Novel folded protein domains generated by combinatorial shuffling of polypeptide segments

It has been proposed that the architecture of protein domains has evolved by the combinatorial assembly and/or exchange of smaller polypeptide segments. To investigate this proposal, we fused DNA encoding the N-terminal half of a beta-barrel domain (from cold shock protein CspA) with fragmented genomic Escherichia coli DNA and cloned the repertoire of chimeric polypeptides for display on filamentous bacteriophage. Phage displaying folded polypeptides were selected by proteolysis; in most cases the protease-resistant chimeric polypeptides comprised genomic segments in their natural reading frames. Although the genomic segments appeared to have no sequence homologies with CspA, one of the originating proteins had the same fold as CspA, but another had a different fold. Four of the chimeric proteins were expressed as soluble polypeptides; they formed monomers and exhibited cooperative unfolding. Indeed, one of the chimeric proteins contained a set of very slowly exchanging amides and proved more stable than CspA itself. These results indicate that native-like proteins can be generated directly by combinatorial segment assembly from nonhomologous proteins, with implications for theories of the evolution of new protein folds, as well as providing a means of creating novel domains and architectures in vitro.     

A thermodynamic definition of protein domains

Protein domains are conspicuous structural units in globular proteins, and their identification has been a topic of intense biochemical interest dating back to the earliest crystal structures. Numerous disparate domain identification algorithms have been proposed, all involving some combination of visual intuition and/or structure-based decomposition. Instead, we present a rigorous, thermodynamically-based approach that redefines domains as cooperative chain segments. In greater detail, most small proteins fold with high cooperativity, meaning that the equilibrium population is dominated by completely folded and completely unfolded molecules, with a negligible subpopulation of partially folded intermediates. Here, we redefine structural domains in thermodynamic terms as cooperative folding units, based on m-values, which measure the cooperativity of a protein or its substructures. In our analysis, a domain is equated to a contiguous segment of the folded protein whose m-value is largely unaffected when that segment is excised from its parent structure. Defined in this way, a domain is a self-contained cooperative unit; i.e., its cooperativity depends primarily upon intrasegment interactions, not intersegment interactions. Implementing this concept computationally, the domains in a large representative set of proteins were identified; all exhibit consistency with experimental findings. Specifically, our domain divisions correspond to the experimentally determined equilibrium folding intermediates in a set of nine proteins. The approach was also proofed against a representative set of 71 additional proteins, again with confirmatory results. Our reframed interpretation of a protein domain transforms an indeterminate structural phenomenon into a quantifiable molecular property grounded in solution thermodynamics.     

Solution NMR structure of the major cold shock protein (CspA) from Escherichia coli: identification of a binding epitope for DNA.

Sequence-specific 1H and 15N resonance assignments have been determined for the major cold shock protein (CspA) from Escherichia coli with recently developed three-dimensional triple-resonance NMR experiments. By use of these assignments, five antiparallel beta-strands were identified from analysis of NMR data. Strands 1-4 have a classical 3-2-1-4 Greek key beta-sheet topology and there are two beta-bulges, at positions Lys10-Trp11 and Gly65-Asn66. Three-dimensional structures of CspA were generated from NMR data by using simulated annealing with molecular dynamics. The overall chain fold of CspA is a beta-barrel structure, with a tightly packed hydrophobic core. Two-dimensional isotope-edited pulsed-field gradient 15N-1H heteronuclear single-quantum coherence spectroscopy was used to characterize the 15N-1H fingerprint spectrum with and without a 24-base oligodeoxyribonucleotide, 5'-AACGGTTTGACGTACAGACCATTA-3'. Protein-DNA complex formation perturbs a subset of the amide resonances that are located mostly on one face of the CspA molecule. This portion of the CspA molecular surface includes two putative RNA-binding sequence motifs which contribute to an unusual cluster of eight surface aromatic side chains: Trp11, Phe12, Phe18, Phe20, Phe31, His33, Phe34, and Tyr42. These surface aromatic groups, and also residues Lys16, Ser44, and Lys60 located on this same face of CspA, are highly conserved in the family of CspA homologues. These isotope-edited pulsed-field gradient NMR data provide a low-resolution mapping of a DNA-binding epitope on CspA.     

Metastability of the folded states of globular proteins.

The possibility that several metastable minima exist in which the folded forms of a polypeptide chain have similar structural characteristics but different energies is suggested. The validity of this hypothesis is illustrated with the aid of simulation methods on a model protein that folds into a beta-barrel structure. Some implications of this hypothesis such as the existence of multiple pathways with intermediates for protein folding are discussed.     

Periodicity of leucine and tandem repetition of a 24-amino acid segment in the primary structure of leucine-rich alpha 2-glycoprotein of human serum.

The complete primary structure of the 3.1S leucine-rich alpha 2-glycoprotein (LRG) present in human plasma has been determined. This protein (Mr approximately 45,000) consists of a single polypeptide chain with one galactosamine and four glucosamine oligosaccharides attached. The polypeptide has two intrachain disulfide bonds and contains 312 amino acid residues of which 66 are leucine. The amino acid sequence can be exactly divided into 13 segments of 24 residues each, eight of which exhibit a periodic pattern in the occurrence of leucine, proline, and asparagine. The consensus sequence for the repeating tetracosapeptide unit is Pro-Xaa-Xaa-Leu-Leu-Xaa-Xaa-Xaa-X aa-Xaa-Leu-Xaa-Xaa-Leu-Xaa-Leu-Xaa-Xaa-Asn-Xaa-Leu-Xaa-Xaa-Leu. This periodicity suggests that the unique structure of LRG arose from a series of unequal crossovers of a precursor oligonucleotide sequence that encoded a building block rich in leucine. Overall, the amino acid sequence of LRG is not significantly homologous to the continuous sequence of any protein in the current data base. However, the consensus tetracosapeptide sequence shows strong homology to segments of many mitochondrial proteins, viral envelope proteins, and oncogene proteins that have a high leucine content and transmembrane domains. Tandem repetition of similar segments also occurs in apolipoproteins that have amphipathic helical potential. Prediction of the secondary structure by the Chou-Fasman rules and calculation of the hydrophilic/hydrophobic profile by several methods confirm the tandem repetition of largely hydrophobic structural units; these begin with a beta-turn that leads into an organized structure with alpha-helical or beta-sheet potential. These structural characteristics and the homology to mitochondrial proteins and apolipoproteins suggest that LRG is a membrane-derived or membrane-associated protein containing a series of domains capable of bipolar surface orientation.     

Liberation of an interaction domain from the phosphotransfer region of CheA, a signaling kinase of Escherichia coli.

The CheA protein of Escherichia coli is a histidine autokinase that donates its phosphate groups to two target proteins, CheY and CheB, to regulate flagellar rotation and sensory adaptation during chemotactic responses. The amino-terminal third of CheA contains the autophosphorylation site, determinants needed to interact with the catalytic center of the molecule, and determinants needed for specific recognition of its phosphorylation targets. To understand the structural basis for these activities, we examined the domain organization of the CheA phosphotransfer region by using DNA sequence analysis, limited proteolytic digestion, and a genetic technique called domain liberation. Comparison of the functionally interchangeable CheA proteins of E. coli and Salmonella typhimurium revealed two extensively mismatched segments within the phosphotransfer region, 22 and 25 aa long, with sequences characteristic of domain linkers. Both segments were readily susceptible to proteases, implying that they have an extended, flexible structure. In contrast, the intervening segments of the phosphotransfer region, designated P1 and P2 (roughly 140 and 65 aa, respectively), were relatively insensitive, suggesting they correspond to more compactly folded structural domains. Their functional properties were explored by identifying portions of the cheA coding region capable of interfering with chemotactic behavior when "liberated" and expressed as polypeptides. P1 fragments were not inhibitory, but P2 fragments blocked the interaction of CheY with the rotational switch at the flagellar motor, leading to incessant forward swimming. These results suggest that P2 contains CheY-binding determinants which are normally responsible for phosphotransfer specificity. Domain-liberation approaches should prove generally useful for analyzing multidomain proteins and their interaction targets.     

Specific binding of a prokaryotic ribosomal protein to a eukaryotic ribosomal RNA: implications for evolution and autoregulation.

Ribosomal protein L1 from the prokaryote Escherichia coli has been shown to form a specific complex with 26S ribosomal RNA from the eukaryote Dictyostelium discoideum. The segment of Dictyostelium rRNA protected from ribonuclease digestion by L1 and the corresponding region in Dictyostelium rDNA were investigated by nucleotide sequence analysis, and an analogous section in rDNA from Xenopus laevis was identified. When the L1-specific segments from eukaryotic rRNA were compared with those from prokaryotic rRNA, striking similarities in both primary and secondary structure were apparent. These conserved features suggest a common structural basis for protein recognition and indicate that such regions became fixed at a very early stage in rRNA evolution. In addition, certain structural elements of the L1 binding sites in rRNA are also found in the initial segment of the polycistronic L11-L1 mRNA, providing support for the hypothesis that L1 participates in the regulation of ribosomal protein synthesis by specific interaction with its own mRNA.     

Native topology determines force-induced unfolding pathways in globular proteins

Single-molecule manipulation techniques reveal that stretching unravels individually folded domains in the muscle protein titin and the extracellular matrix protein tenascin. These elastic proteins contain tandem repeats of folded domains with beta-sandwich architecture. Herein, we propose by stretching two model sequences (S1 and S2) with four-stranded beta-barrel topology that unfolding forces and pathways in folded domains can be predicted by using only the structure of the native state. Thermal refolding of S1 and S2 in the absence of force proceeds in an all-or-none fashion. In contrast, phase diagrams in the force-temperature (f,T) plane and steered Langevin dynamics studies of these sequences, which differ in the native registry of the strands, show that S1 unfolds in an allor-none fashion, whereas unfolding of S2 occurs via an obligatory intermediate. Force-induced unfolding is determined by the native topology. After proving that the simulation results for S1 and S2 can be calculated by using native topology alone, we predict the order of unfolding events in Ig domain (Ig27) and two fibronectin III type domains ((9)FnIII and (10)FnIII). The calculated unfolding pathways for these proteins, the location of the transition states, and the pulling speed dependence of the unfolding forces reflect the differences in the way the strands are arranged in the native states. We also predict the mechanisms of force-induced unfolding of the coiled-coil spectrin (a three-helix bundle protein) for all 20 structures deposited in the Protein Data Bank. Our approach suggests a natural way to measure the phase diagram in the (f,C) plane, where C is the concentration of denaturants.     

Directed hydroxyl radical probing of 16S rRNA using Fe(II) tethered to ribosomal protein S4.

Localized hydroxyl radical probing has been used to explore the rRNA neighborhood around a unique position in the structure of the Escherichia coli 30S ribosomal subunit. Fe(II) was attached to ribosomal protein S4 at Cys-31 via the reagent 1-(p-bromoacetamidobenzyl)-EDTA. [Fe-Cys31]S4 was then complexed with 16S rRNA or incorporated into active 30S ribosomal subunits by in vitro reconstitution with 16S rRNA and a mixture of the remaining 30S subunit proteins. Hydroxyl radicals generated from the tethered Fe resulted in cleavage of the 16S rRNA chain in two localized regions of its 5' domain. One region spans positions 419-432 and is close to the multihelix junction previously placed at the RNA binding site of S4 by chemical and enzymatic protection (footprinting) and crosslinking studies. A second site of directed cleavage includes nucleotides 297-303, which overlap a site that is protected from chemical modification by protein S16, a near neighbor of S4 in the ribosome. These results provide useful information about the three-dimensional organization of 16S rRNA and indicate that these two regions of its 5' domain are in close spatial proximity to Cys-31 of protein S4.     

Dynamic Monte Carlo study of the folding of a six-stranded Greek key globular protein.

To help elucidate the general rules of equilibrium globular protein folding, dynamic Monte Carlo simulations of a model beta-barrel globular protein having the six-stranded Greek key motif characteristic of real globular proteins were undertaken. The model protein possesses a typical beta-barrel amino acid sequence; however, all residues of a given type (e.g. hydrophobic residues) are identical. Even in the absence of site-specific interactions, starting from a high-temperature denatured state, these models undergo an all-or-none transition to a structurally unique six-stranded beta-barrel. These simulations suggest that the general rules of globular protein folding are rather robust in that the overall tertiary structure is determined by the general pattern of hydrophobic, hydrophilic, and turn-type residues, with site-specific interactions mainly involved in structural fine tuning of a given topology. Finally, these studies suggest that loops may play an important role in producing a unique native state. Depending on the stability of the native conformation of the long loop in the Greek key, the conformational transition can be described by a two-state, three-state, or even larger number of multiple equilibrium states model.     

Primary structure of major outer membrane protein II (ompA protein) of Escherichia coli K-12.

The amino acid sequence of major outer membrane protein II (ompA protein) from Escherichia coli K-12 has been determined. The transmembrane polypeptide consists of 325 residues, resulting in a molecular weight of 35,159. The transmembrane part of the protein is located between residues 1 and 177. In this part of the protein a predominantly lipophilic 27-residue segment exists that perhaps spans the membrane in a mostly alpha-helical conformation, or a 19-residue stretch of this segment might traverse the membrane linearly. Inside the outer membrane a sequence -Ala-Pro-Ala-Pro-Ala-Pro-Ala-Pro- exists that, analogous to the -Cys-Pro-Pro-Cys-Pro- sequence in the hinge region of immunoglobulin, could assume the conformation of a polyproline helix. Computer analysis did not reveal a clear overall pattern of internal homology in the protein; besides the -Ala-Pro- repeat, only one local area (two adjacent dodecapeptide segments) shows some repetitiveness. The same analysis did not produce evidence for internal homology in the previously determined sequence of outer membrane protein I (porin) nor was any marked resemblance detected between transmembrane proteins I and II.     

Solution structure of a de novo protein from a designed combinatorial library

Combinatorial libraries of de novo amino acid sequences can provide a rich source of diversity for the discovery of novel proteins. Randomly generated sequences, however, rarely fold into well ordered protein-like structures. To enhance the quality of a library, diversity must be focused into those regions of sequence space most likely to yield well folded structures. We have constructed focused libraries of de novo sequences by designing the binary pattern of polar and nonpolar amino acids to favor structures that contain abundant secondary structure, while simultaneously burying hydrophobic side chains in the protein interior and exposing hydrophilic side chains to solvent. Because binary patterning specifies only the polar/nonpolar periodicity, but not the identities of the side chains, detailed structural features, including packing interactions, cannot be designed a priori. Can binary patterned libraries nonetheless encode well folded proteins? An unambiguous answer to this question requires determination of a 3D structure. We used NMR spectroscopy to determine the structure of S-824, a novel protein from a recently constructed library of 102-residue sequences. This library is "na?ve" in that it has not been subjected to high-throughput screens or directed evolution. The experimentally determined structure of S-824 is a four-helix bundle, as specified by the design. As dictated by the binary-code strategy, nonpolar side chains are buried in the protein interior, and polar side chains are exposed to solvent. The polypeptide backbone and buried side chains are well ordered, demonstrating that S-824 is not a molten globule and forms a unique structure. These results show that amino acid sequences that have neither been selected by evolution, nor designed by computer, nor isolated by high-throughput screening, can form native-like structures. These findings validate the binary-code strategy as an effective method for producing vast collections of well folded de novo proteins.     

Hydrophobic basis of packing in globular proteins.

The self-assembly of globular proteins is often portrayed as a nucleation process in which the hydrogen bonding in segments of secondary structure is the precondition for further folding. We show here that this concept is unlikely because both the buried interior regions and the peptide chain turns of the folded protein (i.e., inside and outside) are predicted solely by the hydrophobicity of the residues, taken in sequential order along the chain. The helices and strands span the protein, and this observed secondary structure is seen to coincide with the regions predicted to be buried from hydrophobicity considerations alone. Our evidence suggests that linear chain regions rich in hydrophobic residues serve as small clusters that fold against each other, with concomitant or even later fixation of secondary structure. A helix or strand would arise in this folding process as one of a few energetically favorable alternatives for a given cluster, followed by a shift in the equilibrium between secondary structure conformers upon cluster association. the linera chain hydrophobicity alternates between locally maximal and minimal values, and these extrema partition the polypeptide chain into structural segments. This partitioning is seen in the x-ray structure as isodirectional segments bracketed between peptide chain-turns, with the segments expressed most often as helices and strands. the segment interactions define the geometry of the molecular interior and the chain-turns describe the predominant features of the molecular coastline. The segmentation of the molecule by linear chain hydrophobicity imposes a major geometric constraint upon possible folding events.     

Massive sequence perturbation of a small protein

Most protein topologies rarely occur in nature, thus limiting our ability to extract sequence information that could be used to predict structure, function, and evolutionary constraints on protein folds. In principle, the sequence diversity explored by a given protein topology could be expanded by introducing sequence perturbations and selecting variant proteins that fold correctly. However, our capacity to explore sequence space is intrinsically limited by the enormous number of sequences generated from the 20 amino acids and the limited number of variants likely to fold. Here we sought to test whether the sequence space for naturally existing proteins can be explored by simple, sequential degeneration of a complete set of short sequence segments of a model protein, without long-range covariation. Using the Raf ras binding domain as a model of a small protein capable of autonomous folding, we degenerated 72 of 76 positions of the primary structure for the 20 amino acids in segments of four to seven residues defined by secondary structure and selected the folded species for interaction with h-ras by using an in vivo survival-selection assay. The methodology presented allowed for rigorous statistical analysis and comparison of sequence diversity. The ensemble of sequence variants of Raf ras binding domain obtained have recaptured the diversity observed for the ubiquitin-roll topology. A signature sequence for this fold and the implication of this strategy to protein design and structure prediction are discussed.     

NMR structure determination of the Escherichia coli DnaJ molecular chaperone: secondary structure and backbone fold of the N-terminal region (residues 2-108) containing the highly conserved J domain.

DnaJ from Escherichia coli is a 376-amino acid protein that functions in conjunction with DnaK and GrpE as a chaperone machine. The N-terminal fragment of residues 2-108, DnaJ-(2-108), retains many of the activities of the full-length protein and contains a structural motif, the J domain of residues 2-72, which is highly conserved in a superfamily of proteins. In this paper, NMR spectroscopy was used to determine the secondary structure and the three-dimensional polypeptide backbone fold of DnaJ-(2-108). By using 13C/15N doubly labeled DnaJ-(2-108), nearly complete sequence-specific assignments were obtained for 1H, 15N, 13C alpha, and 13C beta, and about 40% of the peripheral aliphatic carbon resonances were also assigned. Four alpha-helices in polypeptide segments of residues 6-11, 18-31, 41-55, and 61-68 in the J domain were identified by sequential and medium-range nuclear Overhauser effects. For the J domain, the three-dimensional structure was calculated with the program DIANA from an input of 536 nuclear Overhauser effect upper-distance constraints and 52 spin-spin coupling constants. The polypeptide backbone fold is characterized by the formation of an antiparallel bundle of two long helices, residues 18-31 and 41-55, which is stabilized by a hydrophobic core of side chains that are highly conserved in homologous J domain sequences. The Gly/Phe-rich region from residues 77 to 108 is flexibly disordered in solution.     

Structure and folding of a designed knotted protein

A very small number of natural proteins have folded configurations in which the polypeptide backbone is knotted. Relatively little is known about the folding energy landscapes of such proteins, or how they have evolved. We explore those questions here by designing a unique knotted protein structure. Biophysical characterization and X-ray crystal structure determination show that the designed protein folds to the intended configuration, tying itself in a knot in the process, and that it folds reversibly. The protein folds to its native, knotted configuration approximately 20 times more slowly than a control protein, which was designed to have a similar tertiary structure but to be unknotted. Preliminary kinetic experiments suggest a complicated folding mechanism, providing opportunities for further characterization. The findings illustrate a situation where a protein is able to successfully traverse a complex folding energy landscape, though the amino acid sequence of the protein has not been subjected to evolutionary pressure for that ability. The success of the design strategy--connecting two monomers of an intertwined homodimer into a single protein chain--supports a model for evolution of knotted structures via gene duplication.