Debye ring diffraction elucidation of 2D photonic crystal self-assembly and ordering at the air-water interface.

We fabricate 2D photonic crystals (2DPC) by spreading a dispersion of charged colloidal particles (diameters = 409, 570, and 915 nm) onto the surface of electrolyte solutions using a needle tip flow method. When the interparticle electrostatic interaction potential is large, particles self-assemble into highly ordered hexagonal close packed (hcp) monolayers. Ordered 2DPC efficiently forward diffract monochromatic light to produce a Debye ring on a screen parallel to the 2DPC. The diameter of the Debye ring is inversely proportional to the 2DPC particle spacing, while the Debye ring brightness and thickness depends on the 2DPC ordering. The Debye ring thickness increases as the 2DPC order decreases. The Debye ring ordering measurements of 2DPC attached to glass slides track measurements of the 2D pair correlation function order parameter calculated from SEM micrographs. The Debye ring method was used to investigate the 2DPC particle spacing, and ordering at the air-solution interface of NaCl solutions, and for 2DPC arrays attached to glass slides. Surprisingly, the 2DPC ordering does not monotonically decrease as the salt concentration increases. This is because of chloride ion adsorption onto the anionic particle surfaces. This adsorption increases the particle surface charge and compensates for the decreased Debye length of the electric double layer when the NaCl concentration is below a critical value.

Photochemical incorporation of silver quantum dots in monodisperse silica colloids for photonic crystal applications.

We developed a novel method to fabricate nanocomposite monodisperse SiO2 spheres (approximately 100 nm) containing homogeneously dispersed Ag quantum dots (approximately 2 to 5 nm). The inclusion morphology is controlled through the timing of the photochemical reduction of silver ions during hydrolysis of tetraethoxysilane in a microemulsion. Depending on the timing, Ag quantum dots can be directed to different annuli within the SiO2 spheres, as well as onto the SiO2 sphere surfaces. The embedded Ag quantum dots show a plasmon resonance absorption band at 438 nm. These Ag@SiO2 particles have significant surface charge and readily self-assemble into crystalline colloidal array (CCA) photonic crystals which Bragg-diffract light in the visible region. The magnitude of the plasmon resonance absorption depends on the CCA Bragg diffraction condition. The negative dielectric constant of the silver nanoparticles may be decreasing the silica-silver nanodot composite refractive index below that of the water medium. We may be observing an analogue of the Borrmann effect previously observed in X-ray scattering, where the incident and diffracted electric field standing wave becomes localized in regions of small CCA crystal absorption.

UV Raman studies of peptide conformation demonstrate that betanova does not cooperatively unfold.

We used UV resonance Raman spectroscopy (UVRR) excited within the peptide bond pi --> pi* electronic transitions and within the aromatic amino acid pi --> pi* electronic transitions to examine the temperature dependence of the solution conformation of betanova, a 20-residue beta-sheet polypeptide [Kortemme, T., Ramirez-Alvarado, M., and Serrano, L. (1998) Science 281, 253-256]. The 206.5 nm excited UVRR enhances the amide vibrations and demonstrates that betanova has a predominantly beta-sheet structure between 5 and 82 degrees C. The 229 nm excited UVRR, which probes the tyrosine and tryptophan side chain vibrations, shows an increase in the solvent exposure of the tryptophan side chains as the temperature is increased. Our results are consistent with the existence of an intermediate state similar to that calculated by Bursulaya and Brooks [Bursulaya, B. D., and Brooks, C. L. (1999) J. Am. Chem. Soc. 121, 9947-9951] and exclude the previously proposed two-state cooperative folding mechanism. Betanova's structure appears to be molten globule over the 3-82 degrees C temperature range of our study.

Development of an intelligent polymerized crystalline colloidal array colorimetric reagent.

We have developed a novel colorimetric reagent for the determination of Pb2+, pH, and temperature. This colorimetric reagent consists of a dispersion of approximately 100-microm particles composed of an intelligent polymerized crystalline colloidal array (IPCCA). The IPCCA particles are composed of a hydrogel polymerized around a face-centered cubic (fcc) array of monodisperse, highly charged polystyrene colloidal particles. These IPCCA particles diffract visible light because the (111) planes of the fcc polystyrene colloidal particle array have an approximately 200-nm lattice constant. The IPCCA particles also contain a molecular recognition agent that actuates array volume changes as a result of changes in analyte concentration or temperature. This results in changes in the IPCCA lattice constants, which shifts the wavelength of light diffracted. We report here the use of these sensing materials in a liquid dispersion that can be poured into a sample solution. This diffraction measurement method is analogous to X-ray powder diffraction measurements. The diffraction wavelength is monitored at a defined angle relative to the incident light.

Dihedral psi angle dependence of the amide III vibration: a uniquely sensitive UV resonance Raman secondary structural probe.

UV resonance Raman studies of peptide and protein secondary structure demonstrate an extraordinary sensitivity of the amide III (Am III) vibration and the C(alpha)H bending vibration to the amide backbone conformation. We demonstrate that this sensitivity results from a Ramachandran dihedral psi angle dependent coupling of the amide N-H motion to (C)C(alpha)H motion, which results in a psi dependent mixing of the Am III and the (C)C(alpha)H bending motions. The vibrations are intimately mixed at psi approximately 120 degrees, which is associated with both the beta-sheet conformation and random coil conformations. In contrast, these motions are essentially unmixed for the alpha-helix conformation where psi approximately -60 degrees. Theoretical calculations demonstrate a sinusoidal dependence of this mixing on the psi angle and a linear dependence on the distance separating the N-H and (C)C(alpha)H hydrogens. Our results explain the Am III frequency dependence on conformation as well as the resonance Raman enhancement mechanism for the (C)C(alpha)H bending UV Raman band. These results may in the future help us extract amide psi angles from measured UV resonance Raman spectra.

Transient UV Raman spectroscopy finds no crossing barrier between the peptide alpha-helix and fully random coil conformation.

Transient UV resonance Raman measurements excited within the amide pi --> pi transitions of a 21 unit alpha-helical peptide has for the first time determined a lower bound for the unfolding rate of the last alpha-helical turn to form a fully random coil peptide. A 3 ns T-jump is generated with 1.9 microm laser pulses, which are absorbed by water. Subsequent 3 ns 204 nm UV pulses excite the amide Raman spectra at delay times between 3 ns and 1 ms, to monitor the peptide conformational evolution. We find approximately 180 ns relaxation times which result in a rate constant of >5 x 10(6) s(-1) for unfolding of the last alpha-helical turn. Our data are inconsistent with slow alpha-helix nuclei melting.

UV resonance Raman study of angiotensin II conformation in nonaqueous environments: lipid micelles and acetonitrile.

We used 206.5-nm excited resonance Raman measurements to examine the angiotensin II (AII) secondary structure in H(2)O in the presence of dodecylphosphocholine (DPC) micelles, sodium dodecylsulfate (SDS) monomers and micelles, and in a 70% acetonitrile (ACN-d)-30% water solution. Our AII-SDS titration absorption studies indicate the formation of a 1:2 AII:SDS complex in which two negatively charged SDS molecules attach to the AII positively charged N terminus and to Arg(2). Our 206.5-nm excited Raman results indicate that the 1:2 AII:SDS complexation increases the AII beta-turn composition. We also used 228.9-nm Raman excitation to probe the local solvent accessibility of Tyr(4) (AII) in DPC and SDS micelles. Our Tyr (AII) solvent accessibility studies suggest that the Tyr residue is more exposed to the aqueous environment in SDS micelles than in DPC micelles.

Ultraviolet resonance Raman examination of horse apomyoglobin acid unfolding intermediates.

We have used UV resonance Raman spectroscopy to study the acid-induced denaturation of horse apomyoglobin (apoMb) between pH 7. 0 and 1.8. The 206.5 nm excited Raman spectra are dominated by amide vibrations, which are used to quantitatively determine the apoMb secondary structure. The 229 nm excited Raman spectra are dominated by the Tyr and Trp Raman bands, which are analyzed to examine changes of Tyr and Trp environments and solvent exposures. We observe two partially unfolded apoMb intermediates at pH 4 and pH 2, while we observe only one partially unfolded holoMb intermediate at 2, in which the G and H helices are mainly intact, while the rest of protein is unfolded. This partially unfolded holoMb intermediate at pH 2 is essentially identical to the pH 2 apoMb intermediate. The partially unfolded pH 4 apoMb intermediate is composed of the three folded A, G, and H helices and contains 38% helical structure. The changes in the Trp Raman cross sections during the acid-induced denaturation indicates that Trp 7 is likely to be fully exposed in the apoMb pH 4 intermediate and that the A helix melts with a pKa approximately 3.5.

Ultraviolet Raman examination of the environmental dependence of bombolitin I and bombolitin III secondary structure.

Bombolitin I and III (BI and BIII) are small amphiphilic peptides isolated from bumblebee venom. Although they exist in predominately nonhelical conformations in dilute aqueous solutions, we demonstrate, using UV Raman spectroscopy, that they become predominately alpha-helical in solution at pH > 10, in high ionic strength solutions, and in the presence of trifluoroethanol (TFE) and dodecylphosphocholine (DPC) micelles. In this paper, we examine the effects of electrostatic and hydrophobic interactions that control folding of BI and BIII by systematically monitoring their secondary structures as a function of solution conditions. We determine the BI and BIII secondary structure contents by using the quantitative UV Raman methodology of Chi et al. (1998. Biochemistry. 37:2854-2864). Our findings suggest that the alpha-helix turn in BIII at neutral pH is stabilized by a salt bridge between residues Asp2 and Lys5. This initial alpha-helical turn results in different BI and BIII alpha-helical folding mechanisms observed in high pH and high salt concentrations: BIII folds from its single alpha-helix turn close to its N-terminal, whereas the BI alpha-helix probably nucleates within the C-terminal half. We also used quasielastic light scattering to demonstrate that the BI and BIII alpha-helix formation in 0.2 M Ca(ClO4)2 is accompanied by formation of trimers and hexamers, respectively.

Entropic trapping of macromolecules by mesoscopic periodic voids in a polymer hydrogel.

The separation of macromolecules such as polymers and DNA by means of electrophoresis, gel permeation chromatography or filtration exploits size-dependent differences in the time it takes for the molecules to migrate through a random porous network. Transport through the gel matrices, which usually consist of full swollen crosslinked polymers, depends on the relative size of the macromolecule compared with the pore radius. Sufficiently small molecules are thought to adopt an approximately spherical conformation when diffusing through the gel matrix, whereas larger ones are forced to migrate in a snake-like fashion. Molecules of intermediate size, however, can get temporarily trapped in the largest pores of the matrix, where the molecule can extend and thus maximize its conformational entropy. This 'entropic trapping' is thought to increase the dependence of diffusion rate on molecular size. Here we report the direct experimental verification of this phenomenon. Bragg diffraction from a hydrogel containing a periodic array of monodisperse water voids confirms that polymers of different weights partition between the hydrogel matrix and the water voids according to the predictions of the entropic trapping theory. Our approach might also lead to the design of improved separation media based on entropic trapping.

UV resonance Raman-selective amide vibrational enhancement: quantitative methodology for determining protein secondary structure.

We have directly determined the amide band resonance Raman spectra of the "average" pure alpha-helix, beta-sheet, and unordered secondary structures by exciting within the amide pi-->pi* transitions at 206.5 nm. The Raman spectra are dominated by the amide bands of the peptide backbone. We have empirically determined the average pure alpha-helix, beta-sheet, and unordered resonance Raman spectra from the amide resonance Raman spectra of 13 proteins with well-known X-ray crystal structures. We demonstrate that we can simultaneously utilize the amide I, II, and III bands and the Calpha-H amide bending vibrations of these average secondary structure spectra to directly determine protein secondary structure. The UV Raman method appears to be complementary, and in some cases superior, to the existing methods, such as CD, VCD, and absorption spectroscopy. In addition, the spectra are immune to the light-scattering artifacts that plague CD, VCD, and IR absorption measurements. Thus, it will be possible to examine proteins in micelles and other scattering media.

UV resonance Raman determination of protein acid denaturation: selective unfolding of helical segments of horse myoglobin.

We have used UV resonance Raman spectroscopy to study the acid denaturation of horse heart aquometmyoglobin (Mb) between pH 7.5 and 1.5. Raman spectra excited at 206.5 nm are dominated by amide vibrations, which are analyzed by using a new methodology to quantitatively determine the Mb secondary structure. In contrast, the 229-nm Raman spectra are dominated by the Tyr and Trp Raman bands, which are analyzed to examine changes in Tyr and Trp environments, such as exposure to water, hydrogen bonding, and, for Trp, any alterations of the dihedral angle between the Trp ring and its linkage to the protein backbone. We uniquely determined which Mb alpha-helices melt by combining the amide, Tyr, and Trp Raman spectral information with heme absorption spectral information. We calculate that the Mb alpha-helical composition decreases from approximately 80% at neutral pH to approximately 19% below pH 3.5. The Trp Raman cross sections dramatically decrease at low pH to values which indicate that they are fully exposed to water; this result indicates that the A helix melts. The Tyr Raman bands are pH independent, which indicates that the G and H helices around the Tyr residues do not melt. The dramatic heme absorption acid denaturation changes indicate major alterations of the heme pocket and changes in heme binding. These results indicate that the A, B, C, D, E, and F helices melt in a concerted fashion, while the antiparallel G and H helices only partially melt.

Polymerized colloidal crystal hydrogel films as intelligent chemical sensing materials.

Chemical sensors respond to the presence of a specific analyte in a variety of ways. One of the most convenient is a change in optical properties, and in particular a visually perceptible colour change. Here we report the preparation of a material that changes colour in response to a chemical signal by means of a change in diffraction (rather than absorption) properties. Our material is a crystalline colloidal array of polymer spheres (roughly 100 nm diameter) polymerized within a hydrogel that swells and shrinks reversibly in the presence of certain analytes (here metal ions and glucose). The crystalline colloidal array diffracts light at (visible) wavelengths determined by the lattice spacing, which gives rise to an intense colour. The hydrogel contains either a molecular-recognition group that binds the analyte selectively (crown ethers for metal ions), or a molecular-recognition agent that reacts with the analyte selectively. These recognition events cause the gel to swell owing to an increased osmotic pressure, which increases the mean separation between the colloidal spheres and so shifts the Bragg peak of the diffracted light to longer wavelengths. We anticipate that this strategy can be used to prepare 'intelligent' materials responsive to a wide range of analytes, including viruses.

UV resonance Raman and excited-state relaxation rate studies of hemoglobin.

We have measured the UV resonance Raman (UVRR) spectra of human methemoglobin fluoride (metHbF) and examined the Raman saturation behavior of the metHbF trytophyl (Trp) and tyrosyl (Tyr) residues. Our high-quality UVRR spectra devoid of Raman saturation with 229- and 238.3-nm CW laser excitation allow us to determine small changes in Trp and Tyr residue Raman band frequencies and intensities caused by the hemoglobin R-T quaternary structural change induced by the allosteric effector inositol hexaphosphate. At 238.3-nm excitation, we observe a ca. 15 and 8% intensity increase for the Trp and Tyr bands, respectively, upon the R-T transition. In contrast, a small intensity decrease is observed with 225-nm excitation. These intensity alterations result from Trp and Tyr absorption and Raman excitation profile red-shifts which correlate with a strong 231.5-nm R-T absorption spectral change. These absorption and Raman excitation profile red-shifts and our model compound absorption studies together suggest a T-state increase in the hydrogen bond donation of the Trp-beta(2)37 and Tyr-alpha(1)42 residues at the alpha 1 beta 2 subunit interface. The Tyr-alpha 42 residue appears to be a hydrogen bond donor, rather than an acceptor. We determined the electronic excited-state relaxation rates of the Trp and Tyr residues in hemoglobin by using Raman saturation spectroscopy with 225-nm pulsed laser excitation. The observed average excited-state relaxation rate of the Trp residues is ca. 1/120 ps and is independent of the quaternary structure. This rate is slower that that observed for Trp residues of horse myoglobin. The average excited-state relaxation rate of the Tyr residues is ca. 1/60 ps for both the R and T quaternary forms. These are the first Tyr relaxation rates measured for any heme protein.

Vibrational circular dichroism measurements of ligand vibrations in haem and non-haem metalloenzymes.

We have measured the vibrational circular dichroism (VCD) spectra of the stretching vibrations of azide and cyanide ligated to the Fe3+ atoms of haemoglobin (Hb) and myoglobin (Mb). The antisymmetric azide-stretch of the low-spin haems have an anomalously large g-value of ca. -1 x 10(-3). In contrast, CN- has a g-value of ca. +2.4 x 10(-3). We also show, for the first time, that a significant VCD occurs for the azide ligand antisymmetric stretches of non-haem proteins; we measure a g-value of ca. -1 x 10(-4) for azide bound to haemerythrin. We have examined the mechanism of the VCD phenomenon by: (1) reconstituting Mb with haems substituted such that they insert differently in the haem pocket; (2) replacing the Fe3+ with Mn3+; (3) examining proteins where replacements occur for E-7 His and E-11 Val distal amino acids close to the haem and (4) examining an Mb mutant where the proximal F-8 His is replaced by Gly, and where an imidazole ligand inserts into the resulting crevice and binds to the haem in a way similar to that of the proximal histidine in the native protein. The VCD anisotropy appears insensitive to the haem substituent replacements used in this study. Exchange of the E-7 distal His or the E-11 Val has a dramatic effect on the g-value. Exchange of the F-8 proximal His reverses the sign of the g-value for the azide complex, but not for the cyanide complex. The work to date indicates that VCD has the potential to become a sensitive technique for examining the structure of metalloenzymes. Work is needed to determine the mechanism giving rise to the large g-values and to correlate the VCD spectrum with the metalloenzyme structure at the active site.

Ultraviolet resonance Raman and absorption difference spectroscopy of myoglobins: titration behavior of individual tyrosine residues.

The UV resonance Raman spectra of horse and sperm whale myoglobin excited at 240 nm show bands between 600 and 1700 cm-1 which derive from tyrosyl and tryptophyl residues. No significant contribution from phenylalanine and peptide backbone vibrations occurs at this excitation wavelength. We examine the pH dependence of the UV resonance Raman and UV absorption difference spectra of these myoglobins to correlate the local protein environment of the tyrosyl residues as given by the protein crystal structure to their pKa values, molar absorptivities, and Raman cross sections. Some of our pKa values for the tyrosinate residues of horse Mb differ from those of previous studies. We show that the lambda max values, the molar absorptivities, and the Raman cross sections are sensitive to the local environment of the tyrosinate residues in the protein. We relate differences in the tyrosyl absorption spectra to differences in Raman cross sections. In addition, we discuss the importance to the Raman cross sections of the local electromagnetic field enhancement due to the dielectric environment of the tyrosinate residues in the protein. This local field should scale the Raman cross sections in a way useful as a probe of the average aromatic amino acid residue environment.

Internal intensity standards for heme protein UV resonance Raman studies: excitation profiles of cacodylic acid and sodium selenate.

We examine the utility of SO4(2-), ClO4-, cacodylic acid, and SeO4(2-) as internal intensity standards for Raman spectral measurements of protein structure. We find that 0.1 M SO4(2-) and ClO4- perturb the protein tertiary structure of aquomethemoglobin (met-Hb) and its fluoride (met-HbF) and azide (met-HbN3) complexes. Changes occur for the tryptophan near-UV absorption bands, the iron spin state is altered, and the fluoride ligand affinity decreases. Concentrations of ClO4- and SO4(2-) as low as 0.1 M suppress the met-HbF quaternary R----T transition induced by the allosteric effector inositol hexaphosphate (IHP). In contrast, similar concentrations of cacodylic acid and SeO4(2-) show little effect on the hemoglobin tertiary or quaternary protein structures or upon the R----T transition induced by IHP. We measure the Raman cross sections of cacodylic acid and SeO4(2-) between 218 and 514.5 nm and find that for UV excitation they are ca. 5-fold larger than ClO4- or SO4(2-). Thus, cacodylic acid and selenate can be used at lower concentrations. Cacodylic acid and SeO4(2-) are superior Raman internal intensity standards for protein structural studies.

Resonance Raman evidence for tyrosine involvement in the radical site of galactose oxidase.

Resonance Raman data are reported for the redox-activated form of galactose oxidase from Dactylium dendroides. Excitation within the red (659 nm) and blue (457.9 nm) absorption bands leads to strong resonance enhancement of ligated tyrosine vibrational modes at 550, 1170, 1247, 1484, and 1595 cm-1. The ring mode frequencies are unusually low, indicating a decreased bond order in the ring. The spectra clearly differ in both frequencies and relative intensities from those characteristic of known aromatic pi-radicals. Enhancement of tyrosine ring modes on excitation within absorption bands previously associated with the presence of the radical in the active site suggests that the ligated tyrosine residue is present in the radical site and may stabilize this radical species through formation of a charge transfer complex. A dramatically different Raman spectrum is observed for the N3- adduct of galactose oxidase, exhibiting a single strong 1483 cm-1 feature. The intense visible-near IR absorption bands for galactose oxidase may derive from transitions within a charge transfer complex between an aromatic free radical and a tyrosine-copper complex.