Flow cytometry and mass cytometry are the methods of choice to dissect the cellular makeup of biological samples, including blood and disaggregated tissue. Most of these samples are tremendously heterogeneous and comprise a plurality of specialized cell types that serve distinct functions in health and disease. Flow cytometry and, more recently, mass cytometry have helped to map this heterogeneity and allocate particular functions to particular cell types, thus advancing single-cell systems biology and unveiling novel therapeutic targets. Although early flow cytometry was limited to few (very often four) colors, the past decade has entailed compelling advances in the technical capacity and performance of flow cytometric instruments. This evolution was complemented by the advent of mass cytometry, which uses discrete isotopes of rare-earth elements to detect target Ags through a mass spectrometry readout. Both modern flow cytometry and mass cytometry easily support the analysis of 25+ parameters per individual cell (thus theoretically resolving more than 3 × 107 different populations), and even less sophisticated (e.g., small desktop) devices frequently provide an 18+ parameter configuration allowing in-depth multicolor profiling. Thus, the number of investigated Ags is usually no longer limiting.

Dump gating refers to the process of intentionally excluding cells from a flow cytometry/mass cytometry analysis that stain positive for a particular (often lineage-restricted) marker or marker combination, including viability stains. In the past, this approach was instrumental to overcome limitations in the number of available fluorescence channels by pooling several markers in one channel and gating out these lineage-positive cells (dump gate). Using this strategy, the target population could be better defined or characterized for the expression of additional markers not otherwise possible. However, state-of-the-art multicolor analyses make this strategy largely obsolete, and we will highlight in this commentary that dump gating also holds some inherent biological conceptual pitfalls, such as unforeseen Ag expression across lineage boundaries (potentially leading to target cell exclusion) and nonuniform Ag expression in the population intended to be dumped (potentially leading to off-target cell inclusion).

In humans, the CD20 surface Ag is a well-established marker for cells of the B cell lineage. However, a recent study revealed a distinct subset of circulating CD3+ T cells (∼3–5%) coexpressing CD20 enriched in CD8+, CD45RO+, and CCR7 cells (1). These CD3+CD20+ T cells are also found in lymphoid organs and cerebrospinal fluid and are depleted by CD20-targeting Abs, such as rituximab.

CD56 is an archetypical marker for NK cells and NK T cells, the former of which belong to the group 1 innate lymphoid cells (ILCs). In reality, CD56 is also expressed on myeloid dendritic cell and plasmacytoid dendritic cell subsets (2) as well as on HLA-DR+CD14+ peripheral blood monocytes (3) and γδ T cells (4). CD56 is also aberrantly expressed in a variety of malignant diseases, including blood cancers and carcinomas (4). Moreover, in chronic myeloid leukemia, CD56 expression is detected at the CD34+CD38 stem cell stage (5), suggesting that hematopoietic stem cells might de novo express this surface marker when undergoing leukemic transformation.

Murine B220 (a specific isoform of CD45) is considered a pan–B cell marker and as such is frequently used to exclude (or define) B cells. However, B220 is also expressed by CD19CD11c+ plasmacytoid dendritic cells in bone marrow and spleen (6) and further marks activated T cells primed for apoptosis (7). Considering this and the ambiguity of human CD20 alike (1), many researchers now routinely use the CD19 marker to more accurately identify B cells in flow cytometric experiments. Other examples are the myeloid cell marker CD11b, which is also found on lymphoid cells (811), and the NK cell marker NK1.1, which is not expressed by all NK cell subsets (e.g., NK1.1CD49bCD3CD122+ uterine NK cells) (12).

Recently, a high-dimensional analysis of mucosal tissues from inflammatory bowel diseases performed using a 32-Ab panel oriented to explore major adaptive and innate immune cells unexpectedly revealed the expression of CD11c on a subset of CD56 ILC3-like cells. The ILC scientific community commonly places CD11c in the dump channel, thus precluding the detection of this cellular subset promptly identified by unbiased mass cytometry panel design (13).

In settings other than hematopoiesis, lineage markers also often show a more complex expression pattern. In epithelial tumors (so-called carcinomas), CD326/EpCAM is frequently used to define the malignant cell fraction, yet CD326 also marks nontransformed epithelial cells and may further not be expressed by mesenchymally transitioned tumor cells (14). Similarly, CD31, although commonly perceived as a strictly endothelial marker, is also expressed by a distinct population of bone marrow CD45+ cells that bears high angiogenic/vasculogenic potential (15).

These data concordantly suggest that many of the commonly used lineage markers are not as lineage-specific as their names would imply (Table I). Two principal types of error are associated with this circumstance: First, in exploratory analyses of new cell subsets, the exact identity of the investigated cells is generally unknown. Exclusion of lineage-positive cells by a dump gate may therefore hold back important information about the newly identified cells (e.g., information about population size, cellular makeup/heterogeneity, etc.). Second, nonuniform expression of lineage markers may prohibit the exclusion of all desired cells using a respective dump gate [e.g., CD14dim monocytes (16)]. As a result, unintended carryover of cells may affect (or even distort) the analysis, and this is especially relevant for rare cell subsets, which are naturally susceptible to contamination by more abundant populations. Finally, dump gating also holds some practical drawbacks. When an Ab in a lineage combination is unintentionally omitted (i.e., forgotten) or simply does not work, this will often not be noticed because a signal is detected in the corresponding channel regardless. It is unlikely that this will happen with a one-marker-one-channel approach. Moreover, the exact setting of a dump gate can sometimes be difficult because several populations with inherently different expression levels and/or autofluorescences are artificially forced into one region (one size fits all principle). In theory, this would need to be accounted for by the use of respective fluorescence-minus-one controls within the dump channel to determine the exact threshold for cut-off, but will this ever be done? Dead cell discrimination is also frequently performed using a dump gate of various markers. Although this approach is reasonable for some applications, one should be aware that important information on sample integrity is lost (e.g., the overall viability after tissue disaggregation). Combining a dead cell marker with other markers in a dump channel furthermore precludes inferences on the specific share of the population of interest in the total viable cell fraction.

Table I.
Expression profile of commonly used lineage markers
MarkerUsed To Exclude (or Define)Also Expressed onReference
B220 (mouse)a B cells pDC, activated T cells (6, 7
CD11b (mouse) Myeloid cells B-1/B-2 B cells, NK, NKT (811
CD11c (human) Myeloid cells CD56-, ILC3-like cells (rectal mucosa) (13
CD20 (human)a B cells T cells (1
CD31 (human) Endothelial cells CD45+ bone marrow cells (15
CD56 (human) NK and NK T cells pDC, monocytes, γδ T cells, cancer cells, CML stem cell (25
Marker Used To Exclude (or Define) Not Expressed on Reference 
NK1.1 (mouse) NK cells NK1.1 uterine NK cells (12
CD3 (mouse and human) T cells Activated T cells (22
CD14 (human) Monocytes CD14dim monocytes (16
CD326 (human) Carcinoma cells EMT tumor cells (14
MarkerUsed To Exclude (or Define)Also Expressed onReference
B220 (mouse)a B cells pDC, activated T cells (6, 7
CD11b (mouse) Myeloid cells B-1/B-2 B cells, NK, NKT (811
CD11c (human) Myeloid cells CD56-, ILC3-like cells (rectal mucosa) (13
CD20 (human)a B cells T cells (1
CD31 (human) Endothelial cells CD45+ bone marrow cells (15
CD56 (human) NK and NK T cells pDC, monocytes, γδ T cells, cancer cells, CML stem cell (25
Marker Used To Exclude (or Define) Not Expressed on Reference 
NK1.1 (mouse) NK cells NK1.1 uterine NK cells (12
CD3 (mouse and human) T cells Activated T cells (22
CD14 (human) Monocytes CD14dim monocytes (16
CD326 (human) Carcinoma cells EMT tumor cells (14

This table does not claim completeness.

a

For exclusion (or definition) of B cells, the CD19 marker is to be favored.

CML, chronic myeloid leukemia; EMT, epithelial-to-mesenchymal transition; pDC, plasmacytoid dendritic cell.

Although dump gating has traditionally been an important strategy to overcome limitations in the number of fluorescence channels, modern instruments typically support the measurement of 18 parameters and more, thus making dump gating in many cases dispensable for proper display and analysis of the target cell population. Considering this and the above-mentioned drawbacks, the notorious use of a dump gate seems to base more on historical grounds rather than actual need or experimental logic and should be scrutinized by the individual researcher on a case-by-case basis. In other words, knowledge about the cells of interest under physiological and pathological conditions and the potential pitfalls provided, dump gating can still be productive, for instance, to better demarcate other markers (17), to avoid loss of resolution due to excessive residual spillover from multiple channels following compensation (18), or to reduce background in specific applications such as multimer staining protocols (19). Despite the highest number of available channels, these examples should also be considered in the case of mass cytometry, which may face particular challenges with differential isotope transmission (20) and channel cross-talk (21) as well. Indeed, less bright isotopes may need a better resolution or a background reduction, and the number of markers may need to be reduced and combined in a dump gate to leave important biological questions out of the cross-talk. Systematic investigations of Ag expression on all major leukocyte populations currently under way will further help to define the exclusiveness and comprehensiveness of currently used lineage markers.

In summary, this commentary should raise awareness for the following flow cytometry/mass cytometry reality: the number of channels is no longer limiting, but the (unnecessary) use of dump gating limits scientific discovery. Our article shall encourage researchers to come to an active and well-founded decision whether to use dump gating in their experiments—a decision that ideally goes beyond mere habit or the desire for complexity reduction.

This work was supported by the Austrian Research Promotion Agency under Grant 858057 (HD FACS project), the French government program Investissement d’avenir: Equipements d’Excellence–2010 FlowCyTech under Grant ANR-10-EQPX-02-01, and the Infrastructures Nationales en Biologie et Santé–2011 Infectious Disease Models and Innovative Therapies under Grant ANR-11-INBS-0008. M.B. is supported by an Erwin Schrödinger Fellowship from the Austrian Science Fund (Grant J-3807).

Abbreviation used in this article:

     
  • ILC

    innate lymphoid cell.

1
Schuh
,
E.
,
K.
Berer
,
M.
Mulazzani
,
K.
Feil
,
I.
Meinl
,
H.
Lahm
,
M.
Krane
,
R.
Lange
,
K.
Pfannes
,
M.
Subklewe
, et al
.
2016
.
Features of human CD3+CD20+ T cells
.
J. Immunol.
197
:
1111
1117
.
2
Roothans
,
D.
,
E.
Smits
,
E.
Lion
,
J.
Tel
,
S.
Anguille
.
2013
.
CD56 marks human dendritic cell subsets with cytotoxic potential
.
OncoImmunology
2
:
e23037
.
3
Sconocchia
,
G.
,
K.
Keyvanfar
,
F.
El Ouriaghli
,
M.
Grube
,
K.
Rezvani
,
H.
Fujiwara
,
J. P.
McCoy
Jr.
,
N.
Hensel
,
A. J.
Barrett
.
2005
.
Phenotype and function of a CD56+ peripheral blood monocyte
.
Leukemia
19
:
69
76
.
4
Van Acker
,
H. H.
,
A.
Capsomidis
,
E. L.
Smits
,
V. F.
Van Tendeloo
.
2017
.
CD56 in the immune system: more than a marker for cytotoxicity?
Front. Immunol.
8
:
892
.
5
Janssen
,
J. J.
,
W.
Deenik
,
K. G.
Smolders
,
B. J.
van Kuijk
,
W.
Pouwels
,
A.
Kelder
,
J. J.
Cornelissen
,
G. J.
Schuurhuis
,
G. J.
Ossenkoppele
.
2012
.
Residual normal stem cells can be detected in newly diagnosed chronic myeloid leukemia patients by a new flow cytometric approach and predict for optimal response to imatinib
.
Leukemia
26
:
977
984
.
6
Nikolic
,
T.
,
G. M.
Dingjan
,
P. J.
Leenen
,
R. W.
Hendriks
.
2002
.
A subfraction of B220(+) cells in murine bone marrow and spleen does not belong to the B cell lineage but has dendritic cell characteristics
.
Eur. J. Immunol.
32
:
686
692
.
7
Renno
,
T.
,
A.
Attinger
,
D.
Rimoldi
,
M.
Hahne
,
J.
Tschopp
,
H. R.
MacDonald
.
1998
.
Expression of B220 on activated T cell blasts precedes apoptosis
.
Eur. J. Immunol.
28
:
540
547
.
8
Ghosn
,
E. E.
,
Y.
Yang
,
J.
Tung
,
L. A.
Herzenberg
,
L. A.
Herzenberg
.
2008
.
CD11b expression distinguishes sequential stages of peritoneal B-1 development
.
Proc. Natl. Acad. Sci. USA
105
:
5195
5200
.
9
Park
,
S.
,
H.
Sim
,
H. I.
Kim
,
D.
Jeong
,
G.
Wu
,
S. Y.
Cho
,
Y. S.
Lee
,
H. J.
Kwon
,
K.
Lee
.
2017
.
CD11b regulates antibody class switching via induction of AID
.
Mol. Immunol.
87
:
47
59
.
10
Han
,
Y.
,
Z.
Jiang
,
Z.
Chen
,
Y.
Gu
,
Y.
Liu
,
X.
Zhang
,
X.
Cao
.
2015
.
Pathogen-expanded CD11b+ invariant NKT cells feedback inhibit T cell proliferation via membrane-bound TGF-β1
.
J. Autoimmun.
58
:
21
35
.
11
Hayakawa
,
Y.
,
N. D.
Huntington
,
S. L.
Nutt
,
M. J.
Smyth
.
2006
.
Functional subsets of mouse natural killer cells
.
Immunol. Rev.
214
:
47
55
.
12
Yadi
,
H.
,
S.
Burke
,
Z.
Madeja
,
M.
Hemberger
,
A.
Moffett
,
F.
Colucci
.
2008
.
Unique receptor repertoire in mouse uterine NK cells
.
J. Immunol.
181
:
6140
6147
.
13
van Unen
,
V.
,
N.
Li
,
I.
Molendijk
,
M.
Temurhan
,
T.
Höllt
,
A. E.
van der Meulen-de Jong
,
H. W.
Verspaget
,
M. L.
Mearin
,
C. J.
Mulder
,
J.
van Bergen
, et al
.
2016
.
Mass cytometry of the human mucosal immune system identifies tissue- and disease-associated immune subsets
.
Immunity
44
:
1227
1239
.
14
Boesch
,
M.
,
G.
Spizzo
,
A.
Seeber
.
2018
.
Concise review: aggressive colorectal cancer: role of epithelial cell adhesion molecule in cancer stem cells and epithelial-to-mesenchymal transition
.
Stem Cells Transl. Med.
7
:
495
501
.
15
Kim
,
H.
,
H. J.
Cho
,
S. W.
Kim
,
B.
Liu
,
Y. J.
Choi
,
J.
Lee
,
Y. D.
Sohn
,
M. Y.
Lee
,
M. A.
Houge
,
Y. S.
Yoon
.
2010
.
CD31+ cells represent highly angiogenic and vasculogenic cells in bone marrow: novel role of nonendothelial CD31+ cells in neovascularization and their therapeutic effects on ischemic vascular disease
.
Circ. Res.
107
:
602
614
.
16
Cros
,
J.
,
N.
Cagnard
,
K.
Woollard
,
N.
Patey
,
S. Y.
Zhang
,
B.
Senechal
,
A.
Puel
,
S. K.
Biswas
,
D.
Moshous
,
C.
Picard
, et al
.
2010
.
Human CD14dim monocytes patrol and sense nucleic acids and viruses via TLR7 and TLR8 receptors
.
Immunity
33
:
375
386
.
17
Liechti
,
T.
,
H. F.
Günthard
,
A.
Trkola
.
2018
.
OMIP-047: high-dimensional phenotypic characterization of B cells
.
Cytometry A
93
:
592
596
.
18
Rundberg Nilsson
,
A.
,
D.
Bryder
,
C. J.
Pronk
.
2013
.
Frequency determination of rare populations by flow cytometry: a hematopoietic stem cell perspective
.
Cytometry A
83
:
721
727
.
19
Attig
,
S.
,
L.
Price
,
S.
Janetzki
,
M.
Kalos
,
M.
Pride
,
L.
McNeil
,
T.
Clay
,
J.
Yuan
,
K.
Odunsi
,
A.
Hoos
, et al
CRI-CIC Assay Working Group
.
2011
.
A critical assessment for the value of markers to gate-out undesired events in HLA-peptide multimer staining protocols
.
J. Transl. Med.
9
:
108
.
20
Tricot
,
S.
,
M.
Meyrand
,
C.
Sammicheli
,
J.
Elhmouzi-Younes
,
A.
Corneau
,
S.
Bertholet
,
M.
Malissen
,
R.
Le Grand
,
S.
Nuti
,
H.
Luche
,
A.
Cosma
.
2015
.
Evaluating the efficiency of isotope transmission for improved panel design and a comparison of the detection sensitivities of mass cytometer instruments
.
Cytometry A
87
:
357
368
.
21
Chevrier
,
S.
,
H. L.
Crowell
,
V. R. T.
Zanotelli
,
S.
Engler
,
M. D.
Robinson
,
B.
Bodenmiller
.
2018
.
Compensation of signal spillover in suspension and imaging mass cytometry
.
Cell Syst
6
:
612
620.e5
.
22
Valitutti
,
S.
,
S.
Müller
,
M.
Salio
,
A.
Lanzavecchia
.
1997
.
Degradation of T cell receptor (TCR)-CD3-zeta complexes after antigenic stimulation
.
J. Exp. Med.
185
:
1859
1864
.

The authors have no financial conflicts of interest.