The role of asparagine-32 in forming the receptor-binding epitope of human epidermal growth factor

The highly conserved asparagine residue at position 32 (Asn32)in the 'hinge' region of epidermal growth factor (EGF) separatesthe N- and C-terminal structural motifs of the EGF moleculeand is therefore an appropriate target for structure-functionstudies. Analogs of human EGF (hEGF) were generated in whichAsn32 was substituted with aspartate, glycine, isoleucine, lysine,proUne and tryptophan. The relative affinity of the EGF receptorfor mutant hEGF analogs was determined by radioreceptor competitionassay. A wide range of receptor affinities was observed dependingon the amino acid substitution. N32K and N32W hEGF analogs hadrelatively high receptor affinity, while the N32G and N32D analogsshowed decreased affinity, 35% and 25% respectively, relativeto wild type hEGF. However, no binding of the N32P analog wasdetected by radioreceptor competition assay. The N32P mutantdisplayed an NMR spectrum significantly different from thatof native wild type hEGF, indicating gross structural perturbation.In contrast, the N32K and N32D analogs exhibited spectra similarto that of native wild type hEGF. Genetically combining theN32D hEGF with an hEGF species having either the mutation L26Ghi the N-terminal region or L47A in the C-terminal region, generateddouble-mutant hEGF species whkh had relative affinities essentiallyequal to the product of the relative affinities of the parenthEGF mutants, indicating functionally independent changes inUgand-receptor interaction. These studies indicate the requirementfor H-bond donor functionality in the side chain of residuenumber 32 in forming a fully competent receptor-binding epitope.     

Enhanced stability of recombinant keratinocyte growth factor by mutagenesis

Native sequence keratinocyte growth factor (KGF) is fairly unstable, as manifested by the loss of the monomeric native protein accompanied by the accumulation of aggregated species during storage at moderate temperatures. Several different types of analogs were generated and the storage stability of the protein assessed. In the first type of analog one or more of the five cysteinyl residues in KGF were replaced; in the second class the N-terminal residues that included the first disulfide bond were deleted. Both of these types of analogs involved removal of the disulfide bond between cysteines 1 and 15. The third group involved mutating one of the basic amino acids located in a cluster of positive charges (involved in heparin binding) around Arg144 to a neutral or acidic amino acyl residue. Among the cysteine replacement analogs, the double mutation of Cys1 and 15 to Ser resulted in significantly increased stability without compromising the mitogenic activity, while Cys to Ser mutations at other positions were either destabilizing or had no effect. Deletion of the 15, 23 or 27 N-terminal amino acyl residues also increased the stability of the protein. The activity of the analogs was not affected by the deletion of 15 or 23 amino acids, but it was significantly decreased upon removal of the 27 N-terminal amino acyl residues. Much greater stability was achieved by mutation of the basic amino acids, especially Arg144, to Glu or Gln, but this increase in stability was accompanied by large decrease in activity. The analog with the 23 N-terminal amino acyl residues deleted represents one of the best compromises between increased stability and retention of activity.     

Stability constraints and protein evolution: the role of chain length, composition and disulfide bonds

Stability of the native state is an essential requirement in protein evolution and design. Here we investigated the interplay between chain length and stability constraints using a simple model of protein folding and a statistical study of the Protein Data Bank. We distinguish two types of stability of the native state: with respect to the unfolded state (unfolding stability) and with respect to misfolded configurations (misfolding stability). Several contributions to stability are evaluated and their correlations are disentangled through principal components analysis, with the following main results. (1) We show that longer proteins can fulfil more easily the requirements of unfolding and misfolding stability, because they have a higher number of native interactions per residue. Consistently, in longer proteins native interactions are weaker and they are less optimized with respect to non-native interactions. (2) Stability against misfolding is negatively correlated with the strength of native interactions, which is related to hydrophobicity. Hence there is a trade-off between unfolding and misfolding stability. This trade-off is influenced by protein length: less hydrophobic sequences are observed in very long proteins. (3) The number of disulfide bonds is positively correlated with the deficit of free energy stabilizing the native state. Chain length and the number of disulfide bonds per residue are negatively correlated in proteins with short chains and uncorrelated in proteins with long chains. (4) The number of salt bridges per residue and per native contact increases with chain length. We interpret these observations as an indication that the constraints imposed by unfolding stability are less demanding in long proteins and they are further reduced by the competing requirement for stability against misfolding. In particular, disulfide bonds appear to be positively selected in short proteins, whereas they evolve in an effectively neutral way in long proteins.     

Use of periplasmic target protein capture for phage display engineering of tight-binding protein-protein interactions

Phage display is a powerful tool to study and engineer protein and peptide interactions. It is not without its limitations, however, such as the requirement for target protein purification and immobilization in a correctly folded state. A protein capture method is described here that allows enrichment of tight-binding protein variants in vivo thereby eliminating the need for target protein purification and immobilization. The linkage of genotype to phenotype is achieved by placing both receptor and ligand encoding genes on the same plasmid. This allows the isolation of the tight-binding ligand-receptor pair complexes after their association in the bacterial periplasm. The interaction between the TEM-1-β-lactamase fused to the gene 3 coat protein displayed on the surface of M13 bacteriophage and the β-lactamse inhibitory protein (BLIP) expressed in soluble form with a signal sequence to export it to the periplasm was used as a model system to test the method. The system was experimentally validated using a previously characterized collection of BLIP alanine mutants with a range of binding affinities for TEM-1 β-lactamase and by isolating tight-binding variants from a library of mutants randomized at residue position Tyr50 in BLIP which contacts β-lactamase.     

The N-terminal domain of VirG of Agrobacterium tumefadens: modelling and analysis of mutant phenotypes

Fourteen mutants in the N-terminal domain of virulence factorG (VirG) were obtained by random mutagenesis. Two mutants showedan altered phenotype, all others were nonfunctional.All mutantscan still be phosphorylated and bind to DNA. A 3-D model wasbuilt based on the coordinates of chemotaxis protein Y (CheY).Many of the observed phenotypic changes of VirG are explainedqualitatively. Combination of model building and biochemicalinformation leads to the conclusion that the active sites ofVirG and CheY must partly use different residues to performthe same phosphorylation and dephosphorylation reactions.     

Redesign of the Phosphate Binding Site of L‐Rhamnulose‐ 1‐Phosphate Aldolase towards a Dihydroxyacetone Dependent Aldolase

The aldol addition of unphosphorylated dihydroxyacetone (DHA) to aldehydes catalyzed by L ‐rhamnulose‐1‐phosphate aldolase (RhuA), a dihydroxyacetone phosphate‐dependent aldolase, is reported. Moreover, a single point mutation in the phosphate binding site of the RhuA wild type, that is, substitution of aspartate for asparagine at position N29, increased by 3‐fold the of aldol addition reactions of DHA to other aldehyde acceptors rather than the natural L ‐lactaldehyde. The RhuA N29D mutant modified the optimum enzyme design for the natural substrate and changed its catalytic properties making the aldolase more versatile to other aldol additions of DHA.     

Structure-function analysis of a conserved aromatic cluster in the N- terminal domain of human epidermal growth factor

The importance of a cluster of conserved aromatic residues of human epidermal growth factor (hEGF) to the receptor binding epitope is suggested by the interaction of His10 and Tyr13 of the A-loop with Tyr22 and Tyr29 of the N-terminal beta-sheet to form a hydrophobic surface on the hEGF protein. Indeed, Tyr13 has previously been shown to contribute a hydrophobic determinant to receptor binding. The roles of His10, Tyr22 and Tyr29 were investigated by structure-function analysis of hEGF mutant analogues containing individual replacements of each residue. Substitutions with aromatic residues or a leucine at position 10 retained receptor affinities and agonist activities similar to wild- type indicating that an aromatic residue is not essential. Variants with polar, charged or aliphatic substitutions altered in size and/or hydrophobicity exhibited reduced binding and agonist activities. 1- Dimensional 1H NMR spectra of high, moderate and low-affinity analogues at position 10 suggested only minor alterations in hEGF native structure. In contrast, a variety of replacements were tolerated at position 22 or 29 indicating that neither aromaticity nor hydrophobicity of Tyr22 and Tyr29 is required for receptor binding. CD spectra of mutant analogues at position 22 or 29 indicated a correlation between loss of receptor affinity and alterations in hEGF structure. The results indicate that similar to Tyr13, His10 of hEGF contributes hydrophobicity to the receptor binding epitope, whereas Tyr22 and Tyr29 do not appear to be directly involved in receptor interactions. The latter conclusion, together with previous studies, suggests that hydrophobic residues on only one face of the N-terminal beta-sheet of hEGF are important in receptor recognition.      

Simple intrasequence difference (SID) analysis: an original method to highlight and rank sub-structural interfaces in protein folds. Application to the folds of bovine pancreatic trypsin inhibitor, phospholipase A2, chymotrypsin and carboxypeptidase A

We present Simple Intrasequence Difference (SID) analysis, anovel bioinformatic technique designed to help comprehend theproperties of protein fold topologies. The analysis grades numericallyevery residue position in a given protein 3D structure accordingto the topological situation of the position in the folded chain.This results in an expression of the potential contributionof each residue position and its vicinity towards the integrityof the molecular conformation. Contiguous highly graded residuesdelineate the sub-structural interfaces that arise from thepresence within the molecular fold of discrete domains and sub-domains.This comprehensive rendering of the internal arrangement ofchain interfacing helps predict the potential for site-specificinductions (e.g. via mutations or ligand binding) of conformationalchange in the fold. Whereas SID analysis of single folds canconvey an idea of the basic potential for topological adjustmentin the protein family, comparative SID analysis of related foldsfocuses attention on those areas of the family fold where evolutionarychanges, activation events and ligand binding have had the mosttopological impact. For demonstration, SID analysis is appliedto the folds of pancreatic trypsin inhibitor (Kunitz), phospholipaseA₂, chymotrypsin and carboxypeptidase A. We find that many ofthe potentially vulnerable sub-structural interfaces tend tobe protected in the fold interior, in many cases stabilisedby disulfide bridges spanning the interface. However, the mostprominent interfaces tend to be externally accessible, withoutremedial stabilisation by disulfide bridges. These latter interfacesare associated so closely with the known functional sites thatalterations to the interfacial juxtapositions should influencerecognition and catalytic behaviour directly. This shows howside chain mutations, chemical modifications and binding eventsremote from the sites can nevertheless adjust, via interfacialrealignment, the conformations and emergent properties of thesites. The close association also provides clear opportunitiesfor interfacial rearrangements to follow intermolecular recognitionevents in the sites, facilitating translation of the bindinginto adjustment of the molecular conformation in areas distantfrom the sites. As a direct consequence of the topological arrangements,a large proportion of the molecular structure has the capacityto shape the character of the functional sites and, conversely,binding at these sites has the potential to radiate influenceto the rest of the molecule. For the enzymes considered, theevidence is consistent with the possibility that primary andsecondary binding by the substrate enhances catalytic efficiencyby imposing conformational change upon the catalytic centrevia adjustments to the fold. This influence may be expressedas favourable adjustment of the catalytic geometry, transitionstate ensemble, energy propagation pathway, or as a physicalstrain exerted on the substrate bond to be cleaved. The scaleof the adjustments, and their importance to the mechanisms,may have been seriously underestimated.     

Disulfide bonds, their stereospecific environment and conservation in protein structures

We studied the specificity of the non-bonded interaction in the environment of 572 disulfide bonds in 247 polypeptide chains selected from the Protein Data Bank. The preferred geometry of interaction of peptide oxygen atoms is along the back of the two covalent bonds at the sulfur atom of half cystine. With aromatic residues the geometries that direct one of the sulfur lone pair of electrons into the aromatic pi-system are avoided; an orientation in which the sulfide plane is normal or inclined to the aromatic plane and on top of its edge is normally preferred. The importance of the S...aromatic interaction is manifested in the high degree of its conservation across members in homologous protein families. These interactions, while providing extra overall stability to the native fold and reducing the accessibility of the disulfide bond and thereby preventing exchange reactions, also set the orientation of the conserved aromatic rings for further interactions and binding to another molecule. The conformational features and the mode of interactions of disulfide bridges should be useful for molecular design and protein engineering experiments.     

Assessing the Effect of Loop Mutations in the Folding Space of β2-Microglobulin with Molecular Dynamics Simulations

We use molecular dynamics simulations of a full atomistic Gō model to explore the impact of selected DE-loop mutations (D59P and W60C) on the folding space of protein human β₂-microglobulin (Hβ₂m), the causing agent of dialysis-related amyloidosis, a conformational disorder characterized by the deposition of insoluble amyloid fibrils in the osteoarticular system. Our simulations replicate the effect of mutations on the thermal stability that is observed in experiments in vitro. Furthermore, they predict the population of a partially folded state, with 60% of native internal free energy, which is akin to a molten globule. In the intermediate state, the solvent accessible surface area increases up to 40 times relative to the native state in 38% of the hydrophobic core residues, indicating that the identified species has aggregation potential. The intermediate state preserves the disulfide bond established between residue Cys25 and residue Cys80, which helps maintain the integrity of the core region, and is characterized by having two unstructured termini. The movements of the termini dominate the essential modes of the intermediate state, and exhibit the largest displacements in the D59P mutant, which is the most aggregation prone variant. PROPKA predictions of pK_a suggest that the population of the intermediate state may be enhanced at acidic pH explaining the larger amyloidogenic potential observed in vitro at low pH for the WT protein and mutant forms.     

Alteration in folding efficiency and conformation of recombinant human tumor necrosis factor-alpha by replacing cysteines 69 and 101 with aspartic acid 69 and arginine 101

An analog of human tumor necrosis factor- (TNF-) was createdinvolving the replacement of Cys69 with Asp and Cys101 withArg. The solution structure and behavior of this analog werecompared with the native protein. The analog exhibited a greatlydecreased folding efficiency following dilution from urea, butessentially identical circular dichrok spectra in both the foldedand unfolded states. The Stokes radius of the native and analogTNF- in the folded state were identical, with the analog exhibitinga slight broadening of the eluting peak. The fluorescence emissionspectrum of the native protein exhibits a plateau from 320 to328 nm, while the spectrum of the analog consisted of a singlepeak with a maximum at 335 nm. The analog also had a 1.4-foldincrease in the fluorescence intensity. Limited proteolysisof the analog resulted in only one of the two peptides seenfollowing digestion of the native protein, and this productwas less stable than the equivalent native protein fragment.The analog exhibited a 10-fold lower cytolytic activity thanthe native protein. These results demonstrated that the disulfidebond is not necessary for folding and activity, but are consistentwith the analog having a looser, more flexible structure insolution than the native TNF-.     

Chemical Synthesis and Structure of the Prokineticin Bv8

Bv8, a 77‐residue protein isolated from frogs, is the prototypic member of the prokineticin family of cytokines. Prokineticins (PKs) have only recently been identified in vertebrates (including humans), and they are believed to be involved in a number of key physiological processes, such as angiogenesis, neurogenesis, nociception, and tissue development. We used a combination of Boc solid‐phase peptide synthesis, native chemical ligation, and in vitro protein folding to establish robust chemical access to this molecule. Synthetic Bv8 was obtained in good yield and exhibited full activity in a human neuroblastoma cell line and rat dorsal root ganglion (DRG) neurons. The 3D structure of the synthetic protein was determined by using NMR spectroscopy and it was found to be homologous with that of mamba intestinal toxin 1, which is the only other known prokineticin structure. Analysis of a truncated mutant lacking five residues at the N terminus that are critical for receptor binding and activation showed no perturbation to the core protein structure. Together with the functional data, this suggests that receptor binding is likely to be a highly cooperative process possibly involving major allosterically driven structural rearrangements. The facile and efficient synthesis presented here will enable preparation of unique chemical analogues of prokineticins, which should be powerful tools for modulating the structure and function of prokineticins and their receptors, and studying the many physiological processes that have been linked to them.     

3-D structure of the D153G mutant of Escherichia coli alkaline phosphatase: an enzyme with weaker magnesium binding and increased catalytic activity

The substitution of aspartate at position 153 in Escherichiacoli alkaline phosphatase by glycine results in a mutant enzymewith 5-fold higher catalytic activity (k_cat but no change inKm at pH 8.0 in 50 mM Tris-HCl. The increased k_cat is achievedby a faster release of the phosphate product as a result ofthe lower phosphate affinity. The mutation also affects Mg²⁺binding, resulting in an enzyme with lower metal affinity. The3-D X-ray structure of the D153G mutant has been refined at2.5 Å to a crystallographic Rfactor of 16.2%. An analysisof this structure has revealed that the decreased phosphateaffinity is caused by an apparent increase in flexibility ofthe guanidinium side chain of Argl66 involved in phosphate binding.The mutation of Aspl53 to Gly also affects the position of thewater ligands of Mg²⁺, and the loop Glnl52–Thrl55 is shiftedby 0.3 Å away from the active site. The weaker Mg²⁺ bindingof the mutant compared with the wild type is caused by an alteredcoordination sphere in the proximity of the Mg²⁺ ion, and alsoby the loss of an electrostatic interaction (Mg²⁺.COO^-Aspl53)in the mutant Its ligands W454 and W455 and hydroxyl of Thrl55,involved in the octahedral coordination of the Mg²⁺ ion, arefurther apart in the mutant compared with the wild-type     

A grafting approach to obtain site-specific metal-binding properties of EF-hand proteins

The EF-hand calcium-binding loop III from calmodulin was insertedwith glycine linkers into the scaffold protein CD2.D1 at threelocations to study site-specific calcium binding propertiesof EF-hand motifs. After insertion, the host protein retainsits native structure and forms a 1:1 metal–protein complexfor calcium and its analog, lanthanum. Tyrosine-sensitized Tb³⁺energy transfer exhibits metal binding and La³⁺ and Ca²⁺ competefor the metal binding site. The grafted EF-loop III in differentenvironments has similar La³⁺ binding affinities, suggestingthat it is largely solvated and functions independently fromthe host protein. Received May 25, 2002; revised January 23, 2003; accepted April 25, 2003.     

Molecular dynamics study of Ca2+ binding loop variants of parvalbumin with modifications at the `gateway' position

The helix–loop–helix (i.e. EF-hand) Ca²⁺ ion bindingmotif is characteristic of a large family of high-affinity Ca²⁺ion binding proteins, including the parvalbumins and calmodulins.In this paper we describe a set of molecular dynamics computationson the major parvalbumin from the silver hake (SHPV-B). In allvariants examined, both whole protein and fragments thereof,the ninth loop residue in the Ca²⁺ binding coordination sitein the CD helix–loop–helix region (the so-called`gateway' residue) has been mutated. The three gateway mutationsexamined are arginine, which has never been found at the gatewayposition of any EF-hand protein, cysteine, which is the residueobserved least in natural EF-hand sites, and serine, which isthe most common (by far) non-acidic residue substitution atthis position in EF-hand proteins in general, but never in parvalbumins.Results of the molecular dynamics simulations indicate thatall three modifications are disruptive to the integrity of themutated Ca²⁺ binding site in the whole parvalbumin protein.In contrast, only the arginine and cysteine mutations are disruptiveto the integrity of the mutated Ca²⁺ binding site in the CDfragment of the parvalbumin protein. Surprisingly, the serinevariant of the CD helix–loop–helix fragment exhibitedremarkable stability during the entire molecular dynamics simulation,with retention of the Ca²⁺ binding site. These results indicatethat there are no inherent problems (for Ca²⁺ ion binding) associatedwith the sequence of the CD helix–loop–helix fragmentthat precludes the incorporation of serine at the gateway position.Since the CD site is totally disrupted in the whole proteinserine variant, this indicates that the Ca²⁺ ion binding deficienciesare most likely related to the unique interaction that existsbetween the paired EF-hands in the whole protein. Our theoreticalresults correlate well with previous studies on engineered EF-handproteins and with all of our experimental evidence on the silverhake parvalbumin.     

The dimerization of Pseudomonas putida cytochrome P450cam: practical consequences and engineering of a monomeric enzyme

Cytochrome P450cam dimerizes via the formation of an intermolecular disulfide bond, complicating the storage and handling of the enzyme, particularly at higher concentrations. The dimeric enzyme is 14% less active than the monomer and forms at a slow but significant rate even at 4 degrees C [k = 1.09 x 10(-3) mM(-1) h(-1)]. To eliminate any ambiguity introduced by dimer formation and to simplify handling and storage of the enzyme, site-directed mutagenesis was used to identify C334 as the single cysteine residue responsible for the formation of the disulfide linkage and to engineer a monomeric enzyme by substituting an alanine in its place. The C334A mutant is identical with the wild-type P450cam monomer in terms of optical spectra, camphor binding and turnover activity, but shows no evidence of dimerization and aggregation even at millimolar concentrations. Preliminary 1H NMR investigations also indicate a significant improvement in the quality of spectra obtained with this mutant. (C334A)P450cam is therefore proposed as an alternative to the wild-type enzyme-a base mutant otherwise identical with the wild-type but with improved handling characteristics.      

Mutation of a highly conserved aspartate residue in subdomain IX abolishes Fer protein-tyrosine kinase activity

Before the structure of cAMP-dependent protein kinase had beensolved, sequence alignments had already suggested that severalhighly conserved peptide motifs described as kinase subdomainsI through XI might play some functional role in catalysis. Crystalstructures of several members of the protein kinase superfamilyhave suggested that the nearly invariant aspartate residue withinsubdomain IX contributes to the conformational stability ofthe catalytic loop by forming hydrogen bonds with backbone amideswithin subdomain VI. However, substitution of this aspartatewith alanine or threonine in some protein kinases have indicatedthat these interactions are not essential for activity. In contrast,we show here that conversion of this aspartate to arginine abolishedthe catalytic activity of the Fer protein-tyrosine kinase whenexpressed either in mammalian cells or in bacteria. Structuralmodeling predicted that the catalytic loop of the FerD743R mutantwas disrupted by van der Waal's repulsion between the side chainsof the substituted arginine residue in subdomain IX and histidine-683in subdomain VI. The FerD743R mutant model predicted a shiftin the peptide backbone of the catalytic loop, and an outwardrotation of histidine-683 and arginine-684 side chains. However,the position and orientation of the presumptive catalytic base,aspartate-685, was not substantially changed. The proposed modelexplains how substitutions of some, but not all residues couldbe tolerated at this nearly invariant aspartate in kinase subdomainIX.     

Redesign of the coenzyme specificity in L-Lactate dehydrogenase from Bacillus stearothermophilus using site-directed mutagenesis and media engineering

L-Lactate dehydrogenase (LDH) from Bacillus stearothermophilusis a redox enzyme which has a strong preference for NADH overNADPH as coenzyme. To exclude NADPH from the coenzyme-bindingpocket, LDH contains a conserved aspartate residue at position52. However, this residue is probably not solely responsiblefor the NADH specificity. In this report we examine the possibilitiesof altering the coenzyme specificity of LDH by introducing arange of different point mutations in the coenzyme-binding domain.Furthermore, after choosing the mutant with the highest selectivityfor NADPH, we also investigated the possibility of further alteringthe coenzyme specificity by adding an organic solvent to thereaction mixture. The LDH mutant, I51K:D52S, exhibited a 56-foldincreased specificity to NADPH over the wild-type LDH in a reactionmixture containing 15% methanol. Furthermore, the NADPH turnovernumber of this mutant was increased almost fourfold as comparedwith wild-type LDH. To explain the altered coenzyme specificityexhibited by the D52SI51K double mutant, molecular dynamicssimulations were performed.