How to cite this article: DiGiuseppe JA, Cardinali J. Improved compensation of the fluorochrome AmCyan using cellular controls. Cytometry Part B 2011; 80B: 191–194.
Implementation of polychromatic flow cytometry in the clinical laboratory often requires the use of newer fluorochromes, with which experience may be comparatively limited. In the course of implementing polychromatic flow cytometry in our laboratory, we have observed significant differences in compensation values derived for the violet-excited dye, AmCyan, when cells rather than a commercially available set of polystyrene microparticles (BD CompBeads) are used as compensation controls.
Compensation values were calculated for AmCyan and several other fluorochromes using the BD CompBeads Set according to the manufacturer's protocol, and using unstained and singly stained lymphocytes as compensation controls.
When the BD CompBeads Set was used to determine compensation values, spillover from AmCyan into V450 was overcompensated, while spillover from AmCyan into FITC was undercompensated. In contrast, when compensation values were calculated using unstained and singly stained lymphocytes, spillover into V450 and FITC from cells stained brightly with AmCyan-conjugates was compensated appropriately. Although significant differences were observed in the compensation of spillover from AmCyan into V450 and FITC using cells rather than the BD CompBeads Set as compensation controls (P < 0.0001, two-tailed Wilcoxon signed-rank test), such differences were not observed in control experiments using fluorochromes excited by the blue (FITC and PE) or red (APC) lasers.
Although initially developed to probe the complexity of the immune system (reviewed in Ref.1), polychromatic flow cytometry has more recently been adopted for use in the clinical flow cytometry laboratory (reviewed in Ref.2). Polychromatic flow cytometry offers many potential benefits to the clinical laboratory, including (i) direct identification of immunophenotypically complex cellular populations without recourse to inferential reasoning, (ii) improved ability precisely to define and detect abnormal populations present at low levels, (iii) ability to acquire comparatively large amounts of information from sparsely cellular specimens (e.g., cerebrospinal fluid), and (iv) reduction in the number of tubes required for complete immunophenotypic analysis, which is itself associated with a number of potential efficiencies (e.g., less hand-on technologists' time required for case set-up, less instrument time required for sample acquisition) and cost savings (e.g., elimination of redundant gating antibodies).
A number of technical challenges must be surmounted, though, in order to implement successfully polychromatic flow cytometry in the clinical laboratory. These include the use of newer fluorochromes (3) with which laboratory experience may be comparatively limited and increased complexity in the design (4), compensation (5), and analysis (3) of multicolor antibody combinations. In the course of implementing polychromatic flow cytometry in our laboratory, we have observed significant differences in compensation values derived for the violet-excited dye, AmCyan, when cells rather than a commercially available set of polystyrene microparticles (BD CompBeads) are used as compensation controls. We describe these findings and show that improved compensation of AmCyan is obtained when cells rather than BD CompBeads are used as compensation controls.
MATERIALS AND METHODS
These studies were carried out using BD FACSCanto II flow cytometers equipped with violet, blue, and red lasers (BD Biosciences, San Jose, CA), running BD FACSDiva software v. 6.1.3 (BD Biosciences). Instrument performance was verified daily using the BD Cytometer Setup & Tracking Beads kit (BD Biosciences). Samples were prepared as described previously (6). Spectral compensation was carried out using either a commercially available set of polystyrene microparticles (BD CompBeads, BD Biosciences) according to the manufacturer's protocol or normal peripheral blood lymphocytes gated by FSC and SSC. AmCyan-conjugated monoclonal antibodies recognizing CD8 (clone SK1) and CD45 (clone 2D1) were purchased from BD Biosciences.
In our initial studies, we found that spillover from AmCyan into spectrally adjacent detectors (V450 and FITC) (7) appeared to be incorrectly compensated when compensation values were calculated on the basis of control studies using a commercially available set of polystyrene microparticles (BD CompBeads, BD Biosciences) according to the manufacturer's protocol. Specifically, spillover from AmCyan into V450 appeared to be overcompensated (Fig. 1A), while spillover from AmCyan into FITC appeared to be undercompensated (Fig. 1B). Such anomalous compensation was observed with both AmCyan-conjugated antibodies tested. In contrast, when compensation values were calculated using unstained and singly stained lymphocytes instead of polystyrene microparticles as controls, spillover into V450 and FITC from cells stained brightly with AmCyan-conjugated antibodies appeared to be appropriately compensated (Figs. 1C and 1D). These differences in the compensation of spillover from AmCyan into V450 and FITC using cells rather than polystyrene microparticles as compensation controls were both reproducible (n = 30) and significant (P < 0.0001, two-tailed Wilcoxon signed-rank test) (Fig. 2). Compensation for spillover of AmCyan into PE also differed significantly when cells rather than polystyrene microparticles were used as compensation controls (Fig. 2).
Differences in the compensation of V450 and FITC spillover into AmCyan as calculated using cells rather than polystyrene microparticles as compensation controls were also detectable (Fig. 2); however, they were of a smaller magnitude and were not readily apparent in routine multicolor experiments. To test whether such differences in compensation calculated using cells rather than polystyrene microparticles as controls might extend to fluorochromes that are excited by blue or red lasers, we compared the compensation values of FITC into PE, PE into FITC, and APC into APC-H7 as determined using polystyrene microparticle-based and cellular compensation controls. As shown in Figure 2, compensation values as determined by both methods were indistinguishable (P = N.S., two-tailed Wilcoxon signed-rank test; n = 20).
Because we observed anomalous compensation of a violet-excited dye (AmCyan), but not blue- (FITC and PE) or red-excited dyes (APC) when using polystyrene microparticles as compensation controls, and because cellular autofluorescence is more strongly initiated by violet than red excitation (8), we evaluated the autofluorescence characteristics of the different microparticles that comprise the BD CompBeads set: anti-mouse-IgG-κ particles and negative control particles with no antibody-binding capacity. If the autofluorescence properties of these two different microparticles were different, compensation would be expected to be incorrect (4). As shown in Figure 3, autofluorescence was substantially higher in the violet detectors than the blue and red detectors. However, the autofluorescence properties of the negative control (Fig. 3A) and anti-mouse-IgG-κ (Fig. 3B) microparticles were indistinguishable.
We report that more accurate compensation of spectral overlap involving the violet-excited fluorochrome, AmCyan, may be obtained by using cellular compensation controls rather than a commercially available set of polystyrene microparticles (BD CompBeads). Although the precise mechanism for the apparently anomalous compensation of AmCyan as calculated using BD CompBeads is uncertain, differences in autofluorescence properties between the negative control and anti-mouse-IgG-kappa microparticles that comprise the set would appear not to be responsible for the effect (Fig. 3). Recently, published data suggest that others may have encountered similar difficulties in compensating AmCyan (9). In panel 3 of Figure 1 from that study (9), the signal measured in FITC increases in a coordinated fashion with increasing AmCyan fluorescence, as would be expected if spillover from AmCyan into FITC were undercompensated. However, it is important to note that any potential undercompensation of AmCyan spillover into FITC in that study would not have affected its conclusions, as the authors' multiparameter gating strategy ensured that plasma cells were correctly identified (9).
Although antibody-capture beads are extremely valuable as compensation controls (4), particularly when compensating fluorochrome-labeled antibodies that recognize cellular antigens expressed infrequently or at low levels, our data suggest that polystyrene microparticles may not be ideal surrogates for cells in all cases. Indeed, technical data sheets accompanying conjugates of the recently released dye BD Horizon V500 indicate that cellular compensation controls may be preferable to beads for both AmCyan and V500 (10). Whether this principle holds true for other dyes that emit in this region requires further investigation. Finally, because we have only evaluated one commercially available set of compensation particles in this study, we wish to emphasize that the determination of whether spillover of AmCyan can be correctly compensated using other microparticle-based systems would require additional study.
We gratefully acknowledge the expert advice and guidance of Dr. Brent Wood in the implementation of polychromatic flow cytometry in our laboratory, and the outstanding technical assistance of past and present members of the Special Hematology Laboratory at Hartford Hospital: Darren Brown, Beth Constant, Sheila Fuller, Len Iorio, Sophie Santoro, and Judy Thibeault. We are also grateful to Dr. Charles Goolsby for his encouragement and helpful suggestions for improving the manuscript and to both Reviewers for their constructive criticism and suggestions.