SPECTROSCOPIC STUDIES OF CHLOROPHYLL a AGGREGATES FORMED BY AQUEOUS DIMETHYL SULFOXIDE

Characteristic chlorophyll (Chl) a aggregates formed in aqueous dimethyl sulfoxide (DMSO) were investigated spectroscopically. Four chlorophyll forms were found with increasing DMSO concentration; they are called A-672, A-683, A-695 and A-665 according to the wavelengths of their absorption maxima. Transformation occurred only in this order. Reverse transformation could not be realized. A-683 and A-695 were apparently formed by the interaction of Chl a with DMSO in the linear dimer and linear polymer arrangements, respectively. Coordination of the Mg atom with a DMSO O atom and interaction between the S atom of one DMSO molecule and the O atom of an other DMSO molecule should lead to formation of a sandwich-type complex of partially overlapping chlorophyll macrocycles (Chl a-DMSO)_n. A-672 and A-665 were assigned to Chl a micelles and to dissolved monomeric Chl a in DMSO, respectively. Fluorescence spectra showed that the A-683 was highly fluorescent, while the A-695 was less fluorescent. Energy migration within the A-695 form to a trap with a low fluorescence yield might be responsible for this difference in the emission intensity.     

Convergence process with quasi-canonical localized orbital in all-electron SCF calculation on proteins

A convergence process with a new type of localized orbital (LO) is proposed for self-consistent field (SCF) molecular orbital (MO) calculations on proteins. Recently, an all-electron density functional (DF) calculation on cytochrome c was achieved by a convergence method in which the initial electron density of a peptide was constructed by linking the electron density of small peptides. However, its convergence was slow and insufficient in SCF calculations for long peptides. In this paper a new kind of LO called the ‘quasi-canonical localized orbital (QCLO)’ is formulated and a computational process with QCLOs is proposed to improve the SCF convergence. The QCLO is localized in a certain region of a molecule, but it is also the canonical MO of the region. In test calculations on a seven-residue peptide, the error in the initial total energy calculated with QCLOs was about one twentieth of that by cut and paste of the electron density, and the number of SCF iterations was reduced markedly. Future applications of QCLO to big and complex molecules are also discussed.     

Improved Method of Molecular Imprinting of Cyclodextrin on Silica-gel Surface for the Preparation of Stable Stationary HPLC Phase

By using N-(3-triethoxysilyl)propylacrylamide (TPAAm), vinyl groups were introduced onto the surface of silica-gel. On the surface of this silica-gel, β-CyD was molecularly imprinted by using a redox initiator, and the composite was used as stationary phase of high performance liquid chromatography (HPLC). The pump pressure was sufficiently low and did not increase even after continuous elution for 24 h. In order to prepare still more stable columns, a new polymerization process was developed. There, the redox initiator was first mixed with the surface-modified silica-gel and then vinylated β-CyD, crosslinker, and the template were added. This modification promoted the immobilization of β-CyD copolymer to the silica-gel, resulting in still lower pump pressure. Concurrently, the imprinting efficiency was increased in comparison with previous method where the redox initiator was directly added to the mixture of the β-CyD–template complex, crosslinker, and surface-modified silica-gel. The molecularly imprinted β-CyD column, prepared by this new method, efficiently discriminated the enantiomers of N-benzyloxycarbonyltyrosine.     

On partial sums of a spectral analogue of the Möbius function

Sankaranarayanan and Sengupta introduced the function μ ^*(n) corresponding to the Möbius function. This is defined by the coefficients of the Dirichlet series 1/L _f (s), where L _f (s) denotes the L-function attached to an even Maaß cusp form f. We will examine partial sums of μ ^*(n). The main result is $\sum_{n\leq x}\mu^{*}(n)=O(x\exp(-A\sqrt{\log x}))$ , where A is a positive constant. It seems to be the corresponding prime number theorem.     

Detection of highly oriented aggregation of L-glutamic acid-derived lipids in dilute organic solution

Aggregation structures in organic gels and xerogels formed from L-glutamic acid-derived lipids were investigated by scanning and transmission electron microscopies, X-ray analyses, and 1H NMR and IR spectroscopic methods. These analyses showed that the gels were produced through the formation of highly oriented aggregates based on a single layer and a remarkable development of their fibrous morphology. We also describe how the critical aggregation concentration can be observed at a concentration below the critical gel concentration by using a dye-complexation method with a cyanine dye, NK-77.     

A cDNA CLONE WITH ANTIONCOGENE ACTIVITY

A cDNA clone, p14--6, which has an antioncogene activity on the v--Ki--Ras oncogene-transformed malignant cell line DT, was found. This clone was recovered from the revertantR14 cells, which had been isolated by transfections of DT cells with a normal human fibro-blast cDNA library cloned in pcD2, an Okayama-Berg vector. When transfected into DT cells,p14--6 clone gave rise to phenotypical flat reversion in 5--15% of DT transfectant colo-nies. The p14--6--transfected flat cell line, RR, was proven to be a true revertant with signif-icantly reduced malignancy by in ritro and in riro malignancy tests. All other clones recov-ered from R14 cells were unable to cause this reversion. Molecular hybridizations showedthat the p14--6 was inserted into RR genome as tandem repeats, and no structural changewas found in the D--Ki--Ras oncogene in RR genome. These facts suggest that the antioncogeneactivity of the p14--6 clone on the DT cells may be exerted through expression of thecDNA contained in this clone. Possib     

Finite Element Analysis of Unsteady Natural Convection

In this study, a new finite element method (the MSR-method) is proposed for unsteady three-dimensional thermal-fluid analyses. This method is a combination of a modified Galerkin method (MGM) and the SIMPLER formulation. In the MSR-method the velocity and pressure are computed using the SIMPLER procedure and the approximate velocity and the energy equation are solved using the MGM. In the MGM, the inertia term and the pressure term are considered explicitly, so only the symmetrical matrixes appear. Then an artificial viscosity is introduced through an error analysis approach to improve its accuracy and stability. In this paper, the natural convection problems in a three-dimensional cavity are simulated up to the Rayleigh number of 10⁸, and converged solutions are obtained. Authors confirmed that our proposed method gives reasonable results for these problems comparing with other research works.