Abstract
Immune monitoring of patients on a single-cell level is becoming increasingly important in various diseases. Due to the often very limited availability of human specimens and our increased understanding of the immune systems there is an increasing demand to analyze as many markers as possible simultaneously in one panel. Full spectrum flow cytometry is emerging as a powerful tool for immune monitoring since 5-laser instruments enable characterization of 40 parameters or more in a single sample. Nevertheless, even if only machines with fewer lasers are available, development of novel fluorophore families enables increasing panel sizes. Here, we demonstrate that careful panel design enables the use of 31-color panels on a 3-laser Cytek ® Aurora cytometer for analyzing human peripheral blood leuko cytes, without the need for custom configuration and using only commercially available fluorochromes. The panel presented here should serve as an example of a 31-fluorochrome combination that can be resolved on a 3-laser full spectrum cyt ometer and that can be adapted to comprise other (and possibly more) markers of interest depending on the research focus. KEYWORDS 3-laser system, cytek aurora, full spectrum flow cytometry, panel design 1 | INTRODUCTION Immune monitoring of patients on a single-cell level is becoming increas ingly important to characterize pathomechanisms and identify promising treatment strategies in various diseases (Chattopadhyay et al., 2014; Sanjabi & Lear, 2021). However, the availability of human specimens is often very limited with regards to sample size, that is, how many cells can be analyzed from a single patient. In addition, our increased understanding of immune cell subtypes and their relationships dictates the need to ana lyze as many markers as possible simultaneously in one panel. Therefore, full spectrum flow cytometry is emerging as a powerful tool for immune monitoring (Bonilla et al., 2020; Chen et al., 2023; Jensen & Wnek, 2021; Monneret & Venet, 2016), as 5-laser instruments (usually equipped with a 355, 405, 488, 561, and 640 nm laser) enable characterization of 40 parameters or more in a single sample (Park et al., 2020; Sahir et al., 2020). Nevertheless, even if only machines with fewer lasers are available due to budget restrictions or other limitations, development of novel fluorophore families enables increasing panel sizes. It has recently been demonstrated that 34 colors can be distinguished on a 3-laser (405, 488, 640 nm) full spectrum cytometer by increasing the use of the near-infrared spectrum (Seong et al., 2022). However, this requires access to novel fluorochromes that are not yet commercially available. Here, we demonstrate that careful panel design enables the use of 31-color panels on a 3-laser Cytek ® Aurora cytometer without the Received: 2 January 2023 Revised: 4 May 2023 Accepted: 11 May 2023 DOI: 10.1002/cyto.b.22126 This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes. © 2023 The Authors. Cytometry Part B: Clinical Cytometry published by Wiley Periodicals LLC on behalf of International Clinical Cytometry Society. Cytometry. 2023;1–7. wileyonlinelibrary.com/journal/cytob 1 need for custom configuration and using only commercially available fluorochromes. This approach is applied to analyze immune cell dynamics in peripheral blood from humans. The panel presented here should serve as an example of a 31-fluorochrome combination (rather than a marker combination) that can be distinguished on a 3-laser full spectrum cytometer and that can be adapted to comprise other markers of interest depending on the research focus. 2 | MATERIALS AND METHODS Anti-coagulated (EDTA) whole blood was obtained from healthy vol unteers and processed immediately. 200 μL of whole blood were incubated with a fixable viability dye and fluorochrome-conjugated antibodies (see Table 1) in blocking buffer (PBS +2% BSA +2% normal mouse/human/rabbit/rat serum) for 20 min at room temperature. Lysis of red blood cells was performed by adding 2 mL of 1x BD Pharm Lyse™ Lysing buffer (BD Biosciences, #555899) without a prior washing step and incubating the samples for further 10 min at room temperature. Cells were washed with staining buffer (PBS +2 mM EDTA) and lysis was performed a second time. Subsequently, cells were fixed for 10 min in PBS containing 2% paraformaldehyde at room temperature. After the final wash, cells were resuspended in 200 μL staining buffer for analysis. ### 2.2. | Panel development The panel presented here was developed using antibodies that had already been previously validated and optimized on conventional TABLE 1 Reagents used in the 31-color panel. Specificity Fluorochrome Clone Company Catalog no. Concentration Purpose CD1c BV421 L161 BioLegend 331526 1:200 Classical DC type 2 CD31 Super Bright™ 436 WM-59 Invitrogen 62031942 1:200 Endothelial cells CD66b Pacific Blue™ G10F5 BioLegend 305112 1:200 Granulocytes CD4 BV480 RPA-T4 BD Biosciences 746541 1:200 T-helper cells CD16 V500 3G8 BD Biosciences 561394 1:200 Neutrophils, monocytes CD3 Spark Violet™ 538 UCHT1 BioLegend 300484 1:200 T cells HLA-DR BV570 L243 BioLegend 307638 1:200 Antigen presenting cells TCR Vα7.2 BV605 3C10 BioLegend 351720 1:200 MAIT cells CD15 BV650 W6D3 BioLegend 323033 1:200 Granulocytes CD303 BV711 201A BioLegend 354233 1:100 Plasmacytoid DC CD56 BV750 5.1H11 BioLegend 362556 1:200 NK cells TLR2 BV785 11G7 BD Biosciences 742771 1:200 Pattern recognition receptor CD8 BB515 RPA-T8 BD Biosciences 564526 1:800 Cytotoxic T cells CD11b FITC M1/70 BioLegend 101206 1:200 Pan-myeloid cells CD45 Alexa Fluor™ 532 HI30 Invitrogen 58045942 1:200 Pan-immune cells CX3CR1 PE 2A9-1 BD Biosciences 12609942 1:200 Monocytes CCR2 PE/Dazzle™ 594 K036C2 BioLegend 357222 1:200 Monocytes CD45RO PE/Fire™ 640 UCHL1 BioLegend 304263 1:200 Memory T cells CD19 PE-Cy5 HIB19 BioLegend 302210 1:200 B cells CD127 BB700 HLA-7R-M21 BD Biosciences 566398 1:80 T cell subtypes CD14 PerCP-Cy5.5 M5E2 BioLegend 301824 1:200 Classical monocytes CD45RA PerCP-eFluor™ 710 GRT22 Invitrogen 46046842 1:200 Naive T cells CD161 PE-Cy7 HP-3G10 Invitrogen 25161942 1:200 NK/NKT cells, MAIT cells CXCR3 PE/Fire™ 810 G025H7 BioLegend 353759 1:200 Chemokine receptor CD123 APC 6H6 BioLegend 306011 1:200 Plasmacytoid DC, basophils CD163 Alexa Fluor™ 647 MAC 2–158 Invitrogen 51163742 1:200 Monocytes CD27 Spark NIR™ 685 O323 BioLegend 302855 1:200 T cells CD11c Alexa Fluor™ 700 Bu15 BioLegend 337220 1:200 Classical DC CD33 APC-Cy7 WM53 BioLegend 303442 1:200 Pan-myeloid cells CD25 APC/Fire™ 810 M-A251 BioLegend 356150 1:200 Regulatory T cells Viability Zombie NIR™ / Biolegend 423106 1:5000 Cell viability flow cytometers in our lab. The different iterations of the panel used during panel development can be found in Table S2. In brief, the first color combination (23 fluorochromes) tested was based on the Cytek ® Aurora Fluorochrome Selection Guidelines for 3-laser machines (can be accessed at https://cytekbio.com/pages/ fluorochrome-guides), followed by stepwise addition of fluoro chromes that showed a distinct spectral profile from already used colors. In addition, markers with a low expression profile in healthy individuals (e.g., PD1 or PDL1) were gradually replaced with markers with higher expression (such as CD163 and CD15) to ensure that all fluorochromes were detected in the 31-color panel and therefore allowed assessment, whether all 31 colors could indeed be resolved from each other. In general, antibodies optimized on conventional cytometers showed equally good resolution when analyzed on the Cytek ® Aurora at the same concentration (1:200, see also Table 1). For anti bodies showing low resolution in the 31-color panel, titration as well as fluorescent minus one (FMO) controls was performed to ensure these were used at the optimal concentrations to achieve separation from the background. Figure S1 shows titration information and chosen concentrations for CD303 (Figure S1A; used to identify plas macytoid dendritic cells in Figure 1) and CD127 (Figure S1B; used to identify T-helper cell subsets in Figure 2), the two markers with the lowest resolution in our panel. CD8-BB515 displayed a signal over the detection limit when used on the full spectrum cytometer at the concentration optimized on conventional machines, possibly due to increased sensitivity of the Cytek ® Aurora. This antibody was titrated as well to determine the optimal concentration to achieve a good separation of positive and negative populations without reach ing the detection limit (Figure S1C). TABLE 2 Reference controls used for unmixing. Single stains of each fluorochrome-conjugated antibody were performed on cells (human peripheral blood leukocytes) or Biolegend ® Compensation Beads as indicated. Fluorochrome Marker Ref ctrl type Peak channel BV421 CD1c Biolegend ® Compensation Beads V1 Super Bright™ 436 CD31 Biolegend ® Compensation Beads V2 Pacific Blue™ CD66b Biolegend ® Compensation Beads V3 BV480 CD4 Biolegend ® Compensation Beads V5 V500 CD16 Biolegend ® Compensation Beads V7 Spark Violet™ 538 CD3 Biolegend ® Compensation Beads V7 BV570 CD11b Biolegend ® Compensation Beads V8 BV605 TCR Vα7.2 Biolegend ® Compensation Beads V10 BV650 CD15 Biolegend ® Compensation Beads V11 BV711 CD303 Biolegend ® Compensation Beads V13 BV750 CD56 Biolegend ® Compensation Beads V14 BV785 TLR2 Biolegend ® Compensation Beads V15 BB515 CD8 Biolegend ® Compensation Beads B1 FITC CD11b Biolegend ® Compensation Beads B2 Alexa Fluor™ 532 CD45 Biolegend ® Compensation Beads B3 PE CX3CR1 Biolegend ® Compensation Beads B4 PE/Dazzle™ 594 CCR2 Biolegend ® Compensation Beads B6 PE/Fire™ 640 CD45RO Biolegend ® Compensation Beads B7 PE-Cy5 CD19 Biolegend ® Compensation Beads B8 BB700 CD127 Biolegend ® Compensation Beads B9 PerCP-Cy5.5 CD14 Peripheral blood leukocytes B9 PerCP-eFluor™ 710 CD45RA Biolegend ® Compensation Beads B10 PE-Cy7 CD161 Biolegend ® Compensation Beads B13 PE/Fire™ 810 CXCR3 Biolegend ® Compensation Beads B14 APC CD123 Biolegend ® Compensation Beads R1 Alexa Fluor™ 647 CD163 Biolegend ® Compensation Beads R2 Spark NIR™ 685 CD27 Biolegend ® Compensation Beads R3 Alexa Fluor™ 700 CD11c Peripheral blood leukocytes R4 Zombie NIR™ Viability Peripheral blood leukocytes R6 APC-Cy7 CD33 Biolegend ® Compensation Beads R7 APC/Fire™ 810 CD25 Biolegend ® Compensation Beads R8 Unstained control / Peripheral blood leukocytes / HAMMERICH ET AL. 3 The viability dye (Zombie NIR) showed residual staining in live cells when used at the concentration recommended by the manufac turer (1:1000) and was titrated to achieve a concentration that does not stain live cells while still providing a clear signal in dead cells (1:5000). This information can be found in Figure S2. Samples were analyzed on a standard issue Cytek ® Aurora Cytometer equipped with 3 lasers and 38 detectors (Table S1) and running Spec troflo ® v3.0.3. Instrument QC was performed daily according to the manufacturer's instructions. Samples were acquired on low to medium speed. Data analysis was performed with FCSExpress 7.14.0020 (Denovo Software, Pasadena, CA, USA). FIGURE 1 Gating strategy for myeloid cell subsets. Percentages indicate relative frequencies among all CD45+ cells (for gating strategy to determine the total number of CD45+ cells see Figure S6). cDC, conventional dendritic cells; clMon, classical monocytes; intMono, intermediate monocytes; ncMono, non-classical monocytes; pDC, plasmacytoid dendritic cells. [Color figure can be viewed at wileyonlinelibrary.com] ### 2.5. | Spectral unmixing To generate reference controls for spectral unmixing, single stains of each antibody were performed on human peripheral blood leukocytes or Biole gend ® Compensation Beads (Biolegend, Inc., Catalog #424601) as indi cated in Table 2. In order to create bright reference controls for all markers, including those with low frequency in the blood (such as, e.g., CD303), we opted to use Biolegend ® Compensation Beads for refer ence controls for all fluorochromes that show identical emission spectra on beads and cells. Of note, cells were used for single stains of PerCP Cy5.5 and Alexa Fluor™ 700, because their emission spectra differ signifi cantly on cells compared to compensation beads (Figure S3). For the via bility dye (Zombie NIR™) and the unstained control we used cells as well. Spectral unmixing was performed in SpectroFlo ® 3.0.3. FIGURE 2 Gating strategy for lymphocytes. Percentages indicate relative frequencies among all CD45+ cells (for gating strategy to determine the total number of CD45+ cells see Figure S6). MAIT, mucosal associated invariant T; NK, natural killer cell; NKT, natural killer T cell; Treg, regulatory T cell. [Color figure can be viewed at wileyonlinelibrary.com] HAMMERICH ET AL. 5 1 5 5 2 4 9 5 7 , 3 | RESULTS The complexity index for the full 31-color panel as well as the similar ity indices for all fluorochrome pairs are given in Figure S4. Spectro Flo ® indicated a complexity index of 55.34 and pairwise comparison between all fluorochromes revealed six combinations with a similarity of more than 90%: BV421 versus Super Bright™ 436, BB515 versus FITC, PE-Cy5 versus PE/Fire™ 640, PerCP-Cy5.5 versus BB700, PerCP-Cy5.5 versus PerCP-eFluor™ 710 and APC versus Alexa Fluor™ 647. Nevertheless, spectral unmixing was still able to accu rately distinguish those fluorochromes from each other (Figure S5). However, the high similarity leads to considerable signal spread between both channels and mandates careful panel design. By choos ing mutually exclusive marker pairs that are not co-expressed on the same cell types, we were able to include even the two combinations with the highest similarity, BB515/FITC and BB700/PerCP-Cy5.5, in our panel (Figure S5A,B). The panel used here is able to identify all major immune cell types in human peripheral blood (Table 1), including low frequency populations such as basophilic granulocytes and dendritic cells. Some functional markers, such as chemokine or pattern recognition receptors and lym phocyte activation/memory markers were included as well. An example gating strategy of myeloid cells from fresh human whole blood is shown in Figure 1. After exclusion of doublets, dead cells and endothelial cells (CD31+ CD45 ), lymphocytes were excluded from the CD45+ leukocytes by sequentially gating on CD3 CD19 and CD56 TCRVα7.2 cells. Neutrophils were identified as CD66b+ CD16+ and eosinophils as CD66b+ CD16 CD15+ CD33+. The remaining CD66b- population included various subtypes of monocytes, baso phils and dendritic cells. Classical monocytes were defined as CD14+ CD16 , non-classical monocytes as CD14 CD16+ and intermediate monocytes as CD14+ CD16+; classical monocytes and intermediate monocytes also express CD163. Classical and non-classical monocytes were further analyzed for their expression of CCR2, CX3CR1, TLR2 and CD11b. Among the CD14 CD16 population, basophils were identified as HLA-DR CD123+, plasmacytoid dendritic cells as HLA-DR+ CD123+ CD303+ and conventional dendritic cells as HLA-DR+ CD11c+ CD1c+. Figure 2 shows the gating strategy for lymphocyte subsets from the same staining panel. After gating on SSC-low cells, doublet, dead cell and endothelial cells exclusion and identification of live CD45+ cells, remaining myeloid cells were excluded by gating on CD163 CD66b cells. Subsequently, B cells were identified as CD19+, NK cells as CD56+ CD3 , NKT cells as CD56+ CD3+ and T cells as CD56 CD3+. NK cells were further subdivided into CD56bright NK cells (CD56high CD16low) and CD56dim NK cells (CD56 low CD16high). Among all T cells, we identified MAIT cells (CD161+ TCRVa7.2+), other unconventional T cells (CD161+ TCRVa7.2 ) and conventional T cells (CD161 ). The latter were separated into CD4+ T cells, CD8+ T cells and regulatory T cells (Treg, CD4+ CD25+ CD127 ). CD4+ and CD8+ T cells were then analyzed for their expression of CD45RA, CD45RO, CXCR3 and CD27, which can be used to distinguish functionally distinct subsets, such as naïve and memory T cells. 4 | CONCLUSIONS Here, we demonstrate that upon careful panel design a standard issue 3-laser full spectrum cytometer can resolve 31 fluorochromes that are excited off the violet (405 nm), blue (488 nm) and red (640 nm) laser. Most importantly, fluorochrome pairs with very high similarity and considerable signal spread should not be used for markers that are co-expressed on the same cell types. All fluo rochromes and conjugated antibodies used here are commercially available simplifying modification of the panel and facilitating replacement of markers of interest according to the research needs of the respective project. It is important to note that all reagents should be titrated carefully, as too low concentrations could impair separation from the background while too high concentrations may result in greater signal spread and therefore increased panel com plexity. In addition, it seems plausible that single fluorochromes could be replaced by other fluorochromes with identical or almost identical emission spectra (eg. AlexaFluor™ 488 instead of FITC, or APC/Fire™ 750 instead of APC-Cy7), further increasing the num ber of readily available antibodies. However, this might still influ ence the complexity of the panel and should be tested carefully before being applied. Finally, this panel currently only utilizes 29 of the available 38 detectors as channels for the peak fluorescence of individual fluorochromes. With the ever-expanding number of companies developing novel fluorophore families, it is more than likely that panels for 3-laser systems can be expanded to more than 31 colors in the future. The authors thank Désirée Kunkel and the BIH Cytometry Core Facil ity for technical support and helpful discussions. This work was supported by the German Research Foundation (DFG; CRC1382, SFB/TRR 296 and SPP2306), the Else-Kröner-Fresenius Stiftung (Grant-ID 2021_EKEA.145) and Charité 3Rj Replace— Reduce—Refine. CONFLICT OF INTEREST STATEMENT The authors declare no conflict of interest. ETHICS STATEMENT The study was conducted according to the guidelines of the Declara tion of Helsinki, and approved by the local ethics committee at Char ité Universitätsmedizin Berlin, Germany (ethical approval number EA2/065/21). PATIENT CONSENT STATEMENT Informed consent was obtained from all subjects involved in the study and is kept on file. ORCID Linda Hammerich https://orcid.org/0000-0003-0557-3927 Bonilla, D. L., Reinin, G., & Chua, E. (2020). Full Spectrum flow cytometry as a powerful Technology for Cancer Immunotherapy Research. Fron tiers in Molecular Biosciences, 7, 612801. Chattopadhyay, P. K., Gierahn, T. M., Roederer, M., & Love, J. C. (2014). Single-cell technologies for monitoring immune systems. Nature Immu nology, 15(2), 128–135. Chen, X., Gao, Q., Roshal, M., & Cherian, S. (2023). Flow cytometric assess ment for minimal/measurable residual disease in B lymphoblastic leukemia/lymphoma in the era of immunotherapy. Cytometry. Part B, Clinical Cytometry, 104, 205–223. Jensen, H. A., & Wnek, R. (2021). Analytical performance of a 25-marker spectral cytometry immune monitoring assay in peripheral blood. Cyto metry Part A, 99(2), 180–193. Monneret, G., & Venet, F. (2016). Sepsis-induced immune alterations monitoring by flow cytometry as a promising tool for individual ized therapy. Cytometry. Part B, Clinical Cytometry, 90(4), 376–386. Park, L. M., Lannigan, J., & Jaimes, M. C. (2020). OMIP-069: Forty-color full spectrum flow cytometry panel for deep immunophenotyping of major cell subsets in human peripheral blood. Cytometry. Part A, 97(10), 1044–1051. Sahir, F., Mateo, J. M., Steinhoff, M., & Siveen, K. S. (2020). Development of a 43 color panel for the characterization of conventional and uncon ventional T-cell subsets, B cells, NK cells, monocytes, dendritic cells, and innate lymphoid cells using spectral flow cytometry. Cytometry A, 1–7. Sanjabi, S., & Lear, S. (2021). New cytometry tools for immune monitoring during cancer immunotherapy. Cytometry. Part B, Clinical Cytometry, 100(1), 10–18. Seong, Y., Nguyen, D. X., Wu, Y., Thakur, A., Harding, F., & Nguyen, T. A. (2022). Novel PE and APC tandems: Additional near-infrared fluoro chromes for use in spectral flow cytometry. Cytometry. Part A, 101(10), 835–845. SUPPORTING INFORMATION Additional supporting information can be found online in the Support ing Information section at the end of this article. How to cite this article: Hammerich, L., Shevchenko, Y., Knorr, J., Werner, W., Bruneau, A., & Tacke, F. (2023). Resolving 31 colors on a standard 3-laser full spectrum flow cytometer for immune monitoring of human blood samples. Cytometry Part B: Clinical Cytometry, 1–7. https://doi.org/10.1002/cyto.b. 22126 HAMMERICH ET AL. 7
Showing the abstract — retrieve the full paper via the Exa API.