for Hybridoma Technology
Introduction
Hybridoma technology includes cell fusion, the plating of master cell cultures, and antibody screening. The ultimate goal is to identify a single hybridoma cell producing a monoclonal antibody with a desired specificity or function. Achieving cell clonality is important from a process control as well as a regulatory perspective. Until recently, hybridoma-producing laboratories have pursued a labor-intensive and time-consuming serial path of hybridoma culturing and limiting dilution cloning to achieve clonality.
This Application Note addresses the need to verify hybridoma cell clonality in an automated fashion. The ImageXpress® Velos System offers sensitive, rapid, and robust screening of multi-well plates for the identification of wells containing a single cell as well as characterization of colonies grown from those cells.
Experimental
Hybridoma cells were grown in clear polystyrene tissue culture plates. The hybridoma line, growth medium and the quantitation of human IgG were provided by Medarex (Milpitas, CA). The ImageXpress Velos System was configured with a 488 nm laser. The laser scatter signal was collected in channel 1 and fluorescence was filtered through a 510-540 nm band pass filter for channel 2. A 96-well Costar tissue culture plate was loaded into the plate nest of the ImageXpress Velos System for scanning and all instrument settings including the gains for the PMTs were automatically configured by the ImageXpress Velos System method (see Figure 1). Plates were typically scanned at a resolution of 10µ; scantime at setting is 120 sec.
Figure 1. Flow diagram of data acquisition and data analysis. A ImageXpress Velos System Method controls all instrument parameters that are defined for a type of assay and specific plate. The acquired data is saved as a single file and as an image stack (one image for each parameter) per well. Data analysis of a well or entire plate consists of using BlueImage to run a previously defined “Process” to classify “objects” and generate a comma separated variable (csv) data file containing all measured parameters per well.
Hybridoma Master Cultures
Hybridoma master cultures were established in 96-well clear polystyrene tissue culture plates after cell fusion on Day 0. The ability to use standard clear polystyrene tissue culture plates of 6, 12, 24, 48, or 96-well densities has been demonstrated previously. The 96-well plates were scanned on day 6, 8 and 11. A half-volume media replacement was done on day 7. As shown in Figure 2 the scatter image clearly shows the growth of the hybridoma colony versus number of days in culture. The lid remained on the plate during the scanning to maintain sterility
Day 6
Day 8
Day 11
Figure 2. ImageXpress Velos System Ch1 laser scatter image of a single well from a hybridoma master culture plate. The plate was scanned at 10 micron resolution on day 6, 8, and 11. The image shows the growth of a hybridoma colony (outlined in black) with measured areas of 0.4, 1, and 3.3 x 106 µ2 , for days 6, 8, and 11 respectively. Note the complex laser reflections from the plate sides, lid and the liquid surface (meniscus).
Single Cell Analysis in Tissue-Culture Plates
While it is possible to detect cells with laser scatter, a more robust approach is to stain them with (CMFDA, Molecular Probes, Eugene, OR) and measure fluorescence. Images of a single cell and a colony grown from that cell, both stained with CMFDA, are shown in Figure 3. A complication arises from laser induced fluorescence (LIF) from the clear polystyrene plates (Note: this is not an issue with black wall plates). These signals, seen in the right side of the wells in Figure 3, limit the ability to clearly define cells using an image threshold approach. However, the detection of single cells can be accomplished by using a fast Fourier transform (FFT) to remove the unwanted signal. The results of this process are shown in Figure 4, where the initial image shows a well with two CMFDA-stained cells, followed by a FFT filter process, and finally thresholding and measurement of single cells. Table I shows the number of cells counted per well on Day 0 from plating 0.5 cells/well into a 96-well plate using this technique. The resolution setting of 10µ provides a sufficient number of data points per cell for robust detection and allows a scantime of 120 sec.
Figure 3. ImageXpress Velos System Ch2 fluorescence images of a single CMFDAstained cell in well FO2 (left image; blue circle around a single cell) and the resulting colony stained with a second round of CMFDA (right image). The hybridoma subcloning plate was scanned at 10 micron resolution on day 0, 5, and 7. Fresh growth medium with a final concentration of 0.5 µM CMFDA was added on day 7. The day 7 shows a fluorescence image of the hybridoma colony after labeling with CMFDA for 5 min at room temperature. The other signals on the right and left edges are due to laser induced fluorescence (LIF) from the clear polystyrene plate.
Raw Image 
Image after FFT Filter 
Thresholded Particles 
Figure 4. ImageXpress Velos System fluorescence images of a single well containing two CMFDA-stained cells (blue circles around cells in the left image labeled “Raw Image”). The hybridoma subcloning plate was scanned at 10 micron resolution on day 0. The raw image shows the signal on the left edge due to LIF from the clear polystyrene plate. After the FFT image processing step low spatial frequency signals are removed (middle image labeled “Image after FFT”). The resulting thresholded image (right image labeled “Thresholded Particles”) shows two cells in this well.
Table I. Number of cells per well of a 96-well plate on Day 0
No Effect of CMFDA-labeling on Hybridoma Growth or Colony Formation
The repeated labeling of the hybridoma cells with CMFDA and scanning process had no effect on the cells ability to grow and form colonies, as shown by the representative colony in Figure 5. Moreover, additional studies demonstrated i) that antibody production was not affected by CMFDA-staining and ii) antibody detection could be linked to wells containing a single CMFDAlabeled cell (data not shown).
Figure 5. ImageXpress Velos System fluorescence image from one well containing one CMFDA-stained cell (blue dotted circle around cell in image labeled Day 0). The hybridoma subcloning plate was scanned at 10 micron resolution on day 0. On day 3 the well was manually examined and found to contain 3 cells. On Day 5 the resulting colony was stained with CMFDA and scanned (right image labeled Day 5). The blue dotted circle represents the location of the single cell responsible for the final colony (note evidence for cell migration).
Conclusions
Compared with previously available methods, the use of the ImageXpress Velos System allows the fast analysis of large numbers of wells in multi-well plates for secreted products on the single-cell level. Critical to this is the ability to scan complete wells and perform multiparametric analysis of cells and colonies. Specifically we have demonstrated the capability to perform:
- Rapid cell counting of CMFDA-stained cells in standard tissue culture plates
- High throughput and rapid single cell detection
- Growth monitoring of colonies
- Multi-color fluorescence analysis of live clones (transfectomas)
The ability to rapidly identify wells containing a single cell has the potential of accelerating the antibody discovery process. The technology is broadly applicable for the detection of secreted products as well as the identification of transfected cells expressing the desired product. In addition, the platform provides the ability to normalize the amount of secreted product to the number of cells. These features are especially important for bioprocess development where the selection of high producing cell lines is an important step.