Post Acquisition Cell Compensation Using VenturiOne®

A review of colour cell compensation using VenturiOne®, flow cytometry data analysis software, as endorsed by Ian Dimmick and Rebecca Stewart of the Flow Cytometry Core Facility, Institute of Human Genetics, Newcastle upon Tyne University.

Introduction to fluorochrome emmissions

Due to the increased use of flow with multiple lasers, the choice of fluorochromes available has grown considerably (1). This offers users great benefits, allowing more varied and extensive research by enabling the use of multiple lasers on individual instruments.

This offers the chance to optimise fluorochrome excitation and increase the number of fluorochromes and dyes for simultaneous detection within each experiment. The overall outcome of this being more data with fewer cells for cell compensation analysis.

When using two or more fluorochromes in an experiment, the spectral overlap of the fluorochrome emission spectra needs to be taken into consideration. Collectively the emission spectra of all fluorochromes cover a broad wavelength range (fig.1).

This will lead to spectral overlap where the fluorescence spectrum of one fluorochrome spills over into the detection channel dedicated to another fluorochrome being used simultaneously.

This spectral overlap causes difficulty when simultaneously trying to measure the true fluorescence of each fluorochrome, therefore obtaining a correct representation of the data and a correction must be applied.

This correction is termed colour compensation (1). A subtraction of the spill over from one fluorochrome onto another is applied to compensate for the spectral overlap. This value can be calculated as a percentage spillover of the primary fluorochrome into a detector in which it is not the primary detector.

Individual control samples stained with each fluorochrome to be used in the experiment are used independently to establish spectral overlap. This is in order to determine the amount of compensation required for each fluorochrome.

Researchers can adopt a method to calculate the cell compensation before data is collected by the flow cytometer called pre acquisition compensation.

figure 1 cell compensation venturione
Figure 1. Emission Profiles of Fluorochromes showing Spectral Overlap.

Introduction to Post Acquisition Cell Compensation

Compensation can also be calculated after data collection within the software, called post acquisition compensation and if need be this can be used to check the integrity of the pre acquisition compensation values.

Post acquisition compensation means that adjustments are made away from the instrument environment at leisure, but the basics of compensation remain the same using either pre or post acquisition compensation. Using multiple fluorochromes and dyes means that compensation can be time-consuming, meaning that post acquisition is an ever increasing requirement (2).

In this review the focus is on post acquisition compensation using VenturiOne®, investigating manual and automatic post acquisition compensation methods.

In this review, I will also assess the use of compensation beads and cells which are routinely used at the flow cytometry core facility at the Institute of Human Genetics, and discusses both their advantages and disadvantages.

Pre Acquisition Cell Compensation Method with a Flow Cytometer

Please refer to Instrument User Manual for full instructions

1. Set up the instrument PMT voltages using appropriate cellular baseline controls

2. Use compensation beads to obtain maximal fluorescence of each fluorochrome whilst running under the cell derived instrument settings. The cytometer calculates compensation values. Forward Scatter (FS) and Side Scatter (SS) parameter settings may need changing, but these will not affect the final compensation values. Compensation beads are the preferred choice due to the advantage of giving distinct negative and positive peaks and increased signal intensity when analysed.

3. Verify compensation values by re analysis of the controls to ensure correct compensation values with respect to appropriate X and Y axis negative and positive mean or median values.

Post Acquisition Cell Compensation Method with VenturiOne®

Please refer to VenturiOne® User Guide for full instructions on Compensation Wizard

1. Set the default log decade scaling to match that of the pre acquisition method.

2. Open the compensation cells file into the playlist of VenturiOne®. Replicate the files for each fluorochrome that is to be compensated.

3. Open the Compensation Wizard from the compensation tab and configure the settings for compensation (See fig 2) click Next.

figure 2 cell compensation venturione
Figure 2. Configuration page of the Compensation Wizard

4. On the compensation page, adjust the gating and compensation regions to capture the populations for calculating the coefficients of the fluorochrome. See fig. 3.

figure 3 cell compensation venturione
Figure 3. Compensation page of the wizard that allows adjustment of the regions for compensation.

5. You only need to click Next to repeat this for every fluorochrome to be compensated.

6. Then the last page of the wizard allows you to review, print or save the compensation settings file. See fig 4.

figure 4 cell compensation venturione
Figure 4. Review Page with Options to Save or Print the Compensation Settings.

7. Save compensation file. See fig 5.

figure 5 cell compensation venturione
Figure 5. Compensation Matrix using Compensation Cells.

8. Repeat steps 2 to 7 with the Compensation bead files.

9. See Fig 6 for Compensation Bead matrix.

figure 6 cell compensation venturione
Figure 6. Compensation Matrix using Compensation Beads.

10. Manual compensation is used for minor adjustments of the compensation values.

11. Verify values with positive control using compensation matrix. See fig 7. Use V-log to check for overcompensation.

figure 7 cell compensation venturione
Figure 7. Dual Parameter Plots of Compensated Positive Control Using Compensation Beads Matrix.

Principles of Cell Compensation for Consideration

• More accurate compensation is achieved through using compensation beads, which give brighter signals than cells.

• Autofluoresence of positive and negative populations must be the same for all compensation controls in order to achieve correct compensation (4).

 • The CD19 expression expresses a weak signal. This makes it difficult to determine positive populations from negative populations. Not just a case of the brightest peak being the positive population. See fig. 8 where the uncompensated CD19 histogram shows two positive peaks. The brighter of these two is actually the spectral overlap of CD8 into the CD19 detector whereas the weaker positive peak is CD19 only.

• CD45 expression results in a very bright signal so is a good control

• Ensure to Clear Compensation after calculating each fluorochromes compensation value in VenturiOne® so that you are not calculating compensation on parameters that are already compensated. Please note, this is not required when using the Compensation Wizard.

• Compensation values must always be recalculated if the voltages of the Photomultiplier Tubes (PMT) are altered (3).

• Due to the complexity of multiple colour compensation, use the automatic compensation feature in the software primarily and use the manual feature to make minor adjustments.

• Two dimensional fluorescence plots can help determine more accurate compensation values by giving better visualization of populations requiring compensation.

• Controls for determining the compensation matrix should produce fluorescent signals that are bright as possible.

• In stem cell research, stem cells may not have well expressed surface antigens therefore they do not achieve bright fluorescent signals which we need to calculate optimum compensation values. Instead, compensation beads are recommended for calculating compensation in this scenario, as they will allow us to achieve brighter fluorescent signals.

figure 8 cell compensation venturione
Figure 8- CD19 Histogram Uncompensated Showing the Spectral Overlap of CD8 into CD19 Detector.

Summary

VenturiOne® alleviates the complexity of multiple colour cell compensation, offering a quick and easy approach to post acquisition compensation, with the added flexibility of intuitive automatic and manual cell compensation options.

Try it for yourself, with a free 30-day trial here.

References

1. Ormerod, M (2008). Flow Cytometry- A Basic Introduction

2. Roederer, M (2001). Spectral Compensation for Flow Cytometry: Visualization Artifacts, Limitations and Caveats. Cytometry 45:194-205

3. Setting up 2 or 3 Colour FACS Analysis: http://users.ox.ac.uk/~path0116/tig/ccomp.html 18.12.97

4. Compensation (An informal perspective): http://www.drmr.com/compensation/ May 24, 2000.

Acknowledgements

With thanks to Ian Dimmick and Rebecca Stewart for dedicating the time required to collate the data for this document and for their continued support of VenturiOne®.

Introduction to cell cycle analysis

This article showcases the VenturiOne cell cycle analysis feature in a description of the application of the Watson modelling, in order to overcome the overlap between S phase, G1 and G2/M phases of the cell cycle.

The DNA content of cells varies in relation to the distinct phase of the cell cycle they are currently in. Flow cytometry analyses the different stages of the cell cycle by staining them with a DNA-binding fluorescent dye, such as propidium iodide (Figure 1).

1 venturione flow cytometry data analysis software cell cycle analysis
Figure 1. The relationship between the DNA histogram and the cell cycle.

Frequently, an investigator wants to know the percentage of cells in each phase of the cell cycle particularly in the field of cancer research. Here is where we can observe the effects of cytotoxic drug, radiation and other cancer treatments.

During DNA analysis a problem can arise due to the overlap of cells in G1 and those in early S phase and the overlap between G2/M and late S phase (Figure 2).

2 venturione flow cytometry data analysis software cell cycle analysis
Figure 2. The overlap between S phase, G1 and G2/M phases of the cell cycle. The cells in S phase were labelled with bromodeoxyuridine, which separated the DNA histograms of G1, S and G2/M phases. They then superimposed the three histograms.

To overcome this problem a computer program can separate the phases of the cell cycle by modelling the DNA histogram, and these programs can use several different modelling approaches. One such approach is the modified Watson ‘pragmatic’ algorithm which has been employed in the VenturiOne cell cycle analysis feature.

Experimental procedures

The cells shown in Figure 3 were found to be fixed in 70% ice-cold ethanol, in which they were stored at -20°C where they then, for analysis, were resuspended in PBS with propidium iodide (PI) at 20µg/ml and RNase at 1 mg/ml. For this and other staining protocols see Ormerod, 2000, and Current Protocols in Cytometry, 2007.

Analysis using VenturiOne

The model assumes that G1 and G2/M are normally distributed and fits Gaussian curves to them, returning the peak channel number and the coefficient of variation (CV). It then uses these values to construct the regions of S phase which overlap either with G1 or G2/M, using curves based on the cumulative error function (erf). The model uses the unmodified experimental values for the centre of S phase. Ormerod et al., 1987 describes this in more detail.

When we activate the cell cycle option in the VenturiOne software, a modified Watson algorithm of the cell cycle analysis is automatically applied to the selected histogram plot. The software distinguishes the three phases associated with the DNA cell cycle by colour. Red denotes the G0/G1 phase, blue denotes the S phase and green denotes the G2/M phase.

There are also user adjustable markers present on the cell cycle plots. We have positioned a red marker at the estimated central point of the G0/G1 phase and a green marker at the estimated central point of the G2/M phase. The software displays additional blue markers on the plots to set noise thresholds. It presents events that occur above or below these thresholds on the plots, but does not analyse them.

Some results using the VenturiOne software are shown below. In Figure 3A, a large proportion of cells are in G0/G1 phase as indicated by a large red peak whereas in Figure 3B, the majority of cells are in G2/M phase indicated by a large green peak.

3 venturione flow cytometry data analysis software cell cycle analysis
Figure 3. Cell cycle analysis of (A) a murine leukaemic cell line, (B) the cells 48 hours after incubation with a cytotoxic drug.

Discussion of the Watson model in VenturiOne

Using any program, the goodness of fit will depend on the quality of the DNA histogram. As in the results above, the cell cycle analysis reports an error message which indicates whether the algorithm experienced any problems. Here the message is Fit OK displaying good G1 and G2 CV’s. CV’s across the G1 and G2/M peaks should be =8% should be rejected.

There will be occasions when the program clearly has not fitted the histogram correctly, and the most likely source of the problem is fitting G2 and the S/G2 overlap due to a lot of noise. This is easily detected as the CV G1 and G2/M peaks are vastly out of line, and when this happens you should experiment by changing the starting position of the G2 peak.

The Watson model used in the VenturiOne software is a simple and robust method as it requires the assumption that the data are normally distributed and a recognisable G1 peak; this may be vestigial although from referenced literature, rarely absent. Based on these assumptions, VenturiOne produces fast and simple cell cycle analysis automatically therefore reducing the need for user intervention. As seen in the results, VenturiOne is also used to show differences between typical cell lines and those treated with cytotoxic drugs without having to take anything else into account.

Find out more about VenturiOne®, and to try a free trial today.

Note

This program is for research use only.

References

Current Protocols in Cytometry, 2007. John Wiley & Sons, Inc.

Ormerod, M.G. (2000). Flow Cytometry. A Practical Approach.

Ormerod, M.G. (ed). 3rd edition. Oxford University Press, Oxford. 83-97.

Ormerod, M.G., Payne, A.W.R. and Watson, J.V. (1987) Improved program for the analysis of DNA histograms. Cytometry 8: 637-641.

Watson, J.V., Chambers, S.H. and Smith, P.J. (1987) A pragmatic approach to the analysis of DNA histograms with a definable G1 peak. Cytometry 8: 1-8.