Profiling Enzyme Activities in Human Breast Cancer

Institution: Stanford University
Investigator(s): Stefanie Jeffrey, M.D. - Benjamin Cravatt, Ph.D. -
Award Cycle: 2004 (Cycle 10) Grant #: 10EB-1086 Award: $400,000
Award Type: TRC Full Research Award
Research Priorities
Biology of the Breast Cell>Pathogenesis: understanding the disease

This is a collaboration with: 10EB-0086 -

Initial Award Abstract (2004)
The molecular features special to metastatic breast carcinomas that support their motile and invasive behavior are complex and ill-defined. Although one generally accepted notion assigns proteins, such as proteases and catabolic enzymes, a central role in promoting the aggressive properties of metastatic breast tumors. Indeed, many proteases and their endogenous inhibitors show altered expression patterns (i.e., gene transcription and protein abundance) during breast tumorigenesis; however, the overall functional outcome of such changes remains unclear. The current confusion surrounding the relationship between specific enzymes and cancer arises from an inability to identify those enzymes whose activities actually change during the progression of this disease. For example, the coincidental or coordinated up-regulation of both a protease and its endogenous inhibitor (as occurs in breast cancer for urokinase and its natural inhibitor PAI-1) might result in little net difference in enzyme activity. To better understand how protease and other enzyme activities impact breast cancer, we have developed a novel chemical proteomics strategy, termed activity-based protein profiling (ABPP), that allows complex cell and tissue samples to be studied.

The translational collaboration supported by this CBCRP grant consists of basic science expertise of ABPP-based proteomics (Dr. Cravatt) with breast cancer gene expression profiling and clinical disease classification (Dr. Jeffrey). Previous CBCRP support to Dr. Cravatt successfully applied ABPP probes to the proteomic analysis of several enzyme classes across a panel of human breast cancer lines and mouse xenograft models of human breast cancer. These studies resulted in the identification of multiple proteases, lipases, and glutathione S-transferases, that were differentially expressed in estrogen receptor (ER)(+) and ER(-) breast cancer lines. The critical next step for the association of these enzymes with breast cancer is to determine whether their expression pattern is maintained in primary human tumors. Since genomics-based research by Dr. Jeffrey and others has shown that clinical breast cancers are divided into a variety of sub-types, it becomes important to apply the ABPP methodology to breast cancer sub-types and relate critical, cell invasion-related protease activity to other key cancer genes. Since proteases are thought to be so tightly associated with metastasis, this information will shed light on this aspect of clinical disease progression and outcome. Accordingly, we propose to use ABPP to determine the enzyme activity profiles of a panel of primary human breast tumors and correlate these proteomic data to key molecular and clinical parameters, including 1) gene expression profiles, 2) tumor type and grade, and 3) survival outcome. In addition to ABPP, we will employ biostatistics, in situ hybridization, and immunohistochemistry techniques, so that gene transcript and protein expression data can be compared to protein activity. These studies should not only validate of our cell line-derived breast cancer markers in the clinical setting, but also identify novel tumor-associated enzyme activities that are not expressed in breast cancer lines. Finally, we will determine the cellular and sub-cellular distribution of key breast cancer enzyme activities, which should provide important insights into the potential roles that these proteins play in tumorigenesis.

Progress towards better diagnostic and therapeutic agents to treat breast cancer relies heavily on our ability to identify new proteins that are both: 1) up-regulated in breast cancer samples, and 2) of functional relevance to the progression of this disease. We anticipate that our ABPP method will identify novel enzymes that support the growth and metastatic properties of breast cancer in vivo. These enzymes will in turn represent valuable markers and targets for future studies aimed at understanding, diagnosing, and ultimately treating breast cancer.


Final Report (2007)
The goal of this project was to utilize a novel chemical methodology called activity-based protein profiling (ABPP) developed in our laboratory to identify enzyme activities involved in breast cancer. Toward this end we have developed an advanced functional proteomics strategy that unites the ABPP and multidimensional protein identification technologies (ABPP-MudPIT) for the streamlined analysis of primary human breast tumors and cells. As will be described below, we have successfully applied ABPP-MudPIT to identify novel enzyme activities linked to breast cancer and proceeded to demonstrate a role for one of these enzymes in supporting disease pathogenesis. The project represented a collaboration between the Cravatt lab at Scripps

First, we applied ABPP-MudPIT to analyze a panel of primary human breast tumors. Analysis was carried out using 5-10µm frozen tissue sections from five normal breast and 28 breast tumor samples. Breast tissue specimens were found to cluster into five major groups based on their membrane enzyme activity profiles, with at least three of these groups representing clinically relevant breast tissue subtypes: normal breast, ER(+)/PR(+) breast cancer, and ER(-)/PR(-) breast cancer. The other two groups contained a mixture of tumors that were positive for either ER or PR, as well as some double-positive and double-negative tumors. In contrast, and consistent with previous findings from ABPP of human cancer cell lines, serine hydrolase activity profiles of the soluble proteome did not effectively differentiate the breast tissue specimens and, therefore, were not further examined in this study. Results from this initial phase indicated that 1D gel-based ABPP could generate enzyme activity profiles sufficiently rich in information content to classify primary human tumor specimens into biologically relevant subtypes, and did so while consuming only minute quantities of sample.

Next, two representative tumor specimens analyzed by ABPP-MudPIT identified over 50 unique serine hydrolase activities in the membrane proteomes. These enzymes included proteases (e.g., thrombin, tryptase, elastase), lipases (e.g., hormone-sensitive lipase, monoacylglycerol lipase), esterases (e.g., acetylcholinesterase, esterase D), and at least 15 uncharacterized hydrolases. Several enzymes were identified that exhibited altered levels of activity among the breast cancer specimens. For example, three enzyme activities, fibroblast activation protein (FAP, or seprase), KIAA1363, and platelet-activating factor acetylhydrolase 2 (PAF-AH2) were elevated in ER(-)/PR(-) tumors compared to either ER(+)/PR(+) tumors or normal breast tissue.

We next compared the relative activity levels of KIAA1363 and FAP to their gene expression profiles as measured by cDNA microarrays across the entire set of breast tumor samples. We found a strong overall relationship between KIAA1363 activity and ER(-)/PR(-) status among the tumors, as ten of the eleven specimens expressing the highest levels of KIAA1363 activity were from this tumor class. These studies led to a high-profile publication in Nature Methods [Jessani, N. et al. (2005) Nat. Methods 2, 691-697]. More generally, the ABPP-MudPIT method developed under the support of this TRC award is now being employed by our lab and many others to identify enzyme activities linked to human disease.

Concurrently, with our efforts to localize breast cancer-related enzymes in primary tumor sections, we have embarked on a project to disrupt the function of these enzymes in breast cancer models using pharmacological (e.g., inhibitors) and molecular biology (e.g., RNA interference) techniques. These studies have led to the discovery that KIAA1363 regulates a provocative ether lipid signaling network in human breast and ovarian cancer cells. Using mouse xenograft models, we have shown further that disruption of the KIAA1363-ether lipid network reduces breast and ovarian tumor growth in vivo, suggesting that this enzyme may constitute a novel therapeutic target for the treatment of these diseases. These exciting findings were recently published as a journal cover article [Chiang, K.P. et al. (2006) Chem. Biol. 13, 1041-1050].


Symposium Abstract (2005)
discovery of new protein biomarkers and therapeutic targets. The field of proteomics aims to facilitate this process by developing new methods for the parallel analysis of many proteins in highly complex samples. Achieving information content of satisfactory breadth and depth remains a formidable challenge for proteomics. This problem is particularly acute in the study of primary human specimens, such as tumor biopsies, which are heterogeneous and of finite quantity. Here, we present a functional proteomics strategy that unites the activity-based protein profiling and multidimensional protein identification technologies (ABPP-MudPIT) for the streamlined analysis of human specimens. This convergent platform involves a rapid initial phase where enzyme activity signatures are generated for functional classification of samples, followed by in-depth analysis of representative members from each class. This two-tiered approach enabled the identification of more than 50 enzyme activities in human breast tumors, nearly a third of which represented previously uncharacterized proteins. Comparison with cDNA microarrays identified enzymes whose activity, but not mRNA expression, depicted tumor class, underscoring the power of ABPP-MudPIT for the discovery of novel markers of breast cancer that may evade detection by other molecular profiling methods.

Discovery and validation of breast cancer subtypes.
Periodical:BMC Genomics
Index Medicus: BMC Genomics
Authors: Kapp AV, Jeffrey SS, Langerod A, Borresen-Dale AL, Han W, Noh DY, Bukholm IR, et al.
Yr: 2006 Vol: 11 Nbr: 7 Abs: Pg:231

Disease-specific genomic analysis: Identifying the signature of pathologic biology
Periodical:Bioinformatics
Index Medicus:
Authors: Nicolau M, Tibshirani R, Borresen-Dale A-L, Jeffrey SS
Yr: 2007 Vol: Nbr: Abs: Pg:

A streamlined platform for high-content functional proteomics of primary human specimens.
Periodical:Nature Methods
Index Medicus: Nat Methods
Authors: Jessani N, Niessen S, Yates JR 3rd, Jeffery SJ, and Cravatt BF
Yr: 2005 Vol: 2 Nbr: 9 Abs: Pg:691-7