Single-Cell Mass Cytometry of Differential Immune and Drug Responses Across a Human Hematopoietic Continuum

Bendall SC, Simonds EF, Qiu P, Amir ED, Krutzik PO, Finck R, Bruggner RV, Melamed R, Trejo A, Ornatsky OI, Balderas RS, Plevritis SK, Sachs K, Pe'er D, Tanner SD, Nolan GP

Abstract

Flow cytometry is an essential tool for dissecting the functional complexity of hematopoiesis. We used single-cell “mass cytometry” to examine healthy human bone marrow, measuring 34 parameters simultaneously in single cells (binding of 31 antibodies, viability, DNA content, and relative cell size). The signaling behavior of cell subsets spanning a defined hematopoietic hierarchy was monitored with 18 simultaneous markers of functional signaling states perturbed by a set of ex vivo stimuli and inhibitors. The data set allowed for an algorithmically driven assembly of related cell types defined by surface antigen expression, providing a superimposable map of cell signaling responses in combination with drug inhibition.

Visualized in this manner, the analysis revealed previously unappreciated instances of both precise signaling responses that were bounded within conventionally defined cell subsets as well as more continuous phosphorylation responses that crossed cell population boundaries in unexpected manners yet tracked closely with cellular phenotype. Collectively, such single-cell analyses provide system-wide views of immune signaling in healthy human hematopoiesis, against which drug action and disease can be compared for mechanistic studies and pharmacologic intervention.

Experiment Overview

Purpose:

We sought to use mass cytometry to obtain a system-wide view of immune signaling in healthy human hematopoiesis, against which drug action and disease can be compared for mechanistic studies and pharmacologic intervention. We prepared a set of reagents to capture a system-wide view of immune cell types from a replicate analysis of bone marrow mononuclear cells from two healthy human donors.

Samples were stimulated or inhibited with one or more of the following conditions:

BCR, Flt3L, GCSF, GMCSF, IFNa, IL3, IL7, LPS, PMAionomycin, PVO4, SCF, TNFα, TPO, DMSO, Dasatinib, JAKi, U0126

Surface and intracellular markers measured

CD10, CD117, CD11b, CD11c, CD123, CD13, CD14, CD15, CD16, CD161, CD19, CD20, CD235a/b, CD3, CD33, CD34, CD38, CD4, CD41, CD44, CD45, CD45RA, CD47, CD56, CD61, CD7, CD8a, CD90, CXCR4, HLADR, IgM, IkB alpha, Ki67, Btk/Itk (pY551/pY511), Creb (pS133), CrkL (pY207), Erk1/2 (pT202/pY204), H3 (pS28), MAPKAPK 2 (pT334), NfkB (pS536), p38 (pT180/pY182), PLCg2 (pY759), S6 (pS235/pS236), Shp2 (pY580), SLP-76 (BLNK) (pY128), Src (pY418), STAT3 (pY705), STAT5 (pY694), ZAP70/Syk (pY319/pY352)

Cytometer Used:

DVS Sciences, Inc. CyTOF™ Mass Cytometer

Download the MIFlowCyt report for more information

IL7 --> pStat5 in T cells

IL-7 is a canonical activator of T cell proliferation. Here, IL-7 mediated activation of pSTAT5 in T cells is shown as an example of the cell-type specific signaling responses that can be detected by mass cytometry. Signaling at 15 minutes was detected using an antibody against STAT5 phosphorylated at the Y694 residue.

This same data is summarized in 4 squares of the heatmap highlighted in Figure 3B of the paper.

Marrow1
Mature CD4+ T
Naive CD4+ T
Naive CD8+ T
Mature CD8+ T
Calculated Arcsinh Ratio of Medians by First Row using X-Axis channel(s): 150-CD161, 150-pSTAT5
Mature CD4+ T Naive CD4+ T Naive CD8+ T Mature CD8+ T
Basal1
  • 0.0
  • 0.0
  • 0.0
  • 0.0
IL7
  • 1.83
  • 1.73
  • 1.62
  • 1.28

BCR --> pPLCg2 in B Cells

The B cell receptor (BCR) is a surface immunoglobulin that activates the B cell response to infection. Here, artificial crosslinking of the BCR was used to activate the BCR signaling pathway in B cells for 15 minutes. Activation of phosphorylated PLCgamma2, a second messenger of this pathway, is shown in the histogram overlay.

Different signaling responses were observed across different B cell developmental stages. This same data is summarized in 4 squares of the heatmap highlighted in Figure 3B of the paper.

Marrow1
Mature CD38lo B
Mature CD38mid B
Mature CD4+ T
Mature CD8+ T
Calculated Arcsinh Ratio of Medians by First Row using X-Axis channel(s): 150-CD161, 150-pSTAT5
Mature CD38lo B Mature CD38mid B Mature CD4+ T Mature CD8+ T
Basal1
  • 0.0
  • 0.0
  • 0.0
  • 0.0
BCR
  • 1.33
  • 1.06
  • -0.08
  • -0.03

Figure 3a (Heatmap)

A heatmap summary, ordered developmentally by cell type and stimulation condition, of the status of 18 intracellular functional markers in cells treated with one of 13 biological and chemical stimuli (left; abbreviations refer to recombinant human proteins, except: BCR = B-cell receptor cross-linking; lipopolysaccharide (LPS); PMA/Iono. = Phorbol-12-myristate-13-acetate with ionomycin; PVO4 = Pervanadate). Single-cell data from healthy human bone marrow were manually divided (“gated”) into 24 conventional cell populations based on 13 surface markers and DNA content. Signaling induction was calculated as the difference of inverse hyperbolic sine (arcsinh) medians of the indicated ex vivo stimulus compared to the unstimulated control for each manually assigned cell type (see Experimental Procedures online). Each row within a given stimulus group (gray bars) indicates the signaling induction of one of 18 intracellular functional markers (bottom).

Note: This figure was created in TIBCO Spotfire using data gated in Cytobank. For the figure below, data was normalized to the average of 5 unstimulated controls. The link below will bring you to a heatmap created on Cytobank using the identical data, but normalized to the first of 5 unstimulated controls. You can export the exact gated data and statistical values used to generate this figure.

Figure 3c-e (SPADE trees)

Canonical, cell type-specific signaling functions. Stimulation by IL-7, B cell receptor cross-linking (BCR), or lipopolysaccharide (LPS) uniquely induced phosphorylation of STAT5 in T cells, BLNK (detected with an antibody raised against pSLP-76; see Experimental Procedures online) in B cells, and p38 MAP kinase in monocytes, respectively. Signaling induction for each node in the SPADE diagram was calculated as the difference of arcsinh median intensity of the indicated ex vivo stimulus compared to the unstimulated control.

We hypothesized that the inherent similarity of cell stages and continuity of the transitions between cell differentiation states could be used to organize high-dimensional data into ordered, continuous clusters of similar cell phenotypes which, when projected on a two-dimensional plane, would convey the relatedness of these cells in a higher-dimensional space. We leveraged progressive changes in CD marker expression to organize bone marrow cells in an unsupervised manner, creating a tree-like scaffold for visualization of high-dimensional intracellular signaling behaviors in various cell types present during hematopoietic development in the bone marrow. To accomplish this, we used density normalization, agglomerative clustering, and minimum-spanning tree algorithms to distill multi-dimensional single-cell data down to interconnected clusters of rare, transitional, and abundant cell populations which were organized and displayed as a two-dimensional tree plot. Such a tree plot from healthy bone marrow represented the clustered expression of the cell surface antigens that were used to build the tree in 13-dimensional space based on the core surface markers conserved between our two 34 parameter analysis panels (CD3, 4, 8, 11b, 19, 20, 33, 34, 38, 45, 45RA, 90, 123). Each node of the plot encompasses a cluster of cells that were phenotypically similar in the 13-dimensional space defined by the ‘core’ surface markers. The approach uses a minimum-spanning tree algorithm, wherein each node of cells is connected to its ‘most related’ node of cells as a means to convey the relationships between the cell clusters. The number of nodes and ultimately their boundaries is driven by a user definable value (see Experimental Procedures online for an extended description of the algorithm). Each node describes an n-dimensional boundary encompassing a population of phenotypically similar cells. When connected via the minimum spanning tree this provides a convenient approach to map complex n-dimensional relationships into a representative two-dimensional structure.

For a more objective and fine-grained view of cell type-specific signaling responses, free of the biases of conventional one- and two-dimensional surface marker categorization, we overlaid the signaling behavior of the 18 functional epitopes on the tree structure using a similar approach as described for the ‘immunophenotype’ staining panel allowing the intracellular signaling status to be visualized on the previously annotated tree structure. Nodes were colored according to the magnitude of the difference in their median responses relative to the untreated control. This effectively eliminated the subjectivity of manual classification and improved the resolution of the heatmap, separating the 24 manually assigned cell types into 282 logically connected nodes of phenotypically distinct, but locally similar cell clusters.

Note: This figure was created in a flow cytometry analysis package called SPADE (not yet published) using data gated and exported from Cytobank. The link below will bring you to an illustration from which you can export the exact gated data used to generate this figure.

Figure S7 (Venn Diagram)

Venn diagram of statistically significant signaling responses among manually gated bone marrow populations from two donors. Independent one-sample t-tests were performed to identify statistically significant signaling responses for each of 22 unique experimental conditions in bone marrow samples from each of 2 healthy donors. The effect of each experimental condition was measured using 18 intracellular markers in 24 manually gated cell populations, for a total of 432 signaling responses per condition. After correction for multiple comparisons, an average of 554 significant signaling responses were observed for each donor, of which 248 were significant for both (45% concordance).

Note: This figure was created in Microsoft Excel using data gated in Cytobank. The link below will bring you to an illustration from which you can export the exact gated data and statistical values used to generate this figure.

Figure S4b (Surface-only tube)

Expression of immunophenotype surface markers overlaid onto the SPADE plots of healthy human bone marrow: The expression of an additional 18 surface markers from the 31 surface marker analysis of the same sample was overlaid on the SPADE tree constructed from 13 core surface markers. These 18 surface markers were not used in the SPADE plot and their localized expression is based solely on the shared expression patterns of the 13 core surface markers.

Note: This figure was created in a flow cytometry analysis package called SPADE (not yet published) using data gated and exported from Cytobank. The link below will bring you to an illustration from which you can export the exact gated data used to generate this figure.

IL7 --> pStat5 in T cells (Gating Hierarchy)

Mature CD4+ T

  • All Cells
  • Singlets
  • Singlets
  • CD33hi
  • Singlets
  • CD33lo
  • Singlets
  • CD3+
  • Singlets
  • CD4+
  • Singlets
  • CD8-
  • Singlets
  • CD19-
  • CD33lo CD19- CD3+ CD4+ CD8-
  • CD45RA- CD38-

Naive CD4+ T

  • All Cells
  • Singlets
  • Singlets
  • CD33hi
  • Singlets
  • CD33lo
  • Singlets
  • CD3+
  • Singlets
  • CD4+
  • Singlets
  • CD8-
  • Singlets
  • CD19-
  • CD33lo CD19- CD3+ CD4+ CD8-
  • CD45RA+ CD38-

Naive CD8+ T

  • All Cells
  • Singlets
  • Singlets
  • CD33hi
  • Singlets
  • CD33lo
  • Singlets
  • CD3+
  • Singlets
  • CD4-
  • Singlets
  • CD8+
  • Singlets
  • CD19-
  • CD33lo CD19- CD3+ CD8+ CD4-
  • CD45RA+ CD38-

Mature CD8+ T

  • All Cells
  • Singlets
  • Singlets
  • CD33hi
  • Singlets
  • CD33lo
  • Singlets
  • CD3+
  • Singlets
  • CD4-
  • Singlets
  • CD8+
  • Singlets
  • CD19-
  • CD33lo CD19- CD3+ CD8+ CD4-
  • CD45RA- CD38-

BCR --> pPLCg2 in B Cells (Gating Hierarchy)

Mature CD38lo B

  • All Cells
  • Singlets
  • Singlets
  • CD33hi
  • Singlets
  • CD33lo
  • Singlets
  • CD3-
  • Singlets
  • CD4-
  • CD33lo CD3- CD4-
  • CD19+
  • CD33lo CD3- CD4- CD19+
  • CD8-

Mature CD38mid B

  • All Cells
  • Singlets
  • Singlets
  • CD33hi
  • Singlets
  • CD33lo
  • Singlets
  • CD3-
  • Singlets
  • CD4-
  • CD33lo CD3- CD4-
  • CD19+
  • CD33lo CD3- CD4- CD19+
  • CD8-

Mature CD4+ T

  • All Cells
  • Singlets
  • Singlets
  • CD33hi
  • Singlets
  • CD33lo
  • Singlets
  • CD3+
  • Singlets
  • CD4+
  • Singlets
  • CD8-
  • Singlets
  • CD19-
  • CD33lo CD19- CD3+ CD4+ CD8-
  • CD45RA- CD38-

Mature CD8+ T

  • All Cells
  • Singlets
  • Singlets
  • CD33hi
  • Singlets
  • CD33lo
  • Singlets
  • CD3+
  • Singlets
  • CD4-
  • Singlets
  • CD8+
  • Singlets
  • CD19-
  • CD33lo CD19- CD3+ CD8+ CD4-
  • CD45RA- CD38-

Disclaimer

Cytobank and the Cytobank Logo are property of Cytobank Inc.

© 2011 Cytobank Inc., All rights reserved.