User Ratings

Easy to Follow Achieved Expected Results Overall Rating Add your comments

Practical Issues in High‐Speed Cell Sorting

Larry W. Arnold¹, Joanne Lannigan²

¹The University of North Carolina at Chapel Hill, Chapel Hill, North Carolina
²The University of Virginia, Charlottesville, Virginia

Publication Name:

Current Protocols in Cytometry

Unit Number:

Unit 1.24

DOI:

10.1002/0471142956.cy0124s51

Online Posting Date:

January, 2010

GO TO THE FULL TEXT:

PDF or HTML at Wiley Online Library

Are you the author of this protocol? Login or register and return to this page.

Abstract

Modern flow cytometric cell sorters are all capable of so-called “high-speed sorting.” However, there is confusion about exactly how fast a “high-speed” cell sorter can sort cells. There are many considerations in achieving the fastest sorting speed, as well as the highest quality sort results—cell recovery, purity, and functionality. This requires the same considerations required for “slow-speed sorting”; however, a more precise implementation is required for high-speed sorting. The modern cell sorters enable high-speed sorting because of advances in high-speed electronics and data processing. We discuss the practical considerations of high-speed sorting in terms of the theory and practical aspects of the mechanical and software components of sorting, statistics of sorting, cell preparation and viability, instrument setup, sort strategies, and biosafety. Curr. Protoc. Cytom. 51:1.24.1-1.24.30. © 2010 by John Wiley & Sons, Inc.

Keywords: flow cytometry; sorting; FACS

GO TO THE FULL PROTOCOL:

PDF or HTML at Wiley Online Library

Introduction
What is High-Speed Sorting?
Understanding High-Speed Sorting
Practical Aspects of High-Speed Sorting
Safety
Literature Cited
Figures
Tables

GO TO THE FULL PROTOCOL:

PDF or HTML at Wiley Online Library

Figures

Figure 1.24.1

The histograms shown demonstrate the issue with sorting dim populations from negative populations. Panels A and B show the data from the starting sample. Panel A shows the position of R1 that tries to encompass the majority of what appears to be the positive sort region R1 population. Panel B shows a more conservative position for sort region R1 to emphasize purity. Panel C shows the sorting result using panel A R1 sort region and panel D shows the sort result using the panel B R1 sort region. In panel C, the sort result is heavily contaminated with negatives while in panel D the sort result is much more pure for the positive population.

View Image
Figure 1.24.2

These photos (A and B) from the stream illumination camera show a central waste stream and two deflected side streams. Panel A shows the waste stream after proper defanning has been set. Panel B shows the waste stream before adjustment for charge correction (i.e., defanning). Panel C shows a photo of the setup for 4-way sorting.

View Image
Figure 1.24.3

This illustration demonstrates the effects of setting a purity mode (or mask). The rectangles represent the drops to be formed. Note that at the time these sort decisions are being made, the “drops” are not yet formed but are contained in the fluid stream. The “target drop” will be the drop which is predicted to contain the desired cell to be sorted, the “leading drop” will be the drop to be formed immediately preceding the target drop, and the “trailing drop” will be the drop to be formed following the target drop. These definitions apply to all similar figures. The drops are divided into 1/16ths (for simplicity—current instruments have divisions which range from 1/32nd to 1/1000th) for cell localization. The exclusion zone (or purity mask) that has been set is 0.5 drops (i.e., 0.25 drops to either side of the drop predicted to contain the cell to be sorted. This means that the sort envelope is 1.5 drops). The small (green) dots represent wanted cells and the large (red) dots represent cells that are not wanted. The drops that will be sorted are boxed (light blue).

View Image
Figure 1.24.4

This illustration shows the effects of setting a recovery or yield mask of 0.5 drops (0.25 drops on either side of predicted drop). When the cell is in the middle half of the drop predicted to contain it, only the one drop will be sorted. When the cell is in either of the outside quarters of the drop, the drop predicted to contain it and the drop to the respective side will be sorted as the cell may partition into the neighboring drop. The drops that will be sorted are boxed (light blue).

View Image
Figure 1.24.5

Shown in this figure is the effect of combining both a purity and recovery/yield mode or mask. Both masks are set to 0.5 drop (i.e., ± 0.25 drop). The yield mask is shown in olive (shaded) and the purity mask in light purple (hashed). Wanted cells are represented by small (green) dots and unwanted cells by large (red) dots. The drops that will be sorted are boxed (light blue).

View Image
Figure 1.24.6

This is an illustration of a “single” sort mode (phase mask). The mask of 0.5 drops is shown in shaded (olive). The drops that will be sorted are boxed (light blue). Only when a cell is predicted to be in the middle half of the drop will the drop be sorted. This mode results in the largest loss of cells of any of the modes. Note the single mode can also be combined with a purity sort (not shown here).

View Image
Figure 1.24.7

Shown are an example of a good drop breakoff (A) and a poor breakoff (B). In panel B, the drop drive amplitude and/or frequency are set too high. The neck between the last and second to last attached drops is too short. Compare this to the neck in panel A, which is long and straight. Also visible in panel B is that the first satellite drop is too small and the satellite drop fuses back with the main drops within 3 drops which is too fast. Panel C shows how the satellite drops fuse back at drop 5 with a good breakoff. Also, note the slightly longer first satellite when the drop breakoff is good.

View Image
Figure 1.24.8

Shown here are photographic images from the drop breakoff camera demonstrating the drop breakoff changing over time and how to adjust to return to a good breakoff. The horizontal line is a reference point on the monitor. Panel A shows a good starting breakoff. Panel B shows that the breakoff is creeping shorter and the desired last attached drop (in panel A) is now separating from the neck. Panel C shows how the breakoff is beginning to creep longer as demonstrated by the lengthening of the first satellite drop, as well as the last attached drop. The reference line on the camera monitor shows that although the drop breakoff is changing, the length from the nozzle has not changed. The breakoff change represented by the images in panels B and C should be corrected by decreasing or increasing the drop drive amplitude, respectively. Panels D and E show the effect of increasing and decreasing stream velocity, respectively. While the breakoff pattern looks the same as in panel A, the position of the last attached drop has moved down or up from the reference line. These types of changes should not be corrected by changing the drop drive amplitude.

View Image
Figure 1.24.9

The data plots shown illustrate the utility of adding an autofluorescence parameter to resolve negatives and dim positives. In this case, the cells were single-color stained with two FITC-labeled antibodies (CD29 or CD105) (or isotype control) and plotted vs. FL2 which was autofluorescence only. The fluorescein spillover into FL2 was compensated. The left two plots show the isotype control and the gates that were set by visualizing this “negative” control sample. If the data had been visualized using just a single-parameter histogram (bottom row), it is difficult to tell where the positives separate from the negatives. The percent positives are grossly underestimated by the simplistic method using the single parameter plots. The two-parameter plots easily provide resolution of the positives.

View Image

Literature Cited

Literature Cited
	Alberti, S., Parks, D.R., and Herzenberg, L.A. 1987. A single laser method for subtraction of cell autofluorescence in flow cytometry. Cytometry 8:114-119.
	Lennartz, K., Lu, M., Flasshove, M., Moritz, T., and Kirsten, U. 2005. Improving the biosafety of cell sorting by adaptation of a cell sorting system to a biosafety cabinet. Cytometry 66:119-127.
	Maecker, H.T., Frey, T., Nomura, L.E., and Trotter, J. 2004. Selecting fluorochromes for maximum sensitivity. Cytometry 62:169-173.
	Oberyszyn, A.S. and Robertson, F.M. 2001. Novel rapid method for visualization of extent and location of contamination during high-speed sorting of potential biohazardous samples. Cytometry 43:217-222.
	Perfetto, S.P., Ambrozak, D.R., Koup, R.A., and Roederer, M. 2003. Measuring containment of viable infectious cell sorting in high-velocity cell sorters. Cytometry 52:122-130.
	Perfetto, S.P., Ambrozak, D.R., Roederer, D.R., and Koup, R.A. 2004. Viable infectious cell sorting in a BSL-3 facility. Methods Cell Biol. 263:419-424.
	Peters, D., Branscomb, E., Dean, P., Merrill, T., Pinkel, D., Van Dilla, M., and Gray, J.W. 1985. The LLNL high-speed sorter: Design features, operational characteristics, and biological utility. Cytometry 6:290-301.
	Pinkel, D. and Stovel, R. 1985. "Flow chambers and sample handling". In Flow Cytometry: Instrumentation and Data Analysis (P.N. Dean, ed.) pp. 78-128. Academic Press, London.
	Rayleigh, J.W.S. 1877. Theory of Sound. MacMillan, London. Reprinted 1945. Dover, New York.
	Roederer, M. and Murphy, R.F. 1986. Cell-by-cell autofluorescence correction for low signal-to-noise systems: Applications to epidermal growth factor endocytosis by 3T3 fibroblasts. Cytometry 7:558-565.
	Schmid, I., Nicholson, J.K.A., Giorgi, J.V., Janossy, G., Kunkl, A., Lopez, P.A., Perfetto, S., Seamer, L.C., and Dean, P.N. 1997. Biosafety guidelines for sorting unfixed cells. Cytometry 28:99-117.
	Schmid, I., Lambert, C., Ambrozak, D., Marti, G.E., Moss, D.M., and Perfetto, S.P. 2007. International society for analytical cytology biosafety for sorting of unfixed cells. Cytometry 71:414-437.
	Van den Engh, G. 2000. "High speed cell sorting". In Emerging Tools for Single-Cell Analysis (G. Durack and J.P. Robinson, eds.) pp. 21-48. Wiley-Liss, New York.
	van den Engh, G.J. and Stokdijk, W. 1989. Parallel processing data acquisition system for multi-laser flow cytometry and cell sorting. Cytometry 10:282-293.
	Van Dilla, M.A., Deaven, L.L., Albright, K.L., Bartholdi, N.C., Campbell, E.W., Carrano, A.V., Christensen, M., Clark, L.M., Cram, L.S., Dean, P.N., de Jong, P., Fawcett, J.J., Fuscoe, J.C., Gray, J.W., Hildebrand, C.E., Jackson, P.J., Jett, J.H., Killa, S., Longmire, J.L., Lozes, C.R., Luedemann, M.L., McNinch, J.S., Mendelsohn, M.L., Meyne, J., Meincke, L.J., Moyzis, R.K., Mullikin, J., Munk, A.C., Perlman, J., Pederson, L., Peters, D.C., Silva, A.J., Trask, B.J., van den Engh, G. 1990. "The national laboratory gene library project". In Flow Cytogenetics (J.W. Gray, ed.) pp. 257-274. Academic Press, London.