Metastasis is responsible for most cancer deaths, and tumor cell migration is an important part of the metastatic cascade. Understanding how tumor cells mobilize and invade other tissues, as well as how to modulate this behavior, is of great value to cancer biology and the development of therapeutics.
The most prevalent assay for tumor cell invasion involves modified Boyden chambers in which cells traverse a membrane coated with extracellular matrix (ECM). This approach requires cumbersome manual interventions that limit throughput and yields only limited information on the bulk population of cells that reach the other side of a thin layer of matrix.
In this article, we describe a microfluidics-based approach to patterning cells and ECM that delivers a much more detailed, physiologically relevant assessment of cell behavior and drug action. Cell migration is assayed in a true three-dimensional matrix that is more representative of the in vivo environment, and, because the assays are automatable, throughput sufficient for compound profiling can be readily achieved.
The 3-D tumor-cell invasion assay is performed in collagen-filled channels using the iuvo™ microconduit 5250 array plate from BellBrook Labs. The plate consists of 192 microchannels arrayed onto an SBS/ANSI-compatible plate (Figure 1). Each channel can be accessed through openings or ports on either end. Being only 140 microns tall, the ultrathin assay compartment is designed to allow full access by microscopes and high-content imagers. When the channel is filled with extracellular matrix such as type I collagen, and cells are deposited in the port at one end, cell migration can be observed as cells enter the matrix-filled channel area.
As shown in Figure 2, cell migration in this assay is truly 3-D, with cells distributed across the height of the channel. As expected, cell migration is inhibited by compounds such as the myosin II inhibitor blebbistatin.
In contrast to filter-based methods that rely on counts of cells that traverse an ECM-coated membrane, this type of migration assay has high information content because it allows imaging of actively invading cells in the ECM compartment during the entire assay period. Since each cell is assigned an x,y coordinate, a variety of analysis methods can be employed including those based on distributions, such as median distance traveled, as well as total number of cells migrating (Figure 3).
Analyses based on distribution and cells migrated both yield satisfactory data quality for compound profiling and screening. Because a purely antiproliferative compound would not be expected to alter the distribution of cells within the channel, distribution-based analysis is less sensitive to effects on proliferation. Furthermore, relative effects on migration vs. proliferation can be assessed by measuring the density of cells in the loading area at the end of the assay. Migration then can be normalized for any proliferation and/or cell death that might occur during the course of the assay, allowing compounds to be binned based on their inhibition of migration relative to other activities such as proliferation and cytotoxicity.
In addition to number of cells migrating and distance traveled, imaging of actively invading cells can provide clues to the mode of migration and the underlying mechanisms. For instance, with a clearly visible gap between almost all cells, the migration pattern of PC3M cells (prostatic carcinoma) in type I collagen can be identified as individual, as opposed to collective.
More detailed information on migration mechanisms or pathways can be gleaned using post-assay molecular staining of cells in the channels. Both phalloidin staining and immunocytochemistry have been employed in this way. Though we have just begun to explore these methods, the ability to apply in situ high-content analysis approaches to cells migrating in 3-D ECM is a promising avenue for delineating and targeting specific cell invasion events and pathways.