“Their directionality is still broken,” Irimia says, “their ‘steering wheel’ is still broken, and they cannot make it to the target.”

CHIP-Move

Lab-on-a-chip devices are finding wider use in the cell motility field. Yu Huang, a postdoctoral fellow in John Kuo's lab at the University of Wisconsin, Madison, is using microfluidics to study migration of glioblastoma stem cells, while Eric Balzer, a postdoctoral fellow with Konstantinos Konstantopoulous at Johns Hopkins University, uses them uses them to study chemotaxis.

The classic techniques to address migration and chemotaxis are the wound-healing and Boyden chamber assays, respectively. In the former, a monolayer of cells is gouged with a pipette tip to create a “wound,” which the cells then move in to “repair.” In the latter approach, cells are seeded in small membrane-bottomed cup immersed in a bath of chemoattractant; the assay is read by measuring how many cells follow the attractant and cross the membrane. Trouble is, the wound assay is lengthy, and cells can fill the gash by both moving and dividing, while the Boyden chamber is an endpoint assay: it isn't possible to know what the cells are doing during the assay, and how they get into the receiving chamber. It also isn't possible to image the cells during the experiment, to watch the behavior of individual cells, or to change experimental conditions.

Balzer and his coworkers have constructed a special microfluidic device that could overcome those issues. Made from PDMS, Balzer's device uses three fluid inlets — one for chemokine and two for buffer — to establish a chemical gradient by laminar flow. A series of 100-micron-long “microchannels” of varying width couple that gradient to a “cell-seeding” channel. Over a period of 8 hours or more, cells in the seeding channel work their way into and through the microchannels, moving up the gradient as they go.

According to Konstantopoulos, this design allows researchers to study cell movement in a confined three-dimensional space without all the variability, such as matrix porosity, rigidity, and so on, normally encountered with such assays. “In our device we have one parameter that we can change,” he says: “We can change the physical constriction.”

In their proof of concept paper, the team tested channels as wide as 50 microns and as narrow as 3 microns. The former is akin to a 2D planar surface whereas the latter mirrors the in vivo condition, or at least a more three-dimensional environment. Balzer refers to it as “two-and-a-half-D.”

On the scent: Over 16 hours these MDA MB 231 cells migrated towards (and beneath) an agarose drop laced with epidermal growth factor, following the indicated tracks. (Click to enlarge)

“The beauty of our device lies in two things,” Konstantopoulos says, “the fact that you can directly compare set to set the 2D migration versus migration in confined spaces … [and] that this migration can be dictated by a chemotactic gradient.”

Already, his team is learning new things about cell migration. For instance, the classic protrusion-and-retraction amoebic motion of cells on planar surfaces does not occur when they are confined in narrow spaces; instead, they appear to maintain a relatively constant shape and size. “What we have learned so far on a 2D surface appears to be very different from what is happening in confined spaces,” says Konstantopoulos.

Now, says Balzer, the goal is to use his 2-1/2D device to reconcile how cells behave under both sets of conditions. “That could help to unify the field.”

In Vivo Motility

Of course, another way to unify the field is to observe cells “in the wild”, an approach being adopted by Jeffrey Segall.

Segall, a professor of anatomy and structural biology at the Albert Einstein College of Medicine, studies the role of macrophages in promoting cancer metastasis. To do that, he uses two different methods, both in vivo.

First, there is an in vivo invasion assay. Here a 33-gauge needle, filled with a mixture of Matrigel (a 3D growth matrix) and growth factors such as EGF or CSF1, is inserted into an anesthetized animal's tumor, establishing a chemical gradient. That gradient attracts both tumor cells and macrophages, just as a warm pie inevitably attracts a cat in Saturday morning cartoons. The cells invade the needle, and when removed, those cells can be counted, providing a quantifiable measure of invasion. Segall and his colleagues have used this approach to document a positive-feedback paracrine loop between macrophages and tumor cells in both breast and brain cancers.