Search

Heparan sulfate interacts with a variety of proteins at the cell surface. These proteins are primarily attracted to the high negative charge distribution brought by sulfate, sulfamate, and carboxylate functionalities along the sugar chain. Apart from electrostatic interactions, hydrogen bonding and even hydrophobic interactions contribute to the complex formation. While additional sulfate/sulfamate groups are often tolerated as long as the main structural requirements are met, occasionally, certain extra sulfate groups may be detrimental to the binding affinity. Here, we show these binding characteristics using the binding of fibroblast growth factors and heparin-binding hemagglutinin to synthetic heparan sulfate oligosaccharides as examples. Insights into the binding characteristics of these proteins may benefit future therapeutic interventions., (Copyright © 2018 Elsevier Ltd. All rights reserved.)

Calcineurin (CaN) is a Ca 2+ /calmodulin-dependent Ser/Thr protein phosphatase, which plays essential roles in many cellular and developmental processes. CaN comprises two subunits, a catalytic subunit (CaN-A, 60 kDa) and a regulatory subunit (CaN-B, 19 kDa). CaN-A tightly binds to CaN-B in the presence of minimal levels of Ca 2+ , but the enzyme is inactive until activated by CaM. Upon binding to CaM, CaN then undergoes a conformational rearrangement, the auto inhibitory domain is displaced and thus allows for full activity. In order to elucidate the regulatory role of CaM in the activation processes of CaN, we used NMR spectroscopy to determine the structure of the complex of CaM and the target peptide of CaN (CaNp). The CaM/CaNp complex shows a compact ellipsoidal shape with 8 α-helices of CaM wrapping around the CaNp helix. The RMSD of backbone and heavy atoms of twenty lowest energy structures of CaM/CaNp complex are 0.66 and 1.14 Å, respectively. The structure of CaM/CaNp complex can be classified as a novel binding mode family 1-18 with major anchor residues Ile 396 and Leu 413 to allocate the largest space between two domains of CaM. The relative orientation of CaNp to CaM is similar to the CaMKK peptide in the 1-16 binding mode with N- and C-terminal hydrophobic anchors of target sequence engulfed in the hydrophobic pockets of the N- and C-domain of CaM, respectively. In the light of the structural model of CaM/CaNp complex reported here, we provide new insight in the activation processes of CaN by CaM. We propose that the hydrophobic interactions between the Ca 2+ -saturated C-domain and C-terminal half of the target sequence provide driving forces for the initial recognition. Subsequent folding in the target sequence and structural readjustments in CaM enhance the formation of the complex and affinity to calcium. The electrostatic repulsion between CaM/CaNp complex and AID may result in the displacement of AID from active site for full activity., Competing Interests: The authors declare no conflict of interest.

Heparin-binding hemagglutinin (HBHA) is a 199 amino acid virulence factor at the envelope of Mycobacterium tuberculosis that contributes to latent tuberculosis. The binding of HBHA to respiratory epithelial cells, which leads to extrapulmonary dissemination of the pathogen, is mediated by cell-surface heparan sulfate (HS). We report the structural characterization of the HBHA/HS complex by NMR spectroscopy. To develop a model for the molecular recognition, the first chemically synthesized uniformly 13 C- and 15 N-labeled HS octasaccharide and a uniformly 13 C- and 15 N-labeled form of HBHA were prepared. Residues 180-195 at the C-terminal region of HBHA show large chemical shift perturbation upon association with the octasaccharide. Molecular dynamics simulations conforming to the multidimensional NMR data revealed key electrostatic and even hydrophobic interactions between the binding partners that may aid in the development of agents targeting the binding event., (© 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim.)

A cDNA encoding a cysteine protease inhibitor, cystatin was cloned from sesame (Sesamum indicum L.) seed. This clone was constructed into an expression vector and expressed in E. coli and purified to homogeneous. The recombinant sesame cystatin (SiCYS) showed effectively inhibitory activity toward C1 cysteine proteases. In order to unravel its inhibitory action from structural point of view, multidimensional heteronuclear NMR techniques were used to characterize the structure of SiCYS. The full (1)H, (15)N, and (13)C resonances of SiCYS were assigned. The secondary structure of SiCYS was identified by using the assigned chemical shifts of (1)H(α), (13)C(α), (13)C(β), and (13)CO through the consensus chemical shift index (CSI). The results of CSI analysis of SiCYS suggest eight β-strands (residues 33-46, 51-61, 63-75, 80-87, 150-155, 157-169, 172-183, and 192-195) and two α-helices (residues 16-30, and 120-135).

Human apolipoprotein E (apoE) has been known to play a key role in the transport of plasma cholesterol and lipoprotein metabolism. It is an apolipoprotein of 299 amino acids with a molecular mass, ~34 kDa. ApoE has three major isoforms, apoE2, apoE3, and apoE4 which differ only at residue 112 or 158. ApoE consists of two independently folded domains (N-terminal and C-terminal domain) separated by a hinge region. The N-terminal domain and C-terminal domain of apoE are responsible for the binding to receptor and to lipid, respectively. Since the high resolution structures of apoE in lipids are still unavailable to date, we therefore aim to resolve the structures in lipids by NMR. Here, we reported the resonance assignments and secondary structure distribution of the C-terminal domain of wild-type human apoE (residue 195-299) in the micelles formed by dihexanoylphosphatidylcholine. Our results may provide a novel structural model of apoE in micelles and may shed new light on the molecular mechanisms underlying the apoE related biological processes.

Calmodulin (CaM), the primary receptor for intracellular Ca(2+), regulates a large number of key enzymes and controls a wide spectrum of important biological responses. Olfactory cyclic nucleotide-gated ion channels (OLF channels) mediate olfactory transduction in olfactory receptor neurons. The opening of OLF leads to a rise in cytosolic concentration of Ca(2+), upon binding to Ca(2+), CaM disrupts the open conformation by binding to the CaM-binding domain in the N-terminal region and triggers the close mechanism. In order to unravel the regulatory role of CaM from structural point of view, NMR techniques were used to characterize the structure of CaM in association with the CaM binding domain of rat OLF channel (OLFp, 28 residues). Our data indicated that two distinct CaM/OLFp complexes existed simultaneously with stable structures that were not inter-exchangeable within the NMR time scale. Here, we report the full backbone and side chain resonance assignments of these two complexes of CaM/OLFp.

Cell-penetrating peptides (CPPs) can traverse cellular membranes and deliver biologically active molecules into cells. In this study, we demonstrate that CPPs comprised of nona-arginine (R9) and a penetration accelerating peptide sequence (Pas) that facilitates escape from endocytic lysosomes, denoted as PR9, greatly enhance the delivery of noncovalently associated quantum dots (QDs) into human A549 cells. Mechanistic studies, intracellular trafficking analysis and a functional gene assay reveal that endocytosis is the main route for intracellular delivery of PR9/QD complexes. Endocytic trafficking of PR9/QD complexes was monitored using both confocal and transmission electron microscopy (TEM). Zeta-potential and size analyses indicate the importance of electrostatic forces in the interaction of PR9/QD complexes with plasma membranes. Circular dichroism (CD) spectroscopy reveals that the secondary structural elements of PR9 have similar conformations in aqueous buffer at pH 7 and 5. This study of nontoxic PR9 provides a basis for the design of optimized cargo delivery that allows escape from endocytic vesicles.

Calmodulin (CaM), the primary intracellular Ca(2+) receptor, regulates a large number of key enzymes and controls a wide spectrum of important biological responses. Recognition between CaM and its target sequence in rat olfactory cyclic nucleotide-gated ion channel (OLFp) was investigated by circular dichroism (CD), fluorescence, and NMR spectroscopy. Fluorescence data showed the OLFp tightly bound to CaM with a dissociation constant of 12 nM in a 1:1 stoichiometry. Far-UV CD data showed that approximately 60% of OLFp residues formed α-helical structures when associated with CaM. NMR data showed that most of the (15)N-(1)H HSQC cross-peaks of the (15)N-labeled CaM not only shifted but also split into two sets of peaks upon association with the OLFp. Our data indicated that the two distinct CaM/OLFp complexes existed simultaneously with stable structures that were not interexchangeable within the NMR time scale. In light of the palindromic sequence of OLFp (FQRIVRLVGVIRDW) for CaM targeting, we proposed that the helical OLFp with C2 symmetry may bind to CaM in two orientations. This hypothesis is supported by the observation that only one set of (15)N-(1)H HSQC cross-peaks of the (15)N-labeled CaM was detected upon association with OLFp-M13 chimeric peptide (OLFMp), a mutated OLFp lacking the palindromic feature. The binding specificity of OLFMp to CaM was restored when the palindromic feature was destroyed. Binding modes of CaM/OLFp and CaM/OLFMp simulated by molecular docking were in accord with their distinct patterns observed in HSQC spectra. Our studies suggest that the palindromic residues in OLFp are crucial for the orientation-specific recognition by CaM.

Oil bodies of similar sizes were observed in the cells of embryo and aleurone layer of rice seeds, and remained their structural integrity in vitro after isolation. Comparably, two abundant oleosin isoforms were found in both preparations of oil bodies isolated from the embryo and the aleurone layer. Immunological detection and mass spectrometric analyses indicated that the two oleosin isoforms, termed oleosin-H and oleosin-L, in the embryo and those in the aleurone layer were identical proteins encoded by the same genes (BAF12898.1 and BAF15387.1 for oleosin-H and oleosin-L, respectively). In contrast, one caleosin was found in oil bodies isolated from the embryo but not those isolated from the aleurone layer. Immunological staining of rice seeds confirms that oleosin is present in both embryo and aleurone layer while caleosin exists only in embryo. Caleosin extracted from oil bodies of rice embryo migrated faster on SDS-PAGE in the presence of Ca(2+), in a manner identical to caleosin extracted from sesame oil bodies. Similar to other known monocot caleosins, the rice caleosin possesses an N-terminal appendix that is absent in dicotyledonous caleosins., (Copyright © 2012 Elsevier Masson SAS. All rights reserved.)

Dioscorins, the major storage proteins in yam tubers, exhibit biochemical and immunomodulatroy activities. To investigate the potential application of dioscorins in biomedical research, we expressed the dioscorin genes Dj-dioA3 and Dp-dioA2 from Dioscorea japonica and Dioscorea pseudojaponica, respectively, in E. coli and routinely obtained approximately 15 mg proteins per liter Escherichia coli culture (mg/L) to 30 mg/L of rDj-dioscorinA3 and 4 to 8 mg/L of rDp-dioscorinA2. Western blot analyses revealed that both recombinant dioscorins contained epitopes with similar antigenicities to those of the native dioscorins. Results from dithiothreitol (DTT) treatment followed by monobromobimane (mBBr) staining showed that both recombinant dioscorins, like the native dioscorins, contain an intramolecular disulfide bond between Cys(28) and Cys(187) residues. Circular dichroism spectroscopy findings indicated that the secondary structural contents of the recombinant dioscorins showed high similarity to those of their corresponding native dioscorins. Both recombinant dioscorins, like the native dioscorins, exhibited 1,1-diphenyl-2-picrylhydrazyl (DPPH) radical scavenging and Toll-like receptor 4 signaling activities, and stimulated the phagocytosis of E. coli by macrophage. Overall, our results indicated that substantial amounts of recombinant dioscorins can be purified easily from E. coli and that these recombinant dioscorins are appropriate for application in future investigations of the biomedical functions of dioscorins., (Copyright © 2012 Elsevier Inc. All rights reserved.)

Numerous biomolecules possess α-D-glucosamine as structural component. However, chemical glycosylations aimed at this backbone are usually not easily attained without generating the unwanted β-isomer. We report herein a versatile approach in affording full α-stereoselectivity built upon a carefully selected set of orthogonal protecting groups on a D-glucosaminyl donor. The excellent stereoselectivity provided by the protecting group combination was found independent of leaving groups and activators. With the trichloroacetimidate as the optimum donor leaving group, core skeletons of glycosylphosphatidyl inositol anchors, heparosan, heparan sulfate, and heparin were efficiently assembled. The orthogonal protecting groups were successfully manipulated to further carry out the total syntheses of heparosan tri- and pentasaccharides and heparin di-, tetra-, hexa-, and octasaccharide analogues. Using the heparin analogues, heparin-binding hemagglutinin, a virulence factor of Mycobacterium tuberculosis, was found to bind at least six sugar units with the interaction notably being entropically driven.

A cDNA encoding a cysteine protease inhibitor, cystatin was cloned from pineapple (Ananas comosus L.) stem. This clone was constructed into an expression vector and expressed in E. coli and purified to homogeneous. The recombinant pineapple cystatins (AcCYS) showed effectively inhibitory activity toward cysteine proteases including papain, bromelain, and cathepsin B. In order to unravel its inhibitory action from structural point of view, multidimensional heteronuclear NMR techniques were used to characterize the structure of AcCYS. The full (1)H, (15)N, and (13)C resonance assignments of AcCYS were determined. The secondary structure of AcCYS was identified by using the assigned chemical shift of (1)Hα, (13)Cα, (13)Cβ, and (13)CO through the consensus chemical shift index (CSI). The results of CSI analysis suggest 5 β-strands (residues 45-47, 84-91, 94-104, 106-117, and 123-130) and one α-helix (residues 55-73).

The structure of a recombinant pineapple cystatin (AcCYS) was determined by NMR with the RMSD of backbone and heavy atoms of twenty lowest energy structures of 0.56 and 1.11 Å, respectively. It reveals an unstructured N-terminal extension and a compact inhibitory domain comprising a four-stranded antiparallel β-sheet wrapped around a central α-helix. The three structural motifs (G(45), Q(89)XVXG, and W(120)) putatively responsible for the interaction with papain-like proteases are located in one side of AcCYS. Significant chemical shift perturbations in two loop regions, residues 45 to 48 (GIYD) and residues 89 to 91 (QVV), of AcCYS strongly suggest their involvement in the binding to papain, consistent with studies on other members of the cystatin family. However, the highly conserved W120 appears not to be involved in the binding with papain as no chemical shift perturbation was observed. Chemical shift index analysis further indicates that the length of the α-helix is shortened upon association with papain. Collectively, our data suggest that AcCYS undergoes local secondary structural rearrangements when papain is brought into close contact. A molecular model of AcCYS/papain complex is proposed to illustrate the interaction between AcCYS and papain, indicating a complete blockade of the catalytic triad by AcCYS.

Inhibitor-1 is an acid- and heat-stable protein. It can be turned into a potent inhibitor of protein phosphatase-1 (PP1) after phosphorylation at Thr35 by c-AMP-dependent protein kinase (PKA). Although it has been known that pre-phosphorylation is essential for inhibition of PP1, the structure-function relationship of Thr(35)-phosphorylated inhibitor-1, such as whether or not PKA-phosphorylation pre-triggers conformational changes in inhibitor-1, remains unclear. In this study, we performed structural characterization of Thr(35)-phosphoroylated inhibitor-1 by using multi-dimensional heternuclear NMR spectroscopy. The result of structural comparison between Thr(35)-phosphoroylated and non-phosphorylated inhibitor-1 indicated that PKA-phosphorylation has no significant effect on the global conformation of free-state inhibitor-1. This finding may support the inference that regulation of the interactions between inhibitor-1 and PP1 through PKA-phosphorylation mainly depends on the phosphate group instead of phosphorylation-induced conformational change.

Oil bodies were observed in cells of both embryo and aleurone layers of mature adlay grains (Coix lachryma-jobi L. var. ma-yuen Stapf). Stable oil bodies were successfully isolated from the adlay grains. Thin-layer chromatography revealed that the contents stored in the adlay oil bodies were mainly neutral lipids (>90% triacylglycerols and about 5% diacylglycerols). The integrity of the isolated oil bodies was presumably maintained via electronegative repulsion and steric hindrance provided by their surface proteins. Immunological cross-recognition using antibodies against sesame oil-body proteins indicated that two oleosin isoforms (termed oleosin-H and oleosin-L) and one caleosin were present in the adlay oil bodies. Full-length cDNA fragments encoding these three unique oil-body proteins were obtained by PCR cloning. MALDI-MS analyses confirmed that the three full-length cDNA fragments encoded the two oleosin isoforms and one caleosin observed in the oil bodies isolated from the adlay grains.

Caleosin is a unique calcium binding protein anchoring to the surface of seed oil bodies by its central hydrophobic domain composed of an amphiphatic alpha-helix and a proline-knot subdomain. Stable artificial oil bodies were successfully constituted with recombinant caleosin overexpressed in Escherichia coli. The stability of artificial oil bodies was slightly or severely reduced when the amphiphatic alpha-helix or proline-knot subdomain in the hydrophobic domain of caleosin was truncated. Deletion of the entire central hydrophobic domain substantially increased the solubility of the recombinant caleosin, leading to a complete loss of its capability to stabilize these oil bodies. A recombinant protein engineered with the hydrophobic domain of caleosin replaced by that of oleosin, the abundant structural protein of seed oil bodies, could stabilize the artificial oil bodies, in terms of thermo- and structural stability, as effectively as caleosin or oleosin.

Macromolecular crowding, a common phenomenon in the cellular environments, can significantly affect the thermodynamic and kinetic properties of proteins. A single-molecule method based on atomic force microscopy (AFM) was used to investigate the effects of macromolecular crowding on the forces required to unfold individual protein molecules. It was found that the mechanical stability of ubiquitin molecules was enhanced by macromolecular crowding from added dextran molecules. The average unfolding force increased from 210 pN in the absence of dextran to 234 pN in the presence of 300 g/L dextran at a pulling speed of 0.25 microm/sec. A theoretical model, accounting for the effects of macromolecular crowding on the native and transition states of the protein molecule by applying the scaled-particle theory, was used to quantitatively explain the crowding-induced increase in the unfolding force. The experimental results and interpretation presented could have wide implications for the many proteins that experience mechanical stresses and perform mechanical functions in the crowded environment of the cell.

Jelly curd used for a popular summer drink in Taiwan is prepared by extracting the pericarpial portion of jelly fig (Ficus awkeotsang Makino) achenes. The two most abundant proteins found in jelly curd have been identified as a pectin methylesterase and a chitinase. A method was developed to purify the next abundant protein by 40% ammonium sulfate precipitation and flowing through Mono Q chromatography. In sodium dodecyl sulfate-polyacrylamide gel electrophoresis analyses, the purified protein migrated as a polypeptide of 20 kDa in the absence of beta-mercaptoethanol but split into a minor polypeptide of 20 kDa and a major polypeptide of 27 kDa in the presence of this reducing agent. Two cDNA fragments encoding precursor polypeptides of two putative thaumatin-like protein isoforms were obtained by polymerase chain reaction cloning and subsequently overexpressed in Escherichia coli to generate recombinant proteins for antibody preparations. Immunological detection and mass spectrometric analyses indicated that the two split polypeptides were thaumatin-like protein isoforms encoded by the two cloned cDNA fragments.

Aggregated beta-amyloid (Abeta) peptides are neurotoxic and cause neuronal death both in vitro and in vivo. Although the formation of a beta-sheet structure is usual required to form aggregates, the relationship between neurotoxicity and the Abeta sequence remains unclear. To explore the correlation between Abeta sequence, secondary structure, aggregative ability, and neurotoxicity, we utilized both full-length and fragment-truncated Abeta peptides. Using a combination of spectroscopic and cellular techniques, we demonstrated that neurotoxicity and aggregative ability are correlated while the relationship between these characteristics and secondary structure is not significant. The hydrophobic C-terminus, particularly the amino acids of 17-21, 25-35, and 41-42, is the main region responsible for neurotoxicity and aggregation. Deleting residues 17-21, 25-35 or 41-42 significantly reduced the toxicity. On the other hand, truncation of the peptides at either residues 22-24 or residues 36-40 had little effect on toxicity and aggregative ability. While the N-terminal residues 1-16 may not play a major role in neurotoxicity and aggregation, a lack of N-terminal fragment Abeta peptide, (e.g. Abeta17-35), does not display the neurotoxicity of either full-length or 17-21, 25-35 truncated Abeta peptides.

The conformational change and associated aggregation of beta amyloid (Abeta) with or without metals is the main cause of Alzheimer's disease (AD). In order to further understand the effects of Abeta and its associated metals on the aggregation mechanism, the influence of Abeta conformation on the metal affinity and aggregation was investigated using circular dichroism (CD) spectroscopy. The Abeta conformation is dependent on pH and trifluoroethanol (TFE). The binding of metals to Abeta was found to be dependent on the Abeta conformation. The aggregation induced by Abeta itself or its associated metals is completely diminished for Abeta in 40% TFE. Only in 5% and 25% TFE can Abeta undergo an alpha-helix to beta-sheet aggregation, which involve a three-state mechanism for the metal-free state, and a two-state transition for the metal-bound state, respectively. The aggregation-inducing activity of metals is in the order, Cu2+ > Fe3+ > or = Al3+ > Zn2+.

A cDNA clone, SiMT encoding an Ec type of metallothionein (MT)-like protein, was isolated from maturing seeds of sesame (Sesamum indicum L.), and its deduced protein sequence shared 47-65% similarity to other known Ec type of MT-like proteins with three highly conserved cysteine-rich segments. The transcript of SiMT was exclusively accumulated in maturing seeds from two weeks after flowering to the end of seed maturation. The results of a southern blot analysis suggested that one SiMT and one SiMT-like gene were present in the sesame genome. Recombinant SiMT fused with glutathione-S-transferase (GST) was over-expressed in Escherichia coli, and purified to homogeneity by affinity chromatography. Recombinant SiMT released from GST was harvested after factor Xa cleavage. Migration of the recombinant SiMT during SDS-PAGE was accelerated when its binding metal ions were depleted by EDTA. The metal-binding capability of recombinant SiMT was measured by inductively-coupled plasma atomic emission spectrometry. Our results show that the recombinant SiMT could trap zinc or copper ions, but not manganese ions, with a stoichiometric ratio (metal ion/SiMT) of approximately 2.

A hybrid molecule consisting of calmodulin (CaM) and the CaM-binding domain of olfactory nucleotide-gated ion-channel peptide (CaM-OLFp) was purified and crystallized by the hanging-drop vapour-diffusion method at 298 K. The crystals diffracted to a maximum resolution of 1.85 A at cryogenic temperature (100 K) using X-rays from a rotating anode (Cu, wavelength 1.54 A). The crystal belongs to the monoclinic space group C2, with unit-cell parameters a = 64.76, b = 36.23, c = 70.96 A, alpha = gamma = 90, beta = 109.4 degrees. Analysis of the packing density shows that the asymmetric unit contains one CaM-OLFp hybrid molecule with a solvent content of 36.42%.

Single-molecule manipulation techniques have enabled the characterization of the unfolding and refolding process of individual protein molecules, using mechanical forces to initiate the unfolding transition. Experimental and computational results following this approach have shed new light on the mechanisms of the mechanical functions of proteins involved in several cellular processes, as well as revealed new information on the protein folding/unfolding free-energy landscapes. To investigate how protein molecules of different folds respond to a stretching force, and to elucidate the effects of solution conditions on the mechanical stability of a protein, we synthesized polymers of the protein ubiquitin and characterized the force-induced unfolding and refolding of individual ubiquitin molecules using an atomic-force-microscope-based single-molecule manipulation technique. The ubiquitin molecule was highly resistant to a stretching force, and the mechanical unfolding process was reversible. A model calculation based on the hydrogen-bonding pattern in the native structure was performed to explain the origin of this high mechanical stability. Furthermore, pH effects were studied and it was found that the forces required to unfold the protein remained constant within a pH range around the neutral value, and forces decreased as the solution pH was lowered to more acidic values.

A cDNA fragment encoding the cysteine protease inhibitor, cystatin, was cloned from pineapple (Ananas comosus) stem. This clone was constructed in a fusion vector and was easily over-expressed in Escherichia coli; satisfactory over-expression of non-fusion cystatin was achieved after an additional start codon was inserted prior to its coding sequence. Both recombinant cystatins were predominately found in the soluble fraction of the cell extract, and were demonstrated to be functionally active in a reverse zymographic assay. The fusion and non-fusion cystatins were separately purified to homogeneity via a His-tag or papain-coupling affinity column. Effective inhibitory activity against papain was detected with both the fusion and non-fusion cystatins with comparable K(i) values of 1.18 x 10(-10) M and 9.53 x 10(-11) M, respectively. The recombinant cystatins were found to be thermally stable up to 60 degrees C. Inhibition of the endogenous protease activity in minced fish muscle revealed that the recombinant pineapple cystatins might be an adequate stabilizer to prevent protein degradation during industrial food processing.

Seed oil bodies are lipid storage organelles of 0.5-2 microm in diameter and comprise a triacylglycerol matrix shielded by a monolayer of phospholipids and proteins. These proteins include abundant structural proteins, oleosins, and at least two minor proteins termed caleosin and steroleosin. This study examined if artificial oil bodies (AOBs) composed of triacylglycerol and phospholipid could be stabilized by oleosin, caleosin, or steroleosin. Our results showed that stabilization effects could be realized by oleosin or caleosin but not by steroleosin. The sizes of the AOBs constituted with oleosin (0.5-2 microm) or caleosin (50-200 nm) were similar to or 10 times smaller than those of the native oil bodies. Recombinant caleosin expressed in Escherichia coli also encapsulated AOBs with a size, topology, and stability comparable to those encapsulated with native caleosin. A proteinase K digestion indicated that caleosin anchored the AOBs via its central hydrophobic domain of approximately 4 kDa. Isoelectrofocusing revealed that the isoelectric point of the caleosin-stabilized AOBs was pH 4.0. Aggregation of AOBs was observed at a pH lower than 4.5; thus, their stability and integrity were presumably contributed by surface caleosin via electronegative repulsion and steric hindrance. The caleosin-stabilized AOBs were thermostable up to 70 degrees C and potentially useful for biotechnological applications.

Inhibitor-2 (I2) is a thermostable protein that specifically binds to the catalytic subunit of protein phosphatase-1 (PP1), resulting in the formation of the inactive holoenzyme, ATP-Mg-dependent phosphatase. Phosphorylation of I2 at Thr-72 by glycogen synthase kinase-3 (GSK-3) results in activation of the phosphatase, suggesting that kinase action triggers conformational change in the complex. In this paper, we characterize the effect of GSK-3 phosphorylation on the structure of free state I2[1-172] by nuclear magnetic resonance and circular dichroism spectroscopy, and show that phosphorylation has no significant effect on its conformation. We conclude that the conformational changes of ATP-Mg-dependent phosphatase induced by GSK-3 phosphorylation must depend on the interactions between PP1 and I2.

A partially folded state of guinea pig alpha-lactalbumin (the A-state or molten globule state), formed by denaturation at low pH, has been studied using hydrogen exchange methods. The overall distribution of exchange kinetics, measured by 1-D NMR, suggests that fewer than 20 amides in the structure are involved in highly persistent residual structure, although CD results suggest that many other parts of the chain are folded, for a significant proportion of the time, into less stable structural elements. The pH-jump experiments show that some amides that are strongly protected from exchange in the native state become freely accessible in the A-state but that conversely a majority, at least, of those that are slow to exchange in the A-state retain that protection in the native state. This suggests that the persistent structure in the A-state is native-like although the possibility that nonnative like structural elements persist cannot be eliminated. Resonance assignments for key residues in the NMR spectrum of the native state have enabled us to use the pH-jump method also to identify the majority of the most protected amides in the A-state: they are located in two hydrophobic segments, corresponding to the B- and C-helices of the native structure. This strongly suggests that the most persistent structure of the A-state includes these regions. A variety of lines of evidence, including fluorescence quenching data and, most remarkably, very effective protection from exchange of an indole NH in a tryptophan side chain, suggest that some form of hydrophobic core in the helical domain of the native structure persists in the A-state, although without the stereochemical rigidity of the native tertiary structure. The other domain of the native structure, including the beta-sheet, appears not to contain structural elements which persist to the same extent in the A-state, emphasizing that the molten globule is highly heterogeneous, in terms of the stability and specificity of both backbone and side chain interactions.

1H-NMR assignments have been defined for the aromatic-ring protons of the bovine, guinea pig and human variants of alpha-lactalbumin. Spin-system networks were identified by means of double-quantum-filtered two-dimensional J-correlated spectroscopy and two-dimensional relayed coherence spectroscopy data. Analysis of two-dimensional nuclear-Overhauser-enhancement spectroscopy data of the proteins indicated that in each case two clusters of aromatic residues exist. The two clusters are also evident in the crystal structure of the human protein, and this evidence, in conjunction with sequence differences between the three proteins, permitted sequence-specific assignments to be made for the majority of aromatic residues. Remaining ambiguities in the assignments could be resolved by analysis of photochemically induced dynamic nuclear polarization (PCIDNP) effects. Comparison of the PCIDNP spectra of the three proteins indicated the presence of only minor differences in the surface exposure of conserved aromatic residues. Taken together, these results indicate that the environments of the conserved aromatic residues in bovine, guinea pig and human alpha-lactalbumin in solution are very similar to each other, and that the solution and the crystal forms of at least the human protein are similar.