RESEARCH ARTICLE pubs.acs.org/acscombsci
“One Cell-One Well”: A New Approach to Inkjet Printing Single Cell Microarrays )
)
)
Albert R. Liberski,†, Joseph T. Delaney, Jr.,‡,†, and Ulrich S. Schubert*,†,‡,§, †
)
Laboratory of Organic and Macromolecular Chemistry (IOMC), Friedrich-Schiller-University Jena, Humboldtstrasse 10, Jena D-07743, Germany ‡ Laboratory of Macromolecular Chemistry and Nanoscience, Eindhoven University of Technology, P.O. Box 513, Eindhoven 5600 MB, The Netherlands § Center for Nanoscience (CeNS), Ludwigs-Maximilians-University M€unchen, Amalienstrasse 54, M€unchen D-80799, Germany Dutch Polymer Institute (DPI), Post Office Box 902, Eindhoven 5600 AX, The Netherlands
bS Supporting Information ABSTRACT: A new approach to prepare arrays of sessile droplets of living single cell cultures using a liquid hydrophobic barrier prevents the samples from dehydrating, and allows for spatially addressable arrays for statistical quantitative single cell studies. By carefully advancing a thin layer of mineral oil on the substrate over the droplets during the printing, dehydration of the droplets can be prevented, and the vitality of the cells can be maintained. The net result of this confluence of submerged cell culturing and inkjet printing is facile access to spatially addressable arrays of isolated single cells on surfaces. Such single cell arrays may be particularly useful as high-throughput tools in the rapidly emerging “omics” fields of cell biology. KEYWORDS: one cell-one well, inkjet printing, single cell microarrays, microdroplets
’ INTRODUCTION High-throughput single-cell measurements of cellular responses are of great importance for a variety of applications, including drug testing, toxicology, and basic cell biology,1,2 with single cell biochip applications emerging as a rapidly expanding field of research.1,3,4 Moreover, single-cell analysis approaches have become clinically important for cytotoxicity assays (comet assay),5,6 and the demand for more practical ways of preparing arrays of single living cells will only increase with the advent of the emerging field of cellomics.7-9 Motivated by this, we have developed a facile new approach that allows for high density arrays of isolated single cell cultures to be prepared on unmodified surfaces. In recent years, a wide variety of approaches to produce arrays of single cells have been investigated. One way in which this problem has been addressed is microcontact printing.10-18 This approach can be used with a wide range of surfaces, but because of the inherent nature of contact-based deposition techniques, there are some noteworthy risks and limitations, namely, issues such as sample-to-sample cross-contamination, and the inherent variability of how much material is deposited at any given spot. Moreover, with stamps, if the stamp itself is uneven, not all of the surface will be contacted, and rapid prototyping is proscribed without the use of geometrically highly specialized stamps. Additionally, there is the risk of mechanically damaging the cells during the stamping process. Beyond these initial problems, there remains the issue of the loss of material on drying. During the deposition r 2010 American Chemical Society
process, the droplets loaded with the cell culture medium have an intrinsically high surface-to-volume ratio; as they are exposed to the environment, they have ample opportunity to dehydrate, reducing the cell viability.19 Another approach is the use of a patterned surface, where the surface is modified to include arrays of microwells.2,20,21 In such an approach, surfaces are first patterned with wells that are exactly large enough to house an individual cell, and the surface is then seeded with a cell suspension; ideally, the cells are then automatically deposited into the wells. In some cases, this can be promoted by patterning the surface using PEGylation to create spots where cells will adhere selectively, surrounded by biophobic areas where adhesion is proscribed.22 Alternatively, a positive resist can be used, where cell seeding is directed to specific locations by creating patterns of biophilic materials.23-26 Perhaps the biggest detraction from this approach is the need for highly specialized substrates, which are not yet readily commercially available, and inherently limit experiment design to predetermined layouts, and may not be suitable for all cells. Yet another general category of reported approaches involves the use of sandwiched networks of twodimensionally oriented microfluidic pathways, in which liquid cell suspensions are passed through, and cells are either entrapped in Received: November 10, 2010 Revised: November 15, 2010 Published: December 23, 2010 190
dx.doi.org/10.1021/co100061c | ACS Comb. Sci. 2011, 13, 190–195
ACS Combinatorial Science
RESEARCH ARTICLE
Figure 1. Preparation of a standing well cell plate, (a) microscope glass slide with sealing frame, (b) inkjet printing of cell culture medium (50 drops per spot) and simultaneous spreading paraffin oil for evaporation prevention, (c) inkjet printing of L929 cells (5 drops per spot), (d and e) schematic and real appearance of prepared well plate (720 wells), (f) examples of wells containing cell, (g) appearance of spots containing L929 cells after 7 h incubation, (h) manual screen of microarray using bright field microscopy (10 objective) allowing for the position selection of wells occupied by cells of interest, (i-k) wells containing single cells were identified; wells containing L929 cells exhibiting (from top to bottom) linear, triangular, and circular morphology.
narrow passage ways or stopped by small fluid blockades.27,28 This leads to microarrays, but requires the custom fabrication of microfluidic chips, and the positions are not entirely spatially addressable. Also, since the cells are sandwiched inside a chip, they are not immediately, directly accessible. Inkjet printing of cellular microarrays has also been reported,25,29,30 including variations of drop-on-demand technology employing thermal printing,31 as well as piezoelectric actuation.32,33 Starting with the “cytoscription” work of Klebe et al.,34,35 the application of inkjet as a tool for cell culturing has increased exponentially over the past 22 years, with particular attention on multicellular culturing for tissue engineering. However, in terms of making spatially addressable microarrays of single cells, there have been some technical limitations that need to be overcome; namely, since single cell droplets have small volumes (1-100 pL) and inherently high surface-to-volume ratios, the challenge of maintaining arrays of individual drops on a surface without these dehydrating makes generating arrays problematic. In the work by Demirci and Montesano,36 the use of inkjet printing as a tool for the preparation of single cell arrays is discussed in great detail, including a discussion of the probability issues with printing dilute cell solutions and a discussion of the application of Poisson distributions as a model for the probability of printing a given number of cells as a function of aliquot size. In their study, the focus of the work was investigating the processing conditions that lead to the maximum number of spots with one or more cells, and did not have single cell printing as the exclusive goal. Moreover, while post-processing viability of various cell types was examined by collecting suspensions immediately, there was no direct measurement of viability of cells printed onto a surface, and no quantification of the size of the arrays they attempted. Though being important work, the remaining practical details of how to maintain a large array of isolated single cells as viable and separate remained unresolved. In a recent article,37 the use of mineral oil and silicone oil as a protective layer for standing aqueous droplets was reported. In this approach, macroscopic droplets (1 μL) of cell culture were deposited onto a surface that had been submerged in various hydrophobic liquid media, and was shown to protect the droplets from evaporation. The advantages of this approach are apparent,
but the lack of an automated dispensing technique, coupled with the large size of the droplets used, make this approach difficult to apply as a tool for generating single cell culture arrays. The use of mineral oil or similar materials as a barrier for protecting submerged cell cultures is well-known, and well-established.38-43 Water-in-oil microdroplets in microfluidic systems have also been widely reported, where individual droplets containing single cells are kept separate in microfluidic channels that are otherwise filled with oil.42,44 The use of dual liquid phase inkjet printing has also been reported, in the form of aqueous biphasic systems, where each of the layers is loaded with a high concentration of a non-volatile component that forces the two solutions to remain separate.30 Moreover, the combination of arrays of macrodroplets (g1 μL) of cell cultures submerged in mineral oil has also been recently reported.37 However, the combination of inkjet printing with waterin-oil cell culturing has not been reported, but offers a practical means of overcoming evaporation-related challenges of living cell microarray fabrication, and to miniaturize sample size further into the nanoliter/subnanoliter range of volumes. Because of the sample volumes involved, inkjet is particularly well-suited as a dispensing technique for delivering single living cells to predetermined locations.
’ EXPERIMENTAL SECTION Some special considerations were required to effectively prepare arrays containing isolated living single cells in spatially addressable patterns by inkjet printing. In general, starting from a clean, standard borosilicate microscope slide, an array of droplets of culture medium (RPMA, 50 drops 3 spot-1) was printed in the pattern of the desired array (Figure 1). Using the Microdrop Autodrop system (settings: 96 V, 28 μs PW, 100 Hz), equipped with an ADK-501 glass pipet (nozzle inner diameter: 70 μm). This was the first active step in preparing a well-ordered array that involves printing droplets of sterilized culture medium into the desired pattern. This was found to be necessary because the wetted culture medium droplets ensured that the aqueous phase had good adsorptive contact with the glass surface; with good water-to-glass contact established, a stationary target for subsequent cell addition was realized. As the cell culture medium was being printed, a small 191
dx.doi.org/10.1021/co100061c |ACS Comb. Sci. 2011, 13, 190–195
ACS Combinatorial Science
RESEARCH ARTICLE
amount of mineral oil (
