Purification and characterization of the fructose-inducible HPr-like protein, FPr, and the fructose-specific enzyme III of the phosphoenolpyruvate: sugar phosphotransferase system of Salmonella typhimurium

The proteins comprising the fructose-specific phosphoenolpyruvate:sugar phosphotransferase system were investigated using a strain of Salmonella typhimurium which lacks the general phosphotransferase system proteins, HPr and Enzyme I, synthesizes the fructose phosphotransferase system proteins, FPr, Enzyme IIfru, Enzyme IIIfru, and fructose-1-phosphate kinase, constitutively, and expresses the Enzyme I-like protein Enzyme I. Enzyme I activity was found in the cytoplasmic fraction, Enzyme IIfru in the membrane fraction, and FPr and Enzyme IIIfru activities were distributed between the two fractions. Extraction of membranes with butanol and urea led to quantitative release of the membrane-associated Enzyme IIIfru and FPr activities, while Enzyme IIfru remained with the membranes. FPr was purified to homogeneity using ion exchange chromatography, gel filtration, and reversed phase high pressure liquid chromatography (HPLC), and its amino acid composition and N-terminal sequence were determined. A complex of FPr and Enzyme IIIfru (Mr 50,000) was also purified to near homogeneity using ion exchange chromatography, gel filtration, and chromatography on hydroxylapatite. When the purified complex was analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, it was visualized as two protein bands with mobilities corresponding to molecular weights of about 40,000 (Enzyme IIIfru) and 9,000 (FPr). Neither the FPr and Enzyme IIIfru activities nor the proteins represented by these two bands separated during the above chromatography steps or using any of several other techniques, including reversed phase HPLC, indicating a very tight association. Active Enzyme IIIfru free of FPr was never isolated or observed. The proteins could be separated in denatured form by gel filtration in the presence of guanidine HCl or urea. Free FPr and the FPr-Enzyme IIIfru complex were characterized, and the properties of free and complexed FPr were compared to those of HPr.     

The roles of HPr and FPr in the utilization of fructose by Escherichia coli

A mutant impaired in FPr activity was isolated. The altered gene (fpr), which was located near min. 44 on the E. coli genome, was transferred by phage-mediated transduction to appropriate recipients that lack HPr (ptsH), or Enzyme IIman (ptsM), or neither. The rates of growth on fructose of such transductants indicate that phosphate from PEP is transferred predominantly via FPr to fructose that enters the cells by Enzyme IIfru, but that HPr can play a role in transferring phosphate to fructose taken up via Enzyme IIman.     

Primary structure of the basic nuclear protein from spermatozoa of the mollusk Illex argentinus and its comparison with the structure of sperm proteins in other animals

The spermatic protein of chromatin I2 of squid Illex argentinus was separated by HPLC into two components I2-1 and I2-2. Amino acid sequences of the major portion of protein I2-1 (52 residues) and the N-terminal sequence of protein I2-2 (21 residues) were determined. Arginines in protein I2-1 are arranged in clusters typical of protamines; the first cluster is in the N-terminus, the longest heterogeneous basic cluster is in the central part of the protein chain, the C-terminal part of the molecule contains two clusters of three hydroxyamino acids each. The N-terminal sequences of illexins I2-1 and I2-2 (1-14 residues) are highly homologous. Homologous regions were found in illexin I2-1, tunnin of tuna fish and avian gallin thus defining the notion of proteins of an intermediate type from mollusc spermatozoa chromatin exemplified by the squid protamine-like protein.     

The nucleotide sequence and comparative analysis of the H-2Dp class I H-2 gene

We present the complete nucleotide sequence and the deduced amino acid sequence of the H-2Dp class I gene. This gene, which was cloned from a B10.P genomic DNA library, encodes and intact, functional H-2Dp molecule. Comparative analysis of the Dp sequence with other class I sequences reveals both similarities and differences. This analysis also shows that these genes exhibit D region-specific, locus-specific, as well as allele-specific sequences. The H-2Dp nucleotide sequence is greater than 90% homologous to the H-2Ld and H-2Db genes and only approximately 85% homologous to the H-2Dd gene. The K region and Qa region genes are less homologous. The 3' noncoding sequences appear to be region-specific. All of the previously described D region genes, Db, Ld, and Dd, possess the B2-SINE Alu-like repetitive sequence, as does Dp. Thus, this B2 repeat is a region-specific marker present in all D region genes studied so far. The additional polyadenylation site found in the H-2Dp gene starting at nucleotide 4671, which is homologous to non-D region sequences, as well as unique protein Dp coding sequences, make this gene an interesting model for studying the evolution of polymorphism and structure/function relationships in the class I gene family.     

Analysis in Neisseria meningitidis and other Neisseria species of genes homologous to the FKBP immunophilin family

The immunophilin family of FK506-binding proteins (FKBPs), involved in eukaryotic protein folding and cell regulation, have recently been found to have prokaryotic homologues. Genes with sequences homologous to those encoding human FKBPs were examined in Neisseria species. An FKBP DNA sequence was present, as shown by the polymerase chain reaction and Southern blotting experiments, in the chromosome of Neisseria meningitidis (14 strains) and in all 11 different commensal Neisseria spp. studied, but was not found in Neisseria gonorrhoeae (11 strains tested) or in Moraxella catarrhalis. The nucleotide and predicted protein sequences of the FKBP-encoding domain from five of the meningococcal strains were highly conserved (e.g. ≥97% homologous). The meningococcal nucleotide sequence was ≥93% homologous and the consensus meningococcal protein sequence was ≥97% homologous to FKBP sequences found in seven different commensal Neisseria spp. The meningococcal nucleotide and predicted protein sequences were ≥59% homologous to the conserved C-terminus of the human FKBP gene family. The FKBP nucleotide sequence was present as a single copy in the chromosome of commensal Neisseria spp. and in most strains of N. meningitidis. The FKBP gene was linked to the silent pilin locus, pilS, in class II-piliated meningococcal strains. In meningococcal strains expressing class I pili, the FKBP gene was linked to one of several pilS loci but not the pilE locus present in these strains. FKBP genes found in commensal Neisseria spp. were not linked to known pilin loci.     

Phosphocarrier proteins in an intracellular symbiotic bacterium of aphids.

A GroEL homolog produced by Buchnera, an intracellular symbiotic bacterium of aphids, is not only a molecular chaperone but also a novel phosphocarrier protein, suggesting that this protein plays a role in a signal transducing system specific to bacteria living in an intracellular environment. This prompted us to look into phosphocarrier proteins of Buchnera that may be shared in common with other bacteria. As a result, no evidence was obtained for the presence of sensor kinases of the two-component system in Buchnera, which are found in many bacteria. It is possible that the lack of sensor kinases is compensated for by the mulitifunctional GroEL homolog in this symbiotic bacteria. In contrast, we successfully identified three phosphotransferase system genes, ptsH, ptsI, and crr in Buchnera, and provide evidence for their active expression. While the deduced amino acid sequences of these gene products, histidine-containing phosphocarrier protein, Enzyme I, and Enzyme III were similar to their counterparts in Escherichia coli, the predicted isoelectric points of the Buchnera proteins were strikingly higher. It was also suggested that Buchnera Enzyme I, when produced in E. coli, is able to accept the phosphoryl group from phosphoenolpyruvate, but not from ATP.     

Comparison of the nucleotide sequences at the 3′-terminal region of RNAs from RNA coliphages

In order to study the genealogical relationships among four groups (I to IV) of RNA coliphages, we sequenced 200 to 260 nucleotides from the 3′ termini of 14 phage RNAs according to the method of Sanger et al. (1977), and compared the results. It was found that the sequences of phage RNAs in the same group were extremely homologous (about 90%). On the other hand, when the sequences were compared with those from other groups, they were seen to be only about 50 to 60% homologous between group I and group II, and about 50% homologous between group III and group IV. In other combinations, such as groups I (or II) and III, and groups I (or II) and IV, however, the extent of homology was small. Furthermore, the sequences up to 30 residues from the 3′ end were found to be about 90% homologous between groups I and II, and between groups III and IV.These results confirm our previous findings, that the sequences located in the proximity of the 3′ end of phage RNA in the same group were well-conserved (Inokuchi et al., 1979), and that close relationships exist between groups I and II, and between groups III and IV (Furuse et al., 1979).     

Isolation and partial characterization of the multiple forms of deoxyribonucleic acid-dependent ribonucleic acid polymerase in the fungus Podospora anserina

Three DNA-dependent RNA polymerases have been isolated and partially purified from the mycelium of the fungus Podospora anserina. Separated by DEAE-Sephadex chromatography, they have been designated RNA polymerases I, II, and III according to their order of elution. Their catalytic properties and alpha-amanitin sensitivity are in agreement with those of the homologous enzymes found in other eukaryotic organisms. The three enzymes exhibit rather sharp monophasic ammonium sulfate dependence with optima which are, respectively, 0.035 M, 0.050 M, and 0.075 M. Enzyme I has the largest Mn2+/Mg2+ activity ratio, shows a marked preference for native DNA, and is insensitive to alpha-amanitin. Enzyme III uses poly(dA-dT) in preference to native DNA as template and is only partially sensitive to alpha-amanitin. Enzyme II is sensitive to alpha-amanitin, but high concentrations of the toxin are required for inhibition compared to other eukaryotic class II enzymes. Three similar RNA polymerases with comparable levels of activity were found in the temperature-dependent VR strain when cellular incompatibility, leading to a rapid cessation of RNA synthesis, was induced.     

Functional interactions between proteins of the phosphoenolpyruvate:sugar phosphotransferase systems of Bacillus subtilis and Escherichia coli.

Proteins of the phosphoenolpyruvate:sugar phosphotransferase system (PTS) of Bacillus subtilis were overexpressed, purified to near homogeneity, and characterized. The proteins isolated include Enzyme I, HPr, the glucose-specific IIA domain of the glucose-specific Enzyme II (IIAglc), and the mannitol-specific IIA protein, IIAmtl. Site specific mutant proteins of IIAglc and HPr were also overexpressed and purified, and their properties were compared with those of the wild type proteins. These proteins and their phosphorylated derivatives were characterized with respect to their immunological cross-reactivities employing the Western blot technique and in terms of their migratory behavior during sodium dodecyl sulfate-gel electrophoresis, nondenaturing gel electrophoresis, and isoelectric focusing. The interactions between homologous and heterologous Enzymes I and HPrs, between homologous and heterologous HPrs and the IIAglc proteins, and between homologous and heterologous IIAglc proteins and IIBCscr of B. subtilis as well as IICBglc of Escherichia coli were defined and compared kinetically. The mutant HPrs and IIAglc proteins were also characterized kinetically as PTS phosphocarrier proteins and/or as inhibitors of the phosphotransferase reactions of the PTS. These studies revealed that complexation of IIAglc with the mutant form of HPr in which serine 46 was replaced by aspartate (S46D) did not increase the rate of phosphoryl transfer from phospho Enzyme I to S46D HPr more than when IIAmtl was complexed to S46D HPr. These findings do not support a role for HPr(Ser-P) in the preferential utilization of one PTS carbohydrate relative to another. Functional analyses in E. coli established that IIAglc of B. subtilis can replace IIAglc of E. coli with respect both to sugar transport and to regulation of non-PTS permeases, catabolic enzymes, and adenylate cyclase. Site-specific mutations in histidyl residues 68 and 83 (H68A and H83A) inactivated IIAglc of B. subtilis with respect to phosphoryl transfer and its various regulatory roles.     

Characterization of alcohol dehydrogenase in young soybean seedlings

Molecular properties of alcohol dehydrogenase (ADH) were examined in young soybean seedlings. Soybean radicle tissue is ADH-rich. Enzyme specific activity decreases slowly with the development of roots and becomes almost undetectable when the first true leaves appear. Soybean ADH was not found to be inducible by flooding. 2,4-Dichlorophenoxyacetic acid (2,4-D) treatment increased ADH specific activity as much as 14-fold. Only one ADH isozyme was detected by isoelectric focusing. By DNA-DNA hydridization, soybean ADH genomic sequences were shown to be partly homologous to maize ADH1 cDNA. The presence of more than one Adh gene in soybean is discussed.     

The enzymatic nature of an anonymous protein sequence cannot reliably be inferred from superfamily level structural information alone

As the largest fraction of any proteome does not carry out enzymatic functions, and in order to leverage 3D structural data for the annotation of increasingly higher volumes of sequence data, we wanted to assess the strength of the link between coarse grained structural data (i.e., homologous superfamily level) and the enzymatic versus non-enzymatic nature of protein sequences. To probe this relationship, we took advantage of 41 phylogenetically diverse (encompassing 11 distinct phyla) genomes recently sequenced within the GEBA initiative, for which we integrated structural information, as defined by CATH, with enzyme level information, as defined by Enzyme Commission (EC) numbers. This analysis revealed that only a very small fraction (about 1%) of domain sequences occurring in the analyzed genomes was found to be associated with homologous superfamilies strongly indicative of enzymatic function. Resorting to less stringent criteria to define enzyme versus non-enzyme biased structural classes or excluding highly prevalent folds from the analysis had only modest effect on this proportion. Thus, the low genomic coverage by structurally anchored protein domains strongly associated to catalytic activities indicates that, on its own, the power of coarse grained structural information to infer the general property of being an enzyme is rather limited.     

Cellular protein interactions with herpes simplex virus type 1 oriS.

The herpes simplex virus type 1 (HSV-1) origin of DNA replication, oriS, contains an AT-rich region and three highly homologous sequences, sites I, II, and III, identified as binding sites for the HSV-1 origin-binding protein (OBP). In the present study, interactions between specific oriS DNA sequences and proteins in uninfected cell extracts were characterized. The formation of one predominant protein-DNA complex, M, was demonstrated in gel shift assays following incubation of uninfected cell extracts with site I DNA. The cellular protein(s) that comprises complex M has been designated origin factor I (OF-I). The OF-I binding site was shown to partially overlap the OBP binding site within site I. Complexes with mobilities indistinguishable from that of complex M also formed with site II and III DNAs in gel shift assays. oriS-containing plasmid DNA mutated in the OF-I binding site exhibited reduced replication efficiency in transient assays, demonstrating a role for this site in oriS function. The OF-I binding site is highly homologous to binding sites for the cellular CCAAT DNA-binding proteins. The binding site for the CCAAT protein CP2 was found to compete for OF-I binding to site I DNA. These studies support a model involving the participation of cellular proteins in the initiation of HSV-1 DNA synthesis at oriS.     

Nucleotide sequence of Saccharomyces cerevisiae genes TRP2 and TRP3 encoding bifunctional anthranilate synthase: indole-3-glycerol phosphate synthase

Saccharomyces cerevisiae anthranilate synthase:indole-3-glycerol phosphate synthase is a multifunctional hetero-oligomeric enzyme encoded by genes TRP2 and TRP3. TRP2, encoding anthranilate synthase Component I, was cloned by complementation of a yeast trp2 mutant. The nucleotide sequence of TRP2 as well as that of TRP3 were determined. The deduced anthranilate synthase Component I primary structure from yeast exhibits only limited similarity to that of the corresponding Escherichia coli subunit encoded by trpE. On the other hand, yeast anthranilate synthase Component II and indole-3-glycerol phosphate synthase amino acid sequences from TRP3 are clearly homologous with the corresponding sequences of the E. coli trpG and trpC polypeptide segments and thereby establish the bifunctional structure of TRP3 protein. Based on comparisons of TRP3 amino acid sequence with homologous sequences from E. coli and Neurospora crassa, an 11-amino acid residue connecting segment was identified which fuses the trpG and trpC functions of the bifunctional TRP3 protein chain. These comparisons support the conclusion that the amino acid sequence of connectors in homologous multifunctional enzymes need not be conserved. Connector function is thus not dependent on a specific sequence. Nuclease S1 mapping was used to identify mRNA 5' termini. Heterogeneous 5' termini were found for both TRP2 and TRP3 mRNA. TRP2 and TRP3 5'-flanking regions were analyzed for sequences that might function in regulation of these genes by the S. cerevisiae general amino acid control system. The 9 base pair direct repeat (Hinnebusch, A.G., and Fink, G.R. (1983) J. Biol. Chem. 258, 5238-5247) and inverted repeats were identified in the 5'-flanking sequences of TRP2 and TRP3.     

On the statistical significance of homologous structures among the Escherichia coli ribosomal proteins

Summary Completion of the sequence determination of all 52 Escherichia coli ribosomal proteins enabled a final comparison of their sequences. Similarities in amino acid compositions were compared to the relatedness of the sequences, which was analyzed statistically with the aid of the computer programs RELATE and ALIGN.Among the examined 52×52 possible protein pairs at least 40 pairs were found that can be regarded as distantly related (showing segment comparison score values slightly above 3.0 S.D. units). These protein pairs were further examined with the programs ALIGN and SEEK to locate homologous sequence stretches. In no case were two complete homologous sequences found (with the exception of the known identical pairs L7/L12 and S20/L26). However, short homologous sequence regions were observed. Beside those protein pairs that show significant although distant relatedness, other pairs were slightly below the threshold value of 3.0 S.D. units.Those pairs observed to be distantly related consisted either of two proteins from the same subunit or of one protein from each of the different subunits. A further analysis of these pairs revealed a correlation between their relatedness and their time of incorporation into the ribosome during assembly.     

Characterization of the baiH gene encoding a bile acid-inducible NADH:flavin oxidoreductase from Eubacterium sp. strain VPI 12708.

A cholate-inducible, NADH-dependent flavin oxidoreductase from the intestinal bacterium Eubacterium sp. strain VPI 12708 was purified 372-fold to apparent electrophoretic homogeneity. The subunit and native molecular weights were estimated to be 72,000 and 210,000, respectively, suggesting a homotrimeric organization. Three peaks of NADH:flavin oxidoreductase activity (forms I, II, and III) eluted from a DEAE-high-performance liquid chromatography column. Absorption spectra revealed that purified form III, but not form I, contained bound flavin, which dissociated during purification to generate form I. Enzyme activity was inhibited by sulfhydryl-reactive compounds, acriflavine, o-phenanthroline, and EDTA. Activity assays and Western blot (immunoblot) analysis confirmed that expression of the enzyme was cholate inducible. The first 25 N-terminal amino acid residues of purified NADH:flavin oxidoreductase were determined, and a corresponding oligonucleotide probe was synthesized for use in cloning of the associated gene, baiH. Restriction mapping, sequence data, and RNA blot analysis suggested that the baiH gene was located on a previously described, cholate-inducible operon > or = 10 kb long. The baiH gene encoded a 72,006-Da polypeptide containing 661 amino acids. The deduced amino acid sequence of the baiH gene was homologous to that of NADH oxidase from Thermoanaerobium brockii, trimethylamine dehydrogenase from methylotrophic bacterium W3A1, Old Yellow Enzyme from Saccharomyces carlsbergensis, and the product of the baiC gene of Eubacterium sp. strain VPI 12708, located upstream from the baiH gene in the cholate-inducible operon. Alignment of these five sequences revealed potential ligands for an iron-sulfur cluster, a putative flavin adenine dinucleotide-binding domain, and two other well-conserved domains of unknown function.