Biomedical subjects
Publications and source records attributed to E Stellwagen.
Capillary zone electrophoresis measurements in acidic isoelectric buffers provide a sensitive and rapid method for comparison of the folding and stability of wild type, mutant or post-translationally modified proteins. The potential of the method is illustrated using the small globular protein cytochrome c.
The model peptide XAAAAEAAARAAAARamide is used to examine the contributions of an N-terminal capping interaction to the conformation and stability of a helical ensemble. The reference peptide has an alanine residue at position X while the capping peptide has a serine residue at this position. The helical ensemble was characterized using circular dichroism measurements and carbonyl-carbon chemical shift measurements of selectively enriched residues. The distribution of helicity within the ensemble of the reference peptide at pH 11 and 0 degrees C appears symmetrical, having a uniform central helix and frayed ends. This distribution is truncated at pH 6 by the repulsive electrostatic interaction between the positively charged alpha-amino group and the positively charged end of the helical macrodipole. The capping peptide forms a side-chain/ main-chain hydrogen bond involving the serine residue and amide of alanine 4. The presence of this hydrogen bond generates a unique motif in the chemical shift profile of its helical ensemble. The conformational stabilization contributed by this hydrogen bond, although cooperatively distributed throughout the helical ensemble, is preferentially focused within the first helical turn. The stabilization provided by this hydrogen bond is able to offset the truncation of the helical ensemble generated by the repulsive electrostatic interaction observed at pH 6.
The thermal dependence of the carbonyl carbon chemical shift of each residue in a helical peptide may be analyzed in terms of a two-state helix/coil transition. Such analyses generate values for the chemical shift of each residue in the helical and in the coil conformational ensembles of the peptide. The sequence dependence of the difference in these two values, termed the difference chemical shift, provides a description of the mean distribution of helicity within the helical ensemble. In this report, we improve two aspects of the procedures used to analyze prior chemical shift measurements of the helical peptide acetylW (EAAAR)3Aamide. The new difference chemical shift values for 16 of the 18 residues describe a very uniform central helical ensemble with frayed ends. However, the difference chemical shift values for the two remaining residues, alanines 03 and 08, are significantly diminished relative to this uniform distribution. Each of these two alanine residues is located i-4 to a glutamate residue. It is suggested that the difference chemical shifts for these two alanine residues are diminished by a self-capping interaction within the i + 4 glutamate residues.
Changes in the temperature, pH, ionic strength, or denaturant concentration of aqueous solutions of the monomeric non-alpha-helical peptide acetylYEAAAKEAPAKEAAAKAamide generate changes in its dichroic spectrum characteristic for a conformational transition. This transition has the characteristic features of a residue PII/unstructured conformational equilibrium in which PII denotes an extended left-handed helical conformation and unstructured denotes all the remaining conformations in a random coil ensemble. Replacement of the proline residue facilitates population of residues in an alpha-helical conformation. However, the ellipticity values for these non-proline peptides merge with the ellipticity of the proline peptide as the population of residues in the alpha-helix conformation is diminished. This convergence suggests that all residues in a host/guest peptide series of the same length share a common PII/unstructured conformational equilibrium in a given solvent. We propose that the fractional helix content of peptides within such a series may be estimated by using a two-state calculation in which the ellipticity for the non-alpha-helix conformations is provided by a peptide having a central proline guest residue.
Traditional statistical models for the prediction of peptide helicity are written in terms of the mean fractional helicity of the peptide residues. Far ultraviolet circular dichroic measurements of peptide solutions are converted to mean fractional helicity by partitioning the observed ellipticity between that of a perfect helix and a random coil. This partition does not adequately represent the ensemble of peptide molecules present in solution that populate imperfect helical conformations of quite variable lengths. A new dichroic statistical model has been written in terms of ellipticity rather than fractional helical content that recognizes (1) the source of ellipticity, peptide bond adsorption; (2) the differential ellipticity of peptide bonds in the terminal and interior helical turns; and (3) the contributions of each participant in a conformational ensemble to the observed ellipticity. Comparative analyses of host/guest peptides indicates that significant differences are obtained between residue w and n weights and ellipticity values using the traditional and dichroic statistical models.
The helix/coil equilibrium of a peptide in solution can be modulated by a variety of side-chain interactions that are not incorporated into the standard statistical mechanical models for prediction of peptide helical content. In this report, we describe a recursive formulation of the Lifson-Roig model that facilitates incorporation of specific pairwise side-chain interactions as well as nonspecific individual side-chain capping interactions. Application of this extended model to a series of host/guest peptides indicates that the apparent delta G value for a pairwise apolar interaction is dependent upon the spacing and orientation but not the sequential location of the participating residues. The apparent delta G values for such interactions are about 40% greater than the corresponding apparent delta delta G values obtained from difference measurements.
The mean residue ellipticity of the helical host peptide, acetyl-YEAAAKEAXAKEAAAKA-amide containing guest residues at position X, was measured as a function of pH and ionic strength at 0 degree C. Changes in ellipticity at 222 nm were interpreted in terms of a two-state helix/coil transition of a monomeric peptide. Variable pH measurements in low concentrations of KCl defined changes in helix stability resulting from the ionization of each guest residue. Variable [KCl] measurements at fixed pH generated ellipticity values for the neutral and ionic forms of each guest residue free of electrostatic and lyotropic contributions. These ellipticity values were used to calculate a helix propagation parameter for each form of a guest residue using the Lifson-Roig algorithm and assuming a universal nucleation parameter. In all cases, the propagation parameter of a residue is either unaffected or decreased by ionization of its side chain.
Purified phosphofructokinase 1 from baker's yeast (Saccharomyces cerevisiae) was subjected to proteolysis by thermolysin, endoproteinase lys-C, trypsin and chymotrypsin under defined solvent conditions. In the absence of substrates and allosteric effectors, the catalytic activity of phosphofructokinase rapidly disappeared in the presence of each proteolytic enzyme. The presence of a saturating concentration of ATP protected phosphofructokinase activity from proteolytic inactivation while the collective presence of fructose 6-phosphate, AMP and fructose 2,6-bisphosphate provided transient activation during proteolysis. Changes in the quaternary structure of phosphofructokinase resulting from proteolysis were estimated by high performance size exclusion chromatography while changes in the primary sequence of the individual alpha and beta polypeptide chains were estimated by polyacrylamide-gel electrophoresis in sodium dodecylsulfate. The site(s) of proteolytic cleavage were identified by N-terminal sequence analysis of resolved electrophoretic components. The presence of ATP protects phosphofructokinase from thermolysin proteolysis, while the collective presence of fructose 6-phosphate, AMP and fructose 2,6-bisphosphate restricts proteolysis to one site in each polypeptide chain involving the peptide bonds preceding Leu199 in the alpha chain and Leu192 in the beta chain. The truncated phosphofructokinase retains its octameric structure. The presence of ATP largely restricts endoproteinase lys-C proteolysis to a single site in the alpha chain involving the peptide bond preceding Val914. This cleavage results in the dissociation of the octameric form of phosphofructokinase into two tetramers. The presence of ATP restricts both trypsin and chymotrypsin proteolysis to the N-terminal and C-terminal regions described above, resulting in the preferential stabilization of the tetrameric form of phosphofructokinase. It would appear that the first 200 and last 80 residues which are unique to the sequence of the yeast phosphofructokinase are not directly involved in catalysis or its allosteric regulation. However, the last 80 residues of the alpha polypeptide chain do appear to stabilize an octameric structure which is unique to yeast phosphofructokinase.
Circular dichroic measurements of the host peptide acetyl-Y(EAAAK)3A-amide were obtained in solutions of increasing ionic strength at pH 7.0 and 0 degree C. The changes observed in the dichroic spectra are characteristic for a two-state helix/coil transition. The mean residue ellipticity at 222 nm exhibits a curvilinear dependence on ionic strength which becomes linear at ionic strengths greater than 1 M. The slope of the linear portion is assumed to represent the lyotropic character of the salt, and its extrapolated intercept is assumed to represent the mean residue ellipticity of the peptide solution freed from both electrostatic and lyotropic contributions which affect the helical stability of the host peptide. An extrapolated mean residue ellipticity value was obtained for each host peptide having a different amino acid guest residue at position 9 in the peptide sequence. These values were used to calculate a propagation parameter, s, for each residue using the Lifson-Roig algorithm for peptide helical content and assuming a common nucleation parameter of 0.003. The ability of these minimally determined residue parameters to predict the helical content of a variety of peptides is encouraging. Estimates were also made of the delta G values for the electrostatic interactions within the host peptide and for the additional interactions generated by ionic guest residues.
Size exclusion profiles of proteins with persistent conformations exhibit broad asymmetric peaks whose shape and elution times are dependent on denaturant concentration. The collective elution profiles were precisely simulated by an apparent binding model that treats the denaturant dependence in terms of an apparent matrix binding. The model requires three experimentally measurable parameters: the elution time for the unbound protein, an apparent association equilibrium constant for binding, and an apparent exchange time for binding. The denaturant dependence for each of these parameters is related to the accessible surface area of the protein.
The kinetics of the hydrodynamic volume change associated with the unfolding and refolding of a globular protein can be observed using high performance size exclusion chromatography. Chromatographic profiles that evidence such dynamics can be simulated using equations in which chromatographic partitioning and the conformational transition are described in terms of a finite difference algorithm incorporating an apparent binding model to generate broad and asymmetric peaks. Application of these equations to the simple two-state unfolding transition of ribonuclease A in guanidine hydrochloride indicates that reliable kinetic parameters can be obtained using these equations.
The unfolding and refolding of T4 thioredoxin was observed by equilibrium and kinetic size exclusion chromatographic measurements in guanidine hydrochloride at 4 degrees C and pH 7.0. All the observed chromatographic profiles can be simulated by a cubic mechanism using a consistent set of equilibrium and kinetic parameters describing each of the coupled transitions. The four components in the folded protein and in the unfolded protein are interrelated by configurational transitions having parameters characteristic for proline peptide isomerizations. Only two of the four folded conformations are significantly populated at equilibrium. Each of the four unfolded components can refold by a unique conformational transition. No transiently populated folding intermediates are detected having hydrodynamic volumes intermediate between those characteristics for the folded and unfolded protein.
The kinetics of the hydrodynamic volume change accompanying the reversible unfolding of staphylococcal nuclease have been observed by size-exclusion chromatography at 4 degrees C and pH 7.0 using the denaturant guanidine hydrochloride. The observed chromatographic profiles have been simulated by a six-component unfolding/refolding mechanism using a consistent set of equilibrium and kinetic parameters. The native protein is an equilibrium mixture of the cis and trans isomers of the peptide bond preceding proline-117. The native conformation containing the cis isomer dominates the equilibrium mixture, is more stable, and unfolds more slowly at its transition midpoint. The denatured protein is an equilibrium mixture of at least four components, the cis/trans isomers of proline-117 and one of the five remaining prolines. The dominant refolding pathway is initiated from the denatured component containing the trans isomer of proline-117. The six-component mechanism is consistent with tryptophan fluorescence kinetic measurements of the wild-type protein and with chromatographic measurements of a mutant P117G protein.
Comparative CD measurements were made on the model helical peptides acetylYEAAAKEAXAKEAAAKAamide and acetylYEAAAEKAXAKEAAAKAamide in which X represents a nonaromatic nonionic residue. The former peptide contains three potential i, i + 4 complementary ion pairs at neutral pH, while the latter peptide contains one potential complementary and two potential antagonistic i, i + 4 ion pairs. The effect of pH and ionic strength on the mean residue ellipticity of these peptides was measured at 222 nm and 0 degrees C. These measurements were analyzed assuming a common two-state helix/coil transition and only i, i, + 4 ion-pair interactions. The analyses suggest that the central ion pairs do modulate helical content while the peripheral ion pairs do not, presumably due to the location of the peripheral ion pairs in the frayed ends of the helix. The complementary central ion pair stabilizes the helix by about 0.4 kcal/mole and the antagonistic central ion pair destabilizes the helix by about 0.2 kcal/mole.
The sequence of a model monomeric peptide, acetylA(EAAAK)3Aamide was altered to expedite measurement of peptide concentration and to enhance its fractional helical content. Replacement of the N-terminal alanine residue with a tryptophan residue provides a convenient chromophore for measurement of peptide concentration without diminishing the helical content. Replacement of the three lysine residues with arginine residues enhances the helical content without loss of their electrostatic contributions. Increasing the number of EAAAR sequence units in the peptide acetylW(EAAAR)nAamide from three to five indicates that the spectral features anticipated for a completely helical peptide are closely approached.
The three ER ion pairs in the peptide acetyl-W(EAAAR)3A-amide were replaced in turn with the ion pairs EK, EO, DR, DK, and DO, where O represents an ornithine residue. The far-ultraviolet circular dichroic spectra of the six peptides measured in 10 mM NaCl at pH 2 and 0 degrees C form a nested set having an isodichroic point at 203 nm of -17,000 deg cm2 dmol-1. The ellipticity values of the six peptides at 222 nm range from -31,600 to -7400 deg cm2 dmol-1 in the order listed. Changing the pH of each peptide solution from 2 to 13 also generates a nested set of dichroic spectra with the same isodichroic values. Increasing the pH from 2 to 7 differentially increases the ellipticity at 222 nm in a single transition having an apparent pK of 4.1 for the E-containing peptides are 3.6 for the D-containing peptides. Increasing the pH beyond neutrality differentially decreases the ellipticity at 222 nm in a single transition having an apparent pK of greater than or equal to 13.2 for the R-containing peptides, 11.1 for the K-containing peptides, and 10.7 for the O-containing peptides. It is proposed that the difference in the ellipticity of the six peptides chiefly reflects the helix preferences for the variable residues supplemented by intrahelical electrostatic interactions in the neutral pH range.
The peptide acetylYEAAAKEARAKEAAAKAamide exhibits the dichroic features characteristic of a monomeric helix/coil transition in aqueous solution. Nineteen variants of this peptide each containing a different residue at position 9 were prepared by solid-phase peptide synthesis and purified by reversed-phase chromatography. The thermal dependence of the far-ultraviolet dichroic spectrum of each of these peptides except that containing proline is characteristic for an alpha-helix/coil transition. The relative stability of the helical forms of these peptides does not correlate with the preference of the variable amino acid to occupy a middle position in a protein helix. It is likely that the specific interactions of the variable residue with its local environment obscure any inherent preference of the residue to reside in an alpha-helix.
The 17-residue peptide acetylAEAAAKEAAAKEAAAKAamide, described as an autonomous folding unit (Marqusee & Baldwin, 1987), has been used to examine the effect of amino acid replacements on helix stability. Alanine residues(s) at positions 4, 9, and 14 in the peptide sequence were replaced either singly or multiply by either serine or methionine residues with solid-phase peptide synthesis. The thermal dependence of the helix/coil transition of each peptide was observed by far-ultraviolet circular dichroism. Within experimental variation, all three single replacements exhibit a common thermal transition, and all three double replacements exhibit a different common thermal transition. These results suggest that replacement of the central alanine residue in the repeat EAAAK located in the N-terminus, in the middle, or in the C-terminus of the peptide helix has the same effect on helix stability. The melting temperature of each thermal transition was estimated by assuming a linear van't Hoff plot and a change in molar ellipticity of 33,500 deg cm2 dmol-1. Such analysis indicates that each replacement of an alanine residue by a serine residue diminishes the melting temperature by 11 +/- 1 degrees C and that each replacement of an alanine residue by a methionine residue diminishes the melting temperature by 6 +/- 1 degrees C. These results suggest that the effect of these replacements on helix stability is additive.