The study of protein-protein interactions is of pivotal importance because most biological functions are mediated by protein complexes. In contrast to most other techniques, affinity purification coupled with mass spectrometry (AP-MS) is unbiased in that it does not require knowledge of potential interaction partners and it can be used for systems-wide analysis of protein-protein interaction networks.1
AP-MS has often been performed with the goal of a high degree of purification of protein complexes using tandem affinity purification (TAP) tagging approaches.2
However, this requires large amounts of starting material because of the two purification steps. Furthermore, the stringent washing conditions involved in TAP purifications often lead to loss of weakly bound protein complex members. Quantitative mass spectrometry can overcome these limitations by distinguishing specific interactors from unspecific background binders by the ratios of proteins in bait versus control pull-downs.3−5
This allows single step low stringent purification and high confidence interaction mapping including weak interactors.
There are many different formats for performing AP-MS in a qualitative, semiquantitative and quantitative fashion.6−22
Recently our laboratory has established an integrated quantitative workflow for AP-MS using bacterial artificial chromosomes (BACs) containing the gene of interest fused to the green fluorescent protein (GFP), which leads to expression of the full-length, GFP-tagged proteins from their endogenous promoters.23
This system, termed QUBIC for QUantitative BAC InteraCtomics, has several advantages. Most importantly the bait protein is expressed close to endogenous levels because the entire gene encoding the bait protein, including up- and downstream regulatory elements, is stably integrated into the genome of the cell.24−27
As tagged transcripts and proteins are processed by the cell equally to the endogenous counterpart, different splice isoforms can be expressed and proteins are post-translationally modified in the correct manner. Furthermore, cell lines expressing tagged versions of very large proteins can be created. In contrast to APs of the endogenous proteins, the QUBIC strategy does not rely on the availability of highly specific and immunoprecipiating antibodies for each protein of interest.
Most protein complexes, especially those with regulatory functions, are dynamic structures that form or change their composition and activity in response to cellular perturbations.28
Stimulation-dependent changes in protein conformation, subcellular localization or modification determine the interaction properties of the different complex members. AP-MS using stable isotope labeling with amino acids in cell culture (SILAC)29,30
in a double-labeling format is frequently employed for the characterization of protein interactions. SILAC with three isotope states has previously mainly been used to study the time dimension of the proteome31−33
but has also enabled comparison of bead proteomes,34
differentiation of isoform specific interactors6
and the change in composition of RNA polymerase upon inhibition of transcription.35
Here we wished to establish and characterize a general method for characterizing constitutive and stimulation-dependent dynamic interaction partners of regulatory protein complexes. We combined the QUBIC approach with triple SILAC labeling to differentiate background binders from specific binders and, in the same experiment, constitutive interactors from those that associate with a complex in a stimulus-dependent manner. We applied this method to the analysis of complexes in the Wnt signaling pathway and investigated differential complex formation dependent on stimulation of cells with the Wnt ligand Wnt3a.
The canonical Wnt pathway regulates cell fate, proliferation and self-renewal of adult stem and progenitor cells during the entire lifespan of metazoan organisms.36−40
Aberrant regulation of this pathway leads to different diseases, most prominently sporadic colon cancer. The key step in canonical Wnt signaling is the regulation of β-catenin. In the absence of Wnt ligands, β-catenin levels are low as a result of its continuous phosphorylation by the destruction complex, which triggers ubiquitylation and subsequent proteasomal degradation. Core components of the destruction complex are APC (Adenomatous Polyposis Coli) and Axin-1, which both function as scaffolds, and the kinases glycogen synthase kinase-3β (GSK-3β) and casein kinase I-α (CKI-α). Upon Wnt ligand binding to the receptors Frizzled and LRP5/6, the destruction complex function is attenuated, at least in part through relocalization to the plasma membrane and interactions with Dishevelled (DVL).36−40
Levels of β-catenin then accumulate in the cytoplasm and β-catenin translocates to the nucleus where it binds to TCF/LEF transcription factors and coactivates transcription of target genes.
Because of its central importance, the Wnt pathway is intensively studied and new pathway players that may be potential therapeutic targets are still found using a variety of approaches.41−43
Although canonical Wnt signaling has been investigated in depth, the exact mechanism by which the destruction complex is inhibited and β-catenin is stabilized is still not fully understood. Our Wnt pathway interactome study identifies potential novel Wnt pathway members and sheds light on the dynamics of the complexes involved.