Sweetening The Search For Breast Cancer Biomarkers

When a person develops cancer, changes can occur in the structures of intricate chains of sugar molecules that decorate many proteins. Now researchers have introduced a method to help identify proteins whose sugar modifications are altered in breast cancer (Anal. Chem., DOI: 10.1021/ac2002802). Future blood tests could use these so-called glycoproteins to detect breast cancer at early stages, the researchers say, when therapies are more likely to succeed.

"Changes in glycosylation are a hallmark of cancer," says Marina Hincapie, a biochemist at Northeastern University. "Tumors trigger an inflammatory response, which somehow turns on the glycosylation machinery of the cell." As a result, both tumors and surrounding normal tissues release modified glycoproteins into the bloodstream.

Detecting these altered proteins is challenging, says Hincapie, because serum is a complex mixture of thousands of proteins that vary widely in concentration. In a sea of abundant proteins, scarce proteins are virtually invisible to standard techniques. To identify these proteins by mass spectrometry, Hincapie and her colleagues developed an automated, three-step fractionation method to better sift through serum samples.

In the first step, the researchers used antibodies to remove the most abundant proteins—albumin and immunoglobulins—from the sample so that these proteins wouldn't overwhelm the analysis. Then, they isolated the serum's glycoproteins using a resin modified with proteins called lectins that bind specific sugar chains. In the final step of the purification, the researchers divided the glycoproteins into ten fractions using microscale isoelectric focusing. This technique separates proteins on the basis of their isoelectric point, the pH at which a protein's net charge is zero, which can vary with changes in glycosylation.

Combining the purification steps significantly increased the number of proteins the researchers could identify by mass spectrometry. The isoelectric focusing step was particularly important: Compared to when they left out the step, the researchers identified almost twice as many proteins when they included it. The team also could detect proteins with serum concentrations as low as 5 ng/mL; standard blood proteomics methods would have missed them, Hincapie says.

The technique allowed investigators to identify proteins that differed in abundance, glycosylation, or both in blood samples from patients with breast cancer versus samples from those with lung cancer or from healthy control subjects. The researchers now plan to characterize the specific differences in glycosylation of a subset of these proteins. They intend to test which of these changes is the best predictor of breast cancer. If the potential biomarkers prove robust, Hincapie says, doctors could someday diagnose breast cancer with a simple antibody- or microarray-based blood test that detects abundance and glycosylation changes.

David Lubman, a chemist at the University of Michigan, Ann Arbor, says it's too early to tell whether Hincapie's potential breast cancer biomarkers are any better than those found by previous techniques. But he says the method offers a better opportunity for biomarker discovery, because it has identified many low-abundance proteins that were missed by less sensitive techniques.