2-d Technique Maps Tumor Proteins

David M. Lubman is marching steadily toward personalized treatment of cancer patients. A chemist and professor in the department of surgery at the University of Michigan, Ann Arbor, Lubman hopes the analytical methods he is developing will lead to early diagnosis and personalized management of cancer.

To achieve his goal, Lubman relies on two-dimensional liquid mass mapping of proteins in cancer cells. The technique is similar to 2-D gel electrophoresis of proteins but faster, more accurate, more reproducible, and more amenable to mathematical analysis, he said. He described how it works last month at the American Chemical Society's national meeting in Washington, D.C., in a symposium organized by the Division of Analytical Chemistry.

Two-dimensional liquid mass mapping involves liquid chromatography of denatured proteins from cells. In the first dimension, proteins are separated according to their isoelectric points--the pH at which they exist as electrically neutral species. In the second dimension, the proteins are separated according to their affinity to a hydrophobic column made of nonporous silica particles. According to Lubman, Beckman Coulter has licensed the technology and commercialized it as the ProteomeLab PF 2D automated system.

In Lubman's lab, the 2-D liquid separation is interfaced with mass spectrometry. The proteins eluting from the second chromatograph can go directly to a mass spectrometer for mass analysis. This interfacing is not yet available commercially, Lubman said, but it's only a matter of time.

After manipulation by software, the mass data yield a distribution over 2-D space with one axis defined by the isoelectric point (pH 4-9) and the other by molecular weight (2,000-100,000 daltons). The map, Lubman said, "is highly reproducible. You can use it to compare samples. We'd like to find key protein markers that will allow us to say that a particular ovarian tumor is of a certain type and not another."

Other methods exist to classify cancer subtypes, but most are inadequate, Lubman noted. For example, pathologists classify cancers according to how the cells look, but the results can be imprecise. Gel electrophoresis also is used, but it is time-consuming, limited in the number of proteins it can distinguish, and poorly reproducible.

Gel electrophoresis also cannot match the wealth of information made available by the mass maps, he pointed out. "Not only can you get quantitative information on what proteins are up- or down-regulated, but you can also see subtle structural changes, which often accompany the progression of cancer."

Lubman believes the mass maps can help guide clinicians in devising individualized treatments. "The bottom line is: These cancers are not one disease," he said. "Ovarian cancer has four subtypes at least. Some are more aggressive than others. The molecular profile may tell how aggressive the treatment needs to be.

"It's still early in the game," though, as far as direct impact on cancer treatment, Lubman told C&EN. Yet the technique already has revealed a protein that may be associated with a subtype of ovarian cancer that is highly resistant to certain types of chemotherapy. The link has not yet been validated, but if it holds, clinicians would want to know whether resistance is related to the protein, he explained.

With the help of a mathematician, Lubman is beginning to organize the mass data from different ovarian cancers. Analysis is revealing the proteins that characterize two types of ovarian cancer cells. Typically, these types are differentiated by morphology. In some cases, however, cells that are indistinguishable under the microscope can be differentiated by their protein profiles, he said.

Lubman envisions other uses for 2-D liquid mass mapping. "We're interested in inflammation, autoimmune disease, and other diseases to which protein markers can be applied."