It’s a tough life for a cancer cell. First, there’s all that exhausting, uncontrolled dividing. Then, there’s the peer pressure created by a cell’s rapidly multiplying neighbors. Not to mention being squished by the abundant fluid that accumulates as inflammation spreads in the surrounding tissue.
As if those hassles weren’t enough, in order to travel to other parts of the body—a process known as metastasis—cancer cells have to squeeze themselves through hardened tissue and crevices in the walls of blood vessels to access the bloodstream. Once there, they float through the circulatory system, dock at a distant site—perhaps in a different organ—and slip back through vessel walls and tissue to continue dividing.
This rough-and-tumble existence, of course, doesn’t elicit sympathy from cancer researchers, but the stress that the cells experience and the structural changes they make to adapt to their challenging lifestyle have piqued scientists’ interest in recent years. One technique in particular, atomic force microscopy (AFM), is showing promise as a probe of cancer cell morphology and stress.
Much like a child poking some unknown substance with a stick, AFM probes the elasticity of materials on surfaces by pressing a sharp or spherical tip on the end of a flexible cantilever down onto a sample. A plot of a cell’s pushback on the tip versus the probe’s depth yields a force curve that provides information about the cell’s stiffness.
In 2007, a research group led by chemist James K. Gimzewski of the University of California, Los Angeles, used AFM to quantitate the elasticity differences between metastatic and benign cells from the body fluids of cancer patients (Nat. Nanotechnol., DOI: 10.1038/nnano.2007.388). On average, the researchers found, lung, breast, and pancreatic metastatic cells are 70% softer than benign cells. In addition, they found that the distribution of cell stiffness values measured for a group of normal cells is five times as broad as for a group of diseased cells, indicating that diseased cells are more morphologically similar to one another.
With all the forces they must endure, it’s no wonder that cancer cells reorganize their structural scaffolding, or cytoskeleton, to become more uniformly pliable than normal cells. “The softness aids in the spreading of the metastatic cells” through the body, Gimzewski says, and is due, at least in part, to a thinning of the thicket of protein fibers that make up the cytoskeleton.
That 2007 report made cancer scientists sit up and notice AFM, both as a diagnostic tool and as a technique that might teach them more about cancer progression and metastasis. Traditionally, AFM has been used by physicists and chemists. And it is these scientists who are now debating the best protocols for collecting meaningful nanomechanical data on the deadly disease and successfully transitioning AFM to the clinic.
“Cancer is traditionally viewed as a genetic disease” and studied in terms of gene mutations and chromosome abnormalities, says Subra Muralidharan, a molecular bioscientist at Washington State University, Pullman. But it’s now clear that cancer is more than just genetics, he says, explaining that no single factor seems to cause tumor formation. Cancer is far from being conquered, so it is time to take another tack and “think about how physical stress is transduced to biochemical pathways,” he suggests.
Making a connection between the physics and biochemistry of cancer has also become a goal of the National Cancer Institute. In 2009, the organization launched the Physical Sciences-Oncology Centers program to study the disease from a physics-based perspective. Scientists at 12 institutions around the country, including Scripps Research Institute and Johns Hopkins University, participate, applying tools such as AFM and optical tweezers to investigate the elusive disease.
One center, at Arizona State University, is home to Robert Ros, a physicist who started as a single-molecule force spectroscopist. “I’ve been pulling on molecules for the past 15 years,” Ros quips, adding that developments in AFM cancer cell measurements recently enticed him to switch to pushing on materials instead.
In a new study—Ros’s first venture into the field—his group uses AFM to show that even precancerous cells, which are poised to become cancerous, show signs of softness (Phys. Biol., DOI: 10.1088/1478-3975/8/1/015007). Precancerous esophageal cells are 45% more elastic than normal esophageal cells, the researchers note. They also observed a greater number of “sawtooth” discontinuities in the force curves measured for the precancerous cells than for the benign ones. These so-called breakthroughs are points at which the sharp, conical AFM tip, when pushed deep into the cell, senses a disruption in the cytoskeleton caused by broken fibers, Ros and his group believe. The findings suggest that precancerous cells are structurally weaker than their healthy counterparts.
To make the measurements, Ros’s group combined AFM with a confocal laser scanning microscope. The combination enables the researchers to obtain both fluorescence images of the cells’ organelles and AFM force maps of the cells’ surfaces. Essentially, “we know exactly where we are poking the cell,” Ros says. The overlapping data are important, he adds, because depending on where a cell is probed—particularly with a sharp tip that pinpoints small surface areas—it will likely have a different elasticity.
UCLA’s Gimzewski, who says he also uses sharp tips to gain more detailed information about the cell surface, acknowledges this problem. He argues that AFM measurements with this configuration should be made directly over a cell’s nucleus, its thickest area, so that force curve data can be fit to the standard model of elasticity, which assumes a thick sample. He also believes that researchers should limit the number of times a cell is probed in the same spot because repeated poking might alter the sample’s structure.
The ideal number of points that should be measured on a given cell and the number of cells that should be examined are contentious issues for some in the cell nanomechanics community. Washington State’s Muralidharan, another newcomer to the field, thinks some researchers aren’t collecting enough data to get accurate elasticity values for their cells.
At the spring American Chemical Society national meeting in Anaheim, Calif., Muralidharan’s graduate student Chandima Bandara presented the results of an analysis of 2,000–5,000 AFM force curves collected from each type of prostate cell line studied, ranging from normal to malignant to highly metastatic. That large volume of data is the result of measuring elasticity in a few hundred spots with a sharp tip across the surfaces of 10–12 cells of each type. Although the data are heterogeneous because the researchers probed away from the cells’ nuclei, thereby sampling a larger portion of their surfaces, Muralidharan contends that the distribution of data enables the determination of a more accurate average elasticity value.
With this strategy, the team found that highly metastatic prostate cells are about 67% more elastic than benign cells. But the researchers didn’t see much of a softness difference between normal cells and nonmetastatic cancerous cells.
Igor Sokolov, a physicist at Clarkson University, in New York, has observed a similar phenomenon for nonmetastatic malignant and normal cervical cells (Nat. Nanotechnol., DOI: 10.1038/nnano.2009.77). Having seen no statistical difference between the elasticity of normal and cancerous cells, however, Sokolov and his team looked at the force curve data carefully and fit it to a new model—one that takes into account the membrane protrusions on the outside of cells. This brush of whiskerlike structures called microvilli turns out to be different on the surface of normal and cancerous cells, the researchers found.
Cancerous cells have a brush of long, sparse whiskers interspersed with short, dense whiskers, and normal cells have a dense coating of whiskers that are all one length. When pushing on a cell with an AFM tip, “you deform the cell brush first and then the rest of the cell body,” Sokolov says, so these differences in structure show up in the raw data. He thinks that measuring a cell’s brush with AFM will be a more accurate diagnostic for cancer than simple elasticity measurements.
Sokolov also advocates the use of more accurate models for fitting force curves, the comparison of cancerous and normal cells of a defined age, and the use of a dull, spherical AFM tip when making measurements. All of these factors, he says, influence the cells’ force response. For instance, as normal cells age, they naturally become stiffer, and a sharp AFM tip causes too much localized strain on the cell, Sokolov says.
“There is now a relatively big community of people looking at the mechanics of cells,” he says. “We need a protocol—what we can trust and what we can’t.” When diagnosing cancer, you don’t want to make mistakes, he adds. “Otherwise, we’ll lose credibility in the eyes of the medical community.”
Gimzewski, however, is already attempting to put AFM in the clinic. His group is trying to set up a lab in the cytology department at UCLA’s Medical Center. There, the researchers will automate the process of looking at cells both optically, as pathologists do, and with AFM. Automation is necessary because if AFM “is going to be an applied tool, then it has to be used by clinicians” and not just physicists, says Shivani Sharma, a project scientist at UCLA’s California NanoSystems Institute, where she works with Gimzewski.
At the same time, Gimzewski has cofounded a company, Edeixa, to commercialize the technology. “I really hope such a machine will one day be part of fast-turnaround diagnostics and personalized medicine,” Gimzewski says. It could be used to help pathologists further assess suspicious cells from a biopsy and detect cancer at an early stage. The tool might also be used to measure the effects of cancer therapeutics on cells, he says.
At the spring ACS meeting, Sharma presented work she did with Gimzewski to measure the effects of the cancer drug cisplatin on ovarian cells (C&EN, April 4, page 10). They observed that, when exposed to cisplatin, cancerous cells increased significantly in stiffness. Normal cells also stiffened, but only slightly.
The team also reports that when a variety of metastatic cancer cells are exposed to green tea extract, a similar loss of softness occurs (Nanotechnology, DOI: 10.1088/0957-4484/22/21/215101). This herbal product increased the stiffness of the metastatic cells more than 80%, which the researchers correlated to an increase in the cells’ cytoskeletal filaments. The green tea extract, however, did not affect the normal cells, a promising result considering the damage chemotherapy drugs often inflict on noncancerous cells, Sharma says.
These results fortify Sharma’s conviction that AFM will be a handy tool for clinicians. So far, “AFM has shown a lot of potential, but it remains a very physics-lab-oriented tool,” she says. “Our goal is to bridge this gap.”