Final Activity Report Summary - MS-MODIB (Mass spectrometry-based molecular diagnosis of breast cancer)
In order to combat breast cancer, we have succeeded in assembling a multi-disciplinary team of medics, biologists, analytical chemists and informaticians. Our team has developed essential protocols and novel software for proteomic analyses of breast cancer. We have, and are in, the continuous process of expanding cancer research by two fundamentally new approaches: (a) focus on chromatin and (b) use of cross-linking to study conformations and interactions.
We have identified and addressed one of the major shortfalls of the current molecular definition of breast cancer by developing protocols and undertaking an unbiased analysis of protein changes on chromatin. Whole cell changes of protein abundance have been determined by mass spectrometry-based proteomic studies of MCF7 cells. However, whole cell analyses miss important aspects of cancer biology that are chromatin mediated. Firstly, chromatin composes only 11% of proteins in a cell. Therefore, chromatin protein expression changes are difficult to observe by proteomic methods in the context of all cellular proteins.
Secondly, change in location, not expression, is an important mechanism by which many chromatin regulators act, as seen for oestrogen receptor alpha, a key protein in breast cancer biology and treatment. Increase or decrease in chromatin binding is only visible when focussing on purified chromatin.
A chromatin proteomics approach combines advantages of unbiased analysis with the focus of studying a cellular compartment of specific relevance to endocrine function and resistance. Importantly, there is no proteomic analysis of chromatin to date and there is no purification protocol for interphase chromatin that links with proteomic analysis.
We have developed such a protocol as part of this project and conducted the first proteomic analyses of chromatin. We have applied our targeted and quantitative chromatin analysis to cellular breast cancer models, tissue biopsies of breast cancer patients, and breast cancer stem cells. We have also conducted the analysis of signalling pathways, membranes, and interstitial fluid. All this work has led to many candidate biomarkers which will be followed up in collaboration with the Breast Cancer Campaign unit at the University of Edinburgh.
We have furthermore identified the potential of structural information as biomarkers and developed robust tools that will allow the inclusion of this information as an additional dimension in the future. Current work on biomarkers focuses on the presence of specific molecules such as proteins to correlate with disease state and/or prognosis. This falls short in respecting the importance of structure in form of conformation and interactions for the biology of proteins and thus also for the role any protein may play in the context of a disease. We have analysed the conformation of two serum proteins, C3 and C3b, in vitro and in serum. For this we have developed novel tools including enrichment of cross-linked peptides, controls to recognise cross-linking artefacts that arise from aggregation, tailored acquisition methods for cross-linked peptides on modern mass spectrometers, matching of fragmentation spectra with peptide pairs as candidate cross-links and finally annotation software for fragmentation spectra. These tools were benchmarked by analysing RNA polymerase II, a twelve-protein 500 kDa complex. Using the available crystal structure of RNA pol II allowed establishing that our approach is 99% accurate in its identified cross-links.
In summary, the team has succeeded developing chromatin proteomics and cross-linking as the tools necessary to continue its analysis of breast cancer in the future. Both analytical approaches are of general applicability and will contribute to the progress of basic science in many fields. Importantly, these approaches provide novel diagnostic and therapeutic strategies against cancer.
We have identified and addressed one of the major shortfalls of the current molecular definition of breast cancer by developing protocols and undertaking an unbiased analysis of protein changes on chromatin. Whole cell changes of protein abundance have been determined by mass spectrometry-based proteomic studies of MCF7 cells. However, whole cell analyses miss important aspects of cancer biology that are chromatin mediated. Firstly, chromatin composes only 11% of proteins in a cell. Therefore, chromatin protein expression changes are difficult to observe by proteomic methods in the context of all cellular proteins.
Secondly, change in location, not expression, is an important mechanism by which many chromatin regulators act, as seen for oestrogen receptor alpha, a key protein in breast cancer biology and treatment. Increase or decrease in chromatin binding is only visible when focussing on purified chromatin.
A chromatin proteomics approach combines advantages of unbiased analysis with the focus of studying a cellular compartment of specific relevance to endocrine function and resistance. Importantly, there is no proteomic analysis of chromatin to date and there is no purification protocol for interphase chromatin that links with proteomic analysis.
We have developed such a protocol as part of this project and conducted the first proteomic analyses of chromatin. We have applied our targeted and quantitative chromatin analysis to cellular breast cancer models, tissue biopsies of breast cancer patients, and breast cancer stem cells. We have also conducted the analysis of signalling pathways, membranes, and interstitial fluid. All this work has led to many candidate biomarkers which will be followed up in collaboration with the Breast Cancer Campaign unit at the University of Edinburgh.
We have furthermore identified the potential of structural information as biomarkers and developed robust tools that will allow the inclusion of this information as an additional dimension in the future. Current work on biomarkers focuses on the presence of specific molecules such as proteins to correlate with disease state and/or prognosis. This falls short in respecting the importance of structure in form of conformation and interactions for the biology of proteins and thus also for the role any protein may play in the context of a disease. We have analysed the conformation of two serum proteins, C3 and C3b, in vitro and in serum. For this we have developed novel tools including enrichment of cross-linked peptides, controls to recognise cross-linking artefacts that arise from aggregation, tailored acquisition methods for cross-linked peptides on modern mass spectrometers, matching of fragmentation spectra with peptide pairs as candidate cross-links and finally annotation software for fragmentation spectra. These tools were benchmarked by analysing RNA polymerase II, a twelve-protein 500 kDa complex. Using the available crystal structure of RNA pol II allowed establishing that our approach is 99% accurate in its identified cross-links.
In summary, the team has succeeded developing chromatin proteomics and cross-linking as the tools necessary to continue its analysis of breast cancer in the future. Both analytical approaches are of general applicability and will contribute to the progress of basic science in many fields. Importantly, these approaches provide novel diagnostic and therapeutic strategies against cancer.