Michael T. Stumpp

Designing Repeat Proteins: Well-expressed, Soluble and Stable Proteins from Combinatorial Libraries of Consensus Ankyrin Repeat Proteins

We describe an efficient way to generate combinatorial libraries of stable, soluble and well-expressed ankyrin repeat (AR) proteins. Using a combination of sequence and structure consensus analyses, we designed a 33 amino acid residue AR module with seven randomized positions having a theoretical diversity of 7.2x10(7). Different numbers of this module were cloned between N and C-terminal capping repeats, i.e. ARs designed to shield the hydrophobic core of stacked AR modules. In this manner, combinatorial libraries of designed AR proteins consisting of four to six repeats were generated, thereby potentiating the theoretical diversity. All randomly chosen library members were expressed in soluble form in the cytoplasm of Escherichia coli in amounts up to 200 mg per 1 l of shake-flask culture. Virtually pure proteins were obtained in a single purification step. The designed AR proteins are monomeric and display CD spectra identical with those of natural AR proteins. At the same time, our AR proteins are highly thermostable, with T(m) values ranging from 66 degrees C to well above 85 degrees C. Thus, our combinatorial library members possess the properties required for biotechnological applications. Moreover, the favorable biophysical properties and the modularity of the AR fold may account, partly, for the abundance of natural AR proteins.

Comprehensive survey of proteins targeted by chloroplast thioredoxin

Possible target proteins of chloroplast thioredoxin (Trx) have been investigated in the stroma lysate of spinach chloroplasts. For that purpose, we immobilized a mutant of m-type Trx in which an internal cysteine at the active site was substituted with serine, on cyanogen bromide-activated resin. By using this resin, the target proteins in chloroplast were efficiently acquired when they formed the mixed-disulfide intermediates with the immobilized Trxs. We could acquire Rubisco activase (45 kDa) and 2-Cys-type peroxiredoxin (Prx), which were recently identified as targets of chloroplast Trxs. Glyceraldehyde-3-phosphate dehydrogenase and sedoheputulose 1,7-bisphosphatase, well-known thiol enzymes in the Calvin cycle, also were recognized among the collected proteins, suggesting the method is applicable for our purpose. Furthermore, four proteins were identified from a homology search of the NH2-terminal sequence of the acquired proteins: glutamine synthetase, a protein homologous to chloroplast cyclophilin, a homolog of Prx-Q, and the Rubisco small subunit. The Trx susceptibilities of the recombinant cyclophilin and Prx-Q of Arabidopsis thaliana were then examined. The method developed in the present study is thus applicable to investigate the various redox networks via Trxs and the related enzymes in the cell.

Designed to be stable: Crystal structure of a consensus ankyrin repeat protein

Ankyrin repeat (AR) proteins mediate innumerable protein–protein interactions in virtually all phyla. This finding suggested the use of AR proteins as designed binding molecules. Based on sequence and structural analyses, we designed a consensus AR with fixed framework and randomized interacting residues. We generated several combinatorial libraries of AR proteins consisting of defined numbers of this repeat. Randomly chosen library members are expressed in soluble form in the cytoplasm of Escherichia coli constituting up to 30% of total cellular protein and show high thermodynamic stability. We determined the crystal structure of one of those library members to 2.0-Å resolution, providing insight into the consensus AR fold. Besides the highly complementary hydrophobic repeat–repeat interfaces and the absence of structural irregularities in the consensus AR protein, the regular and extended hydrogen bond networks in the β-turn and loop regions are noteworthy. Furthermore, all residues found in the turn region of the Ramachandran plot are glycines. Many of these features also occur in natural AR proteins, but not in this rigorous and standardized fashion. We conclude that the AR domain fold is an intrinsically very stable and well-expressed scaffold, able to display randomized interacting residues. This scaffold represents an excellent basis for the design of novel binding molecules.

Efficient Tumor Targeting with High-Affinity Designed Ankyrin Repeat Proteins: Effects of Affinity and Molecular Size

Slow-clearing, tumor-targeting proteins such as monoclonal antibodies typically exhibit high tumor accumulation but low tissue contrast, whereas intermediate-sized proteins such as scFvs show faster clearance but only moderate tumor accumulation. For both, tumor targeting does not seem to improve further above an optimal affinity. We show here that with very small high-affinity proteins such as designed ankyrin repeat proteins (DARPins), these limits can be overcome. We have systematically investigated the influence of molecular mass and affinity on tumor accumulation with DARPins with specificity for HER2 in SK-OV-3.ip nude mouse xenografts. DARPins with a mass of 14.5 kDa and affinities between 270 nmol/L and 90 pmol/L showed a strong correlation of tumor accumulation with affinity to HER2, with the highest affinity DARPin reaching 8% ID/g after 24 hours and 6.5% ID/g after 48 hours (tumor-to-blood ratio >60). Tumor autoradiographs showed good penetration throughout the tumor mass. Genetic fusion of two DARPins (30 kDa) resulted in significantly lower tumor accumulation, similar to values observed for scFvs, whereas valency had no influence on accumulation. PEGylation of the DARPins increased the circulation half-life, leading to higher tumor accumulation (13.4% ID/g after 24 hours) but lower tumor-to-blood ratios. Affinity was less important for tumor uptake of the PEGylated constructs. We conclude that two regimes exist for delivering high levels of drug to a tumor: small proteins with very high affinity, such as unmodified DARPins, and large proteins with extended half-life, such as PEGylated DARPins, in which the importance of affinity is less pronounced.

DARPins: a new generation of protein therapeutics.

DARPins (designed ankyrin repeat proteins) are a novel class of binding molecules with the potential to overcome limitations of monoclonal antibodies, hence allowing novel therapeutic approaches. DARPins are small, single domain proteins (14 kDa) which can be selected to bind any given target protein with high affinity and specificity. These characteristics make them ideal agonistic, antagonistic or inhibitory drug candidates. Furthermore, DARPins can be engineered to carry various effector functions or combine multiple binding specificities, enabling completely new drug formats. Taken together, DARPins are a prominent member of the next generation of protein therapeutics with the potential to surpass existing antibody drugs.

Designing repeat proteins: modular leucine-rich repeat protein libraries based on the mammalian ribonuclease inhibitor family.

We present a novel approach to design repeat proteins of the leucine-rich repeat (LRR) family for the generation of libraries of intracellular binding molecules. From an analysis of naturally occurring LRR proteins, we derived the concept to assemble repeat proteins with randomized surface positions from libraries of consensus repeat modules. As a guiding principle, we used the mammalian ribonuclease inhibitor (RI) family, which comprises cytosolic LRR proteins known for their extraordinary affinities to many RNases. By aligning the amino acid sequences of the internal repeats of human, pig, rat, and mouse RI, we derived a first consensus sequence for the characteristic alternating 28 and 29 amino acid residue A-type and B-type repeats. Structural considerations were used to replace all conserved cysteine residues, to define less conserved positions, and to decide where to introduce randomized amino acid residues. The so devised consensus RI repeat library was generated at the DNA level and assembled by stepwise ligation to give libraries of 2-12 repeats. Terminal capping repeats, known to shield the continuous hydrophobic core of the LRR domain from the surrounding solvent, were adapted from human RI. In this way, designed LRR protein libraries of 4-14 LRRs (equivalent to 130-415 amino acid residues) were obtained. The biophysical analysis of randomly chosen library members showed high levels of soluble expression in the Escherichia coli cytosol, monomeric behavior as characterized by gel-filtration, and alpha-helical CD spectra, confirming the success of our design approach.

A novel strategy to design binding molecules harnessing the modular nature of repeat proteins

Repeat proteins, such as ankyrin or leucine‐rich repeat proteins, are ubiquitous binding molecules, which occur, unlike antibodies, intra‐ and extracellularly. Their unique modular architecture features repeating structural units (repeats), which stack together to form elongated repeat domains displaying variable and modular target‐binding surfaces. Based on this modularity, we developed a novel strategy to generate combinatorial libraries of polypeptides with highly diversified binding specificities. This strategy includes the consensus design of self‐compatible repeats displaying variable surface residues and their random assembly into repeat domains. We envision that such repeat protein libraries will be highly valuable sources for novel binding molecules especially suitable for intracellular applications.

Designed Armadillo Repeat Proteins as General Peptide-Binding Scaffolds: Consensus Design and Computational Optimization of the Hydrophobic Core

Armadillo repeat proteins are abundant eukaryotic proteins involved in several cellular processes, including signaling, transport, and cytoskeletal regulation. They are characterized by an armadillo domain, composed of tandem armadillo repeats of approximately 42 amino acids, which mediates interactions with peptides or parts of proteins in extended conformation. The conserved binding mode of the peptide in extended form, observed for different targets, makes armadillo repeat proteins attractive candidates for the generation of modular peptide-binding scaffolds. Taking advantage of the large number of repeat sequences available, a consensus-based approach combined with a force field-based optimization of the hydrophobic core was used to derive soluble, highly expressed, stable, monomeric designed proteins with improved characteristics compared to natural armadillo proteins. These sequences constitute the starting point for the generation of designed armadillo repeat protein libraries for the selection of peptide binders, exploiting their modular structure and their conserved binding mode.

Consensus design of repeat proteins.

Consensus design is a valuable protein‐engineering method that is based on statistical information derived from sequence alignments of homologous proteins. Recently, consensus design was adapted to repeat proteins. We discuss the potential of this novel repeat‐based approach for the design of consensus repeat proteins and repeat protein libraries and summarize recent results from such experiments.

Designed armadillo repeat proteins

Designed Armadillo repeat proteins (ArmRPs) are a novel class of binding proteins intended for general modular peptide binding and have very favorable expression and stability properties. Using a combination of sequence and structural consensus analyses, we generated a 42-amino-acid designed Armadillo repeat module with six randomized positions, having a theoretical diversity of 9.9×10(6) per repeat. Structural considerations were used to replace cysteine residues, to define less conserved positions and to decide where to introduce randomized amino acid residues for potential interactions with the target peptide. Based on these concepts, combinatorial libraries of designed ArmRPs were assembled. The most stable version of designed ArmRP in library format was the N5C format, with three randomized library repeat modules flanked by full consensus repeat modules on either side and, in turn, flanked by N- and C-terminal capping repeats. Unselected members of this library were well expressed in the Escherichia coli cytoplasm, monomeric and showed the expected CD spectra and cooperative unfolding. N5C libraries were used in ribosome display selections against the peptide neurotensin. Highly specific peptide binders were enriched after four rounds of selections using ribosome display. Four peptide side chains were shown to contribute most of the interaction energy, and single alanine mutants could be discriminated. Thus, designed ArmRP libraries can become valuable sources for peptide binding molecules because of their favorable biophysical properties and with a potential for application in general modular peptide recognition.