Functional analysis of ori1 and repA of the R-plasmid pSJ5.6 from Neisseria gonorrhoeae

The functional ori1 of the 5.6kb gonococcal R-plasmid pSJ5.6 contains an A-T rich region followed by four 22bp direct repeats and one 19bp inverted repeat. The replication region of the plasmid also contains a gene encoding for a 39kD RepA protein. We have further assessed the functionality of the replication region in pSJ5.6, an-iteron type plasmid, using in vivo complementation assays in Escherichia coli. A 2.1kb PstI-RsaI fragment containing the ori1 and repA gene of pSJ5.6 was cloned into vector pZErO -2 to obtain pZA-MRR. The pUC origin in pZA-MRR was deleted to render the plasmid dependable on the cis-acting ori1 for replication. The resulting plasmid, pMRR, was capable of replication and maintenance in E. coli. We also cloned the ori1 and repA gene separately to obtain pA-Ori and pZG-Rep, respectively. Using in vivo complementation assays, we demonstrated that the ori1(+) plasmid (pA-Ori) was maintained only when the RepA protein was supplied in trans by the high copy number plasmid pZG-Rep.     

Molecular characterization of a small Haemophilus influenzae plasmid specifying beta-lactamase and its relationship to R factors from Neisseria gonorrhoeae.

The ampicillin-resistant Haemophilus influenzae strain Ve445 which caused purulent meningitis and septicaemia in a newborn child in Germany contained a 4.4 megadalton (Mdal) plasmid (pVe445) and produced a TEM type beta-lactamase. The transformation to ampicillin resistance of a sensitive Escherichia coli strain with isolated pVe445 DNA proved that the structural gene for the beta-lactamase resided on this plasmid genome. Molecular DNA-DNA hybridization studies and electron microscope DNA heteroduplex analysis indicated that pVe445 probably contained 38 to 41% of the ampicillin translocation DNA segment (TnA) found on R factors of enteric origin. The TnA fragment present in pVe445 most likely does not contain both of the inverted repeat sequences of TnA. DNA-DNA polynucleotide sequence studies indicated that the 4.4 Mdal plasmid pVe445 was unrelated to the 30 to 38 Mdal H. influenzae R plasmids but was closely related to the 4.1 Mdal ampicillin resistance specifying H. influenzae plasmid RSF0885 isolated in the U.S.A. The H. influenzae plasmid pVe445 shared 91% of its base sequences with the beta-lactamase specifying Neisseria gonorrhoeae plasmid pMR0360 (4.4 Mdal) and had 85% of its base sequences in common with the beta-lactamase specifying N. gonorrhoeae plasmid pMR0200 (3.2 Mdal). All of the four 3.2 to 4.4 Mdal beta-lactamase specifying R plasmids of H. influenzae and N. gonorrhoeae investigated probably have a common evolutionary origin.     

Mobilization of the gonococcal 5.2 kb beta-lactamase plasmid pSJ5.2 into Escherichia coli by cointegration with several gram-conjugative plasmids

We report the mobilization by cointegration of the gonococcal 5.2 kb beta-lactamase plasmid pSJ5.2 in an Escherichia coli background. Transfer of pSJ5.2 was measured by filter mating assays with five different conjugative plasmids from Enterobacteriaceae and the gonococcal 41 kb tet(M). Plasmid pSJ5.2 was mobilized to E. coli at frequencies of 1.7x10(-6), 9.3x10(-8) and 2.7x10(-5) by the tet(M), R64 drd-33 and N3 conjugative plasmids, respectively. Mobilization of pSJ5.2 by the 41 kb tet(M) conjugative plasmid resulted in stable Amp(R) E. coli transconjugants consisting of pSJ5.2 plasmid with an insertion located in the 2.4 kb BamHI-BamHI fragment. Mobilization of pSJ5.2 by R64drd-33 and N3 conjugative plasmids involved stable cointegrates as detected by Southern Blot with a DIG-labelled PstI-digested pSJ5.2 probe. Restriction analysis of the R64::pSJ5.2 and N3::pSJ5.2 cointegrates and Southern Blot with the pSJ5.2 probe showed that cointegrates formed by deletion of DNA regions within the 1.8 kb BamHI-HindIII fragment of pSJ5.2. The plasmid thus appears to use multiple recombination mechanisms for cointegration with different conjugative plasmids. The complete nucleotide sequence of pSJ5.2 was determined, and will be a useful tool to further investigate the molecular mechanisms leading to its cointegrative transfer.     

Four different integrative recombination events involved in the mobilization of the gonococcal 5.2 kb beta-lactamase plasmid pSJ5.2 in Escherichia coli

We identified and characterized four different recombination mechanisms involved in the cointegrative transfer of the Neisseria gonorrhoeae beta-lactamase plasmid pSJ5.2 by the gonococcal 41 kb tet(M) and the Gram negative self-transmissible plasmids N3 and R64 drd-33 using an Escherichia colirecA-background. Mobilization of pSJ5.2 by the tet(M) plasmid occurred by cointegration through a replicative transposition of two IS1 elements inserted upstream from the beta-lactamase gene of pSJ5.2 and creating a IS1::beta-lactamase hybrid promoter. Two types of recombinational events occurred within the 1.8 kb BamH1-HindIII fragment of pSJ5.2 with the N3 and R64 plasmids. A non-homologous recombination was found at coordinates 1817 and 2849 of pSJ5.2 with sequences from R64. A non-homologous recombination combined with an IS26-mediated one-ended transposition was found at coordinates 1817 and 3010 of pSJ5.2 with N3. In both recombinational events, a deletion of over 1 kb of pSJ5.2 occurred. The fourth recombination event was detected in the 1.0 kb BamH1-HindIII fragment of pSJ5.2 by homologous recombination between DNA from the truncated Tn3 resolvase gene of pSJ5.2 and the resolvase sequences from R64 and N3.     

Penicillinase-producing Neisseria gonorrhoeae: a molecular comparison of 5.3-kb and 7.4-kb beta-lactamase plasmids.

Restriction endonuclease analysis and heteroduplex studies indicate that the only difference between the 5.3-kilobase (kb) and 7.4-kb plasmids from beta-lactamase-producing Neisseria gonorrhoeae is that the latter is the 5.3-kb plasmid with a 2.1-kb insertion. The insertion is bounded by inverted repeats of approximately 300 base pairs. Several plasmids from beta-lactamase-producing N. gonorrhoeae isolated at different times and in different countries were compared. Nine 5.3-kb plasmids were examined and found to be indistinguishable, as were sixteen 7.4-kb plasmids.     

Identification and characterization of a new replication region in the Neisseria gonorrhoeae beta-lactamase plasmid pFA3.

The 7.1-kilobase-pair (kbp) plasmid pFA3 specifies TEM beta-lactamase production in Neisseria gonorrhoeae. We studied the minimal region required for replication of this plasmid in Escherichia coli by constructing a set of nested deletions of the 3.4-kbp PstI-HindIII fragment. The smallest fragment capable of maintenance in E. coli when ligated to a streptomycin-spectinomycin resistance cassette was 2.0 kbp in size and was different from another autonomously replicating fragment of pFA3 reported by K. H. Yeung and J. Dillon (Plasmid 20:232-240, 1988). The fragment contained single BamHI and XbaI sites and specified a 39-K protein. Fragments subcloned from the minimal region or constructed by deletion from the 3' or 5' ends were not capable of autonomous replication. Mutants constructed by end filling and religating DNA cleaved at the BamHI or XbaI sites were not capable of autonomous replication and no longer produced the 39K protein. These results suggest that replication is dependent on the 39K protein. DNA sequence analysis of the region showed an A-T-rich region followed by four 22-bp direct repeats followed by an open reading frame encoding a 39K basic protein.     

Mutational and physical analysis of F plasmid traY protein binding to oriT

F plasmid traY protein binding to wild-type or deleted regions containing the TraY-binding site, sbyA, was studied in vitro. The principal DNA-protein complex was formed with DNA segments including the sbyA site defined by footprinting and (with lesser affinity) with truncated segments that retained the leftward two-thirds of sbyA. This located the major sequence determinants for TraY binding between bp 204 and 227 on the oriT map. For all sequences tested, bound TraY induced bending of approximateiy 50 to 55°, and centred between bp 214 and 221. Thermodynamic and mobility analyses indicated that two TraY protomers bind to sbyA. At higher TraY concentrations, additional TraY bound to the left of the sbyA in a region previously shown to bind IHF (site IHF A). TraY binding to this additional site (sbyC) was inhibited by IHF. Sequence similarities shared by sbyA, sbyB, and SbyC may include the critical base pairs for TraY binding.     

Purification and characterization of the RecA protein from Neisseria gonorrhoeae

The strict human pathogen Neisseria gonorrhoeae is the only causative agent of the sexually transmitted infection gonorrhea. The recA gene from N. gonorrhoeae is essential for DNA repair, natural DNA transformation, and pilin antigenic variation, all processes that are important for the pathogenesis and persistence of N. gonorrhoeae in the human population. To understand the biochemical features of N. gonorrhoeae RecA (RecA(Ng)), we overexpressed and purified the RecA(Ng) and SSB(Ng) proteins and compared their activities to those of the well-characterized E. coli RecA and SSB proteins in vitro. We observed that RecA(Ng) promoted more strand exchange at early time points than RecA(Ec) through DNA homologous substrates, and exhibited the highest ATPase activity of any RecA protein characterized to date. Further analysis of this robust ATPase activity revealed that RecA(Ng) is more efficient at displacing SSB from ssDNA and that RecA(Ng) shows higher ATPase activity during strand exchange than RecA(Ec). Using substrates created to mimic the cellular processes of DNA transformation and pilin antigenic variation we observed that RecA(Ec) catalyzed more strand exchange through a 100 bp heterologous insert, but that RecA(Ng) catalyzed more strand exchange through regions of microheterology. Together, these data suggest that the processes of ATP hydrolysis and DNA strand exchange may be coupled differently in RecA(Ng) than in RecA(Ec). This difference may explain the unusually high ATPase activity observed for RecA(Ng) with the strand exchange activity between RecA(Ng) and RecA(Ec) being more similar.     

Deletion analysis of the F plasmid oriT locus.

Functional domains of the Escherichia coli F plasmid oriT locus were identified by deletion analysis. DNA sequences required for nicking or transfer were revealed by cloning deleted segments of oriT into otherwise nonmobilizable pUC8 vectors and testing for their ability to promote transfer or to be nicked when tra operon functions were provided in trans. Removal of DNA sequences to the right of the central A + T-rich region (i.e., from the direction of traM) did not affect the susceptibility of oriT to nicking functions; however, transfer efficiency for oriT segments deleted from the right was progressively reduced over an 80- to 100-bp interval. Deletions extending toward the oriT nick site from the left did not affect the frequency of transfer if deletion endpoints lay at least 22 bp away from the nick site. Deletions or insertions in the central, A + T-rich region caused periodic variation in transfer efficiency, indicating that phase relationships between nicking and transfer domains of oriT must be preserved for full oriT function. These data show that the F oriT locus is extensive, with domains that individually contribute to transfer, nicking, and overall structure.     

Identification and molecular analysis of a 63-kilodalton stress protein from Neisseria gonorrhoeae.

Iron limitation, glucose deprivation, and growth under low oxygen supply (environmental stress) increased the expression of several proteins of Neisseria gonorrhoeae, including a 63-kilodalton protein identified by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. This gonococcal stress protein (GSP63) was detected in the cytosol and copurified with lithium acetate-derived outer membranes. Successful purification of the protein was achieved by sucrose density gradient centrifugation and by chromatography on phenyl-Sepharose. Gel filtration of the purified protein revealed a molecular weight of approximately 450,000, suggesting that in its native state, the protein consists of a multimer of six to eight subunits. Isoelectric focusing indicated a pI of 5.2. Immunoblotting experiments using a polyclonal antiserum raised against the purified protein demonstrated cross-reactivity with a protein of the same electrophoretic mobility as GSP63 in all eight gonococcal isolates tested. N-terminal amino acid sequencing of the protein revealed up to 65% homology with members of the Hsp60 heat shock protein family, suggesting that GSP63 is related to this group of proteins. This relationship was further substantiated by the immunological cross-reactivity of GSP63 with mycobacterial Hsp60 and the ATP-binding activity of the gonococcal stress protein.     

Molecular principles of antigenic variation in Neisseria gonorrhoeae

The genome of Neisseria gonorrhoeae harbours many gene loci for the production of variant pili. Strain MS11 has two expression genes (pilE) with promoter and complete coding sequences. The remaining genes are silent (pilS) lacking the promoter and the conservative amino terminals coding sequences of pilin. The pilus genes consist of six variable minicassettes (mc's), that are flancked by strictly conserved sequences. Upon phase (P⁺ to P⁺) and antigenic (P⁺ to P^–, or vice versa) transitions minicassettes from silent loci are transferred from silent pilus gene copies to the expression gene by gene conversion. P^– variants resulting from such rearrangements still produce pilin mRNA as well as pilin, but only a few are found on the surface of those gonococci.     

Molecular characterization of the interaction between porins of Neisseria gonorrhoeae and C4b-binding protein

Neisseria gonorrhoeae, the causative agent of gonorrhea, is a natural infection only in humans. The resistance of N. gonorrhoeae to normal human serum killing correlates with porin (Por)-mediated binding to the complement inhibitor, C4b-binding protein (C4BP). The entire binding site for both porin molecules resides within complement control protein domain 1 (CCP1) of C4BP. Only human and chimpanzee C4BPs bind to Por1B-bearing gonococci, whereas only human C4BP binds to Por1A strains. We have now used these species-specific differences in C4BP binding to gonococci to map the porin binding sites on CCP1 of C4BP. A comparison between human and chimpanzee or rhesus C4BP CCP1 revealed differences at 4 and 12 amino acid positions, respectively. These amino acids were targeted in the construction of 13 recombinant human mutant C4BPs. Overall, amino acids T43, T45, and K24 individually and A12, M14, R22, and L34 together were important for binding to Por1A strains. Altering D15 (found in man) to N15 (found in rhesus) introduced a glycosylation site that blocked binding to Por1A gonococci. C4BP binding to Por1B strains required K24 and was partially shielded by additional glycosylation in the D15N mutant. Only those recombinant mutant C4BPs that bound to bacteria rescued them from 100% killing by rhesus serum, thereby providing a functional correlate for the binding studies and highlighting C4BP function in gonococcal serum resistance.     

Molecular characterization of two beta-lactamase-specifying plasmids isolated from Neisseria gonorrhoeae.

The molecular nature of two distinct gonococcal R plasmids, 4.4 X 10(6) and 3.2 X 10(6) daltons, encoding beta-lactamase activity were examined. Both plasmids contained about 40% of the transposable ampicillin resistance sequence Tn2. Deoxyribonucleic acid-deoxyribonucleic acid polynucleotide sequence studies have shown that the two gonococcal plasmids share about 70% of their sequences and are closely related to RSF0885, a 4.1 X 10(6)-dalton plasmid found in a beta-lactamase-producing strain of Haemophilus influenzae. All three of these R plasmids possess a guanine-plus-cytosine content of 0.40 to 0.41 mol fraction and are present as multicopy gene pools in their bacterial hosts.     

Linearization of donor DNA during plasmid transformation in Neisseria gonorrhoeae.

We examined the fate of plasmid DNA after uptake during transformation in Neisseria gonorrhoeae. An 11.5-kilobase plasmid, pFA10, was processed to linear double-stranded DNA during uptake by competent cells, but cleavage of pFA10 was not site specific. A minority of pFA10 entered as open circles. A 42-kilobase plasmid, pFA14, was degraded into small fragments during uptake; no intracellular circular forms of pFA14 were evident. Since pFA10 DNA linearized by a restriction enzyme was not further cut during uptake, the endonucleolytic activity associated with entry of plasmid DNA appeared to act preferentially on circular DNA. Although linear plasmid DNA was taken up into a DNase-resistant state as efficiently as circular DNA, linear plasmid DNA transformed much less efficiently than circular plasmid DNA. These data suggest that during entry transforming plasmid DNA often is processed to double-stranded linear molecules; transformants may arise when some molecules are repaired to form circles. Occasional molecules which enter as intact circles may also lead to transformants.     

Evaluation of four methods for detecting the beta-lactamase activity in Neisseria gonorrhoeae isolated in Cuba

Four methods (chromogenic, acidimetric, inhibition, and iodometric) for demonstration of the beta-lactamase production by 70 isolates of Neisseria gonorrhoeae, were evaluated in Cuba. There was 100% correlation between all beta-lactamase methods and the standardized penicillin dilution susceptibility test for penicillinase-non-producing N. gonorrhoeae. For penicillinase-producing N. gonorrhoeae strains, there was a perfect correlation between the chromogenic method and penicillin susceptibility testing, but one and two strains failed to give a positive result for beta-lactamase with the inhibition/acidimetric and the iodometric methods, respectively. There was a high concordance between the chromogenic method, considered as gold standard and the rest of penicillinase tests evaluated: Kappa Index (KI) = 0.98 for inhibition/acidimetric methods and KI = 0.97 for the iodometric method. The four methods evaluated were accurate, reproducible, easily readable, economical, and ease to use for screening primary isolates of N. gonorrhoeae in Cuba. We recommended the use of the inhibition method, when testing the penicillinase activity in gonococcal isolates in provincial and municipal reference laboratories.     

Genetics of surface protein variation in Neisseria gonorrhoeae

Neisseria gonorrhoeae, the bacterium that causes the sexually transmitted disease gonorrhea, demonstrates extensive antigenic heterogeneity in its surface components. The organism has the capacity to switch on and off the synthesis of different versions of components such as pili, outer membrane proteins, and lipopolysaccharide. Recent studies have shown that the gonococcus uses novel and complex mechanisms, of types not described previously, to store different versions of genetic information for surface proteins, and to regulate expression of those genes.     

Intragenic variation by site-specific recombination in the cryptic plasmid of Neisseria gonorrhoeae.

Cryptic plasmid DNA of Neisseria gonorrhoeae was found integrated into the gonococcal chromosome in both plasmid-bearing strains and plasmid-free strains. At several chromosomal locations only segments of the plasmid were found. However, in at least two strains an intact copy of the plasmid seemed to be present with the joints between the plasmid and the chromosomal DNA being located within the cppB gene of the cryptic plasmid. The cppB gene was shown to undergo a sequence-specific intragenic deletion. The deletion removed 54 base pairs, representing 18 amino acids, and did not affect the reading frame. It is proposed that the cryptic plasmid integrates into the chromosome and other gonococcal plasmids within this site-specific deletion region. Models for the site-specific recombination are presented.     

Molecular epidemiology and novel combinations of auxotype, serovar, and plasmid content in tetracycline-resistant Neisseria gonorrhoeae isolated in Canada

Between October 1987 and June 1989, 84 isolates of Neisseria gonorrhoeae carrying the TetM resistance determinant (TRNG) were received at the Laboratory Centre for Disease Control, Ottawa, from six Canadian provinces and were characterized into classes based on auxotype, serovar and plasmid content. One-fifth (17/84) of the TRNG were also penicillinase producing (PPNG). The PPNG-TRNG isolates comprised six classes based on auxotype, serovar, and plasmid content. Most (16/17) PPNG-TRNG carried 3.2-MDa beta-lactamase plasmids and the 25.2-MDa TetM-containing plasmid. We report, for the first time, the association of a 4.5-MDa beta-lactamase plasmid with the 25.2-MDa plasmid in a clinical TRNG isolate. Non-PPNG TRNG isolates comprised 11 classes based on auxotype, serovar, and plasmid content, including two previously unreported auxotype-serovar classes, P/IB-26 and P/IB-20.     

Preferential uptake of restriction fragments from a gonococcal cryptic plasmid by competent Neisseria gonorrhoeae

Factors involved in the specificity of DNA uptake by competent Neisseria gonorrhoeae were examined. Host-controlled modification did not affect uptake. Certain restriction fragments of the 4.2 kb gonococcal cryptic plasmid pFA1 and of the replicative form of the bacteriophage M13 were taken up in preference to others, independent of differences in fragment size. A 600 bp fragment from the 4.2 kb plasmid was cloned into pLES2, a gonococcal-Escherichia coli shuttle vector; the 600 bp fragment was taken up into a DNAase-I-resistant state in preference to the vector fragment. A second 370 bp fragment in pFA1 was also taken up preferentially. The 600 bp and 370 bp fragments share a 10 bp sequence, which is found in pFA1 only on fragments that were taken up readily. However, a fragment from M13 which was efficiently taken up did not contain this 10 bp sequence. In addition, this sequence was not sufficient to direct preferential DNA uptake by gonococci, since a recombinant plasmid containing this 10 bp sequence was not taken up appreciably better than the vector plasmid or another recombinant plasmid containing an unrelated 10 bp sequence. Sequence comparisons of the three restriction fragments which were preferentially taken up did not yield any consensus sequences greater than 7 bp. Although it is likely that efficient uptake of DNA by gonococci is determined by DNA structure, a single short sequence could not be found that accounted for specific uptake.