In Bradyrhizobium japonicum the common nodulation genes, nodABC, are linked to nifA and fixA

Summary Using cloned Rhizobium phaseoli nodulation (nod) genes as hybridization probes homologous restriction fragments were detected in the genome of the slow-growing soybean symbiont, Bradyrhizobium japonicum strain 110. These fragments were isolated from a cosmid library, and were shown to lie 10 kilobasepairs (kb) upstream from the nifA and fixA genes. Specific nod probes from Rhizobium leguminosarum were used to identify nodA-, nodB-, and nodC-like sequences clustered within a 4.5 kb PstI fragment. A mutant was constructed in which the kanamycin resistance gene from Tn5 was inserted into the nodA homologous B. japonicum region. This insertion was precisely located, by DNA sequencing, to near the middle of the nodA gene. B. japonicum mutants carrying this insertion were completely nodulation deficient (Nod^-).     

Molecular cloning of the Salmonella typhimurium lep gene in Escherichia coli

Summary A system is described which enabled the selection of a heterologous ep gene, encoding signal peptidase I, in Escherichia coli. It is based on complementation of an E. coli mutant, in which the synthesis of signal peptidase I can be regulated. With this system the lep gene of Salmonella typhimurium was cloned and the nucleotide sequence was determined. The S. typhimurium lep gene encodes a protein of 324 amino acids. Expression of the gene in the E. coli mutant resulted in suppression of growth inhibition and in the restoration of processing activity under conditions where synthesis of E. coli signal peptidase I was repressed. The cloned S. typhimurium signal peptidase I had an apparent molecular weight of 36000 daltons, which is in agreement with the calculated molecular weight of 35782 daltons. The system described for selection of the S. typhimurium lep gene may permit the cloning and expression of other heterologous signal peptidase I gen/es.     

A TnphoA insertion within the Bradyrhizobium japonicum sipS gene, homologous to prokaryotic signal peptidases, results in extensive changes in the expression of PBM-specific nodulins of infected soybean (Glycine max) cells

Bradyrhizobium japonicum mutant 132 was obtained by random TnphoA mutagenesis of strain 110spc4. A 6.5 kb BamHI kanamycin-resistance-coding DNA fragment of mutant 132 was used as a hybridization probe to clone the corresponding wild-type fragment. DNA sequence analysis of a 3213 bp BamHI—ClaI fragment revealed that three open reading frames (ORFs) were encoded in the same orientation. Based on sequence similarities to other proteins in the database, the second ORF was called sipS (signal peptidase). The TnphoA insertion in mutant 132 was found to be in frame near the 3′ end of sipS. The resulting SipS—PhoA hybrid polypeptide was shown to be expressed in free-living B. japonicum and in Escherichia coli cultures. An immunoblot analysis with a polyclonal antibody directed against the alkaline phosphatase revealed the appearance of a weak signal of a 70 kDa polypeptide both in B. Japonicum and in E. coli, in agreement with the calculated size of the hybrid polypeptide. A much stronger 52 kDa band was also detected. Mutant 132 was specifically disturbed in the interaction with soybean (Glycine max) when the bacteria were released from the infection threads. The bacteroids were not stably maintained within the symbiosome. Numerous vesicles were found in the plant cytosol, which finally resulted in the formation of large vacuoles within the infected nodule cells. Immunohistochemical analyses with antibodies directed against nodulins of the peribacteroid membrane indicated a lower expression of these proteins. The mutant phenotype was genetically complemented by 4.4 kb BamHI fragment including sipS. Subfragments thereof also complemented a temperature-sensitive E. coli lepB mutant, demonstrating that the B. japonicum fragment was functionally replacing Lep^ts in E. coli. Genetic studies suggested that the three genes are organized in one common operon which is expressed from a promoter upstream of the sequenced region. Inactivation of the gene downstream of sipS did not result in a detectable phenotype.     

Use of Tn KPK2 for sequencing a 10.6-kb PstI DNA fragment of Bradyrhizobium japonicum and for the construction of aspA and ndvA mutants

Transposon TnKPK2 was used to saturate a randomly cloned Bradyrhizobium japonicum PstI fragment and the insertions were used as starting points for the sequence determination. The first gene of the 10.6-kb DNA insert encodes a homologue to ndvA, the product of which is known to be involved in the formation of periplasmic cyclic glucans. Selected TnKPK2 insertions were introduced into the B. japonicum wild-type strain. The resulting mutants were subsequently tested for their symbiotic interactions with soybeans. As in Sinorhizobium meliloti, a B. japonicum ndvA mutant was affected in salt-stress tolerance and exhibited symbiotic defects in that it induced the formation of ineffective soybean nodules. The central nodule tissue was infected by bacteroids, but within the infected cells the mutant was not properly maintained. Another gene was found to be highly similar to bacterial aspartases and thus was named aspA. The putative function of the product of this gene was confirmed by genetic complementation of aspartase-less Escherichia coli strain TK237. The symbiotic phenotype of a B. japonicum aspATnKPK2 mutant consisted of enlarged symbiosomes that made the system ineffective. In general, TnKPK2 is a suitable means for fast sequencing. In combination with pJQ200SK, the resulting recombinant plasmids can be directly used to create genetically defined mutants.     

Sequence analysis and functional studies of a chromosomal region of alkaliphilic Bacillus firmus OF4 encoding an ABC-type transporter with similarity of sequence and Na+ exclusion capacity to the Bacillus subtilis NatAB transporter

A 14.1-kb DNA fragment was cloned from a lambda library containing inserts of DNA from alkaliphilic Bacillus firmus OF4 on the basis of its hybridization to a probe from a previously sequenced alkaliphile homolog of the natA gene from Bacillus subtilis. Sequence analysis of the entire fragment revealed that, as in B. subtilis, the natA gene was part of a putative gene locus encoding an ABC-type transporter. In the alkaliphile, the transporter involved three genes, designated natCAB, that are part of a larger operon of unknown function. This is in contrast to the two-gene natAB operon and to another homolog from B. subtilis, the yhaQP genes. Like natAB, however, the alkaliphile natCAB catalyzes Na⁺ extrusion as assessed in a mutant of Escherichia coli that is deficient in Na⁺ extrusion. The full 14.1-kb fragment of alkaliphile DNA sequenced in this study contained several probable operons as well as likely monocistronic units. Among the 17 predicted ORFs apart from natCAB were acsA, a homolog of a halobacterial gene encoding acetylCoA synthetase; sspA, a homolog of a small acid-soluble spore protein; and malK, an ATP-binding component that was unaccompanied by candidates for other mal transport genes but was able to complement a malK-deficient mutant of E. coli. No strong candidates for genes encoding a secondary Na⁺/H⁺ antiporter were found in the fragment, either from the sequence analysis or from analyses of complementation of E. coli mutants by subclones of the 14.1-kb piece. There were a total of 12 ORFs whose closest and significant homologs were genes from B. subtilis; of these, one-third were in apparently different contexts, as assessed by the sequence of the neighboring genes, than the B. subtilis homologs. Received: August 30, 1998 / Accepted: November 13, 1998     

Cloning the dapA dapB cluster of the lysine-secreting bacterium Corynebacterium glutamicum

Summary The genes encoding the two successive enzymes of the lysine biosynthetic pathway, dihydrodipicolinate synthase (dapA) and dihydrodipicolinate reductase (dapB), have been isolated from Corynebacterium glutamicum by heterologous complementation of Escherichia coli mutants. The two genes reside on a single 3.8-kb chromosomal fragment. They were subcloned as non overlapping fragments on an E. coli/C. glutamicum shuttle vector and introduced into C. glutamicum. This resulted in overexpression of both enzyme activities which was irrespective of the orientation of the inserts and comparable to that obtained with the large 3.8-kb fragment. Therefore, both genes are located in close proximity to each other on the C. glutamicum chromosome, but are apparently independently transcribed.     

Exopolysaccharide (EPS) synthesis in Bradyrhizobium japonicum : sequence, operon structure and mutational analysis of an exo gene cluster

The nucleotide sequence of a 8330-bp DNA fragment from Bradyrhizobium japonicum 110spc4 was determined. Sequence analysis revealed that six ORFs were present and the deduced amino acid sequences were homologous to enzymes involved in exopolysaccharide (EPS) biosynthesis. The genes appear to be organized into at least four different operons. One gene was found to be homologous to exoB, which encodes a UDP-galactose 4′-epimerase. Other ORFs were homologous to UDP-hexose transferases and one ORF showed similarity to Sinorhizobium (Rhizobium) meliloti ExoP, which has been suggested to be involved in EPS chain-length determination. A set of deletion and insertion mutants was constructed and the resulting B. japonicum strains were tested for their symbiotic traits. Deletion mutant ΔP22, which lacks the C-terminal part of ExoP, the UDP-hexose transferase ExoT and the N-terminal part of ExoB, shows a delayed nodulation phenotype and induces symptoms of plant defense reactions; its EPS does not contain galactose and no high molecular weight fraction is synthesized. In contrast, insertion mutant EH3, which expresses an exoP gene product that is truncated in its putative periplasmic domain, produced an EPS containing both HMW and LMW fractions. However, the interaction of EH3 with soybeans was severely perturbed. As a rule, only the initial steps of nodule formation were observed. Received: 2 January 1998 / Accepted: 24 March 1998     

Characterization of a region of the Enterococcus faecalis plasmid pAMβ1 which enhances the segregational stability of pAMβ1-derived cloning vectors in Bacillus subtilis

The nucleotide sequence of a 2.13-kb EcoRI-HindIII, pAMβ1-derived fragment, isolated from the gram-positive cloning vector pHV1431, has been determined and shown to encode two ORFs. ORF H encodes for a protein of 23,930 Da which exhibits substantial homology to bacterial site-specific recombinases, particularly the resolvases of the gram-positive transposons Tn917 (30.3% identity) and Tn552 (31.6% identity) and the clostridial plasmid pIP404 (27.1% identity). The second ORF (I) is incomplete and encodes a polypeptide which has significant homology with Escherichia coli topoisomerase I (26.0% identity). Insertion of either the entire 2.13-kb EcoRI-HindIII fragment or a 0.73-kb EcoRI-DraI subfragment encoding only the resolvase into the pAMβ1-based cloning vector pMTL500E causes a significant enhancement of segregational stability (from 6.5 × 10⁻² to 3.0–4.0 × 10⁻³ plasmid loss per cell per generation). Improved segregational stability is mirrored by a reduction in plasmid polymerization. The introduction of a stop codon into the resolvase coding region negates its ability to promote segregational stability. It is proposed that the identified determinant stabilizes pAMβ1-based vectors in Bacillus subtilis by maintaining the plasmid population in the monomeric state, thereby reducing the chances of producing plasmid-free segregants.     

The pilA gene of Xanthomonas campestris pv. citri is required for infection by the filamentous phage cf

 Host factors that are important for infection of Xanthomonas campestris pv. citri by the filamentous bacteriophage cf were investigated by transposon mutagenesis with Tn5tac1. A mutant, XT501, that was resistant to cf infection was recovered, showing that the gene inactivated by the transposon is required for infection by the phage but not for cf replication or assembly. A 1.7-kb SacI-ApaI DNA fragment from XT501 containing the bacterial DNA flanking one end of the transposon was cloned and shown to be required for cf infection. Nucleotide sequence analysis of the 1.7-kb fragment reveals the presence of an ORF that encodes a protein of 146 amino acids. This protein shows 42% identity to the type 4 prepilin encoded by the pilA genes of other bacteria. The pilA gene of X. campestris pv. citri is thus essential for infection by the bacteriophage cf. Received: 30 November 1998 / Accepted: 21 April 1999     

The Saccharomyces cerevisiae SGE1 gene product: a novel drug-resistance protein within the major facilitator superfamily

Several pleiotropic drug sensitivities have been described in yeast. Some involve the loss of putative drug efflux pumps analogous to mammalian P-glycoproteins, others are caused by defects in sterol synthesis resulting in higher plasma membrane permeability. We have constructed a Saccharomyces cerevisiae strain that exhibits a strong crystal violet-sensitive phenotype. By selecting cells of the supersensitive strain for normal sensitivity after transformation with a wild-type yeast genomic library, a complementing 10-kb DNA fragment was isolated, a 3.4-kb subfragment of which was sufficient for complementation. DNA sequence analysis revealed that the complementing fragment comprised the recently sequenced SGE1 gene, a partial multicopy suppressor of gal11 mutations. The supersensitive strain was found to be a sge1 null mutant. Overexpression of SGE1 on a high-copy-number plasmid increased the resistance of the supersensitive strain. Disruption of SGE1 in a wild-type strain increased the sensitivity of the strain. These features of the SGE1 phenotype, as well as sequence homologies of SGE1 at the amino acid level, confirm that the Sge1 protein is a member of the drug-resistance protein family within the major facilitator superfamily (MFS).     

Cloning and characterisation of the Neisseria gonorrhoeaearoB gene

The gene coding for the 3-dehydroquinate synthetase (aroB) of Neisseria gonorrhoeae has been cloned by functional complementation of an Escherichia coli aroB mutant. The aroB gene isolated from a gonococcal plasmid library encodes a 359 amino acid protein with a molecular mass of 38.6 kDa. Alignment of different prokaryotic and eukaryotic aroB gene products reveals an overall identity ranging from 33 to 55%. An open reading frame coding for an aroK homologue is located immediately upstream of aroB. Downstream of aroB a region of inverted repeats and a gene showing high homology to yafJ of E. coli has been identified. Disruption of aroB generates a gonococcal mutant that is unable to grow in the absence of aromatic compounds. Complementation of the mutant with the intact aroB gene intrans indicates that the gene is responsible for the auxotrophic phenotype. In infection assays with AroB-deficient gonococcal strains, binding, entry and short-term survival in epithelial cells is not affected. The aroB gene might be useful as a selectable marker and target for attenuation of a gonococcal live vaccine strain or as a biosafe laboratory strain. Received: 23 September 1997 / Accepted: 19 November 1997     

Five genetic loci involved in the synthesis of acidic exopolysaccharides are closely linked in the genome of Rhizobium sp strain NGR234

Summary R-prime plasmids were constructed from a derivative of Rhizobium strain NGR234 (ANU280) and were shown to contain overlapping genomic DNA segments involved in biosynthesis of exopolysaccharides (EPS). The R-primes originally constructed carried the mutant allele from Tn5-induced EPS-deficient (Exo^–) mutant ANU2811. This plasmid-located mutant allele was dominant to the corresponding wild-type allele as merodiploid strains were Exo^–. Exo⁺ revertants occurred at a low rate (1×10^-7) and these were shown to result from double reciprocal recombination events, which led to the isolation of R-prime plasmids carrying functional wild-type exo alleles. R-prime plasmids that carry overlapping segments of DNA from parental strain ANU280 complemented 28 of the 30 group 2 Exo^– mutants of strain ANU280. Complementation of these Exo^– mutants also restored their symbiotic abilities of effective nodulation. Subsequent in vivo recombination between the wild-type alleles located on the R-prime and the corresponding mutated allele on the genome, was used to generate a new family of R-primes, which carried mutations in the exo genes. The 30 group 2 Exo^– mutants were classified into 7 distinct genetic groups based upon complementation and physical mapping data. Five of the seven exo loci were gentically linked and located on a 15-kb region of DNA. Mutations at two loci were dominant only when the mutations were R-prime plasmid-located while a mutation at a second locus was cis-dominant to two other exo loci. At least five genes involved in the synthesis of acidic exopolysaccharide synthesis have been identified.     

A molecular linkage map of nitrogenase and nodulation genes in Rhizobium trifolii

Summary A molecular map was constructed linking the nitrogenase structural genes (nif) and nodulation genes (nod) in the white clover symbiont, Rhizobium trifolii. In R. trifolii strain ANU843 these two genetic regions are located some 16 kilobases (kb) apart on the 180 kb symbiotic (Sym) plasmid. The molecular linkage of nod and nif genetic regions was established by hybridization analysis using recombinant plasmids containing overlapping cloned sequences. Nodulation genes were located by means of a Tn5-induced nodulation-defective mutant that failed to induce clover root hair curling (Hac^- phenotype). A cloned wild-type DNA fragment was shown to phenotypically correct the Hac^- mutation by complementation. The nifHDK genes were cloned by positive hybridization to another R. trifolii nif-specific probe. Location of the nif genes relative to the nod genes was established by analysis of a Sym plasmid deletion derivative.     

Cloning of hemA from Rhizobium sp. NGR234 and symbiotic phenotype of a gene-directed mutant in diverse legume genera

Summary The hemA gene which encodes -aminolaevulinic acid synthase (ALAS), was cloned and characterized from the broad host-range Rhizobium strain NGR234. A cosmid, identified by hybridization with the cloned gene of R. meliloti and complementation of an R. meliloti hemA mutant, was subcloned to yield a 5.5 kb fragment containing the entire NGR234 gene. A physical-genetic map was made and the interposon was introduced into a single EcoRI site which bisects the gene. The mutated gene was homogenotized into NGR234 to generate a hemA mutant, with a view to evaluating the role of rhizobial bacteroid ALAS activity for a wide variety of legume symbioses. The mutant strain formed an ineffective (Fix^–) symbiosis with all tested host plants. These included tropical legumes that produce either indeterminate (Leucaena) or determinate (Desmodium, Macroptilium, Lablab, Vigna) root nodules.Abbreviations ALA -aminolaevulinic acid - ALAS aminolaevulinic acid synthase - Lb leghaemoglobin - Lb-haem haem moiety of leghaemoglobin     

Symbiotic mutants of USDA191, a fast-growing Rhizobium that nodulates soybeans

Summary Symbiotic and auxotrophic mutants of Rhizobium japonicum strain USDA191 were isolated using Tn5 mutagenesis and techniques that cause plasmid deletions and plasmid curing. Characterization of several mutants that are unable to nodulate (Nod^-) or unable to fix nitrogen (Fix^_) showed that nod and nif genes are located within one regions of a 200 MD plasmid (pSym191). Blot hybridization analysis of plasmids in other fast-growing R. japonicum strains showed that nod as well as nif sequences are located on plasmids in eight strains but are apparently carried in the chromosome in two strains.     

The gene for indole-3-acetyl-L-aspartic acid hydrolase from Enterobacter agglomerans : molecular cloning, nucleotide sequence, and expression in Escherichia coli

A 5.5-kb DNA fragment containing the indole-3-acetyl-aspartic acid (IAA-asp) hydrolase gene (iaaspH) was isolated from Enterobacter agglomerans strain GK12 using a hybridization probe based on the N-terminal amino acid sequence of the protein. The DNA sequence of a 2.4-kb region of this fragment was determined and revealed a 1311-nucleotide ORF large enough to encode the 45-kDa IAA-asp hydrolase. A 1.5-kb DNA fragment containing iaaspH was subcloned into the Escherichia coli expression plasmid pTTQ8 to yield plasmid pJCC2. Extracts of IPTG-induced E. coli cultures containing the pJCC2 recombinant plasmid showed IAA-asp hydrolase levels 5 to 10-fold higher than those in E. agglomerans extracts. Homology searches revealed that the IAA-asp hydrolase was similar to a variety of amidohydrolases. In addition, IAA-asp hydrolase showed 70% sequence identity to a putative thermostable carboxypeptidase of E. coli. Received: 12 March 1998 / Accepted: 30 March 1998     

A gene, SMP2, involved in plasmid maintenance and respiration in Saccharomyces cerevisiae encodes a highly charged protein

Summary The smp2 mutant of Saccharomyces cerevisiae shows increased stability of the heterologous plasmid pSR1 and YRp plasmids. A DNA fragment bearing the SMP2 gene was cloned by its ability to complement the slow growth of the smp2 smp3 double mutant (smp3 is another mutation conferring increased stability of plasmid pSR1). The nucleotide sequence of SMP2 indicated that it encodes a highly charged 95 kDa protein. Disruption of the genomic SMP2 gene resulted in a respiration-deficient phenotype, although the cells retained mitochondrial DNA, and showed increased stability of pSR1 like the original smp2 mutant. The fact that the smp2 mutant is not always respiration deficient and shows increased pSR1 stability even in a rho ⁰ strain lacking mitochondrial DNA suggested that the function of the Smp2 protein in plasmid maintenance is independent of respiration. The SMP2 locus was mapped at a site 71 cM from lys7 and 21 cM from ilv2/SMR1 on the right arm of chromosome XIII.     

Cloning and analysis of Agrobacterium tumefaciens C58 loci involved in glutamine biosynthesis: Neither the glnA (GSI) nor the glnII (GSII) gene plays a special role in virulence

Summary Using heterologous complementation of a glutamine synthetase deficient (glnA; GS^-) Escherichia coli mutant strain and heterologous DNA hybridization probes from Rhizobium meliloti and Bradyrhizobium japonicum, three distinct Agrobacterium tumefaciens loci involved in glutamine biosynthesis were identified. These loci correspond to the glnA (GSI), glnII (GSII) and a third previously unidentified locus, which is capable of complementing an E. coli glnA mutant, but may be cryptic in A. tumefaciens. The gene products encoded by the cloned glnA and glnII loci were identified using maxicells. Single insertion mutations in the glnA (GSI) and glnII (GSII) genes and a glnA glnII double mutant were constructed using gene replacement techniques. These mutant strains were examined for GSI and II activities, for growth on a variety of nitrogen (N) sources and for virulence properties on Kalanchoë plants. Neither glnA (GSI) nor glnII (GSII) were found to be essential for tumour induction on Kalanchoë nor for opine catabolism.     

Mutational analysis of the Bradyrhizobium japonicum common nod genes and further nod box-linked genomic DNA regions

Summary By insertional and deletional marker replacement mutagenesis the common nod region of Bradyrhizobium japonicum was examined for the presence of additional, essential nodulation genes. An open reading frame located in the 800 bp large intergenic region between nodD1 and nodA did not appear to be essential for nodulation of soybean. Furthermore, a strain with a deletion of the nodI- and nodJ-like genes downstream of nodC had a Nod⁺ phenotype. A mutant with a 1.7 kb deletion immediately downstream of nodD1 considerably delayed the onset of nodulation. This region carried a second copy of nodD (nodD2). A nodD1-nodD2 double mutant had a similar phenotype to the nodD2 mutant. Using a 22-mer oligonucleotide probe partially identical to the nod box sequence, a total of six hybridizing regions were identified in B. japonicum genomic DNA and isolated from a cosmid library. Sequencing of the hybridizing regions revealed that at least three of them represented true nod box sequences whereas the others showed considerable deviations from the consensus sequence. One of the three nod box sequences was the one known to be associated with nodA, whereas the other two were located 60 to 70 kb away from nif cluster I. A deletion of one of these two sequences plus adjacent DNA material mmutant 308) led to a reduced nodulation on Vigna radiata but not on soybean. Thus, this region is probably involved in the determination of host specificity.Dedicated to Prof. Giorgio Semenza on the occasion of his 60th birthday