Complementation of buried lysine and surface polar residues in a designed heterodimeric coiled coil

The coiled coil is an attractive target for protein design. The helices of coiled coils are characterized by a heptad repeat of residues denoted a to g. Residues at positions a and d form the interhelical interface and are usually hydrophobic. An established strategy to confer structural uniqueness to two-stranded coiled coils is the use of buried polar Asn residues at position a, which imparts dimerization and conformational specificity at the expense of stability. Here we show that polar interactions involving buried position-a Lys residues that can interact favorably only with surface e' or g' Glu residues also impart structural uniqueness to a designed heterodimeric coiled coil with the nativelike properties of sigmoidal thermal and urea-induced unfolding transitions, slow hydrogen exchange and lack of ANS binding. The position-a Lys residues do not, however, confer a single preference for helix orientation, likely reflecting the ability of Lys at position a to from favorable interactions with g' or e' Glu residues in the parallel and antiparallel orientations, respectively. The Lys-Glu polar interaction is less destabilizing than the Asn-Asn a-->a' interaction, presumably reflecting a higher desolvation penalty associated with the completely buried polar position-a groups. Our results extend the range of approaches for two-stranded coiled-coil design and illustrate the role of complementing polar groups associated with buried and surface positions of proteins in protein folding and design.     

Mutational analysis and membrane-interactions of the beta-sheet-like N-terminal domain of the pediocin-like antimicrobial peptide sakacin P

To gain insight into how the N-terminal three-stranded beta-sheet-like domain in pediocin-like antimicrobial peptides positions itself on membranes, residues in the well-conserved (Y)YGNGV-motif in the domain were substituted and the effect of the substitutions on antimicrobial activity and binding of peptides to liposomes was determined. Peptide-liposome interactions were detected by measuring tryptophan-fluorescence upon exposing liposomes to peptides in which a tryptophan residue had been introduced in the N-terminal domain. The results revealed that the N-terminal domain associates readily with anionic liposomes, but not with neutral liposomes. The electrostatic interactions between peptides and liposomes facilitated the penetration of some of the peptide residues into the liposomes. Measuring the antimicrobial activity of the mutated peptides revealed that the Tyr2Leu and Tyr3Leu mutations resulted in about a 10-fold reduction in activity, whereas the Tyr2Trp, Tyr2Phe, Tyr3Trp and Tyr3Phe mutations were tolerated fairly well, especially the mutations in position 3. The Val7Ile mutation did not have a marked detrimental effect on the activity. The Gly6Ala mutation was highly detrimental, consistent with Gly6 being in one of the turns in the beta-sheet-like N-terminal domain, whereas the Gly4Ala mutation was tolerated fairly well. All mutations involving Asn5, including the conservative mutations Asn5Gln and Asn5Asp, were very deleterious. Thus, both the polar amide group on the side chain of Asn5 and its exact position in space were crucial for the peptides to be fully active. Taken together, the results are consistent with Val7 positioning itself in the hydrophobic core of target membranes, thus forcing most of the other residues in the N-terminal domain into the membrane interface region: Tyr3 and Asn5 in the lower half with their side chains pointing downward and approaching the hydrophobic core, Tyr2, Gly4 and His8 and 12 in the upper half, Lys1 near the middle of the interface region, and the side chain of Lys11 pointing out toward the membrane surface.     

Buried polar residues in coiled-coil interfaces

Coiled coils, estimated to constitute 3-5% of the encoded residues in most genomes, are characterized by a heptad repeat, (abcdefg)(n), where the buried a and d positions form the interface between multiple alpha-helices. Although generally hydrophobic, a substantial fraction ( approximately 20%) of these a- and d-position residues are polar or charged. We constructed variants of the well-characterized coiled coil GCN4-p1 with a single polar residue (Asn, Gln, Ser, or Thr) at either an a or a d position. The stability and oligomeric specificity of each variant were measured, and crystal structures of coiled-coil trimers with threonine or serine at either an a or a d position were determined. The structures show how single polar residues in the interface affect not only local packing, but also overall coiled-coil geometry as seen by changes in the Crick supercoil parameters and core cavity volumes.     

Heat capacity changes upon burial of polar and nonpolar groups in proteins

In this paper we address the question of whether the burial of polar and nonpolar groups in the protein locale is indeed accompanied by the heat capacity changes, DeltaC(p), that have an opposite sign, negative for nonpolar groups and positive for polar groups. To accomplish this, we introduced amino acid substitutions at four fully buried positions of the ubiquitin molecule (Val5, Val17, Leu67, and Gln41). We substituted Val at positions 5 and 17 and Leu at position 67 with a polar residue, Asn. As a control, Ala was introduced at the same three positions. We also replaced the buried polar Gln41 with Val and Leu, nonpolar residues that have similar size and shape as Gln. As a control, Asn was introduced at Gln41 as well. The effects of these amino acid substitutions on the stability, and in particular, on the heat capacity change upon unfolding were measured using differential scanning calorimetry. The effect of the amino acid substitutions on the structure was also evaluated by comparing the (1)H-(15)N HSQC spectra of the ubiquitin variants. It was found that the Ala substitutions did not have a considerable effect on the heat capacity change upon unfolding. However, the substitutions of aliphatic side chains (Val or Leu) with a polar residue (Asn) lead to a significant (> 30%) decrease in the heat capacity change upon unfolding. The decrease in heat capacity changes does not appear to be the result of significant structural perturbations as seen from the HSQC spectra of the variants. The substitution of a buried polar residue (Gln41) to a nonpolar residue (Leu or Val) leads to a significant (> 25%) increase in heat capacity change upon unfolding. These results indicate that indeed the heat capacity change of burial of polar and nonpolar groups has an opposite sign. However, the observed changes in DeltaC(p) are several times larger than those predicted, based on the changes in water accessible surface area upon substitution.     

An investigation of protein subunit and domain interfaces

Protein structures were collected from the Brookhaven Database of tertiary architectures that displayed oligomeric association (24 molecules) or whose polypeptide folding revealed domains (34 proteins). The subunit and domain interfaces for these proteins were respectively examined from the following aspects: percentage water-accessible surface area buried by the respective associations, surface compositions and physical characteristics of the residues involved in the subunit and domain contacts, secondary structural state of the interface amino acids, preferred polar and non-polar interactions, spatial distribution of polar and non-polar residues on the interface surface, same residue interactions in the oligomeric contacts, and overall cross-section and shape of the contact surfaces. A general, consistent picture emerged for both the domain and subunit interfaces.     

The core protein of alphaviruses. 1. Purification of peptides and complete amino-acid sequence of Semliki Forest virus core protein

The primary structure of the core protein of Semliki Forest virus has been established by protein chemical characterization of 102 peptides, generated by digestion with trypsin, pepsin, thermolysin, and by partial acid cleavage of the protein. Besides a difference in one position, the sequence as established by these experiments is in agreement with the sequence predicted from the nucleotide sequence of the mRNA [Garoff et al. (1980) Proc. Natl Acad. Sci. USA, 77, 6376-6380]. The core protein has a blocked N terminus, consists of 267 amino acid residues, and has the following amino acid composition: Asp12, Asn9, Thr16, Ser10, Glu11, Gln15, Pro23, Gly20, Ala23, Val19, Met8, Ile11, Leu9, Tyr7, Phe6, His7, Lys37, Arg15, Trp5, Cys4, and an Mr of 29919. It contains 22.1% basic amino acids, mainly lysines, compared with a total of 8.6% acidic residues. The resulting surplus of positive charge is located in the N-terminal half of the protein (predominantly arginines at positions 12-21 and lysines at positions 66-114). Other amino acids are also unevenly distributed; proline and glutamine are accumulated in the N-terminal half of the sequence whereas histidine, glycine and the acidic residues are mainly present in the C-terminal part. This distribution suggests that the virus core protein consists of two or more structural domains.     

Comparison of the amino acid sequences of the variable regions of light chains derived from two homogeneous rabbit anti-pneumococcal antibodies

The amino acid sequence of the N-terminal 139 residues of the L (light) chain derived from a homogeneous rabbit antibody to type III pneumococci was determined. This L chain, designated BS-5, exhibits a greater degree of homology with the basic sequence of human kappa chains of subgroup I (72%) than with subgroups II and III. L-chain BS-5 differs from another L chain (BS-1), also derived from an antibody to type III pneumococci (Jaton, 1974), by eight amino acid residues, even though the chains are identical within the N-terminal 30 residues. Six of these eight substitutions are located within the three hypervariable sections of the variable half: Asn/Ser in position 31, Glu/Ala in position 55, Asx/Thr, Thr/Gly, Thr/Gly and Val/Tyr in positions 92, 94, 96 and 97 respectively. The two anti-pneumococcal L chains BS-1 and BS-5 are much more similar to each other than to an anti-azobenzoate L chain (Appella et al., 1973), from which they differ by 30 and 29 residues respectively. Of these interchanges 13-15 are confined to the three hypervariable sections, and 11 occur within the N-terminal 27 positions. The three chains have an identical sequence from residue 98 to residue 139, except for a possible inversion of two residues in positions 130-131 of the anti-azobenzoate chain.     

Structural and conformational changes concomitant with the E1-E2 transition in H(+)K(+)-ATPase: a comparative protein modeling study

Comparative modeling studies on conserved regions of the gastric H(+)K(+)-ATPase reveal that the E1-E2 conformational transition induces significant tertiary structural changes while conserving the secondary structure. The residues 516-530 of the cytoplasmic domain and TM10 within the transmembrane (TM) regions undergo maximum tertiary structural changes. The luminal regions exhibit comparatively lesser tertiary structural deviations. Residues 249-304 show maximum secondary structural deviation in the conformational transition. The Cys-815 and Cys-323 residues involved in inhibitor binding are found to have smaller buried side chain areas in the E1 conformation compared to E2. Retention of activity correlates well with the buried side chain area when selected amino acid residues in TM6 are mutated using modeling techniques with bulkier amino acid residues. Conformational specificity for ion binding is corroborated with the fraction of side chains exposed to polar atoms of the residues E345, D826, V340, A341, V343, and E822.     

The contribution of buried polar groups to the conformational stability of the GCN4 coiled coil

The dimeric interface of the leucine zipper coiled coil from GCN4 has been used to probe the contributions of hydrophobic and hydrogen bonding interactions to protein stability. We have determined the energetics of placing Ile or Asn residues at four buried positions in a two-stranded coiled coil. As expected, Ile is favored over Asn at these buried positions, but not as much as predicted by considering only the hydrophobic effect. It appears that interstrand hydrogen bonds form between the side-chains of the buried Asn residues and these contribute to the conformational stability of the coiled-coil peptides. However, these contributions are highly dependent on the locations of the Asn pairs. The effect of an Ile to Asn mutation is greatest at the N terminus of the peptide and decreases almost twofold as we move the substitution from the N to C-terminal heptads.     

Position-dependence of stabilizing polar interactions of asparagine in transmembrane helical bundles

Recent studies with model peptides and statistical analyses of the crystal structures of membrane proteins have shown that buried polar interactions contribute significantly to the stabilization of the three-dimensional structures of membrane proteins. Here, we probe how the location of these polar groups along the transmembrane helices affect their free energies of interaction. Asn residues were placed singly and in pairs at three positions within a model transmembrane helix, which had previously been shown to support the formation of trimers in micelles. The model helix was designed to form a transmembrane coiled coil, with Val side chains at the "a" positions of the heptad repeat. Variants of this peptide were prepared in which an Asn residue was introduced at one or more of the "a" positions, and their free energies of association were determined by analytical ultracentrifugation. When placed near the middle of the transmembrane helix, the formation of trimers was stabilized by at least -2.0 kcal/mol per Asn side chain. When the Asn was placed at the interface between the hydrophobic and polar regions of the peptide, the substitution was neither stabilizing nor destabilizing (0.0 +/- 0.5 kcal/mol of monomer). Finally, it has previously been shown that a Val-for-Asn mutation in a water-soluble coiled coil destabilizes the structure by approximately 1.5 kcal/mol of monomer [Acharya, A., et al. (2002) Biochemistry 41, 14122-14131]. Thus, the headgroup region of a micelle appears to have a conformational impact intermediate between that of bulk water and the apolar region of micelle. A similarly large dependence on the location of the polar residues was found in a statistical survey of helical transmembrane proteins. The tendency of different types of residues to be buried in the interiors versus being exposed to lipids was analyzed. Asn and Gln show a very strong tendency to be buried when they are located near the middle of a transmembrane helix. However, when placed near the ends of transmembrane helices, they show little preference for the surface versus the interior of the protein. These data show that Asn side chains within the apolar region of the transmembrane helix provide a significantly larger driving force for association than Asn residues near the apolar/polar interface. Thus, although polar interactions are able to strongly stabilize the folding of membrane proteins, the energetics of association depend on their location within the hydrophobic region of a transmembrane helix.     

Functionally acceptable substitutions in two alpha-helical regions of lambda repressor

A method of targeted random mutagenesis has been used to investigate the informational content of 25 residue positions in two alpha-helical regions of the N-terminal domain of lambda repressor. Examination of the functionally allowed sequences indicates that there is a wide range in tolerance to amino acid substitution at these positions. At positions that are buried in the structure, there are severe limitations on the number and type of residues allowed. At most surface positions, many different residues and residue types are tolerated. However, at several surface positions there is a strong preference for hydrophilic amino acids, and at one surface position proline is absolutely conserved. The results reveal the high level of degeneracy in the information that specifies a particular protein fold.     

Correlation of structure with transforming activity of the P21 proteins with substitutions at positions 12 and 16

The structural effects of amino acid substitutions at positions 12 and 16 in the amino-terminal segment (Tyr 4-Ala 18) of the ras-oncogene-encoded P21 proteins have been investigated using conformational energy analysis. The P21 protein with Val at position 12 and Lys at position 16 is known to have high transforming ability, while the P21 protein with Val at position 12 and Asn at position 16 is known to have poor transforming ability, similar to that of the normal protein (with Gly at 12 and Lys at 16.) The current results demonstrate a significant conformational change at position 15 induced by the substitution of Asn for Lys at position 16, which could explain this alteration in transformation potential. These findings are consistent with previous results suggesting the existence of a normal and a malignancy-causing conformation for the P21 proteins and suggest that the critical transforming region may encompass residues 12–15.     

Crystal structure of the coiled-coil dimerization motif of geminin: structural and functional insights on DNA replication regulation

We have determined the crystal structure of the coiled-coil domain of human geminin, a DNA synthesis inhibitor in higher eukaryotes. We show that a peptide encompassing the five heptad repeats of the geminin leucine zipper (LZ) domain is a dimeric parallel coiled coil characterized by a unique pattern of internal polar residues and a negatively charged surface that may target the basic domain of interacting partners. We show that the LZ domain itself is not sufficient to inhibit DNA synthesis but upstream and downstream residues are required. Analysis of a functional form of geminin by density sedimentation indicates an oligomeric structure. X-ray solution scattering experiments performed on a non-functional form of geminin having upstream basic residues and the LZ domain show a tetramer structure. Altogether, these results give a consistent identification and mapping of geminin interacting regions onto structurally important domains. They also suggest that oligomerization properties of geminin may be implicated in its inhibitory activity of DNA synthesis.     

Dimeric Crystal Structure of Rabbit l-Gulonate 3-Dehydrogenase/λ-Crystallin: Insights into the Catalytic Mechanism

l-Gulonate 3-dehydrogenase (GDH) is a bifunctional dimeric protein that functions not only as an NAD⁺-dependent enzyme in the uronate cycle but also as a taxon-specific λ-crystallin in rabbit lens. Here we report the first crystal structure of GDH in both apo form and NADH-bound holo form. The GDH protomer consists of two structural domains: the N-terminal domain with a Rossmann fold and the C-terminal domain with a novel helical fold. In the N-terminal domain of the NADH-bound structure, we identified 11 coenzyme-binding residues and found 2 distinct side-chain conformers of Ser124, which is a putative coenzyme/substrate-binding residue. A structural comparison between apo form and holo form and a mutagenesis study with E97Q mutant suggest an induced-fit mechanism upon coenzyme binding; coenzyme binding induces a conformational change in the coenzyme-binding residues Glu97 and Ser124 to switch their activation state from resting to active, which is required for the subsequent substrate recruitment. Subunit dimerization is mediated by numerous intersubunit interactions, including 22 hydrogen bonds and 104 residue pairs of van der Waals interactions, of which those between two cognate C-terminal domains are predominant. From a structure/sequence comparison within GDH homologues, a much greater degree of interprotomer interactions (both polar and hydrophobic) in the rabbit GDH would contribute to its higher thermostability, which may be relevant to the other function of this enzyme as λ-crystallin, a constitutive structural protein in rabbit lens. The present crystal structures and amino acid mutagenesis studies assigned the role of active-site residues: catalytic base for His145 and substrate binding for Ser124, Cys125, Asn196, and Arg231. Notably, Arg231 participates in substrate binding from the other subunit of the GDH dimer, indicating the functional significance of the dimeric state. Proper orientation of the substrate-binding residues for catalysis is likely to be maintained by an interprotomer hydrogen-bonding network of residues Asn196, Gln199, and Arg231, suggesting a network-based substrate recognition of GDH.     

Buried polar interactions and conformational stability in the simian immunodeficiency virus (SIV) gp41 core

For human (HIV) and simian (SIV) immunodeficiency viruses, the gp41 envelope protein undergoes a receptor-activated conformational change from a labile native structure to an energetically more stable fusogenic conformation, which then mediates viral-cell membrane fusion. The core structure of fusion-active gp41 is a six-helix bundle in which three antiparallel carboxyl-terminal helices are packed against an amino-terminal trimeric coiled coil. Here we show that a recombinant model of the SIV gp41 core, designated N36(L6)C34, forms an alpha-helical trimer that exhibits a cooperative two-state folding-unfolding transition. We investigate the importance of buried polar interactions in determining the overall fold of the gp41 core. We have replaced each of four polar amino acids at the heptad a and d positions of the coiled coil in N36(L6)C34 with a representative hydrophobic amino acid, isoleucine. The Q565I, T582I, and T586I variants form six-helix bundle structures that are significantly more stable than that of the wild-type peptide, whereas the Q575I variant misfolds into an insoluble aggregate under physiological conditions. Thus, the buried polar residues within the amino-terminal heptad repeat are important determinants of the structural specificity and stability of the gp41 core. We suggest that these conserved buried polar interactions play a role in governing the conformational state of the gp41 molecule.     

Crystallization,crystal structure analysis and molecular model of the third domain of Japanese quail ovomucoid,a Kazal type inhibitor

The third domain of Japanese quail ovomucoid, a Kazal type inhibitor, has been crystallized and its crystal structure determined at 2.5 Å resolution using multiple isomorphous replacement techniques. The asymmetric unit contains four molecules. In the crystal the molecules are arranged in two slightly different octamers with approximate D4 symmetry. The molecules are held together mainly by interactions of the N-terminal residues, which form a novel secondary structural element, a β-channel.The molecule is globular with approximate dimensions 35 Å × 27 Å × 19 Å. The secondary structural elements are a double-stranded anti-parallel β-sheet of residues Pro22 to Gly32 and an α-helix from Asn33 to Ser44. The reactive site Lys18-Asp19 is located in an exposed loop. It is close to Asn33 at the N terminus of the helical segment. The polypeptide chain folding of ovomucoid bears some resemblance to other inhibitors in the existence of an anti-parallel double strand following the reactive site loop.     

Peptide inhibitors use two related mechanisms to alter the apparent calcium affinity of the sarcoplasmic reticulum calcium pump

The primary sequence of phospholamban (PLB) has provided a template for the rational design of peptide inhibitors of the sarcoplasmic reticulum calcium ATPase (SERCA). In the transmembrane domain of PLB, there are few polar residues and only one is essential (Asn (34)). Using synthetic peptides, we have previously investigated the role of Asn (34) in the context of simple hydrophobic transmembrane peptides. Herein we propose that the role of Asn in SERCA inhibition is position-sensitive and dependent upon the distribution of hydrophobic residues. To test this hypothesis, we synthesized a series of transmembrane peptides based on a 24 amino acid polyalanine sequence having either an alternating Leu-Ala sequence (Leu 12) or Leu residues at the native positions found in PLB (Leu 9). Asn-containing Leu 9 and Leu 12 peptides were synthesized with a single Asn residue located either one amino acid (N+/-1) or one turn of the helix (N+/-4) in either direction from its native position. Co-reconstitution of these peptides with SERCA into proteoliposomes revealed effects on the apparent calcium affinity and cooperativity of SERCA that correlated with the positions of the Asn and Leu residues. The most inhibitory peptides increased the cooperativity of SERCA as indicated by the Hill coefficients, suggesting that calcium-dependent reversibility is an inherent part of the inhibitory mechanism. Kinetic simulations combined with molecular modeling of the interaction between the peptides and SERCA reveal two related mechanisms of inhibition. Peptides that resemble PLB use the same inhibitory mechanism, whereas peptides that are more divergent from PLB alter an additional step in the calcium transport cycle.     

Mutations in the putative HR-C region of the measles virus F2 glycoprotein modulate syncytium formation

The fusion (F) glycoproteins of measles virus strains Edmonston (MV-Edm) and wtF (MV-wtF) confer distinct cytopathic effects and strengths of hemagglutinin (H) interaction on a recombinant MV-Edm virus. They differ in just two amino acids, V94 and V101 in F-Edm versus M94 and F101 in F-wtF, both of which lie in the relatively uncharacterized F(2) domain. By comparing the sequence of MV F with those of the parainfluenza virus SV5 and Newcastle disease virus (NDV) F proteins, the structures of which are known, we show that MV F(2) also possesses a potential heptad repeat (HR) C domain. In NDV, the N-terminal half of HR-C interacts with HR-A in F(1) while the C-terminal half is induced to kink outward by a central proline residue. We found that this proline is part of an LXP motif conserved in all three viruses. Folding and transport of MV F require this motif to be intact and also require covalent interaction of cysteine residues that probably support the potential HR-A-HR-C interaction. Amino acids 94 and 101, both located in "d" positions of the HR-C helical wheel, lie in the potentially outwardly kinked region. We demonstrate that their effect on MV fusogenicity and glycoprotein interaction is mediated solely by amino acid 94. Substitutions at position 94 with polar or charged amino acids are tolerated poorly or not at all, while changes to smaller and more hydrophilic amino acids are tolerated in both transiently expressed F protein and recombinant virus. MV F V94A and MV F V94G viruses induce extensive syncytium formation and are relatively, or almost completely, resistant to a known inhibitor of MV glycoprotein-induced fusion. We propose that the conformational changes in MV F protein required to expose the fusion peptide involve the C-terminal half of the HR-C helix, specifically amino acid 94.