1479
Electrophoresis 2015, 36, 1479–1484
Alexandra LaLonde Maria F. Romero-Creel Blanca H. Lapizco-Encinas Microscale Bioseparations Laboratory, Biomedical Engineering Department, Rochester Institute of Technology, Rochester, NY, USA
Received July 10, 2014 Revised August 6, 2014 Accepted August 8, 2014
Short Communication
Assessment of cell viability after manipulation with insulator-based dielectrophoresis The effects of insulator-based DEP (iDEP) manipulation on cell viability were investigated by varying operating conditions and the shape of the insulating structures. Experiments were conducted with Escherichia coli, Bacillus subtilis, and Saccharomyces cerevisiae cells by varying the applied potential (300–1000 V), exposure time (1–4 min), and composition of the suspending medium (0–10% glucose); using devices made from polydimethylsiloxane. Cell viability was quantified employing Trypan blue staining protocols. The results illustrated a strong decrease in cell survival at higher applied electric potentials and exposure times; and an increase in cell viability obtained by increasing suspending medium osmolality. The composition and structure of the cell wall also played a major role on cell survival, where prokaryotic Gram-positive B. subtilis was the most resilient cell strain, while eukaryotic S. cerevisiae had the lowest survival rate. Due to the popularity of iDEP in applications with biological cells, characterizing how iDEP operating conditions affect cell viability is essential. Keywords: Cells / Cell viability / Dielectrophoresis / Electric field / Electrokinetics / Microfluidics DOI 10.1002/elps.201400331
Additional supporting information may be found in the online version of this article at the publisher’s web-site
There is a need for the development of inexpensive and portable devices capable of analyzing biological samples in an easy-to-use manner. Microfluidics has revolutionized the field of bioanalytical assessments, by offering the possibility of portable laboratories. The potential applications of bioanalytical microdevices span from food and water safety, environmental monitoring to point-of-care applications. Electrokinetics (EK) techniques, that is, electric field driven methods, have become one of the main pillars in microfluidics, due to their flexibility and simplicity in application. DEP is a leading EK technique that exploits particle polarization effects generated by a nonuniform electric field. DEP possesses great flexibility, since it can be used to manipulate neutral and charged particles with AC or DC electric fields. Depending on the par-
Correspondence: Associate Professor Blanca H. Lapizco-Encinas, Microscale Bioseparations Laboratory, Department of Biomedical Engineering, Rochester Institute of Technology, Institute Hall (Building 73), Room 3103, 160 Lomb Memorial Drive, Rochester, NY 14623, USA E-mail: [email protected]
Abbreviations: eDEP, electrode-based DEP; EK, electrokinetics; EOF, electroosmotic flow; iDEP, insulator-based dielectrophoresis; nDEP, negative dielectrophoresis C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
ticle’s dielectric properties, DEP can be positive or negative. Positive DEP is when particles with higher polarizability than the suspending medium are attracted toward the regions of high electric field gradient, the opposite behavior is negative DEP (nDEP). Traditionally, DEP had been carried out in devices with microelectrode arrays. DEP was discovered employing a rudimentary two-electrode system in 1951 [1]. Electrode-based DEP (eDEP) has been successfully employed for the manipulation of many bioparticles of interest, including macromolecules, viruses, bacteria, parasites, and mammalian cells [2]. However, there are some limitations with eDEP systems: electrodes are prone to fouling and complex fabrication processes may be expensive. Alternatively, insulator-based DEP (iDEP), first proposed by Masuda et al. [3], employs insulating structures (instead of electrodes) to generate nonuniform electric fields. This newer DEP method offers the advantages of requiring simpler device configurations while allowing for liquid and particle pumping by means of electroosmotic flow (EOF). Under DC or low-frequency conditions, biological cells exhibit nDEP behavior originated by the insulating nature of the cells cytoplasmic membrane. Numerous biological particles have been successfully
Colour Online: See the article online to view Figs. 1–3 in colour.
www.electrophoresis-journal.com
1480
A. LaLonde et al.
manipulated in iDEP systems: protein enrichment [4], DNA manipulation [5], virus and bacteria separation and sorting [6, 7], as well as the analysis of cancer cells in clinical applications [8]. Dielectrophoretic applications for bacterial and yeast cell detection, sorting, and separation have received significant attention. Rapid assessment of samples containing bacterial cells is critical for water and food safety, environmental analysis, and clinical tests for the prevention of bacterial infections. Traditional methods that rely on microbiological cultures can take from 24 to 72 h to produce results. In comparison, iDEP methods have the capability to separate and enrich a sample containing bacterial and yeast cells in minutes [9]. Numerous iDEP systems have been used for the analysis of samples containing microorganisms, application range from viability assessments [10, 11], to enrichment and concentration [12], separation of Gram-positive and Gramnegative bacterial cells [13], and strain and serotype discrimination [7,14]. Although many successful applications of iDEP for the manipulation of microbes (such as bacteria and yeast cells) have been reported, little attention has been given to the effects of high electric fields [15], Joule heating [16, 17], and pH changes [18] can have on cell viability. When a cell becomes electrically polarized, in the presence of an electric field under-low frequency conditions, the cell cytoplasm is shielded from the electric field. The cytoplasmic membrane withstands the full magnitude of the electric potential applied to the cell, causing electrical stress on the cell membrane [15]. Depending on the conditions and the duration of the exposure to the electric field, the electrical stress on the cytoplasmic membrane can lead to cell death. In the case of bacteria, which are prokaryotic cells, there are significant structural differences between Gram-negative and Gram-positive bacterial cell walls. These differences can translate to distinctive strength and survival rates after manipulation in iDEP systems [19]. Both cell types possess a cell wall and a cell membrane. The cell wall gives the bacteria shape and provides protection from osmotic lysis. In contrast, the cell membrane functions as a selective. Permeable barrier that controls the movement of substances in and out of the cells and prevents loss of essential components due to leakage [13, 20]. The cell wall of Gram-negative bacteria comprises several layers that include an outer membrane 7–8 nm thick, a thin layer 1–3 nm thick of peptidoglycan (polymer consisting of sugar and amino acids) and the cytoplasmic membrane. Gram-positive cells walls have a simpler structure that features a thick peptidoglycan multilayer (20– 80 nm thick), a cytoplasmic membrane, and no outer membrane [19, 21]. The cytoplasmic membrane in both types of bacteria is similar, composed of a lipid bilayer of phospholipids, glycolipids, and proteins. The main difference between Gram-negative and Gram-positive bacterial cell walls is the fraction of peptidoglycan that they contain, which is around 90% of the dry weight in Gram-positive compared to only 10% of Gram-negative cells. Within the cell wall, peptidoglycan is the layer mainly responsible for preserving cell integrity, providing structural strength, and counteracting changes in os C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Electrophoresis 2015, 36, 1479–1484
motic pressure [20, 22]. Yeast are eukaryotic fungi cells with walls composed of an outer layer of manoproteins, an inner layer of -1,3-glucan (fibrous polysaccharide), and 70to 100-nm-thick layer of chitin (polysaccharide of acetylglucosamine). Yeast cell walls are highly elastic and their mechanical strength is provided by the layer of -1,3-glucan and chitin. For Saccharomyces cerevisiae the cell wall makes up to 15–30% of the dry weight of the cell [23]. There have been a couple of studies that analyzed bacterial and yeast cells viability after dielectrophoretic manipulation, however, these reports were done for eDEP systems [15]. The present study analyzes the effect of iDEP manipulation on cell viability, where the influence of the magnitude of applied potential, exposure time, osmolality of the suspending medium, and device geometry on cell viability were assessed. Samples containing Escherichia coli, Bacillus subtilis, or S. cerevisiae cells were manipulated in devices with circular or diamond-shaped insulating posts. The electric potentials considered were 300, 500, or 1000 V and the exposures times were varied from 1, 2, to 4 min employing DI water, 5 or 10% glucose solutions as the suspending mediums in devices made from PDMS. Mathematical modeling with COMSOL Multiphysics was utilized to predict the magnitude of the gradient of the squared electric field (E2 ) to which the cells were exposed during iDEP manipulation. The results demonstrated, that under specific conditions, cell viability can be significantly decreased after iDEP manipulation, mainly due to direct damage to the cell membrane caused by the electric field combined with Joule heating and pH changes. As expected, increasing the applied electric potential and exposure time reduces cell survival rate, while increasing suspending medium osmolality improved survival rate. Employing sharper diamond geometries also lessens cell viability due to higher magnitude of E2 . Cell strain, which is directly related to the structure and composition of the cell wall and cytoplasmic membrane, also played a major role on cell viability after exposure to iDEP manipulation. The number of studies on dielectrophoretic-based manipulations for the analysis of samples containing biological cells has grown considerably in the last few years. In particular iDEP is an attractive technology that allows analyzing and processing samples, while eDEP is mainly used for analytical purposes only. Therefore, it is important to characterize how iDEP operating conditions affect the viability of the cells of interest. The present study provides examples of detailed quantification of cell viability under conventional iDEP operating conditions. In iDEP, cells are concentrated in devices with insulating structures, usually DC fields are used and EOF is present. Devices made from PDMS were employed in this study, since PDMS has a negative zeta potential; the resulting EOF direction, which dominates EK motion, is from the positive to the negative electrode (Fig. 1A and B). Particle capture in iDEP is achieved when DEP overcomes diffusion and EK motion. An image of trapped cells in an array of diamond-shaped posts is shown in Fig. 1C, where S. cerevisiae cells were captured by nDEP at 300 V after 2 min. The cells are forming a band prior to the constriction between the posts, the band location www.electrophoresis-journal.com
Microfluidics and Miniaturization
Electrophoresis 2015, 36, 1479–1484
1481
Table 1. Prediction of the values for E2 AVG for the two geometries at the three applied potentials employed in this study
Figure 1. Schematic representation of the microchannels employed in this study; (A) circular post array, (B) diamond post array. (C) Image of dielectrophoretic trapping of S. cerevisiae cells in a diamond post array at a potential of 300 V after an exposure time of 2 min.
depends on the relative magnitude of EK and DEP forces. COMSOL Multiphysics version 4.4 (COMSOL, Newton, MA, USA) with the AC/DC module was used to estimate the distribution of the gradient of the electric field. Further details on DEP theory and COMSOL model are included in the Supporting Information of this article and previous publications by our group [24, 25]. E. coli (ATCC 11775), B. subtilis (ATCC 6051), and S. cerevisiae (ATCC 9763) cells were cultured in appropriate medium following standard techniques. Samples of cells were prepared by washing the cells and resuspending them in DI water or glucose solution. The final cell concentrations employed for experimentation were 7 × 108 cells/mL. Cell dimensions were measured using bright-field microscopy, E. coli was 2.38 ± 0.32 m long and 0.96 ± 0.21 m wide; B. subtilis was 4.86 ± 0.41 long and 1.94 ± 0.19 m wide; and S. cerevisiae had a diameter of 6.3 ± 0.4 m. A high-voltage sequencer was used to apply DC electric potentials (HVS6000D, LabSmith, Livermore, CA, USA). Experiment visualization was performed with a microfluidic microscope (SVM340, LabSmith) and an inverted microscope (Axiovert 40 CFL, Carl Zeiss Microscopy, Thornwood, NY, USA). Experiments started with a clean microchannel that was filled with DI water (pH = 8 and m = 20 S/cm) or glucose solution (pH = 8 and m = 30 S/cm). A sample of 10 L containing the cells was introduced, platinum wire electrodes were placed and a DC electric potential was applied. After each experiment, a 50 L sample of cells that were exposed to the electric field was extracted from the channel outlet reservoir and 5 L of 0.4% Trypan blue solution was added to assess cell viability. The sample was analyzed using a hemocytometer and the viable and nonviable cells were counted. For the purpose of this study, cell viability was defined as a function of the membrane integrity, which was assessed by employing the cell capacity to uptake Trypan blue exclusion dye [26, 27]. C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Applied voltage (V)
Geometry
E2 AVG (V2 /m3 )
300 500 1000
Circular geometry
8.892 × 1012 2.470 × 1013 9.880 × 1013
300 500 1000
Diamond geometry
1.593 × 1013 4.427 × 1013 1.771 × 1014
The effect of post shape on the gradient of the square of the electric field (E2 ) was explored using COMSOL modeling (see Supporting Information). Figure 2 and Table 1 depict the average values of E2 obtained for circular and diamond-shaped post geometries at three different applied electric potentials. As it can be seen from the figure, the diamond geometry reaches higher E2 values (as depicted by the darker red color) and these higher values cover a larger region between the insulating posts when compared with the circular posts. Higher values of E2 covering larger regions within the post array means higher exposure of the cells to the effects of the electric field; thus, the diamond geometry could lead to more damaging conditions for the cells. It can also be noted that as the voltage increases, the value of E2 increases significantly, since E2 is of second order with the electric field. As observed from Table 1, the values of E2 AVG for the diamond geometry are 1.8 times greater than those obtained with circular posts; leading to approximately double the magnitude of the DEP forces exerted on cells, which can produce higher cell death. The effect of exposure time, voltage, and post geometry on cell viability was analyzed using S. cerevisiae cells, as shown in Fig. 3A and B. It can be observed that as the voltage and exposure time increased, the percentage of viable cells decreased. It can also be noted that the insulating posts geometry had an effect on cell viability. For example, with circular posts at 300 V for 4 min (Fig. 3A, green bar), about 49% of S. cerevisiae cells remained viable, while at the same conditions with the diamond post device (Fig. 3B, green bar) about 25% of S. cerevisiae cells were viable. The use of the diamond geometry resulted in around half (28%) the number of viable cells compared to the circular geometry. A similar trend was observed for the rest of the results. A threshold of viability was observed as the voltage and exposure time increased; once the voltage was increased past 500 V for 4 minutes, around the same low rate of viable cells (⬍25%) was obtained, regardless of the post geometry employed. In all cases, cell survival rates ⬍50% were obtained at applied potentials of 1000 V, regardless of exposure time and post geometry. This significant low cell viability at 1000 V is the result of the high E2 at which the cells were exposed, which ranges from 0.9 to 1.8 × 1014 V2 /m3 (Table 1). High values of E2 cause direct damage to the cell cytoplasmic membrane due to exposure to the electric field, combined with damage caused by www.electrophoresis-journal.com
1482
A. LaLonde et al.
Electrophoresis 2015, 36, 1479–1484
Figure 2. Simulations of the gradient distributions of the electric field squared (E2 ). Circular insulating post array at (A) 300 V, (B) 500 V, and (C) 1000 V. Diamond insulating post array at (D) 300 V, (E) 500 V and (F) 1000 V.
temperature gradients (originated from Joule heating) and pH variations across the microchannel. Joule heating has been studied for iDEP systems, where temperature variations above 50°C can be obtained. In some cases, systems can quickly reach 100°C at the end of the post array [16], which would significantly affect cells trapped in that location. Changes in pH have also been analyzed for iDEP systems, where electrode reactions produce a pH gradients lowering the pH of the suspending medium to values around 3.0 [18]. These highly acidic conditions can result in denaturation of the proteins in the cell wall and membrane, leading to loss of biological activity and compromising the integrity of the cytoplasmic membrane. The present study is focused on quantifying the decrease in cell viability under conventional DC-iDEP conditions, where the combination of electric field gradients, temperature, and pH changes are present and affect cell viability. Under the set of conditions studied here, viability of S. cerevisiae cells varied from 75 to 5% (Fig. 3A and B), demonstrating that operating conditions strongly influence cell viability. Increasing the osmolality of the suspending medium showed an increase on cell viability, as demonstrated using S. cerevisiae cells in a circular post device at 500 V (Fig. 3C). For example, when using DI water at 500 V for 4 min, about 25% of S. cerevisiae cells remained viable, compared to more than double (57%) obtained using a 10% glucose medium. A similar trend was observed for the rest of the results. It has been reported that early stages of apoptosis in mammalian cells can be reversible if the stress stimulus is removed and there is appropriate intervention such as nutrient feeding [28]. A similar situation might be observed for S. cerevisiae in this study. Cells begin to debilitate when harvested and introduced into the DI water; adding glucose to the media allows cells stay longer in optimal conditions and provides the nutrients that could help the cells to recover from minor damages. It is important to note that the media osmolality does not seem to have an effect on cell viability at a low exposure times. Around 70% of S. cerevisiae cells remained viable in all three media at 1 min; this suggests that some cells will be affected regardless of the suspending medium, but varying the osmolality of the medium will improve cell viability when cells are exposed to higher electric potentials for longer periods of time. These C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
results encourage the use of glucose solutions for iDEP cell manipulation, since glucose protects the cells from damage without significantly affecting medium conductivity. The effect of cell strain on viability employing E. coli, B. subtilis, and S. cerevisiae cells was analyzed at 500 V using the circular and diamond post devices. Figure 3D and E illustrate that B. subtilis remained the most resistant to cell death at 500 V at longer exposure times. As a Grampositive bacterium, B. subtilis contains a very thick peptidoglycan layer [19, 20], which provides significant strength and could explain its higher viability. The Gram-negative E. coli bacterium possesses a much thinner peptidoglycan layer, while S. cerevisiae, which was most susceptible to the electric field, does not contain peptidoglycan. These differences in cell wall structure and composition, including the presence or lack of a peptidoglycan layer, strongly influenced cell survival after iDEP manipulation. The survival rates were 77–46%, 75–35%, and 65–18%, for B. subtilis, E. coli, and S. cerevisiae, respectively. To highlight the effect of cell strain, B. subtilis cells remained twice more viable (56%) compared to S. cerevisiae cells (25%) in the circular post device after 4 min (Fig. 3D). Eukaryotic cells such as S. cerevisiae, which lack the structural strength provided by a peptidoglycan layer, are strongly affected after iDEP manipulation. Bacterial cells are much more resistant, and can preserve above 70% of viability if operating conditions are carefully selected. In conclusion, cell viability can be significantly affected after manipulation with iDEP. This study included bacterial cells (E. coli and B. subtilis) and yeast cells (S. cerevisiae). The parameters considered were: applied voltage, exposure time, geometry of insulating posts, and medium osmolality. Modeling and experiments results were in agreement, demonstrating that higher electric field gradients (diamond posts) resulted in lower cell viability. As the exposure time and applied voltage were increased, cell viability for all three strains decreased drastically. The results also demonstrated that increasing the osmolality of the medium, by adding glucose, doubled the percentage of viable cells. Cell wall composition had a major effect on the viability of cells, where the Gram-positive bacteria B. subtilis, which contains a thick peptidoglycan layer, had the highest survival rate, followed by the Gram-negative E. coli, which contains a thinner
www.electrophoresis-journal.com
Electrophoresis 2015, 36, 1479–1484
Microfluidics and Miniaturization
1483
Figure 3. Percentage of viable S. cerevisiae cells obtained by varying the electric potentials and exposure times using (A) circular insulating post device and (B) diamond insulating post device. (C) Percentage of viable S. cerevisiae obtained by varying suspending medium osmolality (DI water, 5% glucose, and 10% glucose) at 500 V using a circular posts. Percentage of viable cells of different strains (S. cerevisiae, E. coli, B. subtilis) at 500 V using (D) circular posts, and (E) diamond posts.
peptidoglycan layer. S. cerevisiae, which lacks peptidoglycan, had the lowest survival rate. These results, that quantify the viability of biological cells after manipulation with iDEP, are significant for applications that require fast response when handling biological cells, where iDEP could be the technique of choice.
The authors have declared no conflict of interest.
References [1] Pohl, H. A., J. Appl. Phys. 1951, 22, 869–871. [2] Pethig, R., Crit. Rev. Biotechnol. 1996, 16, 331–348.
The authors would like to acknowledge the financial support provided by the National Science Foundation (Award CBET1336160) and the Kate Gleason College of Engineering at Rochester Institute of Technology, through a start-up package to BHLE.
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
[3] Masuda, S., Washizu, M., Nanba, T., IEEE Trans. Ind. Appl. 1989, 25, 732–737. ´ [4] Lapizco-Encinas, B. H., Ozuna-Chacon, S., RitoPalomares, M., J. Chromatogr. A 2008, 1206, 45–51.
www.electrophoresis-journal.com
1484
A. LaLonde et al.
´ [5] Gallo-Villanueva, R. C., Rodr´ıguez-Lopez, C. E., D´ıaz-dela-Garza, R. I., Reyes-Betanzo, C., Lapizco-Encinas, B. H., Electrophoresis 2009, 30, 4195–4205. ´ [6] Moncada-Hernandez, H., Baylon-Cardiel, J. L., Perez´ Gonzalez, V. H., Lapizco-Encinas, B. H., Electrophoresis 2011, 32, 2502–2511. [7] Braff, W. A., Willner, D., Hugenholtz, P., Rabaey, K., Buie, C. R., PLoS One 2013, 8, e76751. [8] Salmanzadeh, A., Romero, L., Shafiee, H., GalloVillanueva, R. C., Stremler, M. A., Cramer, S. D., Davalos, R. V., Lab Chip 2012, 12, 182–189. ´ [9] Moncada-Hernandez, H., Lapizco-Encinas, B. H., Anal. Bioanal. Chem. 2010, 396, 1805–1816. ´ ´ [10] Gallo-Villanueva, R. C., Jesus-P erez, N. M., Mart´ınez´ Lopez, J. I., Pacheco, A., Lapizco-Encinas, B. H., Microfluid. Nanofluidics 2011, 10, 1305–1315.
Electrophoresis 2015, 36, 1479–1484
[16] Gallo-Villanueva, R. C., Sano, M. B., Lapizco-Encinas, B. H., Davalos, R., Electrophoresis 2014, 35, 352–361. [17] Kale, A., Patel, S., Hu, G., Xuan, X., Electrophoresis 2013, 34, 674–683. [18] Gencoglu, A., Camacho-Alanis, F., Nguyen, V. T., Nakano, A., Ros, A., Minerick, A. R., Electrophoresis 2011, 32, 2436–2447. [19] Beveridge, T. J., J. Bacteriol. 1999, 181, 4725–4733. [20] Demchick, P., Koch, A. L., J. Bacteriol. 1996, 178, 768–773. [21] Markx, G. H., Huang, Y., Zhou, X. F., Pethig, R., Microbiology-UK 1994, 140, 585–591. [22] Vollmer, W., Blanot, D., De Pedro, M. A., FEMS Microbiol. Rev. 2008, 32, 149–167. [23] Klis, F. M., Mol, P., Hellingwerf, K., Brul, S., FEMS Microbiol. Rev. 2002, 26, 239–256.
[11] Lapizco-Encinas, B. H., Simmons, B. A., Cummings, E. B., Fintschenko, Y., Anal. Chem. 2004, 76, 1571–1579.
[24] Gencoglu, A., Olney, D., LaLonde, A., Koppula, K. S., Lapizco-Encinas, B. H., Electrophoresis 2014, 35, 363–373.
[12] Lapizco-Encinas, B. H., Davalos, R., Simmons, B. A., Cummings, E. B., Fintschenko, Y., J. Microbiol. Methods 2005, 62, 317–326.
[25] LaLonde, A., Gencoglu, A., Romero-Creel, M. F., Koppula, K. S., Lapizco-Encinas, B. H., J. Chromatogr. A 2014, 1344, 99–108.
[13] Lapizco-Encinas, B. H., Simmons, B. A., Cummings, E. B., Fintschenko, Y., Electrophoresis 2004, 25, 1695–1704.
[26] Braasch, K., Nikolic-Jaric, M., Cabel, T., Salimi, E., Bridges, G. E., Thomson, D. J., Butler, M., Biotechnol. Bioeng. 2013, 110, 2902–2914.
[14] Jones, P., DeMichele, A., Kemp, L., Hayes, M., Anal. Bioanal. Chem. 2014, 406, 183–192.
[27] Garcia, M. M., Ohta, A. T., Walsh, T. J., Vittinghof, E., Lin, G. T., Wu, M. C., Lue, T. F., J. Urol. 2010, 184, 2466–2472.
[15] Pethig, R., Markx, G. H., Trends Biotechnol. 1997, 15, 426–432.
[28] Geske, F. J., Lieberman, R., Strange, R., Gerschenson, L. E., Cell Death Differ. 2001, 8.
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.electrophoresis-journal.com
Electrophoresis 2015, 36, 1479–1484
Alexandra LaLonde Maria F. Romero-Creel Blanca H. Lapizco-Encinas Microscale Bioseparations Laboratory, Biomedical Engineering Department, Rochester Institute of Technology, Rochester, NY, USA
Received July 10, 2014 Revised August 6, 2014 Accepted August 8, 2014
Short Communication
Assessment of cell viability after manipulation with insulator-based dielectrophoresis The effects of insulator-based DEP (iDEP) manipulation on cell viability were investigated by varying operating conditions and the shape of the insulating structures. Experiments were conducted with Escherichia coli, Bacillus subtilis, and Saccharomyces cerevisiae cells by varying the applied potential (300–1000 V), exposure time (1–4 min), and composition of the suspending medium (0–10% glucose); using devices made from polydimethylsiloxane. Cell viability was quantified employing Trypan blue staining protocols. The results illustrated a strong decrease in cell survival at higher applied electric potentials and exposure times; and an increase in cell viability obtained by increasing suspending medium osmolality. The composition and structure of the cell wall also played a major role on cell survival, where prokaryotic Gram-positive B. subtilis was the most resilient cell strain, while eukaryotic S. cerevisiae had the lowest survival rate. Due to the popularity of iDEP in applications with biological cells, characterizing how iDEP operating conditions affect cell viability is essential. Keywords: Cells / Cell viability / Dielectrophoresis / Electric field / Electrokinetics / Microfluidics DOI 10.1002/elps.201400331
Additional supporting information may be found in the online version of this article at the publisher’s web-site
There is a need for the development of inexpensive and portable devices capable of analyzing biological samples in an easy-to-use manner. Microfluidics has revolutionized the field of bioanalytical assessments, by offering the possibility of portable laboratories. The potential applications of bioanalytical microdevices span from food and water safety, environmental monitoring to point-of-care applications. Electrokinetics (EK) techniques, that is, electric field driven methods, have become one of the main pillars in microfluidics, due to their flexibility and simplicity in application. DEP is a leading EK technique that exploits particle polarization effects generated by a nonuniform electric field. DEP possesses great flexibility, since it can be used to manipulate neutral and charged particles with AC or DC electric fields. Depending on the par-
Correspondence: Associate Professor Blanca H. Lapizco-Encinas, Microscale Bioseparations Laboratory, Department of Biomedical Engineering, Rochester Institute of Technology, Institute Hall (Building 73), Room 3103, 160 Lomb Memorial Drive, Rochester, NY 14623, USA E-mail: [email protected]
Abbreviations: eDEP, electrode-based DEP; EK, electrokinetics; EOF, electroosmotic flow; iDEP, insulator-based dielectrophoresis; nDEP, negative dielectrophoresis C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
ticle’s dielectric properties, DEP can be positive or negative. Positive DEP is when particles with higher polarizability than the suspending medium are attracted toward the regions of high electric field gradient, the opposite behavior is negative DEP (nDEP). Traditionally, DEP had been carried out in devices with microelectrode arrays. DEP was discovered employing a rudimentary two-electrode system in 1951 [1]. Electrode-based DEP (eDEP) has been successfully employed for the manipulation of many bioparticles of interest, including macromolecules, viruses, bacteria, parasites, and mammalian cells [2]. However, there are some limitations with eDEP systems: electrodes are prone to fouling and complex fabrication processes may be expensive. Alternatively, insulator-based DEP (iDEP), first proposed by Masuda et al. [3], employs insulating structures (instead of electrodes) to generate nonuniform electric fields. This newer DEP method offers the advantages of requiring simpler device configurations while allowing for liquid and particle pumping by means of electroosmotic flow (EOF). Under DC or low-frequency conditions, biological cells exhibit nDEP behavior originated by the insulating nature of the cells cytoplasmic membrane. Numerous biological particles have been successfully
Colour Online: See the article online to view Figs. 1–3 in colour.
www.electrophoresis-journal.com
1480
A. LaLonde et al.
manipulated in iDEP systems: protein enrichment [4], DNA manipulation [5], virus and bacteria separation and sorting [6, 7], as well as the analysis of cancer cells in clinical applications [8]. Dielectrophoretic applications for bacterial and yeast cell detection, sorting, and separation have received significant attention. Rapid assessment of samples containing bacterial cells is critical for water and food safety, environmental analysis, and clinical tests for the prevention of bacterial infections. Traditional methods that rely on microbiological cultures can take from 24 to 72 h to produce results. In comparison, iDEP methods have the capability to separate and enrich a sample containing bacterial and yeast cells in minutes [9]. Numerous iDEP systems have been used for the analysis of samples containing microorganisms, application range from viability assessments [10, 11], to enrichment and concentration [12], separation of Gram-positive and Gramnegative bacterial cells [13], and strain and serotype discrimination [7,14]. Although many successful applications of iDEP for the manipulation of microbes (such as bacteria and yeast cells) have been reported, little attention has been given to the effects of high electric fields [15], Joule heating [16, 17], and pH changes [18] can have on cell viability. When a cell becomes electrically polarized, in the presence of an electric field under-low frequency conditions, the cell cytoplasm is shielded from the electric field. The cytoplasmic membrane withstands the full magnitude of the electric potential applied to the cell, causing electrical stress on the cell membrane [15]. Depending on the conditions and the duration of the exposure to the electric field, the electrical stress on the cytoplasmic membrane can lead to cell death. In the case of bacteria, which are prokaryotic cells, there are significant structural differences between Gram-negative and Gram-positive bacterial cell walls. These differences can translate to distinctive strength and survival rates after manipulation in iDEP systems [19]. Both cell types possess a cell wall and a cell membrane. The cell wall gives the bacteria shape and provides protection from osmotic lysis. In contrast, the cell membrane functions as a selective. Permeable barrier that controls the movement of substances in and out of the cells and prevents loss of essential components due to leakage [13, 20]. The cell wall of Gram-negative bacteria comprises several layers that include an outer membrane 7–8 nm thick, a thin layer 1–3 nm thick of peptidoglycan (polymer consisting of sugar and amino acids) and the cytoplasmic membrane. Gram-positive cells walls have a simpler structure that features a thick peptidoglycan multilayer (20– 80 nm thick), a cytoplasmic membrane, and no outer membrane [19, 21]. The cytoplasmic membrane in both types of bacteria is similar, composed of a lipid bilayer of phospholipids, glycolipids, and proteins. The main difference between Gram-negative and Gram-positive bacterial cell walls is the fraction of peptidoglycan that they contain, which is around 90% of the dry weight in Gram-positive compared to only 10% of Gram-negative cells. Within the cell wall, peptidoglycan is the layer mainly responsible for preserving cell integrity, providing structural strength, and counteracting changes in os C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Electrophoresis 2015, 36, 1479–1484
motic pressure [20, 22]. Yeast are eukaryotic fungi cells with walls composed of an outer layer of manoproteins, an inner layer of -1,3-glucan (fibrous polysaccharide), and 70to 100-nm-thick layer of chitin (polysaccharide of acetylglucosamine). Yeast cell walls are highly elastic and their mechanical strength is provided by the layer of -1,3-glucan and chitin. For Saccharomyces cerevisiae the cell wall makes up to 15–30% of the dry weight of the cell [23]. There have been a couple of studies that analyzed bacterial and yeast cells viability after dielectrophoretic manipulation, however, these reports were done for eDEP systems [15]. The present study analyzes the effect of iDEP manipulation on cell viability, where the influence of the magnitude of applied potential, exposure time, osmolality of the suspending medium, and device geometry on cell viability were assessed. Samples containing Escherichia coli, Bacillus subtilis, or S. cerevisiae cells were manipulated in devices with circular or diamond-shaped insulating posts. The electric potentials considered were 300, 500, or 1000 V and the exposures times were varied from 1, 2, to 4 min employing DI water, 5 or 10% glucose solutions as the suspending mediums in devices made from PDMS. Mathematical modeling with COMSOL Multiphysics was utilized to predict the magnitude of the gradient of the squared electric field (E2 ) to which the cells were exposed during iDEP manipulation. The results demonstrated, that under specific conditions, cell viability can be significantly decreased after iDEP manipulation, mainly due to direct damage to the cell membrane caused by the electric field combined with Joule heating and pH changes. As expected, increasing the applied electric potential and exposure time reduces cell survival rate, while increasing suspending medium osmolality improved survival rate. Employing sharper diamond geometries also lessens cell viability due to higher magnitude of E2 . Cell strain, which is directly related to the structure and composition of the cell wall and cytoplasmic membrane, also played a major role on cell viability after exposure to iDEP manipulation. The number of studies on dielectrophoretic-based manipulations for the analysis of samples containing biological cells has grown considerably in the last few years. In particular iDEP is an attractive technology that allows analyzing and processing samples, while eDEP is mainly used for analytical purposes only. Therefore, it is important to characterize how iDEP operating conditions affect the viability of the cells of interest. The present study provides examples of detailed quantification of cell viability under conventional iDEP operating conditions. In iDEP, cells are concentrated in devices with insulating structures, usually DC fields are used and EOF is present. Devices made from PDMS were employed in this study, since PDMS has a negative zeta potential; the resulting EOF direction, which dominates EK motion, is from the positive to the negative electrode (Fig. 1A and B). Particle capture in iDEP is achieved when DEP overcomes diffusion and EK motion. An image of trapped cells in an array of diamond-shaped posts is shown in Fig. 1C, where S. cerevisiae cells were captured by nDEP at 300 V after 2 min. The cells are forming a band prior to the constriction between the posts, the band location www.electrophoresis-journal.com
Microfluidics and Miniaturization
Electrophoresis 2015, 36, 1479–1484
1481
Table 1. Prediction of the values for E2 AVG for the two geometries at the three applied potentials employed in this study
Figure 1. Schematic representation of the microchannels employed in this study; (A) circular post array, (B) diamond post array. (C) Image of dielectrophoretic trapping of S. cerevisiae cells in a diamond post array at a potential of 300 V after an exposure time of 2 min.
depends on the relative magnitude of EK and DEP forces. COMSOL Multiphysics version 4.4 (COMSOL, Newton, MA, USA) with the AC/DC module was used to estimate the distribution of the gradient of the electric field. Further details on DEP theory and COMSOL model are included in the Supporting Information of this article and previous publications by our group [24, 25]. E. coli (ATCC 11775), B. subtilis (ATCC 6051), and S. cerevisiae (ATCC 9763) cells were cultured in appropriate medium following standard techniques. Samples of cells were prepared by washing the cells and resuspending them in DI water or glucose solution. The final cell concentrations employed for experimentation were 7 × 108 cells/mL. Cell dimensions were measured using bright-field microscopy, E. coli was 2.38 ± 0.32 m long and 0.96 ± 0.21 m wide; B. subtilis was 4.86 ± 0.41 long and 1.94 ± 0.19 m wide; and S. cerevisiae had a diameter of 6.3 ± 0.4 m. A high-voltage sequencer was used to apply DC electric potentials (HVS6000D, LabSmith, Livermore, CA, USA). Experiment visualization was performed with a microfluidic microscope (SVM340, LabSmith) and an inverted microscope (Axiovert 40 CFL, Carl Zeiss Microscopy, Thornwood, NY, USA). Experiments started with a clean microchannel that was filled with DI water (pH = 8 and m = 20 S/cm) or glucose solution (pH = 8 and m = 30 S/cm). A sample of 10 L containing the cells was introduced, platinum wire electrodes were placed and a DC electric potential was applied. After each experiment, a 50 L sample of cells that were exposed to the electric field was extracted from the channel outlet reservoir and 5 L of 0.4% Trypan blue solution was added to assess cell viability. The sample was analyzed using a hemocytometer and the viable and nonviable cells were counted. For the purpose of this study, cell viability was defined as a function of the membrane integrity, which was assessed by employing the cell capacity to uptake Trypan blue exclusion dye [26, 27]. C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Applied voltage (V)
Geometry
E2 AVG (V2 /m3 )
300 500 1000
Circular geometry
8.892 × 1012 2.470 × 1013 9.880 × 1013
300 500 1000
Diamond geometry
1.593 × 1013 4.427 × 1013 1.771 × 1014
The effect of post shape on the gradient of the square of the electric field (E2 ) was explored using COMSOL modeling (see Supporting Information). Figure 2 and Table 1 depict the average values of E2 obtained for circular and diamond-shaped post geometries at three different applied electric potentials. As it can be seen from the figure, the diamond geometry reaches higher E2 values (as depicted by the darker red color) and these higher values cover a larger region between the insulating posts when compared with the circular posts. Higher values of E2 covering larger regions within the post array means higher exposure of the cells to the effects of the electric field; thus, the diamond geometry could lead to more damaging conditions for the cells. It can also be noted that as the voltage increases, the value of E2 increases significantly, since E2 is of second order with the electric field. As observed from Table 1, the values of E2 AVG for the diamond geometry are 1.8 times greater than those obtained with circular posts; leading to approximately double the magnitude of the DEP forces exerted on cells, which can produce higher cell death. The effect of exposure time, voltage, and post geometry on cell viability was analyzed using S. cerevisiae cells, as shown in Fig. 3A and B. It can be observed that as the voltage and exposure time increased, the percentage of viable cells decreased. It can also be noted that the insulating posts geometry had an effect on cell viability. For example, with circular posts at 300 V for 4 min (Fig. 3A, green bar), about 49% of S. cerevisiae cells remained viable, while at the same conditions with the diamond post device (Fig. 3B, green bar) about 25% of S. cerevisiae cells were viable. The use of the diamond geometry resulted in around half (28%) the number of viable cells compared to the circular geometry. A similar trend was observed for the rest of the results. A threshold of viability was observed as the voltage and exposure time increased; once the voltage was increased past 500 V for 4 minutes, around the same low rate of viable cells (⬍25%) was obtained, regardless of the post geometry employed. In all cases, cell survival rates ⬍50% were obtained at applied potentials of 1000 V, regardless of exposure time and post geometry. This significant low cell viability at 1000 V is the result of the high E2 at which the cells were exposed, which ranges from 0.9 to 1.8 × 1014 V2 /m3 (Table 1). High values of E2 cause direct damage to the cell cytoplasmic membrane due to exposure to the electric field, combined with damage caused by www.electrophoresis-journal.com
1482
A. LaLonde et al.
Electrophoresis 2015, 36, 1479–1484
Figure 2. Simulations of the gradient distributions of the electric field squared (E2 ). Circular insulating post array at (A) 300 V, (B) 500 V, and (C) 1000 V. Diamond insulating post array at (D) 300 V, (E) 500 V and (F) 1000 V.
temperature gradients (originated from Joule heating) and pH variations across the microchannel. Joule heating has been studied for iDEP systems, where temperature variations above 50°C can be obtained. In some cases, systems can quickly reach 100°C at the end of the post array [16], which would significantly affect cells trapped in that location. Changes in pH have also been analyzed for iDEP systems, where electrode reactions produce a pH gradients lowering the pH of the suspending medium to values around 3.0 [18]. These highly acidic conditions can result in denaturation of the proteins in the cell wall and membrane, leading to loss of biological activity and compromising the integrity of the cytoplasmic membrane. The present study is focused on quantifying the decrease in cell viability under conventional DC-iDEP conditions, where the combination of electric field gradients, temperature, and pH changes are present and affect cell viability. Under the set of conditions studied here, viability of S. cerevisiae cells varied from 75 to 5% (Fig. 3A and B), demonstrating that operating conditions strongly influence cell viability. Increasing the osmolality of the suspending medium showed an increase on cell viability, as demonstrated using S. cerevisiae cells in a circular post device at 500 V (Fig. 3C). For example, when using DI water at 500 V for 4 min, about 25% of S. cerevisiae cells remained viable, compared to more than double (57%) obtained using a 10% glucose medium. A similar trend was observed for the rest of the results. It has been reported that early stages of apoptosis in mammalian cells can be reversible if the stress stimulus is removed and there is appropriate intervention such as nutrient feeding [28]. A similar situation might be observed for S. cerevisiae in this study. Cells begin to debilitate when harvested and introduced into the DI water; adding glucose to the media allows cells stay longer in optimal conditions and provides the nutrients that could help the cells to recover from minor damages. It is important to note that the media osmolality does not seem to have an effect on cell viability at a low exposure times. Around 70% of S. cerevisiae cells remained viable in all three media at 1 min; this suggests that some cells will be affected regardless of the suspending medium, but varying the osmolality of the medium will improve cell viability when cells are exposed to higher electric potentials for longer periods of time. These C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
results encourage the use of glucose solutions for iDEP cell manipulation, since glucose protects the cells from damage without significantly affecting medium conductivity. The effect of cell strain on viability employing E. coli, B. subtilis, and S. cerevisiae cells was analyzed at 500 V using the circular and diamond post devices. Figure 3D and E illustrate that B. subtilis remained the most resistant to cell death at 500 V at longer exposure times. As a Grampositive bacterium, B. subtilis contains a very thick peptidoglycan layer [19, 20], which provides significant strength and could explain its higher viability. The Gram-negative E. coli bacterium possesses a much thinner peptidoglycan layer, while S. cerevisiae, which was most susceptible to the electric field, does not contain peptidoglycan. These differences in cell wall structure and composition, including the presence or lack of a peptidoglycan layer, strongly influenced cell survival after iDEP manipulation. The survival rates were 77–46%, 75–35%, and 65–18%, for B. subtilis, E. coli, and S. cerevisiae, respectively. To highlight the effect of cell strain, B. subtilis cells remained twice more viable (56%) compared to S. cerevisiae cells (25%) in the circular post device after 4 min (Fig. 3D). Eukaryotic cells such as S. cerevisiae, which lack the structural strength provided by a peptidoglycan layer, are strongly affected after iDEP manipulation. Bacterial cells are much more resistant, and can preserve above 70% of viability if operating conditions are carefully selected. In conclusion, cell viability can be significantly affected after manipulation with iDEP. This study included bacterial cells (E. coli and B. subtilis) and yeast cells (S. cerevisiae). The parameters considered were: applied voltage, exposure time, geometry of insulating posts, and medium osmolality. Modeling and experiments results were in agreement, demonstrating that higher electric field gradients (diamond posts) resulted in lower cell viability. As the exposure time and applied voltage were increased, cell viability for all three strains decreased drastically. The results also demonstrated that increasing the osmolality of the medium, by adding glucose, doubled the percentage of viable cells. Cell wall composition had a major effect on the viability of cells, where the Gram-positive bacteria B. subtilis, which contains a thick peptidoglycan layer, had the highest survival rate, followed by the Gram-negative E. coli, which contains a thinner
www.electrophoresis-journal.com
Electrophoresis 2015, 36, 1479–1484
Microfluidics and Miniaturization
1483
Figure 3. Percentage of viable S. cerevisiae cells obtained by varying the electric potentials and exposure times using (A) circular insulating post device and (B) diamond insulating post device. (C) Percentage of viable S. cerevisiae obtained by varying suspending medium osmolality (DI water, 5% glucose, and 10% glucose) at 500 V using a circular posts. Percentage of viable cells of different strains (S. cerevisiae, E. coli, B. subtilis) at 500 V using (D) circular posts, and (E) diamond posts.
peptidoglycan layer. S. cerevisiae, which lacks peptidoglycan, had the lowest survival rate. These results, that quantify the viability of biological cells after manipulation with iDEP, are significant for applications that require fast response when handling biological cells, where iDEP could be the technique of choice.
The authors have declared no conflict of interest.
References [1] Pohl, H. A., J. Appl. Phys. 1951, 22, 869–871. [2] Pethig, R., Crit. Rev. Biotechnol. 1996, 16, 331–348.
The authors would like to acknowledge the financial support provided by the National Science Foundation (Award CBET1336160) and the Kate Gleason College of Engineering at Rochester Institute of Technology, through a start-up package to BHLE.
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
[3] Masuda, S., Washizu, M., Nanba, T., IEEE Trans. Ind. Appl. 1989, 25, 732–737. ´ [4] Lapizco-Encinas, B. H., Ozuna-Chacon, S., RitoPalomares, M., J. Chromatogr. A 2008, 1206, 45–51.
www.electrophoresis-journal.com
1484
A. LaLonde et al.
´ [5] Gallo-Villanueva, R. C., Rodr´ıguez-Lopez, C. E., D´ıaz-dela-Garza, R. I., Reyes-Betanzo, C., Lapizco-Encinas, B. H., Electrophoresis 2009, 30, 4195–4205. ´ [6] Moncada-Hernandez, H., Baylon-Cardiel, J. L., Perez´ Gonzalez, V. H., Lapizco-Encinas, B. H., Electrophoresis 2011, 32, 2502–2511. [7] Braff, W. A., Willner, D., Hugenholtz, P., Rabaey, K., Buie, C. R., PLoS One 2013, 8, e76751. [8] Salmanzadeh, A., Romero, L., Shafiee, H., GalloVillanueva, R. C., Stremler, M. A., Cramer, S. D., Davalos, R. V., Lab Chip 2012, 12, 182–189. ´ [9] Moncada-Hernandez, H., Lapizco-Encinas, B. H., Anal. Bioanal. Chem. 2010, 396, 1805–1816. ´ ´ [10] Gallo-Villanueva, R. C., Jesus-P erez, N. M., Mart´ınez´ Lopez, J. I., Pacheco, A., Lapizco-Encinas, B. H., Microfluid. Nanofluidics 2011, 10, 1305–1315.
Electrophoresis 2015, 36, 1479–1484
[16] Gallo-Villanueva, R. C., Sano, M. B., Lapizco-Encinas, B. H., Davalos, R., Electrophoresis 2014, 35, 352–361. [17] Kale, A., Patel, S., Hu, G., Xuan, X., Electrophoresis 2013, 34, 674–683. [18] Gencoglu, A., Camacho-Alanis, F., Nguyen, V. T., Nakano, A., Ros, A., Minerick, A. R., Electrophoresis 2011, 32, 2436–2447. [19] Beveridge, T. J., J. Bacteriol. 1999, 181, 4725–4733. [20] Demchick, P., Koch, A. L., J. Bacteriol. 1996, 178, 768–773. [21] Markx, G. H., Huang, Y., Zhou, X. F., Pethig, R., Microbiology-UK 1994, 140, 585–591. [22] Vollmer, W., Blanot, D., De Pedro, M. A., FEMS Microbiol. Rev. 2008, 32, 149–167. [23] Klis, F. M., Mol, P., Hellingwerf, K., Brul, S., FEMS Microbiol. Rev. 2002, 26, 239–256.
[11] Lapizco-Encinas, B. H., Simmons, B. A., Cummings, E. B., Fintschenko, Y., Anal. Chem. 2004, 76, 1571–1579.
[24] Gencoglu, A., Olney, D., LaLonde, A., Koppula, K. S., Lapizco-Encinas, B. H., Electrophoresis 2014, 35, 363–373.
[12] Lapizco-Encinas, B. H., Davalos, R., Simmons, B. A., Cummings, E. B., Fintschenko, Y., J. Microbiol. Methods 2005, 62, 317–326.
[25] LaLonde, A., Gencoglu, A., Romero-Creel, M. F., Koppula, K. S., Lapizco-Encinas, B. H., J. Chromatogr. A 2014, 1344, 99–108.
[13] Lapizco-Encinas, B. H., Simmons, B. A., Cummings, E. B., Fintschenko, Y., Electrophoresis 2004, 25, 1695–1704.
[26] Braasch, K., Nikolic-Jaric, M., Cabel, T., Salimi, E., Bridges, G. E., Thomson, D. J., Butler, M., Biotechnol. Bioeng. 2013, 110, 2902–2914.
[14] Jones, P., DeMichele, A., Kemp, L., Hayes, M., Anal. Bioanal. Chem. 2014, 406, 183–192.
[27] Garcia, M. M., Ohta, A. T., Walsh, T. J., Vittinghof, E., Lin, G. T., Wu, M. C., Lue, T. F., J. Urol. 2010, 184, 2466–2472.
[15] Pethig, R., Markx, G. H., Trends Biotechnol. 1997, 15, 426–432.
[28] Geske, F. J., Lieberman, R., Strange, R., Gerschenson, L. E., Cell Death Differ. 2001, 8.
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.electrophoresis-journal.com
