Dietary intake of antioxidant vitamins and risk of stroke: the Japan Public Health Center-based Prospective Study.

BACKGROUND/OBJECTIVES: Epidemiologic evidence on the relationship between antioxidant vitamin intake and stroke is limited. We aimed to investigate the association between dietary intake of antioxidant vitamins and the incidence of total stroke and ischemic stroke. SUBJECTS/METHODS: The subjects were 82 044 Japanese men and women aged 45-74 years under the Japan Public Health Center-based Prospective Cohort Study. Between 1995 and 1997, dietary assessment was done using a food frequency questionnaire. During 983 857 person-years of follow-up until the end of 2009 we documented 3541 incident total strokes and 2138 ischemic strokes. RESULTS: Dietary intakes of α-carotene, ß-carotene, α-tocopherol and vitamin C were not inversely associated with the incidence of total stroke and ischemic stroke adjustment for cardiovascular risk factors and selected lifestyle variables. When stratified by current smoking status, the inverse association between dietary vitamin C intake and incidence of total stroke observed among non-smokers but not smokers, with respective multivariable hazard ratios for the highest versus lowest quintiles of vitamin C of 0.81 (95% confidence interval (CI), 0.68-0.96; P-trend=0.03) among non-smokers; and 1.03 (0.84-1.25; P-trend=0.55) among smokers. As for ischemic stroke, the corresponding multivariable hazard ratios were 0.76 (0.60-0.96; P-trend=0.02) among non-smokers; and 1.00 (0.78-1.28; P-trend=0.61) among smokers. CONCLUSIONS: Dietary vitamin C intake was inversely associated with the incidence of total stroke and ischemic stroke among non-smokers.

Structure, backbone dynamics and interactions with RNA of the C-terminal RNA-binding domain of a mouse neural RNA-binding protein, Musashi1.

Musashi1 is an RNA-binding protein abundantly expressed in the developing mouse central nervous system. Its restricted expression in neural precursor cells suggests that it is involved in the regulation of asymmetric cell division. Musashi1 contains two ribonucleoprotein (RNP)-type RNA-binding domains (RBDs), RBD1 and RBD2. Our previous studies showed that RBD1 alone binds to RNA, while the binding of RBD2 is not detected under the same conditions. Joining of RBD2 to RBD1, however, increases the affinity to greater than that of RBD1 alone, indicating that RBD2 contributes to RNA-binding. We have determined the three-dimensional solution structure of the C-terminal RBD (RBD2) of Musashi1 by NMR. It folds into a compact alpha beta structure comprising a four-stranded antiparallel beta-sheet packed against two alpha-helices, which is characteristic of RNP-type RBDs. Special structural features of RBD2 include a beta-bulge in beta2 and a shallow twist of the beta-sheet. The smaller 1H-15N nuclear Overhauser enhancement values for the residues of loop 3 between beta2 and beta3 suggest that this loop is flexible in the time-scale of nano- to picosecond order. The smaller 15N T2 values for the residues around the border between alpha2 and the following loop (loop 5) suggest this region undergoes conformational exchange in the milli- to microsecond time-scale. Chemical shift perturbation analysis indicated that RBD2 binds to an RNA oligomer obtained by in vitro selection under the conditions for NMR measurements, and thus the nature of the weak RNA-binding of RBD2 was successfully characterized by NMR, which is otherwise difficult to assess. Mainly the residues of the surface composed of the four-stranded beta-sheet, loops and C-terminal region are involved in the interaction. The appearance of side-chain NH proton resonances of arginine residues of loop 3 and imino proton resonances of RNA bases upon complex formation suggests the formation of intermolecular hydrogen bonds. The structural arrangement of the rings of the conserved aromatic residues of beta2 and beta3 is suitable for stacking interaction with RNA bases, known to be one of the major protein-RNA interactions, but a survey of the perturbation data suggested that the stacking interaction is not ideally achieved in the complex, which may be related to the weaker RNA-binding of RBD2.

Structure of the C-terminal RNA-binding domain of hnRNP D0 (AUF1), its interactions with RNA and DNA, and change in backbone dynamics upon complex formation with DNA.

Heterogeneous nuclear ribonucleoprotein (hnRNP) D0 has two ribonucleoprotein (RNP) -type RNA-binding domains (RBDs), each of which can specifically bind to the UUAG-sequence. hnRNP D0 also binds specifically to single-stranded d(TTAGGG)(n), the human telomeric DNA repeat. We have already reported the structure and interactions with RNA of the N-terminal RBD (RBD1). Here, the structure of the C-terminal RBD (RBD2) determined by NMR is presented. It folds into a compact alpha beta structure comprising an antiparallel beta-sheet packed against two alpha-helices, which is characteristic of RNP-type RBDs. In addition to the four beta-strands commonly found in RNP-type RBDs, an extra beta-strand, termed beta 4(-), was found just before the fourth beta-strand, yielding a five-stranded beta-sheet. Candidate residues of RBD2 involved in the interactions with RNA were identified by chemical shift perturbation analysis. Perturbation was detected on the beta-sheet side, not on the opposite alpha-helix side, as observed for RBD1. It is notable that the beta 4(-) to beta 4 region of RBD2 is involved in the interactions in contrast to the case of RBD1. The chemical shift perturbation analysis also showed that RBD2 interacts with DNA in essentially the same way as with RNA. Changes in the backbone dynamics upon complex formation with DNA were examined by means of model free analysis of relaxation data. In free RBD2, the beta 4(-) to beta 4 region exhibits slow conformational exchange on the milli- to microsecond time scale. The exchange is quenched upon complex formation. The flexibility of free RBD2 may be utilized in the recognition process by allowing different conformational states to be accessed and facilitating induced fit. Additionally, faster flexibility on the nano- to picosecond time scale was observed for loop 3 located between beta 2 and beta 3 in free RBD2, which is retained by the complex as well.

Crystallization and preliminary X-ray investigation of non-specific complexes of a mutant ribonuclease T1 (Y45W) with 2'AMP and 2'UMP.

We have succeeded in crystallizing complexes of a mutant ribonuclease T1 (Y45W) with the non-cognizable ribonucleotides 2'AMP and 2'UMP by macroscopic seeding of microcrystals of the mutant enzyme complexed with 2'GMP, which is the cognizable nucleotide inhibitor. The mutant enzyme has a tryptophan residue instead of Tyr45 of the wild-type enzyme and thus this mutation enhances the binding of ribonucleotides to the enzyme. The space group is P212121 with unit cell dimensions a = 49.40 A, b = 46.71 A, c = 41.02 A for the complex with 2'AMP and a = 48.97, b = 46.58 A, c = 40.97 A for the complex with 2'UMP, both of which are poorly isomorphous to the mother crystals. Diffraction data for the complexes with 2'AMP and 2'UMP were collected on a diffractometer at 1.7 A and 2.4 A resolution, respectively. The present studies show that crystallization of non-specific complexes of other protein-ligand systems with the dissociation constants around 10(-3) M, or even larger, could be feasible by application of the seeding technique. A comparison of the crystal structures of the complexes with that with 2'GMP may serve as a structural basis for the determination of differences between the specific and non-specific interactions of the enzyme.

An intramolecular quadruplex of (GGA)(4) triplet repeat DNA with a G:G:G:G tetrad and a G(:A):G(:A):G(:A):G heptad, and its dimeric interaction.

The structure of d(GGAGGAGGAGGA) containing four tandem repeats of a GGA triplet sequence has been determined under physiological K(+) conditions. d(GGAGGAGGAGGA) folds into an intramolecular quadruplex composed of a G:G:G:G tetrad and a G(:A):G(:A):G(:A):G heptad. Four G-G segments of d(GGAGGAGGAGGA) are aligned parallel with each other due to six successive turns of the main chain at each of the GGA and GAGG segments. Two quadruplexes form a dimer stabilized through a stacking interaction between the heptads of the two quadruplexes. Comparison of the structure of d(GGAGGAGGAGGA) with the reported structure of d(GGAGGAN) (N=G or T) containing two tandem repeats of the GGA triplet revealed that although the two structures resemble each other to some extent, the extension of the repeats of the GGA triplet leads to distinct structural differences: intramolecular quadruplex for 12-mer versus intermolecular quadruplex for 7-mer; heptad versus hexad in the quadruplex; and three sheared G:A base-pairs versus two sheared G:A base-pairs plus one A:A base-pair per quadruplex. It was also suggested that d(GGAGGAGGAGGA) forms a similar quadruplex under low salt concentration conditions. This is in contrast to the case of d(GGAGGAN) (N=G or T), which forms a duplex under low salt concentration conditions. On the basis of these results, the structure of naturally occurring GGA triplet repeat DNA is discussed.

Three-dimensional structure of a mutant ribonuclease T1 (Y45W) complexed with non-cognizable ribonucleotide, 2'AMP, and its comparison with a specific complex with 2'GMP.

The crystal structure of a mutant ribonuclease T1 (Y45W) complexed with a non-cognizable ribonucleotide, 2'AMP, has been determined and refined to an R-factor of 0.159 using X-ray diffraction data at 1.7 A resolution. A specific complex of the enzyme with 2'GMP was also determined and refined to an R-factor of 0.173 at 1.9 A resolution. The adenine base of 2'AMP was found at a base-binding site that is far apart from the guanine recognition site, where the guanine base of 2'GMP binds. The binding of the adenine base is mediated by a single hydrogen bond and stacking interaction of the base with the imidazole ring of His92. The mode of stacking of the adenine base with His92 is similar to the stacking of the guanine base observed in complexes of ribonuclease T1 with guanylyl-2',5'-guanosine, reported by Koepke et al., and two guanosine bases, reported by Lenz et al., and in the complex of barnase with d(GpC), reported by Baudet & Janin. These observations suggest that the site is non-specific for base binding. The phosphate group of 2'AMP is tightly locked at the catalytic site with seven hydrogen bonds to the enzyme in a similar manner to that of 2'GMP. In addition, two hydrogen bonds are formed between the sugar moiety of 2'AMP and the enzyme. The 2'AMP molecule adopts the anti conformation of the glycosidic bond and C-3'-exo sugar pucker, whereas 2'GMP is in the syn conformation with C-3'-endo-C'-2'-exo pucker. The mutation enhances the binding of 2'GMP with conformational changes of the sugar ring and displacement of the phosphate group towards the interior of the catalytic site from the corresponding position in the wild-type enzyme complex. Comparison of two crystal structures obtained provides a solution to the problem that non-cognizable nucleotides exhibit unexpectedly strong binding to the enzyme, compared with high specificity in nucleolytic activity. The results indicate that the discrimination of the guanine base from the other nucleotide bases at the guanine recognition site is more effective than that estimated from nucleotide-binding experiments so far.

Structure and interactions with RNA of the N-terminal UUAG-specific RNA-binding domain of hnRNP D0.

Heterogeneous nuclear ribonucleoprotein (hnRNP) D0 has two ribonucleoprotein (RNP)-type RNA-binding domains (RBDs), each of which can bind solely to the UUAG sequence specifically. The structure of the N-terminal RBD (RBD1) determined by NMR is presented here. It folds into a compact alphabeta structure comprising a four-stranded antiparallel beta-sheet packed against two alpha-helices, which is characteristic of the RNP-type RBDs. Special structural features of RBD1 include N-capping boxes for both alpha-helices, a beta-bulge in the second beta-strand, and an additional short antiparallel beta-sheet coupled with a beta-turn-like structure in a loop. Two hydrogen bonds which restrict the positions of loops were identified. Backbone resonance assignments for RBD1 complexed with r(UUAGGG) revealed that the overall folding is maintained in the complex. The candidate residues involved in the interactions with RNA were identified by chemical shift perturbation analysis. They are located in the central and peripheral regions of the RNA-binding surface composed of the four-stranded beta-sheet, loops, and the C-terminal region. It is suggested that non-specific interactions with RNA are performed by the residues in the central region of the RNA-binding surface, while specific interactions are performed by those in the peripheral regions. It was also found that RBD1 has the ability to inhibit the formation of the quadruplex structure.

Structural properties and RNA-binding activities of two RNA recognition motifs of a mouse neural RNA-binding protein, mouse-Musashi-1.

mouse-Musashi-1 (m-Msi-1) is an RNA-binding protein, abundantly expressed in the developing mammalian central nervous system (CNS). m-Msi-1 contains two RNA recognition motifs (RRMs). In this study, we found that the N-terminal RRM of m-Msi-1 (MMA) binds strongly to poly(G) and weakly to poly(U) in a way similar to that of the full-length m-Msi-1 protein characterized previously. The C-terminal RRM of m-Msi-1 (MMB), however, does not bind to RNA. In addition, the circular dichroism (CD) spectra of the two RRMs showed that the alpha-helical content of MMA is significantly higher than that of MMB, indicating that some differences in the secondary structure may be responsible for the distinct RNA binding properties of MMA and MMB.

Novel 1H-31P heteronuclear coherence transfer spectroscopy based on spin locking MLEV-17 and its application to signal assignment for a short DNA fragment.

We have applied the MLEV-17 proton spin locking pulse to 1H-31P 2D heteronuclear correlation spectroscopy. Using this technique, long-range correlations between a proton and phosphorus that is up to 5 bonds distant are observed. Thus, the 3'-, 4'- and 5'-proton resonances can be traced 'sequentially' along the chain on the 2D NMR chart. Application of this technique to the complete assignment of the sugar proton and phosphorus resonances for d(ApGpA) is reported. This technique can also be used for convenient assignment of the phosphorus resonances of oligonucleotides as correlations between the 1'-proton and its 3'-side phosphorus are observed.

The role of a conserved guanosine residue in the hammerhead-type RNA enzyme.

We have designed a hammerhead-type RNA system which consists of three RNA fragments for normal and modified complexes which contain a non-cleavable substrate with 2'-O-methylcytidine and a guanosine-to-inosine replaced enzyme. Examination of the RNA-cleaving activity and conformational properties of the complexes suggests that the 2-amino group of a conserved guanosine residue in the loop region plays an important role for maintaining both the activity and loop conformation.

A----Z transition in the synthetic hexanucleotide (dCdGfl)3.

500 MHz proton NMR and NOE measurements on (dCdGfl)3 show that at very low ionic strength the hexanucleotide adopts an A-DNA conformation, whereas at high salt concentrations a Z-form is found. At intermediate salt concentrations the two species are in slow exchange on the proton NMR time scale. This transition was also observed by characteristic changes in the CD spectra.

Secretion of recombinant ribonuclease T1 into the periplasmic space of Escherichia coli with the aid of the signal peptide of alkaline phosphatase.

The ribonuclease T1 (RNase T1) gene was ligated to a synthetic gene for the signal peptide of Escherichia coli alkaline phosphatase. When this fusion gene was expressed in E. coli under the control of the trp promoter, active RNase T1 having the correct N-terminal sequence was secreted into the periplasmic space, indicating that the heterologous signal peptide had been cleaved off correctly. The enzyme could be readily purified from the periplasmic fraction with a yield of 1.8 mg from 1 liter culture. Adopting the same strategy, it was possible to produce a labile mutant of RNase T1 (Glu-58----Ala mutant) in E. coli, the yield of the purified mutant enzyme being 2.0 mg from 1 liter culture.

Combination of Trp and Glu residues for recognition of mRNA cap structure. Analysis of m7G base recognition site of human cap binding protein (IF-4E) by site-directed mutagenesis.

Four mutants of the human cap binding protein (hCBP), in which Trp-102, Glu-103, Asp-104 or Glu-105 was changed to the aliphatic Leu or Ala, were prepared, and their cap binding abilities were examined. Cap binding abilities of two mutants, W102L (Trp-102----Leu) and E105A (Glu-105----Ala), were significantly decreased in comparison with the wild-type hCBP. This result suggests that Trp-102 and Glu-105 are both necessary for the cap binding, and the most probable binding mode with the m7G of cap structure is the combination of the stacking by Trp-102 and the hydrogen-bond pairing by Glu-105, as was already proposed from the model studies.

Non-cognizable ribonucleotide, 2'AMP, binds to a mutant ribonuclease T1 (Y45W) at a new base-binding site but not at the guanine-recognition site.

Complex of a mutant ribonuclease T1 (Y45W) with a non-cognizable ribonucleotide, 2'AMP, has been determined and refined by X-ray diffraction at 1.7 A resolution. The 2'AMP molecule locates at a new base-binding site which is remote from the guanine-recognition site, where 2'GMP was found to be bound. The nucleotide adopts the anti conformation of the glycosidic bond and C3'-exo sugar pucker. There exists a single hydrogen bond between the adenine base and the enzyme, and, therefore, the site found is apparently a non-specific binding site. The results indicate that the binding of 2'AMP to the guanine-recognition site is weaker than that to the new binding site.

Temporal and spatial profile of apoptotic cell death in transient intracerebral mass lesion of the rat.

Apoptosis is involved in the pathogenesis of cerebral ischemia. Previous studies have confirmed that the brain surrounding an intracerebral hematoma develops ischemia. We investigated the number and distribution of cells exhibiting DNA fragmentation with apoptotic morphology in the transient intracerebral mass lesion to determine whether apoptosis contributed to the lesion progress after intracerebral hemorrhage (ICH). Transient intracerebral mass was created by inflation of a microballoon for 10 min (group A) or 2 h (group B) in the caudoputamen in rats, and brains were examined 1, 3, 6, 24, and 48 h after microballoon deflation. The lesion volume was calculated using parallel coronal sections with cresyl violet staining. Terminal deoxynucleotidyl transferase (TdT)-mediated deoxyuridine (dUTP)-biotin nick end labeling (TUNEL) was used to detect cells undergoing DNA fragmentation. Immunohistochemistry for Fas antigen was also done to ascertain molecular mechanisms of apoptosis. Histological examination of hematoxylin and eosin-stained sections showed the typical appearance of neuronal necrosis in the caudoputaminal lesion. Lesion volume in the caudoputamen gradually increased as time advanced from 1 to 48 h. Cells stained heavily by TUNEL with apoptotic morphology were detected in the lesion, but not in the inner boundary zone of the lesion. The number of these cells significantly increased from 6 to 24 h in each experimental group (p < 0.05). The cells with positive immunoreactivity for Fas antigen was prominently observed in the lesion at 6 h. The distribution of apoptotic cells and the rapid increase in the number of apoptotic cells after 24 h propose that apoptotic cell death may contribute to lesion core formation but not to gradual development of the lesion.

Preparation of a new fluorescent analog of ATP, 2'-(5-dimethylaminonaphthalene-1-sulfonyl)amino-2'-deoxy ATP, and its interactions with myosin and actomyosin.

A fluorescent ATP analog, 2'-(5-dimethylaminonaphthalene-1-sulfonyl)amino-2'-deoxy ATP (DNS-ATP), was synthesized. In water, the wavelengths of maximum excitation were 260 and 340 nm, and that of maximum emission was 554 nm. The fluorescence quantum yield with excitation at 340 nm was 0.052. In 80% dioxane-20% water solution, the wavelength of the maximum emission shifted to 527 nm and the quantum yield was about 5.4 times in water. When DNS-ATP was mixed with HMM in the presence of Mg2+ ions, the fluorescence intensity of DNS-ATP was enhanced by about 30%, and the wavelength of maximum emission shifted to 545 nm. The observed second-order rate constant for the change in fluorescence intensity after adding DNS-ATP to HMM was 1.6 x 10(-7) M-1 . s-1, while the observed first-order rate constant for its recovery was 0.17 s-1. When the HMM DNS-ATPase reaction was measured in terms of the TCA-Pi liberation, 1 mol of initial burst of Pi liberation per mol of myosin was observed. In 50 mM KCl and at 20 degrees C, the rate of the HMM DNS-ATPase reaction was increased by F-actin from 0.4 to 1.15 s-1 (in 3 mg/ml F-actin). The observed dissociation constant for the binding of DNS-ATP with HMM increased from 1.2 to 20 microM in the presence of 5 mg/ml F-actin. However, the extent of change in fluorescence intensity at infinite concentration of DNS-ATP was unaffected by the presence of F-actin.

Effects of 2'-substituents of the first deoxyguanosine residue in the recognition sequence of EcoRI restriction endonuclease activity.

The effects of 2'-substituents of the first deoxyguanosine on EcoRI activity were examined using synthetic octadeoxynucleotides D(GG*AATTCC) containing 2'-substituted derivatives (G*), i.e., 2'-fluoro-2'-deoxyguanosine (dGfl), 2'-chloro-2'-deoxyguanosine (dGcl), and guanosine (rG). The overall structures of the octamers were very similar, as shown by CD and UV measurements, although their EcoRI reactivities were very different: 100% in 60 min for d(GGAATTCC) and d(GGflAATTCC), 5% in 24 h for d[G(rG)AATTCC], and no cleavage at all in 24 h for d(GGclAATTCC). However, the kinetics showed the octamers exhibit similar binding-affinity to the enzyme (10(-6)-10(-7) M). 31P-NMR analysis suggested the modified octamers change the phosphate backbone conformation in a duplex, since an unusual downfield-shifted signal in the spectra was commonly observed for the modified octamers at low temperature (i.e., a duplex state), which was shifted upfield at high temperature (i.e., a single strand state). The order of the differences was dGcl > rG > dGfl-containing octamers, coinciding with that of the vdW volume of 2'-substituents (Cl > OH > F) and the cleavage reactivities. These findings suggest that steric hindrance by the 2'-substituents causes of conformational change of the phosphate backbone close to the scissile bond, and then interferes with the EcoRI reaction.

Activation and repression of the activity of a lead ribozyme by the combination of Pb2+ and Mg2+.

The effect of Pb2+ and Mg2+ on the activity of a lead ribozyme with modified sequences has been studied. At low Pb2+ concentrations, cleavage at a previously reported site (site a) is observed. At higher Pb2+ concentrations, cleavage at a new site (site b) adjacent to site a is observed, while the cleavage at site a is repressed. On the addition of a certain amount of Mg2+, the cleavage at site a is enhanced by almost fourfold, while the cleavage at site b is repressed. Further addition of Mg2+ represses the cleavage at both sites. CD analysis indicates that the structure and stability of the lead ribozyme change depending on the metal conditions. Activation and repression of the activity by the combination of Pb2+ and Mg2+ are rationalized by considering that the two metals compete with each other for binding at two metal-binding sites.

Direct expression of a synthetic gene in Escherichia coli: purification and physicochemical properties of human initiation factor 4E.

An artificial synthetic gene coding for human eIF-4E was cloned into an expression vector and direct expression was attempted in Escherichia coli [BL21(DE3) strain] under the control of T7 promoter. The active gene product which was induced in high yield (ca. 4 mg/100 ml) by isopropyl-beta-D-thiogalactopyranoside was purified to homogeneity by a two-step chromatographic procedure with a good yield (ca. 74%), and was confirmed to be recombinant human eIF-4E by amino acid composition and sequence analyses, isoelectric focusing, and absorption spectral measurements. The identity of three-dimensional structures between the recombinant and native human eIF-4Es was confirmed by CD and fluorescence measurements.

Dimerization and DNA binding facilitate alpha-helix formation of Max in solution.

Max is a basic region/helix-loop-helix/leucine zipper (b/HLH/Z) protein that forms a hetero-complex with the Myc family proteins Myc, Mad, and Mxi1, and a homo-complex with itself. These complexes specifically bind to double-stranded DNA containing CACGTG sequences. Here, we report on the structural properties in aqueous solution of a 109-amino-acid protein, Max110, corresponding to the N-terminal domain of Max containing the b/HLH/Z motif (residues 2-110), as characterized by combined use of circular dichroism (CD) and sedimentation equilibrium experiments. The results showed that the alpha-helical content of Max110 increases with increasing protein concentration. The sedimentation equilibrium data indicated that Max110 exists as a monomer at low protein concentration, and forms a dimer at high protein concentration. Further increases in the alpha-helical content of Max110 occur upon addition of DNA with the CACGTG recognition sequence. Thus, dimerization and binding to DNA of Max both favor an increase of the alpha-helical content.