Nucleotide sequence and genetic organization of the Bacillus subtilis comG operon

The complete general secretory pathway in gram-negative bacteria

The unifying feature of all proteins that are transported out of the cytoplasm of gram-negative bacteria by the general secretory pathway (GSP) is the presence of a long stretch of predominantly hydrophobic amino acids, the signal sequence. The interaction between signal sequence-bearing proteins and the cytoplasmic membrane may be a spontaneous event driven by the electrochemical energy potential across the cytoplasmic membrane, leading to membrane integration. The translocation of large, hydrophilic polypeptide segments to the periplasmic side of this membrane almost always requires at least six different proteins encoded by the sec genes and is dependent on both ATP hydrolysis and the electrochemical energy potential. Signal peptidases process precursors with a single, amino-terminal signal sequence, allowing them to be released into the periplasm, where they may remain or whence they may be inserted into the outer membrane. Selected proteins may also be transported across this membrane for assembly into cell surface appendages or for release into the extracellular medium. Many bacteria secrete a variety of structurally different proteins by a common pathway, referred to here as the main terminal branch of the GSP. This recently discovered branch pathway comprises at least 14 gene products. Other, simpler terminal branches of the GSP are also used by gram-negative bacteria to secrete a more limited range of extracellular proteins.

Identification and molecular characterization of the acetyl coenzyme A synthetase gene (acoE) of Alcaligenes eutrophus

The gene locus acoE, which is involved in the utilization of acetoin in Alcaligenes eutrophus, was identified as the structural gene of an acetyl coenzyme A synthetase (acetate:coenzyme A ligase [AMP forming]; EC 6.2.1.1). This gene was localized on a 3.8-kbp SmaI-EcoRI subfragment of an 8.1-kbp EcoRI restriction fragment (fragment E) that was cloned recently (C. Fründ, H. Priefert, A. Steinbüchel, and H. G. Schlegel, J. Bacteriol. 171:6539-6548, 1989). The 1,983 bp acoE gene encoded a protein with a relative molecular weight of 72,519, and it was preceded by a putative Shine-Dalgarno sequence. A comparison analysis of the amino acid sequence deduced from acoE revealed a high degree of homology to primary structures of acetyl coenzyme A synthetases from other sources (amounting to up to 50.5% identical amino acids). Tn5 insertions in two transposon-induced mutants of A. eutrophus, that were impaired in the catabolism of acetoin were mapped 481 and 1,159 bp downstream from the translational start codon of acoE. The expression of acoE in Escherichia coli led to the formation of an acyl coenzyme A synthetase that accepted acetate as the preferred substrate (100% relative activity) but also reacted with propionate (46%) and hydroxypropionate (87%); fatty acids consisting of four or more carbon atoms were not accepted. In addition, evidence for the presence of a second acyl coenzyme A synthetase was obtained; this enzyme exhibited a different substrate specificity. The latter enzyme is obviously required for the activation of propionate, e.g., during the formation of the storage compound poly(3-hydroxybutyric acid-co-3-hydroxyvaleric acid) when propionate is provided as the sole carbon source. An analysis of mutants provided evidence that the expression of the uptake protein for propionate depends on the presence of alternate sigma factor sigma 54.

Analysis of eight out genes in a cluster required for pectic enzyme secretion by Erwinia chrysanthemi: sequence comparison with secretion genes from other gram-negative bacteria

Many extracellular proteins produced by Erwinia chrysanthemi require the out gene products for transport across the outer membrane. In a previous report (S. Y. He, M. Lindeberg, A. K. Chatterjee, and A. Collmer, Proc. Natl. Acad. Sci. USA 88:1079-1083, 1991) cosmid pCPP2006, sufficient for secretion of Erwinia chrysanthemi extracellular proteins by Escherichia coli, was partially sequenced, revealing four out genes sharing high homology with pulH through pulK from Klebsiella oxytoca. The nucleotide sequence of eight additional out genes reveals homology with pulC through pulG, pulL, pulM, pulO, and other genes involved in secretion by various gram-negative bacteria. Although signal sequences and hydrophobic regions are generally conserved between Pul and Out proteins, four out genes contain unique inserts, a pulN homolog is not present, and outO appears to be transcribed separately from outC through outM. The sequenced region was subcloned, and an additional 7.6-kb region upstream was identified as being required for secretion in E. coli. out gene homologs were found on Erwinia carotovora cosmid clone pAKC651 but were not detected in E. coli. The outC-through-outM operon is weakly induced by polygalacturonic acid and strongly expressed in the early stationary phase. The out and pul genes are highly similar in sequence, hydropathic properties, and overall arrangement but differ in both transcriptional organization and the nature of their induction.

Some of the out genes involved in the secretion of pectate lyases in Erwinia chrysanthemi are regulated by kdgR

The out genes of Erwinia chrysanthemi are required for the translocation across the outer membrane of pectate lyases and cellulases. We present the characterization and the nucleotide sequence of five genes of the out cluster. The products of outS, B, C, D and E have significant homology with the PulS, B, C, D and E proteins necessary to the secretion of pullulanase in Klebsiella pneumoniae. An open reading frame, outT, located between outB and outC has no homology with the pul cluster but is involved in secretion. outC, outD and outE form an operon while outS, outB and outT constitute independent transcription units. outT and the outCDE operon are regulated by kdgR, the negative regulatory gene controlling pectinase production. outB and outS seem to be expressed constitutively.

Kinetics and sequence specificity of processing of prepilin by PilD, the type IV leader peptidase of Pseudomonas aeruginosa

PilD, originally isolated as an essential component for the biogenesis of the type IV pili of Pseudomonas aeruginosa, is a unique endopeptidase responsible for processing the precursors of the P. aeruginosa pilin subunits. It is also required for the cleavage of the leader peptides from the Pdd proteins, which are essential components of an extracellular secretion pathway specific for the export of a number of P. aeruginosa hydrolytic enzymes and toxins. Substrates for PilD are initially synthesized with short, i.e., 6- to 8-amino-acid-long, leader peptides with a net basic charge and share a high degree of amino acid homology through the first 16 to 30 residues at the amino terminus. In addition, they all have a phenylalanine residue at the +1 site relative to the cleavage site, which is N methylated prior to assembly into the oligomeric structures. In this study, the kinetics of leader peptide cleavage from the precursor of the P. aeruginosa pilin subunit by PilD was determined in vitro. The rates of cleavage were compared for purified enzyme and substrate as well as for enzyme and substrate contained within total membranes extracted from P. aeruginosa strains overexpressing the cloned pilD or pilA genes. Optimal conditions were obtained only when both PilD and substrate were contained within total membranes. PilD catalysis of P. aeruginosa prepilin followed normal Michaelis-Menten kinetics, with a measured apparent Km of approximately 650 microM, and a kcat of 180 min-1. The kinetics of PilD processing of another type IV pilin precursor, that from Neisseria gonorrhoeae with a 7-amino-acid-long leader peptide, were essentially the same as that measured for wild-type P. aeruginosa prepilin. Quite different results were obtained for a number of prepilin substrates containing substitutions at the conserved phenylalanine at the +1 position relative to the cleavage site, which were previously shown to be well tolerated in vivo. Substitutions of methionine, serine, and cysteine for phenylalanine show that Km values remain close to that measured for wild-type substrate, while kcat and kcat/Km values were significantly decreased. This indicates that while the affinity of enzyme for substrate is relatively unaffected by the substitutions, the maximum rate of catalysis favors a phenylalanine at this position. Interesting, PilD cleavage of one mutated pillin (asparagine) resulted in a lower Km value of 52.5 microM, which indicates a higher affinity for the enzyme, as well as a lower kcat value of 6.1 min m(-1). This suggests that it may be feasible to design peptide inhibitors of PilD.

The Aeromonas hydrophila exeE gene, required both for protein secretion and normal outer membrane biogenesis, is a member of a general secretion pathway

The Aeromonas hydrophila Tn5-751 insertion mutant L1.97 is unable to secrete extracellular proteins, and is fragile because of defective assembly of its outer membrane. A KpnI 4.1 kb fragment, which complements this mutant when supplied with an exogenous promoter, was isolated and sequenced. It contains two complete genes, exeE and exeF, plus fragments of two others and may form part of an operon. The exeE and exeF open reading frames encode 501-residue M(r) 55,882 and 388-residue M(r) 43,431 proteins, respectively. These genes were expressed in vitro and their initiation codons verified by deletion analysis. Tn5-751 had inserted near the centre of the exeE gene in the L1.97 strain. Subclones of the KpnI 4.1 kb fragment which contained only the exeE gene fully complemented the mutation, indicating that its function is required both for extracellular secretion and outer membrane assembly. ExeE and ExeF are highly similar to other proteins which have been shown to be involved in extracellular secretion, suggesting that an additional export apparatus beyond that required for inner membrane translocation may be part of the physiology of many Gram-negative bacteria.

Conjugative transfer functions of broad-host-range plasmid RK2 are coregulated with vegetative replication

The kilB locus (which is unclonable in the absence of korB) of broad-host-range plasmid RK2 (60 kb) lies between the trfA operon (co-ordinates 16.4 to 18.2 kb), which encodes a protein essential for vegetative replication, and the Tra2 block of conjugative transfer genes (co-ordinates 20.0 to 27.0 kb). Promoter probe studies indicated that kilB is transcribed clockwise from a region containing closely spaced divergent promoters, one of which is the trfA promoter. The repression of both promoters by korB suggested that kilB may also play a role in stable maintenance of RK2. We have sequenced the region containing kilB and analysed it by deletion and insertion mutagenesis. Loss of the KilB+ phenotype does not result in decreased stability of mini RK2 plasmids. However insertion in ORFI (kilBI) of the region analysed results in a Tra- phenotype in plasmids which are otherwise competent for transfer, demonstrating that this locus is essential for transfer and is probably the first gene of the Tra2 region. From the kilBI DNA sequence KilBI is predicted to be 34995 Da, in line with M(r) = 36,000 observed by sodium dodecyl sulphate/polyacrylamide gel electrophoresis, and contains a type I ATP-binding motif. The purified product was used to raise antibody which allowed the level of KilBI produced from RK2 to be estimated at approximately 2000 molecules per bacterium. Protein sequence comparisons showed the highest homology score with VirB11, which is essential for the transfer of the Agrobacterium tumefaciens Ti plasmid DNA from bacteria to plant cells. The sequence similarity of both KilBI and VirB11 to a family of protein export functions suggested that KilBI may be involved in assembly of the surface-associated Tra functions. The data presented in this paper provide the first demonstration of coregulation of genes required for vegetative replication and conjugative transfer on a bacterial plasmid.

Two-way chemical signaling in Agrobacterium-plant interactions

The discovery in 1977 that Agrobacterium species can transfer a discrete segment of oncogenic DNA (T-DNA) to the genome of host plant cells has stimulated an intense interest in the molecular biology underlying these plant-microbe associations. This attention in turn has resulted in a series of insights about the biology of these organisms that continue to accumulate at an ever-increasing rate. This excitement was due in part to the notion that this unprecedented interkingdom DNA transfer could be exploited to create transgenic plants containing foreign genes of scientific or commercial importance. In the course of these discoveries, Agrobacterium became one of the best available models for studying the molecular interactions between bacteria and higher organisms. One extensively studied aspect of this association concerns the exchange of chemical signals between Agrobacterium spp. and host plants. Agrobacterium spp. can recognize no fewer than five classes of low-molecular-weight compounds released from plants, and other classes probably await discovery. The most widely studied of these are phenolic compounds, which stimulate the transcription of the genes needed for infection. Other compounds include specific monosaccharides and acidic environments which potentiate vir gene induction, acidic polysaccharides which induce one or more chromosomal genes, and a family of compounds called opines which are released from tumorous plant cells to the bacteria as nutrient sources. Agrobacterium spp. in return release a variety of chemical compounds to plants. The best understood is the transferred DNA itself, which contains genes that in various ways upset the balance of phytohormones, ultimately causing neoplastic cell proliferation. In addition to transferring DNA, some Agrobacterium strains directly secrete phytohormones. Finally, at least some strains release a pectinase, which degrades a component of plant cell walls.

An enzyme with type IV prepilin peptidase activity is required to process components of the general extracellular protein secretion pathway of Klebsiella oxytoca

The last gene (pulO) of the pulC-O pullulanase secretion gene operon of Klebsiella oxytoca codes for a protein that is 52% identical to the product of the pilD/xcpA gene required for extracellular protein secretion and type IV pilus biogenesis in Pseudomonas aeruginosa. The PilD/XcpA protein is known to remove the first six amino acids of the signal sequence of the type IV pilin precursor by cleaving after the glycine residue in the conserved sequence GF(M)XXXE (where X represents hydrophobic amino acids). This prepilin peptidase cleavage site is present in the products of four genes in the pulC-O operon (PulG, PulH, Pull and PulJ proteins). It is shown here that PulO processes the pulG gene product in vivo. Processing was maximal within 15 seconds, but experiments in which the expression of pulO was uncoupled from that of the other genes in the secretion operon suggest that processing can also occur post-translationally. The products of two pulG derivatives with internal inframe deletions were also processed by PulO, but the three PulG-PhoA hybrids, two PulJ-PhoA hybrids and the single PulH-PhoA hybrid tested did not appear to be processed. Sucrose gradient fraction experiments showed that both precursor and mature forms of PulG appear to be associated with low-density, outer membrane vesicles prepared by osmotic lysis of sphaeroplasts. Neither the xcpA gene nor the Bacillus subtilis gene comC, which is also homologous to pulO and codes for a protein with type IV prepilin peptidase activity, can correct the pullulanase secretion defect in an Escherichia coli strain carrying all of the genes required for secretion except pulO. Furthermore, neither XcpA nor ComC is able to process prePulG protein in vivo.

Pullulanase secretion in Escherichia coli K-12 requires a cytoplasmic protein and a putative polytopic cytoplasmic membrane protein

The previously uncharacterized third and fourth genes (pulE and pulF) of the pullulanase secretion gene operon of Klebsiella oxytoca strain UNF5023 are, respectively, predicted to encode a 55 kDa polypeptide with a putative nucleotide-binding site, and a highly hydrophobic 44 kDa polypeptide that probably spans the cytoplasmic membrane several times. Expression of pulE in minicells or under the control of a strong bacteriophage T7 promoter resulted in the production of a c. 58 kDa cytoplasmic protein. A representative PulE-beta-galactosidase hybrid protein created by Tnlac mutagenesis was also found mainly in the cytoplasm. These results are in line with the predicted absence from PulE of a region of sufficient hydrophobicity to function as a signal sequence. The PulF polypeptide could not be detected either in minicells or when the gene was transcribed from the T7 promoter, but the acquirement of three pulF-lacZ gene fusions that encoded hybrid proteins with relatively high levels of beta-galactosidase activity indicates that this gene can be transcribed and translated. Gene disruption experiments indicated that both pulE and pulF are required for pullulanase secretion in Escherichia coli K-12. Both proteins exhibit considerable homology throughout their entire lengths with other proteins involved in protein secretion, pilin assembly, conjugation and transformation competence in a variety of bacteria. In addition, PulE protein has consensus sequences found in a wide variety of nucleotide-binding proteins. This study completes the initial characterization of the pullulanase secretion gene operon, which comprises 13 genes that are all essential for the transport of pullulanase across the outer membrane.

Components of the protein-excretion apparatus of Pseudomonas aeruginosa are processed by the type IV prepilin peptidase

In the Gram-negative pathogen Pseudomonas aeruginosa, mutants in the gene for the prepilin peptidase (pilD) are pleiotropic, as they not only fail to process pilin but also accumulate in the periplasm, in their mature form, several toxins and hydrolytic enzymes that are normally exported to the external medium (excreted). We have suggested that this excretion defect is due to the lack of PilD-dependent processing of proteins that share sequences in common with the prepilin subunit and that are components of a protein-excretion machinery. In this paper we report the isolation and characterization of transposon-induced excretion mutants with phenotypes similar to that of a pilD gene mutant. Using oligonucleotide probes designed to recognize sequences encoding the cleavage site of the type IV prepilins, we have isolated four linked genes with the predicted putative PilD-dependent cleavage site. Site-specific mutations within these genes have shown that they are required for protein excretion, and PilD-dependent processing of at least one of the four encoded proteins was demonstrated. Evidence suggests that similar components play a role in protein excretion in a wide variety of Gram-negative bacteria.

Neisseria gonorrhoeae prepilin export studied in Escherichia coli

The pilE gene of Neisseria gonorrhoeae MS11 and a series of pilE-phoA gene fusions were expressed in Escherichia coli. The PhoA hybrid proteins were shown to be located in the membrane fraction of the cells, and the prepilin product of the pilE gene was shown to be located exclusively in the cytoplasmic membrane. Analysis of the prepilin-PhoA hybrids showed that the first 20 residues of prepilin can function as an efficient export (signal) sequence. This segment of prepilin includes an unbroken sequence of 8 hydrophobic or neutral residues that form the N-terminal half of a 16-residue hydrophobic region of prepilin. Neither prepilin nor the prepilin-PhoA hybrids were processed by E. coli leader peptidase despite the presence of two consensus cleavage sites for this enzyme just after this hydrophobic region. Comparisons of the specific molecular activities of the four prepilin-PhoA hybrids and analysis of their susceptibility to proteolysis by trypsin and proteinase K in spheroplasts allow us to propose two models for the topology of prepilin in the E. coli cytoplasmic membrane. The bulk of the evidence supports the simplest of the two models, in which prepilin is anchored in the membrane solely by the N-terminal hydrophobic domain, with the extreme N terminus facing the cytoplasm and the longer C terminus facing the periplasm.

Activity of the Agrobacterium T-DNA transfer machinery is affected by virB gene products

The oriT (origin of transfer) sequence and mob (mobilization) genes of plasmid RSF1010 can functionally replace transfer DNA (T-DNA) borders to generate an RSF1010 intermediate transferable to plants through activities of the tumor-inducing (Ti)-plasmid virulence (vir) genes of Agrobacterium tumefaciens. Because the Ti plasmid virB gene products are hypothesized to form a membrane-localized T-DNA transport apparatus, we investigated whether specific virB genes were needed for RSF1010 transfer. Here we report that transformation of Nicotiana tabacum leaf explants by an RSF1010-derivative plasmid (pJW323) requires the essential virulence genes virB9, virB10, and virB11, consistent with the hypothesis that both the T-DNA and RSF1010 transfer intermediates utilize the same transport machinery. Further, while pJW323 is transferred into plant cells by Agrobacterium strains harboring both pJW323 and pTiA6, the initiation of crown gall tumors (i.e., T-DNA transfer) is greatly suppressed. Coordinate overexpression of the virB9, virB10, and virB11 gene products relieves pJW323-mediated oncogenic suppression and restores tumorigenicity, but does not increase the transfer frequency of pJW323 into plant cells. We propose that the virB9, virB10, and virB11 gene products function coordinately and stoichiometrically to enhance DNA transfer in a fashion specific for the T-DNA intermediate.

Structural characterization of protein secretion genes of the bacterial phytopathogen Xanthomonas campestris pathovar campestris: relatedness to secretion systems of other gram-negative bacteria

The nucleotide sequence was determined of a 5.3 kb region of the Xanthomonas campestris pathovar campestris genome carrying a gene cluster encoding protein secretion and pathogenicity functions. A putative promoter sequence and five open reading frames (ORF) which may be part of an operon were revealed. The five predicted primary translation products comprise 531, 390, 147, 169 and 138 amino acids with Mr values of 58,854, 42,299, 15,548, 18,214 and 15,108 respectively. A sixth, partial ORF is also present. Between ORF1 and ORF2 is a sequence of unknown function showing 7 bp duplications. The deduced amino acid sequence of ORF1 is related to the Klebsiella pneumoniae PulE protein, to the Bacillus subtilis ComG ORF1 and to the Agrobacterium tumefaciens VirB ORF11 products. In addition, the deduced amino acid sequence of ORF2 showed homology to the PulF and to the ComG ORF2 products. The proteins encoded by ORF3, 4 and 5 showed amino acid homology to PulG, H and I products respectively. The proteins encoded by ORF2, 3, 4 and 5 showed significant hydrophobic domains which may represent membrane-spanning regions. By contrast the protein encoded by ORF1 was largely hydrophilic and had two putative nucleoside triphosphate binding sites.

Biogenesis of the bacterial pilus

The assembly of surface structures in gram-negative bacteria requires specialized secretion and chaperone systems localized on both sides of the cytoplasmic membrane. Major advances have been made over the last year in understanding how these systems form part of a general strategy used by bacteria to cap and target interactive subunits imported into the periplasmic space to outer membrane uncapping and assembly sites.

Cloning, sequence analysis, and functional expression of the acetyl coenzyme A synthetase gene from Methanothrix soehngenii in Escherichia coli

In the acetoclastic methanogen Methanothrix soehngenii, acetate is activated to acetyl coenzyme A by acetyl coenzyme A synthetase (Acs). The acs gene, coding for the single Acs subunit, was isolated from a genomic library of M. soehngenii DNA in Escherichia coli by using antiserum raised against the purified Acs. After introduction in E. coli, the acs gene was expressed, resulting in the production of an immunoreactive protein of 68 kDa, which is approximately 5 kDa smaller than the known size of purified Acs. In spite of this difference in size, the Acs enzymes are produced in similar quantities in E. coli and M. soehngenii and show comparable specific activities. Upstream from the acs gene, consensus archaeal expression signals were identified. Immediately downstream from the acs gene there was a putative transcriptional stop signal. The amino acid sequence deduced from the nucleotide sequence of the acs gene showed homology with those of functionally related proteins, i.e., proteins involved in the binding of coenzyme A, ATP, or both.

Genetic competence in Bacillus subtilis

Genetic competence may be defined as a physiological state enabling a bacterial culture to bind and take up high-molecular-weight exogenous DNA (transformation). In Bacillus subtilis, competence develops postexponentially and only in certain media. In addition, only a minority of the cells in a competent culture become competent, and these are physiologically distinct. Thus, competence is subject to three regulatory modalities: growth stage specific, nutritionally responsive, and cell type specific. This review summarizes the present state of knowledge concerning competence in B. subtilis. The study of genes required for transformability has permitted their classification into two broad categories. Late competence genes are expressed under competence control and specify products required for the binding, uptake, and processing of transforming DNA. Regulatory genes specify products that are needed for the expression of the late genes. Several of the late competence gene products have been shown to be membrane localized, and others are predicted to be membrane associated on the basis of amino acid sequence data. Several of these predicted protein sequences show a striking resemblance to gene products that are involved in the export and/or assembly of extracellular proteins and structures in gram-negative organisms. This observation is consistent with the idea that the late products are directly involved in transport of DNA and is equally consistent with the notion that they play a morphogenetic role in the assembly of a transport apparatus. The competence regulatory apparatus constitutes an elaborate signal transduction system that senses and interprets environmental information and passes this information to the competence-specific transcriptional machinery. Many of the regulatory gene products have been identified and partially characterized, and their interactions have been studied genetically and in some cases biochemically as well. These include several histidine kinase and response regulator members of the bacterial two-component signal transduction machinery, as well as a number of known transcriptionally active proteins. Results of genetic studies are consistent with the notion that the regulatory proteins interact in a hierarchical way to make up a regulatory pathway, and it is possible to propose a provisional scheme for the organization of this pathway. It is remarkable that almost all of the regulatory gene products appear to play roles in the control of various forms of postexponential expression in addition to competence, e.g., sporulation, degradative-enzyme production, motility, and antibiotic production. This has led to the notion of a signal transduction network which transduces environmental information to determine the levels and timing of expression of the ultimate products characteristic of each of these systems.

Sequence and transcriptional regulation of com101A, a locus required for genetic transformation in Haemophilus influenzae

A 2.8-kb EcoRI-BglII fragment cloned from the wild-type Haemophilus influenzae Rd chromosome is shown to increase the transformability of the Com-101 mutant through trans complementation. Deletion and sequence analyses indicate that the active region of the clone carries a 687-bp open reading frame. A 0.3-kb insertion in the corresponding EcoRI-BglII fragment of the Com-101 chromosome is shown to be a partial (331-bp) duplication of this open reading frame. The wild-type sequence produces a peptide of a size that is consistent with the sequence data when this sequence is expressed in Escherichia coli with a T7 promoter-based transcription vector. RNA hybridization analysis using a DNA probe derived from the open reading frame suggests that the sequence is transiently expressed during competence development. On the basis of these observations, it is proposed that the open reading frame corresponds to the com101A gene.

Molecular cloning of two linked loci that increase the transformability of transformation-deficient mutants of Haemophilus influenzae

A plasmid containing a 13.3-kb insert (pER194) was isolated from an EcoRI genomic library of Haemophilus influenzae on the basis of its ability to increase the transformability of the transformation-deficient mutants Com-78 and Com-101. The plasmid failed to increase the transformability of the Rec-1 and Rec-2 mutants, indicating that the mutations producing the Com-78 and Com-101 phenotypes are distinct from those giving rise to the Rec-1 and Rec-2 phenotypes. The physical mapping of the cloned fragment on the H. influenzae chromosome was found to be consistent with the genetic mapping of the Com-101 trait. A 2.8-kb EcoRI-BglII subfragment, representing one end of the 13.3-kb clone, was found to increase the transformation frequency of the Com-78 and Com-101 mutants when supplied in trans, indicating that the subfragment carries one or more loci required for chromosomal transformation. The corresponding region of the Com-101 chromosome was determined by hybridization analysis to contain a 0.3-kb insertion, suggesting that the Com-101 strain may contain an insertion mutation at this locus. A 3.0-kb EcoRI-MluI subfragment, representing the other end of the 13.3-kb EcoRI fragment, was found to increase the transformation frequency of the Com-101 mutant but not of the Com-78 mutant, suggesting that the Com-101 phenotype results from a complex genotype involving mutations at two or more transformation-related loci. This conclusion is consistent with data indicating that the Com-101 trait can be genetically separated into at least two components.

Isolation, transcription, and inactivation of the gene for an atypical alkaline phosphatase of Synechococcus sp. strain PCC 7942

The alkaline phosphatase of Synechococcus sp. strain PCC 7942 is 145 kDa, which is larger than any alkaline phosphatase previously characterized and approximately three times the size of the analogous enzyme in Escherichia coli. The gene for the alkaline phosphatase, phoA, was cloned and sequenced, and the protein that it encodes was found to have little similarity to other phosphatases. Some sequence similarities were observed between the Synechococcus sp. strain PCC 7942 alkaline phosphatase, the alpha subunit of the ATPase from bacteria and chloroplasts, and the UshA sugar hydrolase of E. coli. Also, limited sequence similarity was observed between a region of the phosphatase and a motif implicated in nucleotide binding. Interestingly, although the alkaline phosphatase is transported across the inner cytoplasmic membrane and into the periplasmic space, it does not appear to have a cleavable signal sequence at its amino terminus. The half-life of the mRNA encoding the alkaline phosphatase, measured after inhibition of RNA synthesis, is approximately 5 min. Similar kinetics for the loss of alkaline phosphatase mRNA occur upon the addition of phosphate to phosphate-depleted cultures, suggesting that high levels of this nutrient inhibit transcription from phoA almost immediately. The phoA gene also appears to be the first gene of an operon; the largest detectable transcript that hybridizes to a phoA gene-specific probe is 11 kb, over twice the size needed to encode the mature protein. Other phosphate-regulated mRNAs are also transcribed upstream of the phoA gene. Insertional inactivation of phoA results in the loss of extracellular, phosphate-regulated phosphatase activity but does not alter the capacity of the cell for phosphate uptake.

Characterisation of a Pseudomonas aeruginosa twitching motility gene and evidence for a specialised protein export system widespread in eubacteria

Type-4 fimbriae (pili) are associated with a phenomenon known as twitching motility, which appears to be involved with bacterial translocation across solid surfaces. Pseudomonas aeruginosa mutants which produce fimbriae, but which have lost the twitching motility function, display altered colony morphology and resistance to fimbrial-specific bacteriophage. We have used phenotypic complementation of such mutants to isolate a region of DNA involved in twitching motility. This region was physically mapped to a SpeI fragment around 20 min on the P. aeruginosa PAO chromosome, remote from the major fimbrial locus (around 75 min) where the structural subunit-encoding gene (fimA/pilA) and ancillary genes required for fimbrial assembly (pilB, C and D) are found. A gene, pilT, within the twitching motility region is predicted to encode a 344-amino acid protein which has strong homology to a variety of other bacterial proteins. These include the P. aeruginosa PilB protein, the ComG ORF-1 protein from the Bacillus subtilis comG operon (necessary for competence), the PulE protein from the Klebsiella oxytoca (formerly K. pneumoniae) pulC-O operon (involved in pullulanase export), and the VirB-11 protein from the virB operon (involved in virulence) which is located on the Agrobacterium tumefaciens Ti plasmid. We have also identified other sets of homologies between P. aeruginosa fimbrial assembly (Pil) proteins and B. subtilis Com and K. oxytoca Pul proteins, which suggest that these are all related members of a specialised protein export pathway which is widespread in the eubacteria.

Product of the Pseudomonas aeruginosa gene pilD is a prepilin leader peptidase

The related type IV pilins produced by Pseudomonas aeruginosa, Neisseria gonorrhoeae, Bacteroides nodosus, and Moraxella bovis are synthesized as precursors with short, six- or seven-amino acid N-terminal leader peptides. We have previously observed that P. aeruginosa mutations in pilD, a gene required for pilus biogenesis, result in the accumulation of unprocessed prepilin in the membrane and a general defect in the excretion of a number of extracellular enzymes. An endopeptidase activity has been detected in detergent-solubilized inner membrane of P. aeruginosa and shown to correctly cleave the prepilin of P. aeruginosa and N. gonorrhoeae. It is absent from pilD mutants, increased by pilD overexpression, and conferred on Escherichia coli by the introduction of the pilD gene. The pilD gene product, purified by immunoaffinity chromatography with antibody to a PilD-derived synthetic peptide, was identified with the endopeptidase. PilD appears to be a prototype of a class of enzymes that process not only type IV pilin precursors but also components of a protein-excretion apparatus of Gram-negative bacteria.

DegS-DegU and ComP-ComA modulator-effector pairs control expression of the Bacillus subtilis pleiotropic regulatory gene degQ

Production of a class of both secreted and intracellular degradative enzymes in Bacillus subtilis is regulated at the transcriptional level by a signal transduction pathway which includes the DegS-DegU two-component system and at least two additional regulatory genes, degQ and degR, encoding polypeptides of 46 and 60 amino acids, respectively. Expression of degQ was shown to be controlled by DegS-DegU. This expression is decreased in the presence of glucose and increased under any of the following conditions: growth with poor carbon sources, amino acid deprivation, phosphate starvation, and growth in the presence of decoyinine, a specific inhibitor of GMP synthetase. In addition, expression of degQ is shown to be positively regulated by the ComP-ComA two-component system. Separate targets for regulation of degQ gene expression by DegS-DegU and ComP-ComA were located by deletion analysis between positions -393 and -186 and between positions -78 and -40, respectively. Regulation of degQ expression by amino acid deprivation was shown to be dependent upon ComA. Regulation by phosphate starvation, catabolite repression, and decoyinine was independent of the two-component systems and shown to involve sequences downstream from position -78. The ComP-ComA and DegS-DegU two-component systems seem to be closely related, sharing several target genes in common, such as late competence genes, as well as the degQ regulatory gene. Sequence analysis of the degQ region revealed the beginning of an open reading frame directly downstream from degQ. Disruption of this gene, designated comQ, suggests that it also controls expression of degQ and is required for development of genetic competence.

The conjugative transposon Tn925: enhancement of conjugal transfer by tetracycline in Enterococcus faecalis and mobilization of chromosomal genes in Bacillus subtilis and E. faecalis

The effects of tetracycline on transfer of the conjugative, tetracycline-resistance transposon. Tn925, as well as the ability of the transposon to promote the transfer of chromosomal genes was examined in Enterococcus faecalis and Bacillus subtilis. To test for chromosomal transfer, multiply-marked strains of each organism, each carrying a single chromosomal copy of Tn925, were mated on filters with suitable recipient strains, under conditions where transformation and transduction were precluded. In both cases, transfer of a variety of chromosomal genes, at frequencies comparable to the frequency of Tn925 transfer, was detected readily. The presence of Tn925 in one of the members of the mating pair was absolutely required for chromosomal transfer, but transfer of Tn925 did not accompany every chromosomal transfer event. The results were consistent with a mating event resembling a type of cell fusion, allowing for extensive recombination between the genomes of the mating partners. Growth of Tn925-containing donor cells in the presence of tetracycline increased the transfer frequency of Tn925 by about tenfold in E. faecalis, but not in B. subtilis.

Pilin expression and processing in pilus mutants of Neisseria gonorrhoeae: critical role of Gly-1 in assembly

Spontaneous mutants of Neisseria gonorrheae failing to express pili or having diminished levels of piliation were studied with regard to pilin expression. All mutants displayed altered pilin processing detectable as the release of soluble, truncated pilin molecules (S-pilin). Of particular interest was the finding, in one mutant, that substitution of serine for glycine at position -1 of propilin, a highly conserved residue among N-metPhe and related pilins, abolished pilus expression but not S-pilin release. The degree of S-pilin processing and the levels of membrane-associated pilin varied among the different classes of mutants, suggesting that each was blocked at a distinct step of pilus biogenesis. The data support a model in which increased S-pilin processing is a result of a decreased rate of pilus polymerization.

The regulation of genetic competence in Bacillus subtilis

Genetic competence develops as a global response of Bacillus subtilis to the onset of stationary phase, in glucose-minimal salts-based media. The onset of competence is accompanied by the expression of several late gene products that are required for the binding, processing and uptake of transforming DNA. A number of regulatory genes have been identified that are needed for the appropriate synthesis of the late gene products. The regulatory gene products include a number of known transcription factors, as well as several members of the bacterial two-component regulatory system. Genetic analysis has suggested a scheme for the flow of regulatory information signalling the onset of competence. Most of these regulatory products appear to be involved in the response to nutritional status, while the components responsible for growth stage and cell-type-specific control remain unknown. The general implications of this scheme for post-exponential expression are discussed.

Complementation analysis of Agrobacterium tumefaciens Ti plasmid virB genes by use of a vir promoter expression vector: virB9, virB10, and virB11 are essential virulence genes

The virB gene products of the Agrobacterium tumefaciens tumor-inducing (Ti) plasmid have been proposed to mediate T-DNA transport through the bacterial cell wall into plant cells. Previous genetic analysis of the approximately 9.5-kilobase-pair virB operon has been limited to transposon insertion mutagenesis. Due to the polarity of the transposon insertions, only the last gene in the operon, virB11, is known to provide an essential virulence function. We have now begun to assess the contribution of the other virB genes to virulence. First, several previously isolated Tn3-HoHo1 insertions in the 3' end of the virB operon were precisely mapped by nucleotide sequence analysis. Protein extracts from A. tumefaciens strains harboring these insertions on the Ti plasmid were subjected to immunostaining analysis with VirB4-, VirB10-, and VirB11-specific antisera to determine the effect of the insertion on virB gene expression. In this manner, avirulent mutants containing polar insertions in the virB9 and virB10 genes were identified. To carry out a complementation analysis with these virB mutants, expression vectors were constructed that allow cloned genes to be expressed from the virB promoter in A. tumefaciens. These plasmids were used to express combinations of the virB9, virB10, and virB11 genes in trans in the virB insertion mutants, thereby creating strains lacking only one of these three virB gene products. Virulence assays on Kalanchoe daigremontiana demonstrated that in addition to virB11, the virB9 and virB10 genes are required for tumorigenicity.

Identification of a virB10 protein aggregate in the inner membrane of Agrobacterium tumefaciens

Products of the virB operon are proposed components of a membrane-associated T-DNA transport apparatus in Agrobacterium tumefaciens. Here we identified the virB10 gene product and raised specific antiserum to the protein. While the virB10 reading frame contains two potential ATG translation start sites located 32 codons apart, we found that only the downstream ATG was required for efficient VirB10 synthesis. Cellular localization studies and analysis of translational fusions with the Escherichia coli alkaline phosphatase gene (phoA) indicated that VirB10 was anchored in the inner membrane and contained a periplasmic domain. This work also demonstrated the utility of alkaline phosphatase as a reporter for secreted proteins in A. tumefaciens. Several high-molecular-weight forms of VirB10 were observed after treatment of A. tumefaciens whole cells or inner membranes with protein cross-linking agents, suggesting that VirB10 exists as a native oligomer or forms an aggregate with other membrane proteins. These results provide the first biochemical evidence that a VirB protein complex is membrane associated in A. tumefaciens.

Transcriptional regulation of comC: evidence for a competence-specific transcription factor in Bacillus subtilis

comC specifies a protein product that is required for genetic competence in Bacillus subtilis. The probable transcriptional start site of comC has been localized by high-resolution primer extension analysis and shown to be preceded by an appropriately positioned sequence that resembles the consensus promoter for the sigma A form of RNA polymerase. Low-resolution S1 nuclease transcription mapping was used to identify the comC terminator, which is located near a palindromic element recognizable in the DNA sequence. Deletion analysis of the sequence upstream from the likely promoter identified a region required in cis for the expression of comC. An overlapping, and possibly identical, sequence was shown to inhibit the expression of competence and of several late competence genes, when present in multiple copies. This was interpreted as due to the titration of a positively acting competence transcription factor (CTF) by multiple copies of the promoter-bearing fragment. In crude lysates of B. subtilis grown to competence, a DNA-binding activity that appeared to be specific for the comC promoter fragment was detected by gel retardation assays. This activity, postulated to be due to CTF, was detected only following growth in competence medium, only in the stationary phase of growth, and was dependent on the expression of ComA, a known competence-regulatory factor. In the presence of the mecA42 mutation, the ComA requirement for CTF activity was bypassed, and CTF activity could be detected in lysates prepared from a strain grown in complex medium. This behavior suggested that either the expression or the activation of CTF was regulated in a competence-specific manner. Comparison of the putative CTF-binding site defined by deletion analysis with a similarly positioned sequence upstream from the start site of the late competence gene comG revealed that both sequences contained palindromes, with 5 of 6 identical base pairs in each arm. It is suggested that these palindromic sequences comprise recognition elements for CTF binding and that CTF binding must occur for the appropriate expression of late competence genes.

Identification of cis-acting sequences required for translational autoregulation of the ermC methylase

ermC methylase gene expression has been shown to be limited by translational autorepression, presumably due to methylase binding to ermC mRNA. It was found that this repression occurs in trans, yielding a 50% reduction in translation of an ermC-lacZ fusion mRNA. We investigated the ermC mRNA sequences required for translational repression in vivo. A series of deletions identified sequences in the 5' regulatory region that were required for translational repression. These included sequences of the 5' stem-loop structure that were not required for induction, as well as some that were required. The implications of these results for regulation are discussed.

Growth medium-independent genetic competence mutants of Bacillus subtilis

The development of competence in Bacillus subtilis is normally dependent on the growth medium. Expression of late competence genes occurs in glucose-minimal salts-based media but not in complex media. Expression is also inhibited when glutamine is added to competence medium and when glycerol is substituted for glucose. Mutations have been identified in two regulatory loci, mecA and mecB, which render competence development independent of these variables. Although in mec mutants the expression of late competence genes, as well as of competence itself, occurred in all media tested, this expression was still growth stage regulated. Thus at least some forms of medium-dependent and growth stage-specific regulation are genetically separable. One of the mecB mutations (mecB31) conferred oligosporogenicity. The mecB mutations were tightly linked by transformation to rif, lpm, and std markers and were located between rif-2103 and cysA14. The mecA42 mutant was linked by transduction to argC4.

Characterization of the virB operon of an Agrobacterium tumefaciens Ti plasmid: nucleotide sequence and protein analysis

The virulence regulon of the Agrobacterium tumefaciens TiC58 plasmid is composed of six operons, virA, virB, virG, virC, virD and virE, which direct the transfer of T-DNA into plant cells. The 9.5 kbp virB operon is the largest of these operons and its entire nucleotide sequence was determined and found to contain eleven open reading frames (ORFs). Gene fusions of each VirB ORF to T7 phi 10 were made and overexpressed in Escherichia coli to confirm that they encode proteins of predicted size. Hydrophobic analysis of these peptide sequences revealed nine proteins that contain hydrophobic spanning regions including signal-peptide-like sequences. These data suggest that the majority of VirB proteins may associate with bacterial cell membranes, while the two additional proteins possess a potential ATP-binding site. Strong homologies in amino acid sequences were observed between nopaline- and octopine-type plasmids. Specific differences in amino acid sequence encoded by VirB ORFs of nopaline and octopine Ti plasmid and a functional role of the gene products are discussed.

Suppression of early competence mutations in Bacillus subtilis by mec mutations

Although competence normally develops only in glucose-minimal salts media, mecA and mecB mutations permit the expression of competence and of late competence genes in complex media as well (D. Dubnau and M. Roggiani, J. Bacteriol. 172:4048-4055, 1990). The expression of late competence genes is dependent on the products of the regulatory genes comA, comB, comP, sin, abrB, spo0H, and spo0A. We show here that this list must be extended to include degU, csh-293, and spo0K. mecA and -B mutations bypass most of these requirements, making the expression of late competence genes and of competence itself independent of all of these regulatory genes, with the exceptions of spo0A and spo0K (in the case of mecB). The expression of late competence genes in mec mutants that are deficient for each of the bypassed regulatory functions is still under growth stage-specific regulation. The implications of these findings are discussed, and a provisional scheme for the flow of information during the development of competence is proposed.

A membrane protein with similarity to N-methylphenylalanine pilins is essential for DNA binding by competent Bacillus subtilis

In a cloned copy of comG open reading frame 3 (ORF3), an in-frame deletion was generated by site-directed in vitro mutagenesis, removing the coding sequence for 15 amino acids from the central portion of this pilin-related protein. The mutagenized ORF3 was incorporated into the Bacillus subtilis chromosome, replacing the wild-type ORF3. The presence of the deleted ORF3 in the chromosome, as confirmed by Southern analysis, was associated with the complete loss of competence by the mutant strain. The ability of the mutant cells to bind exogenous radiolabeled DNA was reduced to the level of nonspecific binding of DNA by noncompetent cells. The chromosomal ORF3 mutation was partially complemented in trans by a plasmid-encoded wild-type ORF3 copy under PSPAC control upon induction of the PSPAC promoter. Using antiserum raised against a synthetic 14-mer oligopeptide deduced from the ORF3 sequence, an immunoreactive band of approximately the expected molecular size was obtained in Western blot (immunoblot) experiments with extracts of cells containing the plasmid-encoded inducible gene. A signal was also detected when cells harboring the chromosomal wild-type or mutant ORF3 in single copy were grown in competence medium. This signal was detected only in the light-buoyant-density (competent) cell fraction and only after the transition from the exponential to the stationary growth phase. In cell fractionation experiments with competent cell extracts, the immunoreactive protein was found in both the NaOH-insoluble and -soluble membrane fractions and was sensitive to proteinase K treatment of either protoplasts or whole cells.

A gene required for transfer of T-DNA to plants encodes an ATPase with autophosphorylating activity

The virB operon of the Agrobacterium tume-faciens pTiA6NC plasmid likely plays a role in directing T-DNA transfer events at the bacterial membrane, as determined previously by mutagenesis and cellular fractionation studies and by DNA sequence analysis of the approximately 12-kilobase-pair operon. The DNA sequence analysis also revealed consensus mononucleotide binding domains in the deduced virB5 and virB11 gene products, suggesting that one or both of these proteins couple energy, by means of nucleotide triphosphate (NTP) hydrolysis, to T-DNA transport. In this report, the product of virB11, an essential virulence gene, was overproduced in Escherichia coli and purified by using immunoaffinity chromatography. The immunoaffinity purified protein, as well as NaDodSO4/polyacrylamide gel-eluted protein, bound and hydrolyzed ATP in the absence of DNA effectors. VirB11 protein also demonstrated in vitro autophosphorylation activity. VirB11 protein was localized primarily to the cytoplasmic membrane by immunoblot analysis of membrane fractions. The deduced VirB11 protein exhibits sequence similarity to comG ORF1, a protein required for uptake of DNA by competent Bacillus subtilis cells. These findings suggest that phosphorylation may serve to activate a component(s) of the A. tumefaciens T-DNA transport apparatus and may also represent a general activation mechanism of other bacterial DNA transport systems.

Molecular cloning and characterization of comC, a late competence gene of Bacillus subtilis

comC is a Bacillus subtilis gene required for the development of genetic competence. We have cloned a fragment from the B. subtilis chromosome that carries comC and contains all the information required to complement a Tn917lac insertion in comC. Genetic tests further localized comC to a 2.0-kilobase HindIII fragment. Northern (RNA) blotting experiments revealed that an 800-base-pair comC-specific transcript appeared at the time of transition from exponential to stationary phase during growth through the competence regimen. The DNA sequence of the comC region revealed two open reading frames (ORFs), transcribed in the same direction. The upstream ORF encoded a protein with apparent sequence similarity to the folC gene of Escherichia coli. Insertion of a chloramphenicol resistance determinant into this ORF and integration of the disrupted construct into the bacterial chromosome by replacement did not result in competence deficiency. The downstream ORF, which contained the Tn917lac insertion that resulted in a lack of competence, is therefore the comC gene. The predicted protein product of comC consisted of 248 amino acid residues and was quite hydrophobic. The comC gene product was not required for the expression of any other com genes tested, and this fact, together with the marked hydrophobicity of ComC, suggests that it may be a component of the DNA-processing apparatus of competent cells.

Cloning and characterization of a cluster of linked Bacillus subtilis late competence mutations

We characterized a segment of chromosomal DNA from Bacillus subtilis that was required for the development of genetic competence. The chromosomal DNA was cloned from a group of genetically linked and phenotypically similar Tn917lac insertion mutants deficient in competence. This cluster of mutations defined the comG locus. Chromosomal DNA flanking each of the six insertions was cloned. Restriction maps of the cloned plasmids revealed that their chromosomal inserts consisted of overlapping fragments. These data, together with Southern blots of chromosomal DNA from the comG mutants, showed that the six Tn917lac comG insertions occurred in the following order: comG12, comG39, comG412, comG107, comG56, and comG210. Expression of the comG Tn917lac insertions was from a promoter located upstream from the first insertion, comG12. This was determined genetically and by low-resolution S1 nuclease mapping of the 3' terminus. The comG region spanned about 5 kilobase pairs, based on low-resolution S1 nuclease mapping of the transcription terminator and Northern blotting. The comG12 mutation had a partial epistatic effect on the expression of one other com locus, comE, but none of the other comG mutations affected expression of this or any other com gene tested. Based on these conclusions, and on its size and phenotype, the comG locus must be organized as a polycistronic operon that is subject to competence-specific regulation.

Cloning and characterization of the regulatory Bacillus subtilis competence genes comA and comB

comA and comB are Bacillus subtilis competence genes that are identified by insertions of Tn917lac. They are classified as early genes because of their expression throughout growth; the expression of late com genes increases sharply during the transition to the stationary phase. The comA and comB determinants were cloned, and the 5' and 3' termini of their transcripts were localized by low-resolution S1 nuclease protection experiments. comA and comB were found by Southern blotting to be localized near one another, but they were nevertheless apparently transcribed independently. Epistatic relationships among the com genes were explored by using the beta-galactosidase expressed from transcriptional fusions as a marker. Late com genes were found to be dependent on the products of comA, comB, and sin for their expression. The sin gene is a transcriptional regulator that is required for the development of competence (N. K. Gaur, E. Dubnau, and I. Smith, J. Bacteriol. 168:860-869, 1986).