Cancer gets a close-up


From research to diagnosis to treatment, the ability to visualize cancer has played a big role in understanding and fighting the disease. To highlight this importance, and to show how advanced imaging has become, the National Cancer Institute recently held its “Cancer Close Up” contest. Two UW Carbone Cancer Center graduate students’ images were selected among 24 winners, and their work will be featured at upcoming national conferences. Below, they describe their winning images.

The Genesis of Cancer” by Ryan Denu, Medical ScientistTraining Program student, Mark Burkard’s lab

In a sea of normal cells, one cell produces too many centrosomes, and will lead to an improper separation of DNA to its daughter cells.

What are we looking at? These are retinal pigment epithelial cells. The blue is DNA, the red are microtubules, a protein that gives structure to cells, and the yellow/green is pericentrin, a protein that is part of the centrosome. These cells are overexpressing a protein, Plk4, that leads to centrosome amplification. This cell in the middle is dividing, but it has four centrosomes. A normal dividing cell should have two. I called this image ‘The Genesis of Cancer’ because you have a sea of normal cells and right in the middle of it all you see chaos.

How do centrosomes function? Centrosomes are found at the poles of dividing cells. They are responsible for organizing the microtubules, which are attached to DNA and pull chromosomes to the daughter cells. But if you have more than two centrosomes, who knows where these are going to pull? When this cell divides, those new cells are going to have errors in their DNA content.

How do centrosomes relate to cancer? Higher-risk human cancers have a higher incidence of centrosome amplification. When a cell doesn’t divide and separate its DNA correctly, the resulting cells can evolve and mutate more, they can become more resistant to therapy and become more aggressive. We use these cells as a model to study these high-risk cancers.

Breast Tumor Microenvironment” by Joseph Szulczewski, Molecular and Cellular Pharmacology graduate student, Patricia Keely’s lab

Tumor cells (in blue) and collagen (green) imaged from a live animal show the microenvironment of a breast tumor

What are we looking at? This image is the breast tumor microenvironment of a live animal using a laser microscopy technique called multiphoton microscopy. The green is collagen, the blue is breast tumor cells and the red are macrophages. We wanted to be able to characterize the collagen architecture in mammary tissue because as breast cancer develops, the collagen reorganizes. One hypothesis is that this reorganization allows for cells to migrate out of the area, leading to metastases. Visualizing macrophages, a type of immune cell, is also important because previous studies have shown a correlation between macrophage numbers and how aggressive a tumor is.

What is exciting about this microscopy technique? The really profound part of this is that we are not using any exogenous fluorophores or dyes. Normally you need to label cells with a fluorescent marker or an antibody, but these are all endogenous signals. The red and blue in this image are chemicals that are part of a cell’s normal energy metabolism, and they cycle between a fluorescent and non-fluorescent form depending on the cell’s metabolic state. So we can use the intensities of these signals as a rough metabolic readout. The green is visualized through second harmonic generation, another label-free approach that specifically identifies collagen and its organization.

What implications does this imaging have on diagnosing and treating cancer? The use of all endogenous signals is important, because you can’t fluorescently label people in a clinical setting the way we can with animals in a lab setting. So, maybe not for many years down the road, but this type of imaging could be used to distinguish – non-invasively – disease progression and differentiate cell types in live human tissue without the use of dyes or contrast agents. You can imagine treatment plans would be more aggressive for even a small tumor with significant collagen reorganization or associated with more macrophages than a large tumor where we didn’t notice any of this collagen reorganization.