Binary patterning of polar and nonpolar amino acids has been used as the key design feature for constructing large combinatorial libraries of de novo proteins. Each position in a binary patterned sequence is designed explicitly to be either polar or nonpolar; however, the precise identities of these amino acids are varied extensively. The combinatorial underpinnings of the "binary code" strategy preclude explicit design of particular side chains at specified positions. Therefore, packing interactions cannot be specified a priori. To assess whether the binary code strategy can nonetheless produce well-folded de novo proteins, we constructed a second-generation library based upon a new structural scaffold designed to fold into 102-residue four-helix bundles. Characterization of five proteins chosen arbitrarily from this new library revealed that (1) all are ␣-helical and quite stable; (2) four of the five contain an abundance of tertiary interactions indicative of well-ordered structures; and (3) one protein forms a well-folded structure with native-like features. The proteins from this new 102-residue library are substantially more stable and dramatically more native-like than those from an earlier binary patterned library of 74-residue sequences. These findings demonstrate that chain length is a crucial determinant of structural order in libraries of de novo four-helix bundles. Moreover, these results show that the binary code strategy-if applied to an appropriately designed structural scaffold-can generate large collections of stably folded and/or native-like proteins.

No abstract

We previously reported the design of a library of de novo proteins targeted to fold into 4-helix bundles. 1 The library was created using a "binary code" strategy in which the sequence locations of polar and nonpolar amino acids were specified explicitly, but the identities of these side chains were varied extensively. Combinatorial diversity was made possible by the organization of the genetic code: Positions designed to contain polar amino acids were encoded by the degenerate DNA codon NAN, which codes for Lys, His, Glu, Gln, Asp, or Asn. Positions designed to contain nonpolar amino acids were encoded by NTN, which codes for Met, Leu, Ile, Val, or Phe (N represents a mixture of DNA nucleotides). We subsequently reported that approximately half the sequences in our initial library 1 bind heme. 2 The enormous diversity of sequences within the binary code library presents an opportunity for the isolation of de novo heme-based enzymes. We now establish the catalytic potential of the binary code proteins by demonstrating that several of the de novo heme proteins function as peroxidases.Natural peroxidases such as horseradish peroxidase (HRP) catalyze the reduction of hydrogen peroxide or alkyl peroxides to water or alcohol, respectively. The peroxidase mechanism is described by the following steps: 3 E represents the ferric resting state of the heme enzyme. Compound I is an intermediate two oxidation states above the resting state, and compound II is one oxidation state above the resting state. AH 2 is the reducing agent. Two commonly used reducing agents are 2,2′,5,5′-tetramethyl-benzidine (TMB) and 2,2′-azino-di(3-ethyl-benzthiazoline-6-sulfonic acid) (ABTS). These reagents become colored upon oxidation, thereby allowing peroxidase activity to be monitored spectraphotometrically.To search for novel peroxidases in our library of heme proteins, we developed an activity screen that does not require purification of the de novo proteins from cellular contaminants. Samples were prepared by a rapid freeze/thaw protocol 4 shown previously to produce protein samples of sufficient quality for various studies including NMR spectroscopy 5a and H/D exchange kinetics. 5b Freeze/thaw samples from 37 binary code proteins 6 were screened by monitoring oxidation of TMB. The results (Figure 1) indicate that several of the de novo proteins display activity significantly above the controls.On the basis of this screen, we chose four proteins for detailed analysis. 9 Proteins were purified, reconstituted with heme, and analyzed for peroxidase activity using ABTS as the reducing agent. Experimental data for protein 86 are shown in Figure 2. Neither the apoprotein nor the heme alone 10 were active, thereby demonstrating the heme-protein complex is the active species.Maximal velocities (V max ) were determined by plotting the data according to the Michaelis-Menten model (Figure 3). Turnover numbers (k cat ) were calculated by dividing V max by the concentration of heme protein. More detailed kinetic constants (k 1 and k 3 ) were determine...

The aggregation of the 37-residue polypeptide IAPP, as either insoluble amyloid or as small oligomers, appears to play a direct role in the death of pancreatic β-islet cells in type II diabetes. While IAPP has been known to be the primary component of type II diabetes amyloid, the molecular interactions responsible for this aggregation have not been identified. To identify the aggregation-prone region(s), we constructed a library of randomly generated point mutants of IAPP. This mutant IAPP library was expressed in E. coli as genetic fusions to the reporter protein enhanced green fluorescent protein (EGFP). Because IAPP aggregates rapidly, both independently and when fused to EGFP, the fusion protein does not yield a functional, fluorescent EGFP. However, mutations of IAPP that result in non-amyloidogenic sequences remain soluble and allow EGFP to fold and fluoresce. Using this screen, we identified 22 single mutations, 4 double mutations and 2 triple mutations of IAPP that appear to be less amyloidogenic than wild type human IAPP. A comparison of these sequences suggests residues 13 and 15-17 comprise an additional aggregation-prone region outside of the main amyloidogenic region of IAPP.The aggregation of misfolded proteins into toxic oligomers and fibers has been linked to a variety of diseases such as type II diabetes, Alzheimer's disease and Parkinson's disease. In type II diabetes the amyloid-forming peptide is islet amyloid polypeptide (IAPP, amylin). This 37 amino acid polypeptide misfolds and forms aggregates within the pancreas. This misfolding into toxic aggregates, such as small soluble oligomers or large fibers, is believed to contribute to the loss of pancreatic β-cells. While the exact role of IAPP in type II diabetes is unclear, it is known that IAPP is found as extracellular deposits of amyloid in approximately 95% of patients afflicted with type II diabetes (1-3). IAPP has also been shown to be a toxic agent in vitro when added to human islet β-cells (4).Many of the therapeutic strategies for preventing or slowing the progression of amyloid diseases such as type II diabetes, involve slowing or preventing the aggregation of the amyloidogenic proteins. To this end, a great deal of effort has been made in identifying the amino acids responsible for the aggregation-prone nature of amyloidogenic peptides such as †

Carbon monoxide binding was studied in a collection of de novo heme proteins derived from combinatorial libraries of sequences designed to fold into 4-helix bundles. The design of the de novo sequences was based on the previously reported "binary code" strategy, in which the patterning of polar and nonpolar amino acids is specified explicitly, but the exact identities of the side chains are varied extensively.(1) The combinatorial mixture of amino acids included histidine and methionine, which ligate heme iron in natural proteins. However, no attempt was made to explicitly design a heme binding site. Nonetheless, as reported previously, approximately half of the binary code proteins bind heme.(2) This collection of novel heme proteins provides a unique opportunity for an unbiased assessment of the functional potentialities of heme proteins that have not been prejudiced either by explicit design or by evolutionary selection. To assess the capabilities of the de novo heme proteins to bind diatomic ligands, we measured the affinity for CO, the kinetics of CO binding and release, and the resonance Raman spectra of the CO complexes for eight de novo heme proteins from two combinatorial libraries. The CO binding affinities for all eight proteins were similar to that of myoglobin, with dissociation constants (K(d)) in the low nanomolar range. The CO association kinetics (k(on)) revealed that the heme environment in all eight of the de novo proteins is partially buried, and the resonance Raman studies indicated that the local environment around the bound CO is devoid of hydrogen-bonding groups. Overall, the CO binding properties of the de novo heme proteins span a narrow range of values near the center of the range observed for diverse families of natural heme proteins. The measured properties of the de novo heme proteins can be considered as a "default" range for CO binding in alpha-helical proteins that have neither been designed to bind heme or CO, nor subjected to genetic selections for heme or CO binding.