Application Note

Label-Free/Live Suspension Cell CountingBrightfield

ObjectiveNon-labeled, living U-937 suspension cells were used to demonstrate automated counting using the MIAS-2 truVIEW microscopy reader and eaZYX IMAGING Software (Figure 1) . The MIAS-2 enables to detect cells from seeding densities ranging from 2000 to 28000 cells per well in 96-well plate despite major positional illumination differences resulting from the air-liquid meniscus.

IntroductionThe MIAS-2 is a compact, fully automated, versatile microscopy reader combining 5 brightfield modes and with up to 8 fluorescent modes. For this application the brightfield mode is selected. Using cell seeding experiments with the non-adherent cell line U-937 as an example, this application note describes how cell counting in plates can be executed fully automated and in a fully non-invasive manner at plate cycle times that match high content screening requirements.

Materials/MethodsU-937 cells (ATCC 1593.2) were grown in RPMI medium (Invitrogen, Cat.No. 21875-034) in 175 cm2 T-flasks. The density of the cell suspension was calibrated using a Neubauer counting chamber immediately prior to cell seeding in plates (1). 2000 to 28000 U-937 cells were seeded in 200 µl of medium into 96 well plates (Greiner, CellStar, Cat.No. 655180). Cells were allowed to settle for at least 1 hour in the incubator (95% RH, 37ºC, 5% CO2). Plates were equilibrated to room temperature for 15 minutes and placed in the MIAS-2 reader. Whole well, 5 x 5 tiled images were captured using a 5x objective and 1.0x Optovar settings. Brightfield light settings were adjusted for proper illumination at the centre of the wells. A specific application for suspension cells was selected for analysis. Whole well overviews with image analysis result overlays were created using eaZYX-IMAGING software (2). Data tables were imported in a spreadsheet program (MS-Excel) and graphics were prepared. The images shown in the figures were JPEG compressed and contrast adjusted prior to publication.

Automated Counting of Label-free Live U-937 Suspension Cells in 96-well Plates Using the MIAS®-2 truVIEW Microscopy Reader and eaZYX® Imaging Software

Bieke Govaerts, Leen Geuens, Marc Moeremans, Kris Ver Donck & Johan Geysen

Figure 1. The MIAS-2 reader equipped with eaZYX IMAGING software.

Figure 2. View of a whole well with the overlay of the image analysis result in red. These types of images allow to verify the quality of the image and the cell count result. The eaZYX IMAGING software will display a thumbnail plate overview. The seeding density in this well was 16000.

BA

DC

BA

DC

ResultsA novel, fully automated image analysis application was developed to count suspension cells. The image capture plate cycle time was ~20 min. for whole well scanning (total well counts) and ~5 min. per plate when a single image was captured (cell count in a sample image taken at the center of wells). In Figure 2. it is shown how U-937 cells settle unevenly in plates and how in each image, the background is affected differently by the optical effects of the air-medium meniscus. Despite these variable backgrounds, the overlay image analysis result shows robust object identification from the center to the perimeter of the well (see also details in Figure 3). As shown in Figure 4, the cell count match closely to the numbers of cells seeded and linearity (R2 of 0.9985) proved to extend across the representative range of cell densities tested.

DiscussionThe results support the conclusion that the MIAS-2 reader and the eaZYX analysis software can adequately count living, unlabeled U-937 cells across the well at cell densities ranging from low (typically used when seeding a cell plate) up to high, fully grown (semi) confluent cell densities. The non-invasive nature of this application facilitates several applications in cell based screening. Bad quality wells, plates and batches can be identified and discarded in a timely fashion. The exact cell count prior to or after biochemical or cell-based screening can be used to correct for actual well-to-well, plate-to-plate and day-to-day variability in cultures. Furthermore, the application can be used as a screening assay to screen for compounds or genes that inhibit or stimulate cell proliferation. With this fast and label-free alternative for a biochemical surrogate endpoint assay, a significant reduction in reagent cost and hands-on time can be achieved.

References1. Freshney, R.I., 1987, Culture of Animal Cells: A Manual of Basic Technique, Second

Edition, 227-229.2. Van Osta, P., Geusebroek, J.M., Ver Donck, K., Bols, L., Geysen, J., ter Haar Romeny B.

,ypocsorcim thgil ni ruoloc dna erutcurts ot deilppa ecaps elacs fo selpicnirp ehT ,2002 ,.MProceedings of the Royal Microscopical Society, 37 (3): 161-166

Figure 3. Analysis result details from the center (A), halfway (B, C) and edge (D) of the well shown in fig. 1. Differences in background caused by air-medium meniscus effects shown at image borders in both vertical (B) and horizontal (C) direction. Cells on the walls of the well are not counted (D).

Figure 4. Graphical presentation of the object count versus the actual cell seeding density (n = 6) after whole well scanning. Object count and seeding density are closely correlated and the linearity extends across the density range.

Digilab, Inc. 84 October Hill Road

Holliston, MA 01746Phone: (508) 893-3130

Toll Free: (800) 935-8011 Fax: (508) 893-8011 E-Mail: info@digilabglobal.com

R2 = 0.9985

0

10

20

30

40

0 10 20 30 40Cell Seeding Density (x1000)

Objec

ts Co

unte

d (x

1000

)

ANE.0641.001.01

BA

DC

BA

DC