Intentional crystal-contact-free space in protein crystal

To understand the function of proteins, it is essential to perform the structural analysis of the protein complexes with ligands, such as substrates or partner molecules. The motions of ligands are restricted by the contacts with neighbor protein molecules in the crystal lattice. Here, we propose a new technique to analyze dynamics of a ligand in the bound state preserved in the crystal-contact-free space, which is intentionally created in protein crystals. We used Tom20 as a target protein. Tom20 functions as a general protein import receptor, by recognizing N-terminal signal sequences (presequences) of mitochondrial matrix proteins. Our working hypothesis is that the promiscuous specificity of Tom20 is attributed to the large mobility of the presequneces in the binding groove of Tom20 (1,2). Our aim is to obtain electron density that reflects the large mobility of a presequence in the crystal-contact-free space. In order to create the crystal-contact-free space, we took advantage of a protein fused with maltose binding protein (MBP). The key of the design is the connection of the two proteins firmly. We fused the C-terminal α-helix of MBP and the N-terminal α-helix of Tom20 seamlessly. After a systematic model building study, we decided to use a design with four residues inserted in the linker region. We found smeared electron density in the binding site of presequences in the difference Fourier electron-density map. We attached an iodine atom at the N-terminus of the presequence and confirmed the N-terminal position in the smeared electron density. We performed molecular dynamics simulation without the tethering in solution (3). The electron density simulated from the MD trajectory was fully consistent with the smeared electron density in the crystal contact-free space. We concluded that the smeared electron density corresponded to the partially overlapping region of the multiple states of the bound presequence.

Structural basis for the recognition of Eps15 by the SGIP1 μ homology domain

Clathrin-mediated endocytosis (CME) is a process for eukaryotic cells to internalize extracellular molecules. FCHo1 and FCHo2 are involved in the initial clathrin assembly step of CME [1, 2]. These proteins contain the lipid-binding EFC/F-BAR domain at the N-terminus [3], and the μ homology domain (μHD) at the C-terminus. The μHDs of these proteins interact with a region rich in a repeated sequence motif, Asp-Pro-Phe (DPF), of an endocytic scaffold protein, Eps15. Eps15 contains fifteen DPF motifs. SGIP1 is also involved in CME and contains the μHD highly homologous to those of FCHo1/2 at the C-terminus, which also interacts with Eps15. The μHDs of these proteins share weak amino-acid sequence homology with the μ subunits of the adaptor protein complexes, such as AP-2, which links cargo proteins and clathrin in CME. To investigate the mechanism of Eps15 recognition by the FCHo1/FCHo2/SGIP1 μHDs, we first identified the minimal Eps15 fragment retaining the ability to interact with the μHD by the interaction studies using the SGIP1 μHD. We found that two Eps15-derived 11-residue peptides each containing two DPF motifs connected by a short 2–3 residue linker interact with the SGIP1 μHD with modest affinities (Kd = 15–20 μM). In contrast, peptides containing only one DPF motif did not bind to the μHD. Thus, the SGIP1 μHD requires two DPF motifs for binding. Moreover, the structures of the SGIP1 μHD in complex with the peptides containing two DPF motifs revealed that the SGIP1 μHD extensively recognizes the two adjacent tandem DPF motifs, while not contacting the flanking residues, consistent with the interaction studies. This mode of recognition is distinct from that of the Eps15 DPF motif recognition by the AP-2 appendage domains, providing the rationale of the recruitment of Eps15 by FCHo1/2 to the nascent endocytic sites, while allowing the FCHo1/2-bound Eps15 to recruit AP-2 complex with the different parts of the Eps15 DPF repeat region for clathrin assembly.