Invasion of HeLa 229 cells by virulent Bordetella pertussis.

Phase-dependent invasive behavior of Bordetella pertussis was demonstrated by recovery of viable organisms from gentamicin-treated HeLa cell monolayers and by transmission electron microscopy. Several mutants of B. pertussis with Tn5 or Tn5 lac inserted into various vir-regulated genes were evaluated for differences in their invasive abilities. Mutants lacking filamentous hemagglutinin, pertussis toxin, and two as yet uncharacterized vir-regulated products had levels of invasion significantly lower than that of the parent strain BP338. In contrast, invasion by mutants lacking adenylate cyclase toxin was significantly increased compared with that of wild-type B. pertussis. This increase in invasion was eliminated when concentrations of intracellular cyclic 3'-5' AMP were stimulated by treating HeLa cells with cholera toxin or forskolin. Entry of B. pertussis occurred through a microfilament-dependent phagocytic process, as evidenced by the marked reduction in uptake following treatment of HeLa cells with cytochalasin D. Invasion was inhibited with polyclonal anti-B. pertussis and anti-filamentous hemagglutinin antisera. In addition, a monoclonal antibody against lipooligosaccharide A reduced uptake by 65.5%. The preservation of HeLa cell integrity and the limited replication of intracellular bacteria suggest that invasion may represent a means by which B. pertussis evades an active host immune response.     

Lethal infection by Bordetella pertussis mutants in the infant mouse model.

Different aspects of lethal infection of infant mice with Bordetella pertussis were examined. Mutants deficient in vir-regulated genes were tested for the ability to cause a lethal infection in the infant mouse model. Adenylate cyclase toxin-hemolysin and pertussis toxin were required to cause a lethal infection at low doses. Mixed infection caused by challenging the mice with an equal number of pertussis toxin and adenylate cyclase toxin-hemolysin mutants at a dose at which neither alone was lethal was also unable to cause a lethal infection. Production of the filamentous hemagglutinin and the dermonecrotic toxin was not required to cause a lethal infection. Nine other mutants in vir-regulated genes whose phenotypes have yet to be determined were also tested. Only two of these mutants were impaired in the ability to cause a lethal infection. Expression of fimbriae does not appear to affect the dose required to cause a lethal infection; however, fimbrial expression was correlated with the later stages of a nonlethal, persistent infection. Growth of the bacteria in MgSO4, a condition which reversibly suppresses expression of the genes required for virulence, did not alter the ability of the bacteria to cause a lethal infection. Auxotrophic mutants deficient in leucine biosynthesis were as virulent as the parental strain; however, mutants deficient in methionine biosynthesis were less virulent. A B. parapertussis strain was much less effective in promoting a lethal infection than any of the wild-type B. pertussis strains examined. A persistent infection in the lungs was observed for weeks after challenge for mice given a sublethal dose of B. pertussis, and transmission from infected infants to the mother was never observed.     

Tn5-induced mutations affecting virulence factors of Bordetella pertussis.

Transposon Tn5 was used to isolate mutants of Bordetella pertussis. Strains with Tn5 insertions were screened for loss of virulence-associated factors, including filamentous hemagglutinin, hemolysin, and pertussis toxin. Several mutants deficient for hemolysin production were obtained. All produced dermonecrotic toxin, pertussis toxin, and filamentous hemagglutinin, but were found to vary with respect to adenylate cyclase production. One hemolysin mutant had no detectable adenylate cyclase activity; others had 0.6% or 16% wild-type activity, whereas a fourth seemed to be unaffected in terms of adenylate cyclase activity. Mutants deficient in the ability to hemagglutinate sheep erythrocytes were also isolated. These mutants either failed to synthesize or produced reduced amounts of three protein species of 200,000, 130,000, and 100,000 daltons, all of which reacted with antiserum to filamentous hemagglutinin. Pertussis toxin mutants were identified by screening culture supernatants for failure to induce a clustered growth pattern in Chinese hamster ovary cells, and identification was confirmed by the standard histamine-sensitizing assay in mice. These mutants will be useful to determine the relative contribution of each virulence factor to pathogenicity as well as to determine the identity of the antigens important in protective immunity.     

Characterization of vir-activated TnphoA gene fusions in Bordetella pertussis.

The expression of many of the known virulence determinants of Bordetella pertussis is coordinately regulated by the vir regulatory locus and reduced in response to environmental signals called modulators. We have previously identified eight TnphoA gene fusions in B. pertussis in which the expression of alkaline phosphatase was maximal in the absence of the modulators nicotinic acid and MgSO4. We have termed the genes identified by these fusions vir-activated genes. Here we report the characterization of these TnphoA mutant strains. Four fusion strains were defective in known virulence determinants. For one of these, fusion strain SK39, Southern blot hybridization demonstrated that TnphoA was inserted in the S1 subunit gene of pertussis toxin. Hemagglutination assays, sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and immunoblots identified three fusions strains, SK16, SK75, and SK91, that were defective in filamentous hemagglutinin. Whereas all three filamentous hemagglutinin-defective mutants showed either normal or enhanced colonization, the pertussis toxin-defective mutant showed a marked defect in pulmonary persistence. Of the four other fusion strains, two were deficient in outer membrane proteins. One of these, strain SK8, was defective in a major outer membrane protein of 95 kDa as determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. This strain colonized mouse lungs less well and did not induce lymphocytosis after aerosol challenge. The other strain, SK34, was defective in four outer membrane proteins, three of which were detectable only on a Western blot with polyclonal sera against B. pertussis. Two of our gene fusion strains did not show any defect in identifiable vir-regulated proteins.     

Detection of subunits of pertussis toxin in Tn5-induced Bordetella mutants deficient in toxin biological activity.

Monoclonal antibodies with specificity for pertussis toxin subunits S1, S2, and S4 were used in Western blots to show that the subunits were not secreted into culture medium from Tn5 insertion mutants. The mutants are deficient in toxin biological activities due to an insertion in the S3 subunit structural gene. The Western blots demonstrated that each of the respective subunits was exported in a wild-type strain. Anti-S1 and anti-S2 monoclonal antibodies were capable of detecting subunits in solubilized whole-cell material from a wild-type strain and from the Tn5 mutants lacking only in biologically active toxin (Tox-). Another Tn5 insertion mutant, lacking all known B. pertussis virulence factors (Vir-), did not produce any of the subunits either in whole cellular extracts or in culture supernatants. The data demonstrate that Tn5 Tox- insertion mutants, though defective in toxin activity, synthesize some toxin subunits. The presence of the S3 subunit is most likely a necessity for transport of the toxin from cells. Alternatively, a nonstructural gene coding for a protein involved in transport of the toxin across the membrane may be affected by the Tn5 mutation.     

Mutations in the S1 subunit of pertussis toxin that affect secretion

Pertussis toxin is a member of the AB(5) family of toxins and is composed of five subunits (S1 to S5) present in a 1:1:1:2:1 ratio. Secretion is a complex process. Each subunit has a secretion signal that mediates transport to the periplasm, where processing and assembly occur. Secretion of the assembled 105-kDa toxin past the outer membrane is mediated by the nine proteins encoded in the ptl operon. Previous studies have shown that S1, the catalytically active A subunit of pertussis toxin, is necessary for efficient secretion, suggesting that a domain on S1 may be required for interaction with the secretion apparatus. Previously, recombinant S1 from four different mutants (serine 54 to glycine, serine 55 to glycine, serine 56 to glycine, and arginine 57 to lysine) was shown to retain catalytic activity. We introduced these mutations into Bordetella pertussis and monitored pertussis toxin production and secretion. No pertussis toxin was detected in the serine 54-to-glycine mutant. The other S1 mutants produced periplasmic pertussis toxin, but little pertussis toxin secretion was observed. The arginine 57-to-lysine mutant had the most dramatic secretion defect. It produced wild-type levels of periplasmic pertussis toxin but secreted only 8% as much toxin as the wild-type strain. This phenotype was similar to that observed for strains with mutations in the ptl genes, suggesting that this region may have a role in pertussis toxin secretion.     

Expression of pertussis toxin in Bordetella bronchiseptica and Bordetella parapertussis carrying recombinant plasmids.

Pertussis toxin is produced only by strains of Bordetella pertussis. Cloned genes encoding pertussis toxin from B. pertussis were transferred into Bordetella bronchiseptica and Bordetella parapertussis by conjugation. These transconjugants expressed pertussis toxin at levels comparable to those expressed by B. pertussis. The toxin made by these strains was biologically active in the Chinese hamster cell clumping assay, contained all five subunits, and was mostly periplasmic. Toxin expression appeared to be modulated in the same way as are the vir-regulated genes of B. pertussis. Introduction of these plasmids into B. pertussis failed to produce hypertoxigenic strains. Instead, these transconjugants underwent plasmid loss, gene deletions, or conversion to the avirulent phase.     

Construction and analysis of Bordetella pertussis mutants defective in the production of fimbriae.

Although the role of fimbriae in bacterial disease has been well established, little is known about the function of Bordetella pertussis fimbriae. To study this function, well-defined fimbrial mutants were constructed. B. pertussis harbours three fimbrial genes, fim2, fim3 and fimX, and strains were constructed in which one or more fimbrial genes were inactivated by means of gene replacement. Analysis of these strains by means of immunoblotting suggested the presence of a fourth fimbrial gene, tentatively designated fimY. A fimbrial mutant was analysed in a mouse respiratory infection model, together with a strain harbouring a deletion in the gene for the filamentous haemagglutinin. Both mutants were affected in their ability to persist in the trachea. Persistence in the nasopharynx was only affected by the mutation in the filamentous haemagglutinin gene. Neither the filamentous haemagglutinin nor the fimbrial mutants were affected in their ability to persist in the lung. Our results suggest that the filamentous haemagglutinin plays a more crucial role than fimbriae in the colonization of the upper respiratory tract of the mouse.     

Isolation of Citrobacter sp. mutants defective in decolorizing malachite green

To identify genes involved in the decolorization of malachite green, random mutants generated by transposon insertion in the malachite green-decolorizing bacterium, Citrobacter sp. were isolated. The resulting mutant bank yielded 24 mutants with complete defects in their abilities to decolorize malachite green. Southern hybridization with a Tn5 fragment as a probe showed a single hybridized band in 7 mutants, which appeared to have insertions at different sites of the chromosome. The Tn5-inserted genes were isolated and the DNA sequence flanking Tn5 was determined. Based on a sequence database, the putative protein products encoded by the mg genes were identified as follows. mg3, an ABC transporter homolog; mg6, a LysR-type regulatory protein; m11, an oxidoreductase; mg17, a MalG protein in the maltose transport system; and mg21, a sugar kinase. The deduced sequences from two mg genes (mg7 and mg18) showed no significant similarity to any protein with a known function, suggesting that these two mg genes encode unidentified proteins that are responsible for the decolorization of malachite green.     

Cloning of the adenylate cyclase genetic determinant of Bordetella pertussis and its expression in Escherichia coli and B. pertussis

A recombinant plasmid, pRMB1, identified from a gene library of B. pertussis, restored adenylate cyclase (AC) and haemolysin (HLY) activities to B. pertussis BP348 (a Tn5-insertion mutant deficient in both these activities). B. pertussis BP348 was considerably less virulent than wild type strains of B. pertussis when 3-week-old mice were challenged intranasally; possession of pRMB1 restored virulence. Neither AC nor HLY activities were expressed in E. coli harbouring pRMB1. However, expression of calmodulin-responsive AC was obtained in E. coli when restriction fragments of pRMB1 were subcloned into other vectors; expression depended on the lac and tac promoters of these vectors. The enzyme was not readily solubilized from urea extracts of E. coli and required sonication for efficient release. One plasmid conferred a specific AC enzymic activity to E. coli which was greater than that for wild type B. pertussis strains. Unlike extracts of B. pertussis, extracts from E. coli expressing enzymic AC activity, did not elevate cAMP levels in S49 lymphoma cells. A second plasmid, pRMB2, was identified from the gene library, which contained a trans-acting regulatory determinant required for expression of AC, HLY and other virulence-associated factors in B. pertussis.     

Characterization of fimbrial subunits from Bordetella species

Using antisera raised against serotype 2 and 3 fimbrial subunits from Bordetella pertussis, serologically related polypeptides were detected in Bordetella bronchiseptica, Bordetella parapertussis and Bordetella avium strains. The two B. pertussis fimbrial subunits, and three of the serologically related B. bronchiseptica polypeptides, were shown to be very similar in amino acid composition and N-terminal amino acid sequence. Homology was observed between the N-termini of these polypeptides, and fimbrial subunits from Escherichia coli, Haemophilus influenzae and Proteus mirabilis. A synthetic oligonucleotide probe, derived from the N-terminal sequence of the B. pertussis serotype 2 fimbrial subunit, was used to identify fimbrial genes in genomic Southern blots. The results suggested the presence of multiple fimbrial subunit genes in B. pertussis, B. bronchiseptica and B. parapertussis. The DNA probe was used to clone one of the three tentative fimbrial subunit genes detected in B. pertussis.     

Identification of a virulence-associated determinant, dihydrolipoamide dehydrogenase (lpd), in Mycoplasma gallisepticum through in vivo screening of transposon mutants

To effectively analyze Mycoplasma gallisepticum for virulence-associated determinants, the ability to create stable genetic mutations is essential. Global M. gallisepticum mutagenesis is currently limited to the use of transposons. Using the gram-positive transposon Tn4001mod, a mutant library of 110 transformants was constructed and all insertion sites were mapped. To identify transposon insertion points, a unique primer directed outward from the end of Tn4001mod was used to sequence flanking genomic regions. By comparing sequences obtained in this manner to the annotated M. gallisepticum genome, the precise locations of transposon insertions were discerned. After determining the transposon insertion site for each mutant, unique reverse primers were synthesized based on the specific sequences, and PCR was performed. The resultant amplicons were used as unique Tn4001mod mutant identifiers. This procedure is referred to as signature sequence mutagenesis (SSM). SSM permits the comprehensive screening of the M. gallisepticum genome for the identification of novel virulence-associated determinants from a mixed mutant population. To this end, chickens were challenged with a pool of 27 unique Tn4001mod mutants. Two weeks postinfection, the birds were sacrificed, and organisms were recovered from respiratory tract tissues and screened for the presence or absence of various mutants. SSM is a negative-selection screening technique whereby those mutants possessing transposon insertions in genes essential for in vivo survival are not recovered from the host. We have identified a virulence-associated gene encoding dihydrolipoamide dehydrogenase (lpd). A transposon insertion in the middle of the coding sequence resulted in diminished biologic function and reduced virulence of the mutant designated Mg 7.     

Construction and characterization of Bordetella pertussis toxin mutants.

Pertussis toxin is one of the major virulence determinants produced by Bordetella pertussis. The DNA encoding the structural genes for pertussis toxin was cloned in Escherichia coli, and pertussis toxin subunit S4 was expressed under the control of the tac promoter. Mutations were introduced into the cloned toxin genes, and a conjugative shuttle vector system was devised for delivering the mutations from E. coli back into B. pertussis. The mutations were introduced by allelic exchange into the chromosome of B. pertussis resulting in a series of B. pertussis strains which were isogenic except at the loci encoding the structural genes for pertussis toxin. These B. pertussis strains were utilized to study the biogenesis of pertussis toxin. Polar mutations in the S1 gene led to a lack of detectable S2 or S4 subunits in whole-cell lysates, suggesting a polycistronic arrangement for these genes. Mutations in the S5 subunit gene resulted in a truncated S1 subunit, while mutations in the S4 gene resulted in a lack of detectable S2 subunit, suggesting that physical relationships among the toxin subunits are directly reflected in the stable biogenesis of the subunits.     

Localization of stx, a determinant essential for high-level production of shiga toxin by Shigella dysenteriae serotype 1, near pyrF and generation of stx transposon mutants.

Hfr strains of Shigella dysenteriae serotype 1 were constructed by transient integration of an RP4 plasmid derivative carrying transposon Tn501 into the Shigella chromosome through Tn501-mediated cointegration. The Hfr strains were mated with Escherichia coli K-12 recipients carrying various auxotrophic markers, and E. coli recombinants which had received prototrophic Shigella genes were selected. Some of the E. coli transconjugants produced high levels of a cytotoxin which was neutralized by both polyclonal and monoclonal anti-Shiga toxin sera. The determinant for Shiga toxin production, designated stx, was first transferred to E. coli K-12 and then mapped by Hfr crosses to the trp-pyrF region located at 30 min on the E. coli chromosome. Bacteriophage P1-mediated transduction analysis of stx gave the following gene order: trp-pyrF-stx. The level of Shiga toxin production in E. coli Stx+ transconjugants and transductants was as high as that of the parental S. dysenteriae 1 strain. Stx- mutants of an Stx+ E. coli transductant were generated by random in vivo insertion mutagenesis with a Tn10 derivative transposon, Tn-mini-kan, followed by P1 cotransduction of the kanamycin resistance and PyrF+ markers into a pyrF Stx+ E. coli K-12 recipient. One stx::Tn-mini-kan transposon mutation was transferred by P1 transduction from this E. coli Stx- mutant to an E. coli K-12 Hfr strain and in turn transferred by conjugation to the original S. dysenteriae 1 strain plus two others. All kanamycin-resistant recombinants of S. dysenteriae 1 had lost their ability to produce high levels of Shiga toxin. A gene that specifies high-level Shiga toxin production is thus located near pyrF on the chromosome of S. dysenteriae 1. Stx- mutants of S. dysenteriae 1 exhibited full virulence in the Serény test.     

Salmonella typhimurium mutants lacking flagella or motility remain virulent in BALB/c mice

Nonmotile flagellated (mot) and nonflagellated (fla) mutants of Salmonella typhimurium LT-2 were isolated from a collection of mutants with random Tn10-insertion mutations. Both classes of mutants were resistant to infection by the flagellotropic bacteriophage chi. The nonflagellated (fla::Tn10) mutants did not react with H antigen-specific antisera and did not possess flagella when examined by electron microscopy, and sheared-cell extracts were devoid of flagellin. The nonmotile (mot::Tn10) mutants reacted with H-specific antisera and expressed paralyzed flagella that were indistinguishable from wild-type flagella. The Tn10 insertions in strain LT-2 were mapped to loci in regions II (flh and mot) and III (fli) of the flagellar genes, and the mutations were transduced into the mouse-virulent S. typhimurium strains SR-11 and SL1344. Lack of motility reduced the ability of S. typhimurium to invade Henle cells in vitro, yet the virulence in mice of the nonmotile mutants of SR-11 and SL1344 was unaffected by the inactivity or loss of flagella. Wild-type SR-11 had a 50% lethal dose (LD50) in BALB/c mice following oral (p.o.) challenge of 2.4 x 10(4) CFU. The p.o. LD50 of the SR-11 fli-8007::Tn10 mutant was 4.5 x 10(4) CFU. The mot-8008::Tn10 mutation in SR-11 conferred paralyzed flagella and increased the p.o. LD50 in mice to 2.2 x 10(5) CFU, but this was not statistically significant. A similar increase in the p.o. LD50 was observed when the SL1344 mot-8008::Tn10 mutant was tested in mice. Wild-type SR-11 and the isogenic nonflagellated and nonmotile mutants were equally virulent in mice challenged via intraperitoneal injection.     

Characterization of murine lung inflammation after infection with parental Bordetella pertussis and mutants deficient in adhesins or toxins.

Bordetella pertussis expresses factors such as filamentous hemagglutinin, agglutinogens, pertactin, and pertussis toxin, which participate in bacterial adhesion; pertussis toxin, dermonecrotic toxin, lipopolysaccharide, and tracheal cytotoxin, which are responsible for toxic effects; and adenylate cyclase-hemolysin, which is required to initiate infection. By using a murine respiratory model, we showed that the RGD sequences of filamentous hemagglutinin and pertactin are important for bacterial persistence. However, mutants deficient in filamentous hemagglutinin and agglutinogens or in pertactin and the RGD sequence of filamentous hemagglutinin behaved as did wild-type B. pertussis, i.e., induced bronchopneumonia, alveolitis, and an influx of macrophages, lymphocytes, and polymorphonuclear leukocytes into bronchoalveolar lavage fluids. These results suggest that these adhesins are not involved in the induction of pulmonary lesions following infection. The intensity of inflammation was markedly reduced after infection with mutants deficient in either hemolytic activity or pertussis toxin expression, whereas a mutant devoid of adenylate cyclase activity behaved as did the avirulent mutant. Pertussis toxin and adenylate cyclase-hemolysin may act indirectly by altering immune cell functions and thus allowing other factors, such as filamentous hemagglutinin, agglutinogens, and pertactin, to trigger adhesion and lipopolysaccharide, dermonecrotic toxin, and tracheal cytotoxin to induce their toxic effects. However, it is possible that pertussis toxin is also responsible for the induction of some pulmonary alterations.     

Transposon Tn5 mutagenesis of Brucella abortus.

We demonstrate that transposon-mediated mutagenesis can be used to construct mutations in the pathogen Brucella abortus. We have used both a plasmid and a bacteriophage to introduce either Tn5 or Tn5 lac into the Brucella chromosome. B. abortus is naturally sensitive to kanamycin. We have selected 22 independent kanamycin-resistant colonies in strain US-19 and 19 colonies in strain 2308. We have demonstrated by Southern hybridization that Tn5 was inserted into the chromosome.