Review
Highly multiparametric analysis by mass cytometry

https://doi.org/10.1016/j.jim.2010.07.002Get rights and content

Abstract

This review paper describes a new technology, mass cytometry, that addresses applications typically run by flow cytometer analyzers, but extends the capability to highly multiparametric analysis. The detection technology is based on atomic mass spectrometry. It offers quantitation, specificity and dynamic range of mass spectrometry in a format that is familiar to flow cytometry practitioners. The mass cytometer does not require compensation, allowing the application of statistical techniques; this has been impossible given the constraints of fluorescence noise with traditional cytometry instruments. Instead of “colors” the mass cytometer “reads” the stable isotope tags attached to antibodies using metal-chelating labeling reagents. Because there are many available stable isotopes, and the mass spectrometer provides exquisite resolution between detection channels, many parameters can be measured as easily as one. For example, in a single tube the technique allows for the ready detection and characterization of the major cell subsets in blood or bone marrow. Here we describe mass cytometric immunophenotyping of human leukemia cell lines and leukemia patient samples, differential cell analysis of normal peripheral and umbilical cord blood; intracellular protein identification and metal-encoded bead arrays.

Introduction

Multiparametric analysis of individual cells by flow cytometry has become the method of choice for functional and immunophenotypic analysis of cells in heterogeneous populations, particularly in the diagnosis of hematological malignancies (Jennings and Foon, 1997, Kern et al., 2004, Pedreira et al., 2008, Perfetto and Roederer, 2004, Terstappen et al., 1992, Wood et al., 2007). In traditional flow cytometry, a cell sample is treated with a panel of antibodies, each type labeled with a different fluorescent dye. The cell suspension, at an appropriate dilution, is then introduced into the instrument in such a way that individual cells are registered as they pass through a laser excitation region, with simultaneous detection of their fluorescence emission. Multiple lasers provide excitation of the different dyes at different wavelengths, and multiple detectors capture light scattering and fluorescent signals from the immunostained cells.

Unambiguous identification of cell populations in heterogeneous samples such as blood, bone marrow or tumor biopsy requires quantitative determination of many biomarkers simultaneously in individual cells (Chattopadhyay et al., 2008, Wood, 2007, Weir and Borowitz, 2001, Autissier et al., 2010). Current flow cytometers are generally limited to 10 simultaneous measurements. However, in the research laboratory, polychromatic flow cytometry has reached the level of multiplexing 17 different antigens (Perfetto et al., 2004). While this is an impressive and important step forward, the potential of fluorophore-based highly multiplexed assays is complicated by challenges inherent in fluorescence detection. The emission bands of fluorescent dyes are sufficiently broad that spectral overlap is inevitable when one makes simultaneous measurements with multiple dyes. Quantum dots have more narrow emission bands, especially at shorter wavelengths, which mitigates but does not eliminate the problem (Roederer et al., 2004, Chattopadhyay et al., 2006). As a consequence, complex correction algorithms are needed to deconvolute the overlapped spectra. Furthermore, this type of compensation requires a large number of control samples stained in multiple antibody combinations to optimize the assay (McLaughlin et al., 2008).

To address the increasing need for multiparametric analysis, the method of simultaneous detection of multiple proteins in acidified biological samples by inductively coupled mass spectrometry (ICP-MS) using metal-tagged antibodies, was first suggested and demonstrated by Baranov et al. (Baranov et al., 2002, Quinn et al., 2002) and further investigated by several groups (Careri et al., 2007, Zhang et al., 2002, Zhang et al., 2004, Bettmer et al., 2006, Careri et al., 2009). ICP-MS is a tool designed for the analysis of elements and is widely used in applications (mining and metallurgy, the semiconductor industry), which demand precise quantitation of element abundance. In ICP-MS, samples are atomized and ionized in plasma at temperatures approximating that of the surface of the sun (7000 K). The mass spectrometer then resolves and quantifies the various element isotopes. Key features of traditional ICP-MS instrumentation include the absence of interference between masses and a linear dynamic range of greater than 108 (Tanner et al., 2007).

However, this technique could not be applied to multi-target analysis of individual intact cells for several reasons. Firstly, the quadrupole mass analyzer most common in ICP-MS, has a settling time of ~ 50–200 μs, required for stabilization of the mass filter between individual isotope measurements. This time is comparable to the duration of the ion cloud produced in the argon plasma from an individual microparticle or cell (Stewart and Olesik, 1999). Thus, the measurement of two or more isotopes during a transient event of such short duration is virtually impossible with available quadrupole mass spectrometers. The required frequency of sampling of the cell-induced transient should be 50,000–100,000 spectra/s, to allow for ten or more individual spectra per cell so that the transients from overlapping particles can be discriminated. Thus, a new type of instrument is required to interrogate individual cells (Tanner et al., 2008).

In the ICP-MS solution bioassays that we have reported (Ornatsky et al., 2006, Baranov et al., 2002) directed at immunophenotyping human leukemia cell lines, various antibodies were labeled with different lanthanides, with each metal serving as an element tag for a particular antibody. We focus on the lanthanide (Ln) series of elements because of the large number of stable resolvable isotopes having similar chemistry that facilitates their incorporation into the same tag structure. Lanthanides have low natural abundance, thus exhibiting a very low background signal in this type of analysis. In these experiments, we attached metal-chelating polymers to the antibodies to increase the signal associated with each sample. Each polymer carried about 30 atoms of a lanthanide element, with several polymers attached to each antibody. The large dynamic range was demonstrated by the ability to quantify cell surface markers that differed in abundance by a factor of 500 in a single multiplexed assay of KG1a cells (Lou et al., 2007). Here we report a major step forward with the extension of this methodology to cell-by-cell analysis using a mass cytometer (CyTOFTM), and metal-containing polymer tags (MAXPARTM) both from DVS Sciences Inc., Richmond Hill, Canada. The instrument is based on a non-optical physical principle of detection and a different chemical nature of labels. The fluorescent labels are replaced by specially designed multi-atom elemental tags and detection takes advantage of the high resolution, sensitivity, and speed of analysis of Time-of-Flight Mass Spectrometry (TOF-MS). Since many available stable isotopes can be used as tags, many proteins and gene transcripts can potentially be detected simultaneously in individual cells.

In this review, we describe the basic principles of the mass cytometry instrumentation, elemental tags, and linked immunological methods using as an example highly multiplex (over 20 antibodies in one mixture) assays applied both to cultured human leukemia cell lines and patient samples. The results are compared to those of traditional fluorescence-based cytometric analysis. Furthermore, we outline the next steps in the development of bead-array technology: synthesis, characterization and application of metal-encoded beads as mass cytometry standards and for bead arrays. Mass cytometry can be scaled to higher multiplicity with the development of a broader array of elemental tags. The simplicity and efficiency of the approach demonstrate the potential of mass cytometry for biological research and drug development.

Section snippets

Antibodies and reagents

Primary monoclonal and polyclonal antibodies to cell surface and intracellular antigens were obtained from commercial suppliers (BD Biosciences, San Jose, CA; BioLegend, San Diego, CA; Abcam Inc., Cambridge, MA; Cell Signaling Technology Inc, Danvers, MA.; Santa Cruz; Biotechnology Inc., Santa Cruz, CA; Millipore, Billerica, MA). All antibodies were bought as affinity purified saline solutions without stabilizing proteins. Species-specific isotype immunoglobulins were used for negative

Antibody titration and quantitation

After labeling antibodies with metal polymer tags we tested the conjugates for optimal staining concentrations and specificity on human cell lines and primary cells. An example is shown in Fig. 6. A mononuclear cell fraction of normal bone marrow was divided into four samples (1 × 106 cells/tube) and stained with a mixture of 22 metal-tagged antibodies at 0.313 μg/mL, 0.625 μg/mL, 1.25 μg/mL or 2.5 μg/mL. Washed and fixed cells were then treated with the Ir-intercalator for DNA staining, and 80,000

Discussion

Biomarker measurements relate the effects of therapeutic drugs on molecular and cellular pathways to treatment outcomes and can help explain, for example, differences in drug metabolism during clinical trials (Atkinson et al., 2001). As such, tumor biomarkers contribute greatly to the selection of appropriate personalized cancer therapy in clinical trials. Immunophenotyping of blood biomarkers using flow cytometry has played an important role in the diagnosis of leukemia subtypes and selection

Conflict-of-interest disclosure

Four of the six authors (O.O., D.B., V.B, and S.T.) are also principles of DVS Sciences, Inc., which manufactures the CyTOFTM mass cytometer and the MAXPARTM reagents.

Acknowledgements

This project was funded by Genome Canada through the Ontario Genomics Institute, the Ontario Ministry of Research and Innovation, the National Institutes of Health [NIH grant R01-GM076127], NSERC Canada, and DVS Sciences Inc. Special thanks is extended to Dr. Qing Chang, Ontario Cancer Institute/Princess Margaret Hospital, University of Toronto, Ontario for the Panc-1 cells, and Dr. M. Milavsky and Dr. J. Wang, University Health Network (Toronto General Research Institute) for HSC and leukemia

References (56)

  • E.G. Weir et al.

    Flow cytometry in the diagnosis of acute leukemia

    Seminars in Hematology

    (2001)
  • B.L. Wood

    Myeloid malignancies: myelodysplastic syndromes, myeloproliferative disorders, and acute myeloid leukemia

    Clinics in Laboratory Medicine

    (2007)
  • A.H. Yin et al.

    AC133, a novel marker for human hematopoietic stem and progenitor cells

    Blood

    (1997)
  • A.I. Abdelrahman et al.

    Metal-containing polystyrene beads as standards for mass cytometry

    Journal of Analytical Atomic Spectrometry

    (2010)
  • A.J. Atkinson et al.

    Biomarkers and surrogate endpoints: preferred definitions and conceptual framework

    Clinical Pharmacology and Therapeutics

    (2001)
  • P. Autissier et al.

    Evaluation of a 12-color flow cytometry panel to study lymphocyte, monocyte, and dendritic cell subsets in humans

    Cytometry. Part A

    (2010)
  • D.R. Bandura et al.

    Mass cytometry: technique for real time single cell multitarget immunoassay based on inductively coupled plasma time-of-flight mass spectrometry

    Analytical Chemistry

    (2009)
  • V.I. Baranov et al.

    The potential for elemental analysis in biotechnology

    Journal of Analytical Atomic Spectrometry

    (2002)
  • M.J. Barten et al.

    Biomarkers in transplantation medicine: prediction of pharmacodynamic drug effects

    Transfusion Medicine and Hemotherapy

    (2007)
  • M.J. Barten et al.

    Pharmacodynamics of T-cell function for monitoring immunosuppression

    Cell Proliferation

    (2007)
  • J. Bettmer et al.

    Elemental tagging in inorganic mass spectrometric bioanalysis

    Analytical and Bioanalytical Chemistry

    (2006)
  • K. Braeckmans et al.

    Encoding microcarriers: present and future technologies

    Nature Reviews. Drug Discovery

    (2002)
  • S. Brenner et al.

    Gene expression analysis by massively parallel signature sequencing (MPSS) on microbead arrays

    Nature Biotechnology

    (2000)
  • M. Careri et al.

    ICP-MS as a novel detection system for quantitative element-tagged immunoassay of hidden peanut allergens in foods

    Analytical and Bioanalytical Chemistry

    (2007)
  • M. Careri et al.

    Element-tagged immunoassay with inductively coupled plasma mass spectrometry for multianalyte detection

    Analytical and Bioanalytical Chemistry

    (2009)
  • J.E. Celis et al.

    Loss of adipocyte-type fatty acid binding protein and other protein biomarkers is associated with progression of human bladder transitional cell carcinomas

    Cancer Research

    (1996)
  • P.K. Chattopadhyay et al.

    Quantum dot semiconductor nanocrystals for immunophenotyping by polychromatic flow cytometry

    Nature Medicine

    (2006)
  • P.K. Chattopadhyay et al.

    A chromatic explosion: the development and future of multiparameter flow cytometry

    Immunology

    (2008)
  • Cited by (299)

    View all citing articles on Scopus
    View full text