Unwinding of the Substrate Transmembrane Helix in Intramembrane Proteolysis.

Intramembrane-cleaving proteases (I-CLiPs) activate pools of single-pass helical membrane protein signaling precursors that are key in the physiology of prokaryotic and eukaryotic cells. Proteases typically cleave peptide bonds within extended or flexible regions of their substrates, and thus the mechanism underlying the ability of I-CLiPs to hydrolyze the presumably α-helical transmembrane domain (TMD) of these membrane proteins is unclear. Using deep-ultraviolet resonance Raman spectroscopy in combination with isotopic labeling, we show that although predominantly in canonical α-helical conformation, the TMD of the established I-CLiP substrate Gurken displays 310-helical geometry. As measured by microscale thermophoresis, this substrate binds with high affinity to the I-CLiPs GlpG rhomboid and MCMJR1 presenilin homolog in detergent micelles. Binding results in deep-ultraviolet resonance Raman spectra, indicating conformational changes consistent with unwinding of the 310-helical region of the substrate's TMD. This 310-helical conformation is key for intramembrane proteolysis, as the substitution of a single proline residue in the TMD of Gurken by alanine suppresses 310-helical content in favor of α-helical geometry and abolishes cleavage without affecting binding to the I-CLiP. Complemented by molecular dynamics simulations of the TMD of Gurken, our vibrational spectroscopy data provide biophysical evidence in support of a model in which the transmembrane region of cleavable I-CLiP substrates displays local deviations in canonical α-helical conformation characterized by chain flexibility, and binding to the enzyme results in conformational changes that facilitate local unwinding of the transmembrane helix for cleavage.

Hydrogen bonds are a primary driving force for de novo protein folding. Corrigendum.

The paper by Lee et al. [(2017). Acta Cryst. D73, 955-969] is withdrawn.

Hydrogen bonds are a primary driving force for de novo protein folding.

The protein-folding mechanism remains a major puzzle in life science. Purified soluble activation-induced cytidine deaminase (AID) is one of the most difficult proteins to obtain. Starting from inclusion bodies containing a C-terminally truncated version of AID (residues 1-153; AID153), an optimized in vitro folding procedure was derived to obtain large amounts of AID153, which led to crystals with good quality and to final structural determination. Interestingly, it was found that the final refolding yield of the protein is proline residue-dependent. The difference in the distribution of cis and trans configurations of proline residues in the protein after complete denaturation is a major determining factor of the final yield. A point mutation of one of four proline residues to an asparagine led to a near-doubling of the yield of refolded protein after complete denaturation. It was concluded that the driving force behind protein folding could not overcome the cis-to-trans proline isomerization, or vice versa, during the protein-folding process. Furthermore, it was found that successful refolding of proteins optimally occurs at high pH values, which may mimic protein folding in vivo. It was found that high pH values could induce the polarization of peptide bonds, which may trigger the formation of protein secondary structures through hydrogen bonds. It is proposed that a hydrophobic environment coupled with negative charges is essential for protein folding. Combined with our earlier discoveries on protein-unfolding mechanisms, it is proposed that hydrogen bonds are a primary driving force for de novo protein folding.

Insights into the aggregation mechanism of Aß(25-40).

The hydrophobic fragment of the Alzheimer's related ß-amyloid (Aß) peptide, Aß(25-40), aggregates and forms insoluble amyloid fibrils at a rate similar to the full-length peptide. In order to gain insight into the fibrillization of Aß(25-40) and the ability of the flavonoid myricetin to inhibit its aggregation, the isoleucine at position 32 (I32A) and the glycine at position 37 (G37A) in the full-length peptide were replaced with alanine. Thioflavin T assays indicate that substitution of isoleucine for alanine significantly reduces the rate and extent of fibrillization compared to the Aß(25-40) and G37A peptides. Although all three peptides are fully disordered initially, circular dichroism studies suggest the structure of the I32A and G37A peptides are different from the parent peptide Aß(25-40). Introduction of myricetin to the peptide samples results in modest structural changes for the Aß(25-40) and G37A peptides but not the I32A peptide. Aß(25-40) oligomers were predominantly tetramers, whereas I32A and G37A oligomers were a mixture of trimers and dimers. After 48h of incubation at 37°C, the amount of tetramers and trimers in solution dropped for the Aß(25-40) and G37A peptides but remained similar for the I32A peptide. Incubation of Aß(25-40) with myricetin increased the relative proportion of trimers to tetramers. Ultraviolet resonance Raman studies suggests that the I32A peptide may be more hydrated than the Aß(25-40) and G37A peptides. Taken together, these data indicate the structural changes observed for the Aß(25-40) and G37A peptides upon introduction of myricetin are localized around residue 32 and could arise from hydrophobic interactions between the peptide and the flavonoid or interference with the self-association of the peptide in this region. Substitution of isoleucine at position 32 with alanine had little effect on the peptide's secondary structure but dramatically decreased the propensity of the peptide fibrillize.

"Parallel factor analysis of multi-excitation ultraviolet resonance Raman spectra for protein secondary structure determination".

Protein secondary structural analysis is important for understanding the relationship between protein structure and function, or more importantly how changes in structure relate to loss of function. The structurally sensitive protein vibrational modes (amide I, II, III and S) in deep-ultraviolet resonance Raman (DUVRR) spectra resulting from the backbone C-O and N-H vibrations make DUVRR a potentially powerful tool for studying secondary structure changes. Experimental studies reveal that the position and intensity of the four amide modes in DUVRR spectra of proteins are largely correlated with the varying fractions of α-helix, ß-sheet and disordered structural content of proteins. Employing multivariate calibration methods and DUVRR spectra of globular proteins with varying structural compositions, the secondary structure of a protein with unknown structure can be predicted. A disadvantage of multivariate calibration methods is the requirement of known concentration or spectral profiles. Second-order curve resolution methods, such as parallel factor analysis (PARAFAC), do not have such a requirement due to the "second-order advantage." An exceptional feature of DUVRR spectroscopy is that DUVRR spectra are linearly dependent on both excitation wavelength and secondary structure composition. Thus, higher order data can be created by combining protein DUVRR spectra of several proteins collected at multiple excitation wavelengths to give multi-excitation ultraviolet resonance Raman data (ME-UVRR). PARAFAC has been used to analyze ME-UVRR data of nine proteins to resolve the pure spectral, excitation and compositional profiles. A three factor model with non-negativity constraints produced three unique factors that were correlated with the relative abundance of helical, ß-sheet and poly-proline II dihedral angles. This is the first empirical evidence that the typically resolved "disordered" spectrum represents the better defined poly-proline II type structure.

Effects of fluidity on the ensemble structure of a membrane embedded α-helical peptide.

Melittin, the main hemolytic component of honeybee venom, is unfolded in an aqueous environment and folds into an α-helical conformation in a lipid environment. Membrane fluidity is known to affect the activity and structure of melittin. By combining two structurally sensitive optical methods, circular dichroism (CD) and deep-ultraviolet resonance Raman spectroscopy (dUVRR), we have identified distinct structural fluctuations in melittin correlated with increased and decreased 1,2-dimyristoyl-sn-glycero-3-phosphocholine bilayer fluidities. CD spectra have reduced intensity at temperatures above 22°C and high concentrations of the cholesterol analog 5α-cholestan-3ß-ol indicating distortions in the α-helical structure under these conditions. No increase in the amide S is observed in the temperature-dependent dUVRR spectra, suggesting an increase in 310 -helical structure with increasing temperatures above 22°C. However, incorporation of 25 mol% 5α-cholestan-3ß-ol resulted in a small increase in the amide S intensity indicating partial unfolding of melittin.

Role of bilayer characteristics on the structural fate of aß(1-40) and aß(25-40).

The ß-amyloid (Aß) peptide is derived from the transmembrane (TM) helix of the amyloid precursor protein (APP) and has been shown to interact with membrane surfaces. To understand better the role of peptide-membrane interactions in cell death and ultimately in Alzheimer's disease, a better understanding of how membrane characteristics affect the binding, solvation, and secondary structure of Aß is needed. Employing a combination of circular dichroism and deep-UV resonance Raman spectroscopies, Aß(25-40) was found to fold spontaneously upon association with anionic lipid bilayers. The hydrophobic portion of the disease-related Aß(1-40) peptide, Aß(25-40), has often been used as a model for how its legacy TM region may behave structurally in aqueous solvents and during membrane encounters. The structure of the membrane-associated Aß(25-40) peptide was found to depend on both the hydrophobic thickness of the bilayer and the duration of incubation. Similarly, the disease-related Aß(1-40) peptide also spontaneously associates with anionic liposomes, where it initially adopts mixtures of disordered and helical structures. The partially disordered helical structures then convert to ß-sheet structures over longer time frames. ß-Sheet structure is formed prior to helical unwinding, implying a model in which ß-sheet structure, formed initially from disordered regions, prompts the unwinding and destabilization of membrane-stabilized helical structure. A model is proposed to describe the mechanism of escape of Aß(1-40) from the membrane surfaces following its formation by cleavage of APP within the membrane.

Bilayer surface association of the pHLIP peptide promotes extensive backbone desolvation and helically-constrained structures.

Despite their presence in many aspects of biology, the study of membrane proteins lags behind that of their soluble counterparts. Improving structural analysis of membrane proteins is essential. Deep-UV resonance Raman (DUVRR) spectroscopy is an emerging technique in this area and has demonstrated sensitivity to subtle structural transitions and changes in protein environment. The pH low insertion peptide (pHLIP) has three distinct structural states: disordered in an aqueous environment, partially folded and associated with a lipid membrane, and inserted into a lipid bilayer as a transmembrane helix. While the soluble and membrane-inserted forms are well characterized, the partially folded membrane-associated state has not yet been clearly described. The amide I mode, known to be sensitive to protein environment, is the same in spectra of membrane-associated and membrane-inserted pHLIP, indicating comparable levels of backbone dehydration. The amide S mode, sensitive to helical structure, indicates less helical character in the membrane-associated form compared to the membrane-inserted state, consistent with previous findings. However, the structurally sensitive amide III region is very similar in both membrane-associated and membrane-inserted pHLIP, suggesting that the membrane-associated form has a large amount of ordered structure. Where before the membrane-associated state was thought to contain mostly unordered structure and reside in a predominantly aqueous environment, we have shown that it contains a significant amount of ordered structure and rests deeper within the lipid membrane.

Observation of persistent α-helical content and discrete types of backbone disorder during a molten globule to ordered peptide transition via deep-UV resonance Raman spectroscopy.

The molten globule state can aide in the folding of a protein to a functional structure and is loosely defined as an increase in structural disorder with conservation of the ensemble secondary structure content. Simultaneous observation of persistent secondary structure content with increased disorder has remained experimentally problematic. As a consequence, modeling how the molten globule state remains stable and how it facilitates proper folding remains difficult due to a lack of amenable spectroscopic techniques to characterize this class of partially unfolded proteins. Previously, deep-UV resonance Raman (dUVRR) spectroscopy has proven useful in the resolution of global and local structural fluctuations in the secondary structure of proteins. In this work, dUVRR was employed to study the molten globule to ordered transition of a model four-helix bundle protein, HP7. Both the average ensemble secondary structure and types of local disorder were monitored, without perturbation of the solvent, pH, or temperature. The molten globule to ordered transition is induced by stepwise coordination of two heme molecules. Persistent dUVRR spectral features in the amide III region at 1295-1301 and 1335-1338 cm-1 confirm previous observations that HP7 remains predominantly helical in the molten globule versus the fully ordered state. Additionally, these spectra represent the first demonstration of conserved helical content in a molten globule protein. With successive heme binding significant losses are observed in the spectral intensity of the amide III3 and S regions (1230-1260 and 1390 cm-1, respectively), which are known to be sensitive to local disorder. These observations indicate that there is a decrease in the structural populations able to explore various extended conformations, with successive heme binding events. DUVRR spectra indicate that the first heme coordination between two helical segments diminishes exploration of more elongated backbone structural conformations in the inter-helical regions. A second heme coordination by the remaining two helices further restricts protein motion.

Influence of the lipid environment on valinomycin structure and cation complex formation.

Carrier-type molecular ionophores, such as the cyclic dodecadepsipeptide valinomycin, often must undergo structural changes during the binding and transport of a cation across the lipid membrane. Observing the structural fluctuations that occur during this process experimentally has proven extremely difficult due to the complexities of spectroscopic analysis of protein structure/dynamics in native lipid bilayer environments. Currently, our understanding of how valinomycin selectively transports ions across membranes is derived from atomic structures solved of the cyclic macromolecule solvated in various organic solvents and complimentary in silico dynamics experiments. We have shown recently that deep-UV excited resonance Raman spectroscopy (DUVRR) has a unique ability to characterize secondary structure content and simultaneously provide information about the relative solvation of the probed peptide backbone C.M. Halsey, J. Xiong, O. Oshokoya, J.A. Johnson, S. Shinde, J.T. Beatty, G. Ghirlanda, R.D. JiJi, J.W. Cooley, Simultaneous observation of peptide backbone lipid solvation and a-helical structure by deep-UV resonance Raman spectroscopy, ChemBioChem 12 (2011) 2125-2128, [16]. Interpretation of DUVRR spectra of valinomycin in swelled lipid and unilamellar lipid bilayer environments indicate that the uncomplexed valinomycin molecule dynamically samples both the open and closed conformations as described for the structures derived from polar and non-polar organic solvents, respectively. Upon introduction of potassium, the structure of valinomycin in swelled lipid environments resembles more closely that of the open conformation. The shift in structure upon complexation is accompanied by a significant decrease in the valinomycin DUVRR spectral amide I intensity, indicating that the open conformation is more water solubilized and is seemingly "trapped" or predominantly located close to the lipid-water interface. The trapping of the valinomycin in the act of complex of potassium at the bilayer-solvent interface and its analysis by DUVRR represents the first spectroscopic description of this state. Conversely, an opposite trend is observed in the amide I intensity upon potassium complexation in unilamellar (or extruded) vesicles, implying the predominant conformation upon potassium binding in native bilayers is one where the peptide backbone of valinomycin is desolvated as would be expected if the molecule were more readily able to traverse a bilayer interior. Interpretation of the DUVRR spectral features is also consistent with the loss or formation of hydrogen bonds observed in the open and closed structures, respectively. Valinomycin must then sample several conformations in the absence of appropriate ions depending upon its locale in the lipid bilayer until potassium causes a greater degree of closure of the open conformer and an increased residency within the more non-polar interior. The potassium induced decreased solubility enables diffusion across the membrane where potassium release can occur by equilibration at the opposite lipid water interface.

Evolution of quantitative methods in protein secondary structure determination via deep-ultraviolet resonance Raman spectroscopy.

Deep-ultraviolet resonance Raman (DUVRR) spectra is sensitive to secondary structural motifs but, similar to circular dichroism (CD) and infrared spectroscopy, requires the application of multivariate and advanced statistical analysis methods to resolve the pure secondary structure Raman spectra (PSSRS) for determination of secondary structure composition. Secondary structure motifs are selectively enhanced by different excitation wavelengths, a characteristic that inspired the first methods for quantifying secondary structures by DUVRR. This review traces the evolution of multivariate methods and their application to secondary structure composition analyses of proteins by DUVRR spectroscopy from the first experiments using two-wavelengths, and culminating with recent studies utilizing time-resolved DUVRR measurements.

Deep-UV resonance Raman analysis of the Rhodobacter capsulatus cytochrome bc1complex reveals a potential marker for the transmembrane peptide backbone.

Classical strategies for structure analysis of proteins interacting with a lipid phase typically correlate ensemble secondary structure content measurements with changes in the spectroscopic responses of localized aromatic residues or reporter molecules to map regional solvent environments. Deep-UV resonance Raman (DUVRR) spectroscopy probes the vibrational modes of the peptide backbone itself, is very sensitive to the ensemble secondary structures of a protein, and has been shown to be sensitive to the extent of solvent interaction with the peptide backbone [ Wang , Y. , Purrello , R. , Georgiou , S. , and Spiro , T. G. ( 1991 ) J. Am. Chem. Soc. 113 , 6368 - 6377 ]. Here we show that a large detergent solubilized membrane protein, the Rhodobacter capsulatus cytochrome bc(1) complex, has a distinct DUVRR spectrum versus that of an aqueous soluble protein with similar overall secondary structure content. Cross-section calculations of the amide vibrational modes indicate that the peptide backbone carbonyl stretching modes differ dramatically between these two proteins. Deuterium exchange experiments probing solvent accessibility confirm that the contribution of the backbone vibrational mode differences are derived from the lipid solubilized or transmembrane α-helical portion of the protein complex. These findings indicate that DUVRR is sensitive to both the hydration status of a protein's peptide backbone, regardless of primary sequence, and its secondary structure content. Therefore, DUVRR may be capable of simultaneously measuring protein dynamics and relative water/lipid solvation of the protein.

Spectroscopic detection of ß -sheet structure in nascent Aß oligomers.

Deep-UV resonance Raman (UVRR) spectroscopy and circular dichroism (CD) were employed to study the secondary structure of Aß(1-42) in fresh samples with increasing fractions of oligomeric peptide. A feature with a minimum at ~217 nm appeared in CD spectra of samples containing oligomeric Aß(1-42). UVRR spectra more closely resembled those of disordered proteins. The primary difference between UVRR spectra was the ratio of the 1236 cm(-1) to 1260 cm(-1) amide III peak intensities, which shifted in favor of the 1236 cm(-1) band as the fraction of oligomeric peptide increased.

Resolution of localized small molecule-Aß interactions by deep-ultraviolet resonance Raman spectroscopy.

The mechanism by which flavonoids prevent formation of amyloid-ß (Aß) fibrils, as well as how they associate with non-fibrillar Aß is still unclear. Fresh, un-oxidized myricetin exhibited excitation and emission fluorescence maxima at 481 and 531 nm, respectively. Introduction of either Aß(1-42) or Aß(25-40) resulted in a fluorescence decrease, when measured at 481 nm, suggesting formation of a myricetin-Aß complex. Circular dichroism (CD) and ultraviolet resonance Raman (UVRR) studies indicate that the association of myricetin with the Aß peptide or its hydrophobic fragment, Aß(25-40), leads to subtle changes in each peptide's conformation. Aß(25-40) formed amyloid fibrils at a similar rate, when compared to the full-length peptide, Aß(1-42), using thioflavin T (ThT) fluorescence. Studies also indicated that myricetin was equally effective at preventing the formation of both Aß(1-42) and Aß(25-40) fibrils. Although ThT assays indicated that Aß(1-16) did not form amyloid fibrils, CD studies of the hydrophilic fragment, Aß(1-16), suggest possible interactions between myricetin and aromatic side chains. UVRR studies of the full-length peptide and Aß(1-16) showed increases in the intensity of the aromatic modes upon introduction of myricetin. Our findings suggest that myricetin interacts with soluble Aß via two mechanisms, association with the hydrophobic C-terminal region and interactions with the aromatic side chains.

Pre-processing of ultraviolet resonance Raman spectra.

The application of UV excitation sources coupled with resonance Raman have the potential to offer information unavailable with the current inventory of commonly used structural techniques including X-ray, NMR and IR analysis. However, for ultraviolet resonance Raman (UVRR) spectroscopy to become a mainstream method for the determination of protein secondary structure content and monitoring protein dynamics, the application of multivariate data analysis methodologies must be made routine. Typically, the application of higher order data analysis methods requires robust pre-processing methods in order to standardize the data arrays. The application of such methods can be problematic in UVRR datasets due to spectral shifts arising from day-to-day fluctuations in the instrument response. Additionally, the non-linear increases in spectral resolution in wavenumbers (increasing spectral data points for the same spectral region) that results from increasing excitation wavelengths can make the alignment of multi-excitation datasets problematic. Last, a uniform and standardized methodology for the subtraction of the water band has also been a systematic issue for multivariate data analysis as the water band overlaps the amide I mode. Here we present a two-pronged preprocessing approach using correlation optimized warping (COW) to alleviate spectra-to-spectra and day-to-day alignment errors coupled with a method whereby the relative intensity of the water band is determined through a least-squares determination of the signal intensity between 1750 and 1900 cm(-1) to make complex multi-excitation datasets more homogeneous and usable with multivariate analysis methods.

MCR-ALS analysis of two-way UV resonance Raman spectra to resolve discrete protein secondary structural motifs.

The ability of ultraviolet resonance Raman (UVRR) spectroscopy to monitor a host of structurally sensitive protein vibrational modes, the amide I, II, III and S regions, makes it a potentially powerful tool for the visualization of equilibrium and non-equilibrium secondary structure changes in even the most difficult peptide samples. However, it is difficult to unambiguously resolve discrete secondary structure-derived UVRR spectral signatures independently of one another as each contributes an unknown profile to each of the spectrally congested vibrational modes. This limitation is compounded by the presence of aromatic side chains, which introduce additional overlapping vibrational modes. To address this, we have exploited an often overlooked tool for alleviating this spectral overlap by utilizing the differential excitability of the vibrational modes associated with alpha-helices and coil moieties, in the deep UV. The differences in the resonance enhancements of the various structurally associated vibrational modes yields an added dimensionality in the spectral data sets making them multi-way in nature. Through a 'chemically relevant' shape-constrained multivariate curve resolution-alternating least squares (MCR-ALS) analysis, we were able to deconvolute the complex amide regions in the multi-excitation UVRR spectrum of the protein myoglobin, giving us potentially useful 'pure' secondary structure-derived contributions to these individual vibrational profiles.

Effects of cyproheptadine and cetirizine on eosinophilic airway inflammation in cats with experimentally induced asthma.

OBJECTIVE: To determine whether oral administration of cyproheptadine or cetirizine blocks the action of serotonin and histamine, respectively, and results in diminished eosinophilic airway inflammation in cats with experimentally induced asthma. ANIMALS: 9 cats in which asthma was experimentally induced through exposure to Bermuda grass allergen (BGA) during a 3-month period. PROCEDURES: Cats were randomized to receive monotherapy with each of 3 treatments for 1 week: placebo (flour in a gelatin capsule, PO, q 12 h), cyproheptadine (8 mg, PO, q 12 h), or cetirizine (5 mg, PO, q 12 h). A 1-week washout period was allowed to elapse between treatments. Prior to and following each 1-week treatment period, blood and bronchoalveolar lavage fluid (BALF) samples were collected. The percentage of eosinophils in BALF was evaluated to determine treatment efficacy. Serum and BALF BGA-specific immunoglobulin contents and plasma and BALF histamine concentrations were determined via ELISAs. Plasma and BALF serotonin concentrations were measured by use of a fluorometric method. RESULTS: The mean +/- SD percentage of eosinophils in BALF did not differ significantly among treatment groups (placebo, 40 +/- 22%; cyproheptadine, 27 +/- 16%; and cetirizine, 31 +/- 20%). Among the treatment groups, BGA-specific immunoglobulin content and histamine and serotonin concentrations were not significantly different. CONCLUSIONS AND CLINICAL RELEVANCE: In cats with experimentally induced asthma, cyproheptadine and cetirizine were not effective in decreasing airway eosinophilic inflammation or in altering several other measured immunologic variables. Neither cyproheptadine nor cetirizine can be advocated as monotherapy for cats with allergen-induced asthma.

Intermediacy of poly(L-proline) II and beta-strand conformations in poly(L-lysine) beta-sheet formation probed by temperature-jump/UV resonance Raman spectroscopy.

Ultraviolet resonance Raman spectroscopy (UVRR) in combination with a nanosecond temperature jump (T-jump) was used to investigate early steps in the temperature-induced alpha-helix to beta-sheet conformational transition of poly(L-lysine) [poly(K)]. Excitation at 197 nm from a tunable frequency-quadrupled Ti:sapphire laser provided high-quality UVRR spectra, containing multiple conformation-sensitive amide bands. Although un-ionized poly(K) (pH 11.6) is mainly alpha-helical below 30 degrees C, there is a detectable fraction (approximately 15%) of unfolded polypeptide, which is mainly in the poly(L-proline) II (PPII) conformation. However, deviations from the expected amide I and II signals indicate an additional conformation, suggested to be beta-strand. Above 30 degrees C un-ionized poly(K) forms a beta-sheet at a rate (minutes) which increases with increasing temperature. A 22-44 degrees C T-jump is accompanied by prompt amide I and II difference signals suggested to arise from a rapid shift in the PPII/beta-strand equilibrium. These signals are superimposed on a subsequently evolving difference spectrum which is characteristic of PPII, although the extent of conversion is low, approximately 2% at the 3 micros time limit of the experiment. The rise time of the PPII signals is approximately 250 ns, consistent with melting of short alpha-helical segments. A model is proposed in which the melted PPII segments interconvert with beta-strand conformation, whose association through interstrand H-bonding nucleates the formation of beta-sheet. The intrinsic propensity for beta-strand formation could be a determinant of beta-sheet induction time, with implications for the onset of amyloid diseases.