Shinji Sueda

Structure of the biotin carboxylase subunit of pyruvate carboxylase from Aquifex aeolicus at 2.2 Å resolution

Pyruvate carboxylase (PC) is distributed in many eukaryotes as well as in some prokaryotes. PC catalyzes the ATP-dependent carboxylation of pyruvate to form oxalacetate. PC has three functional domains, one of which is a biotin carboxylase (BC) domain. The BC subunit of PC from Aquifex aeolicus (PC-beta) was crystallized in an orthorhombic form with space group P2(1)2(1)2, unit-cell parameters a = 92.4, b = 122.1, c = 59.0 A and one molecule in the asymmetric unit. Diffraction data were collected at 100 K on BL24XU at SPring-8. The crystal structure was determined by the molecular-replacement method and refined against 20.0-2.2 A resolution data, giving an R factor of 0.199 and a free R factor of 0.236. The crystal structure revealed that PC-beta forms a dimeric quaternary structure consisting of two molecules related by crystallographic twofold symmetry. The overall structure of PC-beta is similar to other biotin-dependent carboxylases, such as acetyl-CoA carboxylase (ACC). Although some parts of domain B were disordered in ACC, the corresponding parts of PC-beta were clearly determined in the crystal structure. From comparison between the active-site structure of ACC with ATP bound and a virtual model of PC-beta with ATP bound, it was shown that the backbone torsion angles of Glu203 in PC-beta change and some of water molecules in the active site of PC-beta are excluded upon ATP binding.

Detection of higher-ordered DNA sequence by using terbium(III) luminescence

The detection of higher-ordered sequences of DNA such as the dyad symmetrical sequence was investigated using Tb(III) luminescence, based on energy transfer from excited bases in a single stranded DNA probe. An oligonucleotide carrying an iminodiacetic acid function as metal chelating moiety was used as the probe. The Tb(III) complex with this probe gave luminescence characteristic of Tb(III). The intensity of this luminescence was considerably suppressed when bound to the targeted double stranded DNA to form a triple helix. This behavior was especially pronounced when the probe forms a 2:1 (probe strand: Tb(III)) complex on the target double strand carrying a dyad symmetrical base sequence. When the analyte DNA strand does not contain a complementary sequence to the probe, the intensity of the luminescence was scarcely influenced. This characteristic luminescence change will be useful for detecting target genes having a higher-ordered sequence.

Pyridoxine biosynthesis in yeast: participation of ribose 5-phosphate ketol-isomerase.

To identify the genes involved in pyridoxine synthesis in yeast, auxotrophic mutants were prepared. After transformation with a yeast genomic library, a transformant (A22t1) was obtained from one of the auxotrophs, A22, which lost the pyridoxine auxotrophy. From an analysis of the plasmid harboured in A22t1, the RKI1 gene coding for ribose 5-phosphate ketol-isomerase and residing on chromosome no. 15 was identified as the responsible gene. This notion was confirmed by gene disruption and tetrad analysis on a diploid prepared from the wild-type and the auxotroph. The site of mutation on the RKI1 gene was identified as position 566 with a transition from guanine to adenine, resulting in amino acid substitution of Arg-189 with lysine. The enzymic activity of the Arg189-->Lys (R189K) mutant of ribose 5-phosphate ketolisomerase was 0.6% when compared with the wild-type enzyme. Loss of the structural integrity of the protein seems to be responsible for the greatly diminished activity, which eventually leads to a shortage of either ribose 5-phosphate or ribulose 5-phosphate as the starting or intermediary material for pyridoxine synthesis.

Synthetic DNA Ligands Conjugated with Metal Binding Moiety. Regulation of the Interaction with DNA by Metal Ions and the Ligand Effect on Metal Assisted DNA Cleaving

Abstract Various functionalized molecules were synthesized by covalently conjugating DNA intercalators with metal chelators which carry such metal-binding functional groups as ethers, polyamines, and carboxylates. Systematic studies were made on the interaction of these conjugates with DNA in the presence and in the absence of metal ions. DNA binding of the conjugates were strongly affected by coexisting metal ions. This can be accounted for by the change in the net charge and in the conformation of the conjugates through complexation with metal ions. In some of the complexes thus obtained (DNA-intercalator-metal ion ternary complexes), DNA underwent a metal assisted cleavage of the strand, the activities of which were highly dependent on the nature of metal chelating moieties adopted. It is likely that a residual Lewis acidity on the metal in the complex and the strain in the DNA backbone induced by intercalation contribute to facilitate DNA cleavage.

Development of a heme sensor using fluorescently labeled heme oxygenase-1

Free heme, the protein-unbound form of heme, has both toxic and regulatory effects on cells. To detect free heme at low concentrations, we developed a heme sensor using fluorescently labeled heme oxygenase-1 (HO-1), an enzyme that catalyzes oxidative heme degradation and has a high affinity for heme. The response of the heme sensor is based on the fluorescence quenching that occurs when heme binds to the enzyme. Each of the three fluorescently labeled HO-1s exhibits a 1:1 binding stoichiometry and an absorption spectrum similar to that of the heme complex of the wild-type HO-1. Titration of the labeled proteins with hemin resulted in fluorescence quenching in a hemin concentration-dependent manner, presumably due to an energy transfer from the fluorophore to the heme bound to HO-1. The sensor showed a potent affinity for heme with a dissociation constant in the low nanomolar range and a high selectivity for heme. Based on the linear response of the sensor to heme, we performed a fluorometric microplate assay. The sensor was able to selectively detect free heme but did not respond to heme bound to native hemoglobin. This assay will be a useful tool for determination of free heme in biological samples containing protein-bound heme.

Structure of the biotin carboxylase domain of pyruvate carboxylase from Bacillus thermodenitrificans

The biotin carboxylase (BC) domain of pyruvate carboxylase (PC) from Bacillus thermodenitrificans (BC-bPC) was crystallized in an orthorhombic form (space group P2(1)2(1)2(1)), with unit-cell parameters a = 79.6, b = 116.0, c = 115.7 A. Two BC protomers are contained in the asymmetric unit. Diffraction data were collected at 100 K and the crystal structure was solved by the molecular-replacement method and refined against reflections in the 20.0-2.4 A resolution range, giving an R factor of 0.235 and a free R factor of 0.292. The overall structure of BC-bPC is similar to those of the BC subunits of Aquifex aeolicus PC (BC-aPC) and Escherichia coli ACC (BC-eACC). The crystal structure revealed that BC-bPC forms a unique dimeric quaternary structure, which might be caused as a result of the division of the BC domain from the rest of the protein. The position of domain B in BC-bPC differs from those in other enzymes of similar structure (BC-aPC and BC-eACC).

A unique biotin carboxyl carrier protein in archaeon Sulfolobus tokodaii

Biotin carboxyl carrier protein (BCCP) is one subunit or domain of biotin‐dependent enzymes. BCCP becomes an active substrate for carboxylation and carboxyl transfer, after biotinylation of its canonical lysine residue by biotin protein ligase (BPL). BCCP carries a characteristic local sequence surrounding the canonical lysine residue, typically ‐M‐K‐M‐. Archaeon Sulfolobus tokodaii is unique in that its BCCP has serine replaced for the methionine C‐terminal to the lysine. This BCCP is biotinylated by its own BPL, but not by Escherichia coli BPL. Likewise, E. coli BCCP is not biotinylated by S. tokodaii BPL, indicating that the substrate specificity is different between the two organisms.

Immobilization of immunoglobulin-G-binding domain of Protein A on a gold surface modified with biotin ligase.

Protein A from Staphylococcus aureus specifically binds to the Fc region of immunoglobulin G (IgG) and is widely used as a scaffold for the immobilization of IgG antibodies on solid supports. It is known that the oriented immobilization of Protein A on solid supports enhances its antibody-binding capability in comparison with immobilization in a random manner. In the current work, we developed a novel method for the oriented immobilization of the IgG-binding domain of Protein A based on the biotinylation reaction from archaeon Sulfolobus tokodaii. Biotinylation from S. tokodaii has a unique property in that the enzyme, biotin protein ligase (BPL), forms a stable complex with its biotinylated substrate protein, biotin carboxyl carrier protein (BCCP). Here, BCCP was fused to the IgG-binding domain of Protein A, and the resulting fusion protein was immobilized on the BPL-modified gold surface of the sensor chip for quartz crystal microbalance through complexation between BCCP and BPL. The layer of the IgG-binding domain prepared in this way successfully captured the antibody, and the captured antibody retained high antigen-binding capability.

A biotin-based protein tagging system.

Biotin protein ligase (BPL) mediates covalent attachment of biotin to a specific lysine residue of biotin carboxyl carrier protein (BCCP) of biotin-dependent enzymes. We recently found that the biotinylation reaction from thermophilic archaeon Sulfolobus tokodaii has a unique characteristic that the enzyme BPL forms a tight complex with the product, biotinylated BCCP (169 amino acid residues). In the current work, we attempted to apply this characteristic to a novel protein tagging system. Thus, the N terminus of S. tokodaii BCCP was truncated and the interaction of the resulting BCCP, BCCPDelta100 and BCCPDelta17 (with 69 and 152 residues, respectively), with BPL was investigated by surface plasmon resonance (SPR). It was found that the binding of BPL to the biotinylated BCCPDelta100 is extremely tight with a dissociation constant (K(D)) of 1.2 nM, whereas that to the unbiotinylated counterpart was moderate with a K(D) of 3.3 microM. Furthermore, chimeric proteins of glutathione S-transferase (GST) and green fluorescence protein (GFP) with BCCPDelta100 fused to their C terminus were prepared. The resulting fusion proteins were successfully biotinylated and captured on the BPL-modified SPR sensor chip or BPL-modified magnetic beads. The function of GST and GFP was hardly impaired on fusion with BCCPDelta100 and biotinylation of the latter.

Site-specific labeling of proteins by using biotin protein ligase conjugated with fluorophores.

Biotin protein ligase (BPL) mediates the covalent attachment of biotin to a specific lysine residue of biotin carboxyl carrier protein (BCCP). This biotinylation in Sulfolobus tokodaii is unique in that BPL forms a tight complex with the product, biotinylated BCCP, and this property was exploited for fluorescent labeling of a membrane protein. Thus, the truncated form of BCCP (BCCPΔ100, 69 residues) was fused to either the N or C terminus of the bradykinin B2 receptor (B2R). The resulting fusion proteins, BCCPΔ100–B2R and B2R–BCCPΔ100, respectively, were separately expressed in mammalian HEK293 cells, and labeled with BPL conjugated with a fluorophore: either fluorescein, DyLight549 or green fluorescent protein. The fusion proteins were biotinylated and bound to BPL, thereby giving rise to strong fluorescence along the periphery of the cell. Some were capable of binding bradykinin and an antagonist. When stimulated with the former, the receptor translocated to the cytosol; this suggests that the labeled receptor retains its integrity in terms of ligand‐binding and translocation.