Cloning vectors, mutagenesis, and gene disruption (ermR) for the erythromycin-producing bacterium Aeromicrobium erythreum.

Genetic systems for study of Aeromicrobium erythreum, a gram-positive, G + C-rich (72%) bacterium with the capacity for erythromycin biosynthesis, are described. High-copy-number plasmids suitable as gene cloning vectors include derivatives of the Streptomyces plasmids pIJ101, pVE1, and pJV1. pIJ101 derivatives with missense substitutions at the rep gene BamHI site do not replicate in A. erythreum. Ethyl methanesulfonate treatment generated several amino acid auxotrophs and non-erythromycin-producing (Ery-) strains. Using the Ery- strain AR1807 as a recipient for plasmid-directed integrative recombination, the chromosomal ermR gene (encoding 23S rRNA methyltransferase) was disrupted. Phenotypic characterizations demonstrated that ermR is the sole determinant of macrolide antibiotic resistance in A. erythreum.     

Description of the erythromycin-producing bacterium Arthrobacter sp. strain NRRL B-3381 as Aeromicrobium erythreum gen. nov., sp. nov

Arthrobacter sp. strain NRRL B-3381T (T = type strain) is a nonmycelial, nonsporulating actinomycete that produces the macrolide antibiotic erythromycin. This bacterium differs in many ways from the type species of the genus Arthrobacter (Arthrobacter globiformis), suggesting that a taxonomic revision is appropriate. The G + C content of strain NRRL B-3381T DNA is 71 to 73 mol%, and the peptidoglycan of this organism contains LL-diaminopimelic acid. Evolutionary distance data obtained from 16S rRNA sequences identified NRRL B-3381T as the deepest branching member of the Nocardioides group of actinomycetes. The principal long-chain fatty acids which we identified that distinguished strain NRRL B-3381T from related G + C-rich bacteria were 10-methyloctadecanoic (tuberculosteric), octadecenoic, and hexadecanoic acids. These characteristics, together with phage typing and biochemical characteristics, form the basis for our recommendation that strain NRRL B-3381 should be the type strain of a new taxon, for which we propose the name Aeromicrobium erythreum.     

Cloning, mutagenesis, and nucleotide sequence of a siderophore biosynthetic gene (amoA) from Aeromonas hydrophila.

Many isolates of the Aeromonas species produce amonabactin, a phenolate siderophore containing 2,3-dihydroxybenzoic acid (2,3-DHB). An amonabactin biosynthetic gene (amoA) was identified (in a Sau3A1 gene library of Aeromonas hydrophila 495A2 chromosomal DNA) by its complementation of the requirement of Escherichia coli SAB11 for exogenous 2,3-DHB to support siderophore (enterobactin) synthesis. The gene amoA was subcloned as a SalI-HindIII 3.4-kb DNA fragment into pSUP202, and the complete nucleotide sequence of amoA was determined. A putative iron-regulatory sequence resembling the Fur repressor protein-binding site overlapped a possible promoter region. A translational reading frame, beginning with valine and encoding 396 amino acids, was open for 1,188 bp. The C-terminal portion of the deduced amino acid sequence showed 58% identity and 79% similarity with the E. coli EntC protein (isochorismate synthetase), the first enzyme in the E. coli 2,3-DHB biosynthetic pathway, suggesting that amoA probably encodes a step in 2,3-DHB biosynthesis and is the A. hydrophila equivalent of the E. coli entC gene. An isogenic amonabactin-negative mutant, A. hydrophila SB22, was isolated after marker exchange mutagenesis with Tn5-inactivated amoA (amoA::Tn5). The mutant excreted neither 2,3-DHB nor amonabactin, was more sensitive than the wild-type to growth inhibition by iron restriction, and used amonabactin to overcome iron starvation.     

Cloning and mutagenesis of the Rhizobium meliloti isocitrate dehydrogenase gene.

The gene encoding Rhizobium meliloti isocitrate dehydrogenase (ICD) was cloned by complementation of an Escherichia coli icd mutant with an R. meliloti genomic library constructed in pUC18. The complementing DNA was located on a 4.4-kb BamHI fragment. It encoded an ICD that had the same mobility as R. meliloti ICD in nondenaturing polyacrylamide gels. In Western immunoblot analysis, antibodies raised against this protein reacted with R. meliloti ICD but not with E. coli ICD. The complementing DNA fragment was mutated with transposon Tn5 and then exchanged for the wild-type allele by recombination by a novel method that employed the Bacillus subtilis levansucrase gene. No ICD activity was found in the two R. meliloti icd::Tn5 mutants isolated, and the mutants were also found to be glutamate auxotrophs. The mutants formed nodules, but they were completely ineffective. Faster-growing pseudorevertants were isolated from cultures of both R. meliloti icd::Tn5 mutants. In addition to lacking all ICD activity, the pseudorevertants also lacked citrate synthase activity. Nodule formation by these mutants was severely affected, and inoculated plants had only callus structures or small spherical structures.     

Selectable cassettes for simplified construction of yeast gene disruption vectors

Cassettes based on a hisG-URA3-hisG insert have been modified by the addition of a Km^R-encoding gene and flanking polylinker sites, greatly simplifying construction of gene disruption vectors in Escherichia coli. After gene disruption in yeast, URA3 can then be excised by recombination between the hisG repeats flanking the gene, permitting reuse of the URA3 marker     

Cloning and expression of the isocitrate lyase gene from a nitrogen-fixing bacterium, Azotobacter vinelandii, and functional analysis of the enzyme by site-directed mutagenesis

The gene encoding isocitrate lyase (ICL) from a nitrogen-fixing mesophilic bacterium, Azotobacter vinelandii strain IAM1078, was cloned, and the gene expression was examined. When sodium acetate or glucose was used as carbon source, similar growth was observed in this bacterium, but the ICL activity of cells grown with the former source was 43-hold higher than those with the latter. In addition, northern blot analysis revealed that expression of the ICL gene was induced by acetate. Based on a comparison of the amino acid sequences of the ICLs of various organisms, the ICL of this bacterium was found to be classifiable into subfamily 3, one of two phylogenetic groups of eubacteial ICLs. Replacement of the Ile504 in the ICL by Met, which is conserved in the corresponding position of cold-adapted ICLs of psychrophlic bacteria, resulted in decreased thermostability of activity, indicating that this amino acid residue is involved in thermal properties of this enzyme.     

Knockout of the Erythromycin Biosynthetic Cluster Gene, eryBI, Blocks Isoflavone Glucoside Bioconversion during Erythromycin Fermentations in Aeromicrobium erythreum but Not in Saccharopolyspora erythraea

Isoflavone glucosides are valuable nutraceutical compounds and are present in commercial fermentations, such as the erythromycin fermentation, as constituents of the soy flour in the growth medium. The purpose of this study was to develop a method for recovery of the isoflavone glucosides as value-added coproducts at the end of either Saccharopolyspora erythraea or Aeromicrobium erythreum fermentation. Because the first step in isoflavone metabolism was known to be the conversion of isoflavone glucosides to aglycones by a β-glucosidase, we chose to knock out the only β-glucosidase gene known at the start of the study, eryBI, to see what effect this had on metabolism of isoflavone glucosides in each organism. In the unicellular erythromycin producer A. erythreum, knockout of eryBI was sufficient to block the conversion of isoflavone glucosides to aglycones. In S. erythraea, knockout of eryBI had no effect on this reaction, suggesting that other β-glucosidases are present. Erythromycin production was not significantly affected in either strain as a result of the eryBI knockout. This study showed that isoflavone metabolism could be blocked in A. erythreum by eryBI knockout but that eryBI knockout was not sufficient to block isoflavone metabolism in S. erythraea.     

Cloning and sequencing of Schizosaccharomyces pombe DNA topoisomerase I gene, and effect of gene disruption.

We cloned the structural gene topl+ for Schizosaccharomyces pombe DNA topoisomerase I (topo I) by hybridization. An eight-fold increase of topo I relaxing activity was obtained in S. pombe cells transformed with multicopy plasmid with topl+ insert. Nucleotide sequence determination showed a hypothetical coding frame interrupted by two short introns, encoding a 812 residue polypeptide (M.W. 94,000), 43 residues longer than and 47% homologous to Saccharomyces cerevisiae topo I. We show that the topl (null) strain made by gene disruption is viable, although its generation time is 20% longer than that of wild type. The topl locus is mapped in the long arm of chromosome II, using the Leu+ marker integrated with the cloned topl+ sequence. We constructed a double mutant topl (null) top2 (ts) and found its defective phenotype similar to that of previously obtained topl (heat sensitive) top2 (ts). The other double mutant topl (null) top2 (cs), however, was lethal. Our results suggest that topl+ gene of S. pombe is dispensable only if topo II activity is abundant.     

Cloning and expression of the catalase gene from the anaerobic bacterium Desulfovibrio vulgaris (Miyazaki F)

We identified a gene encoding a catalase from the anaerobic bacteria Desulfovibrio vulgaris (Miyazaki F), and the expression of its gene in Escherichia coli. The 3.3-kbp DNA fragment isolated from D. vulgaris (Miyazaki F) by double digestion with EcoRI and SalI was found to produce a protein that binds protoheme IX as a prosthetic group in E. coli. This DNA fragment contained a putative open reading frame (Kat) and one part of another open reading frame (ORF-1). The amino acid sequence of the amino terminus of the protein purified from the transformed cells was consistent with that deduced from the nucleotide sequence of Kat in the cloned fragment of D. vulgaris (Miyazaki F) DNA, which may include promoter and regulatory sequences. The nucleotide sequence of Kat indicates that the protein is composed of 479 amino acids per monomer. The recombinant catalase was found to be active in the decomposition of hydrogen peroxide, as are other catalases from aerobic organisms, but its K(m) value was much greater. The hydrogen peroxide stress against D. vulgaris (Miyazaki F) induced the activity for the decomposition of hydrogen peroxide somewhat, so the catalase gene may not work effectively in vivo.     

Cloning, nucleotide sequence, and disruption of Streptococcus mutans glutathione reductase gene (gor).

We cloned and sequenced the glutathione reductase gene (gor) of an oxygen-tolerant Streptococcus mutans, and constructed a gor-disruption mutant by homologous recombination. The gor gene consisted of 1,350 bp, coding for a protein of 450 amino acid residues. The deduced amino acid sequence of the S. mutans gor gene product showed extensive similarity with those of glutathione reductases from prokaryotes and eukaryotes. Although the mutant could grow aerobically, it showed no growth in the presence of 2 mM diamide, a thiol-specific oxidant. In contrast, growth of the wild-type strain was not significantly inhibited by 2 mM diamide, and glutathione reductase activity was increased 2.2-fold under these conditions. In addition, the level of glutathione reductase activity in the wild-type strain was increased 3.6-fold upon exposure to air, and the elevated level of the enzyme was retained throughout the aerobic growth. Thus, glutathione reductase may be important in protection of S. mutans against oxidative stress.     

Cloning, nucleotide sequence, and mutagenesis of a gene (irpA) involved in iron-deficient growth of the cyanobacterium Synechococcus sp. strain PCC7942.

We describe the cloning and sequencing of a gene from the cyanobacterium Synechococcus sp. strain PCC7942, designated irpA (iron-regulated protein A), that encodes for a protein involved in iron acquisition or storage. Polyclonal antibodies raised against proteins which accumulate during iron-deficient growth were used as probes to isolate immunopositive clones from a lambda gt11 genomic expression library. The clone, designated lambda gtAN26, carried a 1.7-kilobase (kb) chromosomal DNA insert and was detected by cross-reactivity with antibody against a 36-kilodalton protein. It was possible to map a 20-kb portion of the chromosome with various DNA probes from lambda gt11 and lambda EMBL-3 clones, and Southern blot analysis revealed that the irpA gene was present in a single copy and localized within a 1.7-kb PstI fragment. DNA sequencing revealed an open reading frame of 1,068 nucleotides capable of encoding 356 amino acids which yields a protein with a molecular weight of 38,584. The hydropathy profile of the polypeptide indicated a putative N-terminal signal sequence of 44 amino acid residues. IrpA is a cytoplasmic membrane protein as determined by biochemistry and electron microscopy immunocytochemistry. The upstream region of the irpA gene contained a consensus sequence similar to the aerobactin operator in Escherichia coli. This fact, plus a mutant with a mutation in irpA that is unable to grow under iron-deficient conditions, led us to suggest that irpA is regulated by iron and that the gene product is involved in iron acquisition or storage.     

Shuttle cloning vectors for the marine bacterium Vibrio parahaemolyticus.

Two cosmid cloning vectors containing lambda cos sequences and a 42-base-pair multipurpose cloning sequence were constructed. pAD22 also contains a 1.4-kilobase TRP-ARS fragment from Saccharomyces cerevisiae. These cosmids transformed Escherichia coli and S. cerevisiae cells and could be mobilized into Vibrio parahaemolyticus strains with a conjugative plasmid, pRK2013. The cosmid pAD22 was genetically and structurally stable during passage through V. parahaemolyticus and E. coli strains.     

Cloning, sequencing, and disruption of a levanase gene of Bacillus polymyxa CF43.

The Bacillus polymyxa CF43 lelA gene, expressing both sucrose and fructan hydrolase activities, was isolated from a genomic library of B. polymyxa screened in Bacillus subtilis. The gene was detected as expressing sucrose hydrolase activity; B. subtilis transformants did not secrete the lelA gene product (LelA) into the extracellular medium. A 1.7-kb DNA fragment sufficient for lelA expression in Escherichia coli was sequenced. It contains a 548-codon open reading frame. The deduced amino acid sequence shows 54% identity with mature B. subtilis levanase and is similar to other fructanases and sucrases (beta-D-fructosyltransferases). Multiple-sequence alignment of 14 of these proteins revealed several previously unreported features. LelA appears to be a 512-amino-acid polypeptide containing no canonical signal peptide. The hydrolytic activities of LelA on sucrose, levan, and inulin were compared with those of B. subtilis levanase and sucrase, confirming that LelA is indeed a fructanase. The lelA gene in the chromosome of B. polymyxa was disrupted with a chloramphenicol resistance gene (cat) by "inter-gramic" conjugation: the lelA::cat insertion on a mobilizable plasmid was transferred from an E. coli transformant to B. polymyxa CF43, and B. polymyxa transconjugants containing the lelA::cat construct replacing the wild-type lelA gene in their chromosomes were selected directly. The growth of the mutant strain on levan, inulin, and sucrose was not affected.     

Cloning, sequencing, and disruption of the gene encoding sterol C-14 reductase in Saccharomyces cerevisiae.

A sterol C-14 reductase (erg24-1) mutant of Saccharomyces cerevisiae was selected in a fen1, fen2, suppressor background on the basis of nystatin resistance and ignosterol (ergosta-8,14-dienol) production. The erg24-1 allele segregated genetically as a single, recessive gene. The wild-type ERG24 gene was cloned by complementation onto a 12-kb fragment from a yeast genomic library, and subsequently subcloned onto a 2.4-kb fragment. This was sequenced and found to contain an open reading frame of 1,314 bp, predicting a polypeptide of 438 amino acids (M(r) 50,612). A 1,088-bp internal region of the ERG24 gene was excised, replaced with a LEU2 gene, and integrated into the chromosome of the parental strain, FP13D (fen1, fen2) by gene replacement. The ERG24 null mutant produced ergosta-8,14-dienol as the major sterol, indicating that the delta 8-7 isomerase, delta 5-desaturase and the delta 22-desaturase were inactive on sterols with the C14 = 15 double bond.     

Identification and disruption analysis of the recN gene in the extremely radioresistant bacterium Deinococcus radiodurans.

We isolated a radiosensitive mutant strain, KR4128, from a wild-type strain of Deinococcus radiodurans, which is known as a extremely radioresistant bacterium. The gene that restore the defect of the mutant in DNA repair was cloned, and it turned out to be the homolog of the recN gene of Escherichia coli. The recN gene encoded a protein of 58 kDa, and, in its N-terminal region, a potential ATP binding domain was conserved as expected for a prokaryotic RecN protein. An analysis of sequence of the mutant recN gene revealed a G:C to T:A transversion near the 3' end of the coding region. This alteration causes an ochre mutation, and results in the truncation of 47 amino acids from the C-terminal region of the RecN protein. The null mutant of recN gene was constructed by insertional mutagenesis, and it showed substantial sensitivities to various types of DNA damaging agents, indicating that a single defect in the recN gene can directly affect the DNA damage resistant phenotype in D. radiodurans. The recN locus of KR4128 was also disrupted and the disruptant indicated the sensitivity that was indistinguishable from its progenitor. The result indicate that the transversion in the recN gene of KR4128 cells causes a complete loss of function of the RecN protein and thus the C-terminal region of the RecN protein includes domain essential to its function.     

Cloning and disruption of a putative NaH-antiporter gene of Enterococcus hirae.

When growing in a sodium-rich environment, wild-type Enterococcus hirae extrudes sodium by two mechanisms, ATP-driven sodium extrusion, and NaH-antiport. Mutant 7683 is unable to grow on sodium-rich media. This is due to two mutations, one inactivating ATP-driven sodium transport and a second rendering NaH-antiport inoperative. 7683 was transformed by electroporation with a gene bank, derived from E. hirae, in an Escherichia coli-E. hirae shuttle vector. Transformants which had regained the ability to grow on sodium-rich media were selected for and the transforming plasmids analyzed. A gene able to restore NaH-antiport activity in 7683 was identified. This gene was named napA. It codes for an extremely hydrophobic protein of 383 amino acids. Hydropathy analysis of this protein indicates that it probably forms 12 transmembraneous helices. In a mutant, possessing only the NaH-antiporter, the napA gene was disrupted by homologous recombination. The resultant strain failed to grow in sodium-rich media, and vesicles isolated from these cells exhibited a defect in sodium proton antiport activity. We conclude that the napA gene codes for a NaH-antiporter. The NapA protein does not exhibit significant homology to any protein in the EMBL genetic data bank.     

Cloning, sequencing, and expression of a xylanase gene from the anaerobic ruminal bacterium Butyrivibrio fibrisolvens.

A gene coding for xylanase activity, xynA, from the anaerobic ruminal bacterium Butyrivibrio fibrisolvens 49 was cloned into Escherichia coli JM83 by using plasmid pUC19. The gene was located on a 2.3-kilobase (kb) DNA insert composed of two adjacent EcoRI fragments of 1.65 and 0.65 kb. Expression of xylanase activity required parts of both EcoRI segments. In E. coli, the cloned xylanase enzyme was not secreted and remained cell associated. The enzyme exhibited no arabinosidase, cellulase, alpha-glucosidase, or xylosidase activity. The isoelectric point of the cloned protein was approximately 9.8, and optimal xylanase activity was obtained at pH 5.4. The nucleotide sequence of the 1,535-base-pair EcoRV-EcoRI segment from the B. fibrisolvens chromosome that included the xynA gene was determined. An open reading frame was found that encoded a 411-amino-acid-residue polypeptide of 46,664 daltons. A putative ribosome-binding site, promoter, and leader sequence were identified. Comparison of the XynA protein sequence with that of the XynA protein from alkalophilic Bacillus sp. strain C-125 revealed considerable homology, with 37% identical residues or conservative changes. The presence of the cloned xylanase gene in other strains of Butyrivibrio was examined by Southern hybridization. The cloned xylanase gene hybridized strongly to chromosomal sequences in only two of five closely related strains.     

Heat labile ribonuclease HI from a psychrotrophic bacterium: gene cloning, characterization and site-directed mutagenesis.

The rnhA gene encoding RNase HI from a psychrotrophic bacterium, Shewanella sp. SIB1, was cloned, sequenced and overexpressed in an rnh mutant strain of Escherichia coli. SIB1 RNase HI is composed of 157 amino acid residues and shows 63% amino acid sequence identity to E.coli RNase HI. Upon induction, the recombinant protein accumulated in the cells in an insoluble form. This protein was solubilized and purified in the presence of 7 M urea and refolded by removing urea. Determination of the enzymatic activity using M13 DNA-RNA hybrid as a substrate revealed that the enzymatic properties of SIB1 RNase HI, such as divalent cation requirement, pH optimum and cleavage mode of a substrate, are similar to those of E.coli RNase HI. However, SIB1 RNase HI was much less stable than E.coli RNase HI and the temperature (T(1/2)) at which the enzyme loses half of its activity upon incubation for 10 min was approximately 25 degrees C for SIB1 RNase HI and approximately 60 degrees C for E.coli RNase HI. The optimum temperature for the SIB1 RNase HI activity was also shifted downward by 20 degrees C compared with that of E.coli RNase HI. Nevertheless, SIB1 RNase HI was less active than E.coli RNase HI even at low temperatures. The specific activity determined at 10 degrees C was 0.29 units/mg for SIB1 RNase HI and 1.3 units/mg for E.coli RNase HI. Site-directed mutagenesis studies suggest that the amino acid substitution in the middle of the alphaI-helix (Pro52 for SIB1 RNase HI and Ala52 for E.coli RNase HI) partly accounts for the difference in the stability and activity between SIB1 and E.coli RNases HI.