Article pubs.acs.org/ac
Monitoring Dynamic Interactions of Tumor Cells with Tissue and Immune Cells in a Lab-on-a-Chip Verena Charwat,†,∥ Mario Rothbauer,†,∥ Sandro F. Tedde,‡ Oliver Hayden,‡ Jacobus J. Bosch,£ Paul Muellner,† Rainer Hainberger,† and Peter Ertl*,† †
AIT Austrian Institute of Technology GmbH, Donau-City Straβe 1, 1220 Vienna, Austria Siemens AG, Corporate Technology, 91054 Erlangen, Germany £ Department of Internal Medicine 5, Hematology and Medical Oncology, University Hospital Erlangen, 91054 Erlangen, Germany ‡
S Supporting Information *
ABSTRACT: A complementary cell analysis method has been developed to assess the dynamic interactions of tumor cells with resident tissue and immune cells using optical light scattering and impedance sensing to shed light on tumor cell behavior. The combination of electroanalytical and optical biosensing technologies integrated in a lab-on-a-chip allows for continuous, labelfree, and noninvasive probing of dynamic cell-to-cell interactions between adherent and nonadherent cocultures, thus providing real-time insights into tumor cell responses under physiologically relevant conditions. While the study of adherent cocultures is important for the understanding and suppression of metastatic invasion, the analysis of tumor cell interactions with nonadherent immune cells plays a vital role in cancer immunotherapy research. For the first time, the direct cell-to-cell interactions of tumor cells with bead-activated primary T cells were continuously assessed using an effector cell to target a cell ratio of 10:1.
T
contact between tumor and stromal cells. Examples of these coculture invasion assays include organotypic skin models where the vertical invasion of squamous epithelial cells into a collagen matrix harboring fibroblasts is determined, and spheroid assays where the invasion of one cell type into a spheroid built of cells from another cell type is optically detected.4 Cell-based immunotherapies, on the other hand, involve the isolation of immune cells that are enriched ex vivo and then injected back into the patient to fight off cancer cells. Commonly used methods for the ex vivo generation of cytotoxic T cells for anticancer treatment are based on the isolation of primary peripheral blood mononuclear cells (PBMC) followed by the expansion and stimulation/activation of T cells, using either anti-CD3/anti-CD28 monoclonal antibodies immobilized on planar substrates5 or artificial antigen-presenting cells such as HLA-Ig-coated beads6 or paramagnetic anti-CD3/anti-CD28-coated beads.7 Prior to
umor cells, in their in vivo microenvironment, interact with and respond to resident stromal cells, including immune cells and stromal cells, such as fibroblasts and endothelial cells (EC).1 The dynamic interactions of tumor cells in their microenvironment is known to influence the evasion of immune surveillance, cancer growth, and metastasis.2 The development of an in vitro cell analysis method to monitor the dynamic interactions of tumor cells with adherent resident stromal cells and nonadherent immune cells, would therefore advance cancer research, particularly for the development of personalized cancer therapies.3 To date, a methodological divide exists in the application of in vitro cell-based assays in cancer research since no single analysis platform can be employed to detect both tumor invasion and evasion of immune surveillance. To investigate these interrelated processes, a variety of methodologies are employed to either detect tumor invasion using adherent coculture systems or determine the cytotoxic potential of activated T cells for immunotherapies. Tumor cell invasion is commonly investigated using transwell plates, including disease models based on coculture assays to assess physical cell-to-cell © 2013 American Chemical Society
Received: August 15, 2013 Accepted: November 12, 2013 Published: November 12, 2013 11471
dx.doi.org/10.1021/ac4033406 | Anal. Chem. 2013, 85, 11471−11478
Analytical Chemistry
■
injecting back into the patient, however, ex vivo enriched and activated T cells are assessed for their cytotoxic potential to recognize and attack malignant tumor cells in vitro.8 Cytotoxic T cell assays are routinely performed using chromium-51 labeling and release assay,9 fluorescence-activated cell sorting (FACS) combined with annexin-V staining,10 and enzymatic assays such as lactate dehydrogenase activity.11 The drawbacks of existing cell-based assays in cancer research are their tedious, time-consuming, and expensive cell staining procedures that include end-point detection methods to identify cell phenotypes.4 Additionally, end-point detection methods often underestimate labeling artifacts, require complex handling steps, and multiple reagents leading to low repeatability and accuracy.12 To bridge this methodological divide and to obtain deeper insights into tumor malignancy, we have developed a cell analysis method to directly assess the dynamic interactions of adherent and nonadherent cocultures under physiologically relevant conditions.13 The recent trend toward more biologically relevant cell-based assays has afforded new opportunities for label-free technologies based on thermal, electrical, and optical sensing methods to obtain time-resolved information on the phenotypic changes of cell cultures.14 Our goal was to combine two noninvasive and label-free technologies with orthogonal sensing properties using microchip technology to generate biorelevant data without the spatial interference, autofluorescence, or quenching effects of labels. Two well-established label-free cell analysis methods that are normally used independently were chosen to detect the dynamic interactions of tumor cells with tissue and immune cells in a lab-on-a-chip. Cellular impedance sensing is considered a powerful cell analysis tool and has already been successfully applied in cancer research to determine the effects of novel drug candidates on apoptosis,15 proliferation,16 morphology,17 and cellular activity.18 EIS is also used to distinguish between normal and malignant cells, as well as to identify highly and poorly metastatic cancer cells and to determine the ability of breast epithelial cancer cells to disrupt the cytoskeleton of dermal fibroblasts.19−21 Light scattering detection, on the other hand, has long been applied in flow cytometry as the gold standard for the label-free cell analysis of cell suspensions with single-cell resolution. With the use of forward scattering (
