Protease Fingerprinting to Diagnose Breast Cancer

Institution: The Burnham Institute for Medical Research
Investigator(s): Jeffrey Smith, Ph.D. -
Award Cycle: 1999 (Cycle V) Grant #: 5JB-0033 Award: $285,405
Award Type: IDEAS II
Research Priorities
Detection, Prognosis and Treatment>Imaging, Biomarkers, and Molecular Pathology: improving detection and diagnosis



Initial Award Abstract (1999)
There is still a great need to develop more convenient, cost effective, and accurate methods of detecting and tracking breast cancer. The availability of simple and rapid blood tests to screen for breast cancer could vastly increase the number women who are routinely screened. A blood test that detects breast cancer could, like cholesterol screening and PSA for prostate cancer, potentially be used to detect disease stage, changes in disease progression, and to monitor therapy. It would be particularly important for high-risk women, breast cancer survivors at risk for recurrence, and younger women who are difficult to screen accurately using mammography. The two hurdles that must be overcome, 1) the test must recognize many parameters, and 2) the test must be highly sensitive.

This study proposes a method for identifying the "fingerprint" of breast cancer. We plan to study changes in a group of proteins called proteases, which cut other proteins into smaller segments causing them to break down. This tumor-associated proteolytic activity causes the localized changes in the surrounding extracellular matrix, which are key to blood vessel formation (angiogenesis), tumor cell migration, and spread to other parts of the body (metastasis). Our hypothesis is that some of these tumor proteases are released into the blood, and are potentially detectable using emerging technologies. Thus, we will test blood samples from tumor-bearing animals for their ability to cleave peptides that are known substrates for cell proteases. In another phase of the project we will confirm that proteases of interest are associated with the tumor and not originating from other sources. A protease 'fingerprint' would be diagnostic for the clinical parameters associated with breast cancer.

In this phase of the project, we hope to identify 50+ unique proteases indicating the presence of breast cancer. We also suspect that another 50+ will indicate that a tumor has metastasized, or is capable of metastasizing. We anticipate that the results of this study will point the way for a future test of this approach in human breast cancer patients.


Final Report (2001)
The objective of this study was to identify the protease fingerprint of breast cancer. Proteases are enzymes that regulate key biological processes by cleaving other proteins. Protease activity is key to tumor progression and metastasis. The objective or our study was to develop a biochemical profiling method to identify the set of proteases linked to progression of breast cancer. We reasoned that such a capability would allow us to identify "signatures" of protease activity in biological fluids, like the blood of patients. Because protease are well known to be aberrantly regulated, and activated, in breast cancer, we hypothesized that a unique protease signature is present in breast cancer patients, and that uncovering this signature could lead to new insights into the disease, new diagnostic targets, and perhaps new therapeutic strategies.

Each proteases exhibits a characteristic "recognition specificity", it will only cleave proteins with a particular structure. Our study sought to develop methods of measuring this recognition specificity for a large number of proteases at one time using of a powerful biochemical technique called substrate phage display. Using this method, we were able to display tens of millions of different peptide structures (called a library) on a common scaffold. When this library of substrates was exposed to individual proteases, we were able to identify a select few peptides that were cleaved by the protease. Using this strategy we uncovered unique recognition specificity for three proteases associated with breast cancer, matrix metalloproteinases-2 and -9, along with membrane type-1 matrix metalloproteinase. Each of these enzymes is involved in tumor invasion, tumor metastasis, and perhaps more significantly, with tumor angiogenesis. The substrates that we identified for these proteases provide a signature set of peptides that can now be used to identify the activity of these proteases in blood. We envision that these substrates can be incorporated into diagnostic assays, or into imaging agents to image the position of distant metastases, or to monitor disease progression.

In addition to these advances, we were able to push the substrate phage technology to a level that is likely to help reveal the substrate recognition specificity of all proteases in the human genome.

Substrate hydrolysis by matrix metalloproteinase-9
Periodical:Journal of Biological Chemistry
Index Medicus: J Biol Chem
Authors: Kridel SJ, Chen E, Kotra LP, Howard EW, Mobashery S, and Smith JW
Yr: 2001 Vol: 276 Nbr: 23 Abs: Pg:20572-20578

A substrate phage enzyme-linked immunoabsorbant assay to profile panels of proteases
Periodical:Analytical Biochemistry
Index Medicus: Anal Biochem
Authors: Kridel SJ, Chen E, and Smith JW
Yr: 2001 Vol: 294 Nbr: 2 Abs: Pg:176-184

A unique substrate recognition profile for matrix metalloproteinase-2.
Periodical:Journal of Biological Chemistry
Index Medicus: J Biol Chem
Authors: Chen EI, Kridel SJ, Howard EW, Li W, Godzik A, Smith JW.
Yr: 2002 Vol: 277 Nbr: 6 Abs: Pg:4485-91