Structure-function studies of Escherichia coli biotin synthase via a chemical modification and site-directed mutagenesis approach.

In Escherichia coli, biotin synthase (bioB gene product) catalyzes the key step in the biotin biosynthetic pathway, converting dethiobiotin (DTB) to biotin. Previous studies have demonstrated that BioB is a homodimer and that each monomer contains an iron-sulfur cluster. The purified BioB protein, however, does not catalyze the formation of biotin in a conventional fashion. The sulfur atom in the iron-sulfur cluster or from the cysteine residues in BioB have been suggested to act as the sulfur donor to form the biotin molecule, and yet unidentified factors were also proposed to be required to regenerate the active enzyme. In order to understand the catalytic mechanism of BioB, we employed an approach involving chemical modification and site-directed mutagenesis. The properties of the modified and mutated BioB species were examined, including DTB binding capability, biotin converting activity, and Fe(2+) content. From our studies, four cysteine residues (Cys 53, 57, 60, and 97) were assigned as the ligands of the iron-sulfur cluster, and Cys to Ala mutations completely abolished biotin formation activity. Two other cysteine residues (Cys 128 and 188) were found to be involved mainly in DTB binding. The tryptophan and histidine residues were suggested to be involved in DTB binding and dimer formation, respectively. The present study also reveals that the iron-sulfur cluster with its ligands are the key components in the formation of the DTB binding site. Based on the current results, a refined model for the reaction mechanism of biotin synthase is proposed.

Molecular cloning and characterization of a unique 60 kDa/72 kDa antigen gene encoding enzyme I of the phosphoenolpyruvate: sugar phosphotransferase system (PTS) of Mycoplasma hyopneumoniae.

The recombinant clone expressing a 60 kDa (P60) antigen was isolated from Escherichia coli by screening a lambda EMBL3 genomic library using rabbit produced antiserum against Mycoplasma hyopneumoniae. Sequence analysis revealed that an interrupted (by a UGA codon) open reading frame coding for a 72 kDa protein (P72) may contain the P60 antigen gene. Western blot analysis with an anti-P60 monospecific antibody confirmed the presence of a P72 antigen from the total protein of M. hyopneumoniae, and a 72 kDa protein was also expressed in E. coli after changing the codon (UGA to UGG) by site-directed mutagenesis. BLAST (Basic Local Alignment Search Tool) comparison showed that the amino acid sequences of P72 share approximately 70% homology with the phosphotransferase enzyme I (PTSI) of bacteria and other mycoplasma species. The biological function of the P72 cytosolic protein was further confirmed by complementation using an E. coli ptsI mutant. The bacterial phosphoenolpyruvate-sugar phosphotransferase system (PTS) is known to mediate the uptake and phosphorylation of carbohydrates and to be involved in signal transduction. The immune responses of specific pathogen free (SPF) pigs and farm animals toward this unique antigen were observed. The transcription start positions of the PTSI gene were determined in M. hyopneumoniae and E. coli by primer extension experiments and the promoter site was also predicted.

Bioassay of biotin concentration with a Escherichia coli bio deletion mutant.

Biotin concentration was determined unequivocally with the E. coli bio mutant. The results demonstrate that this simple and efficient method can determine biotin concentration in the range of 10 pg to 50 ng/ml. The present method can also clearly distinguish biotin from its precursor and analog, dethiobiotin.

Nucleotide sequencing of S-RNA segment and sequence analysis of the nucleocapsid protein gene of the newly isolated Akabane virus PT-17 strain.

The nucleotide sequences of the S-RNA of Akabane viruses JaGAr-39, OBE-1, Iriki and the newly isolated PT-17 strains and the Aino virus were determined and compared. The results reveal that the S-RNAs of the four Akabane strains share 96.9% homology in nucleotide sequences. Only one amino acid difference out of the 233 amino acids of the nucleocapsid protein (N) and three amino acid differences in the 91 amino acids of the nonstructural protein (NSs) were found among the Akabane viruses. Amino acid sequences of N and NSs proteins of the Aino virus have approximately 80% identity as compared with the Akabane viruses. The results also demonstrate that the four Akabane viruses and the Aino virus can be clearly differentiated by RFLP (restriction fragments length polymorphism) analysis using RT-PCR generated nucleocapsid protein genes and digested with HaeIII and HindIII. The phylogenetic tree based on the UPGMA (Unweighted Pair Group Method with Arithmetic Mean) analysis of the sequences of nucleocapsid protein genes and the S-DNAs revealed that the newly isolated PT-17 strain is most closely related to Iriki strain, than the JaGAr-39 or OBE-1 strains.

Functional consequence of mutating conserved residues of the yeast farnesyl-protein transferase beta-subunit Ram1(Dpr1).

Ras proteins, fungal mating pheromones, and other proteins terminating in the sequence CaaX (where C is Cys, a is any aliphatic amino acid, and X is the C-terminal residue) are posttranslationally prenylated. Farnesyl-protein transferase (FPTase) transfers the farnesyl moiety of farnesyl pyrophosphate (FPP) to the thiol of the CaaX box cysteine in a reaction that requires Zn2+ and Mg2+. We have created mutations in conserved amino acids of the yeast Ram1 protein to identify residues important for Zn2+-dependent FPTase activity. Wild-type and mutant Ram1 proteins were expressed as operon fusions in bacteria, and FPTase activity was measured. Mutations in conserved residues Glu256, His258, Asp307, Cys309, Asp360, and His363 reduce FPTase activity. Asp307, Cys309, and His363 correspond to the residues that have been shown to coordinate Zn2+ in mammalian FPTase. The H258N mutant enzyme exhibited an increased sensitivity to the Zn2+ chelator 1,10-phenanthroline, required higher concentrations of Zn2+ to restore activity to the apoenzyme, and had a 10-fold reduction in catalytic efficiency. The decreases in FPTase activity observed do not appear to be caused by major structural perturbations because the mutants were stably expressed and retained the ability to interact with Ram2p during purification. The FPTase activity of the mutants measured in vitro correlated well with their ability to complement the mating and growth defects of a ram1Delta strain in vivo.

Farnesylation and proteolysis are sequential, but distinct steps in the CaaX box modification pathway.

Membrane localization of Ras proteins requires posttranslational modification of a conserved C-terminal sequence motif known as the CaaX box (C is Cys, a is any aliphatic amino acid, and X is the carboxyl terminal residue). The modification steps include farnesylation, removal of the three C-terminal amino acids, carboxyl-methylation, and palmitoylation. In yeast, the farnesyltransferase (FTase) is encoded by the RAM1(DPR1) and RAM2 genes, and the methyltransferase is the product of STE14. The gene encoding the protease(s) that is responsible for modification of the CaaX has not been identified. We have used in vitro-synthesized Ras2p and synthetic peptide substrates to investigate the relationship between farnesylation and proteolysis. Addition of yeast cytosolic extracts to rabbit reticulocyte extracts programmed to synthesize Ras2p led to prenylation of Ras2p and a change in electrophoretic mobility similar to that observed during Ras maturation in vivo. However, it was not possible to determine if the mobility shift is the result of prenylation, proteolysis or a combination of both steps. Therefore, we examined the relationship between farnesylation and proteolysis directly using extracts prepared from bacteria overexpressing the genes for the yeast FTase (RAM1 and RAM2) and synthetic CaaX box peptides. Extracts from bacteria expressing RAM1/RAM2 efficiently prenylate CaaX box peptides, but do not proteolyze the -aaX residues. However, addition of yeast extracts from wild type, ram1, or ste14 mutants resulted in the removal of the -aaX residues from prenylated CaaX box peptides.

A polybasic domain allows nonprenylated Ras proteins to function in Saccharomyces cerevisiae.

Ras proteins undergo a series of posttranslational modifications prior to association with the cytoplasmic surface of the plasma membrane. The modification steps include farnesylation, proteolysis, methylesterification, and palmitoylation. A 4-amino acid residue motif known as the CaaX box (C is cysteine, a is generally aliphatic, and X is the carboxyl-terminal residue) is the sequence recognized by the prenyl transferase that initiates the modification pathway. As part of our studies to define the requirements for Ras membrane association, we directed mutagenesis to the yeast Ras2 protein CaaX box to assess the relative importance of prenylation, palmitoylation, and stretches of basic amino acids on the function of the protein. The wild type yeast Ras2 protein terminates in the sequence Cys-Cys-Ile-Ile-Ser. We have identified mutations that do not contain a CaaX box but still encode functional Ras proteins. These mutations replace the terminal serine of the CaaX box with the sequence -Lys-Leu-Ile-Lys-Arg-Lys. Three mutants have been analyzed in detail. Ras2(CCIIKLIKRK) functions at a level similar to wild type Ras2, whereas cells expressing only Ras2(SCIIKLIKRK) and Ras2(SSIIKLIKRK) forms of Ras2 protein grow more slowly at 30 degrees C. In addition, strains expressing only Ras2(SSIIKLIKRK) protein fail to grow at 37 degrees C. Replacement of the basic residues with neutral amino acids (Ras2(CCIISIIS)) completely abolishes their ability to support Ras-dependent growth. The extension mutants are not prenylated, but Ras2(CCIIKLIKRK) and Ras2(SCIIKLIKRK) are palmitoylated. These results demonstrate that a diverse set of carboxyl-terminal sequence motifs and posttranslational modifications lead to functional Ras proteins in yeast.

Normal mitochondrial structure and genome maintenance in yeast requires the dynamin-like product of the MGM1 gene.

The isolation and characterization of MGM1, an yeast gene with homology to members of the dynamin gene family, is described. The MGM1 gene is located on the right arm of chromosome XV between STE4 and PTP2. Sequence analysis revealed a single open reading frame of 902 residues capable of encoding a protein with an approximate molecular mass of 101 kDa. Loss of MGM1 resulted in slow growth on rich medium, failure to grow on non-fermentable carbon sources, and loss of mitochondrial DNA. The mitochondria also appeared abnormal when visualized with an antibody to a mitochondrial-matrix marker. MGM1 encodes a dynamin-like protein involved in the propagation of functional mitochondria in yeast.