−25. (32) Tani, H.; Kamidate, T.; Watanabe, H. J. Chromatogr. A 1997, 780, 229−241.

■ ■

Letter

AUTHOR INFORMATION

Corresponding Author

*Phone: +81-3-5734-2238. Fax: +81-3-5734-2238. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS A part of this work has been supported by KAKENHI (Grantin-Aid for Challenging Exploratory Research Grant 26620116). M.F. acknowledges Grant-in-Aid for JSPS Fellows. 3565

DOI: 10.1021/acs.analchem.5b00155 Anal. Chem. 2015, 87, 3562−3565