Fig. 1: The modular FLiP–MS approach.

a, Generating the FLiP marker library. A lysate is subjected to serial ultrafiltration to separate large protein complexes from their monomeric subunits. The fractions are separately subjected to limited proteolysis under native conditions, during which the sequence-unspecific protease PK cleaves surface-accessible residues. Regions at PBIs (red) should be accessible to PK cleavage in the monomeric form but sterically shielded in the complex-bound form, thus LiP should generate differential cleavage patterns. Many surface-accessible regions not located at the PBI (blue) should be equally accessible in both monomeric and complex-bound forms, thus LiP should generate the same cleavage patterns. Regions of proteins not at the interface but that also change protease susceptibility in the two assembly states (orange) will also generate different cleavage patterns. After the LiP step, protein fragments from each fraction are denatured, digested with trypsin and analyzed by mass spectrometry. Differential PK cleavage between the monomeric and the complex-bound forms of a protein is reflected in differential abundances of peptides between fractions. Consequently, differential peptide abundance analysis between fractions identifies markers reporting on PPI changes on a proteome-wide scale and enriches for PBIs (PDB ID 4DSS is shown as an example structure). Figure adapted from ref. ²⁶. b, Combining the FLiP library with LiP–MS. Conventional LiP–MS experiments report on global structural protein alterations between two conditions. The overlap with the FLiP marker library identifies which structural alterations may represent changes in PPIs. c, Network analysis for a global view of interactome dynamics. The identified changes in PPIs are projected on publicly available interaction networks, followed by network propagation and clustering to identify regions of the interactome that change on a given perturbation.