Article pubs.acs.org/ac
Digital Microfluidic Approach for Efficient Electroporation with High Productivity: Transgene Expression of Microalgae without Cell Wall Removal Do Jin Im,†,‡ Su-Nam Jeong,†,§ Byeong Sun Yoo,†,§ Bolam Kim,∥ Dong-Pyo Kim,*,§,∥ Won-Joong Jeong,⊥ and In Seok Kang§ ‡
Department of Chemical Engineering, Pukyong National University, 365 Sinseon-ro, Nam-gu, Busan 608-739, South Korea Department of Chemical Engineering and ∥Department of Interdisciplinary Bioscience and Bioengineering, Pohang University of Science and Technology (POSTECH), San 31, Hyoja-Dong, Nam-Gu, Pohang, Gyeongbuk 790-784, South Korea ⊥ Sustainable Bioresource Research Center, Korea Research Institute of Bioscience and Biotechnology (KRIBB), 125 Gawhak-ro, Yuseong-gu, DaeJeon 305-806, South Korea §
S Supporting Information *
ABSTRACT: A unique digital microfluidic electroporation (EP) system successfully demonstrates higher transgene expression than that of conventional techniques, in addition to reliable productivity and feasible integrated processes. By systematic investigations into the effects of the droplet EP conditions for a wild-type microalgae, 1 order of magnitude higher transgene expression is accomplished without cell wall removal over the conventional bulk EP system. In addition, the newly proposed droplet EP method by a droplet contact charging phenomena shows a great potential for the integration of EP processes and on-chip cell culture providing easy controllability of each process. Finally, the implications of the accomplishments and future directions for development of the proposed technology are discussed.
I
distortion, and sample contamination.13 Furthermore, macroscale bulk EP systems suffer from low cell viability (due to the use of relatively high voltages) and relatively low transfection efficiency. In particular, the transfection efficiency of microalgae (2 kV/ cm). Although there was no direct connection between the two electrodes by the droplet, very small currents (several tens of nA, bottom right of Figure 5a) flowed at the short moment of droplet charging, which is believed to have EP effects on the cells inside the droplet. After a certain period of time of bouncing (from a minute to several hours), the processed droplet was gathered using a syringe. Similar to the static droplet EP, there are mainly two parameters for the bouncing droplet EP: applied voltage and application time. The applied voltage range (500−1000 V) was determined on the basis of the typical range of ECD actuation (2−4 kV/cm).35,52 Because we cannot separately investigate pulse duration and number, as is possible in the static EP, only the total electric field application time was changed from a minute to several hours. Although the applied voltage was quite high, the cell viabilities of all the cases were very good even at a 4 h application time under 800 V (87 ± 9%) on account of extremely small current flows, as shown in Figure 5. For the given application time, the transgene expression efficiency had a linear dependency on the applied voltage (Figure 5b). However, for the given applied voltage, the efficiency depended
Figure 4. YFP transgene expression of wild-type algae (cc-125) by the static droplet EP without cell wall removal. (a−c) In each graph, the reference case is 48 V, n = 8, and 50 ms, i.e., other conditions remained the same as the reference case, as indicated in the graphs. (a) Effect of applied voltage. (b) Effect of the number of pulses. (c) Effect of pulse duration. (d) Comparison with the bulk EP. Except for the applied voltage (bulk EP 192 V/4 mm, static EP 48 V/1 mm), other conditions were the same (n = 8, 100 ms). The efficiency was 0.7 ± 0.4% for the control group (with no electric field), 2.2 ± 1.1% for the bulk EP, and 20.7 ± 6.2% for the static EP, respectively. The error bars represent standard deviation.
application time (pulse number and duration) show a steep change at extreme conditions. Furthermore, even for the same extreme field application time (8 × 100 ms = 16 × 50 ms = 800 ms), the increase in pulse duration (100 ms, 20.7 ± 6.2%) was slightly more efficient than the increase in pulse number (n = 16, 15.4 ± 4.2%), although the difference was not statistically significant (the P value of the t test is 0.287) and the overall dependencies showed a pattern similar to that shown in Figure 4b,c. When it comes to cell viability, longer pulse duration (100 ms) was also efficient. Therefore, for wild-type algae (cc-125) transgene expression, increasing the pulse duration was more efficient than increasing the voltage or pulse number. Here, it should be pointed out that the static droplet EP yields a remarkably high efficiency up to 21% (the 100 ms case in Figure 4c), which is outstandingly improved over the previous work.1 Because we skipped the cell wall removal step, this outstanding efficiency shows the effectiveness of the proposed droplet EP method and its usefulness for directly transforming the cell wall strain without any additional pretreatment. In addition, the CFP transgene expression (Figure S4 of the SI) showed several times higher efficiency than bulk EP system, although the efficiency of CFP (7.8 ± 2.4%) is rather lower compared to that of YFP (20.7 ± 6.2%), presumably due to the larger size of the CFP gene plasmid DNA. Compared with the bulk EP conducted under equivalent conditions, the cell viability of both cases was nearly identical, with little difference, while the static droplet EP showed 1 order of magnitude higher efficiency (2.2% for bulk and 20.7% for static droplet EP), as shown in Figure 4d. The applied voltage in the bulk system is four times greater due to a longer distance between electrodes. Due to this high voltage and 100× larger electrode cross sectional area, the electric current in the bulk system is several hundred times higher (several amperes). This high current also makes problems such as Joule heating and gas bubble formation, which have negative effects on transgene expression.17 The electrochemical reaction at the anode decreases the pH, resulting in cell necrosis. On the other E
DOI: 10.1021/acs.analchem.5b00725 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
It is believed that the bouncing EP has an EP mechanism similar to that of the electrospray that involves ejection of charged droplets under a high electric field without direct connection to the counter electrode and used only for the transfection of adherent mammalian cells.55,56 As shown in Figures 2a and 5a, the applied electric field of the bouncing EP is about 10× higher than that of the static EP. But the current flow of the bouncing EP is much smaller than that of the static EP. When a droplet is charged, most charges will be on the surface of the droplet. Therefore, cells near the surface of the droplet will be affected by those charges. In addition, due to the circulating flow inside the moving droplet, the location of cells will be continuously changed. Therefore, as the bouncing motion of the droplet continues, more and more cells can be affected by the surface charges. This explains the positive correlation of EP effects on bouncing time (Figure 5c). Because this positive correlation holds only for sufficiently high electric field (500 vs 800 V results of Figure 5c), the basic mechanism of the bouncing EP is similar to that of the static EP rather than chemical or other effects. Meanwhile, having almost no contact with solid electrode surfaces is a strong point of the bouncing droplet EP, which enables low contamination EP and cell culture inside a waterin-oil droplet. By simply adjusting the applied voltage, we can easily select the function of the ECD actuation, whether we utilize it for the EP of cells or just use it as a droplet manipulation for cell culture inside the droplet with minimal cell damage from the long-term use of strong electric field. This selective feature provides an easy and efficient way for the combination of genetic engineering and culture of cells using droplet microfluidic technology. In order to demonstrate the potential features for on-chip genetic engineering and cell culture, an on-chip cell culture followed by a bouncing EP was conducted as illustrated in Figure 6. The nine droplets were consecutively dispensed and processed in parallel. After a 1 h operation in a bouncing EP system with 800 V, the whole fluidic component containing the nine droplets was disassembled from the base and was put inside an incubation chamber for a 24 h on-chip culture without any medium exchange. To address potential problems with no culture medium exchange and a highly concentrated droplet environment on cell transgene expression and viability, an additional control group of droplet for an on-chip culture was prepared by employing the same conditions with no application of electric field. The transgene expression efficiencies of the onchip culture (both the control 1.0 ± 0.5% and experimental 6.1 ± 3.6%) were doubled with insignificant change in viability (the P value of the t test is 0.59 for the control groups and 0.39 for the experimental groups, respectively), as shown in Figure 6c, when compared with an off-chip culture (the control 0.4 ± 0.4% and experimental 3.1 ± 1.7%). The higher efficiency seemed to be due to an increased DNA concentration by longterm confinement in a small droplet. The proposed ECD-based water-in-oil droplet cell handling approach has great potential for future cell culture applications. One prominent application is a 3-dimensional (3D) cell culture technique that is one step closer to an in vivo cell growth environment. Culturing cells inside a small hanging droplet has been proposed as a new approach for a 3D cell culture method for tissue engineering,57,58 and the ECD-based system can be utilized in this field. An additional advantage of the proposed system over previous microfluidic or droplet cell culture systems is that the transformation function can be easily
Figure 5. YFP transgene expression of wild-type algae (cc-125) by the bouncing droplet EP without cell wall removal. (a) Schematic illustration of the bouncing EP and an experimental demonstration. A constant voltage (800 V) was applied between the two bottom electrodes, and each time the droplet contacted the electrode, a tiny current peak signal (around 20 nA) was detected by an electrometer. (b) Effects of applied voltage under a constant application time (1, 10 min). (c) Effects of application time under a constant applied voltage (500, 800 V). The error bars represent standard deviation.
on the bouncing time to a different extent (Figure 5c): for a low voltage (500 V), the efficiency was barely affected by the application time, whereas for a higher voltage (800 V), it changed linearly with time. This implies that there is a threshold electric field for the functioning of the bouncing droplet EP. In addition, under a lower electric field (500 V), the continuous droplet bouncing had little effect on the viability, even for a 4 h application time (94.6 ± 4.5%). Therefore, we have controllability for the bouncing EP depending on the objective of ECD actuation. Compared with the static droplet EP, the transgene expression efficiency of the bouncing droplet EP was relatively low; however, it is interesting to perform the same experiments with the bouncing EP because of its unique advantages such as controllability on EP function, low contamination risk, and flexible applicability for cell culture inside a droplet. Although the efficiency of the bouncing EP was relatively low, it was comparable to bulk EP. For sufficiently high voltage (1 kV), even under a 1 min application time, the efficiency (2.5 ± 1.9%) was higher than that of bulk EP (2.2 ± 1.1%). When the application time was long enough (4 h) with moderate voltage (800 V), the efficiency was much higher (5.8 ± 2.6%) than that of bulk EP. When it comes to speed and efficiency, the static droplet EP is superior to the bouncing droplet EP, but the bouncing droplet EP has beneficial features for low contamination and cell culture applications. F
DOI: 10.1021/acs.analchem.5b00725 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
(the bouncing droplet EP) had lower transgene expression efficiency than the static droplet EP, it showed great potential for the integration of the EP process and on-chip cell culture providing easy controllability (voltage control) of each process. In the future, the proposed digital EP unit can be promoted to a fully automated digital electroporation microfluidic system by developing an automated on-demand droplet dispensing and splitting unit.
■
ASSOCIATED CONTENT
* Supporting Information S
The details of ECD chip fabrication procedure and circuit connections; Transgene expression results for wall-less algal cell line cc-503; Information on a plasmid DNA used in the experiments; Results of CrCFP transgene expression; Results of control experiments; A short video demonstrating digital microfluidic control of droplets and static EP operation. The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.5b00725.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Tel.: +82-54-279-2272. Fax: +82-54-279-5598. Author Contributions †
These authors contributed equally to this work (D.J.I., S.-N.J., and B.S.Y.). Notes
The authors declare no competing financial interest.
■
Figure 6. Parallel processing of the bouncing EP and on-chip cell culture demonstration for YFP transgene expression. (a) Through the holes in the top cover, nine droplets (1 μL each) were consecutively dispensed by a micropipette and moved to each position by digital control of the polarities of the bottom electrodes. After the bouncing EP process, for off-chip culture, the processed droplets were collected by syringe. For the on-chip culture, the processed droplets remained with no collection. (b) Disassembly of the fluidic component containing droplets from the base, for on-chip culture inside an incubation chamber. (c) Comparison of the results from the off-chip and on-chip culture. For the on-chip culture control, nine DNA/cell mixture droplets were dispensed in an oil bath with no application of electric field. After the EP process, the two fluidic components were put inside an incubation chamber for 24 h. The error bars represent standard deviation.
ACKNOWLEDGMENTS This research was supported by the grant NRF2014K2A2A2000944, NRF-2014M1A8A1074941, NRF2015R1D1A3A01019112, and the grant No. 2008-0061983. The plasmid DNA CrYFP and the empty vector were provided by Prof. Y.S. Lee (POSTECH).
■
REFERENCES
(1) Qu, B.; Eu, Y. J.; Jeong, W. J.; Kim, D. P. Lab Chip 2012, 12, 4483−4488. (2) Guo, S. L.; Zhao, X. Q.; Tang, Y.; Wan, C.; Alam, M. A.; Ho, S. H.; Bai, F. W.; Chang, J. S. J. Biotechnol. 2013, 163, 61−68. (3) Gimpel, J. A.; Specht, E. A.; Georgianna, D. R.; Mayfield, S. P. Curr. Opin. Chem. Biol. 2013, 17, 489−495. (4) Basiouni, S.; Fuhrmann, H.; J, S. Biotechniques 2012, 1−4. (5) Gong, Y.; Hu, H.; Gao, Y.; Xu, X.; Gao, H. J. Ind. Microbiol. Biotechnol. 2011, 38, 1879−1890. (6) Wu-Scharf, D.; Jeong, B. R.; Zhang, C. M.; Cerutti, H. Science 2000, 290, 1159−1162. (7) Neupert, J.; Karcher, D.; Bock, R. Plant J. 2009, 57, 1140−1150. (8) Shimogawara, K.; Fujiwara, S.; Grossman, A.; Usuda, H. Genetics 1998, 148, 1821−1828. (9) Dev, S. B.; Rabussay, D. P.; Widera, G.; Hofmann, G. A. IEEE Trans. Plasma Sci. 2000, 28, 206−223. (10) Weaver, J. C.; Chizmadzhev, Y. A. Bioelectrochem. Bioenerg. 1996, 41, 135−160. (11) Gehl, J. Acta Physiol. Scand. 2003, 177, 437−447. (12) Kim, J. A.; Cho, K. C.; Shin, M. S.; Lee, W. G.; Jung, N. C.; Chung, C. I.; Chang, J. K. Biosens. Bioelectron. 2008, 23, 1353−1360. (13) Lee, W. G.; Demirci, U.; Khademhosseini, A. Integr. Biol. 2009, 1, 242−251. (14) Fox, M. B.; Esveld, D. C.; Valero, A.; Luttge, R.; Mastwijk, H. C.; Bartels, P. V.; van den Berg, A.; Boom, R. M. Anal. Bioanal. Chem. 2006, 385, 474−485.
implemented just by adjusting the actuation voltage with the addition of the desired DNA to the cell suspension. If automated on-demand dispensing, medium exchange, and sample gathering units are integrated into the proposed system, its usefulness will be much increased.
■
CONCLUSIONS We successfully demonstrated a unique digital microfluidic electroporation system based on ECD with much higher transgene expression than that of conventional techniques as well as the advantages of comparable productivity and convenience. The high transgene expression efficiency of the droplet EP system is mainly attributed to the small size which results in a much lower electric current. By systematic investigations into the effects of the droplet EP conditions for a wild-type algae, greater transgene expression of YFP gene (21%) was accomplished without cell wall removal over the conventional bulk EP system (2.2%). Although the newly proposed droplet EP mechanism by the discrete ECD actuation G
DOI: 10.1021/acs.analchem.5b00725 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
(49) Harrls, E. H. The Chlamydomonas Sourcebook: A Comprehensive Guide to Biology and Laboratory Use; Academic Press: San Diego, 1989. (50) Lim, J. M.; Ahn, J. W.; Hwangbo, K.; Choi, D. W.; Park, E. J.; Hwang, M. S.; Liu, J. R.; Jeong, W. J. Plant Biotechnol. Rep. 2013, 7, 407−414. (51) Yordanova, Z. P.; Woltering, E. J.; Kapchina-Toteva, V. M.; Iakimova, E. T. Ann. Bot. 2013, 111, 191−205. (52) Im, D. J.; Yoo, B. S.; Ahn, M. M.; Moon, D.; Kang, I. S. Anal. Chem. 2013, 85, 4038−4044. (53) Yoshihara, C.; Inoue, K.; Schichnes, D.; Ruzin, S.; Inwood, W.; Kustu, S. Mol. Plant 2008, 1, 1007−1020. (54) Teissié, J.; Rols, M. P. Biophys. J. 1993, 65, 409−413. (55) Okubo, Y.; Ikemoto, K.; Koike, K.; Tsutsui, C.; Sakata, I.; Takei, O.; Adachi, A.; Sakai, T. Angew. Chem., Int. Ed. 2008, 47, 1429−1431. (56) Ikemoto, K.; Sakata, I.; Sakai, T. Sci. Rep. 2012, 2, 289. (57) Seiler, A. E. M.; Spielmann, H. Nat. Protocols 2011, 6, 961−978. (58) Tung, Y.-C.; Hsiao, A. Y.; Allen, S. G.; Torisawa, Y.-s.; Ho, M.; Takayama, S. Analyst 2011, 136, 473−478.
(15) Kim, J.; Hwang, I.; Britain, D.; Chung, T. D.; Sun, Y.; Kim, D. H. Lab Chip 2011, 11, 3941−3948. (16) Geng, T.; Lu, C. Lab Chip 2013, 13, 3803−3821. (17) Wang, S. N.; Lee, L. J. Biomicrofluidics 2013, 7, 011301. (18) Khine, M.; Lau, A.; Ionescu-Zanetti, C.; Seo, J.; Lee, L. P. Lab Chip 2005, 5, 38−43. (19) Boukany, P. E.; Morss, A.; Liao, W. C.; Henslee, B.; Jung, H. C.; Zhang, X. L.; Yu, B.; Wang, X. M.; Wu, Y.; Li, L.; Gao, K. L.; Hu, X.; Zhao, X.; Hemminger, O.; Lu, W.; Lafyatis, G. P.; Lee, L. J. Nat. Nanotechnol. 2011, 6, 747−754. (20) Valero, A.; Post, J. N.; van Nieuwkasteele, J. W.; ter Braak, P. M.; Kruijer, W.; van den Berg, A. Lab Chip 2008, 8, 62−67. (21) Wang, S. N.; Zhang, X. L.; Wang, W. X.; Lee, L. J. Anal. Chem. 2009, 81, 4414−4421. (22) Wei, Z. W.; Zhao, D. Y.; Li, X. M.; Wu, M. X.; Wang, W.; Huang, H.; Wang, X. X.; Du, Q.; Liang, Z. C.; Li, Z. H. Anal. Chem. 2011, 83, 5881−5887. (23) Adamo, A.; Arione, A.; Sharei, A.; Jensen, K. F. Anal. Chem. 2013, 85, 1637−1641. (24) Zhan, Y. H.; Cao, Z. N.; Bao, N.; Li, J. B.; Wang, J.; Geng, T.; Lin, H.; Lu, C. J. Controlled Release 2012, 160, 570−576. (25) Geng, T.; Zhan, Y. H.; Wang, H. Y.; Witting, S. R.; Cornetta, K. G.; Lu, C. J. Controlled Release 2010, 144, 91−100. (26) Geng, T.; Zhan, Y. H.; Wang, J.; Lu, C. Nat. Protoc. 2011, 6, 1192−1208. (27) Zhan, Y. H.; Wang, J.; Bao, N.; Lu, C. Anal. Chem. 2009, 81, 2027−2031. (28) Luo, C. X.; Yang, X. J.; Fu, O.; Sun, M. H.; Ouyang, Q.; Chen, Y.; Ji, H. Electrophoresis 2006, 27, 1977−1983. (29) Jebrail, M. J.; Bartsch, M. S.; Patel, K. D. Lab Chip 2012, 12, 2452−2463. (30) Elvira, K. S.; Solvas, X. C. I.; Wootton, R. C. R.; deMello, A. J. Nat. Chem. 2013, 5, 905−915. (31) Link, D. R.; Grasland-Mongrain, E.; Duri, A.; Sarrazin, F.; Cheng, Z.; Cristobal, G.; Marquez, M.; Weitz, D. A. Angew. Chem., Int. Ed. 2006, 45, 2556−2560. (32) Niu, X.; Gielen, F.; DeMello, A. J.; Edel, J. B. Anal. Chem. 2009, 81, 7321−7325. (33) Takinoue, M.; Atsumi, Y.; Yoshikawa, K. Appl. Phys. Lett. 2010, 96, 104105. (34) Wang, W.; Yang, C.; Liu, Y.; Li, C. M. Lab Chip 2010, 10, 559− 562. (35) Im, D. J.; Noh, J.; Moon, D.; Kang, I. S. Anal. Chem. 2011, 83, 5168−5174. (36) Ahn, B.; Lee, K.; Panchapakesan, R.; Oh, K. W. Biomicrofluidics 2011, 5, 024113. (37) Im, D. J.; Noh, J.; Yi, N. W.; Park, J.; Kang, I. S. Biomicrofluidics 2011, 5, 044112. (38) Choi, K.; Im, M.; Choi, J. M.; Choi, Y. K. Microfluid. Nanofluid. 2012, 12, 821−827. (39) Im, D. J.; Ahn, M. M.; Yoo, B. S.; Moon, D.; Lee, D. W.; Kang, I. S. Langmuir 2012, 28, 11656−11661. (40) Lee, D. W.; Im, D. J.; Kang, I. S. Appl. Phys. Lett. 2012, 100, 221602. (41) Lee, C. P.; Chang, H. C.; Wei, Z. H. Appl. Phys. Lett. 2012, 101, 014103. (42) Lee, D. W.; Im, D. J.; Kang, I. S. Langmuir 2013, 29, 1875− 1884. (43) Lee, D. W.; Im, D. J.; Kang, I. S. J. Phys. Chem. C 2013, 117, 3426−3430. (44) Zhou, H. B.; Yao, S. H. Lab Chip 2013, 13, 962−969. (45) Drews, A. M.; Lee, H. Y.; Bishop, K. J. M. Lab Chip 2013, 13, 4295−4298. (46) Ahn, M. M.; Im, D. J.; Kang, I. S. Analyst 2013, 138, 7362− 7368. (47) Choi, D.; Lee, H.; Im, D. J.; Kang, I. S.; Lim, G.; Kim, D. S.; Kang, K. H. Sci. Rep. 2013, 3, 2037. (48) Ahn, M. M.; Im, D. J.; Kim, J. G.; Lee, D. W.; Kang, I. S. J. Phys. Chem. Lett. 2014, 5, 3021−3025. H
DOI: 10.1021/acs.analchem.5b00725 Anal. Chem. XXXX, XXX, XXX−XXX
