Toshiaki Takei

Formation of circular polyribosomes in wheat germ cell-free protein synthesis system.

We report a morphological study of functioning ribosomes in a efficient and robust cell‐free protein synthesis system prepared from wheat embryos. Sucrose density gradient analysis of translated mixtures programmed with luciferase mRNAs having different 5′ and 3′ untranslated regions showed formation of large polysomes. Electron microscopic examination of translation mixtures programmed with those of capped and polyadenylated mRNA revealed that ribosomes assemble into a circular‐type polysome in vitro. Furthermore, a series of experiments using mRNAs lacking either cap, poly(A) tail or both also resulted in the formation of circular polysomes, which are indistinguishable from those with the original mRNA. The wheat germ cell‐free system may provide a good experimental system for understanding functional ribosomes at the molecular level.

Electron microscopy and computational studies of Ebh, a giant cell-wall-associated protein from Staphylococcus aureus.

Ebh, a giant protein found in staphylococci, contains several domains, including a large central region with 52 imperfect repeats of a domain composed of 126 amino acids. We used electron microscopy to observe the rod-like structure of a partial Ebh protein containing 10 repeating units. This is the first report of the direct observation of an Ebh structure containing a large number of repeating units, although structures containing one, two, or four repeating units have been reported. The observed structure of the partial Ebh protein was distorted and had a length of ca. 520A and a width of ca. 21A. The observed structures were consistent with those deduced from crystal structure analysis, suggesting that the Ebh domains are connected to form a rod-like structure. The crystal structure data revealed distorted, string-like features in the simulated structure of the whole-length Ebh protein. Superposition of fragments of the simulated whole-length structure of the Ebh protein onto each electron micrograph showed a high level of correlation between the observed and calculated structures. These results suggest that Ebh is composed of highly flexible filate molecules. The highly repetitive structure and the associated unique structural flexibility of Ebh support the proposed function of this protein, i.e. binding to sugars in the cell wall. This binding might result in intra-cell-wall cross-linking that contributes to the rigidity of bacterial cells.

Effects of chain length of an amphipathic polypeptide carrying the repeated amino acid sequence (LETLAKA)(n) on α-helix and fibrous assembly formation.

Polypeptide α3 (21 residues), with three repeats of a seven-amino-acid sequence (LETLAKA)(3), forms an amphipathic α-helix and a long fibrous assembly. Here, we investigated the ability of α3-series polypeptides (with 14-42 residues) of various chain lengths to form α-helices and fibrous assemblies. Polypeptide α2 (14 residues), with two same-sequence repeats, did not form an α-helix, but polypeptide α2L (15 residues; α2 with one additional leucine residue on its carboxyl terminal) did form an α-helix and fibrous assembly. Fibrous assembly formation was associated with polypeptides at least as long as polypeptide α2L and with five leucine residues, indicating that the C-terminal leucine has a critical element for stabilization of α-helix and fibril formation. In contrast, polypeptides α5 (35 residues) and α6 (42 residues) aggregated easily, although they formed α-helices. A 15-35-residue chain was required for fibrous assembly formation. Electron microscopy and X-ray fiber diffraction showed that the thinnest fibrous assemblies of polypeptides were about 20 Å and had periodicities coincident with the length of the α-helix in a longitudinal direction. These results indicated that the α-helix structures were orientated along the fibrous axis and assembled into a bundle. Furthermore, the width and length of fibrous assemblies changed with changes in the pH value, resulting in variations in the charged states of the residues. Our results suggest that the formation of fibrous assemblies of amphipathic α-helices is due to the assembly of bundles via the hydrophobic faces of the helices and extension with hydrophobic noncovalent bonds containing a leucine.

Role of positions e and g in the fibrous assembly formation of an amphipathic α‐helix‐forming polypeptide

We previously characterized α3, a polypeptide that has a three times repeated sequence of seven amino acids (abcdefg: LETLAKA) and forms fibrous assemblies composed of amphipathic α‐helices. Upon comparison of the amino acid sequences of α3 with other α‐helix forming polypeptides, we proposed that the fibrous assemblies were formed due to the alanine (Ala) residues at positions e and g. Here, we characterized seven α3 analog polypeptides with serine (Ser), glycine (Gly), or charged residues substituted for Ala at positions e and g. The α‐helix forming abilities of the substituted polypeptides were less than that of α3. The polypeptides with amino acid substitutions at position g and the polypeptide KEα3, in which Ala was substituted with charged amino acids, formed few fibrous assemblies. In contrast, polypeptides with Ala replaced by Ser at position e formed β‐sheets under several conditions. These results show that Ala residues at position e and particularly at position g are involved in the formation of fibrous assemblies.

Type 1 IFN inhibits the growth factor deprived apoptosis of cultured human aortic endothelial cells and protects the cells from chemically induced oxidative cytotoxicity

It has been shown that the genesis of atherosclerotic lesions is resulted from the injury of vascular endothelial cells and the cell damage is triggered by oxygen radicals generated from various tissues. Human vascular endothelial cells can survive and proliferate depending on growth factors such as VEGF or basic FGF and are induced apoptosis by the deprivation of growth factor or serum. It was found that type 1 IFN inhibits the growth factor deprived cell death of human aortic endothelial cells (HAEC) and protects the cells from chemically induced oxidative cytotoxicity. The anti‐apoptotic effects of type 1 IFN were certified by flow cytometry using annexin‐V‐FITC/PI double staining and cell cycle analysis, fluorescence microscopy using Hoechst33342 and PI, colorimetric assay for caspase‐3 activity, p53 and bax mRNA expressions, and cell counts. It was considered that IFN‐β inhibits the executive late stage apoptosis from the results of annexin‐V‐FITC/PI double staining and the inhibition of caspase‐3 activity, and that the anti‐apoptotic effect might be owing to the direct inhibition of the apoptotic pathway mediated by p53 from the transient down‐regulation of bax mRNA expression. Whereas, type 1 IFN protected the cells from the oxidative cytotoxicity induced by tertiary butylhydroperoxide (TBH) under the presence of Ca2+. The effects of IFN‐β is more potent inhibitor of cell death than IFN‐α. These results indicate that type 1 IFN, especially IFN‐β may be useful for the diseases with vascular endothelium damage such as atherosclerosis or restenosis after angioplasty as a medical treatment or a prophylactic. J. Cell. Biochem. 113: 3823–3834, 2012.

pH responsiveness of fibrous assemblies of repeat‐sequence amphipathic α‐helix polypeptides

We reported previously that our designed polypeptide α3 (21 residues), which has three repeats of a seven‐amino‐acid sequence (LETLAKA)3, forms not only an amphipathic α‐helix structure but also long fibrous assemblies in aqueous solution. To address the relationship between the electrical states of the polypeptide and its α‐helix and fibrous assembly formation, we characterized mutated polypeptides in which charged amino acid residues of α3 were replaced with Ser. We prepared the following polypeptides: 2Sα3 (LSTLAKA)3, in which all Glu residues were replaced with Ser residues; 6Sα3 (LETLASA)3, in which all Lys residues were replaced with Ser; and 2S6Sα3 (LSTLASA)3; in which all Glu and Lys residues were replaced with Ser. In 0.1M KCl, 2Sα3 formed an α‐helix under basic conditions and 6Sα3 formed an α‐helix under acid conditions. In 1M KCl, they both formed α‐helices under a wide pH range. In addition, 2Sα3 and 6Sα3 formed fibrous assemblies under the same buffer conditions in which they formed α‐helices. α‐Helix and fibrous assembly formation by these polypeptides was reversible in a pH‐dependent manner. In contrast, 2S6Sα3 formed an α‐helix under basic conditions in 1M KCl. Taken together, these findings reveal that the charge states of the charged amino acid residues and the charge state of the Leu residue located at the terminus play an important role in α‐helix formation.

Production and characterization of genetically modified human IL-11 variants.

Interleukin-11 (IL-11) has been expected as a drug on severe thrombocytopenia caused by myelo-suppressive chemotherapy. Whereas, development of IL-11 inhibitor is also expected for a treatment against IL-11 related cancer progression. Here, we will demonstrate the creation of various kinds of genetically modified hIL-11s. Modified vectors were constructed by introducing N- or O-glycosylation site on the region of hIL-11 that does not belong to the core α-helical motif based on the predicted secondary structure. N-terminal (N: between 22 to 23 aa), the first loop (M1:70 to 71 aa), the second loop (M2:114-115 aa), the third loop (M3:160-161 aa) and C-terminal (C: 200- aa) were selected for modification. A large scale production system was established and the characteristics of modified hIL-11s were evaluated. The structure was analyzed by amino acid sequence and composition analysis and CD-spectra. Glycan was assessed by monosaccharide composition analysis. Growth promoting activity and biological stability were analyzed by proliferation of T1165 cells. N-terminal modified proteins were well glycosylated and produced. Growth activity of 3NN with NASNASNAS sequence on N-terminal was about tenfold higher than wild type (WT). Structural and biological stabilities of 3NN were also better than WT and residence time in mouse blood was longer than WT. M1 variants lacked growth activity though they are well glycosylated and secondary structure is very stable. Both of 3NN and OM1 with AAATPAPG on M1 associated with hIL-11R strongly. These results indicate N-terminal and M1 variants will be expected for practical use as potent agonists or antagonists of hIL-11.