Zurich Open Repository and Archive University of Zurich Main Library Strickhofstrasse 39 CH-8057 Zurich www.zora.uzh.ch Year: 2018 Curvature of designed armadillo repeat proteins allows modular peptide binding Hansen, Simon ; Ernst, Patrick ; König, Sebastian L B ; Reichen, Christian ; Ewald, Christina ; Nettels, Daniel ; Mittl, Peer R E ; Schuler, Benjamin ; Plückthun, Andreas Abstract: Designed armadillo repeat proteins (dArmRPs) were developed to create a modular peptide binding technology where each of the structural repeats binds two residues of the target peptide. An essential prerequisite for such a technology is a dArmRP geometry that matches the peptide bond length. To this end, we determined a large set (n=27) of dArmRP X-ray structures, of which 12 were previously unpublished, to calculate curvature parameters that define their geometry. Our analysis shows that con- sensus dArmRPs exhibit curvatures close to the optimal range for modular peptide recognition. Binding of peptide ligands can induce a curvature within the desired range, as confirmed by single-molecule FRET experiments in solution. On the other hand, computationally designed ArmRPs, where side chains have been chosen with the intention to optimally fit into a geometrically optimized backbone, turned outto be more divergent in reality, and thus not suitable for continuous peptide binding. Furthermore, we show that the formation of a crystal lattice can induce small but significant deviations from the curva- ture adopted in solution, which can interfere with the evaluation of repeat protein scaffolds when high accuracy is required. This study corroborates the suitability of consensus dArmRPs as a scaffold for the development of modular peptide binders. DOI: https://doi.org/10.1016/j.jsb.2017.08.009 Posted at the Zurich Open Repository and Archive, University of Zurich ZORA URL: https://doi.org/10.5167/uzh-150343 Journal Article Accepted Version The following work is licensed under a Creative Commons: Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0) License. Originally published at: Hansen, Simon; Ernst, Patrick; König, Sebastian L B; Reichen, Christian; Ewald, Christina; Nettels, Daniel; Mittl, Peer R E; Schuler, Benjamin; Plückthun, Andreas (2018). Curvature of designed armadillo repeat proteins allows modular peptide binding. Journal of Structural Biology, 201(2):108-117. DOI: https://doi.org/10.1016/j.jsb.2017.08.009 *Revised Manuscript (Unmarked) Click here to view linked References Curvature of designed armadillo repeat proteins allows modular 1 2 3 4 peptide binding 5 6 Authors: Simon Hansena,1 †, Patrick Ernsta, †, Sebastian L.B. Königa, Christian 7 8 a,2 a,3 a a 9 Reichen , Christina Ewald , Daniel Nettels , Peer R.E. Mittl , 10 11 Benjamin Schulera, Andreas Plückthuna, * 12 13 14 Affiliation: a Department of Biochemistry, University Zürich, Winterthurerstrasse 15 16 190, 8057 Zürich, Switzerland 17 18 † 19 equal contribution 20 1 21 Present address: Department of Early Discovery Biochemistry, 22 23 Genentech, 1 DNA Way, South San Francisco, CA 94080, USA 24 25 2 26 Present address: , , - 27 Schlieren, Switzerland 28 29 30 3 Present address: Flow Cytometry Facility, University of Zurich, 31 32 Winterthurerstrasse 190, 8057 Zürich, Switzerland 33 34 Correspondence: *Andreas Plückthun Tel. +41-44-635 5570, Fax. +41-44-635 5712 35 36 e-mail: [email protected] 37 38 39 Manuscript: Total pages: 28 40 41 Supporting information: 42 43 - Supplementary figures and tables: Fig. S1-S4; Tables ST1 & 44 45 ST2 (12 pages) 46 47 - Supplementary raw data table: Curvature parameters (Excel 48 49 file) 50 51 Tables: 1 52 53 54 Figures: 6 55 56 Coordinates: Coordinates and structure factors have been deposited at the PDB with 57 58 accession codes 5MFB, 5MFE, 5MFF, 5MFG, 5MFH, 5MFI, 5MFJ, 59 60 5MFK, 5MFL, 5MFM, 5MFN and 5MFO 61 62 63 1 64 65 1 2 Abstract 3 4 5 Designed armadillo repeat proteins (dArmRP) were developed to create a modular peptide 6 7 binding technology where each of the structural repeats binds two residues of the target 8 9 10 peptide. An essential prerequisite for such a technology is a dArmRP geometry that matches 11 12 the peptide bond length. To this end, we determined a large set (n=27) of dArmRP X-ray 13 14 15 structures, of which 14 were previously unpublished, to calculate curvature parameters that 16 17 define their geometry. Our analysis shows that consensus dArmRPs exhibit curvatures close 18 19 20 to the optimal range for modular peptide recognition. Binding of peptide ligands can induce a 21 22 curvature within the desired range, as confirmed by single-molecule FRET experiments in 23 24 solution. On the other hand, computationally designed ArmRPs, where side chains have been 25 26 27 chosen with the intention to optimally fit into a geometrically optimized backbone, turned out 28 29 to be more divergent in reality, and thus not suitable for continuous peptide binding. 30 31 32 Furthermore, we show that the formation of a crystal lattice can induce small but significant 33 34 deviations from the curvature adopted in solution, which can interfere with the evaluation of 35 36 37 repeat protein scaffolds when high accuracy is required. This study corroborates the suitability 38 39 of consensus dArmRP as a scaffold for the development of modular peptide binders. 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 2 64 65 1 2 3 4 5 Keywords 6 7 8 Armadillo repeat proteins, peptide binding, protein crystallography, protein engineering, 9 10 single-molecule FRET 11 12 13 14 Abbreviations 15 16 17 AU: asymmetric unit, CoM: center of mass, dArmRP: designed armadillo repeat proteins, 18 19 nArmRP: natural armadillo repeat proteins 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 3 64 65 1 2 Introduction 3 4 5 Armadillo repeat proteins consist of a repetitive protein fold found in many proteins, most 6 7 prominently in importin-α d β-catenin, which are involved in cellular processes including 8 9 10 nuclear import, cell adhesion and signaling (Tewari et al., 2010). The name derives from the 11 12 appearance of a Drosophila mutant with a defect in a segment polarity gene later identified as 13 14 15 β-catenin (Nüsslein-Volhard and Wieschaus, 1980). Natural armadillo repeat proteins 16 17 (nArmRP) are abundant in the human genome and usually consist of 4-12 consecutive repeats 18 19 (Gul et al., 2016) where the repeats at either terminus (called N- and C-cap, respectively) are 20 21 22 modified to protect the hydrophobic core from solvent exposure. Each repeat is usually made 23 24 from 42 residues that fold into three triangularly arranged helices (H1, H2 and H3), and the 25 26 27 angular stacking of several repeats creates a binding groove of adjacent H3 helices (Huber et 28 29 al., 1997). The best described members of this protein family, importin-α d β-catenin, were 30 31 32 shown to bind peptides in a stretched conformation in their binding groove (Conti et al., 1998; 33 34 Graham et al., 2000). This stretched conformation of the bound peptide is enforced by 35 36 37 bidentate hydrogen bonds between the backbone NH and O atoms of every second peptide 38 37 39 bond and conserved asparagine residues on H3 (N , superscripted numbers refer to residue 40 41 numbering within one repeat), while the specificity is established by interactions with the 42 43 44 peptide side chains. 45 46 47 nArmRPs bind peptide ligands of up to six amino acid units in length in a modular fashion, 48 49 i.e., two residues per repeat are recognized (Conti and Kuriyan, 2000; Conti et al., 1998). 50 51 52 Longer modular peptide binding has not been observed in nArmRP, because they do not 53 54 possess a regular curvature that fits the binding register dictated by the peptide bond lengths 55 56 57 (Reichen et al., 2016b). Designed armadillo repeat proteins (dArmRP) were developed to 58 59 expand this modular peptide recognition by engineering a completely regular scaffold that fits 60 61 62 63 4 64 65 the required register along the whole scaffold length and would therefore allow binding of, in 1 theory, infinitely long peptides (Alfarano et al., 2012; Madhurantakam et al., 2012; 2 3 4 Parmeggiani et al., 2008) (Fig. 1a). Based on this scaffold, surface-randomized repeats might 5 6 be developed that specifically bind a certain dipeptide in the context of a longer peptide, and 7 8 9 most likely this can be achieved by directed evolution methods. Such preselected repeats 10 11 could then in turn be assembled to create specific peptide binders against novel primary 12 13 peptide sequences without additional selections (Reichen et al., 2014a). 14 15 16 17 The correct curvature of the scaffold is a fundamental prerequisite for the development of 18 19 such a technology. The curvature of a repeat protein can be described by three parameters that 20 21 define a superhelix which is formed by solenoid proteins: The rise (h), 22 ( Ω) b w 23 24 neighboring repeats and the radius of the superhelix (r). For the parametrization of 25 26 experimental structures, these values are usually calculated between the center of mass (CoM) 27 28 29 of each repeat and the central axis of the superhelix (DiMaio et al., 2011; Kobe and Kajava, 30 31 2000; Park et al., 2015). A peptide in an extended conformation has a distance of 6.7-7.0 Å 32 33 α α 34 between the C atom of one residue and the C two residues further toward the C-terminus 35 36 (termed Cα(P/P+2) distance).