Protein Design: Methods and Applications

Protein

Design

Valentin Köhler Editor

Methods and Applications

Second Edition

Methods in

Molecular Biology 1216

METHODS I N MOLECULAR BIOLOGY

Series Editor

John M. Walker

School of Life Sciences

University of Hertfordshire

Hat fi eld, Hertfordshire, AL10 9AB, UK

For further volumes:

http://www.springer.com/series/7651

Protein Design

Methods and Applications

Second Edition

Edited by

Valentin Köhler

Department of Chemistry, University of Basel, Switzerland

ISSN 1064-3745 ISSN 1940-6029 (electronic)

ISBN 978-1-4939-1485-2 ISBN 978-1-4939-1486-9 (eBook)

DOI 10.1007/978-1-4939-1486-9

Springer New York Heidelberg Dordrecht London

Library of Congress Control Number: 2014947803

© Springer Science+Business Media New York 2014

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is

concerned, specifi cally the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction

on microfi lms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation,

computer software, or by similar or dissimilar methodology now known or hereafter developed. Exempted from this

legal reservation are brief excerpts in connection with reviews or scholarly analysis or material supplied specifi cally for

the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work.

Duplication of this publication or parts thereof is permitted only under the provisions of the Copyright Law of the

Publisher’s location, in its current version, and permission for use must always be obtained from Springer. Permissions

for use may be obtained through RightsLink at the Copyright Clearance Center. Violations are liable to prosecution

under the respective Copyright Law.

The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not

imply, even in the absence of a specifi c statement, that such names are exempt from the relevant protective laws and

regulations and therefore free for general use.

While the advice and information in this book are believed to be true and accurate at the date of publication, neither

the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be

made. The publisher makes no warranty, express or implied, with respect to the material contained herein.

Printed on acid-free paper

Humana Press is a brand of Springer

Springer is part of Springer Science+Business Media (www.springer.com)

Editor

Valentin Köhler

Department of Chemistry

University of Basel

Switzerland

v

The second edition of protein design in the Methods in Molecular Biology series aims at

providing the reader with practical guidance and general ideas on how to approach a poten￾tial protein design project. Considering the complexity of the subject and its attention in

the scientifi c community it is apparent that only a selection of subjects, approaches, meth￾ods, studies, and ideas can be presented.

The design of well-folded peptidestructures and the redesign of existing proteins serve

multiple purposes from potentially unlimited and only just developing applications in medi￾cine, material science, catalysis, the realization of systems chemistry, and synthetic biology

to a deeper understanding of molecular evolution.

The book is roughly organized in increasing complexity of the systems studied.

Additional emphasis is put on metals as structure-forming elements and functional sites of

proteins towards the end.

A computational algorithm for the design of stable alpha helices is discussed in the fi rst

chapter and is accessible in the form of a web-based tool. An extensive review on mono￾meric β-hairpin and β-sheet peptides follows. In the design of these species any tendency to

self-assemble has to be carefully considered. In contrast, Chapter 3exploits just this phe￾nomenon—peptides engineered to self-assemble into fi brils.

Subsequently, some possibilities and aspects resulting from the incorporation of unnat￾ural amino acids are outlined. In the practical methods chapter on the redesign of RNase

A, a variable α-helical fragment is reassembled with the remainder of the protein structure,

generated by enzymatic cleavage. Chapter 5discusses the design and characterization of

fl uorinated proteins, which are entirely synthetic. Comparisons to non-fl uorinated analo￾gous structures are included and practical advice is offered.

This is followed by an overview of considerations for the generation of binary- patterned

protein libraries leading on to library-scale computational protein design for the engineer￾ing of improved protein variants. The latter is exemplifi ed for cellobiohydrolase II and a

study aimed at changing the co-substrate specifi city of a ketol-acid reductoisomerase.

Chapter 8focuses on the elaboration of symmetric protein folds in an approach termed

“top-down symmetric deconstruction,” which prepares the folds for subsequent functional

design studies.

The identifi cation of a suitable scaffold for design purposes by means of the scaffold

search program ScaffoldSelection is the topic of Chapter 9 .

The computational design of novel enzymes without cofactor is demonstrated for a

Diels-Alderase in Chapter 10 .

The fi nal four chapters deal with metal involvement in the designed or redesigned

structures, either as structural elements or functional centers. The begin is madewith a

tutorial review that imparts general knowledge for the design of peptide scaffolds as novel

pre-organized ligands for metal-ion coordination and then exemplifi es these further in a

respective case study. This is followed by an introduction on the computational design of

metalloproteins, which encompasses metal incorporation into existing folds, fold design by

Pref ace

vi

exploiting symmetry, and fold design in asymmetric scaffolds. The potential power of cofac￾tor exchange is addressed with the focus on a practical protocol for the preparation of apo￾myoglobin and the incorporation of zinc porphyrin in the penultimate chapter. The book

concludes with a case study on the computational redesign of metalloenzymes carried out

with the aim to assign a new enzymatic function.

This volume of Methods in Molecular Biology contains a number of practical proto￾cols, but compared to other volumes of the series, a larger contribution of reviews or gen￾eral introductions is provided. Those, however, are presented in a tutorial fashion to

communicate principles that can be applied to individual research projects.

I sincerely do hope that the reader fi nds this edition of protein design helpful for devis￾ing their own experiments.

I warmly thank all the authors for their very valuable contributions, their dedication,

and not least their patience.

Basel, Switzerland Valentin Köhler

Preface

vii

Preface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v

Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix

1 De Novo Design of Stable α-Helices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Alexander Yakimov, Georgy Rychkov,and Michael Petukhov

2 Design of Monomeric Water-Soluble β-Hairpin and β-Sheet Peptides . . . . . . . 15

M. Angeles Jiménez

3 Combination of Theoretical and Experimental Approaches

for the Design and Study of Fibril-Forming Peptides. . . . . . . . . . . . . . . . . . . . 53

Phanourios Tamamis, Emmanouil Kasotakis, Georgios Archontis,

and Anna Mitraki

4 Posttranslational Incorporation of Noncanonical Amino Acids

in the RNase S System by Semisynthetic Protein Assembly . . . . . . . . . . . . . . . 71

Maika Genzand Norbert Sträter

5 Design, Synthesis, and Study of Fluorinated Proteins. . . . . . . . . . . . . . . . . . . . 89

Benjamin C. Buerand E. Neil G. Marsh

6 High-Quality Combinatorial Protein Libraries Using the Binary

Patterning Approach. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117

Luke H. Bradley

7 Methods for Library-Scale Computational Protein Design. . . . . . . . . . . . . . . . 129

Lucas B. Johnson, Thaddaus R. Huber,and Christopher D. Snow

8 Symmetric Protein Architecture in Protein Design:

Top- Down Symmetric Deconstruction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161

Liam M. Longoand Michael Blaber

9 Identification of Protein Scaffolds for Enzyme Design

Using Scaffold Selection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183

André C. Stiel, Kaspar Feldmeier,and Birte Höcker

10 Computational Design of Novel Enzymes Without Cofactors . . . . . . . . . . . . . 197

Matthew D. Smith, Alexandre Zanghellini,

and Daniela Grabs-Röthlisberger

11 De Novo Design of Peptide Scaffolds as Novel Preorganized Ligands

for Metal-Ion Coordination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211

Aimee J. Gambleand Anna F. A. Peacock

12 Computational Design of Metalloproteins. . . . . . . . . . . . . . . . . . . . . . . . . . . . 233

Avanish S. Parmar, Douglas Pike,and Vikas Nanda

Contents

viii

13 Incorporation of Modified and Artificial Cofactors into Naturally

Occurring Protein Scaffolds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251

Koji Oohoraand Takashi Hayashi

14 Computational Redesign of Metalloenzymes for Catalyzing

New Reactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265

Per Jr. Greisenand Sagar D. Khare

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275

Contents

ix

GEORGIOS ARCHONTIS • Department of Physics , University of Cyprus , Nicosia , Cyprus

MICHAEL BLABER • Department of Biomedical Sciences, College of Medicine , Florida State

University , Tallahassee , FL , USA

LUKE H. BRADLEY • Departments of Anatomy and Neurobiology, Molecular

and Cellular Biochemistry, and the Center of Structural Biology , University of Kentucky

College of Medicine , Lexington , KY , USA

BENJAMIN C. BUER • Department of Chemistry , University of Michigan , Ann Arbor , MI , USA

KASPAR FELDMEIER • Max Planck Institute for Developmental Biology , Tübingen , Germany

AIMEE J. GAMBLE • School of Chemistry , University of Birmingham , Birmingham , UK

MAIKA GENZ • Faculty of Chemistry and Mineralogy, Center for Biotechnology

and Biomedicine, Institute of Bioanalytical Chemistry , University of Leipzig , Leipzig , Germany

DANIELA GRABS-RÖTHLISBERGER • Arzeda Corp. , Seattle , WA , USA

PER JR. GREISEN • Department of Biochemistry , University of Washington , Seattle , WA , USA

TAKASHI HAYASHI • Department of Applied Chemistry, Graduate School of Engineering ,

Osaka University , Suita , Osaka , Japan

BIRTE HÖCKER • Max Planck Institute for Developmental Biology , Tübingen , Germany

THADDAUS R. HUBER • Department of Chemical and Biological Engineering ,

Colorado State University , Fort Collins , CO , USA

M. ANGELES JIMÉNEZ • Consejo Superior de Investigaciones Científi cas (CSIC) ,

Instituto de Química Física Rocasolano (IQFR) , Madrid , Spain

LUCAS B. JOHNSON • Department of Chemical and Biological Engineering ,

Colorado State University , Fort Collins , CO , USA

EMMANOUIL KASOTAKIS • Department of Materials Science and Technology ,

University of Crete , Heraklion, Crete , Greece

SAGAR D. KHARE • Department of Chemistry and Chemical Biology, Center for Integrative

Proteomics Research , Rutgers University , Piscataway , NJ , USA

LIAM M. LONGO • Department of Biomedical Sciences, College of Medicine , Florida State

University , Tallahassee , FL , USA

E. NEIL G. MARSH • Department of Chemistry , University of Michigan , Ann Arbor , MI ,

USA ; Department of Biological Chemistry , University of Michigan Medical School ,

Ann Arbor , MI , USA

ANNA MITRAKI • Department of Materials Science and Technology , University of Crete ,

Heraklion, Crete , Greece ; Institute for Electronic Structure and Laser, Foundation

for Research and Technology- Hellas (IESL-FORTH) , Heraklion, Crete , Greece

VIKAS NANDA • Department of Biochemistry and Molecular Biology, Center for Advanced

Biotechnology and Medicine, Robert Wood Johnson Medical School , University of Medicine

and Dentistry of New Jersey , Piscataway , NJ , USA

KOJI OOHORA • Department of Applied Chemistry, Graduate School of Engineering ,

Osaka University , Suita , Osaka , Japan

Contributors

x

AVANISH S. PARMAR • Department of Biochemistry and Molecular Biology,

Center for Advanced Biotechnology and Medicine, Robert Wood Johnson Medical School ,

University of Medicine and Dentistry of New Jersey , Piscataway , NJ , USA

ANNA F. A. PEACOCK • School of Chemistry , University of Birmingham , Birmingham , UK

MICHAEL PETUKHOV • Department of Molecular and Radiation Biophysics, Petersburg

Nuclear Physics Institute , NRC Kurchatov Institute , Gatchina , Russia ; Saint Petersburg

State Polytechnical University , Saint Petersburg , Russia

DOUGLAS PIKE • Department of Biochemistry and Molecular Biology, Center for Advanced

Biotechnology and Medicine, Robert Wood Johnson Medical School , University of Medicine

and Dentistry of New Jersey , Piscataway , NJ , USA

GEORGY RYCHKOV • Department of Molecular and Radiation Biophysics,

Petersburg Nuclear Physics Institute , NRC Kurchatov Institute , Gatchina , Russia ;

Saint Petersburg State Polytechnical University , Saint Petersburg , Russia

MATTHEW D. SMITH • Molecular and Cellular Biology Program , University of Washington ,

Seattle , WA , USA

CHRISTOPHER D. SNOW • Department of Chemical and Biological Engineering ,

Colorado State University , Fort Collins , CO , USA

ANDRÉ C. STIEL • Max Planck Institute for Developmental Biology , Tübingen , Germany

NORBERT STRÄTER • Faculty of Chemistry and Mineralogy, Center for Biotechnology

and Biomedicine, Institute of Bioanalytical Chemistry , University of Leipzig ,

Leipzig , Germany

PHANOURIOS TAMAMIS • Department of Physics , University of Cyprus , Nicosia , Cyprus

ALEXANDER YAKIMOV • Department of Molecular and Radiation Biophysics,

Petersburg Nuclear Physics Institute , NRC Kurchatov Institute , Gatchina , Russia ;

Saint Petersburg State Polytechnical University , Saint Petersburg , Russia

ALEXANDRE ZANGHELLINI • Arzeda Corp. , Seattle , WA , USA

Contributors

1

Valentin Köhler (ed.), Protein Design: Methods and Applications, Methods in Molecular Biology, vol. 1216,

DOI 10.1007/978-1-4939-1486-9_1, © Springer Science+Business Media New York 2014

Chapter 1

De Novo Design of Stable α-Helices

Alexander Yakimov, Georgy Rychkov, and Michael Petukhov

Abstract

Recent studies have elucidated key principles governing folding and stability of α-helices in short peptides

and globular proteins. In this chapter we review briefly those principles and describe a protocol for the

de novo design of highly stable α-helixes using the SEQOPT algorithm. This algorithm is based on

AGADIR, the statistical mechanical theory for helix-coil transitions in monomeric peptides, and the tunneling

algorithm for global sequence optimization.

Key words α-Helix, Stability, Sequence optimization, Solubility

1 Introduction

The α-helix is one of the most abundant elements of protein

secondary structure. Numerous studies of α-helical peptides not

only contributed to a better understanding of protein folding but

also represent an increasing pharmacological interest in their prac￾tical utility for the development of novel therapeutics to modulate

protein-protein interactions in vivo [1].

A large amount of information on α-helix folding and stability

has been gathered since the early 1990s [2, 3]. The data show that

sequences of protein helices are not, in general, optimized for high

conformational stability. This may be an important factor in prevent￾ing the accumulation of nonnative intermediates in protein folding

[4–6]. Nevertheless, designing short α-helical peptides and proteins

with sufficient conformational stability under given environmental

conditions (temperature, pH, and ionic strength) still remains an

area of intense investigation in protein engineering [1].

Furthermore a large body of information has been accumulated

regarding the factors which govern the stability of α-helices in

proteins and the helical behavior of both isolated protein fragments

and designed helical sequences in solution [4]. These factors

include interactions between amino acid side chains [7–9], the

helix macrodipole [10], and terminal capping [11].

2

All these factors have been considered separately in attempts to

increase the conformational stability of α-helices in peptides and in

natural proteins [12, 13]. However, the design of peptide sequences

with the optimal implementation of all these factors can often not

be achieved even for short peptides, since they can be mutually

exclusive. The stability of the α-helix is controlled by diverse and

accurately balanced interactions. For example a positively charged

amino acid at position i prefers that the i+3, i+4 and also the i−3,

i−4 positions of the helix (Fig. 1) are occupied by negatively

charged residues that may on the other hand be unfavorable for

helix formation if they occur close to the carboxy-terminus where

they lead to negative interactions with the helix macrodipole [10].

The problem increases rapidly with peptide length, since it deter￾mines the number of interactions to be considered.

Several de novo protein design methods, based on RosettaDesign

[14], EGAD [15], Liang-Grishin [16], and RosettaDesign-SR [17]

programs, have been developed during the past decade. These meth￾ods can also be applied for the design of α-helix-forming peptides

[18]. Unlike these approaches, the AGADIR method is based on

free energy contributions, obtained from experimental data.

The number of possible sequences of a peptide with N amino

acid residues equals 20N. Thus, it is computationally impossible to

calculate the helical content for a complete permutation library

even for short peptides as short as ten amino acids. To overcome

this problem we used the tunneling algorithm for global optimiza￾tion of multidimensional functions [19]. The main advantage of

this approach is that it does not require an examination of all pos￾sible sequences to find a suitable solution for most practical pur￾poses. The method is simple and robust and requires only the

calculation of the first derivatives of the goal function. It has been

reported that the method was successfully applied to identify global

minima to many problems with many thousands of local minima

[19]. However all available global optimization techniques can be

described as random walkers which cover to a greater or lesser

Fig. 1 Schematic view of the physical interactions stabilizing the α-helix segment

Alexander Yakimov et al.

3

extent a significant region of phase space spanned by the task at

hand. None of them can claim the true globality of a found solu￾tion. Besides taking into account imperfectness of theoretical

approximations employed to predict helix stability, it is unlikely

that the solution for any peptide sequence above a certain length

(5–7 amino acids) can be globally optimized currently and in the

near future. The inability of theoretical models to guarantee

convergence to a globally optimized peptide sequence motivates

the development of efficient tools for protein helix optimization,

even if the inherent problem itself cannot be overcome. For protein

engineering applications sufficiently optimized sequences are

employed instead of truly globally optimized ones. Creating and

testing such a tool on short peptide helices was the main goal of

the work presented in the form of a practical method.

Recently we developed a new method for the design of

α-helices in peptides and proteins using AGADIR (located at

http://agadir.crg.es/) [20], the statistical mechanical theory for

helix-coil transitions in monomeric peptides, and the tunneling

algorithm of global optimization of multidimensional functions

[19] for optimization of amino acid sequences [5]. Unlike tradi￾tional approaches that are often used to increase protein stability

by adding a few favorable interactions to the protein structure, this

method deals with all possible sequences of protein helices and

selects a suitable one. Under certain conditions the method can be

a powerful practical tool not only for the design of highly stable

peptide helices but also for protein engineering purposes. In the

study for the design of peptide helices we used an approach com￾bining statistical mechanical calculations based on the AGADIR

model [12] including several of its more recent modifications

[21–27] and the global optimization algorithm [19].

In work [5] we used one sequence approximation of the

AGADIR model (AGADIR1s) for helix-random coil transitions in

monomeric peptides. As any other theoretical model it has its own

simplifications and limitations. Most importantly it includes the

AGADIR partition function physical interactions only within heli￾cal segments and those from a few flanking residues at both N- and

C-termini (the so-called N- and C-capping interactions). The

SEQOPT sequence optimization is not only applicable for short

monomeric peptides in an aqueous environment but also for

solvent-exposed parts of protein alpha-helices which show only

intrahelical residue interactions. As another important simplifica￾tion AGADIR1s ignores the possible existence of multiple helical

segments in each peptide conformation. Multiple sequence approx￾imation (AGADIRms) of the AGADIR model has also been devel￾oped [28] and its predictions of peptide conformational stability

were compared with results of AGADIR1s as well as with Zimm￾Bragg and Lifson-Roig classic models for helix-coil transition in

peptides. It was shown that for all tested peptides having less than

De Novo Design of Stable α-Helices

4

56 residues the helical contents predicted by AGADIR1s are within

0.3 % error with those of AGADIRms. In addition AGADIR1s is

computationally much faster.

In the mid-1970s it was predicted by Finkelstein and Ptitsyn that

short peptides consisting of amino acids with high α-helix propen￾sity should have a fairly stable α-helical conformation in aqueous

solution [29–33]. Later this theory has been verified experimen￾tally by examining synthetic peptide sequences of ribonuclease A

[34, 35]. The theoretical model developed by Finkelstein and

Ptitsyn describes the probability of the formation of α-helices and

β-structures and turns in short peptides and globular proteins based

on the modified classical Zimm-Bragg model. It takes into account

some additional physical interactions, including hydrophobic inter￾actions of a number of amino acid side chains, electrostatic interac￾tions between the charged side chains themselves, as well as the

α-helix macrodipole. The computer program (ALB) based on this

theoretical model was shown to successfully predict not only an

approximate level of the conformational stability of α-helical

peptides [2] but also, with a probability of ~65 %, the distribution

of secondary structure elements in globular proteins.

Beginning in the late 1980s and increasing in the 1990s, a

large number of experiments with amino acid substitutions in short

synthetic peptides exploring different interactions in α-helices have

been described in the literature [3]. We would like to point out the

approach proposed by Scholtz and Baldwin, which enables the

accumulation of sufficient experimental data to proceed to a quan￾titative description of the cooperative mechanisms of conforma￾tional transitions of α-helical conformations in peptides with

random sequences.

Collected data allowed to establish the principle of intrinsic

helical propensity of any amino acid to populate the α-helix forma￾tion. This propensity [22] has been attributed to changes of con￾figurational entropy [36] and solvent electrostatic screening of

amino acid side chains [37]. For instance methionine, alanine, leu￾cine, uncharged glutamic acid, and Lys have high intrinsic helical

propensities, whereas proline and glycine have poor ones. Proline

residues either break or kink a helix, both because they cannot pro￾vide an amide hydrogen for hydrogen bonding (having no amide

hydrogen), and also because its side chain interferes sterically with

the backbone of the preceding turn; inside a helix, this forces a

bend of about 30° in the helix axis [38]. Nevertheless due to its

rigid structure proline is often found to be the first N-terminal resi￾due in protein α-helices [39]. On the other hand glycine also tends

to disrupt helices because its high conformational flexibility makes

it entropically expensive to adopt the relatively constrained α-helical

structure. Nevertheless it often plays a role as N- and C-cap residue

of protein helices [40].

1.1  α-Helix Structure

and Stability

Alexander Yakimov et al.