Cross-Linking Furan-Modified Kisspeptin-10 to the KISS Receptor

Jul 17, 2017 - Chemical cross-linking is well-established for investigating protein–protein interactions. Traditionally, photo cross-linking is used...
0 downloads 0 Views 4MB Size
Articles pubs.acs.org/acschemicalbiology

Cross-Linking Furan-Modified Kisspeptin-10 to the KISS Receptor Willem Vannecke,†,‡ Christophe Ampe,‡ Marleen Van Troys,‡ Massimiliano Beltramo,§ and Annemieke Madder*,† †

Organic and Biomimetic Chemistry Research Group, Ghent University, Krijgslaan 281 S4, B-9000 Ghent, Belgium Department of Biochemistry, Faculty of Medicine and Health Sciences, Ghent University, B-9000 Ghent, Belgium § Equipe Neuroendocrinologie Moleculaire de la Reproduction, Physiologie de la Reproduction et des Comportements, Centre INRA Val de Loire, 37380 Nouzilly, France ‡

S Supporting Information *

ABSTRACT: Chemical cross-linking is well-established for investigating protein−protein interactions. Traditionally, photo cross-linking is used but is associated with problems of selectivity and UV toxicity in a biological context. We here describe, with live cells and under normal growth conditions, selective cross-linking of a furan-modified peptide ligand to its membranepresented receptor with zero toxicity, high efficiency, and spatio-specificity. Furan-modified kisspeptin-10 is covalently coupled to its glycosylated membrane receptor, GPR54(KISS1R). This newly expands the applicability of furan-mediated cross-linking not only to protein−protein cross-linking but also to cross-linking in situ. Moreover, in our earlier reports on nucleic acid interstrand cross-linking, furan activation required external triggers of oxidation (via addition of N-bromo succinimide or singlet oxygen). In contrast, we here show, for multiple cell lines, the spontaneous endogenous oxidation of the furan moiety with concurrent selective cross-link formation. We propose that reactive oxygen species produced by NADPH oxidase (NOX) enzymes form the cellular source establishing furan oxidation.

I

can be rapidly internalized and degraded. Immunoaffinity techniques for isolating ligand−receptor complexes from cell lysates are therefore rarely successful and can only be applied for high affinity interactions.4,5 Furthermore, interactomics techniques such as the yeast two-hybrid screen, are unsuitable for the identification of extracellular protein−protein/peptide interactions.6 Elucidation of such interactions thus requires a technique compatible with live cells under physiological conditions. In recent years, a whole range of bio-orthogonal chemistries was developed, allowing selective modification of biomolecules, in their natural environment. The introduced functional groups can react with a presented probe in an orthogonal way. Examples of such bio-orthogonal reactions include azide− alkyne cycloadditions, Staudinger ligation, or Diels−Alder reactions.7−9 Despite the elegance and efficiency of these methods in labeling a wide variety of biomolecules, the need for modification of both binding partners with a specific unnatural reactive group represents a hurdle for general applicability.10−13

nteractions between ligands, such as peptides or small molecules, and plasma membrane receptors are crucial for numerous key processes in living organisms. Approximately one third of the mammalian genome encodes for membrane proteins. Receptor activation and signaling are often involved in disease processes including malignant transformation. Identification of selective target membrane proteins can be used in the identification of biomarkers for early detection and prognosis of cancer but can also boost the discovery of new therapies. 1,2 Moreover, mass spectrometry analysis has generated huge catalogs of possible bioactive peptides by the so-called peptidomics approaches.3 For many of these peptides that produce a biological effect, it is currently not clear what their targets are or which molecular mechanisms they initiate. Indeed for such orphan peptides, the (cell-surface) receptors or other protein partners generally remain unknown. Analysis of non-covalently bound cellular assemblies is, in particular for membrane receptors, difficult for several reasons. Their hydrophobic nature and relatively low abundance precludes easy upscaling as they typically need to be in their natural environment to maintain binding properties. Activation of receptors by their ligands at the cell surface is based on the formation of a transient complex that, due to dynamic turnover, © XXXX American Chemical Society

Received: May 11, 2017 Accepted: June 29, 2017

A

DOI: 10.1021/acschembio.7b00396 ACS Chem. Biol. XXXX, XXX, XXX−XXX

Articles

ACS Chemical Biology Alternatively, photoaffinity cross-linking is based on the introduction of a photoreactive group which is able to form a cross-link with an unmodified natural binding partner upon activation with UV light. Benzophenones, aryl azides, and diazirines are among the most widely used groups. Although previously applied in the characterization of ligand−receptor complexes, these chemistries bear several disadvantages. Phototoxicity needs to be strictly monitored. Therefore, experiments are generally carried out with cell lysates or, when working with living cells, in cold buffers. Furthermore, the formation of highly reactive intermediates reduces selectivity of cross-linking. Benzophenones have bulky groups, which may negatively influence biological activity of the used probes. Aryl azides are much smaller, but the short-wave UVlight needed for their activation (