[1 Arrays of Transfected Mammalian Cells for High Content Screening Microscopy

By Holger Erfle and Rainer Pepperkok


In this chapter we describe protocols for reverse transfection to generate mammalian cell arrays for systematic gene knock-downs by RNAi or knock-ins by ectopic cDNA expression. The method is suitable for high content screening microscopy at a high spatial and temporal resolution allowing even time-lapse analysis of hundreds of samples in parallel.


The information of complete genome sequences and the identification and systematic cloning of human cDNAs provide the challenging opportunity to analyze the complexity of biological processes on a large scale. For this purpose high-throughput techniques such as protein analysis by mass spectrometry (Smith et al., 2001) or expression and transcription profiling by protein or DNA microarrays (DeRisi et al., 1996; MacBeath et al., 2000; Schena et al., 1995) have been developed and successfully applied to diverse biological questions. An elegant high-throughput method allowing parallel analysis of gene function in intact living cells has been recently introduced by Ziauddin and Sabatini (Ziauddin et al., 2001). In this method expression plasmids encoding for example, GFP-tagged proteins, are printed together with the appropriate transfection reagents in a gelatin matrix at defined locations on glass slides. Tissue culture cells are subsequently plated on these slides resulting in clusters of living cells expressing the respective cDNAs at each location. The approach is called ''reverse transfection,'' as in comparison to conventional transfection the order of addition of DNA and cells is reversed.

The method, originally introduced for ectopically expressing genes, has recently been adapted to transfections of siRNAs to knock down target genes of interest for functional analysis (Elbashir et al., 2001; Erfle et al., 2004; Kumar et al., 2003; Mousses et al., 2003). Several other extensions of the original transfected cell array protocol have been introduced such as the arraying of smaller grids of siRNAs or cDNAs into the wells of 96 well plates (Mishina et al., 2004), enabling ''ultra-high''-throughput applications like drug screens in living cells. The method has also been extended to perform transfections of human primary (Yoshikawa et al., 2004) and non-adherent cells (Kato et al., 2004).

Recent advances in automated fluorescence scanning microscopy and image processing (see e.g., Liebel et al., 2003; Starkuviene et al., 2004) allow now rapid analysis of transfected cell arrays in large scale screening applications. In the following we describe the method of reverse transfection on cell arrays as we use it in our laboratory to examine gene function by RNAi or overexpression of plasmid DNAs with high content screening microscopy.


The method comprises five individual steps (see Fig. 1), including the preparation of the transfection solutions, followed by their spotting onto a cell substrate (e.g., Lab-Tek culture dishes, Nalge Nunc International, Rochester, NY), plating of the cells onto the arrays of spotted transfection solutions, preparation of the transfected cells for functional analysis, and finally the analysis of transfected cells by high content screening microscopy.

Although we describe the method for adherent tissue culture cells on noncoated chambered cover-glass tissue culture dishes, Lab-Tek chambered cover-glass, the very same protocols also work well with different cell substrates such as MatTek (MatTek, Ashland, MA) culture dishes or glass slides. The protocol has been equally successful for transfections of synthetic siRNAs or plasmid cDNAs.

As transfection reagent we use Effectene (Qiagen, Hilden, Germany) or Lipofectamine 2000 (Invitrogen, La Jolla, CA), which give optimal transfection efficiencies for both siRNAs and plasmid DNAs in MCF7, HeLa, COS7L, or HEK 293 cells.

Preparation of Transfection Solutions

The siRNA (plasmid cDNA) gelatin transfection solutions are prepared in 384-well plates (Nalge-Nunc).

1. Add 1 ul of the respective siRNA stock solution (20 uM in RNA dilution buffer as supplied by the manufacturer) to each well. For plasmid transfections 1 ul of plasmid DNA at a stock concentration of 500 ng/ul is added.

2. Add 7.5 ul EC buffer (EC Buffer is part of the Effectene Transfection kit, Qiagen) containing 0.2 M sucrose and mix thoroughly by pipetting three times up and down.

3. Incubate the mixture for 10 min at room temperature.

Fig. 1. The five steps to produce arrays of transfected mammalian cells for high content screening microscopy. 1. Preparation of the transfection solutions on an automated liquid handler. 2. Spotting of the transfection solutions with a spotting Robot, for example, ChipWriter Compact on a cell substrate, for example, Lab-Tek tissue culture dishes. 3. Plating of the cells on dishes with dried transfection solutions. 4. Preparation of samples for functional analysis, for example, immunostaining. 5. Analysis of samples by high content screening microscopy.

Fig. 1. The five steps to produce arrays of transfected mammalian cells for high content screening microscopy. 1. Preparation of the transfection solutions on an automated liquid handler. 2. Spotting of the transfection solutions with a spotting Robot, for example, ChipWriter Compact on a cell substrate, for example, Lab-Tek tissue culture dishes. 3. Plating of the cells on dishes with dried transfection solutions. 4. Preparation of samples for functional analysis, for example, immunostaining. 5. Analysis of samples by high content screening microscopy.

4. Add 4.5 of the Effectene transfection reagent (Qiagen).

5. Incubate for 10 min at room temperature.

6. Add 7.25 of 0.08% gelatin (G-9391, Sigma-Aldrich, St. Louis, MO) containing 3.5 x 10~4% fibronectin (Sigma-Aldrich). The final solution is now ready for the spotting process.


It is important to mix the transfection components just prior to the spotting, to achieve a high reproducibility in transfection efficiency. The optimal incubation times and amounts of transfection reagent are determined empirically for different transfection reagents. However, for optimizing the transfection mix for transfection reagents different from Effectene, the protocol described previously is a good starting point for optimization by simply replacing Effectene with equal amounts of the alternative transfection reagent. The EC buffer from the Effectene kit can be replaced by water without significant loss of transfection efficiencies, when tranfection reagents different from Effectene are used. The presence of sucrose in the EC buffer reduces the loss in transfection efficiencies when the dried arrays are stored prior to their use (see below). Sucrose also facilitates considerably the transfer of the siRNA (cDNA)-gelatin transfection solution to the substrate during the spotting procedure.

The presence of fibronectin in the gelatin solution increases cell adherence to the spot region and reduces the migration of transfected cells away from it.

In order to retrieve the spot regions and to highlight successfully transfected cells for siRNA transfections a Cy3-labeled DNA oligonucleotide is used as a transfection marker. In this case 0.5 ^l of a 40 ^M marker solution is included in step 1 of the protocol above resulting in a total oligonucleotide volume of 1.5 ^l (1 ^l siRNA plus 0.5 ^l Cy3 labeled oligonucleotide).

With the protocol described previously it is possible to cotransfect plas-mid cDNA and siRNA. In this case 1 ^l siRNA plus 1 ^l plasmid cDNA are added in step 1, resulting in a total oligo-nucleotide volume of 2 ^l.

Spotting the Transfection Solution on Lab-Tek Dishes

For the spotting of the transfection solutions onto chambered glass coverslips, we use a ChipWriter Compact robot (Bio-Rad Laboratories, Hercules, CA) equipped with either SNS10 (TeleChem International, Sunnyvale, CA) or PTS 600 (Point Technologies, Boulder, CO) solid pins. These pins show a high reliability of the spotting process, in particular in the spot geometry achieved. Also, they deliver a sufficiently high volume (approximately 4 nl) to give rise to spots of approximately 400 ^m in size, which each can be overlaid with approximately 100 to 200 Hela or MCF7 cells. Transfecting fewer cells per spot suffers in our hands from poor statistics of the results achieved in functional assays. The spot-to-spot distance is adjusted to 1125 ^m and allows all 384 samples of a 384

well plate to be delivered onto one Lab-Tek chambered coverglass (NalgeNunc). The spotted solutions on the Lab-Tek chamber are then dried at room temperature for at least 12 hours after printing before cells are plated onto them. The solutions of one 384-well plate (see preceding) are sufficient to spot at least 50 identical Lab-Tek chambers, which can be stored for later use in a dry environment for several months without significant loss in transfection efficiency. After the tranfection solutions have dried they become visible on the LabTek chambers, which we routinely use to monitor the successful spotting.


It is crucial that the spotting robot used has to be able to pass the walls of the Lab-Tek chamber. The spotting procedure for 384 samples on 50 Lab-Tek chambers in parallel using 8 solid pins typically lasts 6 hours. In order to avoid evaporation of the small sample volumes in the 384 well plate during spotting, it is cooled with an in-house built water-cooled plate. In order to deposit more sample volume, which for some cell types improves transfection efficiencies, we spot repeatedly to the same position.

Plating of Cells on Lab-Tek Dishes with Dried Transfection Solution

The density of the cells plated on the spotted Lab-Tek chambers is always a compromise between the improved statistics that can be achieved with high cell densities and the quality of microscopic analyses that one expects.

Typically, we plate 1.25 x 105 actively growing HeLa, MCF7, COS7L, or HEK 293 cells in 2.5 ml culture medium (DMEM containing 10% heat-inactivated fetal calf serum, 2 mM glutamine, 100 U/ml penicillin, and 100 yug/ml streptomycin) on the dried spots of one Lab-Tek culture dish. This results in about 100 to 200 cells residing on one spot.

Typical incubation times (at 37° and 5% CO2) for the successful expression of plasmid cDNAs vary between 12 and 24 h. The incubation time for RNAi experiments varies between 20 to 50 h and strongly depends on the stability of the proteins targeted by the siRNAs spotted. For long-term experiments lasting several days, for example, RNAi experiments targeting stable proteins, the cell density needs to be lowered as with the cell density typically used (see preceding), the cells may stop growing due to contact inhibition, which makes experiments addressing cell cycle or signal transduction related questions difficult to interpret.

Transfection efficiencies of plasmid cDNAs depend strongly, as with standard transfection protocols, on the protein expressed. However, tranfection efficiencies of up to 90% can be achieved in some cases for plasmid cDNAs in HEK293 cells. Those for the transfection of siRNAs show less variation and are close to 100% as determined by the presence of rhodamine-labeled siRNA oligo-nucleotides in cells residing on a spot.

Preparation of Samples for Functional Analysis

For functional analysis involving high content screening microscopy, we frequently use immunofluorescence to monitor molecule specific morphological and biochemical parameters. It always has the advantage that features of different genes can be analyzed in parallel (Pepperkok et al., 2000). The immunostaining procedure in Lab-Tek chambered glass coverslips is very effective and cheap as it can be performed with the same antibody for hundreds of target genes in parallel.

Specifically for Lab-Teks, we apply 250 ^l of the corresponding antibody by carefully distributing the fluid over the spotted area. We incubate for 10 min with the lid closed followed by 2 washes with 2 ml PBS (30 min each). We routinely include reliable stains highlighting cell nuclei (e.g., Hoechst or Dapi) which facilitates automated focusing and image acquisition (Liebel et al., 2003). The stained samples are stored at 4° either embedded in Mowiol or in PBS solution containing azide after a brief poststaining fixation of the samples with paraformaldehyde for 2 min.

Analysis of the Samples by High Content Screening Microscopy

In principle, images of the cells on the spots can be acquired with any commercially available inverted microscope. We use a ScanR (Olympus Biosystems, Planegg, Germany; Liebel et al., 2003) scanning microscope, with automated focus, allowing time-lapse data acquisition. This microscope is equipped with a 10 x /0.4 air and 40 x /0.95 air PlanApo objective (Olympus, Melville, NY). The 10x objective allows the imaging of all cells of one spot in one image with a reasonable resolution. Both objectives used are air objectives, as this facilitates the sample scanning process in a reliable manner, which is often difficult when oil immersion objectives are used. A key point of the whole imaging process is to find the first spot of the array. For this purpose we use the cotransfected Cy3 DNA oligo-nucleotide. In addition we mark the first spot manually with a thin and water-resistant black marker pen on the opposite side of the coverglass before cells are seeded.

Additional Comments

As the accuracy of the spot-to-spot distance is extremely important, precise positioning of the spots is needed, demanding high-resolution spotting robots. Starting cell plating densities resulting in 100 to 200

cells per spot is a good compromise between the demands on sample statistics for functional assays and cells remaining in the logarithmic growth phase.

Optimizing Conditions and Troubleshooting

To optimize transfection efficiencies for different cell lines, a labeled siRNA resulting in a known phenotype or a GFP tagged cDNA plasmid with known subcellular localization can be used. The ratio of siRNA (cDNA) to transfection reagent for different transfection reagents needs to be adjusted individually. A good starting point for optimization is the ratio recommended by the supplier of the transfection reagent. Thorough mixing of transfection reagent and siRNA (cDNA) is crucial for successful transfection. A possible source for variations in transfection efficiency is batch-to-batch variations in the quality of the transfection reagent.

Summary and Perspectives

We present in this article a protocol to perform parallel analysis of hundreds of different genes on a single Lab-Tek chamber. We focused on a transfection protocol, as it has been applied successfully in our laboratory in combination with high content screening microscopy analysis. It may serve as a basis for any subsequent adaptations of the method to more complicated analysis methods or cell systems.


DeRisi, J., Penland, L., Brown, P. O., Bittner, M. L., Meltzer, P. S., Ray, M., Chen, Y., Su, Y. A., and Trent, J. M. (1996). Use of a cDNA microarray to analyse gene expression patterns in human cancer. Nat. Genet. 14, 457-460. Elbashir, S. M., Harborth, J., Lendeckel, W., Yalcin, A., Weber, K., and Tuschl, T. (2001). Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature 411, 494-498.

Erfle, H., Simpson, J. C., Bastiaens, P. I., and Pepperkok, R. (2004). siRNA cell arrays for high-content screening microscopy. Biotechniques 37, 454-462. Kato, K., Umezawa, K., Miyake, M., Miyake, J., and Nagamune, T. (2004). Transfection microarray of nonadherent cells on an oleyl poly(ethylene glycol) ether-modified glass slide. Biotechniques 37, 444-452. Kumar, R., Conklin, D. S., and Mittal, V. (2003). High-throughput selection of effective RNAi probes for gene silencing. Genome Res. 13, 2333-2340. Liebel, U., Starkuviene, V., Erfle, H., Simpson, J. C., Poustka, A., Wiemann, S., and Pepperkok, R. (2003). A microscope-based screening platform for large scale functional analysis in intact cells. FEBS Lett. 554, 394-398.

MacBeath, G., and Schreiber, S. L. (2000). Printing proteins as microarrays for high-throughput function determination. Science 289, 1760-1763.

Mishina, Y. M., Wilson, C. J., Bruett, L., Smith, J. J., Stoop-Myer, C., Jong, S., Amaral, L. P., Pedersen, R., Lyman, S. K., Myer, V. E., Kreidler, B. L., and Thompson, C. M. (2004). Multiplex GPCR assay in reverse transfection cell microarray. J.Biomol. Screening 9, 196-207.

Mousses, S., Caplen, N. J., Cornelison, R., Weaver, D., Basik, M., Hautaniemi, S., Elkahloun, A. G., Lotufo, R. A., Choudary, A., Dougherty, E. R., Suh, E., and Kallioniemi, O. (2003). RNAi microarray analysis in cultured mammalian cells. Genome Res. 13, 2341-2347.

Pepperkok, R., Girod, A., Simpson, J. C., and Rietdorf, J. (2000). Imunofluorescence microscopy. In ''Monoclonal Antibodies: A practical approach'' (P. Shepard and C. Dean, eds.), pp. 355-370. Oxford University Press, New York.

Schena, M., Shalon, D., Davis, R. W., and Brown, P. O. (1995). Quantitative monitoring of gene expression patterns with a complementary DNA microarray. Science 270, 467-470.

Smith, R. D., Pasa-Tolic, L., Lipton, M. S., Jensen, P. K., Anderson, G. A., Shen, Y., Conrads, T. P., Udseth, H. R., Harkewicz, R., Belov, M. E., Masselon, C., and Veenstra, T. D. (2001). Rapid quantitative measurements of proteomes by Fourier transform ion cyclotron resonance mass spectrometry. Electrophoresis 22, 1652-1668.

Starkuviene, V., Liebel, U., Simpson, J. C., Erfle, H., Poustka, A., Wiemann, S., and Pepperkok, R. (2004). High-content screening microscopy identifies novel proteins with a putative role in secretory membrane traffic. Genome Res. 14, 1948-1956.

Yoshikawa, T., Uchimura, E., Kishi, M., Funeriu, D. P., Miyake, M., and Miyake, J. (2004). Transfection microarray of human mesenchymal stem cells and on-chip siRNA gene knockdown. J. Control Rel. 96, 227-232.

Ziauddin, J., and Sabatini, D. M. (2001). Microarrays of cells expressing defined cDNAs. Nature 411, 107-110.

0 0

Post a comment