Search

Galectin-9 is a lectin, which has various biological functions such as T-cell differentiation and apoptosis. Multivalency of carbohydrate binding is required for galectin-9 to function. Although galectin-1 (a proto-type galectin) forms an oligomer to obtain its multivalency, galectin-9 (a tandem-repeat-type one) has two carbohydrate recognition domains (CRD) in one polypeptide. However, a single CRD of galectin-9, especially the C-terminal one, exhibited pro-apoptotic activity suggesting oligomer formation capability. In this study, we monitored the nuclear magnetic resonance (NMR) signals of the backbone atoms of the galectin-9 C-terminal CRD (G9CCRD). Protein concentration dependence of the signals suggested that a region (F1-F4 strands) opposite to the ligand-binding site was involved in the self-association of G9CCRD. Site-directed mutagenesis in this region (Leu210, Trp277 and Leu279 to Thr; G9CCRD-3T) inhibited the self-association of G9CCRD, and improved the solubility, whereas it reduced its pro-apoptotic activity towards T cells. The high pro-apoptotic activity of G9CCRD seems to be due to the ability to form an oligomer. In addition, the same substitution in two-CRD-containing galectin-9 (G9Null-3T) also diminished the self-association and improved its solubility, although it hardly reduced the anti-proliferative and pro-apoptotic activities. G9CCRD contributes the self-association of full-length galectin-9 at high protein concentrations.

We previously showed that tandem-repeat type galectin-8, which has two covalently linked carbohydrate recognition domains (CRDs), induces neutrophil-adhesion through binding to integrin alphaM. Here, we analysed the function of galectin-8 in Jurkat T-cells. Galectin-8, as well as tandem-repeat galectin-9, and several multivalent plant lectins, induced Jurkat T-cell adhesion to a culture plate, whereas single-CRD galectins-1 and -3 did not. Galectin-8 also induced the adhesion of peripheral blood leucocytes to human umbilical vein endothelial cells. These results suggest that the di- or multi-valent structure of galectin-8 is essential for the induction of cell adhesion and that this ability exhibits broad specificity for leucocytes. The galectin-8-induced cell adhesion was accompanied by stress fibre formation, which suggests that intracellular signalling is required. We have identified integrin alpha4 as one of the candidate target molecules associated with the induction of cell adhesion. Indeed, inhibition of the function of integrin alpha4 by treating cells with a blocking-antibody reduced the sensitivity to galectin-8. Also, the phosphorylation of Pyk and ERK1/2, indicators of integrin-mediated signalling, was up-regulated on treatment with galectin-8. Thus, a primary target of galectin-8 must be the sugar chains on members of the integrin family, which are abundantly expressed on the surface of leucocytic cells.

Galectin-9, a mammalian lectin with affinity for beta-galactosides, is known as an apoptosis inducer of activated T lymphocytes. In the present study, we examined the properties of galectin-9-mediated cell death of Jurkat T cells. Galectin-9NC (wild-type), consisting of two CRDs (N-terminal and C-terminal carbohydrate recognition domains), and derivatives of it, galectins-9-NN and -9-CC, induced Jurkat T-cell apoptosis. However, a single CRD (galectin-9NT or -CT) had no effect, suggesting the stable dimeric structure of two CRDs is required for the activity. The apoptosis was inhibited by pretreatment with an N-glycan synthesis inhibitor, indicating that the expression of N-glycans in the cells is essential for galectin-9-induced apoptosis. We previously showed that the apoptosis of MOLT-4 cell is mediated by galectin-9 via a Ca(2+)-calpain-caspase-1-dependent pathway. In Jurkat cells, the cell death by galectin-9, was insufficiently suppressed by caspase inhibitors, Ca(2+)-chelator or calpain inhibitor. Furthermore, we observed the loss of mitochondrial membrane potential and significant AIF release in galectin-9-treated cells. These findings suggest that caspase-dependent and-independent death pathways exist in Jurkat cells, and the main pathway might vary with the T-cell type.

An artificial gene coding for the human initiation factor (eIF) 4E-binding protein 1 (4E-BP1) was chemically synthesized and cloned. Although the expression of the 4E-BP1 gene alone has not yet been accomplished, the gene was expressed in Escherichia coli [BL21(DE3)] as a fusion gene with the glutathione-S-transferase (GST) gene using a prokaryotic gene fusion vector (pGEX-4T-2), which contains a gene sequence coding the cleavage site for a specific protease, alpha-thrombin. The fusion gene product was purified to homogeneity by glutathione Sepharose-4B affinity column chromatography. It was shown by m7GTP- and glutathione-affinity chromatography that the binding ability of 4E-BP1 to eIF-4E is nearly the same as that to the eIF-4E x m7GTP complex, implying different binding sites of eIF-4E and its nonallosteric obligation for 4E-BP1 and mRNA cap structure. In contrast with the binding of eIF-4E to the mRNA cap structure, where some functional amino acids play an important role in the binding, the binding to 4E-BP1 was suggested to occur via multiple nonspecific interactions.

As reported previously, the synthetic heptapeptide having the amino acid sequence around the reactive Cys (SH1) of myosin heavy chain, IRICRKG-NH2, inhibited acto-myosin subfragment-1 (S-1) ATPase activity and half inhibition (K1/2) was observed at a peptide concentration of 0.06 mM. The inhibitory ability of the peptide was found to be decreased to one-fifth by acetylation of its N-terminal alpha-amino group. A similar effect of N-acetylation was observed with a nonapeptide, EGIRICRKG-NH2, and an undecapeptide, VLEGIRICRKG-NH2. These results indicate that N-terminal-free synthetic peptides do not act as proper analogs of the corresponding segment of S-1 heavy chain against F-actin. We isolated a longer peptide extending from Thr682 to Lys709 in S-1 heavy chain, with two Cys residues corresponding to SH1 and SH2. This peptide, having 28 residues (28peptide), inhibited acto-S-1 ATPase activity with a K1/2 of 0.23 mM. A cosedimentation binding assay indicated that the 28peptide completely dissociated acto-S-1 in the presence of ATP. This behavior is different from that observed with the N-terminal-free synthetic heptapeptide, and thus the 28peptide might be an analog of the corresponding segment. There is a possibility that the region corresponding to the 28peptide in S-1 heavy chain may bind directly with F-actin and may be involved in determining the acto-S-1 link during the steady state of the acto-S-1 ATPase reaction.

Spotted mackerel protamine, scombrine, was isolated from the sperm of a spotted mackerel (Scomber australasicus) by extraction with sulfuric acid and fractionated into one major (scombrine II) and one minor (scombrine I) components by chromatography on CM-Sephadex C-25. Scombrine II gave a single band, whereas scombrine I gave three bands upon PAGE. Scombrine II (scombrine gamma) consists of 34 amino acid residues, and its sequence is: Pro-Arg-Arg-Arg-Arg-Arg-Ala-Ser-Arg-Pro-Val-Arg-Arg-Arg-Arg-Arg- Ala-Arg-Arg-Ser-Thr-Ala-Val-Arg-Arg-Arg-Arg-Arg-Val-Val-Arg-Arg-Arg-Arg. The ion spray mass spectrum shows that scombrine gamma has a molecular mass of 4,532.13 Da. The other minor component (scombrine I) is considered to be a mixture of degradation products of scombrine gamma based on the results of PAGE and ion spray mass spectrometry. Scombrine gamma has a similar sequence to sardaine Z2 from striped bonito except for two positions.

Striped bonito protamine, sardaine, was isolated from the sperm of striped bonito (Sarda orientalis) by extraction with sulfuric acid followed by ion-exchange chromatography. The preparation gave a single band upon polyacrylamide gel electrophoresis. Sardaine consists of 34 amino acid residues, and its sequence is: Pro-Arg-Arg-Arg-Arg-Arg-Ser(Ala)-Ser-Arg-Pro-Val-Arg-Arg-Arg-Arg-Arg-Tyr -Arg- Arg-Ser-Thr-Ala-Ala-Arg-Arg-Arg-Arg-Arg-Val-Val-Arg-Arg-Arg-Arg. At position 7, serine (sardaine Z1) is partially replaced by alanine (sardaine Z2). The ion spray mass spectrum shows that sardaines Z1 and Z2 have molecular masses of 4,612.49 and 4,596.09 Da, respectively. The sequence of sardaine Z1 is 100% identical with that of thynnine Z2 from tuna fish (both fish belong to Scombridae, Perciformes).

The heptapeptide Ile-Arg-Ile-Cys-Arg-Lys-Gly-OEt is the analog of the S-site, one of the actin-binding sites in myosin [Suzuki et al. (1987) J. Biol. Chem. 262, 11410-11412]. Various substituted heptapeptides were synthesized, and the dissociation constants of each acto-heptapeptide complex was measured. Comparison of the dissociation constants indicated that the hydrophobic side chain of Ile-1 was critical for the binding with F-actin, but not that of Ile-3. The positive charge and the side chain length of Arg-2 were also important. The presence of a sulfur atom in the Cys-4 was also necessary. The affinity of the N-terminal Ile-Arg-Ile part for F-actin was influenced by the kind of residues in the C-terminal tetrapeptide part. Based on these results, the side chains of Ile(702), Arg(703), and Cys(SH1)(705) in myosin subfragment-1 heavy chain were assigned to be critical for the binding with F-actin. The amino acid sequence of S-1 heavy chain containing these critical residues for the S-site from residue number 700 to 717 can be predicted as an analogue of the segment B of the ATP-binding site [Walker et al. (1982) EMBO J. 1, 945-951]. The actin-binding S-site possibly shares a part of the ATP-binding site in myosin. We discuss the possibility that the S-site is an inhibitory site of myosin ATPase and the so-called actin-activation of myosin ATPase is a deinhibition induced by transient binding of F-actin to the S-site.

It was previously reported that in chromatophores of Rhodospirillum rubrum, reaction center, which consists of three kinds of protein (Mm, about 78K), is a small fragment of a large protein complex (PRU; photoreaction unit), which contains six other kinds of protein including light-harvesting bacteriochlorophyll protein, has Mm of about 700K and is free of phospholipid [J. Biochem. 86, 1211-1224 (1979); 94, 1815-1826 (1983(]. In the present study, the photosynthetic, cyclic electron transport system sensitive to antimycin A was effectively reconstructed by incubating 60 nM PRU (which contained 1 mol of reaction center and 2 mol of ubiquinone-10 per mol) with 300 nM each of oxidized ubiquinone-10 protein, reduced cytochrome c2 and lipoamino acid (which were all purified from Rhodospirillum rubrum) in the presence of low concentrations of cholate and deoxycholate (pH 8.0). In the light, the cytochrome was oxidized while the quinone was reduced. The oxidation and reduction each progressed rapidly at first, then slowly, reaching maxima (steady states) 1-2 min after the light had been turned on. At the steady states, 30% of the cytochrome was oxidized while 11% of the total quinone was reduced. When the light was turned off, the original oxidation-reduction states of the cytochrome and quinone were restored at rapid rates initially then at slow rates. Antimycin A stimulated the slow rates in the light-on state and depressed them in the light-off state, but did not influence the fast rates. Ubiquinone-10 protein was required for the antibiotic-sensitive, slow oxidation reactions. This indicates that the slow rates were due to cyclic electron transport. Cytochrome c2 was tightly bound to PRU at a molar ratio of 1:1. This cytochrome as well as the quinone bound to PRU was responsible for the fast rates. PRU had other sites able to bind cytochrome c2 and ubiquinone-10 protein with Km of 0.4 and 0.1 microM, respectively. Of the polar lipids tested, lipoamino acid was the most effective for reconstruction, and its effect was maximal at 300 nM, which is far below its critical micelle concentration.

It is known that the coupling factor purified from the acetone powder of chromatophores from Rhodospirillum rubrum shows ATPase activity in the presence of Ca(2)+, but not in the presence of Mg(2)+ or Mn(2)+. The present study deals with conditions, under which the Ca(2)+-ATPase activity is reversibly converted into Mg(2)+- and Mn(2)+-ATPase activites with the purified coupling factor. 1. Of the pH indicators tested, 6 kinds coverted the Ca(2)+-ATPase activity into Mg(2)+- and Mn(2)+-ATPase activities in the order, ethyl orange greater than tropaeolin 000 greater than or equal to metanil yellow greater than tropaeolin 00 greater than ethyl red greater than or equal to bromthymol blue. 2. Of the detergents tested, those other than Triton X-100 and Brij 58 caused the conversion described above; dodecylsulfonate was most effective, whereas dodecylpyridinium chloride was moderately effective. 3. 2,4-Dinitrophenol stimulated approximately two-fold the Ca(2)+-ATPase activity, but not the Mg(2)+- or Mn(2)+-ATPase activity at all. However, in the presence of dodecylpyridinium chloride, the pH indicator remarkably stimulated the Mg(2)+- and Mn(2)+-ATPase activities, accompanied with a partial inhibition of the Ca(2)+-ATPase activity. Methyl red and ethyl red showed similar effects. 4. All the nucleoside triphosphates tested can serve as the substrate. ATP was most effective for the Ca(2)+-ATPase activity, whereas dATP was most effective for the Mg(2)+- and Mn(2)+-ATPase activities induced by ethyl orange. 5. In the presence of ethyl orange, the ATPase activity was induced by various divalent cations in the following order of effectiveness, Mg(2)+ greater than Zn(2)+ greater than CO(2)+ greater than Mn(2)+ greater than Ni(2)+. 6. The mechanism of the reversible conversion from the Ca(2)+-ATPase activity to the Mg(2)+- and Mn(2)+-ATPase activities by pH indicators and detergents is discussed.

1. The light-induced pH change of chromatophore suspensions from Rhodospirillum rubrum was stimulated significantly and similarly by KCl, NaCl, LiCl, RbCl, CsCl, MgCl2, MnCl2, and CaCl2. In the dark, the pH of chromatophore suspensions decreased immediately and markedly on adding these salts. 2. The light-induced pH change stimulated by KCl plus valinomycin was inhibited by LiCl and NaCl, but not by RbCl. 3. The optimum pH values for light-induced pH change and photosynthetic ATP formation were around 5 and 8, respectively. The amount of chromatophore-bound ubiquinone-10 reduced in the light was independent of pH from 5 to 9. At pH 8, the number of protons incorporated into chromatophores in the light was one-half of the number of ubiquinone-10 molecules reduced in the light. 4. Among several pH indicators tested, bromothymol blue (BTB) and neutral red (NR) showed absorbance changes on illumination of chromatophores. Although the pH change indicated by the absorbance change was opposite to the light-induced pH change of the medium, the effect of KCl on the absorbance changes of BTB and NR, and the effect of valinomycin on that of NR, but not on that of BTB, were similar to those on the light-induced pH change. 5. The light-induced absorbance change of BTB was significantly inhibited by NR, whereas that of NR was hardly influenced by BTB. 6. Oligomycin stimulated the light-induced absorbance change of BTB under either non-phosphorylating or phosphorylating conditions. On the other hand, that of NR under phosphorylating conditions was 50% of that under non-phosphorylating conditions, and was increased by oligomycin.

1. The membrane of Rhodospirillum rubrum chromatophores was disintegrated with mild detergents (cholate and deoxycholate) in order to study the spatial arrangement of the functional proteins in the photochemical apparatus and the electron transport system in the membrane. 2. The components solubilized from the membrane by a mixture of cholate and deoxycholate (C-DOC) were separated into four fractions by molecular-sieve chromatography in the presence of C-DOC; they were designated as F1, F2, F3, and F4 in the order of elution. The fractions were further purified by repeated molecular-sieve chromatography in the presence of C-DOC until each fraction was chromatographically homogeneous. 3. F1 appeared to be conjugated forms of F2. 4. The purified F2 was composed of a rigid complex having a weight of 7 X 10(5) daltons, containing approximately 10 different kinds of protein species with molecular weights of 3.8 X 10(4), 3.6 X 10(4), 3.5 X 10(4), 2.8 X 10(4), 2.7 X 10(4), 2.6 X 10(4), 1.3 X 10(4), 1.2 X 10(4), 1.1 X 10(4), and 1.0 X 10(4). The complex contained 33 bacteriochlorophylls, 4 iron atoms, and 90 phosphates, but no cytochrome, ubiquinone, or phospholipid. It showed the same reaction center activity as chromatophores, indicating that the complex was a unit of the photochemical apparatus (photoreaction unit). Each chromatophore of average size was estimated to possess about 24 photoreaction units. 5. The purified F3 showed an absorbance spectrum characteristic of reaction centers, and contained 3.4 bacteriochlorophylls, 2.0 bacteriopheophytins, and 1.9 acid-labile iron atoms, but no cytochrome or ubiquinone (C-DOC reaction center). It had a weight of 1.2 X 10(5) daltons, and the main components were 4 protein species with molecular weights of 2.8 X 10(4), 2.7 X 10(4), 2.6 X 10(4), and 1.0 X 10(4). 6. The purified F4 showed a molecular weight of about 11,000, and contained one mole of ubiquinone-10 per mole (ubiquinone-10 protein). 7. The reaction center activity of C-DOC reaction centers was stimulated by ubiquinone-10 protein. In addition, the reaction center oxidized reduced cytochrome c2 in the light, provided that ubiquinone-10 protein was present (photo-oxidase activity).

The substrate specificity in the hydrolysis of L-, DL-, and D-BAPA (benzoylarginine-p-nitro-anilide) by copoly (L-Cys, L-Glu) and copoly (D-Cys, D-Glu) was studied, and enzyme-like stereospecific hydrolyses by poly-alpha-amino acids were identified for the first time. The L-type copolymer hydrolyzed L-BAPA faster than D-BAPA and the rates (v) of BAPA hydrolyses by L-type copolymer were found to be in the order vL greater than vDL greater than vD. On the other hand, the D-type copolymer hydrolysed D-BAPA faster than L-BAPA and the rates of BAPA hydrolyses by D-type copolymer were in the order vD greater than vDL greater than vL. In all cases, the reaction followed Michaelis-Menten kinetics when the substrate concentration was corrected, and the optimum conditions of the reaction were pH 6.0 and 40 degrees. The activity appeared after a certain amount of BAPA had combined with the polymer. D- and L-substrates combine competitively with the polymer and the different rates of hydrolysis are presumably due to the different substrate configurations in relation to the conformation of the active site in the polymer. The polymer shows activity near the range of random coil conformation, where some alpha-helical conformation is still present. Only some of the cysteine residues in the copolymer are involved in the hydrolytic activity.

A new arginine derivative, N-benzyloxycarbonyl-L-phenylalanyl-L-valyl-L-arginine-p-nitroanilide hydrochloride (ZPVAPA.HCl) was synthesized by the condensation of N-benzyloxy-carbonyl-L-phenylalanyl-L-valine and L-arginine-p-nitroanilide dihydrochloride using dicyclohexylcarbodiimide as a coupling reagent and 1-hydroxy-benzotriazole as an additive. L-ZPVAPA.HCl was split by trypsin more readily than Na-benzyloxycarbonyl-L-arginine-p-nitroanilide hydrochloride (L-ZAPA, HCl), Na-benzoyl-L-arginine-p-nitroanilide hydrochloride (L-BAPA.HCl), Na-tosyl-L-arginine-p-nitroanilide hdyrochloride (L-TAPA.HCl) and Na-benzoyl-DL-arginine-p-nitroanilide hydrochloride (DL-BAPA.HCl) by factors of 100, 400, 600, and 1,200, respectively. Low concentrations of dimethyl formamide (DMF) enhanced the trypsin-catalyzed hydrolyses of L-ZAPA.HCl and L-TAPA.HCl, contrary to the findings of other authors that DMF has no effect on the tryptic hydrolysis.