Alpha-helical coiled coils and bundles: how to design an alpha-helical protein

De novo design and structural characterization of an alpha-helical hairpin peptide: a model system for the study of protein folding intermediates

The de novo design and structural characterization of an alpha-helical hairpin peptide (alpha-helix/turn/alpha-helix, alpha t alpha) are reported. The peptide is intended to provide a model system for the study of the interactions of secondary structural elements during protein folding. Both the diffusion-collision and framework models of protein folding envision that the earliest intermediates in protein folding are transient secondary structures or microdomains which interact and become mutually stabilizing. Design principles for the alpha t alpha peptide were drawn from the large body of work on the structure of peptides in solution. Computer modeling was not used in the design process. Study of alpha t alpha by circular dichroism and two-dimensional nuclear magnetic resonance indicates that the designed peptide is monomeric, helical, and stable in aqueous solution at room temperature. Analysis of two-dimensional nuclear magnetic resonance experiments indicates that the two helices and the turn form in the intended positions and that the helices associate in the designed orientation. Development of alpha t alpha represents an advance in protein design in that both the secondary structural elements and designed tertiary interactions have been realized and can be detected in solution by nuclear magnetic resonance. The resulting model system resembles a protein folding intermediate and will allow the study of interacting helices in a context that approximates an early stage in protein folding.

The local and global unfolding of coiled-coil tropomyosin

The thermal unfolding of a two-stranded alpha-helical coiled coil of tropomyosin was studied using circular dichroism and excimer fluorescence of N-(1-pyrenyl)iodoacetamide-labeled tropomyosin. Tropomyosin unfolds with two transitions, namely local and global unfolding at high salt (greater than 0.1 M NaCl) and pH 7.5. The local unfolding was masked by the global unfolding at low salt (less than 0.1 M NaCl), at high pH (greater than pH 9.0), and in the presence of methanol, where the global unfolding temperature was similar to or lower than the local unfolding temperature. The local and global unfolding are different in nature. A comparison of the helix thermal unfolding of N-(1-pyrenyl)iodoacetamide-tropomyosin with unlabeled tropomyosin showed that tropomyosin had an inherent less-stable region, when Cys190 was N-(1-pyrenyl)iodoacetamide-labeled, disulfide cross-linked, or reduced. Instead, the chemical state of Cys190, determined the stability of the local unfolding region, because a strain created by the disulfide cross-link or the pyrene/pyrene interaction decreases the stability of the local unfolding region. Thus, these data show that the excimer fluorescence of N-(1-pyrenyl)iodoacetamide-labeled tropomyosin is useful for studying the local and global unfolding of tropomyosin.

Identification of MYO4, a second class V myosin gene in yeast

We have isolated a fourth myosin gene (MYO4) in yeast (Saccharomyces cerevisiae). MYO4 encodes a approximately 170 kDa (1471 amino acid) class V myosin, using the classification devised by Cheney et al. (1993a; Cell Motil. Cytoskel. 24, 215–223); the motor domain is followed by a neck region containing six putative calmodulin-binding sites and a tail with a short potential ‘coiled-coil’ domain. A comparison with other myosins in GenBank reveals that Myo4 protein is most closely related to the yeast Myo2 protein, another class V myosin. Deletion of MYO4 produces no detectable phenotype, either alone or in conjunction with mutations in myo2 or other myosin genes, the actin gene, or secretory genes. However, overexpression of MYO4 or MYO2 results in several morphological abnormalities, including the formation of short strings of unseparated cells in diploid strains, or clusters of cells in haploid strains. Alterations of MYO4 or MYO2 indicate that neither the motor domains nor tails of these myosins are required to confer the overexpression phenotype, whereas the neck region may be required. Although this phenotype is similar to that seen upon MYO1 deletion, we provide evidence that the overexpression of Myo4p or Myo2p is not simply interfering with Myo1p function.

The effect of N-terminal acetylation on the structure of an N-terminal tropomyosin peptide and alpha alpha-tropomyosin

We have used a synthetic peptide consisting of the first 30 residues of striated muscle alpha-tropomyosin, with GlyCys added to the C-terminus, to investigate the effect of N-terminal acetylation on the conformation and stability of the N-terminal domain of the coiled-coil protein. In aqueous buffers at low ionic strength, the reduced, unacetylated 32mer had a very low alpha-helical content (approximately 20%) that was only slightly increased by disulfide crosslinking or N-terminal acetylation. Addition of salt (> 1 M) greatly increased the helical content of the peptide. The CD spectrum, the cooperativity of folding of the peptide, and sedimentation equilibrium ultracentrifugation studies showed that it formed a 2-chained coiled coil at high ionic strength. Disulfide crosslinking and N-terminal acetylation both greatly stabilized the coiled-coil alpha-helical conformation in high salt. Addition of ethanol or trifluoroethanol to solutions of the peptide also increased its alpha-helical content. However, the CD spectra and unfolding behavior of the peptide showed no evidence of coiled-coil formation. In the presence of the organic solvents, N-terminal acetylation had very little effect on the conformation or stability of the peptide. Our results indicate that N-terminal acetylation stabilizes coiled-coil formation in the peptide. The effect cannot be explained by interactions with the "helix-dipole" because the stabilization is observed at very high salt concentrations and is independent of pH. In contrast to the results with the peptide, N-terminal acetylation has only small effects on the overall stability of tropomyosin.

Tryptophan photophysics in rabbit skeletal myosin rod

The fibrous region of myosin (myosin rod) is an alpha-helical, two-stranded coiled-coil made up of identical hydrophobic d sites in the heptad repeat that forms the basis for hydrophobic dimerization. The fluorescence excitation and emission spectra of rod in high salt buffer (where the rod exists as a coiled-coil monomer) at 20 degrees C are red- and blue-shifted, respectively, from the comparable spectra of N-acetyl-tryptophanamide or L-tryptophan. These spectral shifts, as well as red-shifts in the emission spectra induced by excitation on the red edge of the absorption or by increases in temperature, indicate that (on average) the tryptophans are partially exposed to aqueous solvent yet in contact with the protein matrix. The tryptophan intensity decays show an unusual bimodal distribution; the major species has a discrete lifetime of about 5.2 ns while the minor species exhibits a complex decay with a broad (3.4 ns full width at half maximum) Gaussian distribution of lifetimes centered around 1.3 ns. The long lifetime species has a blue-shifted excitation and red-shifted emission characteristic of the indole chromophore in a polar (probably aqueous) environment while the short lifetime species has the spectral parameters characteristic of indole in a non-polar environment. Although assignment of these lifetime species to particular tryptophans in the rod is problematic, this study indicates that the coiled-coil interface presents a complex heterogeneous environment that may undergo rapid conformational mobility.

Pericentrin, a highly conserved centrosome protein involved in microtubule organization

Antisera from scleroderma patients that react widely with centrosomes in plants and animals were used to isolate cDNAs encoding a novel centrosomal protein. The nucleotide sequence is consistent with a 7 kb mRNA and contains an open reading frame encoding a protein with a putative large coiled-coil domain flanked by noncoiled ends. Antisera recognize a 220 kd protein and stain centrosomes and acentriolar microtubule-organizing centers, where the protein is localized to the pericentriolar material (hence, the name pericentrin). Anti-pericentrin antibodies disrupt mitotic and meiotic divisions in vivo and block microtubule aster formation in Xenopus extracts, but do not block gamma-tubulin assembly or microtubule nucleation from mature centrosomes. These results suggest that pericentrin is a conserved integral component of the filamentous matrix of the centrosome involved in the initial establishment of organized microtubule arrays.

Rational design of a three-heptad coiled-coil protein and comparison by molecular dynamics simulation with the GCN4 coiled coil: presence of interior three-center hydrogen bonds

alpha-Helical coiled coils have a 7-residue repeating pattern (abcdefg) where a and d are usually hydrophobic. We have designed a 2-stranded 44-residue coiled-coil protein (P44) consisting of 2 22-residue alpha-helices linked by 2 terminal disulfide groups to test whether the disulfide bridges could stabilize a 3-heptad coiled coil. P44 should be stabilized by intrahelical hydrogen bonds, interhelical disulfide and salt bridges, and interior hydrophobic interactions. A computer model of P44 was built and its stability was studied by molecular dynamics simulation with explicit water. This doubly crosslinked 3-heptad coiled coil did not unfold during a 300-ps simulation with explicit water. This doubly crosslinked 3-heptad coiled coil did not unfold during a 300-ps simulation. But reduced P44 with 4 thiol groups did unfold. For comparison, the 62-residue crystal structure of the 4-heptad coiled coil of transcription activator GCN4 did not unfold during a 300-ps simulation. Thus P44 may be a stable folded protein in aqueous solution. These simulations revealed the presence of 2 local hydrogen bond networks involving intra-helical 3-center hydrogen bonds in the hydrophobic interior of the coiled coils of GCN4 and P44. The NH hydrogen at d makes a 3-center hydrogen bond whose major component is to the i - 4 C = O oxygen at g and minor component is to the solvent-inaccessible i - 3 C = O oxygen at a. Likewise, the NH hydrogen at g makes a 3-center hydrogen bond with the i - 4 C = O oxygen at c and the buried i - 3 C = O oxygen at d.

OS9: a novel olfactory gene of Drosophila expressed in two olfactory organs

A novel olfactory gene, OS9, has been identified in Drosophila by subtractive hybridization. OS9 transcripts have been detected in the third antennal segment, the primary olfactory organ in Drosophila, and in the maxillary palp, which has recently been shown to have olfactory function. The OS9 gene thus represents a molecular link between two distinct olfactory tissues in Drosophila. Little if any OS9 expression has been detected in other segments of the antenna or in other tissues examined. The gene is located in region 38AB on the cytogenetic map. OS9 encodes a protein of 159 amino acids, which contains a putative leucine zipper. Polyclonal antibodies raised against the C-terminal half of the encoded protein react with a 24 kD antennal protein. Antisera raised against both C-terminal and N-terminal halves of the OS9 protein appear to react with cell nuclei in both the third antennal segment and the maxillary palp; interestingly, both antisera also stain cells in the head, including photoreceptor nuclei, as if OS9 were an olfactory-specific member of a family of nuclear proteins, possibly transcription factors.

Unfolding/refolding studies of the myosin rod

The effect of guanidine hydrochloride on the gel-filtration chromatography, viscosity, far ultraviolet circular dichroism and fluorescence emission intensity of the myosin rod was studied under equilibrium conditions. The normalized transition curves for each of these methods were comparable with a midpoint at a guanidine hydrochloride concentration of 1.75-2 M. The curves were not, however, superposable, suggesting that the loss of helix content and the dissociation of the two chains of the myosin rod were not tightly linked. Furthermore, they were unexpectedly independent of the protein concentration over 0.05-20 microM. These phenomena are interpreted taking into account the large size of the molecule. A step-wise process is proposed as a model for the unfolding of the myosin rod.

A spacer protein in the Saccharomyces cerevisiae spindle poly body whose transcript is cell cycle-regulated

Monoclonal antibodies against the 110-kD component of the yeast spindle pole body (SPB) were used to clone the corresponding gene SPC110. SPC110 is identical to NUF1 (Mirzayan, C., C. S. Copeland, and M. Synder. 1992. J. Cell Biol. 116:1319-1332). SPC110/NUF1 has an MluI cell cycle box consensus sequence in its putative promoter region, and we found that the transcript was cell cycle regulated in a similar way to other MluI-regulated transcripts. Spc110p/Nuflp has a long central region with a predicted coiled-coil structure. We expressed this region in Escherichia coli and showed by rotary shadowing that rods of the predicted length were present. The 110-kD component is localized in the SPB to the gap between the central plaque and the sealed ends of the nuclear microtubules near the inner plaque (Rout, M., and J. V. Kilmartin. 1990. J. Cell Biol. 111:1913-1927). We found that rodlike structures bridge this gap. When truncations of SPC110 with deletions in the coiled-coil region of the protein replaced the wild-type gene, the gap between the central plaque and the ends of the microtubules decreased in proportion to the size of the deletion. This suggests that Spc110p connects these two parts of the SPB together and that the coiled-coil domain acts as a spacer element.

A switch between two-, three-, and four-stranded coiled coils in GCN4 leucine zipper mutants

Coiled-coil sequences in proteins consist of heptad repeats containing two characteristic hydrophobic positions. The role of these buried hydrophobic residues in determining the structures of coiled coils was investigated by studying mutants of the GCN4 leucine zipper. When sets of buried residues were altered, two-, three-, and four-helix structures were formed. The x-ray crystal structure of the tetramer revealed a parallel, four-stranded coiled coil. In the tetramer conformation, the local packing geometry of the two hydrophobic positions in the heptad repeat is reversed relative to that in the dimer. These studies demonstrate that conserved, buried residues in the GCN4 leucine zipper direct dimer formation. In contrast to proposals that the pattern of hydrophobic and polar amino acids in a protein sequence is sufficient to determine three-dimensional structure, the shapes of buried side chains in coiled coils are essential determinants of the global fold.

Immunoelectron microscopic localization of the HPC-1 antigen in rat cerebellum

HPC-1 antigen is a neuron-specific 34 kDa protein, identical to p35A (syntaxin), and is thought to play important roles in docking or fusion of synaptic vesicles to presynaptic active zones. In the present study we analyze the distribution of HPC-1 antigen in rat cerebellum by a cryoimmunogold technique using an antibody against the fusion protein of beta-galactosidase and the HPC-1 antigen. HPC-1 antigen was detected at high density on the plasma membranes and synaptic vesicles of presynaptic boutons which formed synapses with dendrites of Purkinje cells, and on the plasma membranes of parallel fibres in the cerebellar molecular layer. In the granule cell layer, gold particles were also detected on the endoplasmic reticulum, nuclear membranes and the plasma membranes of granule cells. Presynaptic membranes and synaptic vesicles in glomeruli were also labelled by gold particles. To determine the topology of HPC-1 antigen on the membranes, the synaptosome fraction prepared from rat cerebellum was embedded in agarose, and processed for the pre-embedding protein A-gold technique. Intact synaptosomes were not labelled by gold particles. However, when fixed in hypotonic fixative to rupture plasma membranes, or when ruptured after fixation in normotonic fixative, the cytoplasmic surfaces of presynaptic membranes and synaptic vesicles were labelled by gold particles. These results suggest that most of the epitopes of HPC-1 antigen are located on the cytoplasmic surface of plasma membranes and synaptic vesicle membranes.

Tryptophan fluorescence quenching in rabbit skeletal myosin rod

The solvent accessibility of the four tryptophans of rabbit skeletal muscle myosin rod was investigated using steady-state and time-resolved fluorescence quenching by iodide, acrylamide, and cesium. The quenching by iodide and acrylamide was biphasic; the discrete, long lifetime component was quenched with bimolecular collision constants (kq) of 1 x 10(9) M-1 s-1 and 1.6 x 10(9) M-1 s-1, respectively, while the Gaussian distributed, short lifetime component was quenched with a kq value of 0.3 x 10(9) M-1 s-1 and 0.04 x 10(9) M-1 s-1, respectively. Comparison with kq values for N-acetyl-tryptophanamide indicated that the fractional solvent accessibility was about 25% for the long and less than 10% for the short lifetime component. Cesium quenching was monophasic and provided evidence of an excess of positive charge around these tryptophans. Our findings cast doubt on the general application of the simple coiled-coil model to describe coiled-coil interactions in this protein in solution.

Expression of foreign proteins on gram-positive commensal bacteria for mucosal vaccine delivery

Non-pathogenic Gram-positive oral commensal bacteria expressing recombinant fusion proteins on their cell surface have been successfully used to raise both a mucosal and a systemic immune response to foreign antigens while colonizing the oropharynx. In this system, fusion-protein vaccines are delivered and anchored to the surface of a commensal, which occupies the mucosal niche invaded by a particular pathogen. Surface expression of these foreign proteins is achieved by exploiting the common mechanism employed by Gram-positive bacteria for translocating and anchoring proteins to the cell surface. The process offers a safe alternative to the use of engineered pathogens as live vaccine delivery vehicles.

Peptide ‘Velcro’: Design of a heterodimeric coiled coil

BACKGROUND

The leucine zipper is a protein structural motif involved in the dimerization of a number of transcription factors. We have previously shown that peptides corresponding to the leucine-zipper region of the Fos and Jun oncoproteins preferentially form heterodimeric coiled coils, and that simple principles involving electrostatic interactions are likely to determine the pairing specificity of coiled coils. A critical test of these principles is to use them as guidelines to design peptides with desired properties.

RESULTS

Based on studies of the Fos, Jun and GCN4 leucine zippers, we have designed two peptides that are predominantly unfolded in isolation but which, when mixed, associate preferentially to form a stable, parallel, coiled-coil heterodimer. To favor heterodimer formation, we chose peptide sequences that would be predicted to give destabilizing electrostatic interactions in the homodimers that would be relieved in the heterodimer. The peptides have at least a 10(5)-fold preference for heterodimer formation, and the dissociation constant of the heterodimer in phosphate-buffered saline is approximately 30 nM at pH 7 and 20 degrees C. Studies of the pH and ionic strength dependence of stability confirm that heterodimer formation is favored largely as a result of electrostatic destabilization of the homodimers.

CONCLUSIONS

Our successful design strategy supports previous conclusions about the mechanism of interaction between the Fos and Jun oncoproteins. These results have implications for protein design, as they show that it is possible to design peptides with simple sequences that have a very high preference to pair with one another. Finally, these sequences with 'Velcro'-like properties may have practical applications, including use as an affinity reagent, in lieu of an epitope tag, or as a way of bringing together two molecules in a cell.

A leucine zipper domain of the suppressor of Hairy-wing protein mediates its repressive effect on enhancer function

The suppressor of Hairy-wing [su(Hw)] protein mediates the mutagenic effect of the gypsy retrotransposon by repressing the function of transcriptional enhancers controlling the expression of the mutant gene. A structural and functional analysis of su(Hw) was carried out to identify domains of the protein responsible for its negative effect on enhancer action. Sequence comparison among the su(Hw) proteins from three different species allows the identification of evolutionarily conserved domains with possible functional significance. An acidic domain located in the carboxy-terminal end of the Drosophila melanogaster protein is not present in su(Hw) from other species, suggesting a nonessential role for this part of the protein. A second acidic domain located in the amino-terminal region of su(Hw) is present in all species analyzed. This domain is dispensable in the D. melanogaster protein when the carboxy-terminal acidic domain is present, but the protein is nonfunctional when both regions are simultaneously deleted. Mutations in the zinc fingers result in su(Hw) protein unable to interact with DNA in vivo, indicating a functional role for this region of the protein in DNA binding. Finally, a region of su(Hw) homologous to the leucine zipper motif is necessary for the negative effect of this protein on enhancer function, suggesting that su(Hw) might exert this effect by interacting, directly or indirectly, with transcription factors bound to these enhancers.

Synthetic model peptides for apolipoproteins. I. Design and properties of synthetic model peptides for the amphipathic helices of the plasma apolipoproteins

Amphipathic helical peptides are the lipid-binding motives of the plasma apolipoproteins, and synthetic peptide analogs have been used to unravel the mechanism of lipid association within this class of proteins. Hydrophobic interactions between the apolar amino acid residues belonging to the hydrophobic face of the amphipathic helices and the lipids are the major driving forces in the peptide-lipid association to form discoidal complexes. Ionic interactions and salt bridge formation between contiguous peptide chains in the complex can, however, contribute to the overall stability of the lipid-protein particle. This was studied by designing peptide analogs to the helical repeats of the apolipoproteins with variable degrees of salt bridge formation between adjacent peptide chains. The most stable conformation for pairs of synthetic peptides was calculated by energy minimisation together with the energy of interaction between peptides. The sequence of the peptides was derived from that of the 18A peptide synthesized by Segrest et al., and the theoretical calculations confirmed that ionic interactions between residues close to each other, along the edge of two adjacent anti-parallel peptides, can significantly contribute towards the stability of a peptide-phospholipid complex.

Transcriptional regulator Leu3 of Saccharomyces cerevisiae: separation of activator and repressor functions

The Leu3 protein of Saccharomyces cerevisiae binds to specific DNA sequences present in the 5' noncoding region of at least five RNA polymerase II-transcribed genes. Leu3 functions as a transcriptional activator only when the metabolic intermediate alpha-isopropylmalate is also present. In the absence of alpha-isopropylmalate, Leu3 causes transcription to be repressed below basal levels. We show here that different portions of the Leu3 protein are responsible for activation and repression. Fusion of the 30 C-terminal residues of Leu3 to the DNA-binding domain of the Gal4 protein created a strong cross-species activator, demonstrating that the short C-terminal region is not only required but also sufficient for transcriptional activation. Using a recently developed Leu3-responsive in vitro transcription assay as a test system for repression (J. Sze, M. Woontner, J. Jaehning, and G. B. Kohlhaw, Science 258:1143-1145, 1992), we show that mutant forms of the Leu3 protein that lack the activation domain still function as repressors. The shortest repressor thus identified had only about 15% of the mass of the full-length Leu3 protein and was centered on the DNA-binding region of Leu3. Implications of this finding for the mechanism of repression are discussed.

Temperature dependence of the interaction of alamethicin helices in membranes

The interaction of the voltage-dependent channel-forming peptide alamethicin with dioleoylphosphatidylcholine (DOPC) small unilamellar vesicles (SUV) has been studied using circular dichroism spectroscopy over a range of wavelengths and temperatures. Evidence is presented for the existence of two distinct membrane-bound states of the peptide which reflect different extents of peptide-peptide interaction. An elevated temperature is found to diminish the apparent peptide-peptide interaction. These results provide insight into the general problem of helix-helix interaction in membranes and provide experimental support for the proposal [Popot, J. L., & Engelman, D. M. (1990) Biochemistry 29, 4031-4037] that these interactions can be enthalpically favorable.

Transcriptional regulator Leu3 of Saccharomyces cerevisiae: separation of activator and repressor functions

The Leu3 protein of Saccharomyces cerevisiae binds to specific DNA sequences present in the 5' noncoding region of at least five RNA polymerase II-transcribed genes. Leu3 functions as a transcriptional activator only when the metabolic intermediate alpha-isopropylmalate is also present. In the absence of alpha-isopropylmalate, Leu3 causes transcription to be repressed below basal levels. We show here that different portions of the Leu3 protein are responsible for activation and repression. Fusion of the 30 C-terminal residues of Leu3 to the DNA-binding domain of the Gal4 protein created a strong cross-species activator, demonstrating that the short C-terminal region is not only required but also sufficient for transcriptional activation. Using a recently developed Leu3-responsive in vitro transcription assay as a test system for repression (J. Sze, M. Woontner, J. Jaehning, and G. B. Kohlhaw, Science 258:1143-1145, 1992), we show that mutant forms of the Leu3 protein that lack the activation domain still function as repressors. The shortest repressor thus identified had only about 15% of the mass of the full-length Leu3 protein and was centered on the DNA-binding region of Leu3. Implications of this finding for the mechanism of repression are discussed.

The N-terminal coiled-coil domain of beta is essential for gamma association: a model for G-protein beta gamma subunit interaction

We have identified the N terminus of the beta subunit as an essential domain for G-protein beta gamma assembly. A C-terminal fragment, beta 1-(130-340), fails to bind gamma unless coexpressed with the complementary N-terminal fragment, beta 1-(1-129). Deletion of the N-terminal 33 residues of beta 1, a region identified by computer algorithm to favor coiled-coil formation, abolishes gamma 2 association. On the basis of these findings, we propose a coiled-coil model of beta gamma interaction and refine this by computer-assisted molecular modeling. The model is tested by further mutagenesis: reversing the charge of residues in beta 1 that are hypothesized to be involved in interhelical salt bridges precludes gamma association. Insertions in the coiled-coil region, which disrupt the proposed hydrophobic interface, prevent gamma association. This structural basis for beta gamma dimerization provides a starting point for the design of beta and gamma mutants that can be used to map regions in beta gamma critical for interactions with the alpha subunit, receptors, and effectors.

Determinants of binding-site specificity among yeast C6 zinc cluster proteins

Related DNA binding proteins often recognize similar DNA sites but can distinguish among them with the use of different protein-DNA contacts. Here, it is shown that members of the C6 zinc cluster family of yeast transcriptional activators distinguish related DNA sites by a different mechanism. The DNA binding site for each of these proteins contains identical nucleotide triplets (CGG ... CCG) but differs in the spacings between the triplets. It is shown that zinc clusters of these proteins work interchangeably to recognize the conserved triplets and that the region 19 amino acids to the carboxyl-terminal side of the zinc cluster, comprising the linker and the beginning of a dimerization element as inferred from the GAL4 crystal structure, directs the protein to its preferred site.

Conformational intermediates in the folding of a coiled-coil model peptide of the N-terminus of tropomyosin and alpha alpha-tropomyosin

Circular dichroism was used to study the folding of alpha alpha-tropomyosin and AcTM43, a 43-residue peptide designed to serve as a model for the N-terminal domain of tropomyosin. The sequence of the peptide is AcMDAIKKKMQMLKLDVENLLDRLEQLEADLKALEDRYKQLEGGC. The peptide appeared to form a coiled coil at low temperatures (< 25 degrees C) in buffers with physiological ionic strength and pH. The folding and unfolding of the peptide, however, were noncooperative. When CD spectra were examined as a function of temperature, the apparent degree of folding differed when the ellipticity was followed at 222, 208, and 280 nm. Deconvolution of the spectra suggested that at least three component curves contributed to the CD in the far UV. One component curve was similar to the CD spectrum of the coiled-coil alpha-helix of native alpha alpha-tropomyosin. The second curve resembled the spectrum of single-stranded short alpha-helical segments found in globular proteins. The third was similar to that of polypeptides in the random coil conformation. These results suggested that as the peptide folded, the alpha-helical content increased before most of the coiled coil was formed. When the CD spectrum of striated muscle alpha alpha-tropomyosin was examined as a function of temperature, the unfolding was also not totally cooperative. As the temperature was raised from 0 to 25 degrees C, there was a decrease in the coiled coil and an increase in the conventional alpha-helix type spectrum without formation of random coil. The major transition, occurring at 40 degrees C, was a cooperative transition characterized by the loss of all of the remaining coiled coil and a concomitant increase in random coil.

Paramyosin of Echinococcus granulosus: cDNA sequence and characterization of a tegumental antigen

A lambda ZAPII cDNA library of Echinococcus granulosus larvae was expressed in Escherichia coli SURE cells. Screening of the library with a rabbit antiserum raised against total larval antigen yielded several immunoreactive clones. For analysis of the nucleotide sequence, in vivo excision into pBlueskript was carried out and the 3' end of the cloned insert was sequenced. Three of these clones exhibited identical nucleotide sequences, suggesting expression of identical genes. The complete nucleotide sequence of the largest clone, EG36, with a 3.4-kb insert was determined, presenting an open reading frame of 2.59 kb. The predicted amino acid sequence showed 71.4% identity to the Schistosoma mansoni paramyosin and a significant homology to a 17 amino-acid peptide sequence from antigen B of Taenia solium. From these data we conclude that EG36 is the paramyosin of E. granulosus. For protein purification, the coding sequence of the cDNA was amplified by polymerase chain reaction and ligated in frame into the expression vector pGEX-3X. Affinity-chromatography-purified GST fusion protein was used to induce a polyclonal rabbit antiserum. Immunoblot analysis revealed the expression of a 97-kDa protein by the E. coli clone and that of a protein with a similar molecular weight in protoscolices from E. granulosus and E. multilocularis as well as in E. granulosus cyst fluid. Immunofluorescence studies showed that EG36 was localized throughout the tegument of E. granulosus and E. multilocularis larvae. Sera from patients suffering from echinococcosis, schistosomiasis, and neurocysticercosis reacted with the purified fusion protein when tested in an enzyme-linked immunosorbent assay.

Molecular dynamics study of structure and stability of a model coiled coil

This paper employs methods used earlier to study helix propensity in a model alpha-helix. The methods are extended to simulations of a motif structure of the alpha-helical coiled coil, i.e., a structure with a simple amino acid sequence, containing only alanine, leucine, and valine, with leucine and valine forming hydrophobic contacts in the helix interface (positions "d" and "a"). Dynamic simulations of the model coiled-coil structure reproduce characteristic features of the coiled-coil motif seen in experimental studies. Free energy simulations were used to assess the change in stability of the model when a leucine pair or a valine pair in the helix interface was replaced with an alanine pair. A leucine pair at position d was found to contribute 3.4 kcal/mol to the stability of the coiled coil relative to an alanine pair, and a valine pair at position a was found to contribute 0.8 kcal/mol relative to an alanine pair. The value for the leucine pair agrees with reports in two experimental studies with molecules having different amino sequence. The value for the valine pair is reasonable given the smaller size of the valine side chain and the intrinsic low helix propensity of valine. No experimental value was available for comparison.

Structure, function and application of the coiled-coil protein folding motif

Recent X-ray analyses and synthetic model studies of the coiled-coil motif have clarified roles for hydrophobic core residues and ionic interactions in determining stability, selectivity, stoichiometry and orientation of alpha-helices in this structure. Although much remains to be learnt, current knowledge now enables this motif to be used in novel constructs and points the way to a more explicit understanding of native coiled-coil formation and protein folding in general.

Probing the roles of residues at the e and g positions of the GCN4 leucine zipper by combinatorial mutagenesis

Combinatorial mutagenesis with an alphabet limited to alanine, glutamic acid, lysine, and threonine was used to probe the role of interactions involving surface residues in stabilizing a short alpha-helical coiled coil. The residues at eight e and g positions in the leucine zipper of the Saccharomyces cerevisiae transcription factor GCN4 were randomized to these four residues in a lambda repressor-leucine zipper fusion protein, resulting in 65,536 possible residue combinations. Roughly three-fourths of these combinations allowed stable coiled-coil formation as assayed by DNA binding by the fusion protein. To understand the basis for the activity differences, functional and non-functional mutants were sequenced and statistical tests were applied to identify structure/function correlations. Helix-forming propensity and favorable intrasubunit and intersubunit charge-charge interactions were positively correlated with activity. These studies suggest that the identities of surface side chains at the e and g positions of coiled coils contribute modestly to stability; by comparison with previous work, however, the e and g positions are far less critical than residues at the a and d positions, which form the hydrophobic core of the dimer interface.

Dimerization specificity of the leucine zipper-containing bZIP motif on DNA binding: prediction and rational design

We propose an interhelical salt bridge rule to explain the dimerization specificity between the two amphipathic alpha-helices in the leucine zipper structure. Using the bZIP class of DNA-binding proteins as a model system, we predicted and designed novel dimerization partners. We predicted that ATF4, a member of the ATF/CREB family of transcription factors, would preferentially form heterodimers with IGEBP1, a member of the C/EBP superfamily. These predictions were verified using a gel mobility-shift assay. To further test the value of this interhelical salt bridge rule, we modified the bZIP protein C/EBP attempting to design molecules that would form preferentially heterodimers with C/EBP or molecules that would not interact with C/EBP. These designed molecules behaved as predicted. Therefore, we conclude that this interhelical salt bridge rule is useful in understanding the dimerization specificity of bZIP proteins. In addition, we suggest that this rule could be used to design novel "dominant-negative" molecules to specifically inhibit the function of target leucine zipper proteins in vivo.

Correlation between protein stability and crystal properties of designed ROP variants

Six variants of the ROP protein, designed with the aim to analyze by X-ray crystallography loop formation and core packing interactions in 4-alpha-helical bundles, have been purified and a search of their crystallization conditions has been carried out. Five mutants yield crystals that are suitable for medium to high resolution X-ray diffraction studies. For all mutants crystal size, sensitivity to X-irradiation and diffraction limit are correlated to their stability as determined by differential scanning calorimetry, in a manner which is not yet understood in detail.

A spring-loaded mechanism for the conformational change of influenza hemagglutinin

Influenza hemagglutinin (HA) undergoes a conformational change that induces viral fusion with the cellular membrane. The structure of HA in the fusogenic state is unknown. We have identified a sequence in HA that has a high propensity for forming a coiled coil. Surprisingly, this sequence corresponds to a loop region in the X-ray structure of native HA: the loop is followed by a three-stranded, coiled-coil stem. We find that a 36 residue peptide (LOOP-36), comprising the loop region and the first part of the stem, forms a three-stranded coiled coil. This coiled coil is extended and stabilized in a longer peptide, corresponding to LOOP-36 plus the residues of a preceding, short alpha helix. These findings lead to a model for the fusogenic conformation of HA: the coiled-coil stem of the native state extends, relocating the hydrophobic fusion peptide, by 100 A, toward the target membrane.

Interactions of basic amphiphilic peptides with dimyristoylphosphatidylcholine small unilamellar vesicles: optical, NMR, and electron microscopy studies and conformational calculations

The interactions of DMPC small unilamellar vesicles with four amphiphilic polypeptides [(LKKL)n, (LRRL)n, (LKKL)4, and (YKKY)n] have been investigated by circular and infrared dichroism, turbidimetry, electron microscopy, and fluorescence, 1H, and 31P nuclear magnetic resonance spectroscopy. The main results obtained are the following: (1) Well-defined complexes are formed by the association of one amino acid residue with approximately two lipid molecules. (2) In the presence of polypeptides fusions are observed between SUVs when the molar ratio p is less than 0.05, and a clearance effect is observed when p is higher than 0.05. (3) The anchoring sites of the polypeptides on DMPC molecules are the negative phosphate groups through electrostatic interactions with the terminal NH3+ of lysine residues. (4) The polypeptides adopt an alpha-helical conformation with their axis parallel to the membrane surface. The hydrophobic part of the amphiphilic alpha helix can penetrate the outer lipid leaflet down to the C5 position. (5) Choline methyl groups are not involved in the interactions between lipid molecules and amino acid residues. (6) Phosphorus atom mobility around the P-O-glycerol bond is strongly reduced whereas that of methylene groups is progressively weakened when going up from C13 to C1. Finally, using modeling and energy calculations a model of possible Ac(LKKL)4NHEt-DMPC SUV complexes is presented.

Bovine filensin possesses primary and secondary structure similarity to intermediate filament proteins

The cDNA coding for calf filensin, a membrane-associated protein of the lens fiber cells, has been cloned and sequenced. The predicted 755-amino acid-long open reading frame shows primary and secondary structure similarity to intermediate filament (IF) proteins. Filensin can be divided into an NH2-terminal domain (head) of 38 amino acids, a middle domain (rod) of 279 amino acids, and a COOH-terminal domain (tail) of 438 amino acids. The head domain contains a di-arginine/aromatic amino acid motif which is also found in the head domains of various intermediate filament proteins and includes a potential protein kinase A phosphorylation site. By multiple alignment to all known IF protein sequences, the filensin rod, which is the shortest among IF proteins, can be subdivided into three subdomains (coils 1a, 1b, and 2). A 29 amino acid truncation in the coil 2 region accounts for the smaller size of this domain. The filensin tail contains 6 1/2 tandem repeats which match analogous motifs of mammalian neurofilament M and H proteins. We suggest that filensin is a novel IF protein which does not conform to any of the previously described classes. Purified filensin fails to form regular filaments in vitro (Merdes, A., M. Brunkener, H. Horstmann, and S. D. Georgatos. 1991. J. Cell Biol. 115:397-410), probably due to the missing segment in the coil 2 region. Participation of filensin in a filamentous network in vivo may be facilitated by an assembly partner.

Identification and characterization of a myosin heavy chain gene (mhcA) from the human parasitic pathogen Entamoeba histolytica

The mhcA gene from the human pathogen Entamoeba histolytica was identified using the polymerase chain reaction. It is a single copy gene expressed as a 6.4-kb mRNA. The deduced MhcA protein sequence is highly similar to myosin II from both Dictyostelium discoideum and Acanthamoeba castellanii. The globular head domain of MhcA contains the specific regions involved in ATP binding, actin binding, and interaction with myosin light chain. The tail domain is organized in an alpha-helical coiled coil structure, which suggests that MhcA is an alpha-fibrous protein. The coiled coil is interrupted by two prolines indicating that like other myosins, either from smooth muscle or from non-muscle cells, the tail of MhcA folds twice on itself. In addition, MhcA presents sequence similarities with the heavy chain phosphorylation sites of smooth and non-muscle vertebrate myosins.

Molecular variation in Plasmodium falciparum: polymorphic antigens of asexual erythrocytic stages

Numerous polymorphic antigens of the asexual erythrocytic stages of P. falciparum are now well characterized. Diversity in some of these antigens, including MSA-1, MSA-2 and the S-antigen is associated with changes in the repeat sequences which are frequently a prominent structural feature of malaria antigens. It is not known whether the variation in repeats causes allelic gene products to adopt different conformations but variation in and around the repeats in SPAM, a newly characterized secreted antigen, preserve the unusual alanine-heptad repeats which we assume generate a helical bundle in this protein. Mutations in non-repetitive regions of the S-antigen and in AMA-1, an antigen lacking repeats, are strongly biased towards those which alter the amino acid sequence. This and other evidence indicates the operation of biological selection but the role of immune responses as a selection pressure operating on these diverse antigens remains to be established.

Structure of chicken annexin V at 2.25-A resolution

The crystal structure of chicken annexin V has been solved by molecular replacement and refined at 2.25 A. The final R factor is 19.7% with good geometry. The chicken annexin V structure is very similar to the human annexin V structure, with four similar domains each containing five helices. The structure includes three calcium ions in domains I, II, and IV, each bound by the characteristic K-G-X-G-T-(38 residues)-D/E motif. In view of the structural similarity between human and chicken annexin V, we suggest that they have a common vital function which developed early in evolutionary history.

Subunit interactions provide a significant contribution to the stability of the dimeric four-alpha-helical-bundle protein ROP

Detailed thermodynamic and spectroscopic studies were carried out on the ColE1-ROP protein in order to establish a quantitative basis for the contribution of noncovalent interactions to the stability of four-helix-bundle proteins. The energetics of both heat- and GdnHCl-induced denaturation were measured by differential scanning microcalorimetry (DSC) and/or by following the change in circular dichroism in the far-UV range. Sedimentation equilibrium analyses were performed to characterize the state of aggregation of the protein. No intermediate species could be detected during thermal unfolding of the dimer in the absence of GdnHCl. Under these conditions ROP unfolding exhibits a strict two-state behavior. The thermodynamic parameters for the reaction N2<->2D are delta HD = 580 +/- 20 kJ.(mol of dimer)-1, delta Cp = 10.3 +/- 1.3 kJ.(mol of dimer)-1.K-1, and Tm = 71.0 +/- 0.5 degrees C. The corresponding Gibbs energy change of unfolding is delta GD degree = 71.7 kJ.(mol of dimer)-1 at 25 degrees C and pH 6. In the presence of 2.5 M GdnHCl, however, ROP dissociates into monomers at elevated temperatures, as the loss of the concentration dependence of Tm and the decreased molecular weight demonstrate. The corresponding transition parameters are delta HD (2.5 M GdnHCl) = 130 +/- 10 kJ.(mol of monomer)-1 and Tm = 51.6 +/- 0.3 degrees C. Isothermal unfolding studies at 19 degrees C using GdnHCl as denaturant yielded a Gibbs energy change of unfolding of 22.4 kJ.(mol of monomer)-1. This extrapolated value is 38% lower than the corresponding delta GD degree value of 35.85 kJ.(mol of monomer)-1 calculated from thermal unfolding for the monomer in the absence of GdnHCl, where the protein is known to be a dimer. These results suggest that subunit interactions are an important source of stabilization of the native four-helix-bundle structure of ROP.

Packing and recognition of protein structural elements: a new approach applied to the 4-helix bundle of myohemerythrin

We present a novel search strategy for determining the optimal packing of protein secondary structure elements. The approach is based on conformational energy optimization using a predetermined set of side chain rotamers and appropriate methods for sampling the conformational space of peptide fragments having fixed backbone geometries. An application to the 4-helix bundle of myohemerythrin is presented. It is shown that the conformations of the amino acid side chains are largely determined at the level of helix pairs and that superposition of these results can be used to construct the full bundle. The final solution obtained, taking into account restrictions due to the lateral amphiphilicity of the helices, differs from the native structure by only a 20 degrees rotation of a single helix.

Disulfide bond contribution to protein stability: positional effects of substitution in the hydrophobic core of the two-stranded alpha-helical coiled-coil

To investigate the positional effect of the disulfide bond on the structure and stability of a two-stranded alpha-helical coiled-coil, an interchain disulfide bond was systematically introduced into the hydrophobic core of a de novo designed model coiled-coil at the N-terminus (position 2), C-terminus (position 33), and nonterminal positions a (positions 9, 16, 23, and 30) and d (positions 5, 12, 19, and 26). The rate of formation of a disulfide bond is faster at position d compared to at the corresponding position a under nondenaturing conditions, suggesting that position d is more suitable for engineering a disulfide bond. The structure and stability of the reduced and oxidized coiled-coils were determined by circular dichroism studies in the absence and presence of guanidine hydrochloride. Our results demonstrate that the improvement of protein stability by introduction of a disulfide bond is very relevant to its location and the most effective disulfide bonds are those that can be introduced in the hydrophobic core without any disruption of the protein structure. The disulfide bond at position d with near-optimal geometry does not perturb the coiled-coil structure and makes the largest contribution to coiled-coil stability. In contrast, the inappropriate geometry of the disulfide bond at nonterminal position a introduces a high strain energy on the disulfide bond which disrupts the coiled-coil structure. At positions a, the closer the disulfide bridge is to the center of the coiled-coil, the larger the disruption on the coiled-coil structure and the smaller the contribution the disulfide bond makes to coiled-coil stability. The computer modeling results also suggest that an insertion of an interchain disulfide bond at position a in the GCN4 leucine zipper X-ray structure has a higher potential energy than insertion at position d. The energy-minimized coiled-coil structure with an interchain disulfide bond at position a has a larger root mean square difference from the X-ray structure of GCN4 than the coiled-coil with a disulfide bond at position d. Because interhelical interactions are common in globular proteins as well as coiled-coils, the results obtained in this study will have general utility for selecting the sites for engineering disulfide bonds between alpha-helices.

Packing and hydrophobicity effects on protein folding and stability: effects of beta-branched amino acids, valine and isoleucine, on the formation and stability of two-stranded alpha-helical coiled coils/leucine zippers

The aim of this study was to examine the differences between hydrophobicity and packing effects in specifying the three-dimensional structure and stability of proteins when mutating hydrophobes in the hydrophobic core. In DNA-binding proteins (leucine zippers), Leu residues are conserved at positions "d," and beta-branched amino acids, Ile and Val, often occur at positions "a" in the hydrophobic core. In order to discern what effect this selective distribution of hydrophobes has on the formation and stability of two-stranded alpha-helical coiled coils/leucine zippers, three Val or three Ile residues were simultaneously substituted for Leu at either positions "a" (9, 16, and 23) or "d" (12, 19, and 26) in both chains of a model coiled coil. The stability of the resulting coiled coils was monitored by CD in the presence of Gdn.HCl. The results of the mutations of Ile to Val at either positions "a" or "d" in the reduced or oxidized coiled coils showed a significant hydrophobic effect with the additional methylene group in Ile stabilizing the coiled coil (delta delta G values range from 0.45 to 0.88 kcal/mol/mutation). The results of mutations of Leu to Ile or Val at positions "a" in the reduced or oxidized coiled coils showed a significant packing effect in stabilizing the coiled coil (delta delta G values range from 0.59 to 1.03 kcal/mol/mutation). Our results also indicate the subtle control hydrophobic packing can have not only on protein stability but on the conformation adopted by the amphipathic alpha-helices. These structural findings correlate with the observation that in DNA-binding proteins, the conserved Leu residues at positions "d" are generally less tolerant of amino acid substitutions than the hydrophobic residues at positions "a."

Dynamics of intermediate filaments. Recent progress and unanswered questions

Intermediate filaments (IFs) have always been considered as the most static and 'skeletal' cellular elements. This view is now changing: new information reveals that IFs exchange subunits at steady-state, that IF networks can be assembled de novo, and that IF proteins are subject to elaborate chemical modification and de-modification during mitosis. I describe below some of the key observations which have made us realize that IFs are dynamic structures. I also discuss some of the remaining questions pertinent to the pathways of IF assembly under in vivo conditions.