Two forms of vitellogenin (Vg), Vg A and Vg B, were identified in serum from estrogen-treated barfin flounder (Verasper moseri). Structural changes of lipovitellins (Lvs) derived from the two Vgs were examined during vitellogenesis and oocyte maturation. Two Lvs, vLv A and vLv B, were identified electrophoretically and immunologically in postvitellogenic oocytes. Each appeared to be composed of distinct heavy chains (vLvH A, M(r) 107,000, and vLvH B, M(r) 94,000) and light chains (vLvL A, M(r) 30,000, and vLvL B, M(r) 28,000) when analyzed by SDS-PAGE. Results from N-terminal amino acid sequencing and Western blotting using antisera to vLvH A and vLvH B verified that there are two Vg polypeptides in serum from estrogen-treated fish, Vg A (M(r) 168,000) and Vg B (M(r) 175,000), which give rise to vLvH A-vLvL A and vLvH B-vLvL B, respectively. N-terminal sequencing revealed two sequences for both phosvitin and beta'-component, supporting the concept of duality for all three classes of Vg-derived yolk proteins. During oocyte maturation, native dimeric vLv B was dissociated into a native M(r) 170,000 monomer (oLv B). Meanwhile, vLv A was extensively cleaved including complete degradation of vLvH A into free amino acids. We propose that the quantitative ratio of vLv A to vLv B in postvitellogenic oocytes regulates the buoyancy of the spawned pelagic eggs by controlling availability of free amino acids which function as osmotic effectors during oocyte hydration. The vLv A/vLv B ratio likely also controls the proportional availability of different types of nutrients, free amino acids versus Lv, for use during embryonic development.

The deposition of amyloid beta-protein (Abeta) is an invariable feature of Alzheimer's disease (AD); however, the biological mechanism underlying Abeta assembly into fibrils in the brain remains unclear. Here, we show that a high-density cluster of GM1 ganglioside (GM1), which was detected by the specific binding of a novel peptide (p3), appeared selectively on synaptosomes prepared from aged mouse brains. Notably, the synaptosomes bearing the high-density GM1 cluster showed extraordinary potency to induce Abeta assembly, which was suppressed by an antibody specific to GM1-bound Abeta, an endogenous seed for AD amyloid. Together with evidence that Abeta deposition starts at presynaptic terminals in the AD brain and that GM1 levels significantly increase in amyloid-positive synaptosomes prepared from the AD brain, our results suggest that the age-dependent high-density GM1 clustering at presynaptic neuritic terminals is a critical step for Abeta deposition in AD.

Influenza is an infectious disease caused by the influenza virus, and each year many people suffer from this disease. Hemagglutinin (HA) in the membrane of type A influenza viruses recognizes sialylglycoconjugate receptors on the host cell surface at an initial step in the infection process; consequently, HA inhibitors are considered potential candidates for antiviral drugs. We identified peptides that bind to receptor-binding sites through a multiple serial selection from phage-displayed random peptide libraries. Using the HA of the H1 and H3 strains as target proteins, we obtained peptides that bind to both HAs. The binding affinities of peptides for these HAs were improved by secondary and tertiary selections from the corresponding sublibraries. A docking simulation suggested that, similar to sialic acid, the peptides are recognized by the receptor-binding site in HA, which indicates that these peptides mimic the sialic acid structure. N-stearoyl peptides inhibited infections by the A/Puerto Rico/8/34 (H1N1) and A/Aichi/2/68 (H3N2) strains of influenza virus. Such HA-inhibitors are promising candidates for novel antiviral drugs.

Three complete cDNAs encoding different forms of vitellogenin (Vtg) were isolated from a white perch (Morone americana) liver cDNA library and characterized with respect to immunobiochemical and functional features of the three Vtgs and their product yolk proteins (YPs) in this species and in the congeneric striped bass (Morone saxatilis). The two longest cDNAs encoded Vtgs with a complete suite of yolk protein domains that, based on comparisons with vtg sequences from other species, were categorized as VtgAa and VtgAb using the current nomenclature for multiple teleost Vtgs. The shorter cDNA encoded a Vtg that lacked a phosvitin domain, had a shortened C-terminus, and was categorized as VtgC. Mapping of peptide sequences from the purified Vtgs and their derived YPs to Vtg sequences deduced from the cDNAs definitively identified the white perch VtgAa, VtgAb, and VtgC proteins. Detailed comparisons of the primary structures of each Vtg with partial or complete sequences of Morone yolk proteins or of Vtgs from other fishes revealed conserved and variant structural elements of teleost Vtgs with functional significance, including, as examples, signal peptide cleavage sites, dimerization sites, cathepsin D protease recognition sites, and receptor-binding domains. These comparisons also yielded an interim revision of the classification scheme for multiple teleost Vtgs.

Full-length cDNAs encoding three forms of vitellogenin (Vg) were obtained from a liver cDNA library of estrogen-treated red seabream, Pagrus major. Two of the three Vg sequences had high homology with type-A and -B Vgs (VgA and VgB) of other teleosts. The third red seabream Vg was classified as a type-C or phosvitinless (Pvl) Vg due to its lack of a phosvitin (Pv) domain. Two Vg preparations (610 and 340 kDa) from blood serum of estradiol-treated fish were biochemically characterized. Analyses of precursor-product relationships by examination of N-terminal amino acid sequences verified cleavage of the 610 kDa Vg into a 540 kDa lipovitellin (Lv) and a 32 kDa beta'-component. Each of these yolk preparations comprising both VgA- and VgB-derived polypeptides. The 340 kDa Vg, which was immunologically verified to be a PvlVg, was accumulated by vitellogenic oocytes with no alterations to its native molecular mass. During oocyte maturation, the VgA- and VgB-derived yolk proteins were differentially processed, presumably to generate a pool of free amino acids for oocyte hydration or for allocation of specific types of nutrients, amino acids, and proteins, to the developing embryo. Conversely, the 340 kDa Vg-derived yolk protein is unlikely to contribute to oocyte hydration or diffusible nutrients since the molecule underwent only minor proteolytic nicking during oogenesis. The present study elucidates for the first time specific functions of three different forms of Vg and their product yolk proteins in a higher taxonomic group of marine teleosts that spawn pelagic eggs.