Mapping protein 'interactomes' within membrane trafficking pathways: combining mass spectrometry and SILAC labelling with targeted tyramine tagging

Eukaryotic cells (ie those cells with nuclei) contain a rich collection of internal membrane-bound compartments. These compartments often contain specialized proteins and serve many purposes vital for the well-being of the cell. For example, we are beginning to appreciate that protein mis-targeting within these pathways underlie a number of important diseases. A major area of active research within modern cell biology is to understand how these targeting events take place. In general terms, proteins are selectively targeted to different locations because they interact with 'coat proteins' that capture and direct their targets to different internal compartments. During this process, the coat proteins transiently assemble onto the internal membranes. Furthermore, many of the proteins that are selectively targeted also occur in discrete 'patches' within the targeted membrane themselves, and so the local molecular neighbours may also affect these interactions.

Unfortunately, there is still much uncertainty concerning the nature of some of these protein complexes. Among the many technical problems we face when trying to characterize them is the fact that many interact with their targets only fleetingly and with relatively low concentrations and/or low affinity for targets. It should be noted that this is a general problem that occurs in other aspects of cell biology. For example, hormones such as insulin trigger the assembly of specific proteins onto internal membrane compartments and again these have been difficult to characterize for similar reasons.

Here we propose a method to address this general problem. It develops and extends techniques that have been successfully used in other contexts, but not yet in this combination. Briefly, we will use a suitable protein labeled with an enzyme called peroxidase. This enzyme can convert a chemical called tyramine into a very unstable reagent that will only 'tag' molecules in the immediate vicinity of the enzyme. A suitable protein, labeled with peroxidase will be introduced into cells. Then tyramine reagent will be added to 'tag' both the protein and its immediate neighbours. These molecules can then be recognized and purified by their specific 'tyramine tag'. Once purified, they will be identified by a method called mass spectrometry that can successfully characterize large numbers of proteins in complex mixtures. In addition, we propose to include specifically adapted software that will enable the easy and accurate analysis of any data obtained with the technique. We believe that our process, that we call 'targeted tyramine tagging' will offer a significant improvement in the ability to identify transiently-interacting protein-protein partners within the cell.

Emergent behavior in cell biology often relies on the selective interactions of protein assemblies on internal membrane surfaces. Examples include the formation of coat proteins onto endosomes, and the assembly of protein complexes in response to signal transduction events. These interactions are difficult to examine using current proteomic methods because they are transient, dynamic and occur with low affinity on two dimensional membrane surfaces that are not stable following cell lysis.

A general proteomic method to address this problem would be of considerable interest to the biochemical, cell biology and proteomics communities. Here we propose the combined use of 'target tyramine tagging' together with high throughput SILAC mass spectrometry. This method is based on our recent experience to selectively label protein assemblies at the cell surface. We believe it can now be generalized to examine other protein 'interactomes' throughout the cell.

As 'proof of principle' we will examine the recycling pathway taken by transferrin through endosomes. We will add peroxidase-labeled transferrin to cells so as to fill internal compartments. Then we will add a membrane-permeant tyramine-biotin or tyramine-fluorescein derivative. When acted on by peroxidase, tyramine covalently labels tyrosine residues in proteins immediately adjacent to the peroxidase. Hence, the 'tyramine tagged' proteins can be isolated by standard affinity capture and characterized by mass spectrometry. We will combine this 'targeted tyramine tagging' with SILAC labeling to discriminate specific and non-specifically tagged proteins. We will explore the potential of this method when combined with SILAC labeling, to provide kinetic data on different proteins encountered by transferrin on its sequential journey through endosome compartments.

An integral part of our proposal is also to adapt associated software for the facile analysis of data generated by this method.

The man impacts expected form this project will be:

(i)Biochemistry/cell biology community: The development of the 'targeted tyramine tagging' methodology in conjunction with SILAC mass spectrometry will facilitate the analysis of intramembrane protein clusters. Such localized interactions lie at the explanatory heart of many key biochemical and cell biological processes including, but not limited to: membrane trafficking, cell metabolism and signal transduction. Developing accurate, sensitive and robust proteomic methods, including associated software for the analysis and characterisation of these complexes is a major priority.

(ii)Proteomics community (both academic and industrial): There are already close collaborations between the University of Cambridge and several mass spectrometry vendors and proteomics software vendors including Thermo Finnigan, Applied Biosystems, Waters and Matrix Science. The proteomics community as a whole will benefit from the development of such widely applicable methodology and associated software. KSL has collaborated with Matrix Science and Applied Biosystems for many years including Applied Biosystems contribution to a BBSRC award which resulted in the first developments of organelle proteomics methods. We envisage that such collaborations with mass spectrometry vendors will continue throughout the period of this grant, and will result in us working with the appropriate vendors to incorporate them into the workflow.

(iii)Pharmaceutical industry: The APJ and KSL have given talks frequently to the pharmaceutical industry on cell biology, ion channels and proteomics, including Glaxo Smith Kline and Pfizer. We have numerous contacts with these industrial research groups, including the newly established Pfizer research groups at Granta Park Cambridge. KSL has discussed proteomic methodology with Glaxo Smith Kline, Astra Zeneca and also Genentech. The methodologies developed within the context of this proposal are so fundamental that they could underpin many area of pharmaceutical research.

(iv)Full descriptions of this work will be published in peer reviewed literature and presented at academic conferences.

(v) Software developed as part of this project will be freely disseminated to the scientific community.