In vertebrates, genomic DNA is often methylated at the 5 th position of cytosine in the sequence of CpG, and this is the only chemical modification that genomic DNA of vertebrates allows under physiological conditions. During evolution, vertebrates acquired CpG methylation as a new tool for controlling gene expression in addition to the varieties of transcription factors. In mammals, the methylation pattern of genomic DNA is erased and reset in germ line and at the early stage of embryogenesis. Maintenance-type methylation activity ensures clonal transmission of the lineage-specific methylation pattern in somatic cells. The methylation pattern is dynamic and changes during cell differentiation. Prior to the expression of tissue-specific genes, specific sites of the promoters are demethylated. In general, the methylation of a gene suppresses its expression. However, not much is known about the mechanisms that regulate the methylation state and the gene expression by DNA methylation.

View full abstract

Galactosyltransferases are enzymes which transfer galactose from UDP-Gal to various acceptors with either retention of the anomeric configuration to form α1, 2-, α1, 3-, α1, 4-, and α1, 6-linkages, or inversion of the anomeric configuration to form β1, 3-, β1, 4-, and β1-ceramide linkages. During the last few years, several (c) DNA sequences coding for galactosyltransferases became available. We have retrieved these sequences and conducted sequence similarity studies. On the basis of both the nature of the reaction catalyzed and the protein sequence identity, these enzymes can be classified into twelve groups. Using a sensitive graphics method for protein comparison, conserved structural features were found in some of the galactosyltransferase groups, and other classes of glycosyltransferases, resulting in the definition of five families. The lengths and locations of the conserved regions as well as the invariant residues are described for each family. In addition, the DxD motif that may be important for substrate recognition and/or catalysis is demonstrated to occur in all families but one.

View full abstract

The intracellular sorting of newly synthesized precursor proteins (preproteins) to mitochondria depends on the “mitochondria-targeting sequence” (MTS), which is located at the amino termini of the preproteins. MTS is required, however, not only for targeting newly synthesized preproteins to mitochondria, but also for all the following steps along the mitochondrial protein import pathway. MTS of nascent preproteins is first recognized by a cytoplasmic molecular chaperone, MSF, and then by Tom7O and Tom20 of the mitochondrial outer membrane receptor complex, Tom5 and Tom40 of the outer membrane protein translocation machinery, Tim23 of the inner membrane protein translocation machinery, and finally the processing peptidase, MPP, in the matrix. MTS is a multi-role sorting sequence which specifically interacts with various components along the mitochondrial protein import pathway. Recognition of MTS at multiple steps during the import of preproteins may contribute to the strict sorting of proteins destined for mitochondria.

View full abstract

The preparation of hepatocyte spheroids by adding a water-soluble synthetic polymer as an artificial matrix was performed in a cell suspension system. Cell-aggregation was promoted without cytotoxicity by adding Eudragit (a copolymer of methacrylic acid and methylmethacrylate) to the culture medium. Spheroid-like cell aggregates, whose liver functions were enhanced, were effectively formed in the presence of 0.1% Eudragit, independent of the cultural substratum. Moreover, the mass preparation of spheroids could be achieved with a high production yield by means of a suspension culture in a spinner flask. In this case, the polymer protected the cells from damage due to agitation. The spheroids induced with Eudragit expressed high liver functions, such as albumin secretion, ammonia removal, and urea synthesis. On histological observation, the spheroids showed a well-developed cell adhesion apparatus and bile canaliculi. In addition, a higher calcium ion concentration in the cells of spheroids was observed compared with in monolayer cells.

View full abstract

It has been shown that G proteins are potential regulatory molecules in the transmembrane signaling cascade. The aim of this study was to examine the possibility of equivalent G-protein redistribution and/or down-regulation in a target cell upon agonist stimulation. Short-term (0-80 min) incubation of rat pituitary GH₄C₁ cells with vasoactive intestinal peptide (VIP, 0.1 μM) induced a decrease in the levels of G_sα in the membrane fraction, whereas immunoblot analysis and reconstitution assay of adenylyl cyclase clearly showed an increase in the amount of G_sα in the supernatant (cytosolic) fraction. The VIP-induced release of G proteins α subunits from membranes was specific for G_sα. The VIP-dependent release of G_sα from membranes was blocked by a VIP-receptor antagonist, (N-Ac-Tyr, D-Phe)-GRF(1-29)-NH₂ (10 μM). Pituitary adenylate cyclase-activating polypeptide (PACAP) also stimulated the release of G_sα from membranes of GH₄C₁ cells. Furthermore, prolonged exposure of cells to VIP (0.1 μM) for 2-24 h caused a 21-40% decrease in G, a from membranes and a 6% increase in total G_sα in the cytosolic fraction. The effect of VIP was dose-dependent with ED₅₀ values of 81.6±20.0 nM for down-regulation and 2.5±0.3 nM for translocation of G_sα. Concurrent treatment of GH₄C₁ cells with VIP and cycloheximide indicated that suppression of protein synthesis de novo did not mimic the effect of VIP. Moreover, the chase experiment of ³⁵S-labeled G_sα clearly demonstrated a more rapid rate of decay in the cells maintained in the presence of the agonist. These data indicate that VIP-receptor activates G_sα protein and induces the release of G_sα from membranes along with its down-regulation in cellular levels.

View full abstract

Full-length cDNAs for extracellular signal-regulated kinases (ERK1 and ERK2) were isolated from a carp ovary cDNA library. The deduced amino acid sequences of carp ERK1 (cERK1) and ERK2 (cERK2) exhibited high degrees of homology to the known sequences of the ERK group. Northern blot analysis showed that cERK1 mRNA was not expressed in a tissue-specific manner, though the level of expression of cERK2 mRNA varied among tissues. Western blot analysis of the brain, kidney, and ovary confirmed the expression of cERK1 and cERK2 in carp. Our findings indicate that two isoforms of ERK, ERK1 and ERK2, exist in fish.

View full abstract

We examined the substrate specificity of two mutated geranylgeranyl diphosphate synthases, I-9 and I-11, with respect to several artificial substrates. These mutated enzymes have replacements in the amino acid sequences from positions 170 to 173, which are thought to be a part of the putative substrate binding region. The wild-type enzyme catalyzes the condensation of IPP with a series of (2E)-3-methyl-2-alkenyl diphosphates to give products with carbon numbers between 14 and 21. On the other hand, the mutated enzymes show lower activities for artificial substrates with short alkyl chains than those of the wild-type enzyme though the carbon numbers of the products are similar to those in the case of the wild-type. The mutated enzyme I-11 never accepts artificial substrates shorter than C₈. Analysis of additional mutated en_??_ymes revealed that the characteristics of the mutated enzymes arise from a few substitutions within positions 171 to 173. These results indicate that the amino acids in the positions 171 to 173 of the geranylgeranyl diphosphate synthase from Sulfolobus acidocaldarius are involved in recognition of short allylic substrates, such as dimethylallyl diphosphate, but not in recognition of the chain length of the products.

View full abstract

CH-296, a recombinant fragment of human fibronectin (FN) composed of the cell-binding domain (C-domain), heparin-binding domain II (H-domain), and CS1 site, enhances the retrovirus-mediated gene transduction (GT) of hematopoietic stem cells. The RGD sequence in the C-domain is recognized by a variety of cell types through integrin VLA-5, and the LDV sequence in the CSI-site is recognized by integrin VLA-4. Retrovirus particles were also found to bind to the H-domain. Consequently, the CH-296 fragment can enhance GT through binding to both retrovirus particles and target cells that express integrins VLA-5 and/or VLA-4. In this study, we found that the GT efficiency can be maintained at levels comparable to that of CH-271, a FN fragment similar to CH-296 but lacking the CS1 site, when a cocktail of separated functional domains of CH-271 is used. When a dish was coated with a mixture of the C-domain and H-domain (molar ratio, 1:10), the GT efficiency of NIH3T3 cells reached the same level as that of the mother fragment, CH-271. The H-domain in the cocktail can be replaced with other virus-binding components, polylysine, FGF, and the insulin-binding domain of Co1V, without the loss of GT efficiency. With other than FN fragments, a cocktail of erythropoietin and polylysine caused higher GT efficiency of Epo-receptor expressing TF-1 cells than in the case of each component alone.

View full abstract

The DNA region responsible for the induction of the catalase gene of Schizosaccharomyces pombe in response to oxidative stress was determined by constructing a series of deletions in the 5'-flanking region of the gene. Cells having deletion -672 (numbered with the transcription start site as +1) to -111 showed no significant difference in catalase expression from the wild-type cells. Cells having deletion -672 to -89 showed reduced basal expression of the catalase mRNA, but retained the ability of induction in response to oxidative stress. Cells having deletion -672 to -55 completely lost the ability to express the catalase mRNA. These results suggested that two regions, -89 to -55 and -111 to -89, are involved in expression of the catalase gene. The DNA region of -89 to -55 overlapped with the Atf1 binding sequence. The Atfi is a bZIP transcription factor with an important role in stress response under the control of the Spc1 mitogen activated protein (MAP) kinase. Introduction of the atf1^- or spc1^- mutation into the mutant having a deletion in -672 to -89 completely abolished the expression of the catalase mRNA. This result indicated that the Spc1-Atfl cascade is involved in expression of the catalase gene through the region of -89 to -55. In mutants spc1^- and atfl^-, basal expression and induction by hydrogen peroxide of catalase mRNA were observed. These results revealed that not only the Atf1 binding site but also another DNA element independent of the Spc1-Atf1 pathway is involved in the expression of the catalase gene in response to oxidative stress in S. pombe. Proteins that bound specifically to each DNA element existed in the cell extract of the wild-type S. pombe.

View full abstract

We and others recently isolated a human TPRD gene, possessing a motif of the tetratrico-peptide repeat (TPR), from the Down syndrome-critical region (DCR) of chromosome 21q22.2. In this study, we isolated a mouse homologue of TPRD eDNA, mtprd, and examined its expression profile in mouse embryos. The gene was mapped to mouse chromosome 16C3.3-4, consistent with the location of DCR, and encodes 1, 979 amino acid residues with 76% identity to TPRD. The mtprd protein has three units of the TPR motif with 91% homology to TPRD. The protein also has two regions homologous to several matrix proteins with 86 and 70% identities to those of TPRD. Several splicing variants of the 5' portion of the open reading frame of mtprd were identified by RT-PCR and sequencing of mRNAs. In situ hybridization showed that mtprd is ubiquitously expressed in mouse embryos but predominantly in the central nervous system, including the telencephalon, mesencephalon, and metencephalon. These results suggest that the TPRD gene is one of the genes responsible for not only the morphological anomalies but also the neurological abnormalities observed in Down syndrome. The presence of splicing variants indicates that the protein may also have several isoforms in mice.

View full abstract

The role of intramolecular disulfide bonds in the fusion (F) protein of Sendai virus was studied. The 10 cysteine residues were changed to serine residues using site-directed mutagenesis. None of the cysteine mutant F proteins reacted with a monoclonal antibody specific for the mature conformation of the F protein, but eight of ten mutants reacted with an immature conformation-specific monoclonal antibody. The transport of these mutant proteins to the cell surface was drastically reduced. All of the cysteine mutant F proteins remained sensitive to endoglycosidase H (endo H) for 3 h after their synthesis. Moreover, cell surface transport of the hemagglutinin-neuraminidase (HN) protein co-expressed with each of these cysteine mutant F proteins was also reduced. These results suggest that all cysteine residues participate in the formation of intramolecular disulfide bonds, that co-translational disulfide bond formation is crucial to the correct folding and intracellular transport of the F protein, and that interaction of the F and HN proteins takes place intracellulary.

View full abstract

Treatment of Chinese hamster ovary (CHO) cells with phosphatidylserine (PS) caused cell death in a dose-dependent manner. Other phospholipids, such as phosphatidylcholine, phosphatidylethanolamine, phosphatidylinositol, and phosphatidic acid, had no effect on cell viability. The cells incubated with PS became round and underwent a dramatic reduction of cellular volume while maintaining the membrane containment of cellular contents. The PS-treatment induced chromatin condensation and extensive DNA fragmentation, with a pattern characteristic of internucleosomal fragmentation on agarose gel electrophoresis. These results indicate that PS-treatment induces apoptosis of CHO cells. This apoptosis-inducing activity was highly specific for PS, and neither of the synthetic PS analogs 1, 2-diacyl-sn-glycero-3-phospho-D-serine (D-PS) and 2, 3-diacyl-sn-glycero-l-phospho-L-serine induced apoptosis. Analysis using fluorescence-labeled phospholipids showed that both PS and n-PS were taken up equally and then transported to intracellular membranes, suggesting that the PS-specific induction of apoptosis was not the result of its specific internalization. These observations suggest that certain molecules which may recognize the stereo-specific configuration of PS are involved in the apoptotie process triggered by PS.

View full abstract

The migration inhibitory factor-related proteins (MRPs) MRP-8 and MRP-14 were detected in differentiated human leukemia cell lines (THP-1 and HL-60) by immunocytochemical analysis. They were induced and colocalized in the cytoplasm and in lesser amounts in the nucleus when THP-1 and HL-60 cells were induced to differentiate by 1α, 25-dihydroxy-vitamin D₃ or retinoic acid. In a search for a protein capable of binding MRP_s, both MRP_s were individually produced in insect cells (Sf21) infected with recombinant baculovirus. The purified recombinant MRPs were electrophoretically and antigenically indistinguishable from the native proteins, and their ability to form the MRP8/14 complex was retained. The presence of MRP binding sites was investigated by a binding assay using recombinant MRPs and specific monoclonal antibodies. MRP binding sites were detected on the cell membrane of the human leukemia cell lines THP-1, Raji, and MOLT-4. HL-60 cells treated with 1α, 25-dihydroxyvitamin D₃ did not express MRP binding sites on the cell membrane, but a high level of MRPs accumulated in the cells. The occurrence of MRP binding sites on the cell surface of leukemia cell lines of monocyte and lymphocyte origin suggests that MRPs, released from neutrophils under certain conditions, may contribute to the activation and recruitment of effector cells to inflammatory lesions.

View full abstract

Although isopentenyl diphosphate is a precursor of isoprenoids in Escherichia coli, the genes and enzymes involved in its biosynthesis have not been identified. Thus, we tried to isolate E. coli mutants deficient in the biosynthesis and their complementary genes by use of an artificial phenotypic screening system employing three carotenoid biosynthetic genes, crtE crtB, and crtl. Cells were mutagenized with ethylmethanesulfonate, then transformed with a plasmid for expression of the carotenogenic genes. Mutants deficient in biosynthesis of isopentenyl diphosphate were expected to form white colonies, because they are unable to produce enough lycopene, whereas wild-type cells form red colonies. Among large numbers of red colonies, we identified 117 white colonies. Next, we transformed each mutant with an E. coli genomic library. Twenty-nine complementary genes that restore red color of host colonies were isolated. A homology search and further complementation study using subcloned genes revealed that the true complementary genes encode isopentenyl diphosphate isomerase, subunits of ATP synthase, enzymes of the Krebs cycle, some aldehyde dehydrogenases, phosphate acetyltransferase, and enzymes which relate to the biosynthesis of ubiquinones and menaquinones. Two unknown genes were also found, designated elb1 and 2, which may be involved in the early steps of isoprenoid biosynthesis.

View full abstract

L-Phenylalanine oxidase from Pseudomonas sp. P-501 was irreversibly inactivated by the arginine-specific reagents, phenylglyoxal (PGO) and p-hydroxyphenylglyoxal (HPG). The inactivation by PGO and HPG follows pseudo-first-order kinetics with second-order rate constants of 10.6 and 15.1M^1-•min^-1, respectively, and a single arginyl residue was modified specifically. The effective protection by substrate L-phenylalanine against the inactivation by these reagents strongly suggests that the arginyl residue is located in the substrate binding site. SDS/PAGE analysis of the enzyme modified with [¹⁴C] PGO revealed that the arginyl residue was in the β subunit of the enzyme. The fragment containing the ¹⁴C-labeled arginyl residue was purified from the enzymatic digests of the labeled β subunit by HPLC and sequenced. The modification of Arg-35 in the β subunit was identified. The sequence around Arg-35 shows homology to the corresponding regions of tryptophan-2-monooxygenases.

View full abstract

Fluorescence resonance energy transfer (FRET) spectroscopy has been used to determine spatial relationships between residues on tropomyosin and actin in reconstituted muscle thin filament, and to detect a positional change of tropomyosin relative to actin on the thin filament in the presence and absence of Ca²⁺ ions. In addition to Cys-190 which is a single cysteine residue in rabbit skeletal muscle α-tropomyosin, a new site, Cys-87 which is a unique cysteine residue in a mutant α-tropomyosin, was labeled with a resonance energy donor molecule, 5-(2-iodoacetylaminoethyl)aminonaphthalene 1-sulfonic acid (IAED-ANS). On the other hand, Gln-41, Lys-61, Cys-374, and the ATP-binding site of actin were selectively labeled with acceptor probes: fluorescein cadaverine, fluorescein 5-isothiocyanate, 4-dimethyl-aminophenylazophenyl 4'-maleimide, and TNP-ATP (or TNP-ADP), respectively. The distances between probes attached to position 87 of the mutant tropomyosin and Gln-41, Lys-61, Cys-374, or the nucleotide-binding site of actin on the reconstituted thin filament in the presence of Ca²⁺ ion were measured to be 43.2, 49.7, 45.4, and 35.2 Å, respectively, and the distance between probes attached to position 190 of tropomyosin and Gin-41 or the nucleotide-binding site of actin were 51.6 and 43.1 Å, respectively. The transfer efficiencies between these donor and acceptor molecules were large, so that the efficiency should be very sensitive to changes in distance between probes attached to tropomyosin and actin. However, the transfer efficiency did not change appreciably upon removal of Ca²⁺ ions, suggesting that tropomyosin does not change its position on the reconstituted thin filament in response to a change in Ca²⁺ ion concentration. The present results do not support the notion of tropomyosin movement on skeletal muscle thin filaments as proposed in the steric blocking theory.

View full abstract

The plant hormone ethylene is generated from a unique precursor, l-aminocyclopropane-l-carboxylate (ACC). In previous studies, ACC deaminase, which degrades ACC to α-keto-butyrate and ammonia, was found in four strains of Pseudomonas, characterized, and sequenced. To verify the wider distribution of ACC deaminase in microorganisms, we purified and sequenced ACC deaminase from the yeast Hansenula saturnus. The purified enzyme was active toward ACC, D-serine and dl-coronamic acid, indicating the same stereospecificity as the Pseudomonas enzyme, but unlike the bacterial enzyme it was not active toward β-chloro-D-alanine and O-acetyl-D-serine. Analyses of peptides from proteolytic digests of the purified and modified ACC deaminase covered more than 90% of its amino acid sequence and showed a blocked N-terminal residue as N-acetylserine. A eDNA encoding the ACC deaminase was isolated from H saturnus cells incubated in α-aminoisobutyrate medium, and sequenced. The yeast enzyme has 441 amino acid residues, of which 60 to 63% are identical to those of reported Pseudomonas enzymes. The open reading frame encoding ACC deaminase was subcloned into pET-lid and expressed in Escherichia coli BL21 (DE3) as an active enzyme.

View full abstract

Lysophosphatidylcholine (lysoPC), a component of oxidatively modified lipoproteins, is present in atherosclerotic lesions, and its proatherogenic properties have been demonstrated. To gain an insight into lysoPC-mediated endothelial gene expression, we applied nonradioactive differential display analysis of mRNA from lysoPC-treated and untreated human umbilical vein endothelial cells. We identified 12 up-regulated distinct genes including 5 cell growth-related genes (two phosphatases CL100 and B23/hVH-3, graven, activating transcription factor-4, and heparin-binding epidermal growth factor-like growth factor), 3 thrombosis-related genes (plasminogen activator inhibitor-1, tissue plasminogen activator, and thrombomodulin), and 4 others (stanniocalcin, NAD-dependent methylenetetrahydrofolate dehydrogenase/methenyltetrahydrofolate cyclohydrolase, BENE, and reducing agents and tunicamycin-responsive protein). We isolated a full-length cDNA of human gravin. The cDNA sequence of gravin was homologous with rat mitogenic regulatory gene or rat protein kinase C binding protein and substrate, suggesting that gravin would regulate cell growth. Thus, lysoPC apparently accelerates atherosclerosis by regulating the expression of a wide variety of genes. Our data suggest the involvement in atherogenesis of the genes hitherto regarded as atherosclerosis-unrelated.

View full abstract

The removal of N-terminally located clusters of basic amino acids (N-tail) or C-terminally located clusters of basic amino acids (C-tail) from the midkine (MK) molecule severely reduced its neurite-promoting activity. However, experiments involving chemically synthesized MK derivatives revealed that the roles of the N-tail and C-tail were mostly indirect ones, i.e. they probably maintain the steric arrangements of the N-terminal and C-terminal halves. In particular, the C-domain, which is the C-terminal half devoid of the C-tail, retained considerable neurite-promoting activity when it was uniformly coated on a dish. The removal of the N-tail or C-tail also reduced the enhancing activity of plasminogen activator (PA) in aortic endothelial cells, although the effect was lower. There are two heparin-binding sites in the C-domain, Clusters I and II. A mutation in Cluster I [R⁷⁸→Q] affected the PA-enhancing activity only slightly, and a mutation in Cluster II [K⁸³K⁸⁴→QQ] abolished the activity, while both mutations are known to reduce the neurite-promoting activity moderately. Therefore, the two heparin-binding sites in the C-domain play different roles in these two activities. Indeed, heparin exhibited different effects on these two activities. We also observed that intact MK was required for ordered neurite-promotion along the path of MK; one possible interpretation of this is that the N-terminal half is necessary for the stability of the molecule. Furthermore, K⁷⁶ and K⁹⁹ were found to be required for the secretion of MK; i.e. mutants in which one of these K residues was changed to Q were produced in the host cells, but not found in the medium.

View full abstract

During a cyanocysteine-mediated dissection study of dihydrofolate reductase, a peptide fragment with a molecular mass of 18 Da less than expected was found as a major reaction product when the dissection reaction was applied to a Lys-cyanocysteine linkage. Detailed characterization of the dissection products by protease digestion, peptide sequencing, liquid chromatography/electrospray ionization mass spectrometry, and capillary electro-phoresis suggested that the by-product was generated via a lactam ring formation through the intramolecular nucleophilic attack of the ε-amino group on the carbonyl carbon of the Lys-cyanocysteine linkage. We have also demonstrated the occurrence of intermolecular attack of an α-amino group of glycine on the carbonyl carbon of the X-cyanocysteine linkage to form a new X-Gly linkage, which should be a useful reaction for specific modification of proteins at the C-terminal.

View full abstract

Perfringolysin O (θ-toxin) is a cholesterol-binding and pore-forming toxin that shares with other thiol-activated cytolysins a highly conserved sequence, ECTGLAWEWWR (residues 430-440), near the C-terminus. To understand the membrane-insertion and pore-forming mechanisms of the toxin, we evaluated the contribution of each Trp to the toxin conformation during its interaction with liposomal membranes. Circular dichroism (CD) spectra of Trp mutant toxins indicated that only Trp436 has a significant effect on the secondary structure, and that Trp436, Trp438, and Trp439 make large contributions to near-UV CD spectra. Quenching the intrinsic Trp fluorescence of the wild-type and mutant toxins with brominated lecithin/cholesterol liposomes revealed that Trp438 and probably Trp436, but not Trp439, contributes to toxin insertion into the liposomal membrane. Near-UV CD spectra of the membrane-associated mutant toxins indicated that both Trp438 and Trp439 are required for the CD peak shift from 292 to 300 nm, a signal related to θ-toxin oligomerization and/or pore formation, suggesting a conformational change around Trp438 and Trp439 in these processes.

View full abstract

We found glutamate racemase activity in cell extracts of Bacillus subtilis IFO 3336, which abundantly produces poly-γ-glutamate. The highest activity was obtained in the early stationary phase of growth. The racemase was purified to homogeneity. The enzyme was a monomer with a molecular mass of about 30 kDa and required no cofactor. It almost exclusively catalyzed the racemization of glutamate; other amino acids, including alanine and aspartate but not homocysteinesulfinate, were inactive as either substrates or inhibitors. Although the V_max value of the enzyme for L-glutamate is 21-fold higher than that for n-glutamate, the V_max/K_m value for L-glutamate is almost equal to that for the D-enantiomer. The racemase gene, glr, was cloned into Escherichia coli cells and sequenced. The racemase was overproduced in the soluble fraction of the E. coli clone cells with the substitution of ATG for TTG, the initial codon of the glr gene. D-Amino acid aminotransferase activity was not detected in Bacillus subtilis IFO 3336 cells. B. subtilis CU741, a leuC7 derivative of B. subtilis 168, showed lower glutamate racemase activity and lower productivity of poly-γ-glutamate than B. subtilis IFO 3336. These results suggest that the glutamate racemase is mainly concerned in D-glutamate synthesis for poly-γ-glutamate production in B. subtilis IFO 3336.

View full abstract

We previously detected a fucosylagalactobiantenna with a bisecting GlcNAc residue (BA-2) and one lacking the GIcNAc residue linked to the Manα1-3 residue of BA-2 (BA-1), which were enriched specifically in mouse brain [Shimizu, H., Ochiai, K., Ikenaka, K., Mikoshiba, K., and Hase, S. (1993) J. Biochem. 114, 334-338]. Pyridylamino sugar chains were prepared from mouse brains of various ages, and BA-1 and BA-2 were quantified after separation by HPLC. In cerebrum, BA-1 was scarcely expressed in newborn brain but gradually increased in amount during development, while expression of BA-2 reached a maximum 1 week after birth followed by a rapid decrease; in adult mice, the amount of BA-1 was almost the same as that of BA-2. In cerebellum, expression of BA-1 was lower than that of BA-2 at all stages. Glycoproteins with the BA-1 and BA-2 structures were enriched in the membrane fraction, and the glycoproteins solubilized were purified by lectin-affinity chromatography and gel filtration. The results indicated that BA-1 and BA-2 occurred in glycoproteins of more than 20 kDa in cerebellum, but most BA-1 and BA-2 were found in a 80-200 kDa fraction in cerebrum. These results show that the two brain-specific sugar chains are developmentally regulated and linked to the membrane-associated glycoproteins of subcellular organelles.

View full abstract

Polar flagellum-defective mutants (Pof^- Laf^-) have been isolated from a lateral flagelladefective mutant (Po⁺ Laf^-). Among these Pof^- Laf^- mutants, polar-filamentless mutants, which have the hook structure but not the filament, were identified by electron microscopy. Their hooks were covered with a sheath structure which is contiguous to the outer membrane. The filament proteins, flagellins, were shed into the culture medium of these mutants. These flagellins could be sedimented by high-speed centrifugation even after heat or low pH treatment whereas the depolymerized flagellin of the Pof⁺ strain was degraded by these treatments. After Triton X-100 treatment, most flagellin of the filamentless mutants could no longer be sedimented, and was degraded. We observed vesicle-like structures on the tips of the hooks and in the flagellin fraction sedimented by high speed centrifugation. These results suggest that flagellin of the filamentless mutants is not assembled into the tip of the hook, but is excreted together with a membrane structure which is probably the sheath of polar flagella.

View full abstract

Direct repeat motifs composed of two hexamer half-sites spaced by a single nucleotide (DR-1) are recognized by several members of the nuclear hormone receptor superfamily. We examined, by means of gene transfection assays, the interplay between the DR-1-binding nuclear receptors commonly expressed in liver, peroxisome proliferator-activated receptor α (PPARα), hepatocyte nuclear factor-4 (HNF-4), and chicken ovalbumin upstream transcription factor I (COUP-TFI). Both PPARα and HNF-4 efficiently bound to the acyl-CoA oxidase gene enhancer element, but PPARa exhibited much stronger transactivation than HNF-4. As a result, HNF-4 suppressed the gene-activating function of PPARα, when they were expressed together, due to competition for a common binding site. On the other hand, HNF-4, but not PPARα, effectively bound to the apolipoprotein CIII gene element, and activated gene transcription. PPARα had no effect even when co-expressed with HNF-4. COUP-TFI bound to both elements, and suppressed the gene activation by PPARα and HNF-4. Thus, these nuclear receptors have individual functions in gene regulation, and exhibit complex compound effects when they co-exist.

View full abstract

To elucidate the diversity of troponin C (TnC) during chordate evolution, we determined the organization of TnCs from the amphioxus, the lamprey, and the frog. Like the ascidian, the amphioxus possesses a single gene of TnC, and the fundamental gene structure is identical with the ascidian TnC. However, because alternative splicing does not occur in amphioxus, the potential for generation of TnC isoforms through this event arises only in the ascidian lineage. From the frog Xenopus laevis, two distinct cDNAs encoding fTnC isoforms and a single s/cTnC cDNA were determined. The duplication of the fTnC gene may be a character of only Xenopus or closely related species. The lamprey possesses two cDNAs each encoding fTnC and s/cTnC. The lamprey is the earliest diverged species among vertebrates, and thus it is supposed that the presence of both fTnC and s/cTnC is universal among vertebrate species, and that the gene duplication might have occurred at a vertebrate ancestor after the protochordate/vertebrate divergence. The position of the 4 th intron is 3.24/0 in protochordate TnC genes, but at 3.11/2 in vertebrate fTnCs and s/cTnCs. It is suggested that the 4 th intron sliding might have occurred prior to the gene duplication.

View full abstract

The expression of the human SREBP-2 gene is transcriptionally regulated in a cooperative manner by sterol regulatory element-binding proteins (SREBPs) and the general transcription factor NF-Y [Sato, R., Inoue, J., Kawabe, Y., Kodama, T., Takano, T., and Maeda, M. (1996) J. Biol. Chem. 271, 26461-26464]. To understand the sterol-dependent transcriptional regulation by these factors in detail, we have examined the regulation of the 3-hydroxy-3-methylglutaryl coenzyme A (HMG CoA) synthase and squalene synthase genes, whose promoters have multiple potential sterol regulatory elements (SRE, SREBP binding site) and NF-Y binding sites. The promoter of the human HMG CoA synthase gene was cloned, sequenced, and functionally characterized by means of reporter gene assays. The results indicate that an inverted CCAAT box, two SRE motifs and two Sp1 sites localized in a 90-bp region coordinately regulate the transcription. In the case of the human squalene synthase promoter, two SRE motifs and an inverted CCAAT box between the motifs localized in a 51-bp region are responsible for the sterol-regulated transcription of the gene. Gel mobility shift assay reveals that these two inverted CCAAT boxes are recognized by NF-Y. The involvement of multiple responsive elements in the transcription of HMG CoA synthase and squalene synthase seems to induce a higher level of sterol-dependent regulation (3.5 to 5.8-fold) compared with that of the SREBP-2 promoter, which contains a single pair of SRE motif and CCAAT box (1.8 to 2.6-fold). Reporter gene assays using constructs containing various nucleotide spacing lengths between the SRE motif and the CCAAT box demonstrate that the 16 to 20-bp spacing range is required for maximal transcriptional regulation. These results agree with the findings that the distances between the two motifs in the known sterol responsive elements in several genes, including the human HMG CoA synthase and squalene synthase genes, are in this range.

View full abstract

Recent evidence suggests a critical role for Syk in mast cell activation upon high affinity IgE receptor (FceRI) aggregation. A rat basophilic leukemia cell line, RBL-2113, expresses similar levels of two Syk isoforms that differ with respect to the presence of a 23-amino acid insert within the “linker” region located between the second Src homology 2 and the catalytic domain. Although they exhibit comparable intrinsic enzymatic activity, functional differences between the two isoforms are unknown. Here we report that the deleted Syk isoform can mediate signal transduction in RBL-2H3 cells. Aggregation of chimeric kinase, consisting of either form of Syk fused to the transmembrane and extracellular domains of guinea pig type II IgG Fc receptor, on RBL transfectants resulted in degranulation, release of leukotrienes, and enhanced gene expression of tumor necrosis factor-α. The chimeras as well as phospholipase C-γ1 and Vav became tyrosine-phosphorylated upon aggregation of chimeras. We also found that both Syk isoforms from transiently transfected COS-7 cells were capable of binding to phosphorylated FcERI, and their kinase activities were similarly up-regulated in the presence of tyrosine-phosphorylated synthetic peptides based on the sequence of the y subunit of FceRI. Thus, these results establish that both isoforms of Syk can mediate signal transduction in mast cells and suggest that the 23-amino acid insert in the linker region of Syk may not be obligatory for FceRI signaling.

View full abstract

It was shown that proteins modified with advanced glycation end products (AGE) are effectively endocytosed by macrophages or macrophage-derived cells in vitro, and immunohistochemical studies involving anti-AGE antibodies demonstrated the accumulation of AGE-modified proteins (AGE-proteins) in macrophage-derived foam cells in human atherosclerotic lesions in situ, suggesting the involvement of AGE-modified LDL in the atherogenic process in vivo. To examine this suggestion, LDL was modified with glycolaldehyde, a highly reactive intermediate of the Maillard reaction. Physicochemically, glycolaldehyde-modified LDL (GA-LDL) was characterized by increases in negative charge, fluorescence intensity, and reactivity to anti-AGE antibodies, properties highly similar to those of AGE-proteins. The cellular interaction of GA-LDL with mouse peritoneal macrophages showed that GA-LDL was specifically recognized and endocytosed, followed by lysosomal degradation. The endocytic uptake of GA-LDL by these cells was competitively inhibited by acetylated LDL (acetyl-LDL), and the endocytic degradation of acetyl-LDL was also competed for by GA-LDL. Furthermore, incubation of GA-LDL with these macrophages and Chinese hamster ovary cells overexpressing the macrophage scavenger receptor (MSR), but not with peritoneal macrophages from MSR-knockout mice, led to the intracellular accumulation of cholesteryl esters (CE). These results raised the possibility that AGE-modified LDL, if available in situ, is taken up by macrophages mainly via MSR and then contributes to foam cell formation in early atherosclerotic lesions.

View full abstract