The Motion of a Brain Tumor in a Gel


For several years, members of the Weitz lab have been working in collaboration with several other groups (Tom Deisboeck at Massachusetts General Hospital, Mike Berens at Tgen, Len Sander at the University of Michigan, and Antonio Chiocca at Ohio State University) to understand details about the motion of a particular type of brain tumor cell, Glioblastoma Multiforme (GBM). GBM is a brain cancer that kills its victims quickly because it is highly invasive and because surgery to remove such tumors inevitably leaves much of the tenuous network of invasive cancer cells behind. There are many different aspects of this project, all aimed at understanding the motion of GBM and ways of curtailing it. The Weitz group has been involved in one aspect of this work: measuring mechanical properties of gels that resemble the extracellular matrix the tumors would encounter in vivo and studying how the brain tumor interacts with these gels in the early stages of its growth.

In early days of this study, the brain tumor spheroid was implanted into a commercially available gel (Matrigel) meant to mimic the ECM. Significant progress was made in ascertaining how the growing tumor spheroid and invasive tips were interacting with their surroundings by implanting beads in the gel and correlating their motion with the growth of the tumor spheroid and its invasive tips. This work is described here .
Two of the basic results from this work are:


The figures below illustrate the points above. Both figures contain particle tracks over several hours with early times in blue and late times in red. The first image is from the day on which the tumor was implanted, at which point it had not yet developed invasive tips. The second image is taken the next day, further from the spheroid, where one invasive tip had developed, as can be seen faintly in the brightfield image.




This data tells us several things: that the tumor is exerting significant force outward on its surroundings as it volumetrically expands, but that it also exerts tractional, inward forces as the tips invade. This data also suggests that the tumor is remodelling the matrix of the surrounding gel.

To directly study how the invasive cells were remodelling the gel matrix we moved from the Matrigel to a Collagen 1 gel. Collagen 1 forms long fibrils of collagen that are easily imaged in a variety of ways including by simple confocal reflectance microscopy. A second advantage of working with collagen gels is that we can control the concentration of protein in the gel and vary its mechanical properties significantly and see how the tumor responds to these differences.





These are some images of collagen 1 gels at different concentrations and a graph showing the elastic and viscous moduli for some of these same gels. We find that the elastic modulus of these gels scales as the inverse of the mesh size cubed. When we implant tumor spheroids into these different gels we find they grow at different rates, both in volume of the spheroid, and in area of the invasive tips. Indeed, when we implant spheroids into these gels we see not only interesting trends in their overall growth depending on the characteristics of the matrix, but also interesting details of how invasive tips migrate from the spheroid out into the surrounding Collagen 1 gel.









We have collected data that captures both the three-dimensional collagen matrix and the migrating cells. We do this by using two forms of microscopy simultaneously. For low magnification images, we collect brightfield or phase contrast images of the tumor together with reflective confocal images of the collagen fibers. For high resolution images, we use CARS microscopy to image particular cell tips and reflectance confocal to image the collagen fibers. CARS microscopy is a third-order nonlinear microscopy in which the contrast is due to the particular vibrational structure of the entitity being probed. You can find more information about CARS microscopy here . What is seen in these images and movies is that the tips definitely do remodel the collagen matrix and they align the collagen fibers around them and collect a cone of fibrils from the edge of the tip. We find that the speed and force exerted by these invasive tips are on the order of 10 microns/hour and 10nN, respectively.



We are now trying to study in detail the mechanics of the collagen I gels and correlate their local properties with the local growth of the tumor cells: Can the invasive tips invade a very stiff gel? Can the invasive tips invade if there very few collagen fibrils to exert tractional forces on? Do the brain tumors cause (with proteases they secrete) and/or take advantage of local inhomogeneitites in the gel matrix, and how is this reflected in their shape and growth rate?




For more information send email to Laura Kaufman , Clifford Brangwynne , or Karen Kasza. Other contributors to this work include Vernita Gordon and Emma Filippidi . After August 1, 2004, please find me here.