Nucleotide sequence and genetic organization of the Bacillus subtilis comG operon

First steps from a two-dimensional protein index towards a response-regulation map for Bacillus subtilis

Data on the identification of proteins of Bacillus subtilis on two-dimensional (2-D) gels as well as their regulation are summarized and the identification of 56 protein spots is included. The pattern of proteins synthesized in Bacillus subtilis during exponential growth, during starvation for glucose or phosphate, or after the imposition of stresses like heat shock, salt- and ethanol stress as well as oxidative stress was analyzed. N-terminal sequencing of protein spots allowed the identification of 93 proteins on 2-D gels, which are required for the synthesis of amino acids and nucleotides, the generation of ATP, for glycolyses, the pentose phosphate cycle, the citric acid cycle as well as for adaptation to a variety of stress conditions. A computer-aided analysis of the 2-D gels was used to monitor the synthesis profile of more than 130 protein spots. Proteins performing housekeeping functions during exponential growth displayed a reduced synthesis rate during stress and starvation, whereas spots induced during stress and starvation were classified as specific stress proteins induced by a single stimulus or a group of related stimuli, or as general stress proteins induced by a variety of entirely different stimuli. The analysis of mutants in global regulators was initiated in order to establish a response regulation map for B. subtilis. These investigations demonstrated that the alternative sigma factor sigma B is involved in the regulation of almost all of the general stress proteins and that the phoPR two-component system is required for the induction of a large part but not all of the proteins induced by phosphate starvation.

Identification and characterization of xcpR encoding a subunit of the general secretory pathway necessary for dodecane degradation in Acinetobacter calcoaceticus ADP1

A mutant of Acinetobacter calcoaceticus ADP1 unable to grow on alkanes was complemented for growth on hexadecane with a DNA fragment encoding a protein with homology to XcpR, a subunit of the general secretion pathway for exoproteins in Pseudomonas aeruginosa. Insertional inactivation of xcpR in A. calcoaceticus ADP1 by transcriptional fusion to lacZ abolishes secretion of lipase and esterase and leads to lack of growth on dodecane and slower growth on hexadecane. We, therefore, propose the participation of a secreted protein in alkane degradation.

Uptake and processing of DNA by Acinetobacter calcoaceticus--a review

In natural transformation, DNA in the form of macromolecular fragments can be translocated across the cell envelope of prokaryotic microorganisms. During the past two decades, several, largely mutually contradictory, hypotheses have been forwarded to explain the molecular mechanism and bioenergetics of this translocation process. Other biomacromolecules are translocated across the bacterial cell envelope as well, such as polysaccharides and proteins, the latter for instance in the process of the assembly of type-IV pili. This brings up the question whether or not common components are involved. Here, we review analyses of DNA translocation in Acinetobacter calcoaceticus, a Gram-negative eubacterium that is able to migrate through twitching motility, and also shows a high frequency of natural transformation. DNA uptake in this organism is an energy-dependent process. Upon entry into the cells, the DNA fragments are integrated into the resident chromosome when a sufficiently large region of mutual homology is available (200 to 400 bp). However, this process is rather inefficient, and on the average 500 bp of each incoming fragment is degraded through exonuclease activity. Upon covalent attachment of a bulky protein molecule to the transforming DNA, the DNA-translocation machinery becomes blocked in further translocation activity. Since A. calcoaceticus is not well suited for transposon mutagenesis, a random mutagenesis procedure has been developed, based on the ligation of an antibiotic-resistance marker to random fragments of chromosomal DNA. This method was used to generate several mutants impaired in the natural transformation process. Three of these have been characterized in detail. No components, common to the translocation of macromolecules through the cell envelope of Acinetobacter, have been detected in this screen.

Binding and transport of transforming DNA by Bacillus subtilis: the role of type-IV pilin-like proteins--a review

The pathway for binding, processing and transport of transforming DNA into competent cells of Bacillus subtilis is described. The known proteins involved in mediating these processes are reviewed in turn, including several that resemble proteins required for type-IV pilus assembly and function, and those involved in protein secretion. Based on the phenotypes of null mutations in the cognate genes for these proteins, on similarities to other proteins and on membrane localization and topology data, proposals are made for the roles of the individual proteins in the transformation process. A dynamic model is suggested for the presentation of transforming DNA to the transport apparatus.

The pilus colonization factor of pathogenic neisserial species: organelle biogenesis and structure/function relationships--a review

Type-IV pilus expression plays a critical role in the interactions between Neisseria gonorrhoeae, Neisseria meningitidis and their human host. We have focused on experiments designed to elucidate the mechanisms of organelle biogenesis as one means of understanding the complexities of pilus biology in these species. Employing a variety of approaches, genes and gene products essential to pilus biogenesis have been identified and characterized. The findings indicate that the neisserial type-IV pilus biogenesis machinery is most closely related to that operating in Pseudomonas aeruginosa and other pseudomonad species. This interrelatedness is documented at the levels of gene organization, DNA homologies and identities between the primary structures of the components. Despite these similarities, the biological correlates of pilus expression in the pathogenic Neisseria are quite unique. The current status of our embryonic understanding of the factors influencing organelle biogenesis is presented. In the context of this workshop, emphasis has been placed on specific contributions made through studies of gonococci and meningococci to the field as a whole..

Genes involved in the biogenesis and function of type-4 fimbriae in Pseudomonas aeruginosa

Type-4 fimbriae are filamentous polar organelles which are found in a wide variety of pathogenic bacteria. Their biogenesis and function is proving to be extremely complex, involving the expression and coordinate regulation of a large number of genes. Type-4 fimbriae mediate attachment to host epithelial tissues and a form of surface translocation called twitching motility. In Pseudomonas aeruginosa they also appear to function as receptors for fimbrial-dependent bacteriophages. Analysis of mutants defective in fimbrial function has allowed the identification of many of the genes involved in the biogenesis of these organelles. Thus far over 30 genes have been characterized, which fall into two broad categories: those encoding regulatory networks that control the production and function of these fimbriae (and other virulence determinants such as alginate) in response to alterations in environmental conditions; and those encoding proteins involved in export and assembly of these organelles, many of which are similar to proteins involved in protein secretion and DNA uptake. These systems all appear to be closely related and to function in the assembly of surface-associated protein complexes that have been adapted to different biological functions.

comYA, a gene similar to comGA of Bacillus subtilis, is essential for competence-factor-dependent DNA transformation in Streptococcus gordonii

Tn4001 mutagenesis identified a new competence gene in Streptococcus gordonii Challis designated comYA. A comYA mutant was completely deficient in transformation and exhibited decreased levels of DNA binding and hydrolysis. The deduced 319-amino-acid ComYA protein exhibited 57% similarity and 33% identity to the ComGA transporter protein of Bacillus subtilis and contained the Walker A-box motif conserved in ATP-binding proteins as well as aspartic acid boxes Asp-1 and Asp-2 present in some components of the general secretory pathway of gram-negative bacteria. comYA appeared to be part of a putative operon encompassing a comGB homolog, designated comYB, together with sequences that could encode ComGC- and ComGD-like peptides designated ComYC and ComYD, respectively, as well as other components. The putative ComYC and ComYD peptides had leader sequences similar to the type IV N-methylphenylalanine pilins of gram-negative bacteria, but unlike other examples in this class, including B. subtilis, they contained an alanine at position -1 of the leader instead of the usual glycine residue. Northern analysis identified a single 6.0-kb comYA-containing transcript strictly dependent on exogenous competence factor for expression in ComA1 cells. An identical pattern of expression was seen in wild-type Challis cells grown under conditions of maximal competence but not in cells that were noncompetent.

Interactions between VirB9 and VirB10 membrane proteins involved in movement of DNA from Agrobacterium tumefaciens into plant cells

The 11 VirB proteins from Agrobacterium tumefaciens are predicted to form a membrane-bound complex that mediates the movement of DNA from the bacterium into plant cells. The studies reported here on the possible VirB protein interactions in such a complex demonstrate that VirB9 and VirB10 can each form high-molecular-weight complexes after treatment with a chemical cross-linker. Analysis of nonpolar virB mutants showed that the formation of the VirB10 complexes does not occur in a virB9 mutant and that VirB9 and VirB10 are not components of the same cross-linked complex. VirB9, when stabilized by the concurrent expression of VirB7, was shown to be sufficient to permit VirB10 to cross-link into its usual high-molecular-weight forms in the absence of other Vir proteins. Randomly introduced single point mutations in virB9 resulted in Agrobacterium strains with severely attenuated virulence. Although some of the mutants contained wild-type levels of VirB9 and displayed an unaltered VirB9 cross-linking pattern, VirB10 cross-linking was drastically reduced. We conclude that specific amino acid residues in VirB9 are necessary for interaction with VirB10 resulting in the capacity of VirB10 to participate in high-molecular-weight complexes that can be visualized by chemical cross-linking.

Fimbrial biogenesis genes of Pseudomonas aeruginosa: pilW and pilX increase the similarity of type 4 fimbriae to the GSP protein-secretion systems and pilY1 encodes a gonococcal PilC homologue

Type 4 fimbriae of Pseudomonas aeruginosa are surface filaments involved in host colonization. They mediate both attachment to host epithelial cells and flagelia-independent twitching motility. Four additional genes, pilW, pilX, pilY1 and pilY2, are located on Spel fragment E in the 5 kb intergenic region between the previously characterized genes pilV and pilE, which encode prepilin-like proteins involved in type 4 fimbrial biogenesis. The phenotypes of a transposon insertion and other mutations constructed by allelic exchange show that these genes are involved in the assembly of type 4 fimbriae. The PilW and PilX proteins are membrane located, possess the hydrophobic N-terminus characteristic of prepilin-like proteins, and appear to belong to the GspJ and GspK group of proteins that are required for protein secretion in a wide range of Gram-negative bacteria. These findings increase the similarities between the fimbrial biogenesis and the Gsp-based protein-secretion supersystems. PilY1 is a large protein with C-terminal homology to the PilC2 protein of Neisseria gonorrhoeae, thought to be a fimbrial tip-associated adhesin, and which, like PilY1, is involved in fimbrial assembly. PilY1 appears to be located in both the membrane and the external fimbrial fractions. PilY2 is a small protein that appears to play a subtle role in fimbrial biogenesis and represents a new class of protein.

Enteropathogenic Escherichia coli: identification of a gene cluster coding for bundle-forming pilus morphogenesis

Sequence flanking the bfpA locus on the enteroadherent factor plasmid of the enteropathogenic Escherichia coli (EPEC) strain B171-8 (O111:NM) was obtained to identify genes that might be required for bundle-forming pilus (BFP) biosynthesis. Deletion experiments led to the identification of a contiguous cluster of at least 12 open reading frames, including bfpA, that could direct the synthesis of a morphologically normal BFP filament. Within the bfp gene cluster, we identified open reading frames that share homology with other type IV pilus accessory genes and with genes required for transformation competence and protein secretion. Immediately upstream of the bfp gene cluster, we identified a potential replication origin including genes that are predicted to encode proteins homologous with replicase and resolvase. Restriction fragment length polymorphism analysis of DNA from six additional EPEC serotypes showed that the organization of the bfp gene cluster and its juxtaposition with a potential plasmid origin of replication are highly conserved features of the EPEC biotype.

Compilation and analysis of Bacillus subtilis sigma A-dependent promoter sequences: evidence for extended contact between RNA polymerase and upstream promoter DNA

Sequence analysis of 236 promoters recognized by the Bacillus subtilis sigma A-RNA polymerase reveals an extended promoter structure. The most highly conserved bases include the -35 and -10 hexanucleotide core elements and a TG dinucleotide at position -15, -14. In addition, several weakly conserved A and T residues are present upstream of the -35 region. Analysis of dinucleotide composition reveals A2- and T2-rich sequences in the upstream promoter region (-36 to -70) which are phased with the DNA helix: An tracts are common near -43, -54 and -65; Tn tracts predominate at the intervening positions. When compared with larger regions of the genome, upstream promoter regions have an excess of An and Tn sequences for n > 4. These data indicate that an RNA polymerase binding site affects DNA sequence as far upstream as -70. This sequence conservation is discussed in light of recent evidence that the alpha subunits of the polymerase core bind DNA and that the promoter may wrap around RNA polymerase.

Translation of the open reading frame encoded by comS, a gene of the srf operon, is necessary for the development of genetic competence, but not surfactin biosynthesis, in Bacillus subtilis

A small open reading frame, comS of the srf operon, is the site of mutations that impair competence development in Bacillus subtilis. comS open reading frame translation was required for competence, as was confirmed by the suppression of a comS amber mutation [comS(Am)] by the nonsense suppressor sup-3. comS(Am), when introduced into the srf operon, eliminated late competence gene expression but had no significant effect on surfactin production.

ComEA, a Bacillus subtilis integral membrane protein required for genetic transformation, is needed for both DNA binding and transport

The competence-related phenotypes of mutations in each of the four open reading frames associated with the comE locus of Bacillus subtilis are described. comEA and comEC are required for transformability, whereas the products of comEB and of the overlapping comER, which is transcribed in the reverse direction, are dispensable. Loss of the comEA product decreases the binding of DNA to the competent cell surface and the internalization of DNA, in addition to exhibiting a profound effect on transformability. The comEC product is required for internalization but is dispensable for DNA binding. ComEA is shown to be an integral membrane protein, as predicted from hydropathy analysis, with its C-terminal domain outside the cytoplasmic membrane. This C-terminal domain possesses a sequence with similarity to those of several proteins known to be involved in nucleic acid transactions including UvrC and a human protein that binds to the replication origin of the Epstein-Barr virus.

The major role of Spo0A in genetic competence is to downregulate abrB, an essential competence gene

We show that the major role for Spo0A in the development of genetic competence is to downregulate expression of abrB. AbrB is both a negative regulator and a positive regulator of competence. The negative effects are exerted at multiple points in competence regulation. A regulatory mechanism that is independent of mecA and abrB operates on comK expression.

Characterization of the pilF-pilD pilus-assembly locus of Neisseria gonorrhoeae

Expression of Type IV pili by the bacterial pathogen Neisseria gonorrhoeae appears to be essential for colonization of the human host. Several N. gonorrhoeae gene products have been recently identified which bear homology to proteins involved in pilus assembly and protein export in other bacterial systems. We report here the isolation and characterization of transposon insertion mutants in N. gonorrhoeae whose phenotypes indicate that the N. gonorrhoeae pilF and pilD gene products are required for gonoccocal pilus biogenesis. Mutants lacking the pilD gene product, a pre-pilin peptidase, were unable to process the pre-pilin subunit into pilin and thus were non-piliated. pilF mutants processed pilin but did not assemble the mature subunit. Both classes of mutants released S-pilin, a soluble, truncated form of the pilin subunit previously correlated with defects in pilus assembly. In addition, mutants containing transposon insertions in pilD or in a downstream gene, orfX, exhibited a severely restricted growth phenotype. Deletion analysis of pilD indicated that the poor growth phenotype observed for the pilD transposon mutants was a result of polar effects of the insertions on orfX expression. orfX encodes a predicted polypeptide of 23 kDa which contains a consensus nucleotide-binding domain and has apparent homologues in Pseudomonas aeruginosa, Pseudomonas putida, Thermus thermophilus, and the eukaryote Caenorhabditis elegans. Although expression of orfX and pilD appears to be transcriptionally coupled, mutants containing transposon insertions in orfX expressed pili. Unlike either pilF or pilD mutants, orfX mutants were also competent for DNA transformation.

Identification and characterization of pilG, a highly conserved pilus-assembly gene in pathogenic Neisseria

Expression of type IV pili appears to be a requisite determinant of infectivity for the strict human pathogens Neisseria gonorrhoeae and Neisseria meningitidis. The assembly of these colonization factors is a complex process. This report describes a new pilus-assembly gene, pilG, that immediately precedes the gonococcal (Gc) pilD gene encoding the pre-pilin leader peptidase. The nucleotide sequence of this region revealed a single complete open reading frame whose derived polypeptide displayed significant identities to the pilus-assembly protein PilC of Pseudomonas aeruginosa and other polytopic integral cytoplasmic membrane constituents involved in protein export and competence. A unique polypeptide of M(r) 38 kDa corresponding to the gene product was identified. A highly related gene and flanking sequences were cloned from a group B polysaccharide-producing strain of N. meningitidis (Mc). The results indicate that the pilG genes and genetic organization at these loci in Gc and Mc are extremely conserved. Hybridization studies strongly suggest that pilG-related genes exist in commensal Neisseria species and other species known to express type IV pili. Defined genetic lesions were created by using insertional and transposon mutagenesis and moved into the Gc and Mc chromosomes by allelic replacement. Chromosomal pilG insertion mutants were devoid of pili and displayed dramatically reduced competence for transformation. These findings could not be ascribed to pilin-gene alterations or to polarity exerted on pilD expression. The results indicated that PilG exerts its own independent role in neisserial pilus biogenesis.

Convergent sensing pathways mediate response to two extracellular competence factors in Bacillus subtilis

Development of genetic competence in Bacillus subtilis is regulated by extracellular signaling molecules, including the ComX pheromone, a modified 9- or 10-amino-acid peptide. Here, we present characterization of a second extracellular competence stimulating factor (CSF). CSF appears to be, at least in part, a small peptide of between 520 and 720 daltons. Production of CSF requires several genes that are needed both for initiation of sporulation and development of competence (spo0H, spo0A, spo0B, and spo0F). Although both peptide factors regulate competence, two different sensing pathways mediate the response to the ComX pheromone and CSF. Analysis of double mutants indicated that ComX pheromone is on the same genetic pathway as the membrane-bound histidine protein kinase encoded by comP and that CSF is on the same genetic pathway as the oligopeptide permease encoded by spo0K. Furthermore, the cellular response to partly purified ComX pheromone requires the ComP histidine protein kinase, whereas the response to partly purified CSF requires the Spo0K oligopeptide permease. These two sensing pathways converge to activate competence genes. Both factors and their convergent sensing pathways are required for normal development of competence and might function to integrate different physiological signals.

Characterization of single strand origins of cryptic rolling-circle plasmids from Bacillus subtilis

In this paper we describe the isolation and characterization of single strand origins (SSOs) of several cryptic Bacillus subtilis plasmids which use the rolling-circle mechanism of replication. The plasmids used in this study involved pTA1015, pTA1020, pTA1030, pTA1040, pTA1050 and pTA1060. The SSO of pTA1015 was isolated by shotgun cloning in a specially designed vector, pWM100, which has no SSO of its own. Sequence analysis revealed that the SSO of pTA1015 is almost identical to formerly described palT type SSOs. Also pTA1020 and pTA1060 were shown to contain SSOs highly homologous to palT. Using Southern hybridization with the palT of pTA1015 as a probe, the SSO of pTA1040 was cloned. Sequence analysis revealed a region of 200 bp which is 77% identical to the palT of pTA1015. The plasmids pTA1030 and pTA1050 contain an SSO which is highly homologous to the SSO of pTA1040. The majority of the SSOs of rolling-circle plasmids from B.subtilis seem to belong to two related families which we denote as palT1 (present on pTA1015, pTA1020 and pTA1060) and palT2 (present on pTA1030, pTA1040 and pTA1050). Both families of SSOs are highly efficient single-strand-conversion signals in B.subtilis.

ComC is required for the processing and translocation of comGC, a pilin-like competence protein of Bacillus subtilis

ComGC is a cell surface-localized protein required for DNA binding during transformation in Bacillus subtilis. It resembles type IV prepilins in its N-terminal domain, particularly in the amino acid sequence surrounding the processing cleavage sites of these proteins. ComC is another protein required for DNA binding, which resembles the processing proteases that cleave type IV prepilins. We show here that ComGC is processed in competent cells and that this processing requires ComC. We also demonstrate that the PilD protein of Neisseria gonorrhoeae, a ComC homologue, can process ComGC in Escherichia coli, and that the ComC protein itself is the only B. subtilis protein needed to accomplish cleavage of ComGC in the latter organism. Based on NaOH-solubility studies, we have shown that in the absence of ComC, but in the presence of all other competence proteins, B. subtilis is incapable of correctly translocating ComGC to the outer face of the cell membrane. Finally, we show that ComGC can be cross-linked to yield a form with higher molecular mass, possibly a dimer, and present evidence suggesting that formation of the higher mass complex takes place in the membrane, prior to translocation. Formation of this complex does not require ComC or any of the comG products, other than ComGC itself.

Differential expression of two closely related deoxyribonuclease genes, nucA and nucB, in Bacillus subtilis

Despite the lack of involvement of the competence-specific, membrane-associated deoxyribonuclease (DNase) in competence development, the expression of the gene encoding this protein, nucA, was shown to be dependent on the competence signal transduction pathway, and in particular on ComK, the competence transcription factor, which was shown to bind to the DNA region upstream of nucA. The expression of nucB, specifying an extracellular DNase, which was cloned on the basis of its homology to nucA, was shown to be sporulation-specific and dependent on the gene products of spo0A and spoIIG, the latter constituting an operon responsible for the synthesis of the mother-cell-specific sigma factor sigma E. The observed differential expression of nucA and nucB demarcates the appearance of DNase activities which are either associated with the cytoplasmic membrane or secreted into the medium during different post-exponential growth-phase processes.

Escherichia coli contains a set of genes homologous to those involved in protein secretion, DNA uptake and the assembly of type-4 fimbriae in other bacteria

A specialised system involved in a diverse array of functions, including the biogenesis of fimbriae, protein secretion and DNA uptake, has recently been found to be widespread in the eubacteria. These systems have in common several sets of related genes, including those encoding proteins containing leader sequences homologous to that of the type-4 fimbrial subunit (prepilin), a prepilin-type leader peptidase, a cytoplasmic nucleotide-binding protein, and other proteins located in the inner and outer membranes [Hobbs, M. and Mattick, J.S., Mol Microbiol. 10 (1993) 233-243]. Here, we show that Escherichia coli contains at least nine homologs of this system, and present complete sequence data for five of the genes involved (ppdD. hopB, hopC, hopD and pshM), as well as for an adjacent gene (nadC), which encodes quinolic acid phosphoribosyltransferase. Insertional mutagenesis of hopB and hopD failed to reveal any obvious effects on cell viability, morphogenesis of M13 phage, conjugative transfer of the F plasmid, or protein secretion.

Genetic characterization of mannose-sensitive hemagglutinin (MSHA)-negative mutants of Vibrio cholerae derived by Tn5 mutagenesis

El Tor biotype Vibrio cholerae strains express a cell-associated mannose-sensitive hemagglutinin (MSHA) which is a putative attachment factor. Several MSHA-negative mutants from V. cholerae strain JBK70 were previously generated by Tn5 mutagenesis [Finn et al., Infect. Immun. 55 (1987) 942-946]. The chromosomal DNA regions containing the Tn5 insertions were isolated from eight strains for further analysis. Nucleotide sequencing of the insertional junctions and corresponding clones containing the intact chromosomal region from the parental strain revealed the presence of several contiguous ORFs. Only two ORFs of this region had received insertions, and these showed remarkable homology to genes involved in the general secretory pathway found in several Gram- bacterial species. Proteins corresponding to the observed ORFs were visualized with the T7 promoter/RNA polymerase expression system. Marker exchange mutagenesis was used to insert kanamycin-resistance cassettes and TnphoA insertions into different locations of this region in the chromosome of wild-type V. cholerae strains. The phenotypes of these mutants showed that this DNA region is involved in MSHA production, but is not required for general extracellular protein secretion.

Identification of comS, a gene of the srfA operon that regulates the establishment of genetic competence in Bacillus subtilis

Genetic competence (the ability to internalize exogenous DNA) in Bacillus subtilis is dependent on a regulatory pathway that activates the expression of a battery of competence-specific genes. The srfA operon, encoding the subunits of surfactin synthetase, which catalyzes the nonribosomal synthesis of the peptide antibiotic surfactin, also functions in the competence regulatory pathway. The DNA encoding only one of the seven amino acid-activating domains of surfactin synthetase, the valine-activating domain (srfAB1), is necessary for competence. Deletion analysis revealed that a 569-bp fragment of srfAB1, fused to the srfA promoter, complements a srfA deletion mutation (delta srfA) with respect to competence. This fragment contains an open reading frame consisting of 46 amino acids (orf46), which is out of frame with srfAB1. A frameshift mutation in srfAB upstream of orf46 has no effect on competence but a frameshift and nonsense mutation in orf46 resulted in failure to complement the delt srfA mutation. These results indicate that orf46 encodes the srfA-associated competence regulatory factor. Computer-aided analysis of the putative orf46 product (ComS) shows similarity to the homeodomain of the POU domain class of eukaryotic transcriptional regulators.

The regulation of competence transcription factor synthesis constitutes a critical control point in the regulation of competence in Bacillus subtilis

comK, which encodes the competence transcription factor, is itself transcriptionally activated at the transition from exponential growth to stationary phase in Bacillus subtilis. MecA, a negative regulator of competence, also inhibits comK transcription when overexpressed, and a mecA null mutation results in comK overexpression. Although null mutations in mecA, as well as in another gene, mecB, are known to bypass the requirements for nearly all of the competence regulatory genes, the comK requirement is not suppressed by mecA inactivation. Various competence regulatory genes (comA, srfA, degU, abrB, sin, and spo0A) are shown to be required for the expression of comK. srfA transcription is shown to occur equally in cells destined for competence and those destined not to become competent. In contrast, comK transcription is restricted to the presumptive competent cells. These and other results are combined to describe a regulatory pathway for competence.

Translocation of DNA across bacterial membranes

DNA translocation across bacterial membranes occurs during the biological processes of infection by bacteriophages, conjugative DNA transfer of plasmids, T-DNA transfer, and genetic transformation. The mechanism of DNA translocation in these systems is not fully understood, but during the last few years extensive data about genes and gene products involved in the translocation processes have accumulated. One reason for the increasing interest in this topic is the discussion about horizontal gene transfer and transkingdom sex. Analyses of genes and gene products involved in DNA transfer suggest that DNA is transferred through a protein channel spanning the bacterial envelope. No common model exists for DNA translocation during phage infection. Perhaps various mechanisms are necessary as a result of the different morphologies of bacteriophages. The DNA translocation processes during conjugation, T-DNA transfer, and transformation are more consistent and may even be compared to the excretion of some proteins. On the basis of analogies and homologies between the proteins involved in DNA translocation and protein secretion, a common basic model for these processes is presented.

Membrane association and role in DNA uptake of the Bacillus subtilis PriA analogue ComF1

The late competence protein ComF1 is required for genetic transformation in Bacillus subtilis. Because of the sequence similarities of ComF1 to known ATP-dependent DNA helicases and translocases, we have hypothesized that this protein either unwinds bound double-stranded DNA or helps in the translocation of the transforming single-stranded DNA across the cell membrane. Two important implications of this hypothesis (the association of ComF1 with the membrane and its specific requirement for DNA uptake) have been tested in this report. Using cell fractionation techniques and Western blotting analysis, we show that ComF1 is located almost exclusively on the cell membrane and that it is membrane-targeted independently of other competence proteins. Moreover, ComF1 behaves like an integral membrane protein in extractability and detergent partition assays. We also show that this protein is required for the DNA-uptake step during transformation but not for DNA binding to the cell surface. DNA uptake is blocked in strains with null mutations or in-frame deletions in comF1 but also in strains that overproduce the ComF1 protein under competence conditions. This last observation suggests that ComF1 expression must be balanced with that of other competence proteins, with which it may interact to form a multisubunit complex for DNA uptake.

Regulation of competence-specific gene expression by Mec-mediated protein-protein interaction in Bacillus subtilis

The expression of competence genes in Bacillus subtilis is controlled by a signal transduction cascade which increases the expression of a competence transcription factor (CTF, encoded by comK) during the transition from exponential growth to stationary phase. The transcription of CTF (ComK) is decreased by the product of the mecA gene, and this inhibition is relieved in response to an unknown signal received from upstream in the regulatory pathway. Inactivation of either mecA or another gene, mecB, results in overproduction of ComK. We show here that the concentration of MecA protein does not vary markedly with culture medium, as a function of growth stage, or in competent and noncompetent cells. We also show that MecA can interact directly with ComK. Finally, evidence is presented suggesting that MecB functions prior to MecA in the signaling pathway. A model is discussed which involves the sequestration of ComK by MecA binding and the release of the transcription factor when an appropriate signal is relayed to MecA by MecB.

Identification, localization, and distribution of the PilT protein in Neisseria gonorrhoeae

A monoclonal antibody (MAb) directed against a highly conserved protein of Neisseria gonorrhoeae with a molecular size of 40 kDa was isolated and characterized. The protein antigen detected by this MAb was detected by enzyme-linked immunosorbent assay and immunoblotting in all strains of N. gonorrhoeae tested across a wide range of serovars. The 40-kDa protein was found to be expressed at relatively low levels and localized to both the cytosolic and cytoplasmic membrane fractions. Screening of a lambda gt11 expression library derived from gonococcal genomic DNA with the anti-40-kDa MAb and DNA sequence analysis suggested that the 40-kDa protein and the product of the gonococcal pilT gene were identical. Immunoblotting analysis of gonococcal mutants carrying defined mutations in the pilT gene confirmed that the 40-kDa protein was indeed PilT. The N-terminal sequence derived by microsequencing of the protein purified from gonococci led to the correction of the previously published pilT gene sequence. Sequencing of the pilT gene from three different strains revealed an extremely high degree of conservation at both the amino acid and DNA levels.

beta-Lactamase topology probe analysis of the OutO NMePhe peptidase, and six other Out protein components of the Erwinia carotovora general secretion pathway apparatus

The out gene cluster of Erwinia spp. encodes the proteins of the general secretory pathway (GSP) apparatus that is required for pectinase and cellulase secretion. We have used fusions between Erwinia carotovora subsp. carotovora (Ecc) out genes and the topology probe blaM to assess the ability of Out protein regions to export BlaM across the cytoplasmic membrane in Escherichia coli and Ecc. For the outO gene product (an NMePhe peptidase), seven transmembrane regions have been identified and one more is predicted. The region of OutO with the highest level of hydrophilicity is likely to exist as a large cytoplasmic loop, located between two hydrophobic domains, and is positioned towards the N-terminus of the protein. When BlaM was fused on the C-terminal side of the last hydrophobic stretch of OutO, the resulting hybrid protein transferred the BlaM moiety to the periplasm whilst retaining OutO activity. Removal of a portion of this hydrophobic stretch resulted in the loss of OutO activity, suggesting that there are tight constraints on the topological integrity of OutO for maintaining catalytic function. When outG, -H, -I, -J, -K and -N were fused to blaM, the resulting phenotype suggested that the majority of each protein was targeted to the periplasm. Our results indicate that these six Out proteins, when produced by E. coli or Ecc, each adopt, at least temporarily, a type II bitopic conformation in the cytoplasmic membrane. For OutG, -H, -I and -J this probably represents the membrane topology prior to processing by OutO in Ecc. When produced in vivo from a T7 gene 10 promoter construct, the outG product was processed in Ecc whereas the outO mutant RJP249 failed to process pre-OutG. BlaM fusions positioned on the C-terminal side of the hydrophobic stretches of pre-OutG, -H, -I, and -J were processed by wild-type Ecc but not RJP249 or E. coli DH1. Thus the periplasmic domains of these proteins play no role in the peptidase cleavage reaction. An OutG-BlaM fusion construct was used to demonstrate NMePhe peptidase activity in other bacterial strains including E. carotovora subsp. carotovora (ATCC39048), E. carotovora subsp. atroseptica (SCRI1043) and Erwinia chrysanthemi (3937).

Biochemical and genetic characterization of a competence pheromone from B. subtilis

We have purified and characterized a modified peptide pheromone that accumulates in culture medium as B. subtilis grows to high density. This pheromone is required for the development of genetic competence. When added to cells at low density, the pheromone induces the premature development of competence. The peptide moiety of the pheromone matches nine of the last ten amino acids predicted from a 55 codon open reading frame, comX. comX and comQ, the gene immediately upstream of comX, are required for production of the pheromone. Response to the pheromone requires the comP-comA two-component regulatory system and the oligopeptide permease encoded by spo0K. Spo0K could transport the pheromone into the cell, or function as a receptor, binding the pheromone and sending a transmembrane signal, leading to activation of the ComA transcription factor and induction of competence development.

Molecular characterization of PulE, a protein required for pullulanase secretion

pulE, one of 14 genes specifically required for pullulanase secretion in Klebsiella oxytoca, codes for a putative nucleotide-binding protein. Subcellular fractionation indicated that the majority of PulE in Escherichia coli cells expressing all 14 secretion genes is mainly associated with the cytoplasmic membrane through both hydrophobic and non-hydrophobic interactions. Mutational analysis revealed that one of the two regions of PulE that are conserved in many nucleotide-binding proteins (Walker box A) is essential for pullulanase secretion. Likewise, mutations that removed aspartate residues from each of two regions immediately downstream from the Walker box A also reduced secretion. These aspartate-rich regions are highly conserved in all 16 known PulE homologues but not in any other nucleotide-binding proteins. Altogether, these results indicate that PulE might belong to a new family of nucleotide-binding proteins. The protein could not be cross-linked to the photoactivatable ATP analogue azido-ATP, however. Most pulE point or deletion mutations which prevented pullulanase secretion exhibited transdominance when expressed at high levels in cells producing wild-type PulE protein. Evidence presented suggests that PulE might be a homodimer.

Bioenergetic aspects of the translocation of macromolecules across bacterial membranes

Bacteria are extremely versatile in the sense that they have gained the ability to transport all three major classes of biopolymers through their cell envelope: proteins, nucleic acids, and polysaccharides. These macromolecules are translocated across membranes in a large number of cellular processes by specific translocation systems. Members of the ABC (ATP binding cassette) superfamily of transport ATPases are involved in the translocation of all three classes of macromolecules, in addition to unique transport ATPases. An intriguing aspect of these transport processes is that the barrier function of the membrane is preserved despite the fact the dimensions of the translocated molecules by far surpasses the thickness of the membrane. This raises questions like: How are these polar compounds translocated across the hydrophobic interior of the membrane, through a proteinaceous pore or through the lipid phase; what drives these macromolecules across the membrane; which energy sources are used and how is unidirectionality achieved? It is generally believed that macromolecules are translocated in a more or less extended, most likely linear form. A recurring theme in the bioenergetics of these translocation reactions in bacteria is the joint involvement of free energy input in the form of ATP hydrolysis and via proton sym- or antiport, driven by a proton gradient. Important similarities in the bioenergetic mechanisms of the translocation of these biopolymers therefore may exist.

Posttranslational processing of type IV prepilin and homologs by PilD of Pseudomonas aeruginosa

We have described the characterization of a protein initially identified as having an essential function in biogenesis of polar pili of P. aeruginosa by processing precursors of pilin. Other findings have also expanded the range of substrates for PilD to include a set of proteins that are essential components of the extracellular secretion machinery. Direct demonstration of prepilin processing necessitates use of purified substrates and enzymes, and we present general protocols for purification of both enzymes and substrates, as well as an assay for prepilin peptidase activity. For a source of enzyme and substrates, mutants of P. aeruginosa defective in pilin processing as well as clones overexpressing the pilin gene and PilD were developed. These methods are applicable to other bacterial systems that express Type IV pili and/or possess the PilD-dependent machinery of extracellular protein secretion. PilD is a bifunctional enzyme, which carries out not only cleavage but also amino-terminal methylation of the mature pilin. Cleavage and N-methylation of the pilin-like Xcp proteins involved in extracellular protein secretion have also been shown to be dependent on PilD. The leader peptidase activity of PilD is inhibited by sulfhydryl blocking reagents such as NEM and PCMB, whereas the methyltransferase activity of the purified enzyme is dependent on reduction with dithiothreitol. The conserved region containing the cysteine residues lies within the largest hydrophilic domain of the protein as predicted from hydrophobicity analysis, and it is probably exposed to the cytoplasmic side of the cytoplasmic membrane. Identification of the active site residues involved in recognition of the substrates for processing and subsequent methylation is currently underway. Studies on substrate specificities of PilD, with respect to its leader peptidase and methyltransferase activity, may prove to be useful in designing inhibitors which would interfere with maturation of Type IV prepilins and components of the extracellular protein secretion machinery. In light of the fact that an increasing number of both mammalian and plant pathogens are being shown to have extracellular secretion pathways homologous to that seen for P. aeruginosa, such inhibitors may be useful tools in the study of the role these peptidases play in bacterial virulence.

Cloning, sequence analysis, and hyperexpression of the genes encoding phosphotransacetylase and acetate kinase from Methanosarcina thermophila

The genes for the acetate-activating enzymes, acetate kinase and phosphotransacetylase (ack and pta), from Methanosarcina thermophila TM-1 were cloned and sequenced. Both genes are present in only one copy per genome, with the pta gene adjacent to and upstream of the ack gene. Consensus archaeal promoter sequences are found upstream of the pta coding region. The pta and ack genes encode predicted polypeptides with molecular masses of 35,198 and 44,482 Da, respectively. A hydropathy plot of the deduced phosphotransacetylase sequence indicates that it is a hydrophobic polypeptides; however, no membrane-spanning domains are evident. Comparison of the amino acid sequences deduced from the M. thermophila and Escherichia coli ack genes indicate similar subunit molecular weights and 44% identity (60% similarity). The comparison also revealed the presence of several conserved arginine, cysteine, and glutamic acid residues. Arginine, cysteine, and glutamic acid residues have previously been implicated at or near the active site of the E. coli acetate kinase. The pta and ack genes were hyperexpressed in E. coli, and the overproduced enzymes were purified to homogeneity with specific activities higher than those of the enzymes previously purified from M. thermophila. The overproduced phosphotransacetylase and acetate kinase migrated at molecular masses of 37,000 and 42,000 Da, respectively. The activity of the acetate kinase is optimal at 65 degrees C and is protected from thermal inactivation by ATP. Diethylpyrocarbonate and phenylglyoxal inhibited acetate kinase activity in a manner consistent with the presence of histidine and arginine residues at or near the active site; however, the thiol-directed reagents 5,5'-dithiobis (2-nitrobenzoic acid) and N-ethylmaleimide were ineffective.

Genetic and biochemical analysis of an endonuclease encoded by the IncN plasmid pKM101

The IncN plasmid pKM101 nuc gene encodes a periplasmically localized endonuclease. DNA sequence analysis indicates that this gene encodes a hydrophilic protein of about 19.5 kDa containing a hydrophobic signal sequence. nuc is homologous to a partially sequenced open reading frame adjacent to the sog gene of the plasmid CollB-P9, a plasmid known to encode an endonuclease similar to that of pKM101. A partially sequenced tra gene directly upstream of nuc is homologous to the virB11 gene of Agrobacterium tumefaciens. We have partially purified the pKM101 nuclease by osmotic shock and cation exchange chromatography, and used this enzyme preparation to sequence the protein's amino terminus. The first 13 amino acids of the mature protein match amino acids 23 to 35 of the predicted sequence, indicating that the protein is proteolytically processed to a molecular mass of approximately 17 kDa, probably during export to the periplasmic space. The enzyme was able to attack many sites along an end labelled duplex DNA substrate, but showed clearly preferred cleavage sites, and may cleave preferentially at purine-rich regions.

Characterization of comE, a late competence operon of Bacillus subtilis required for the binding and uptake of transforming DNA

The binding and transport of DNA by competent Bacillus subtilis requires the assembly of a specialized apparatus. We present here the characterization of comE, an operon under competence control that is required for both DNA binding to the competent cell surface, and for uptake. comE contains three open reading frames (ORF1-3) read in the forward direction, preceded by a long untranslated leader sequence and an apparent E sigma A promoter. A minor promoter also is responsible for transcription of ORF2 and -3. A transcript containing a single ORF is produced in the reverse direction. The reverse ORF overlaps ORF1 and the untranslated comE leader. The comE transcript is present at a very low level during growth and at an elevated level in stationary-phase cells. Conversely, the reverse transcript is present during exponential growth and disappears during stationary phase. The reverse ORF resembles prokaryotic and eukaryotic pyrroline-5'-carboxylate reductases, while ORF2 is similar to several dCMP deaminases. ORF1 and ORF3 are predicted to be integral membrane proteins. The latter is specifically required for DNA uptake but not for binding.

Elucidation of regulatory elements that control damage induction and competence induction of the Bacillus subtilis SOS system

A novel consensus sequence (GAAC-N4-GTTC) has been identified within the promoter regions of DNA damage-inducible (din) genes from Bacillus subtilis. This sequence has been proposed to function as an operator site that is required for regulation of the SOS system of B. subtilis. To test this hypothesis, a deletion analysis of the dinA and recA promoter regions was utilized. A single consensus sequence is sufficient and necessary for damage-inducible regulation of the dinA and recA promoters. Deletion of the consensus sequences upstream of these promoters derepressed their expression under uninduced conditions. In addition, this deletion analysis has further defined sequences upstream of the recA promoter that are required for expression of the recA gene in cells that have differentiated to the state of natural competence. Northern (RNA) hybridization and S1 nuclease protection experiments have demonstrated that the damage-inducible and competence-inducible recA-specific transcripts initiate from a single promoter. Mutations within the comA, srfA, and degU loci each completely abolish the competence-inducible expression of the recA gene.

Characterization of pilQ, a new gene required for the biogenesis of type 4 fimbriae in Pseudomonas aeruginosa

Type 4 fimbriae are produced by a variety of pathogens, in which they appear to function in adhesion to epithelial cells, and in a form of surface translocation called twitching motility. Using transposon mutagenesis of Pseudomonas aeruginosa, we have identified a new locus required for fimbrial assembly. This locus contains the gene pilQ which encodes a 77 kDa protein with an N-terminal hydrophobic signal sequence characteristic of secretory proteins. pilQ mutants lack the spreading colony morphology characteristic of twitching motility, are devoid of fimbriae, and are resistant to the fimbrial-specific bacteriophage PO4. The pilQ gene was mapped to Spel fragment 2, which is located at 0-5 minutes on the P. aeruginosa PAO1 chromosome, and thus it is not closely linked to the previously characterized pilA-D, pilS,R or pilT genes. The pilQ region also contains ponA, aroK and aroB-like genes in an organization very similar to that of corresponding genes in Escherichia coli and Haemophilus influenzae. The predicted amino acid sequence of PilQ shows homology to the PulD protein of Klebsiella oxytoca and related outer membrane proteins which have been found in association with diverse functions in other species including protein secretion, DNA uptake and assembly of filamentous phage. PilQ had the highest overall homology to an outer membrane antigen from Neisseria gonorrhoeae, encoded by omc, that may fulfil the same role in type 4 fimbrial assembly in this species.

Subcellular localization of seven VirB proteins of Agrobacterium tumefaciens: implications for the formation of a T-DNA transport structure

Plant cell transformation by Agrobacterium tumefaciens involves the transfer of a single-stranded DNA-protein complex (T-complex) from the bacterium to the plant cell. One of the least understood and important aspects of this process is how the T-complex exits the bacterium. The eleven virB gene products have been proposed to specify the DNA export channel on the basis of their predicted hydrophobicity. To determine the cellular localization of the VirB proteins, two different cell fractionation methods were employed to separate inner and outer membranes. Seven VirB-specific antibodies were used on Western blots (immunoblots) to detect the proteins in the inner and outer membranes and soluble (containing cytoplasm and periplasm) fractions. VirB5 was in both the inner membrane and cytoplasm. Six of the VirB proteins were detected in the membrane fractions only. Three of these, VirB8, VirB9, and VirB10, were present in both inner and outer membrane fractions regardless of the fractionation method used. Three additional VirB proteins, VirB1, VirB4, and VirB11, were found mainly in the inner membrane fraction by one method and were found in both inner and outer membrane fractions by a second method. These results confirm the membrane localization of seven VirB proteins and strengthen the hypothesis that VirB proteins are involved in the formation of a T-DNA export channel or gate. That most of the VirB proteins analyzed are found in both inner and outer membrane fractions suggest that they form a complex pore structure that spans both membranes, and their relative amounts in the two membrane fractions reflect their differential sensitivity to the experimental conditions.

Sequence and properties of mecA, a negative regulator of genetic competence in Bacillus subtilis

The development of competence in Bacillus subtilis is regulated by growth conditions and several regulatory genes. In complex media competence development is poor, and there is little or no expression of late competence genes. mec mutations permit competence development and late competence gene expression in complex media, and bypass the requirements for many of the competence regulatory genes. In this paper we describe the cloning and characterization of mecA. The mecA gene product acts negatively in the development of competence. Null mutations in mecA allowed expression of a late competence gene comG, under conditions where it is not normally expressed, including in complex media and in cells mutant for several competence regulatory genes. Overexpression of MecA from a multicopy plasmid resulted in inhibition of comG transcription. The DNA sequence of mecA was determined and the predicted gene product showed no significant similarity to any protein in the database. Expression of a mecA-lacZ translational fusion was constitutive during growth and did not vary significantly in the different media tested. The role of mecA in competence development and other stationary phase phenomena is discussed.

Cleavage, methylation, and localization of the Pseudomonas aeruginosa export proteins XcpT, -U, -V, and -W

Four components of the apparatus of extracellular protein secretion of Pseudomonas aeruginosa, Xcpt, -U, -V, and -W (XcpT-W), are synthesized as precursors with short N-terminal leader peptides that share sequence similarity with the pilin subunit of this organism. A specialized leader peptidase/methylase, product of the pilD gene, has been shown to cleave the leader peptide from prepilin and to methylate the N-terminal phenylalanine of the mature pilin. Antibodies were prepared against XcpT-W and used to purify each of these proteins. Sequence analysis of XcpT-W has shown that these proteins, like mature pilin, contain N-methylphenylalanine as the N-terminal amino acid. Analysis of cellular fractions from wild-type and pilD mutant strains of P. aeruginosa showed that the precursor forms of XcpT-W are located predominantly in the bacterial inner membrane, and their localization is not altered after PilD-mediated removal of the leader sequence. These studies demonstrate that the biogenesis of the apparatus of extracellular protein secretion and that of type IV pili share a requirement for PilD. This bifunctional enzyme, acting in the inner membrane, cleaves the leader peptides from precursors of pilins and XcpT-W and subsequently methylates the amino group of the N-terminal phenylalanine of each of its substrates.

Amino-acylation site mutations in amino acid-activating domains of surfactin synthetase: effects on surfactin production and competence development in Bacillus subtilis

The part of the srfA operon of Bacillus subtilis that contains the region required for competence development is composed of the first four amino acid-activating domains which are responsible for the incorporation of Glu, Leu, D-Leu, and Val into the peptide moiety of the lipopeptide surfactin. Ser-to-Ala substitutions were made in the amino-acylation site of each domain, and their effects on surfactin production and competence development were examined. All of the mutations conferred a surfactin-negative phenotype, supporting the finding that the conserved Ser in the amino-acylation site is required for peptide synthesis. However, none of the mutations affected significantly competence development or the expression of a lacZ fusion to the late competence operon comG. This, coupled with recent findings that only the fourth, Val-activating, domain is required for competence, suggests that some activity, other than amino-acylation and perhaps unrelated to peptide synthesis, possessed by the fourth domain is involved in the role of srfA in regulating competence development.

Molecular cloning and characterization of 13 out genes from Erwinia carotovora subspecies carotovora: genes encoding members of a general secretion pathway (GSP) widespread in gram-negative bacteria

The chemical mutagen ethylmethanesulphonate (EMS) has been used to generate mutants of Erwinia carotovora subspecies carotovora which are defective in the secretion of pectinases (Pel) and cellulases (Cel) but unaltered for protease (Prt) secretion. Such mutants, called Out-, still synthesize Pel and Cel but these enzymes accumulate within the periplasm. Cosmid clones carrying wild-type E. carotovora ssp. carotovora DNA, identified by their ability to restore the Out+ phenotype when transferred to some Out- mutants, were classified into six complementation groups using cosmids and cosmid derivatives. Analysis of the nucleotide sequence of a 12.7 kb DNA fragment, encompassing complementing cosmid inserts, revealed a coding capacity for 13 potential open reading frames (ORFs), and these were designated outC-outO. Some of the out gene products were visualized using a T7 gene 10 expression system. The predicted Out proteins are highly similar to components of extracellular enzyme secretion systems from a diverse range of eubacteria including Erwinia chrysanthemi, Klebsiella oxytoca, Aeromonas hydrophila, Pseudomonas aeruginosa and Xanthomonas campestris. Lower levels of similarity exist between Ecc Out proteins and components of macromolecular trafficking systems from Bacillus subtilis, Haemophilus influenzae, Agrobacterium tumefaciens, Yersinia pestis and a protein involved in the morphogenesis of filamentous bacteriophages such as M13.

Genetic organization and sequence of the promoter-distal region of the tcp gene cluster of Vibrio cholerae

The nucleotide sequence of the promoter distal region of the major operon associated with biosynthesis of TCP, the toxin coregulated pilus of Vibrio cholerae has been determined. The genes tcpR, tcpD, tcpS, tcpT, tcpE and tcpF are organized to permit translational coupling and are followed by an inverted repeat structure which is likely to act as a strong Rho-independent terminator. TcpS and TcpF possess N-terminal signal sequences and would be expected to be periplasm and outer membrane located, respectively. TcpT and, to a lesser extent, TcpE show homology to protein transport and secretion proteins identified in a number of other bacteria. TcpR and TcpD are also predicted to be localized in the membrane based on their hydrophobicity profiles. In the case of TcpD, this may be in the outer membrane, as the N terminus is reminiscent of the TcpA signal sequence processed by TcpJ. Taken together with other data on the tcp region, it is clear that the types of proteins involved in the biogenesis of TCP are, at least in some cases, related to those associated with type-4 pilus biosynthesis but differ markedly from those for other well-described fimbrial systems, such as Type I, K88, K99 or Pap pili. The complete sequence data for TCP biosynthesis-controlling genes are now available and represent the only complete example of a type-4 pilus-like system.

Biogenesis and regulation of the Vibrio cholerae toxin-coregulated pilus: analogies to other virulence factor secretory systems

Biogenesis of the toxin-coregulated pilus (TCP) of Vibrio cholerae 01 is essential for successful bacterial colonization of the small intestine. Pilus assembly requires the products of at least seven genes located on the chromosome adjacent to the pilin-encoding gene, tcpA. Previously reported TnphoA insertions in the TCP-assembly-deficient V. cholerae strains, KP2.21 and KP4.2, were isolated from the chromosome for further analysis. Nucleotide sequencing of the tcpE::phoA and tcpF::phoA fusions and corresponding clones of the region containing the intact genes revealed the presence of two open reading frames (ORFs) of 340 and 338 amino acids, designated TcpE and TcpF, respectively. The partial sequence of an ORF downstream from the TcpF coding sequence was determined to correspond to the global virulence regulator, ToxT. Proteins corresponding to the observed ORFs were visualized with the T7 promoter/RNA polymerase expression system. Computer-generated alignment algorithms predict that a homology exists between TcpE and the Klebsiella pneumoniae pullulanase secretion proteins PulD and PulF, the Xanthomonas campestris extracellular enzyme secretion factor XpsF, the Bacillus subtilis DNA competence protein ComG-ORF2, and the Yersinia enterocolitica Yop secretion determinant YscC. These observations provide a model to investigate further the relationship between the secretion mechanisms utilized by these seemingly diverse virulence determinants. Additionally, an extreme C-terminal segment of TcpE shows striking homology to the transmembrane segment of the eukaryotic integrin beta-1 chain, which could imply a role for TcpE in not only TCP secretion, but also host cell interaction.

Conservation of genes encoding components of a type IV pilus assembly/two-step protein export pathway in Neisseria gonorrhoeae

Three gonococcal genes have been identified which encode proteins with substantial similarities to known components of the type IV pilus biogenesis pathway in Pseudomonas aeruginosa. Two of the genes were identified based on their hybridization with a DNA probe derived from the pilB gene of P. aeruginosa under conditions of reduced stringency. The product of the gonococcal pilF gene is most closely related to the pilus assembly protein PilB of P. aeruginosa while the product of the gonococcal pilT gene is most similar to the PilT protein of P. aeruginosa which is involved in pilus-associated twitching motility and colony morphology. The products of both of these genes display canonical nucleoside triphosphate binding sites and are predicted to be to cytoplasmically localized based on their overall hydrophilicity. The gonococcal pilD gene, identified by virtue of its linkage to the pilF gene, is homologous to a family of prepilin leader peptidase genes. When expressed in Escherichia coli, the gonococcal PilD protein functions to process gonococcal prepilin in a manner consistent with its being gonococcal prepilin peptidase. These results suggest that Neisseria gonorrhoeae is capable of expressing many of the essential elements of a highly conserved protein translocation system and that these gene products are probably involved in pilus biogenesis.

A single bifunctional enzyme, PilD, catalyzes cleavage and N-methylation of proteins belonging to the type IV pilin family

Precursors of the type IV pilins of a number of bacterial pathogens, as well as related proteins involved in extracellular protein export and DNA uptake, are synthesized with short basic leader sequences. Maturation of these proteins involves two consecutive posttranslational modifications. The leader sequence is first proteolytically removed by specialized endopeptidases, of which the prototype is encoded by the pilD gene of Pseudomonas aeruginosa. Subsequently, the amino termini of these proteins are methylated. Here we demonstrate that PilD, in addition to cleaving the amino-terminal leader sequences of prepilin, also catalyzes N-methylation of the amino-terminal phenylalanine of the mature pilin, using S-adenosyl-L-methionine as a methyl donor. Thus, to our knowledge, PilD is the first characterized bacterial N-methyltransferase. Complete inhibition of N-methylation, but not peptide cleavage, by structural analogues of S-adenosyl-L-methionine suggests that PilD is a bifunctional enzyme with proteolytic and methylation activities carried out within two distinct active sites.

The Agrobacterium tumefaciens virB4 gene product is an essential virulence protein requiring an intact nucleoside triphosphate-binding domain

Products of the approximately 9.5-kb virB operon are proposed to direct the export of T-DNA/protein complexes across the Agrobacterium tumefaciens envelope en route to plant cells. The presence of conserved nucleoside triphosphate (NTP)-binding domains in VirB4 and VirB11 suggests that one or both proteins couple energy, via NTP hydrolysis, to T-complex transport. To assess the importance of VirB4 for virulence, a nonpolar virB4 null mutation was introduced into the pTiA6NC plasmid of strain A348. The 2.37-kb virB4 coding sequence was deleted precisely by oligonucleotide-directed mutagenesis in vitro. The resulting delta virB4 mutation was exchanged for the wild-type allele by two sequential recombination events with the counterselectable Bacillus subtilis sacB gene. Two derivatives, A348 delta B4.4 and A348 delta B4.5, sustained a nonpolar deletion of the wild-type virB4 allele, as judged by Southern blot hybridization and immunoblot analyses with antibodies specific for VirB4, VirB5, VirB10, and VirB11. Transcription of wild-type virB4 from the lac promoter restored virulence to the nonpolar null mutants on a variety of dicotyledonous species, establishing virB4 as an essential virulence gene. A substitution of glutamine for Lys-439 and a deletion of Gly-438, Lys-439, and Thr-440 within the glycine-rich NTP-binding domain (Gly-Pro-Iso-Gly-Arg-Gly-Lys-Thr) abolished complementation of A348 delta B4.4 or A348 delta B4.5, demonstrating that an intact NTP-binding domain is critical for VirB4 function. Merodiploids expressing both the mutant and wild-type virB4 alleles exhibited lower virulence than A348, suggesting that VirB4, a cytoplasmic membrane protein, may contribute as a homo- or heteromultimer to A. tumefaciens virulence.