Biological Implications of Polymeric Microdevices ... - ACS Publications

Nov 10, 2009 - Georgette B. Salieb-Beugelaar, Giuseppina Simone, Arun Arora, Anja Philippi and Andreas Manz . Latest Developments in Microfluidic Cell...
2 downloads 0 Views 3MB Size
Anal. Chem. 2009, 81, 9828–9833

Biological Implications of Polymeric Microdevices for Live Cell Assays Donald Wlodkowic,† Shannon Faley,† Joanna Skommer,‡ Dagmara McGuinness,§ and Jonathan M Cooper*,† Department of Electronics and Electrical Engineering, The Bioelectronics Research Centre, University of Glasgow, Glasgow, U.K. G12 8LT, Queen’s Medical Research Institute, Edinburgh, U.K., and Women’s Reproductive Health Research Center, Medical Center North, Vanderbilt University, Nashville, TN Lab-on-a-chip technologies have the potential to deliver significant technological advances in modern biomedicine, through the ability to provide appropriate low-cost microenvironments for screening cells. However, to date, few studies have investigated the suitability of poly(dimethylsiloxane) (PDMS) for live cell culture. Here, we describe an inexpensive method for production of reusable, optical-grade PDMS microculture chips which provide a static and self-contained microwell system analogous to conventional polystyrene multiwell plates. We use these structures to probe the effects of PDMS upon live cell culture bioassays, using time-lapse fluorescence imaging to explore the toxicity of the substrate. We use three model systems to explore the efficacy of the microstructured devices: (i) live cell culture, (ii) adenoviral gene delivery to mammalian cells, and (iii) gravity enforced formation of multicellular tumor spheroids (MCTS). Results show that PDMS is nontoxic to cells, as their viability and growth characteristic in PDMS-based platforms is comparable to that of their polystyrene counterparts. In vitro cell and tissue culture underpin much of modern biomedicine, including many pharmaceutical studies, putting an increasing demand on real-time (dynamic) fluorescence-based live imaging at the single-cell level as the method of choice to study signaling pathways.1,2 High-resolution imaging, however, often requires expensive optical grade microplates or other proprietary cell culture systems. Alternatively, novel lab-on-a-chip (LOC) technologies can be utilized to combine miniaturized perfusion culture with low cost sample and cell consumables.3-6 In this respect, poly(dimethylsiloxane) (PDMS) is an optically clear, gas permeable polymeric substance widely used in both cell-based and lab-on-a-chip applications. It has been largely assumed that * Corresponding author. E-mail: [email protected]. † University of Glasgow. ‡ Queen’s Medical Research Institute. § Vanderbilt University. (1) Svahn, H. A.; van den Berg, A. Lab Chip 2007, 7, 544–546. (2) Sims, C. E.; Allbritton, N. L. Lab Chip 2007, 7, 423–440. (3) El-Ali, J.; Sorger, P. K.; Jensen, K. F. Nature 2006, 442, 403–411. (4) Manz, A.; Dittrich, P. S. Nat. Drug Discovery 2006, 5, 210–218. (5) Yin, H.; Zhang, X.; Pattrick, N.; Klauke, N.; Cordingley, H.; Haswell, S. J.; Cooper, J. M. Anal. Chem. 2008, 80, 179–185. (6) Regehr, K. J.; Domenech, R. M.; Koepsel, J. T.; Carver, K. C.; Ellison-Zelski, S. J.; Murphy, W. L.; Schuler, L. A.; Alarid, E. T.; Beebe, D. J. Lab Chip 2009, 9, 2132-2139.

9828

Analytical Chemistry, Vol. 81, No. 23, December 1, 2009

PDMS is biologically inert and well suited for cell-based assays, although only recently is this being carefully explored.5,6 In this paper, we describe the use of PDMS molding to produce reusable, optical-grade microculture chips. We compare the use of these PDMS-based cell culture devices with conventional polystyrene vessels. Despite the rapidly growing usage of PDMS microdevices for biological applications, as stated, surprisingly few studies have been performed to examine their suitability for such uses.5,6 In order to explore this, we used PDMS bonded to glass coverslips, which is a static and self-contained system that can support mammalian cell growth for up to 7 days. We provide evidence that such reusable microculture devices permit prolonged cell culture, high-content fluorescence imaging on live cells, dynamic pharmacological profiling of drug activities, and low cost viral gene delivery, as well as the rapid generation of multicellular tumor spheroids (MCTS). Importantly, we show that PDMS is well-suited for this wide variety of long-term cell culture assays based upon cell viability and growth assays. MATERIALS AND METHODS Fabrication of Microculture Devices. Microculture devices were produced in biologically compatible elastomer poly(dimethylsiloxane) (PDMS, Sylgard 184, Dow Corning).7 PDMS was mixed at a 10:1 (w/w) ratio of elastomer base to curing agent and degassed at 40 Torr to remove any residual air bubbles. PDMS was then poured on silicon wafers to achieve 3-5 mm thickness and cured thermally at 70 °C for 2 h. Cured devices were mechanically diced and removed from the silicone wafers. Microculture wells were bored using a stainless steel punch hole. To fabricate a self-sealing cover, a 1 mm layer of PDMS was cast and cured, as a lid, using the general methods as described above. PDMS devices were reversibly sealed to glass coverslips using a standard conformal bonding principle.7 Where necessary, microfluidic connects were formed by mechanically boring apertures with a steel punch. Cell Culture and Labeling with Fluorescent Probes. The origin, characteristics, and culture of human tumor cell lines U2OS and MCF-7 were as previously described.8 Human leukemic K562 and U937 and breast cancer MCF-7 cell lines were obtained from ATCC (Manassas, VA). Human osteosarcoma U2OS cell line was (7) Whitesides, G. M. Nature 2006, 42, 368–372. (8) McDonald, J. C.; Whitesides, G. M. Acc. Chem. Res. 2002, 35, 491–499. 10.1021/ac902010s CCC: $40.75  2009 American Chemical Society Published on Web 11/10/2009

a kind gift from Dr. Simon Wilkinson (Beatson Intitute for Cancer Research, U.K.). All cell cultures were maintained at 37 °C in a 5% CO2 humidified atmosphere. Human K562 and U937 cells were cultured in a complete RPMI 1640 medium whereas MCF-7 and U2OS cells were cultured in a DMEM medium. All media were supplemented with 10% fetal bovine serum (FBS). During experiments, cells were always in asynchronous and exponential phase of growth. Adenovirus containing LC3-GFP construct was a kind gift from Dr. Cindy Miranti (Van Andel Institute, Grand Rapids, Michigan) and was propagated using HEK293AD cell line, directly derived from HEK293 cell line to improve cell adherence. Briefly, cells are infected with LC3-GFP adenovirus, and upon plaque formation, infected cells are collected, washed with PBS, resuspended in 10 mM TRIS pH 8.1, and lysed by three cycles of freezing (dry ice)-thawing (37 °C). Cell debris was collected by centrifugation, and aliquots of supernatant with viral particles were stored at -80 °C. Labeling with fluorescent probes SYTO 16 (250 nM, 15 min at 37 °C), TMRM (150 nM, 15 min at 37 °C), and Hoechst 33342 (1 µM, 15 min at 37 °C) from Molecular Probes, Eugene, OR, was described in detail earlier.8 To uncouple the mitochondrial trans-membrane potential, U2OS cells were exposed to 100 µM of FCCP. Loss of mitochondrial membrane potential was assessed using tetramethylrhodamine methyl ester (TMRM; Molecular Probes) probe as previously described.9 Adenoviral Gene Delivery. To investigate the formation of autophagosomes, human osteosarcoma U2OS and breast cancer MCF-7 cells were seeded in microculture chambers and infected with adenovirus containing LC3-GFP construct. Following overnight infection, cells were washed and exposed to brefeldin A (BFA) to induce autophagosome formation. Live U2OS/LC3-GFP and MCF-7/LC3-GFP cells were counterstained with Hoechst 33341 and propidium idodide (PI). 3D Spheroid Formation. Multicellular tumor spheroids (MCTS) were generated by a modified hanging droplet method which facilitates cell-to-cell attachment and formation of 3D structures. Briefly, 4 × 106 of MCF-7 cells was loaded into microwells, and chips were sealed with a layer of PDMS. Chips were reversed so that hanging droplets were formed inside microwells and maintained by capillary and cohesion forces. Chips were cultured at 37 °C, and the size of MCTS was assessed using bright-field microscopy at 24 h intervals, measured using a microscopic standard (calibrated digital ruler). Live Cell Imaging. For experiments in which an extremely short time was expected between the addition of a substance and the cell response, it is desirable to be able to add compounds directly into the culture chambers without removing the device from the imaging station. To provide a simple experimental setup, we adapted our self-contained microculture chip-based system by addition of polytetrafluoroethylene (PTFE) fluidic lines. PTFE lines were mounted into a PDMS lid of the device and connected to syringes controlled manually and/or mounted onto computerized syringe pumps. Using automated time-resolved imaging, we

performed a quantitative on-chip analysis of dissipation of mitochondrial transmembrane potential (∆Ψm) in real time. Flow Cytometry. Analysis of cell cycle profile (DNA content) was carried out using flow cytometry as described elsewhere.10 Briefly, cells were harvested from PDMS chips, and polystyrene plates were washed twice with PBS, fixed with 70% ethanol, and incubated with RNase A (Sigma) and PI (1 µg/mL; Sigma) for 1 h at 37 °C. The cell cycle profile was estimated on a BD FACSCalibur analyzer equipped with a 15 mW argon-ion laser (BD Biosciences). A linear amplification scale for the 488 nm excitation line and emission at 610 nm (BP) was used. Acquisition was performed in 1024 channels (resolution scale) using CellQuest Pro software (BD Biosciences). A typical run used a sample with 5 000 cells so that a triplicate experiment involved ca. 15 000 cells. Data analysis was performed using Summit v3.1 (Dako Cytomation, Fort Collins, CO) and an open access WinMDI ver.2.8 (http://facs.scripps.edu/software.html) software. Data Collection and Statistical Analysis. Fluorescence images were acquired using a motorized Zeiss Axiovert 200M epifluorescence microscope. Data analysis and presentation was performed using ImageJ (available at http://rsb.info.nih.gov/ij/) and CellProfiler software (available at www.cellprofiler.org). The student’s t test was applied for comparison between groups with significance set at p < 0.05. All control measurements are provided in detail in the figure legends, where appropriate, but in general, they involved making direct comparisons between the PDMS devices and standard polystyrene microtiter cell culture plate measurements.

(9) Wlodkowic, D.; Skommer, J.; Faley, S.; Darzynkiewicz, Z.; Cooper, J. M. Exp. Cell Res. 2009, 315, 1706–1714.

(10) Wlodkowic, D.; Skommer, J.; Pelkonen, J. Cytometry, Part A 2007, 71, 61– 72.

RESULTS AND DISCUSSION Microculture System Design and Performance. Elastomeric polymers such as PDMS are reported as being inexpensive, nontoxic, and permeable to gases under physiological conditions, making them particularly suitable for inexpensive production of chip-based devices for biomedical studies.10 In this paper, we now demonstrate the feasibility of the use of a PDMS static culture device that can support long-term growth of both adherent and suspension human cell lines utilizing an array of cell culture microchambers with an effective culture area (ECA) of 6.5 mm2 and culture volume (ECV) of up to 10 µL (Figure 1A,B). To evaluate influence of microenvironmental conditions on cell physiology, human osteosarcoma U2OS cells were loaded and cultured under static conditions in a complete DMEM medium. Chips were subsequently sealed and placed in a humidified atmosphere at 37 °C. Cell proliferation was assessed by counting nuclei stained with a cell permeable fluorescent Hoechst 33343 probe (1.0 µM) at 24 h intervals (Figure 1C,D). At the same time, cell viability was assessed by addition of 0.25 µg/mL PI to cell culture medium (Figure 1E). After 48 h of culture, the cell cycle profile was also evaluated by flow cytometry (Figure 1F). As inferred from the results in Figure 1, no adverse effects on cell viability or cell growth retardation were observed during microscale static cell culture in PDMS devices. Importantly, results also corresponded to those obtained in a conventional cell culture performed on a 96-well polystyrene plate, R2 g 0.98 for p < 0.01 in a Pearson and Lee linear correlation test, indicating that the

Analytical Chemistry, Vol. 81, No. 23, December 1, 2009

9829

Figure 1. PDMS-based, static microculture devices allow long-term culture of human cells. Control measurements involved making direct comparisons between the PDMS devices and standard polystyrene microtiter cell culture plate. (A) Overview of chip design. Microculture chambers are 2 mm in diameter and hold approximately 10-15 µL of cell culture medium. Multiple cell culture chambers can be simultaneously analyzed on a motorized microscope stage. A reusable PDMS lid prevented evaporative water loss and maintained chip sterility. (B) Magnification of single microculture chamber, mechanically formed in PDMS. (C) Cell growth of human osteosarcoma U2OS cells continuously maintained on-chip under static conditions for up to 5 days. (D) Proliferative capacity of U2OS cells cultured on-chip and on a standard 96-well cell culture plate. (E) Comparison between cell viability during a 5-day cell culture of U2OS cells. Note comparable values for microscale and conventional cell culture were achieved. (F) Cell cycle profile (DNA content) after 48 h of static cell culture on-chip compared with the polystyrene cell culture plate.

microenvironment during static cell culture does not affect normal cell physiology. Similar results were also obtained for suspension cells such as human leukemic K562 and U937 cell lines (R2 g 0.97 for p < 0.01 in a Pearson and Lee linear correlation test; Figure 1 in the Supporting Information). Electrostatically bonded to glass, these polymeric devices maintained adequate integrity for up to 7 days of continuous cell culture with no observable cross-contamination between adjacent culture chambers (Figure 2 in the Supporting Information). We found that the PDMS devices were particularly suitable for high-resolution multicolor imaging on both live cells and fixed cells and dynamic analysis of drug induced cytotox9830

Analytical Chemistry, Vol. 81, No. 23, December 1, 2009

icity (Figure 2). Considering the small volumes used, the evaporation from open-access chip-based devices might be considered a particular challenge during long-term experiments. Our devices, when used in conjunction with a conformally bonded polymeric lid, resulted in water loss by evaporation being always reduced to