the virulence factors of bordetella pertussis: a matter of control

The virulence factors of Bordetella pertussis : a matter of control

Adam M. Smith a;1, Carlos A. Guzman b, Mark J. Walker a;*a Department of Biological Sciences, University of Wollongong, Wollongong, N.S.W. 2522, Australia

b Department of Microbial Pathogenicity and Vaccine Research, Division of Microbiology, GBF-National Research Centre for Biotechnology,Braunschweig, Germany

Received 3 April 2000; received in revised form 8 November 2000; accepted 19 December 2000

Abstract

Bordetella pertussis is the causative agent of whooping cough, a contagious childhood respiratory disease. Increasing public concern overthe safety of whole-cell vaccines led to decreased immunisation rates and a subsequent increase in the incidence of the disease. Research intothe development of safer, more efficacious, less reactogenic vaccine preparations was concentrated on the production and purification ofdetoxified B. pertussis virulence factors. These virulence factors include adhesins such as filamentous haemagglutinin, fimbriae andpertactin, which allow B. pertussis to bind to ciliated epithelial cells in the upper respiratory tract. Once attachment is initiated, toxinsproduced by the bacterium enable colonisation to proceed by interfering with host clearance mechanisms. B. pertussis co-ordinatelyregulates the expression of virulence factors via the Bordetella virulence gene (bvg) locus, which encodes a response regulator responsible forsignal-mediated activation and repression. This strict regulation mechanism allows the bacterium to express different gene subsets indifferent environmental niches within the host, according to the stage of disease progression. ß 2001 Federation of European Micro-biological Societies. Published by Elsevier Science B.V. All rights reserved.

Keywords: Virulence factor expression; Pertussis toxin; Adenylate cyclase; Fimbria; Filamentous hemagglutinin; Pertactin; Bordetella ; bvg

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3102. The genus Bordetella . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3103. Genetic control of virulence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 310

3.1. Bordetella virulence gene regulation system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3113.2. Transcriptional regulation by BvgA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3133.3. Involvement of RNA polymerase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3143.4. Di¡erential activation of virulence factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3143.5. `Bvgi' . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3153.6. The Ris system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3153.7. Bvg accessory factor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315

4. Virulence factors of B. pertussis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3154.1. Filamentous haemagglutinin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3154.2. Serotype-speci¢c ¢mbriae . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3184.3. Pertactin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3204.4. Pertussis toxin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3214.5. Adenylate cyclase toxin/haemolysin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3224.6. Dermonecrotic toxin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3234.7. Tracheal cytotoxin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3244.8. Lipopolysaccharide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3244.9. Tracheal colonisation factor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324

0168-6445 / 01 / $20.00 ß 2001 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved.PII: S 0 1 6 8 - 6 4 4 5 ( 0 1 ) 0 0 0 5 6 - 0

* Corresponding author. Tel. : +61 (2) 4221 3439; Fax: +61 (2) 4221 4135; E-mail : [email protected]

1 Present address: GADI Research Centre, University of Canberra, Canberra, A.C.T., Australia.

FEMSRE 716 1-5-01

FEMS Microbiology Reviews 25 (2001) 309^333

www.fems-microbiology.org

4.10. Serum resistance locus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3255. Elicitation of protective immune responses against B. pertussis . . . . . . . . . . . . . . . . . . . . . 325

5.1. Cell-mediated immunity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3255.2. Development of vaccine resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325

6. Concluding remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 326

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 326

1. Introduction

Pertussis or whooping cough is a highly contagious dis-ease of the human respiratory tract, which is particularlysevere in infants. The disease is characterised by bron-chopneumonia, paroxysmal coughing and the distinctive`whooping' intake of air. Pertussis is especially prevalentin developing countries where appropriate medical assis-tance is often unavailable and disease progression is un-impeded due to the lack of suitable antimicrobials. Thereadily transmissible nature and the severity of the diseasein these countries have seen worldwide deaths fromwhooping cough increase to approximately 346 000 peryear [1]. Whooping cough is caused by the obligate humanpathogen, Bordetella pertussis, a Gram-negative coccoba-cillus originally isolated in 1906 by Bordet and Gengou [2].

The normal course of infection begins with the bacte-rium entering the host via the airways, contained withinairborne droplets derived from the cough of an infectedindividual. The pathogen proceeds down the respiratorytract, adhering to ciliated epithelial cells in the tracheaand nasopharynx. Once attachment is initiated, the bacte-rium begins to replicate and colonises adjacent areas. Tox-ins secreted by the micro-organism damage the epitheliallining, resulting in the loss of ciliated cells which inducesthe characteristic coughing. These toxins also allow thebacterium to evade the host immune response by interfer-ing with clearance mechanisms. Halting ciliary function,short-circuiting host G protein signalling apparatus andinhibiting immune cell function by upregulating cAMPlevels are all examples of the damage caused by the vari-ous toxins of B. pertussis.

The incidence of whooping cough has been greatlyreduced since the advent of pertussis vaccines in the1940s. These `whole-cell' vaccines consist of chemicallyor heat-killed B. pertussis cells. These vaccines, as partof the diphtheria, tetanus, pertussis (DTP) immunisationregime, proved extremely e¡ective at preventing the symp-toms of whooping cough. Unfortunately, a loose associa-tion with rare neurologic complications and deaths hasled to a decline in public con¢dence, a subsequent reduc-tion in immunisation rates, and an increase in the inci-dence of the disease. The other constituents of the DTPvaccine are puri¢ed diphtheria and tetanus toxoids. Thesevaccine components have attained a comparatively lesscontroversial reputation, and signi¢cant research hasbeen carried out in an attempt to bring the pertussis com-

ponent into line with the less reactogenic constituents ofthe vaccine.

2. The genus Bordetella

Historically, the bacteria currently classi¢ed within thisgenus have had a very dynamic taxonomic existence, notalways being classi¢ed together. It was Lopez who pro-posed that a new group, the genus Bordetella, be createdto house the then three species, pertussis, parapertussis andbronchiseptica [3]. More recent molecular evidence, includ-ing DNA sequencing analysis, multilocus enzyme electro-phoresis and restriction fragment length polymorphismtyping, has justi¢ed the grouping of these three speciesinto the same genus [4,5]. B. pertussis is exquisitely tunedto a single environment and has only been found to infecthumans.

B. parapertussis causes a milder `pertussis-like' diseasestate in humans and has also been found associated withovine respiratory infections [6]. Most other mammals canbe colonised by B. bronchiseptica with many commerciallyimportant diseases caused by this micro-organism [7], themost notable being an association with atrophic rhinitis inswine which costs the meat industry considerable pro¢ts,and kennel cough in dogs [8,9]. The bird pathogen Borde-tella avium, a later addition to the genus (Table 1) [10],causes turkey coryza and other respiratory diseases of fowl[11].

More recently, a number of new members of the genushave been proposed. Bordetella hinzii is the name assignedto a novel species recently isolated from a bacteraemicindividual with HIV/AIDS [12]. Associated with humansepticaemia, another species, designated Bordetella holme-sii, has also been identi¢ed [13]. The name Bordetella tre-matum has recently been proposed for a novel speciesisolated from human wounds and ear infections [14].Although none of these new species were associated withrespiratory tract infections, they are similar to other mem-bers of the genus based on phylogenetic analyses [12^14].Since its initial discovery, B. hinzii has also been isolatedfrom the respiratory tract of infected poultry [15].

3. Genetic control of virulence

The goal of any pathogenic bacterium, including B. per-

FEMSRE 716 1-5-01

A.M. Smith et al. / FEMS Microbiology Reviews 25 (2001) 309^333310

tussis, is to colonise and multiply by exploiting its environ-ment to the fullest extent. This is achieved by the bacte-rium controlling the production of the speci¢c factors thatenable it to infect its host.

3.1. Bordetella virulence gene regulation system

It was realised many years ago that B. pertussis existedin more than one phase [16]. `Phase variation' occurs whenspontaneous mutations during DNA replication give riseto avirulent variants. This process occurs at a frequency ofone in a thousand to one in a million cells and producesavirulent bacteria which do not have the ability to invadeand colonise the host. However, if, when culturing wild-type B. pertussis, the growth conditions are altered, thebacteria undergo a totally reversible conversion to oneof principally two states, called X and C modes by Laceyin 1960 [17]. Termed àntigenic modulation', this phenom-enon may be important in the survival of this micro-or-ganism. This landmark study also recognised a third state(I mode), with growth characteristics intermediate betweenX and C modes, and `àn in¢nitude of other antigenicstates''.

More recent research has dealt almost entirely with thetwo extreme modes of growth, virulent (X) and avirulent(C). However, studies on the related species B. bronchisep-tica have provided new insight into the intricacies of anti-genic modulation. Due to the expression of a set of genesunder modulating growth conditions which are easilymonitored, B. bronchiseptica may be used as a model forother members of the genus. An intermediate-phase B.bronchiseptica mutant similar to Lacey's I mode has re-cently been isolated which expresses phenotypes character-istic of both virulent and avirulent states [18].

The members of the genus Bordetella share a geneticlocus which encodes a biological `switch' enabling thesespecies to oscillate between di¡erent phenotypic states de-pending on their environment. In the human respiratory

tract, a subset of B. pertussis genes are expressed whichallow the bacterium to colonise the host. The genes for anumber of adhesins, such as ¢lamentous haemagglutinin(FHA), serotype-speci¢c ¢mbriae, and a 69-kDa outermembrane protein known as pertactin, are expressed, asare the genes encoding products involved in evasion of thehost immune system: pertussis toxin (PT), adenylate cy-clase toxin/haemolysin and dermonecrotic toxin.

In the laboratory, the virulence control switch can bemanipulated to bring Bordetella species into an avirulentstate by the addition of sulfate ions or nicotinic acid to theculture medium, or by reducing the culture temperaturefrom 37³C to 25³C [19].

The expression of virulence factors in Bordetella speciesis co-ordinately regulated by what was originally termedthe vir locus [20] and is now known as the Bordetellavirulence gene (bvg) locus. This gene locus encodes twoproteins which enable the bacterium to `sense' the prevail-ing environmental conditions, and then àct' accordinglyby controlling the expression of speci¢c genes and geneloci. The sensor protein is known as BvgS and the activa-tor as BvgA. Together these gene products constitute atwo-component signal transduction system and make uppart of a signal transduction superfamily [21]. This type ofsystem is utilised to respond to many di¡erent environ-mental stimuli by many prokaryotes, and recently hasbeen described in some eukaryotes [21^24].

The genes encoding BvgA and BvgS are contained with-in an operon [25] (Fig. 1) [26]. The original sequencingdata unfortunately contained mistakes in and downstreamof the bvgS gene, confusing the exact arrangement of theoperon and obscuring the presence of the bvgR open read-ing frame (ORF) [25]. The discovery and analysis of athird gene, bvgR, has extended both the size and scopeof the bvg locus [27]. While bvgA and bvgS are transcrip-tionally linked, bvgR is transcribed separately from theopposing DNA strand (Fig. 1). Transcription of bvgASis controlled by three self-regulated promoters, P1, P3

Table 1Comparison of the gene products synthesised by the four most studied members of the genus Bordetellaa

B. pertussis B. parapertussis B. bronchiseptica B. avium

Pertussis toxin (PT) +b 3 3 3Adenylate cyclase toxin/haemolysin + + + 3Filamentous haemagglutinin (FHA) + + + 3Fimbriae + + + +Pertactin + + + +Tracheal cytotoxinc + + + +Lipopolysaccharide (LPS) + + + +Dermonecrotic toxin + + + +Flagella 3 3 + +Tracheal colonisation factor (Tcf) + 3 3 3BrkA + + +/3 3Urease 3 +/3 + 3

aUpdated and modi¢ed from Gross et al. [10].bAbbreviations: +, gene product expressed; +/3, gene expressed in some strains; 3, gene cryptic or absent.cExpression of tracheal cytotoxin is not regulated by the bvg locus.

FEMSRE 716 1-5-01

A.M. Smith et al. / FEMS Microbiology Reviews 25 (2001) 309^333 311

and P4, and one bvg-independent promoter, P2 (Fig. 1)[28,29]. Transcription from P2 is constitutive, therebymaintaining low-level production of intracellular BvgASproteins. While P1, P2 and P3 are all involved in the tran-scription of bvgAS, P4 produces an antisense RNA strandcomplementary to transcripts expressed from the otherthree promoters. The exact in vivo function of this pro-moter remains unclear; however, it may be involved instabilisation of the sense transcripts via the prevention ofsecondary structure formation. Hybridisation betweenmRNA from P1 and P4 may lead to a more e¤cient in-teraction with ribosomes [29].

The gene encoding FHA is transcribed from a regionjust upstream of these four promoters [29] at Pfha (Fig. 1).There has since been discovered a bvg-independent FHApromoter, Pfha�, which has been shown to express lowlevels of the FHA precursor protein, FhaB, in vitro[30,31].

3.1.1. BvgAThe transcriptional activator, BvgA, which can directly

and indirectly a¡ect the expression of a number of viru-lence-controlled genes, is a 23-kDa cytoplasmic protein.The two-domain structure of BvgA (Fig. 2), consistingof a receiver at the N-terminus and a C-terminal helix-turn-helix (HTH) motif, is characteristic of response reg-ulators [25,32]. It is this HTH module which facilitatesspeci¢c DNA sequence binding at the promoters of bvg-activated genes [33]. Although not part of the HTH do-main, the last 20 amino acid residues are also required forbinding, probably by stabilising the HTH^DNA bindinginteraction [33,34].

The binding of DNA by proteins is a common phenom-enon within cells, and the C-terminal HTH domain ofBvgA shares sequence homology with a number of di¡er-

ent protein families. As expected, this domain is homolo-gous to corresponding areas of other response regulatorssuch as FixJ, UhpA and NarL [34]. Other DNA bindingproteins such as LuxR from Vibrio ¢scheri [35] as well asthe C-termini of selected bacterial c factors [36] also sharehomology with this evolutionarily conserved domain.

Direct control over the transcription of a number ofbvg-activated genes by BvgA is achieved via binding tospeci¢c sequences at target promoter regions [37,38]. Theheptameric sequence-speci¢c BvgA binding siteTTTCCTA ¢rst proposed by Roy and Falkow, or homol-ogous sequences, is present as either direct or invertedrepeats upstream of various bvg-regulated genes [38^40].These homologous heptameric repeats are involved in thebinding of BvgA to virulence-activated promoters, a pointsubstantiated in a number of virulence-activated pro-moters by DNase I protection studies [39^42].

3.1.2. BvgSIn a process that is not yet fully understood, a histidine

kinase residue in the transmitter domain of BvgS is auto-phosphorylated, a state which can be altered if the envi-ronment changes. Following a series of intramolecularphosphorylation events, the phosphoryl group is eventu-ally transferred to BvgA [43]. If there is a change in theenvironment to modulating conditions, the initial phos-phorylation does not occur, preventing the activation ofbvgA.

The 135-kDa sensory component of the signal transduc-tion system, BvgS, is often termed an unorthodox sensorprotein because of the number of extra functional domains

Fig. 2. Domain structure of the 209-amino acid transcriptional activatorBvgA. The receiver domain (open box) is shown, with the phosphoryla-tion site (aspartic acid, D54) indicated. The C-terminal domain (shadedbox) incorporating the helix-turn-helix domain (¢lled box; amino acids168^185) is also shown. Modi¢ed from Uhl and Miller [46].

Fig. 1. Schematic representation of the genetic region encompassing the bvg operon and the fhaB gene, highlighting the promoter organisation of thebvg/fha intergenic region. The genes encoding bvgA, bvgS and bvgR (shaded boxes) and fhaB (open box) are shown, and the direction of transcriptionfrom each promoter is indicated by arrows. The genes for bvgA and bvgS are transcriptionally linked. bvgR is transcribed separately in the opposite di-rection. Modi¢ed from Merkel and Stibitz [48] and Coote [26].

Fig. 3. Domain structure of the 1236-amino acid protein BvgS. The514-aa sensor domain (single line), the 159-aa linker region (single line),232-aa transmitter domain (open box), 131-aa receiver domain (lightlyshaded box) and the 126-aa C-terminal output (heavily shaded box) re-gions are shown. Phosphorylation sites (histidine, H729 ; aspartic acid,D1023 ; and histidine, H1172) are indicated. The transmembrane domains(TM; black boxes) are also depicted. Modi¢ed from Uhl and Miller[46].

FEMSRE 716 1-5-01


it contains. The membrane spanning nature of this proteinpredicts it possesses both periplasmic and transmembranedomains, and sequence homology studies con¢rm the ab-solutely conserved histidine kinase domain [25,32]. How-ever, BvgS also contains a linker section and regions usu-ally reserved for response regulators such as BvgA, areceiver domain and a C-terminal output domain (Fig.3) [25,32]. The deletion of these `extra' domains has beenshown to make BvgS non-functional [25,43,44].

Experiments with BvgS^PhoA fusion proteins haveshown that the N-terminal region is located in the peri-plasm [32]. Mutations in the linker region render the bac-terium incapable of sensing the environment providingevidence that this region acts as a molecular relay between

periplasmic and C-terminal domains located in the cyto-plasm [39,45].

The transmitter region of BvgS is necessary and su¤-cient for autophosphorylation [43]; however, transfer ofthis phosphoryl group to the C-terminal (output) domaincan only be e¡ected via the receiver domain [46]. Not onlydoes the receiver domain positively regulate the phospho-transfer to the C-terminal output domain, it also has theability to reverse this transfer, thereby rephosphorylatingitself and postponing BvgA activation (Fig. 4). It can alsototally dephosphorylate BvgS from the receiver itself,thereby completely halting the activation of BvgA (Fig.4) [46]. It is the output domain (particularly His1172) ofthe C-terminus which is responsible for the phosphoryla-tion of BvgA [47].

3.1.3. BvgRWhile BvgA exercises direct control over the expression

of bvg-activated genes by binding to promoter sequences,control over virulence repressed genes (vrgs) is indirect,implicating a Bvg-activated repressor protein. The genewhich controls repression of the bvg-repressed genes hasbeen identi¢ed and termed bvgR [48].

The DNA sequence of the 5P end of four of the ¢ve bvg-repressed genes (vrg6, vrg18, vrg24, vrg53 and vrg73) hasbrought to light a conserved consensus sequence elementpresent in the coding region of four of these genes [49,50].Using a Southwestern analysis with vrg6 DNA, this se-quence was recognised by a 34-kDa protein [50]. Whenanalysed, none of the native promoter regions of any ofthe bvg-repressed genes seem necessary for modulative re-pression to occur [50].

The bvgR gene lies downstream of and adjacent tobvgAS [48]. Further characterisation led to the identi¢ca-tion of the exact location of the bvgR gene (Fig. 1) [27].Transcribed from its own BvgA-regulated promoter, bvgRlies 43 bp downstream of bvgS and is transcribed in theopposite orientation. Transcriptional analysis has shownthat this ORF produces a predicted protein of 32 kDa,close to the molecular mass (34 kDa) obtained by Beattieand associates [27].

It has been shown that B. pertussis mutants lackingbvgR are less e¤cient colonisers than wild-type strains ina mouse aerosol challenge model, demonstrating thatBvgR regulation of the bvg-repressed genes makes a sig-ni¢cant contribution to infection in mice [51].

3.2. Transcriptional regulation by BvgA

As the ¢nal part of an intricate phosphorelay process,BvgS phosphorylates BvgA. When compared to its un-phosphorylated counterpart, this activated `BvgAVP'has a far greater a¤nity for virulence factor promoterBvgA binding sites [30,33,41,42]. Indeed, BvgA must bephosphorylated for bvg-activated promoters to becomefully operational in in vitro transcription assays [30,31].

Fig. 4. The BvgS^BvgA phosphorylation cascade model. The sensor do-main receives an environmental signal that causes the autophosphoryla-tion of the histidine (H729) residue of the transmitter domain (openbox). The phosphoryl group is then transferred to an aspartic acid resi-due (D1023) of the receiver domain (lightly shaded box). The phosphorylgroup can then be relayed to the histidine (H1172) residue of the C-ter-minal output (heavily shaded box), or can react with water to dephos-phorylate the BvgS molecule. The output domain then donates thephosphoryl group to BvgA, leading to the transcriptional initiation ofvirulence genes. The short transmembrane domains (TM; black boxes)are also depicted. Modi¢ed from Cotter and Miller [18].

FEMSRE 716 1-5-01


Although BvgA-independent transcription of the fhaBgene has been observed in vitro, expression from this al-ternative promoter, Pfha�, only occurs at low levels [30,31].

It was theorised that the positioning of BvgAVP at apromoter region may induce BvgA oligomerisation [29]and raise the a¤nity of the response regulator for thetarget DNA [33]. It has been suggested that BvgAVPmay bind co-operatively to bvg-activated promoters, form-ing multimers along the promoter template [31,40^42]. In-itially, high-a¤nity binding occurs far upstream of thetranscriptional start site, then weaker sites are progres-sively bound until the site adjacent to the RNA polymer-ase recognition sequence is occupied.

3.3. Involvement of RNA polymerase

When bound to the promoter DNA at the correct startsite, RNA polymerase catalyses the initiation of transcrip-tion. RNA polymerase is composed of four subunits, K, L,LP and c. The K subunit is generally the portion of RNApolymerase that interacts with transcription factors at pos-itively regulated prokaryotic promoters [52,53]. The c sub-unit is involved in promoter recognition, interaction withtranscriptional activators, DNA unwinding and setting upthe initiation complex [54]. Much research into the impor-tant roles played by these two RNA polymerase subunitsis currently under way.

The K and major c factor of B. pertussis have both beencloned and characterised [55,56]. The major c factor of B.pertussis is larger than the Escherichia coli c70 and hasbeen designated c80 [56]. The B. pertussis RNA polymer-ase c80 subunit confers enhanced expression of fhaB in E.coli [56]. The authors suggest that the role played by c80

may be in positioning the RNA polymerase/BvgAVPcomplex at the correct site along the promoter template.By interacting directly with BvgAVP, the RNA polymer-ase c subunit is provided with an interface point throughwhich it can initiate transcription [56].

Studies on response regulator molecules related to BvgAsuch as LuxR from V. ¢scheri and UhpA from E. colihave demonstrated that while still able to bind targetDNA, transcription from promoters regulated by theseproteins is severely hindered by C-terminal deletions[23,57]. Similar experiments carried out on BvgA haveshown that a deletion of as few as two C-terminal aminoacids resulted in an avirulent phenotype [58]. When thistruncated bvgA gene was presented on a plasmid (in trans)in otherwise wild-type B. pertussis strains, a reducedgrowth rate was observed. Mutations found to remedythe slowed growth rate were traced to the K subunit ofRNA polymerase, further suggesting that an interactionbetween RNA polymerase and the C-terminus of BvgAis required for the transcriptional activation of virulence-associated genes in B. pertussis [58]. These authors postu-late that the slowed growth rate was caused by an inap-propriate interaction between the truncated BvgA protein

and the K subunit of RNA polymerase. This may have ledto a signi¢cant loss of free RNA polymerase resulting inslowed transcription of normal housekeeping genes, hencea slowed growth rate.

In the absence of BvgAVP, RNA polymerase bindsupstream of the in vivo transcription start sites of thePfha promoter leading to low-level expression from theBvgA-independent promoter, Pfha� [30,31]. However, theaddition of BvgAVP allows transcription to be initiatedat the normal in vivo site. This is further evidence thatBvgA acts upon bvg-activated promoters by correctly po-sitioning RNA polymerase for transcription [30,31].

Recent DNase I protection assays at bvg-activated pro-moters suggest that the promoter DNA £exes to looparound the BvgAVP/RNA polymerase complex[31,40,41]. This may increase the binding a¤nity of theBvgAVP/RNA polymerase complex and stabilise the en-tire transcriptional machinery [31].

3.4. Di¡erential activation of virulence factors

It has been suggested that the expression of bvg-acti-vated genes proceeds in two stages [59,60]. This di¡erentialactivation of `early' and `late' promoters allows the bacte-rium to produce adhesin molecules soon after entering thehost, whereas the `second-wave' bvg-activated genes arenot expressed until attachment is achieved.

The transcription of PT requires the presence of athreshold concentration of BvgAVP [60]. There is ac-tually a 10-fold higher concentration of BvgAVP requiredto initiate transcription at the Pptx promoter than eitherPfha or bvg P1 [40]. Demonstration of the functional di¡er-ence between early and late expressed genes came whenmutations in the C-terminus of BvgA allowed normal lev-els of FHA expression while abolishing PT expression [61].This region has been proposed as the region of the BvgAmolecule that interacts with RNA polymerase [58,61].

Binding studies at the Pptx promoter have found aninitial high-a¤nity binding site far upstream of the 335promoter motif [41]. This region includes a 20-bp directrepeat which in turn encompasses heptameric inverted re-peats [39,41]. Genetic analysis of this region has demon-strated that it is the heptanucleotide repeats that are cru-cial for promoter activation [39]. Due to the distance ofthese repeats from the actual start site, Pptx may requireBvgAVP to multimerise along the promoter templatewith weaker sites progressively bound until the site adja-cent to the RNA polymerase recognition sequence isreached.

A similar analysis has been undertaken at the Pfha pro-moter [31] at which the binding of BvgAVP to an extentmirrors that at the Pptx promoter. Again a high-a¤nity siteis bound initially, including an inverted repeat region ho-mologous to that at Pptx, however it lies further down-stream than the corresponding element at the Pptx pro-moter [31,40]. The architecture of the bvg P1 promoter,

FEMSRE 716 1-5-01


another early promoter, is very similar to that of Pfha andinteracts with BvgAVP and RNA polymerase in a similarmanner [40]. Since the expression of PT is temporallylinked to that of adenylate cyclase toxin/haemolysin it isnot surprising that the transcriptional machinery of thesetwo toxins is alike; the Pcya promoter has been shown tobehave in a manner which parallels that of Pptx [42].

Di¡erences in BvgA binding sites at the promoters ofthe two classes of virulence genes suggest that the speci¢c-ity of BvgA^RNA polymerase interactions may controldi¡erences in transcription at these promoters [31,40^42,62].

3.5. `Bvgi'

Some of the recent research involving the co-ordinateregulation by the bvg locus has been performed in theclosely related species B. bronchiseptica. While the bvg-re-pressed genes of B. pertussis were seldom studied and littleis known about their function, B. bronchiseptica has anarray of important functional bvg-repressed genes includ-ing motility, iron scavenging, urease and phosphatase ac-tivity which are at least in part repressed by the bvg locus[63^67].

Due to this expression of functional genes under mod-ulating growth conditions, B. bronchiseptica is sometimesused as a model for B. pertussis. An intermediate-phase B.bronchiseptica mutant has recently been identi¢ed whichexpresses phenotypes characteristic of both bvg-activatedand bvg-repressed states [18]. Further investigation usingpolyclonal antisera from animals infected by the inter-mediate mutant revealed that under growth conditionsdescribed as semi-modulating, B. bronchiseptica expressedelevated levels of a novel class of gene products. Thephase-locked `Bvgi' mutant has a C to T transition inthe bvgS gene resulting in the threonine at position 733of wild-type BvgS being changed to a methionine. Theclose proximity of this mutation to the putative phosphor-ylation site (His729) in the transmitter domain is thought toa¡ect kinase activity [18].

3.6. The Ris system

The recent discovery in B. bronchiseptica of a secondtwo-component regulatory system distinct from bvg hasfurthered our knowledge of genetic regulation in Bordetel-la spp. Like bvg, this system, designated the ris locus (reg-ulator of intracellular response), also consists of genesencoding a response regulator (RisA) and a sensor kinase(RisS) [68]. The ris locus is essential for bacterial resistanceto oxidative stress and the production of acid phospha-tase, as well as in vivo persistence. DNA sequence analysisof the B. pertussis and B. bronchiseptica ris locus hasshown that there is only a single base di¡erence in eachof the risA and risS genes. However, the function of ris inB. pertussis has not yet been determined.

3.7. Bvg accessory factor

The temporally di¡erential regulation employed byBvgAS to control virulence factor expression in B. pertus-sis may require the presence of accessory factors. In E.coli, the activation of either fhaB or bvgAS merely requiresthe presence of bvgAS in trans [37,69,70]. It was thoughtthat this was not su¤cient however for expression of PTor adenylate cyclase toxin [37,69,71]. When investigatingthe factors a¡ecting the expression of PT, it was foundthat the activation of the bvg locus may depend on thetopology of the DNA [72]. These authors also alluded tothe possibility that the action of an as yet undiscoveredauxiliary protein may be required to activate expressionfrom the ptx promoter [72]. Stibitz [61] proposed a modelsuggesting the presence of an accessory protein, whichinteracts with the C-terminus of BvgA and may be re-quired for cya and ptx activation. This protein has beencharacterised and termed the Bvg accessory factor (Baf).The expression of a ptx-lacZ fusion in E. coli requiresBvgAS and Baf in trans [73]. The expression from Pptx

in E. coli has also been demonstrated without the presenceof Baf, thereby contradicting the previous work and ques-tioning the need for Baf [74].

4. Virulence factors of B. pertussis

The set of genes expressed by B. pertussis which allow itto invade and persist inside the host are termed virulencefactors or virulence determinants. These include adhesinswhich facilitate attachment to target host cells and toxinswhich enable the bacterium to evade the host immunesystem. Mutagenesis studies have shown many of theseadhesins and toxins are required for in vivo persistencein animal models [75^78]. At present the consensus withregard to the strategy for developing whooping coughvaccines seems to include a few puri¢ed immunogens inan acellular, multicomponent vaccine. Many of the viru-lence factors of B. pertussis have been considered for thispurpose. Candidate antigens include FHA, serotype-spe-ci¢c ¢mbriae, pertactin, PT and adenylate cyclase toxin.Other virulence determinants of B. pertussis which haveyet to be considered as vaccine antigens include dermone-crotic toxin, tracheal cytotoxin, lipopolysaccharide (LPS),tracheal colonisation factor (Tcf) and the serum resistancelocus.

4.1. Filamentous haemagglutinin

The critical step in B. pertussis infection is attachment ofthe pathogen to host cells. FHA is consistently referred toas the major adhesin of this bacterium. It is a 220-kDasurface-associated protein secreted to the extracellular en-vironment to facilitate adherence to ciliated respiratoryepithelial cells, thereby initiating the pathogenic cycle.

FEMSRE 716 1-5-01


FHA can also work as a bridge adhesin, thereby facilitat-ing the attachment of other micro-organisms [79]. Thismight in turn explain the superinfections which usuallycomplicate the clinical course of whooping cough.

As the name suggests, FHA has a ¢lamentous structure,supported by electron microscopy studies giving the di-mensions of the molecule as 2 nm wide and 45^50 nmlong [80,81]. An extensive study by Makhov and cowork-ers [81] culminated in the proposal of a model of FHA: apolypeptide chain folded into a monomeric hairpin. Thehairpin comprises head, shaft and tail regions. The modelpredicts that whilst the head contains the N- and C-termi-ni, and the tail the important RGD (arginine-glycine-as-partic acid) sequence, the shaft is composed of tandem 19-amino acid residue repeat regions R1 (38 cycles) and R2(13 cycles) which maintain the structural integrity of themolecule.

This protein is unusual in a number of respects. Firstly,the gene for FHA, fhaB, has a coding potential of 367kDa [82,83]. Over one third of this precursor protein islater cleaved in a complex post-translational maturationprocess to unveil the biologically active polypeptide (Fig.5) [84]. FHA also possesses a number of possible bindingsites and functions. A further interesting point regardingFHA is the level of regulation involved in producing andsecreting the polypeptide.

4.1.1. Binding activitiesThe di¡erent domains of FHA encompass various bind-

ing capabilities, including integrin-mediated attachment tophagocytes, lectin-like binding to sulfated sugars found inthe extracellular matrix (ECM) and on epithelial cells, anda carbohydrate recognition domain which allows attach-ment to ciliated cells present in the respiratory epithelium.

The main target binding site for B. pertussis is the cili-ated epithelium of the respiratory tract [85]. Attachment tocells in this region is mediated by the carbohydrate bind-ing domain of FHA [67,86], which has a unique a¤nityfor glycolipids (speci¢cally lactosylceramide residues) andciliated cells [87]. This carbohydrate recognition domainwas narrowed down to amino acid residues 1141^1279using monoclonal antibodies [87]. Recent studies of thisregion have enabled high-level, soluble expression in E.coli and further characterisation of this 18-kDa polypep-tide, designated fragment A. Anti-fragment A antibodieswere shown to inhibit binding of B. pertussis to ciliatedrabbit cells in the same fashion as antibodies elicited bywhole FHA, raising speculation that with further investi-gation, the more easily obtained fragment A may replaceFHA in component vaccines against whooping cough[88].

The heparin binding activity of FHA may allow B. per-tussis to bind to non-ciliated cells as demonstrated usingWiDr cells, HeLa cells, and Vero cells [89,90], or perhapsfacilitate interactions with the ECM. The lectin-like bind-ing of FHA to heparin and dextran sulfate was character-

ised when it was shown that binding to epithelial cellscould be inhibited by the addition of heparin [91]. Hannahand associates [92] extended this research by revealing thatFHA also binds speci¢cally to sulfated, but not unchargedglycolipids and mapped the heparin binding domain toresidues 442^863 of the amino-terminal end of FHA.

The tail region of FHA includes a binding moiety spe-ci¢c for complement receptor type 3 (CR3, KML2, CD11b/CD18) integrins present on the surface of macrophages[93]. The FHA tail region contains an RGD sequencewhich is present in a number of other bacterial adhesinsand is a region of attachment in many eukaryotic integrinbinding proteins [94,95]. It has been determined that it isthis RGD tripeptide which facilitates binding to CR3 [93].B. pertussis docks with integrin CR3, which may initiateentry into phagocytes without an oxidative burst. It hassince then been shown that the binding process involvesthe formation of a leukocyte signal transduction complexwhich upregulates CR3 binding activity, suggesting thatFHA could regulate the binding activity of its own recep-tor. This leukocyte signal transduction complex is com-posed of a receptor protein called leukocyte response in-tegrin (LRI) and integrin-associated protein (CD47) [96].These authors suggest that the carbohydrate/glycoconju-gate binding capacity of FHA may also be involved inthe primary stages of the CR3^LRI^CD47 complex for-mation. This hypothesis was formulated using the exampleof leukocyte^endothelial cell recognition [97] whereby atransitory state of attachment is mediated by lectins andthis weak connection is strengthened by CR3-dependentbinding. Hazenbos and associates [98] found that a macro-phage surface ¢bronectin receptor, very late antigen-5(VLA-5), may also be involved and proposed that bindingand cross-linking of VLA-5 by B. pertussis activates CR3,facilitating attachment to FHA. Further study led to the¢nding that VLA-5 was acting as a ligand for the minor¢mbrial subunit FimD [99]. These authors suggest that¢mbriae and FHA may act co-operatively to bind to mac-rophages, leading to uptake and intracellular survival ofB. pertussis.

The advantages gained by B. pertussis in binding to`professional phagocytes' are concerned with evasion ofthe immune system and may be threefold. Severing cellularcommunication links to phagocytic cells by obstructing avital receptor site (CR3, VLA-5 or C4BP) may reduce theimpact of the immune system. Secondly, by attaching toCR3, B. pertussis has taken the preliminary step in entryand persistence in phagocytes, and `hand-to-hand combat'using CR3 as a site from which to deliver bacterial toxinsat close range may kill macrophages and neutrophils moreeconomically [100,101]. Finally, B. pertussis FHA bindsthe serum protein C4BP [102]. This protein is known asa regulator of complement activation and acts to inhibitthe classical complement pathway, ceasing the formationof the membrane attack complex. How B. pertussis is ableto utilise this binding capacity remains unclear, but once

FEMSRE 716 1-5-01


again demonstrates the extent and complexity with whichB. pertussis interacts with the human immune system.

4.1.2. Genetic regulation and biogenesisBeing a virulence determinant of B. pertussis, FHA

undergoes a high level of genetic regulation. The bvg locuscontrols expression of fhaB depending on the prevailingenvironmental conditions. When B. pertussis is in suitableconditions for colonisation, FHA is one of the ¢rst pro-teins produced, detected in a matter of minutes [60]. Pro-duction of FHA is also dependent on the expression of atleast one accessory protein, FhaC, the gene for which liesdownstream of fhaB in a cluster of accessory genes in-volved in the export of both FHA and ¢mbriae (Fig. 6)[103,104]. FhaC is a pore-forming outer membrane proteinthat interacts with FhaB allowing for secretion of the ad-hesin [105^107].

4.1.3. Post-translational maturationThe DNA sequence of the structural gene fhaB has been

determined [82,83,86]. The gene consists of approximately10.1 kb and encodes a protein of 367 kDa termed FhaB.This precursor must undergo a series of modi¢cationswhich are still only partially understood before matureFHA is secreted to the extracellular environment.

Toward the N-terminus of FhaB lies a domain of 115residues, bearing homology to regions of the haemolysinsof Serratia marcescens and Proteus mirabilis, and whichare required for secretion of these proteins [82]. Secretionof each of these proteins is also dependent on the presenceof a homologous accessory protein (FhaC in B. pertussis)[107]. Recent studies have shown that the structure ofFhaC contains 19 membrane spanning L-strands thatprobably allow the molecule to form a L-barrel in theouter membrane [106,108] (Fig. 5). An FhaC-like outer

Fig. 5. Simpli¢ed representation of the current model of FHA secretion in B. pertussis. The large signal peptide (open box), haemolysin homologous re-gion (lightly shaded box) and C-terminal domain (¢lled box) are shown. The remainder of the protein, which contains all of the binding domains, in-cludes the repeat regions (striped boxes), the RGD sequence (open box) and the immunodominant region of the molecule (heavily shaded box). Theatypical signal peptide is cleaved from FhaB probably via a leader peptidase-dependent process as it translocates to the inner plasma membrane. Thenew N-terminal glutamine residue is then converted to a pyroglutamine, indicated by the asterisk. The C-terminus is still present in the periplasmicspace, presumably to prevent incorrect folding. An interaction is then established between FhaC and the haemolysin-homologous region of FhaB. Asthe precursor molecule is translocated through the FhaC L-barrel, the C-terminal domain is cleaved. Once through the outer membrane (outer) the ma-ture FHA molecule (FHA) is then able to fold correctly into the predicted hairpin structure. Modi¢ed from Locht et al. [84] and Renauld-Mongenie etal. [112].

FEMSRE 716 1-5-01


membrane protein, HMWB, has since been found in Hae-mophilus in£uenzae, where its function is also to aid in theexport of an adhesin, HMWA [109]. The precursors ofFHA and HMWB share an extra region of homologycomprising 22 amino acid residues, followed by a posi-tively charged region and a hydrophobic segment. Thisacts as an atypical signal peptide and is cleaved from themolecule prior to export [83,105]. Although this is an un-usually large signal peptide (71 amino acid residues), it hasrecently been shown to undergo proteolysis at a leaderpeptidase (Lep) consensus cleavage site, suggesting thatFhaB crosses the inner membrane via the Sec-dependentpathway [110]. The B. pertussis lep gene has recently beencloned and characterised [111], and future studies maydetermine whether Lep is an absolute requirement forFHA secretion in B. pertussis.

The C-terminal third of the precursor protein is alsoremoved leaving mature FHA comprising the N-terminal220 kDa of FhaB. Although not part of the mature pro-tein, the C-terminus plays an important role in secretion[112] possibly as a type of intramolecular chaperone tostabilise the protein and prevent incorrect folding, allow-ing the required secretory interactions to proceed. Thusthe current model of FHA maturation and secretion in-volves the precursor FhaB translocating the inner mem-brane to the periplasmic space, probably utilising the B.pertussis leader peptidase (Lep) and Sec machinery. The115-residue, haemolysin-homologous domain then inter-acts with FhaC (Fig. 5). As the N-terminal portion passesthrough the FhaC-derived L-barrel outer membrane pore,the 150-kDa C-terminal domain is cleaved liberating ma-ture FHA, which then folds into the rigid hairpin structureproposed by Makhov et al. [81].

4.2. Serotype-speci¢c ¢mbriae

Fimbriae, also known as pili and agglutinogens, arelong ¢lamentous protrusions which extend from the bac-terial cell surface and facilitate a variety of binding capa-bilities. The ¢mbriae of B. pertussis incorporate both ma-jor and minor subunits. The major subunits form the¢mbrial strand, being grouped into pentameric repeatunits, each 13 nm in length and comprising two full helicalturns [113]. This ¢nding has been supported and extendedby more recent research [114] which also obtained an ac-curate value of 5.7 nm for the diameter of B. pertussis¢mbriae.

B. pertussis produces two distinct types of ¢mbriae, se-rotype 2 and serotype 3 (also called serotype 6). These areconstructed using major subunits, expressed from the ¢m2and ¢m3 genes respectively, and minor ¢mbrial subunit,encoded by ¢mD [103,104]. Major subunits are stacked toform the long ¢lamentous structure characteristic of ¢m-briae and the minor subunit is located at the tip [115].

4.2.1. Major subunitsThe major ¢mbrial subunits which form the structural

helices, Fim2 and Fim3, are proteins of 22 kDa and 22.5kDa respectively. The ¢m2 and ¢m3 genes which encodethese proteins have both been cloned and sequenced[116,117] and exhibit considerable homology at both thenucleotide and amino acid level. Two other pseudogenesrelated to the major ¢mbrial subunits have also beenfound in B. pertussis, ¢mX and ¢mA [104,118]. The ¢mXgene encodes a protein of approximately 20 kDa which isexpressed at very low levels if at all [119]. The other silent¢mbrial gene, ¢mA, is located in the cluster of ¢mbrial andFHA accessory genes (Fig. 6) [104]. This is a region ofDNA with sequence homology to ¢m2, ¢m3 and ¢mX,however the 5P end of the gene corresponding to the N-terminal region of FimA is absent in B. pertussis. Re-cently, it was found that B. bronchiseptica expresses a fullyintact version of FimA [120].

The low-level expression of FimA seen in B. bronchisep-tica is bvg-regulated. It has been postulated that the twofunctional major ¢mbrial subunit genes and the pseudo-gene ¢mX may have arisen via gene duplication. The dis-covery of ¢mA, its chromosomal location within the ¢m-brial operon and its expression in B. bronchiseptica suggestthat ¢mA may actually be the original, ancestral major¢mbrial subunit from which ¢m2 and ¢m3 arose [104,120].

Since the discovery of the minor subunit and adhesinFimD, the role of the major subunits in adherence hasbeen somewhat downgraded to almost being regarded asmerely the sca¡olding on which sits the important adhesinmolecule. Recently it has been shown, however, that themajor ¢mbrial subunit has an a¤nity for sulfated sugars[121], an ability shared by the other principal adhesinmolecule, FHA. The association between B. pertussis ¢m-briae and derivatives of heparan sulfate is highly depen-dent on both the number of sulfate groups present andtheir placement around the disaccharide molecule [121].The two Fim2 heparin binding regions (H1 and H2)have since been fully characterised [122]. These regions

Fig. 6. Diagrammatic representation of the ¢m/fha gene cluster. Genes involved in FHA production are shown as shaded boxes and those involved in¢mbrial production are shown as open boxes. Modi¢ed from Willems et al. [107].

FEMSRE 716 1-5-01


share sequence and structural homology with each otherand with similar heparin binding sites of ¢bronectin, aeukaryotic ECM molecule.

The major antigenic domains present on Fim2 andFim3 ¢mbriae have been identi¢ed using monoclonal anti-bodies to screen various synthetic peptides [123]. Theseepitopes were again con¢rmed as highly immunogenicwhen they were exposed to, and recognised serum anti-body from, whooping cough-infected patients [124]. Anumber of these antigenic epitopes in fact correspond tothe recently identi¢ed heparin binding domains [122].

4.2.2. Minor subunit and accessory proteinsThe other genes found in the ¢mbrial cluster (Fig. 6) are

designated ¢mB, ¢mC and ¢mD [103,104]. The proteinspredicted within this operon are similar to proteins essen-tial for ¢mbrial biogenesis in both E. coli (Pap) and Kleb-siella pneumoniae (Mrk). The genetic organisation of theentire gene cluster is very similar to the mrk operon [125],perhaps suggesting a common origin. Another similar¢mbrial gene cluster has also been found in H. in£uenzaetype b [126]. An even greater level of homology to ¢mB,¢mC and ¢mD is demonstrated by these hif genes. Themost interesting aspect of this report was the inferencethat the entire cluster represents an ancestral mobile genet-ic element and was originally inserted via a transpositionevent.

The predicted amino acid sequence of FimB is homol-ogous to the PapD superfamily of periplasmic chaperones[103,104]. A putative signal peptide sequence directs thenewly translated FimB to the periplasm, where the signalsequence is cleaved, leaving a mature polypeptide of ap-proximately 24 kDa [103,104]. Southern hybridisation ofB. pertussis, B. parapertussis and B. bronchiseptica DNAdisclosed regions homologous to ¢mB [104] in these closelyrelated species, suggesting that all three may utilise a sim-ilar process to escort ¢mbrial proteins to the cell surface.

FimC is a protein with predicted homology to manyouter membrane ¢mbrial accessory proteins, includingPapC, MrkC and HifC, all of which are essential for thebiogenesis of ¢mbriae in their respective organisms. Themolecular mass of this predicted protein is approximately91 kDa after signal peptide cleavage [103,104]. Based onhomology data, it was proposed that FimC is probablylocated in the outer membrane, involved in ¢mbrial trans-portation and anchorage [103,104].

Much of the more recent research into the ¢mbriae of B.pertussis has concentrated on the minor ¢mbrial subunitand primary adhesin, FimD. Lying directly upstream of¢mC, the ¢mD gene encodes a ¢mbrial adhesin proteinwith a molecular mass of approximately 40 kDa [127].Like its genomic neighbours, ¢mD exhibits homology togenes in the ¢mbrial clusters of H. in£uenzae (hifE), K.pneumoniae (mrkD) and E. coli (¢mH) [103,104,126,128,129]. A ¢mD homologue is also present in B. para-pertussis and B. bronchiseptica [127]. In fact, the corre-

sponding gene in B. bronchiseptica only di¡ers by a singlebase pair change. There are 20 di¡erent amino acids in theFimD polypeptide of B. parapertussis resulting from 34base pair di¡erences. The di¡erences present in theFimD adhesins of B. pertussis and B. parapertussis maycreate an evolutionary advantage for both by eliminatingbinding site competition while sharing the same humanhost [127].

Binding studies involving ¢mbrial mutant strains lack-ing FimD have demonstrated that B. pertussis binds tohuman monocytes via FimD [99]. Known to interactwith B. pertussis ¢mbriae, VLA-5 has since been con¢rmedas the FimD receptor on the surface of monocytes [130].The construction of `true' ¢mbrial mutants, devoid of bothmajor and minor ¢mbrial subunits, allowed further inves-tigation [115]. A mouse colonisation model was utilised toshow that FimD plays a vital role in colonisation. TheFim23/33 ;FimD3 mutant colonised the lungs and tracheaof mice signi¢cantly less e¤ciently than wild-type, FHA3

mutant, and Fim23/33 ;FimD� mutant strains [115].These researchers also demonstrated the binding ofFimD to the sulfated sugar heparin, also a possible targetmolecule for FHA and major ¢mbrial subunits [91,121,131].

4.2.3. Genetic regulationLike that of FHA, ¢mbrial expression in B. pertussis is

regulated at a number of levels. As with other virulencefactors employed by B. pertussis to attach to host cells, theexpression of ¢mbriae is controlled by the bvg locus. How-ever, the correct processing and secretion of ¢mbrial sub-units requires the expression of a number of accessorygenes, which are located in the ¢mbrial operon (Fig. 6).B. pertussis produces two distinct ¢mbriae: serotype 2 andserotype 3. Some strains (serotype 2,3) express both simul-taneously [132]. The serotype expressed depends on thelevel of transcription of ¢mbrial major subunit genes. Al-teration of serotype involves switching promoter activa-tion between the two ¢mbrial subunit genes in a processtermed ¢mbrial phase variation. The genes encoding themajor ¢mbrial subunits, ¢m2 and ¢m3, with which the¢mbriae are constructed have been fully characterised[116,117] as have the homologous but transcriptionallyinactive ¢mX and ¢mA genes [103,104,118]. Sequencedata have revealed a high level of homology between allof these genes which extends into the promoter regions,unveiling some interesting features of ¢mbrial promoterarchitecture (Fig. 7).

The most noticeable aspect of the ¢mbrial promoters isthe `C stretch', a long run of cytosine residues which hasbeen implicated in ¢mbrial phase variation [133]. Insertionor deletion of extra cytosine residues in this C stretch isthe source of ¢mbrial phase variation [133]. It is thoughtthat the distance between the putative 310 box and theactivator (BvgA) binding site is vital for transcriptionalactivation, and the alteration in the number of Cs in the

FEMSRE 716 1-5-01


stretch changes this distance [133]. The reason for the lackof transcription from the ¢mX pseudogene may be theshort C stretch in the promoter region.

The expression of the ¢mbrial accessory genes is alsorequired for the correct delivery and presentation of B.pertussis ¢mbriae at the cell surface [103,104]. A mutation-al analysis of the ¢m/fha gene cluster has provided evi-dence that ¢mC, ¢mD and fhaC are translationally linked.These three genes overlap slightly and there is a putativeribosome binding (Shine^Dalgarno) site directly upstreamof the ¢rst gene, ¢mC [127].

4.3. Pertactin

Also known by the aliases p.69 and OMP 69 because ofits electrophoretic mobility in SDS^PAGE, pertactin isactually a 60-kDa outer membrane protein involved inbacterial adherence [134,135]. Similar molecules are pro-duced by other members of the genus, p.70 in B. para-pertussis [136] and p.68 in B. bronchiseptica [137].

The crystal structure of pertactin has been elucidated,revealing a 16-stranded parallel L-helix with a V-shapedcross-section [138]. The secondary structure of pertactincontains two immunodominant direct repeat regions.The ¢rst, (Gly.Gly.Xaa.Xaa.Pro)5, is located directly afterthe important RGD tripeptide and the other,(Pro.Gln.Pro)5, thought to be the major immunoprotectiveepitope, lies toward the C-terminal end [139,140].

Pertactin is transcribed from the prn gene, which en-codes a 93.5-kDa polypeptide comprising 910 amino acids[139]. This precursor, termed p.93, later undergoes the re-moval of a 34-amino acid N-terminal signal peptide [135]and cleavage of a 30-kDa polypeptide (p.30) from the C-terminus. Although the precise role of p.30 is unclear, it isdetected in outer membrane fractions and is likely to beinvolved in the export of pertactin to the outer membrane[141]. In a comparison of the prn gene sequences of

B. pertussis, B. parapertussis and B. bronchiseptica, theprecursors of the pertactins were found to be extremelyhomologous and interestingly, the most highly conservedregion is the C-terminus, suggesting the functional rele-vance of p.30 to these organisms [142].

The mechanism by which pertactin promotes adherenceto eukaryotic cells is unknown and no receptor has beenfound. The amino acid sequence revealed an RGD motif(amino acids 225^227), a known integrin binding moietypresent in many bacterial adhesins such as FHA and eu-karyotic ECM proteins including ¢bronectin and vitronec-tin [93,143]. It was assumed that this region facilitatesbinding to mammalian cells. Mutants de¢cient in pertactinexpression adhered 30^40% less well to CHO (Chinesehamster ovary) cells and HeLa cells [134]. Synthetic pep-tides containing RGD derived from pertactin were foundto inhibit pertactin binding to epithelial cells and reduce B.pertussis entry into HeLa cells [134,144]. However, a re-cent study has found no role for the pertactin RGD se-quence as a mediator of eukaryote cell invasion [145].These researchers used site-directed mutagenesis (asparticacid to glutamic acid) to obtain a pertactin molecule withan RGE site rather than RGD. Strains expressing themutation displayed no di¡erence in their ability to pro-mote adhesion to HEp-2 or CHO cells. It would be usefulto determine the role of pertactin in wild-type B. pertussisinfection, which would lead to a better understanding ofwhooping cough and the vaccines which prevent it.

It has been demonstrated by a number of groups thatpertactin is an immunoprotective antigen, being used insubunit and live oral vaccines to protect mice from respi-ratory challenge with virulent B. pertussis [146^150]. Manyhuman vaccine trials suggest that the inclusion of pertactinin the formulations is imperative. Although the role ofpertactin in vaccine-based immunoprotection is unknown,preparations including pertactin have attained promisingvaccine e¤cacy results [151,152].

Fig. 7. Promoter regions of the three ¢mbrial genes found in B. pertussis. The 310 boxes and putative activator (BvgA) binding sites are indicated inbold. Dashes have been introduced to increase homology. Di¡erences between strains in the equivalent promoter regions are underlined. The strainfrom which each sequence was derived is named in parentheses. Sequence of ¢m2 (Wellcome 28 strain) is from Livey et al. [117]. Modi¢ed from Riboliet al. [119] and Willems et al. [133].

FEMSRE 716 1-5-01


4.4. Pertussis toxin

A member of the A-B bacterial toxin superfamily, PT isa 106-kDa, hexameric protein comprising ¢ve distinct sub-units [153]. PT is an ADP ribosyltransferase [154] with theability to cause a plethora of e¡ects on host cells. It is theA protomer or S1 subunit which possesses enzymatic ca-pabilities, catalysing the transfer of the ADP-ribose moietyof NAD� to a family of G proteins, which are involved insignal transduction pathways within the host cell. Thisshort-circuits the cell signalling machinery of epithelialand immune system cells by breaking down G proteininteractions [154]. The B oligomer comprises the remain-ing subunits; S2, S3, S4 and S5 make up the B oligomer inthe ratio 1:1:2:1 [153]. This region of the protein facili-tates attachment of the toxin to host cells, delivering thetoxic action of the S1 subunit [153].

The genes encoding the PT subunits are clustered to-gether in an operon typical of many other bacterial toxins(Fig. 8). Genetic analysis has revealed that each subunit istranslated separately with an amino-terminal signal se-quence which is cleaved during transport to the periplasmwhere the holotoxin is then assembled and secreted[155,156].

A major hurdle yet to be overcome in the preparation ofacellular vaccines is the inability to obtain puri¢ed anti-gens in large amounts. The incompatibility of PT geneexpression signals with respect to E. coli makes the pro-duction of PT in this organism di¤cult to achieve[157,158]. There are also problems associated with obtain-ing high yields of puri¢ed PT from virulent B. pertussis.These include the slow growth rate, fastidious nutritionalrequirements and the inherent low PT production levels ofthe bacterium. There is also the risk that the co-puri¢ca-tion of low levels of other virulence-regulated toxins mayoccur in vaccine preparations. Of the Bordetella species,only B. pertussis produces PT. A cryptic PT operon ispresent in B. parapertussis and B. bronchiseptica, howevera concentration of mutations in the promoter region leavesthese operons transcriptionally silent [159].

An expression system for pertussis holotoxin has beenthe goal of a number of groups, particularly stimulated bythe cloning and sequencing of the PT operon. The S1 sub-unit gene has upstream regions analogous to E. coli ribo-somal binding sites. The remaining subunits, however,contain novel ribosome binding site sequences upstreamof the structural genes, which bear no resemblance to

the E. coli consensus ribosome binding site. This, andthe fact that the PT operon is positively regulated by thebvg locus, are major reasons why pertussis holotoxin is yetto be expressed in E. coli. Expression of PT subunits hasbeen achieved in E. coli and Bacillus subtilis [158,160,161].Unfortunately, expression of PT subunits in E. coli causedthe majority of the protein to congregate in inclusionbodies, and not be secreted into the periplasm [158]. Theother major problem with expression of subunits for vac-cine purposes is their inability to assemble into holotoxinin vitro. The individual PT subunits have been shown toinduce antibody responses in immunised mice. However,these mice were not protected when challenged intracere-brally with B. pertussis [158]. Simultaneous expression ofall ¢ve PT subunits has been achieved in E. coli and Sal-monella typhimurium aroA [162,163]. However, varyinglevels of expression and post-translational processing, to-gether with a lack of protection in orally immunised mice,left these strains unsuitable for vaccine production. In asigni¢cant recent study the PT S1 subunit expressed inMycobacterium bovis BCG has been shown to induce pro-tection against live B. pertussis in the mouse intracerebralchallenge model [164]. This reinforces that the system ofdelivery and the induction of an appropriate immune re-sponse are very important elements of vaccine design.

E¤cient expression of pertussis holotoxin has to dateonly been achieved in Bordetella species. The major con-cern here is that PT production takes place only in virulentstrains of B. pertussis. Co-puri¢cation of other toxins andthus contamination of vaccine preparations may thereforeoccur. Lee and associates [165] have used recombinantplasmids to express PT in B. parapertussis and B. bronchi-septica. Yields were comparable to those of B. pertussisand although the majority of the PT produced was local-ised in the periplasm, PT was detected in the culture super-natant. B. parapertussis and B. bronchiseptica have alsobeen used to produce genetically altered detoxi¢ed PT[166,167]. Bacterial strains engineered to produce a safepertussis `toxoid' are very good candidates for the produc-tion of acellular vaccines. Overexpression and secretion ofgenetically detoxi¢ed PT has been achieved in B. pertussis[168] by using allelic exchange techniques to introducemultiple copies of a genetically altered PT operon intothe chromosome. Fully inducible production of PT in B.bronchiseptica has resulted in improved growth rates andPT yields when compared to B. pertussis [169]. Expressionof PT was achieved in a permanently avirulent strain of

Fig. 8. Schematic representation of the ptx/ptl operon. PT subunit genes (open boxes; S1^S5) and PT liberation genes (shaded boxes; A^I) are indi-cated. The ptx/ptl promoter (¢lled arrow; Pwt) is also depicted. Modi¢ed from Smith and Walker [171] ; Farizo et al. [173].

FEMSRE 716 1-5-01


B. bronchiseptica using mini-transposons to clone a pro-moterless PT operon in front of a strong constitutiveB. bronchiseptica promoter [170]. Unfortunately, the toxinwas not secreted to the growth medium, making this strainless than suitable as a vaccine expression system. Furtherstudy demonstrated that neither virulent nor avirulentstrains of B. bronchiseptica were capable of PT secretion[171].

The entire B. pertussis ptl operon, the products of whichdirect the secretion of PT to the culture supernatant, hasbeen cloned and sequenced [172]. There were originallythought to be eight ORFs present in the operon, desig-nated ptlA^ptlH [172]. Evidence for the presence of a ninthgene, ptlI, has since been found (Fig. 8) [173].

Partial ptl sequence data of B. parapertussis and B. bron-chiseptica can be compared to the equivalent genomic re-gion of B. pertussis [155,156,172]. Fusion proteins ex-pressed in E. coli have been used to raise polyclonalantibodies against the predicted PtlA, B, C, E, F, G andH proteins [174]. Immunoreactive bands in Western blotsof B. pertussis whole-cell extracts were only detected usingpolyclonal antibodies against PtlE, F and G. Replacementof the B. bronchiseptica Pptx promoter with the B. pertussisPptx promoter results in detectable expression of PtlF inWestern blots in B. bronchiseptica [175]. These results sug-gested that Pptx actually drives the transcription of thecomplete ptx/ptl region. This was con¢rmed when the11-kb mRNA for the entire ptx/ptl operon was detectedin B. pertussis [176]. The introduction of the B. pertussisPptx promoter together with a portion of the PT S1 sub-unit gene into B. bronchiseptica and B. parapertussis viahomologous recombination results in the expression andsecretion of this heterologous toxin [177].

4.5. Adenylate cyclase toxin/haemolysin

The adenylate cyclase-haemolysin of B. pertussis is abifunctional protein belonging to the RTX (repeat in tox-in) family. This assemblage also includes the much studiedK-haemolysin of E. coli, together with haemolysins andleukotoxins of various pathogenic Gram-negative bacteriaincluding Pasteurella haemolytica, and Actinobacillus spe-cies [178]. Both the adenylate cyclase and the haemolyticactivities have been shown to be essential for the initiationof B. pertussis infection [179]. Adenylate cyclase-haemoly-sin is a secreted polypeptide chain consisting of 1706 ami-no acid residues [180], with a molecular mass of between175 and 220 kDa [181^183].

The adenylate cyclase activity of this protein is locatedwithin the ¢rst 400 amino acids at the amino-terminal endof the protein and is activated by calmodulin [180,184].Although the high a¤nity of adenylate cyclase toxin/hae-molysin for calmodulin was initially thought to be in-volved in the translocation of the toxin into host cells,recent evidence has proven this to be incorrect [185]. Thecalmodulin is, however, crucial for the enzyme once inside

the cell, where the formation of ultraphysiological levels ofcyclic AMP is catalysed [186,187].

The remaining 1300 residues contain an independentlyfunctional haemolysin [180,188] among other structurallyand functionally important domains. A repeat domain ofadenylate cyclase toxin/haemolysin contains 41 repeats ofthe glycine- and aspartic acid-rich regions indicative of theRTX family. Homology with other RTX members sug-gests that this region is involved in host cell recognitionand calcium binding [189,190]. In fact, each repeat motifbinds a single calcium ion, leading to a major conforma-tional change and possibly the translocation of the cata-lytic portion of the molecule into host cells [191].

The gene encoding the adenylate cyclase toxin/haemoly-sin (cyaA) is contained within an operon, together with anumber of other factors involved in toxin activation andsecretion [180] (Fig. 9) [192]. Toxin secretion from B. per-tussis will only proceed with the expression of cyaB, cyaDand cyaE [193]. The ¢nal member of the cya operon, cyaC,is required to activate adenylate cyclase toxin/haemolysin,and is involved in post-translational modi¢cation of thetoxin [194] (Fig. 10). This protein causes the acylation ofadenylate cyclase toxin/haemolysin resulting in the addi-tion of a palmitoyl fatty acid residue to Lys983 of the poly-peptide chain [195] (Fig. 10).

Host cell invasion by adenylate cyclase toxin/haemolysinis not e¡ected via a receptor-mediated endocytotic path-way [196], and instead uses a specialised calcium- andtemperature-dependent entry system which is not yet fullyunderstood [197].

The use of a detoxi¢ed form of adenylate cyclase toxin/haemolysin in acellular vaccine preparations has been ad-vocated by a number of groups [198^200]. Immunisationwith adenylate cyclase toxin/haemolysin has been shownto be protective against bacterial colonisation in animalmodels [199,201,202]. Humans infected with B. pertussisare known to produce antibodies against adenylate cyclasetoxin/haemolysin to ¢ght the infection [198]. The interac-tion between adenylate cyclase toxin/haemolysin and as-pects of the human immune system has been the focus ofmany research groups. Invasion and survival of B. pertus-sis into many di¡erent immune cells is now well docu-mented [100,203,204]. The inhibition of antigen-dependentT-cell proliferation by B. pertussis has recently been dem-

Fig. 9. Operon structure of the genes encoding adenylate cyclase toxin/haemolysin and accessory proteins. The adenylate cyclase gene (lightlyshaded box; cyaA) and accessory genes cyaB, cyaD and cyaE (heavilyshaded boxes; B, D, E) are indicated. The adenylate cyclase gene pro-moter (arrow; Pcya) is depicted. The cyaC gene and the cyaB, cyaD andcyaE genes are transcribed from two other promoters (indicated by ar-rows) respectively. Modi¢ed from Goyard et al. [192].

FEMSRE 716 1-5-01


onstrated and necessitates the presence of adenylate cy-clase toxin/haemolysin [120]. The potential of recombinantadenylate cyclase toxoids as vaccine carrier proteins hasrecently been highlighted due to their ability to deliverheterologous epitopes fused to their N-terminal domaininto the cytosol of antigen presenting cells [205]. This sys-tem is still being re¢ned, however it has been used todeliver CD8+ and CD4+ T-cell epitopes to antigen pre-senting cells to induce antiviral responses [205,206]. Thismay be particularly useful for the presentation of vaccineantigens when a polarised Th1 immune response is re-quired for protection.

The role which adenylate cyclase toxin/haemolysin playsin apoptosis has been demonstrated in vitro and in vivo ina number of di¡erent cell types [207^209]. It has beenproposed that this may be a method by which B. pertussismight escape the antibacterial e¡ects of macrophages[207]. Alternatively, it could be an action facilitated bythe immune system to remove the bacterium from theirpossibly safer intracellular niche and expose it to the rig-ours of the extracellular environment, patrolled by killercells and antibodies.

4.6. Dermonecrotic toxin

Although known for many years [210], this toxin hasnot been as extensively studied as other Bordetella viru-

lence determinants. This may be due, in part, to di¤cultyin the puri¢cation process. Termed heat-labile toxin be-cause it becomes inactive at 56³C, it was given its currentname due to its ability to cause necrotic skin lesions wheninjected subcutaneously into mice [211,212]. This injectionis lethal when given intravenously. Dermonecrotic toxinsare produced by Bordetella species and are regulated bythe bvg locus [210,213^215]. The toxin is not secreted tothe extracellular environment and has been localised to thecytoplasm [216]. This and its very low expression levels areprobably the major puri¢cation problems. Later studiesagree on a molecular mass of 140 kDa and, contrary toprevious reports, a single polypeptide chain [217,218].

The gene encoding the B. pertussis dermonecrotic toxinwas sectionally cloned and sequenced [219]. The toxin canbe placed by homology and functionality into a family oftoxins including the cytotoxic necrotising factors of E. coli,botulinum C3 toxin and Pasteurella multocida toxin [220^223]. All members of this family induce dormant cells tobegin DNA synthesis, leading to either increased cell divi-sion or multinuclear cells, as is the case with dermone-crotic toxin. The mechanism of action has recently beendetermined. The C-terminal part of dermonecrotic toxindeamidates a glutamine residue (Glu-63) of Rho protein[224]. Although dermonecrotic toxin is considered a viru-lence factor, mutants de¢cient in dermonecrotic toxin donot exhibit any di¡erences in virulence in mouse studies

Fig. 10. Model for the activation of adenylate cyclase toxin/haemolysin. The domain structure of the 1706-amino acid residue toxin is shown. The ¢rst400 amino acid residues contain the adenylate cyclase moiety (open box; adenylate cyclase domain). The haemolytic activity is contained in the remain-ing 1300 residues, which are also important for delivery into host cells. This region comprises a 200-residue stretch of hydrophobic amino acids (lightlyshaded box; hydrophobic region), glycine-rich, calcium binding repeat regions (heavily shaded boxes; repeat regions), and a C-terminal signal sequenceinvolved in secretion of the molecule (¢lled box; secretion signal). The palmitoylation site is shown (Lys983). The palmitoyl residue is represented by thezigzag symbol. Modi¢ed from Goyard et al. [192].

FEMSRE 716 1-5-01


[78]. The exact role of dermonecrotic toxin in the patho-genesis of B. pertussis infection is still unknown, howeverrecent technological advances which have allowed the pu-ri¢cation of the toxin and the cloning of its gene may shedfurther light onto the role played by dermonecrotic toxin.

4.7. Tracheal cytotoxin

The major biological activity endowed by this toxin isthe ability to damage ciliated respiratory epithelial cells[225]. Tracheal cytotoxin is a disaccharide-tetrapeptide de-rivative of the cell wall peptidoglycan layer. Both structur-ally and functionally, tracheal cytotoxin falls into the mur-amyl peptide family. Gram-negative bacteria produce alayer of peptidoglycan, however only Bordetella and Neis-seria release fragments extracellularly, in the form of mur-amyl peptides, normally during exponential growth[226,227].

The destruction of cilia and ciliated epithelial cells bytracheal cytotoxin causes ciliostasis and forces the infectedindividual to cough relentlessly in order to remove mucus,a task ordinarily performed by ciliated cells. It has beendemonstrated in hamster trachea epithelial cells that thelactyl tetrapeptide portion of the molecule is responsiblefor its full toxic activity [228]. The toxicity conferred bytracheal cytotoxin is indirect, being caused by the induc-tion of host cells to produce interleukin-1 [229]. This acti-vates host cell nitric oxide synthase leading to high levelsof nitric oxide radicals [230]. It is still not certain whetherit is the tracheal cytotoxin or the interleukin-1 which stim-ulates nitric oxide synthase. The nitric oxide acts by de-stroying iron-dependent enzymes, eventually inhibiting mi-tochondrial function and DNA synthesis in nearby hostcells [229]. Since B. pertussis is attached to the ciliatedepithelial cells, it is these which undergo the most damage.Tracheal cytotoxin also has a toxic e¡ect on other cells,impairing neutrophil function at low concentrations andconferring toxic activity in larger quantities [231]. Evolu-tionarily, the production of tracheal cytotoxin by B. per-tussis may merely represent a breakdown by-product ofpeptidoglycan manufacture. However, it is considered anintegral component in the pathogenesis of B. pertussis in-fection.

4.8. Lipopolysaccharide

The LPS molecule of B. pertussis is smaller than manyother bacterial LPS structures and is therefore often re-ferred to as a lipooligosaccharide. The role which LPSplays in the pathogenesis of B. pertussis infection is un-clear; however, biological activities exhibited by B. pertus-sis LPS include antigenic and immunomodulating proper-ties [232,233]. The most common action associated withLPS is the potent endotoxin activity [234].

Present in most Gram-negative bacteria, LPS generallycomprises lipid A, core oligosaccharide and a long poly-

saccharide O antigen. The structure of B. pertussis LPSdi¡ers in that it lacks the O antigen. B. pertussis actuallyproduces two types of LPS, designated A and B [235].These molecules possess di¡ering electrophoretic mobili-ties due to the presence of an extra trisaccharide moietyon LPS-A [236]. All of the species of Bordetella expressdi¡erent LPS molecules, which may be a factor in the highlevel of species speci¢city demonstrated within this genus[237]. B. parapertussis strains isolated from humans andsheep display distinct LPS pro¢les [237] adding to thehost-speci¢c feature of this molecule [238,239]. The LPSproduced by B. bronchiseptica is similar to that of B. para-pertussis, as they both express temperature-dependent Oantigen [237,240]. The charged O antigen of B. bronchisep-tica is thought to be the element which confers protectionagainst antimicrobial peptides since B. pertussis, whichlacks O antigen, is not protected [241]. B. bronchisepticaalso produces host-speci¢c LPS molecules, with isolatesfrom dogs having more heterogeneous LPS structuresthan isolates from pigs [237]. A human isolate of B. bron-chiseptica displays a di¡erent LPS pro¢le than that of arabbit isolate [207,242]. Subsequent isolation of B. bron-chiseptica from the same human patient over 2 years hasshown a variation in LPS pro¢les during the course ofinfection [207].

Transposon mutagenesis has allowed the identi¢cationof a gene cluster encoding LPS biosynthesis in B. pertussis[243]. Many of these genes are similar to polysaccharideand LPS synthesis genes present in other bacteria, howeverthe B. pertussis LPS operon is arranged in a unique man-ner [243].

Unfortunately, our knowledge of the B. pertussis LPSremains limited and further research is necessary before weare able to de¢ne the involvement of this molecule in boththe pathogenesis of disease caused by this bacterium, andthe immunogenicity and adjuvanticity of vaccines targetedagainst whooping cough.

4.9. Tracheal colonisation factor

Originally discovered with the aid of a transposon-mu-tated strain which was de¢cient in four outer membraneproteins [244], Tcf is a bvg-regulated protein, exclusivelyproduced among the Bordetellae by B. pertussis. The geneencoding Tcf, tcfA, was ¢rst mapped using a TnphoA in-sertional mutation and was termed virulence-activatedgene #34 (vag34) [244]. The original TnphoA mutantswere 10-fold less able to colonise and persist in the tracheaof mice, but were not signi¢cantly disadvantaged in thelungs [245]. The tcfA gene has now been cloned and se-quenced, with the derived amino acid sequence predictinga 68.6-kDa precursor, the ¢rst 39 amino acids of whichcomprise a likely signal peptide, leaving a 64.4-kDa pro-tein [245]. The C-terminal half of Tcf exhibits over 50%homology to the C-terminus of the precursors of the per-tactins of Bordetella species, including the Lys-Arg puta-

FEMSRE 716 1-5-01


tive proteolytic cleavage site. Another Bordetella protein,the serum resistance factor BrkA, also possesses a C-ter-minus with signi¢cant homology to that of Tcf [246]. TheN-terminal half of Tcf contains three RGD motifs and hasa high (16.5%) proline content, which may a¡ect migrationin SDS^PAGE gels, perhaps explaining the discrepancybetween the predicted 68.6-kDa protein and the presumedV90 kDa seen in PAGE gels [245]. The authors havehypothesised that the maturation processes between per-tactin and Tcf may also share common features. Thus,their model of Tcf secretion involves the precursor mole-cule (68.6 kDa) being transported to the periplasm afterthe cleavage of the typical signal peptide. The remaining64.4-kDa precursor is then translocated across the outermembrane, like pertactin, assisted by its own C-terminalregion. Cleavage at the Lys-Arg proteolysis site then oc-curs, liberating the mature 34-kDa Tcf protein. As a con-sequence of the limited research literature on Tcf, the elu-cidation of the exact role played by this protein in thepathogenesis of pertussis will require subsequent scienti¢cscrutiny.

4.10. Serum resistance locus

Mutant strains of B. pertussis de¢cient in the expressionof the serum resistance (brk) locus are signi¢cantly lessvirulent in infant mice than wild-type strains [78]. Thesestrains have also been shown to be more susceptible tocomplement-activated killing in the presence of humanserum [246]. The brk locus, which has homologues in B.parapertussis and B. bronchiseptica, consists of two genestranscribed in opposite directions, brkA and brkB[246,247]. The proteins predicted by these genes areBrkA, 103 kDa, and BrkB, 32 kDa, both of which arerequired for serum resistance. BrkA, translated as a pre-cursor molecule, is proteolytically cleaved to form 73-kDaand 30-kDa products. This cleavage pattern is reminiscentof pertactin, the protein to which BrkA is most closelyrelated, and Tcf [245]. The presence of two RGD motifsand possible roles in adherence and invasion also indicatethe similarity between these proteins.

A recent study has demonstrated that B. pertussis clin-ical isolates were serum-resistant, suggesting that circulat-ing strains gain an advantage in vivo by evading comple-ment-activated killing. It was also shown that strainscarrying an extra genetic copy of the brk locus exhibitedincreased levels of serum resistance [248].

5. Elicitation of protective immune responses againstB. pertussis

The development of whooping cough vaccines has con-centrated on the elicitation of high antibody titres andprotective immunity, and on the reduction of side-e¡ects.Recently, a number of issues have been brought to light

which also bear serious consideration in the preparation ofwhooping cough vaccines.

5.1. Cell-mediated immunity

One of the major challenges in developing superior vac-cines is our limited understanding of the mechanisms in-volved in the development of immunity against pertussisfollowing either natural infection or vaccination. Whichcomponents of the human immune system become primedto ¢ght B. pertussis infection after vaccination? Defenceagainst intracellular pathogens is usually dependent on theactivation of a strong Th1 helper response. B. pertussis hasbeen shown to invade an increasing number of cell types[100,249,250] and is now considered by some a facultativeintracellular pathogen. Intracellular bacteria may be pro-tected from host clearance mechanisms resulting in theinfection persisting. It is therefore reasonable to assumethat cell-mediated immunity (CMI) may play a key rolein eliminating B. pertussis infection, at the very least byfacilitating the release of bacteria surviving within phago-cytes. Studies have demonstrated the importance of CMIin natural and vaccine-based immunity [251,252]. Immu-nisation of mice with whole-cell vaccine or recovery fromnatural B. pertussis infection induces a Th1 response,whereas immunisation with an acellular vaccine elicits aTh2 response [252]. Production of interleukin-12 is in-duced by native infection or whole-cell vaccination, butnot by acellular vaccination [253].

Recent studies have shown that protection after respi-ratory challenge in the murine model is an excellent cor-relate for e¤cacy in humans of acellular and whole-cellvaccines, and that humoral and cell-mediated immunityplay complementary roles in vaccine-elicited protection[254]. Using di¡erent knockout mice, Mills and associates[255] found that both humoral and cell-mediated responsesare required for the stimulation of protective immunityfollowing vaccination. The initial response may be anti-body-mediated and is involved in restricting the extentof infection and minimising the damage to epithelial andimmune cells. To achieve complete bacterial clearancehowever, the induction of a CMI response is also required.The Th2 response elicited after immunisation with an acel-lular vaccine results in delayed bacterial clearance in themurine respiratory challenge model. In contrast, naturalinfection or immunisation with killed whole cells, inducinga Th1 response, was associated with rapid clearance [255].However, analyses of the CMI responses of children inrecent vaccine e¤cacy trials revealed that acellular vac-cines induce Th1 responses [256,257] equivalent to re-sponses exhibited after natural infection [258].

5.2. Development of vaccine resistance

Despite the time spent on research into mechanisms ofimmunity, e¤cacy and reactogenicity of vaccines against

FEMSRE 716 1-5-01


B. pertussis, the bacterium itself has been ¢ghting back.Recent data from The Netherlands reported a dramaticincrease in the incidence of pertussis in that country in1996, despite high levels of immunisation [259]. An exten-sive study has traced the problem to the changing popu-lation structure of B. pertussis itself [260]. The antigenicvariation of the protective antigens pertactin and the S1subunit of PT was analysed using circulating clinical iso-lates collected over a long period. Three di¡erent variantseach of pertactin and S1 were discovered and the presenceof these variants showed temporal shifts. Polymorphism inpertactin was found to occur in the midst of a repeatregion (Gly.Gly.Xaa.Xaa.Pro)5 termed region 1. The mu-tations in the S1 subunit were similarly con¢ned to a smallregion of the polypeptide. The mutations in S1 and per-tactin are concentrated in regions de¢ned as T-cell and B-cell epitopes respectively [140,261]. This highlights the im-portance of maintaining a constant global monitoring ofcirculating isolates and the necessity to adapt present for-mulations according to temporal antigenic shifts.

6. Concluding remarks

B. pertussis is exquisitely ¢ne-tuned to inhabit the hu-man respiratory tract. This has been achieved via an evo-lutionary process. The twentieth century had caught upwith this pathogen somewhat, with the advent of antimi-crobials and immunisation, and it was on the way out.Now it is again evolving and re-emerging as a dangerouspathogen. Together with the attention to vaccine reacto-genicity, the induction of appropriate immune responsemechanisms and raising the protective e¤cacy of futureanti-pertussis vaccines, we must also consider the rapidevolution of B. pertussis in the development of futurewhooping cough vaccines.

An exciting development in B. pertussis research is theongoing genome sequencing project of this bacterium bythe Sanger Centre, details of which are available on theirwebsite [262]. This will undoubtedly facilitate the develop-ment of better vaccines against whooping cough.

References

[1] World Health Organization (1999) The World Health Report, 1999:Making a Di¡erence. World Health Organization, Geneva.

[2] Bordet, J. and Gengou, O. (1906) Le microbe de la coqueluche. Ann.Inst. Pasteur (Paris) 20, 731^741.

[3] Lopez, M.M. (1952) El genero Bordetella. Microbiol. Esp. 5, 177^183.

[4] van der Zee, A., Mooi, F., van Embden, J. and Musser, J. (1997)Molecular evolution and host adaptation of Bordetella spp. ^ phylo-genetic analysis using multilocus enzyme electrophoresis and typingwith three insertion sequences. J. Bacteriol. 179, 6609^6617.

[5] Arico, B. and Rappuoli, R. (1987) Bordetella parapertussis and Bor-detella bronchiseptica contain transcriptionally silent pertussis toxingenes. J. Bacteriol. 169, 2847^2853.

[6] Cullinane, L.C., Alley, M.R., Marshall, R.B. and Manktelow, B.W.(1987) Bordetella parapertussis from lambs. NZ Vet. J. 35, 175.

[7] Goodnow, R.A. (1980) Biology of Bordetella bronchiseptica. Micro-biol. Rev. 44, 722^738.

[8] Switzer, W.P. (1956) Studies on infectious atrophic rhinitis. V. Con-cept that several agents may cause turbinate atrophy. Am. J. Vet.Res. 17, 478^484.

[9] Thompson, H., McCandlish, I.A. and Wright, N.G. (1976) Experi-mental respiratory disease in dogs due to Bordetella bronchiseptica.Res. Vet. Sci. 20, 16^23.

[10] Gross, R., Arico, B. and Rappuoli, R. (1989) Genetics of pertussistoxin. Mol. Microbiol. 3, 119^124.

[11] Kersters, K., Hinz, K.-H., Hertle, A., Segers, P., Lievens, A., Sieg-mann, O. and De Ley, J. (1984) Bordetella avium sp. nov., isolatedfrom the respiratory tracts of turkeys and other birds. Int. J. Syst.Bacteriol. 34, 56^70.

[12] Cookson, B.T., Vandamme, P., Carlson, L.C., Larson, A.M., Shef-¢eld, J.V.L., Kersters, K. and Spach, D.H. (1994) Bacteremia causedby a novel Bordetella species, B. hinzii. J. Clin. Microbiol. 32, 2569^2571.

[13] Weyant, R.S., Hollis, D.G., Weaver, R.E., Amin, M.F.M., Steiger-walt, A.G., Oconnor, S.P., Whitney, A.M., Daneshvar, M.I., Moss,C.W. and Brenner, D.J. (1995) Bordetella holmesii sp. nov., a newGram-negative species associated with septicemia. J. Clin. Microbiol.33, 1^7.

[14] van Damme, P., Heyndrickx, M., Vancanneyt, M., Hoste, B., Devos,P., Falsen, E., Kersters, K. and Hinz, K.H. (1996) Bordetella trem-atum sp. nov., isolated from wounds and ear infections in humans,and reassessment of Alcaligenes denitri¢cans Ruger and Tan 1983.Int. J. Syst. Bacteriol. 46, 849^858.

[15] van Damme, P., Hommez, J., Vancanneyt, M., Monsieurs, M.,Hoste, B., Cookson, B., Wirsing von Konig, C.H., Kersters, K.and Blackall, P.J. (1995) Bordetella hinzii sp. nov., isolated frompoultry and humans. Int. J. Syst. Bacteriol. 45, 37^45.

[16] Leslie, P.H. and Gardner, A.D. (1931) The phases of Haemophiluspertussis. J. Hyg. 31, 423^434.

[17] Lacey, B.W. (1960) Antigenic modulation of Bordetella pertussis.J. Hyg. 58, 57^93.

[18] Cotter, P.A. and Miller, J.F. (1997) A mutation in the Bordetellabronchiseptica bvgS gene results in reduced virulence and increasedresistance to starvation, and identi¢es a new class of bvg-regulatedantigens. Mol. Microbiol. 24, 671^685.

[19] Melton, A.R. and Weiss, A.A. (1993) Characterization of environ-mental regulators of Bordetella pertussis. Infect. Immun. 61, 807^815.

[20] Weiss, A.A. and Falkow, S. (1984) Genetic analysis of phase changein Bordetella pertussis. Infect. Immun. 43, 263^269.

[21] Parkinson, J.S. and Kofoid, E.C. (1992) Communication modules inbacterial signalling proteins. Annu. Rev. Genet. 26, 71^112.

[22] Chang, C., Kwok, S.F., Bleecker, A.B. and Meyerowitz, E.M. (1993)Arabidopsis ethylene-response gene ETR1 : similarity of product oftwo-component regulators. Science 262, 539^544.

[23] Choi, S.H. and Greenberg, E.P. (1991) The C-terminal region of theVibrio ¢scheri LuxR protein contains an inducer-independent luxgene activating domain. Proc. Natl. Acad. Sci. USA 88, 11115^11119.

[24] Ota, I.M. and Varshavsky, A. (1993) A yeast protein similar to bac-terial two-component regulators. Science 262, 566^569.

[25] Arico, B., Miller, J.F., Roy, C., Stibitz, S., Monack, D., Falkow, S.,Gross, R. and Rappuoli, R. (1989) Sequences required for expressionof Bordetella pertussis virulence factors share homology with pro-karyotic signal transduction proteins. Proc. Natl. Acad. Sci. USA86, 6671^6675.

[26] Coote, J.G. (1991) Antigenic switching and pathogenicity: environ-mental e¡ects on virulence gene expression in Bordetella pertussis.J. Gen. Microbiol. 137, 2493^2503.

[27] Merkel, T.J., Barros, C. and Stibitz, S. (1998) Characterisation of thebvgR locus of Bordetella pertussis. J. Bacteriol. 180, 1682^1690.

FEMSRE 716 1-5-01


[28] Roy, C.R., Miller, J.F. and Falkow, S. (1990) Autogenous regulationof the Bordetella pertussis bvgABC operon. Proc. Natl. Acad. Sci.USA 87, 3763^3767.

[29] Scarlato, V., Prugnola, A., Arico, B. and Rappuoli, R. (1990) Pos-itive transcriptional feedback at the bvg locus controls expression ofvirulence factors in Bordetella pertussis. Proc. Natl. Acad. Sci. USA87, 6753^6757.

[30] Ste¡en, P., Goyard, S. and Ullmann, A. (1996) Phosphorylated BvgAis su¤cient for transcriptional activation of virulence-regulated genesin Bordetella pertussis. EMBO J. 15, 102^109.

[31] Boucher, P.E., Murakami, K., Ishihama, A. and Stibitz, S. (1997)Nature of DNA binding and RNA polymerase interaction of theBordetella pertussis BvgA transcriptional activator at the fha pro-moter. J. Bacteriol. 179, 1755^1763.

[32] Stibitz, S. and Yang, M.S. (1991) Subcellular localization and immu-nological detection of proteins encoded by the vir locus of Bordetellapertussis. J. Bacteriol. 173, 4288^4296.

[33] Boucher, P.E., Menozzi, F.D. and Locht, C. (1994) The modulararchitecture of bacterial response regulators. Insights into the activa-tion mechanism of the BvgA transactivator of Bordetella pertussis.J. Mol. Biol. 241, 363^377.

[34] Kahn, D. and Ditta, G. (1991) Modular structure of FixJ: homologyof the transcriptional activator domain with the 335 binding domainof sigma factors. Mol. Microbiol. 5, 987^997.

[35] Choi, S.H. and Greenberg, E.P. (1992) Genetic dissection of DNAbinding and luminescence gene activation by the Vibrio ¢scheri LuxRprotein. J. Bacteriol. 174, 4064^4069.

[36] Helmann, J.D. and Chamberlin, M.J. (1988) Structure and functionof bacterial sigma factors. Annu. Rev. Biochem. 57, 839^872.

[37] Roy, C.R., Miller, J.F. and Falkow, S. (1989) The bvgA gene ofBordetella pertussis encodes a transcriptional activator required forcoordinate regulation of several virulence genes. J. Bacteriol. 171,6338^6344.

[38] Roy, C.R. and Falkow, S. (1991) Identi¢cation of Bordetella pertussisregulatory sequences required for transcriptional activation of thefhaB gene and autoregulation of the bvgAS operon. J. Bacteriol.173, 2385^2392.

[39] Marques, R.R. and Carbonetti, N.H. (1997) Genetic analysis of per-tussis toxin promoter activation in Bordetella pertussis. Mol. Micro-biol. 24, 1215^1224.

[40] Zu, T., Manetti, R., Rappuoli, R. and Scarlato, V. (1996) Di¡erentialbinding of BvgA to two classes of virulence genes of Bordetella per-tussis directs promoter selectivity by RNA polymerase. Mol. Micro-biol. 21, 557^565.

[41] Boucher, P.E. and Stibitz, S. (1995) Synergistic binding of RNApolymerase and BvgA phosphate to the pertussis toxin promoter ofBordetella pertussis. J. Bacteriol. 177, 6486^6491.

[42] Karimova, G., Bellalou, J. and Ullmann, A. (1996) Phosphorylation-dependent binding of BvgA to the upstream region of the cyaA geneof Bordetella pertussis. Mol. Microbiol. 20, 489^496.

[43] Uhl, M.A. and Miller, J.F. (1994) Autophosphorylation and phos-photransfer in the Bordetella pertussis BvgAS signal transductioncascade. Proc. Natl. Acad. Sci. USA 91, 1163^1167.

[44] Beier, D., Schwarz, B., Fuchs, T.M. and Gross, R. (1995) In vivocharacterization of the unorthodox BvgS two-component sensor pro-tein of Bordetella pertussis. J. Mol. Biol. 248, 596^610.

[45] Manetti, R., Arico, B., Rappuoli, R. and Scarlato, V. (1994) Muta-tions in the linker region of BvgS abolish response to environmentalsignals for the regulation of the virulence factors in Bordetella per-tussis. Gene 150, 123^127.

[46] Uhl, M.A. and Miller, J.F. (1996) Central role of the BvgS receiver asa phosphorylated intermediate in a complex two-component phos-phorelay. J. Biol. Chem. 271, 33176^33180.

[47] Uhl, M.A. and Miller, J.F. (1996) A new type of phosphodonor intwo-component signal transduction systems: the Bordetella pertussisBvgAS phosphorelay. EMBO J. 15, 1028^1036.

[48] Merkel, T.J. and Stibitz, S. (1995) Identi¢cation of a locus required

for the regulation of bvg-repressed genes in Bordetella pertussis.J. Bacteriol. 177, 2727^2736.

[49] Beattie, D.T., Knapp, S. and Mekalanos, J.J. (1990) Evidence thatmodulation requires sequences downstream of the promoters of twovir-repressed genes of Bordetella pertussis. J. Bacteriol. 172, 6997^7004.

[50] Beattie, D.T., Mahan, M.J. and Mekalanos, J.J. (1993) Repressorbinding to a regulatory site in the DNA coding sequence is su¤cientto confer transcriptional regulation of the vir-repressed genes (vrggenes) in Bordetella pertussis. J. Bacteriol. 175, 519^527.

[51] Merkel, T.J., Stibitz, S., Keith, J.M., Leef, M. and Shahin, R. (1998)Contribution of regulation by the bvg locus to respiratory infection ofmice by Bordetella pertussis. Infect. Immun. 66, 4367^4373.

[52] Busby, S. and Ebright, R.H. (1994) Promoter structure, promoterrecognition, and transcription activation in prokaryotes. Cell 79,743^746.

[53] Ishihama, A. (1992) Role of RNA polymerase K subunit in transcrip-tion activation. Mol. Microbiol. 6, 3283^3288.

[54] Ishihama, A. (1993) Protein-protein communication within the tran-scription apparatus. J. Bacteriol. 175, 2483^2489.

[55] Carbonetti, N.H., Fuchs, T.M., Patamawenu, A.A., Irish, T.J., Dep-pisch, H. and Gross, R. (1994) E¡ect of mutations causing overex-pression of RNA polymerase alpha subunit on regulation of virulencefactors in Bordetella pertussis. J. Bacteriol. 176, 7267^7273.

[56] Ste¡en, P., Goyard, S. and Ullmann, A. (1997) The Bordetella per-tussis sigma subunit of RNA polymerase confers enhanced expressionof fha in Escherichia coli. Mol. Microbiol. 23, 945^954.

[57] Webber, C.A. and Kadner, R.J. (1995) Action of receiver and acti-vator modules of UhpA in transcriptional control of the Escherichiacoli sugar phosphate transport system. Mol. Microbiol. 15, 883^893.

[58] Stibitz, S. (1998) Mutations a¡ecting the alpha subunit of Bordetellapertussis RNA polymerase suppress growth inhibition conferred byshort C-terminal deletions of the response regulator BvgA. J. Bacter-iol. 180, 2484^2492.

[59] Scarlato, V. and Rappuoli, R. (1991) Di¡erential response of the bvgvirulence regulon of Bordetella pertussis to MgSO4 modulation.J. Bacteriol. 173, 7401^7404.

[60] Scarlato, V., Arico, B., Prugnola, A. and Rappuoli, R. (1991) Se-quential activation and environmental regulation of virulence genesin Bordetella pertussis. EMBO J. 10, 3971^3975.

[61] Stibitz, S. (1994) Mutations in the bvgA gene of Bordetella pertussisthat di¡erentially a¡ect regulation of virulence determinants. J. Bac-teriol. 176, 5615^5621.

[62] Karimova, G. and Ullmann, A. (1997) Characterisation of DNAbinding sites for the BvgA protein of Bordetella pertussis. J. Bacteriol.179, 3790^3792.

[63] Akerley, B.J., Monack, D.M., Falkow, S. and Miller, J.F. (1992) ThebvgAS locus negatively controls motility and synthesis of £agella inBordetella bronchiseptica. J. Bacteriol. 174, 980^990.

[64] Chhatwal, G.S., Walker, M.J., Yan, H.R., Timmis, K.N. and Guz-man, C.A. (1997) Temperature dependent expression of an acid phos-phatase by Bordetella bronchiseptica-role in intracellular survival. Mi-crob. Pathogen. 22, 257^264.

[65] Giardina, P.C., Foster, L.A., Musser, J.M., Akerley, B.J., Miller, J.F.and Dyer, D.W. (1995) Bvg repression of alcaligin synthesis in Bor-detella bronchiseptica is associated with phylogenetic lineage. J. Bac-teriol. 177, 6058^6063.

[66] McMillan, D.J., Shojaei, M., Chhatwal, G.S., Guzman, C.A. andWalker, M.J. (1996) Molecular analysis of the bvg-repressed ureaseof Bordetella bronchiseptica. Microb. Pathogen. 21, 379^394.

[67] Tuomanen, E., Towbin, H., Rosenfelder, G., Braun, D., Larson, G.,Hansson, G.C. and Hill, R. (1988) Receptor analogs and monoclonalantibodies that inhibit adherence of Bordetella pertussis to humanciliated respiratory epithelial cells. J. Exp. Med. 168, 267^277.

[68] Jungnitz, H., West, N.P., Walker, M.J., Chhatwal, G.S. and Guz-man, C.A. (1998) A second two-component regulatory system of

FEMSRE 716 1-5-01


Bordetella bronchiseptica required for bacterial resistance to oxidativestress, production of acid phosphatase, and in vivo persistence. In-fect. Immun. 66, 4640^4650.

[69] Miller, J.F., Roy, C.R. and Falkow, S. (1989) Analysis of Bordetellapertussis virulence gene regulation by use of transcriptional fusions inEscherichia coli. J. Bacteriol. 171, 6345^6348.

[70] Stibitz, S., Weiss, A.A. and Falkow, S. (1988) Genetic analysis of aregion of the Bordetella pertussis chromosome encoding ¢lamentoushemagglutinin and the pleiotropic regulatory locus vir. J. Bacteriol.170, 2904^2913.

[71] Goyard, S. and Ullmann, A. (1991) Analysis of Bordetella pertussiscya operon regulation by use of cya-lac fusions. FEMS Microbiol.Lett. 61, 251^256.

[72] Scarlato, V., Arico, B. and Rappuoli, R. (1993) DNA topology af-fects transcriptional regulation of the pertussis toxin gene of Borde-tella pertussis in Escherichia coli and in vitro. J. Bacteriol. 175, 4764^4771.

[73] DeShazer, D., Wood, G.E. and Friedman, R.L. (1995) Identi¢cationof a Bordetella pertussis regulatory factor required for transcriptionof the pertussis toxin operon in Escherichia coli. J. Bacteriol. 177,3801^3807.

[74] Uhl, M.A. and Miller, J.F. (1995) BvgAS is su¤cient for activationof the Bordetella pertussis ptx locus in Escherichia coli. J. Bacteriol.177, 6477^6485.

[75] Gross, M.K., Au, D.C., Smith, A.L. and Storm, D.R. (1992) Tar-geted mutations that ablate either the adenylate cyclase or hemolysinfunction of the bifunctional cyaA toxin of Bordetella pertussis abolishvirulence. Proc. Natl. Acad. Sci. USA 89, 4898^4902.

[76] Mooi, F.R., Jansen, W.H., Brunings, H., Gielen, H., van der Heide,H.G., Walvoort, H.C. and Guinee, P.A. (1992) Construction andanalysis of Bordetella pertussis mutants defective in the productionof ¢mbriae. Microb. Pathogen. 12, 127^135.

[77] Weiss, A.A., Hewlett, E.L., Myers, G.A. and Falkow, S. (1984) Per-tussis toxin and extracytoplasmic adenylate cyclase as virulence fac-tors of Bordetella pertussis. J. Infect. Dis. 150, 219^222.

[78] Weiss, A.A. and Goodwin, M.S. (1989) Lethal infection by Bordetellapertussis mutants in the infant mouse model. Infect. Immun. 57,3757^3764.

[79] Tuomanen, E. (1986) Piracy of adhesins: attachment of superinfect-ing pathogens to respiratory cilia by secreted adhesins of Bordetellapertussis. Infect. Immun. 54, 905^908.

[80] Arai, H. and Sato, Y. (1976) Separation and characterisation of twodistinct haemagglutinins contained in puri¢ed leukocytosis promotingfactor from Bordetella pertussis. Biochim. Biophys. Acta 444, 765^782.

[81] Makhov, A.M., Hannah, J.H., Brennan, M.J., Trus, B.L., Kocsis, E.,Conway, J.F., Wing¢eld, P.T., Simon, M.N. and Steven, A.C. (1994)Filamentous hemagglutinin of Bordetella pertussis. A bacterial adhe-sin formed as a 50-nm monomeric rigid rod based on a 19-residuerepeat motif rich in beta strands and turns. J. Mol. Biol. 241, 110^124.

[82] Delisse-Gathoye, A., Locht, C., Jacob, F., Raaschou-Nielsen, M.,Heron, I., Ruelle, J.L., de Wilde, M. and Cabezon, T. (1990) Cloning,partial sequence, expression, and antigenic analysis of the ¢lamentoushemagglutinin gene of Bordetella pertussis. Infect. Immun. 58, 2895^2905.

[83] Domenighini, M., Relman, D., Capiau, C., Falkow, S., Prugnola,A., Scarlato, V. and Rappuoli, R. (1990) Genetic characterizationof Bordetella pertussis ¢lamentous haemagglutinin: a protein pro-cessed from an unusually large precursor. Mol. Microbiol. 4, 787^800.

[84] Locht, C., Bertin, P., Menozzi, F.D. and Renauld, G. (1993) The¢lamentous haemagglutinin, a multifaceted adhesion produced byvirulent Bordetella spp. Mol. Microbiol. 9, 653^660.

[85] Tuomanen, E. and Weiss, A. (1985) Characterization of two adhesinsof Bordetella pertussis for human ciliated respiratory-epithelial cells.J. Infect. Dis. 152, 118^125.

[86] Relman, D.A., Domenighini, M., Tuomanen, E., Rappuoli, R. andFalkow, S. (1989) Filamentous hemagglutinin of Bordetella pertus-sis : nucleotide sequence and crucial role in adherence. Proc. Natl.Acad. Sci. USA. 86, 2637^2641.

[87] Prasad, S.M., Yin, Y., Rodzinski, E., Tuomanen, E.I. and Masure,H.R. (1993) Identi¢cation of a carbohydrate recognition domain in¢lamentous hemagglutinin from Bordetella pertussis. Infect. Immun.61, 2780^2785.

[88] Liu, D.-F., Phillips, E., Wizemann, T.M., Siegel, M.M., Tabei, K.,Cowell, J.L. and Tuomanen, E. (1997) Characterisation of a re-combinant fragment that contain a carbohydrate recognition do-main of the ¢lamentous haemagglutinin. Infect. Immun. 65, 3465^3468.

[89] Sato, Y., Izumiya, K., Sato, H., Cowell, J.L. and Manclark, C.R.(1981) Role of antibody to leukocytosis-promoting factor haemag-glutinin and to ¢lamentous haemagglutinin in immunity to pertussis.Infect. Immun. 31, 1223^1231.

[90] Urisu, A., Cowell, J.L. and Manclark, C.R. (1986) Filamentoushemagglutinin has a major role in mediating adherence of Bordetellapertussis to human WiDr cells. Infect. Immun. 52, 695^701.

[91] Menozzi, F.D., Boucher, P.E., Riveau, G., Gantiez, C. and Locht,C. (1994) Surface-associated ¢lamentous hemagglutinin induces au-toagglutination of Bordetella pertussis. Infect. Immun. 62, 4261^4269.

[92] Hannah, J.H., Menozzi, F.D., Renauld, G., Locht, C. and Brennan,M.J. (1994) Sulfated glycoconjugate receptors for the Bordetellapertussis adhesin ¢lamentous hemagglutinin (FHA) and mappingof the heparin-binding domain on FHA. Infect. Immun. 62, 5010^5019.

[93] Relman, D., Tuomanen, E., Falkow, S., Golenbock, D.T., Saukko-nen, K. and Wright, S.D. (1990) Recognition of a bacterial adhesionby an integrin: macrophage CR3 (alpha M beta 2, CD11b/CD18)binds ¢lamentous hemagglutinin of Bordetella pertussis. Cell 61,1375^1382.

[94] Hynes, R.O. (1992) Integrins: versatility, modulation, and signal-ling. Cell 69, 11^25.

[95] Ruoslahti, E. and Pierschbacher, M.D. (1986) Arg-Gly-Asp: a ver-satile cell recognition signal. Cell 238, 491^497.

[96] Ishibashi, Y., Claus, S. and Relman, D.A. (1994) Bordetella pertussis¢lamentous hemagglutinin interacts with a leukocyte signal trans-duction complex and stimulates bacterial adherence to monocyteCR3 (CD11b/CD18). J. Exp. Med. 180, 1225^1233.

[97] Butcher, E.C. (1991) Leukocyte-endothelial cell recognition: three(or more) steps to speci¢city and diversity. Cell 67, 1033^1042.

[98] Hazenbos, W.L., van den Berg, B.M. and van Furth, R. (1993) Verylate antigen-5 and complement receptor type 3 cooperatively medi-ate the interaction between Bordetella pertussis and human mono-cytes. J. Immunol. 151, 6274^6282.

[99] Hazenbos, W.L., Geuijen, C.A., van den Berg, B.M., Mooi, F.R.and van Furth, R. (1995) Bordetella pertussis ¢mbriae bind to hu-man monocytes via the minor ¢mbrial subunit FimD. J. Infect. Dis.171, 924^929.

[100] Saukkonen, K., Cabellos, C., Burroughs, M., Prasad, S. and Tuo-manen, E. (1991) Integrin-mediated localization of Bordetella per-tussis within macrophages: role in pulmonary colonization. J. Exp.Med. 173, 1143^1149.

[101] Steed, L.L., Setareh, M. and Friedman, R.L. (1991) Intracellularsurvival of virulent Bordetella pertussis in human polymorphonu-clear leukocytes. J. Leukoc. Biol. 50, 321^330.

[102] Bergga®rd, K., Johnsson, E., Mooi, F.R. and Lindahl, G. (1997)Bordetella pertussis binds the human complement regulator C4BP:role of ¢lamentous haemagglutinin. Infect. Immun. 65, 3638^3643.

[103] Locht, C., Geo¡roy, M.C. and Renauld, G. (1992) Common acces-sory genes for the Bordetella pertussis ¢lamentous hemagglutininand ¢mbriae share sequence similarities with the papC and papDgene families. EMBO J. 11, 3175^3183.

[104] Willems, R.J., van der Heide, H.G. and Mooi, F.R. (1992) Charac-

FEMSRE 716 1-5-01


terization of a Bordetella pertussis ¢mbrial gene cluster which islocated directly downstream of the ¢lamentous haemagglutiningene. Mol. Microbiol. 6, 2661^2671.

[105] Jacob-Dubuisson, F., Buisine, C., Mielcarek, N., Clement, E., Me-nozzi, F.D. and Locht, C. (1996) Amino-terminal maturation of theBordetella pertussis ¢lamentous hemagglutinin. Mol. Microbiol. 19,65^78.

[106] Jacob-Dubuisson, F., Buisine, C., Mielcarek, N., Clement, E., Me-nozzi, F.D. and Locht, C. (1999) Channel formation by FhaC, theouter membrane protein involved in the secretion of the Bordetellapertussis ¢lamentous haemagglutinin. J. Biol. Chem. 274, 37731^37735.

[107] Willems, R.J., Geuijen, C., van der Heide, H.G., Renauld, G., Ber-tin, P., van den Akker, W.M., Locht, C. and Mooi, F.R. (1994)Mutational analysis of the Bordetella pertussis ¢m/fha gene cluster:identi¢cation of a gene with sequence similarities to haemolysinaccessory genes involved in export of FHA. Mol. Microbiol. 11,337^347.

[108] Guedin, S., Willery, E., Tommassen, J., Fort, E., Drobecq, H.,Locht, C. and Jacob-Dubuisson, F. (2000) Novel topological fea-tures of FhaC, the outer membrane transporter involved in thesecretion of the Bordetella pertussis ¢lamentous haemagglutinin.J. Biol. Chem. 275, 30202^30210.

[109] Barenkamp, S.J. and St Geme III, J.W. (1994) Genes encoding highmolecular weight adhesion proteins of non-typeable Haemophilusin£uenzae are part of gene clusters. Infect. Immun. 62, 3320^3328.

[110] Lambert-Buisine, C., Willery, E., Locht, C. and Jacob-Dubuisson,F. (1998) N-terminal characterization of the Bordetella pertussis ¢l-amentous haemagglutinin. Mol. Microbiol. 28, 1283^1293.

[111] Smith, A.M., Yan, H., Groves, N., Dalla Pozza, T. and Walker,M.J. (2000) Co-expression of the Bordetella pertussis leader pepti-dase I results in enhanced processing and expression of the pertussistoxin S1 subunit in Escherichia coli. FEMS Microbiol. Lett. 191,177^182.

[112] Renauld-Mongenie, G., Mielcarek, N., Cornette, J., Schacht, A.-M.,Capron, A., Riveau, G. and Locht, C. (1996) Induction of mucosalimmune responses against a heterologous antigen fused to ¢lamen-tous hemagglutinin after intranasal immunisation with recombinantBordetella pertussis. Proc. Natl. Acad. Sci. USA 93, 7944^7949.

[113] Steven, A.C., Bisher, M.E., Trus, B.L., Thomas, D., Zhang, J.M.and Cowell, J.L. (1986) Helical structure of Bordetella pertussis ¢m-briae. J. Bacteriol. 167, 968^974.

[114] Heck, D.V., Trus, B.L. and Steven, A.C. (1996) Three-dimensionalstructure of Bordetella pertussis ¢mbriae. J. Struct. Biol. 116, 264^269.

[115] Geuijen, C.A., Willems, R.J., Bongaerts, M., Top, J., Gielen, H. andMooi, F.R. (1997) Role of the Bordetella pertussis minor ¢mbrialsubunit, FimD, in colonization of the mouse respiratory tract. In-fect. Immun. 65, 4222^4228.

[116] Mooi, F.R., ter Avest, A. and van der Heide, H.G. (1990) Structureof the Bordetella pertussis gene coding for the serotype 3 ¢mbrialsubunit. FEMS Microbiol. Lett. 54, 327^331.

[117] Livey, I., Duggleby, C.J. and Robinson, A. (1987) Cloning and nu-cleotide sequence analysis of the serotype 2 ¢mbrial subunit gene ofBordetella pertussis. Mol. Microbiol. 1, 203^209.

[118] Pedroni, P., Riboli, B., de Ferra, F., Grandi, G., Toma, S., Arico, B.and Rappuoli, R. (1988) Cloning of a novel pilin-like gene fromBordetella pertussis : homology to the ¢m2 gene. Mol. Microbiol.2, 539^543.

[119] Riboli, B., Pedroni, P., Cuzzoni, A., Grandi, G. and de Ferra, F.(1991) Expression of Bordetella pertussis ¢mbrial (¢m) genes in Bor-detella bronchiseptica : ¢mX is expressed at a low level and vir-regu-lated. Microb. Pathogen. 10, 393^403.

[120] Boschwitz, J.S., van der Heide, H.G.J., Mooi, F.R. and Relman,D.A. (1997) Bordetella bronchiseptica expresses the ¢mbrial structur-al subunit gene ¢mA. J. Bacteriol. 179, 7882^7885.

[121] Geuijen, C.A., Willems, R.J. and Mooi, F.R. (1996) The major¢mbrial subunit of Bordetella pertussis binds to sulfated sugars. In-fect. Immun. 64, 2657^2665.

[122] Geuijen, C.A., Willems, R.J., Hoogerhout, P., Puijk, W.C., Meloen,R.H. and Mooi, F.R. (1998) Identi¢cation and characterization ofheparin binding regions of the Fim2 subunit of Bordetella pertussis.Infect. Immun. 66, 2256^2263.

[123] Pearce, A.M., Seabrook, R.N., Irons, L.I., Ashworth, L.A., Atkin-son, T. and Robinson, A. (1994) Localization of antigenic domainson the major subunits of Bordetella pertussis serotype 2 and 3 ¢m-briae. Microbiology 140, 205^211.

[124] Williamson, P. and Matthews, R. (1996) Epitope mapping the Fim2and Fim3 proteins of Bordetella pertussis with sera from patientsinfected with or vaccinated against whooping cough. FEMS Immu-nol. Med. Microbiol. 13, 169^178.

[125] Allen, B.L., Gerlach, G.F. and Clegg, S. (1991) Nucleotide sequenceand functions of mrk determinants necessary for expression of type3 ¢mbriae in Klebsiella pneumoniae. J. Bacteriol. 173, 916^920.

[126] van Ham, S.M., van Alphen, L., Mooi, F.R. and van Putten, J.P.M.(1994) The ¢mbrial gene cluster of Haemophilus in£uenzae type b.Mol. Microbiol. 13, 673^684.

[127] Willems, R.J., Geuijen, C., van der Heide, H.J.G., Matheson, M.,Robinson, A., Versluis, L.F., Ebberink, R., Theelen, J. and Mooi,F.R. (1993) Isolation of a putative ¢mbrial adhesin from Bordetellapertussis and the identi¢cation of its gene. Mol. Microbiol. 9, 623^634.

[128] Gerlach, G.F., Clegg, S. and Allen, B.L. (1989) Identi¢cation andcharacterisation of the genes encoding the type 3 and type 1 ¢mbrialadhesins of Klebsiella pneumoniae. J. Bacteriol. 171, 1262^1270.

[129] Klemm, P. and Christiansen, G. (1987) Three ¢m genes required forthe regulation of length and mediation of adhesion of Escherichiacoli type 1 ¢mbriae. Mol. Gen. Genet. 201, 439^445.

[130] Hazenbos, W.L., van den Berg, B.M., Geuijen, C.W., Mooi, F.R.and van Furth, R. (1995) Binding of FimD on Bordetella pertussis tovery late antigen-5 on monocytes activates complement receptortype 3 via protein tyrosine kinases. J. Immunol. 155, 3972^3978.

[131] Menozzi, F.D., Gantiez, C. and Locht, C. (1991) Interaction of theBordetella pertussis ¢lamentous hemagglutinin with heparin. FEMSMicrobiol. Lett. 62, 59^64.

[132] Ashworth, L.A.E., Fitzgeorge, R.B., Irons, L.I., Morgan, C.P. andRobinson, A. (1982) Rabbit nasopharyngeal colonisation by Borde-tella pertussis : the e¡ects of immunisation on clearance and onserum and nasal antibody levels. J. Hyg. 88, 475^479.

[133] Willems, R., Paul, A., van der Heide, H.G.J., ter Avest, A.R. andMooi, F.R. (1990) Fimbrial phase variation in Bordetella pertussis :a novel mechanism for transcriptional regulation. EMBO J. 9, 2803^2809.

[134] Leininger, E., Roberts, M., Kenimer, J.G., Charles, I.G., Fair-weather, N., Novotny, P. and Brennan, M.J. (1991) Pertactin, anArg-Gly-Asp-containing Bordetella pertussis surface protein thatpromotes adherence of mammalian cells. Proc. Natl. Acad. Sci.USA 88, 345^349.

[135] Mako¡, A.J., Oxer, M.D., Ballantine, S.P., Fairweather, N.F. andCharles, I.G. (1990) Protective surface antigen P69 of Bordetellapertussis : its characterization and very high level expression in Es-cherichia coli. BioTechnology 8, 1030^1033.

[136] Li, L.J., Dougan, G., Novotny, P. and Charles, I.G. (1991) P.70pertactin, an outer-membrane protein from Bordetella parapertussis :cloning, nucleotide sequence and surface expression in Escherichiacoli. Mol. Microbiol. 5, 409^417.

[137] Montaraz, J.A., Novotny, P. and Ivanyi, J. (1985) Identi¢cation of a68-kilodalton protective protein antigen from Bordetella bronchisep-tica. Infect. Immun. 47, 744^751.

[138] Emsley, P., Charles, I.G., Fairweather, N.F. and Isaacs, N.W.(1996) Structure of Bordetella pertussis virulence factor P.69 pertac-tin. Nature 381, 90^92.

[139] Charles, I.G., Dougan, G., Pickard, D., Chat¢eld, S., Smith, M.,

FEMSRE 716 1-5-01


Novotny, P., Morrissey, P. and Fairweather, N.F. (1988) Molecularcloning and characterization of protective outer membrane proteinP.69 from Bordetella pertussis. Proc. Natl. Acad. Sci. USA 86, 3554^3558.

[140] Charles, I.G., Li, J.L., Roberts, M., Beesley, K., Romanos, M.,Pickard, D.J., Francis, M., Campbell, D., Dougan, G. and Brennan,M.J. et al. (1991) Identi¢cation and characterization of a protectiveimmunodominant B cell epitope of pertactin (P.69) from Bordetellapertussis. Eur. J. Immunol. 21, 1147^1153.

[141] Charles, I., Fairweather, N., Pickard, D., Beesley, J., Anderson, R.,Dougan, G. and Roberts, M. (1994) Expression of the Bordetellapertussis P.69 pertactin adhesin in Escherichia coli : fate of the car-boxy-terminal domain. Microbiology 140, 3301^3308.

[142] Li, J., Fairweather, N.F., Novotny, P., Dougan, G. and Charles,I.G. (1992) Cloning, nucleotide sequence and heterologous expres-sion of the protective outer-membrane protein P.68 pertactin fromBordetella bronchiseptica. J. Gen. Microbiol. 138, 1697^1705.

[143] Hynes, R.O. (1987) Integrins: a family of cell surface receptors. Cell48, 549^554.

[144] Leininger, E., Ewanowich, C.A., Bhargava, A., Peppler, M.S., Ke-nimer, J.G. and Brennan, M.J. (1992) Comparative roles of the Arg-Gly-Asp sequence present in the Bordetella pertussis adhesins per-tactin and ¢lamentous hemagglutinin. Infect. Immun. 60, 2380^2385.

[145] Everest, P., Li, J., Douce, G., Charles, I., De Azavedo, J., Chat¢eld,S., Dougan, G. and Roberts, M. (1996) Role of the Bordetella per-tussis P.69/pertactin protein and the P.69/pertactin RGD motif inthe adherence to and invasion of mammalian cells. Microbiology142, 3261^3268.

[146] Novotny, P., Chubb, A.P., Cownley, K. and Charles, I.G. (1991)Biologic and protective properties of the 69-kDa outer membraneprotein of Bordetella pertussis : a novel formulation for an acellularpertussis vaccine. J. Infect. Dis. 164, 114^122.

[147] Roberts, M., Cropley, I., Chat¢eld, S. and Dougan, G. (1993) Pro-tection of mice against respiratory Bordetella pertussis infection byintranasal immunization with P.69 and FHA. Vaccine 11, 866^872.

[148] Romanos, M.A., Clare, J.J., Beesley, K.M., Rayment, F.B., Ballan-tine, S.P., Mako¡, A.J., Dougan, G., Fairweather, N.F. andCharles, I.G. (1991) Recombinant Bordetella pertussis pertactin(P69) from the yeast Pichia pastoris : high-level production and im-munological properties. Vaccine 9, 901^906.

[149] Shahin, R.D., Brennan, M.J., Li, Z.M., Meade, B.D. and Manclark,C.R. (1990) Characterization of the protective capacity and immu-nogenicity of the 69-kD outer membrane protein of Bordetella per-tussis. J. Exp. Med. 171, 63^73.

[150] Strugnell, R., Dougan, G., Chat¢eld, S., Charles, I., Fairweather,N., Tite, J., Li, J.L., Beesley, J. and Roberts, M. (1992) Character-ization of a Salmonella typhimurium aro vaccine strain expressing theP.69 antigen of Bordetella pertussis. Infect. Immun. 60, 3994^4002.

[151] Greco, D., Salmaso, S., Mastrantonio, P., Giuliano, M., Tozzi,A.E., Anemona, A., Cio¢ Degli Atti, M.L., Giammanco, A., Panei,P., Blackwelder, W.C., Klein, D.L., Wassilak, S.G. andthe ProgettoPertosse Working Group (1996) A controlled trial of two acellularvaccines and one whole-cell vaccine against pertussis, New Engl. J.Med. 334, 341^348.

[152] Gustafsson, L., Hallander, H.O., Olin, P., Reizenstein, E. and Stor-saeter, J. (1996) A controlled trial of a two-component acellular, a¢ve-component acellular, and a whole-cell pertussis vaccine. NewEngl. J. Med. 334, 349^355.

[153] Tamura, M., Nogimori, K., Murai, S., Yajima, M., Ito, K., Katada,T., Ui, M. and Ishii, S. (1982) Subunit structure of islet-activatingprotein, pertussis toxin, in conformity with the A-B model. Bio-chemistry 21, 5516^5522.

[154] Katada, T., Tamura, M. and Ui, M. (1983) The A protomer of islet-activating protein, pertussis toxin, as an active peptide catalysingADP-ribosylation of a membrane protein. Arch. Biochem. Biophys.224, 290^298.

[155] Locht, C. and Keith, J.M. (1986) Pertussis toxin gene: nucleotidesequence and genetic organisation. Science 232, 1258^1263.

[156] Nicosia, A., Perugini, M., Franzini, C., Casagli, M.C., Borri, M.G.,Antoni, G., Almoni, M., Neri, P., Ratti, G. and Rappuoli, R. (1986)Cloning and sequencing of the pertussis toxin genes: operon struc-ture and gene duplication. Proc. Natl. Acad. Sci. USA 83, 4631^4635.

[157] Nicosia, A., Bartoloni, A., Perugini, M. and Rappuoli, R. (1987)Expression and immunological properties of the ¢ve subunits ofpertussis toxin. Infect. Immun. 55, 963^967.

[158] Burnette, N., Mar, V.L., Cieplak, W., Kaljot, K.T., Marchitto,K.S., Sachdev, R.K., Locht, C. and Keith, J.M. (1988) Direct ex-pression of B. pertussis toxin subunits to high levels in E. coli. Bio-Technology 6, 699^706.

[159] Arico, B., Gross, R., Smida, J. and Rappuoli, R. (1987) Evolution-ary relationships in the genus Bordetella. Mol. Microbiol. 1, 301^308.

[160] Runeberg-Nyman, K., Engstrom, O., Lofdahl, S., Ylostalo, S. andSarvas, M. (1987) Expression and secretion of pertussis toxin sub-unit S1 in Bacillus subtilis. Microb. Pathogen. 3, 461^468.

[161] Saris, P., Taira, S., Airaksinen, U., Palva, A., Sarvas, M., Palva, I.and Runeberg-Nyman, K. (1990) Production and secretion of per-tussis toxin subunits in Bacillus subtilis. FEMS Microbiol. Lett. 68,143^148.

[162] Dalla Pozza, T., Yan, H. and Walker, M.J. (1997) Expression of asynthetic pertussis toxin operon in Escherichia coli. Vaccine 15, 968^975.

[163] Dalla Pozza, T., Yan, H., Meek, D., Guzman, C.A. and Walker,M.J. (1998) Construction and characterisation of Salmonella typhi-murium aroA simultaneously expressing the ¢ve pertussis toxin sub-units. Vaccine 16, 522^529.

[164] Nascimento, I.P., Dias, W.O., Mazzantini, R.P., Miyaji, E.N., Gam-berini, M., Quintilio, W., Gebara, V.C., Cardoso, D.F., Ho, P.L.,Raw, I., Winter, N., Gicquel, B., Rappuoli, R. and Leite, L.C.C.(2000) Recombinant Mycobacterium bovis BCG expressing pertussistoxin subunit S1 induces protection against an intracerebral chal-lenge with live Bordetella pertussis in mice. Infect. Immun. 68, 4877^4883.

[165] Lee, C.K., Roberts, A. and Perrin, S. (1989) Expression of pertussistoxin in Bordetella bronchiseptica and Bordetella parapertussis carry-ing recombinant plasmids. Infect. Immun. 57, 1413^1418.

[166] Nencioni, L., Pizza, M., Bugnoli, M., deMagistris, T., diTommaso,A., Giovannoni, F., Manetti, R., Marsili, I., Matteucci, G. andNucci, D. et al. (1990) Characterization of genetically inactivatedpertussis toxin mutants: candidates for a new vaccine againstwhooping cough. Infect. Immun. 58, 1308^1315.

[167] Pizza, M., Covacci, A., Bartoloni, A., Perugini, M., Nencioni, L.,De, M.M., Villa, L., Nucci, D., Manetti, R. and Bugnoli, M. (1989)Mutants of pertussis toxin suitable for vaccine development. Science246, 497^500.

[168] Zealey, G.R., Loosmore, S.M., Yacoob, R.K., Cockle, S.A., Her-bert, A.B., Miller, L.D., Mackay, N.J. and Klein, M.H. (1992) Con-struction of Bordetella pertussis strains that overproduce geneticallyinactivated pertussis toxin. Appl. Environ. Microbiol. 58, 208^214.

[169] Suarez, A., Staendner, L.H., Rhode, M., Piatti, G., Timmis, K.N.and Guzman, C.A. (1997) Stable expression of pertussis toxin inBordetella bronchiseptica under the control of a tightly regulatedpromoter. Appl. Environ. Microbiol. 63, 122^127.

[170] Walker, M.J., Rohde, M., Wehland, J. and Timmis, K.N. (1991)Construction of minitransposons for constitutive and inducible ex-pression of pertussis toxin in bvg-negative Bordetella bronchiseptica.Infect. Immun. 59, 4238^4248.

[171] Smith, A.M. and Walker, M.J. (1996) Transfer of a pertussis toxinexpression locus to isogenic bvg-positive and bvg-negative strains ofBordetella bronchiseptica using an in vivo technique. Microb. Patho-gen. 20, 263^273.

[172] Weiss, A.A., Johnson, F.D. and Burns, D.L. (1993) Molecular char-

FEMSRE 716 1-5-01


acterization of an operon required for pertussis toxin secretion.Proc. Natl. Acad. Sci. USA 90, 2970^2974.

[173] Farizo, K.M., Cafarella, T.G. and Burns, D.L. (1996) Evidence fora ninth gene, ptlI in the locus encoding the pertussis toxin secretionsystem of Bordetella pertussis and formation of a PtlI-PtlF complex.J. Biol. Chem. 271, 31643^31649.

[174] Johnson, F.D. and Burns, D.L. (1994) Detection and subcellularlocalization of three Ptl proteins involved in the secretion of pertus-sis toxin from Bordetella pertussis. J. Bacteriol. 176, 5350^5356.

[175] Kotob, S.I., Hausman, S.Z. and Burns, D.L. (1995) Localization ofthe promoter for the ptl genes of Bordetella pertussis, which encodeproteins essential for secretion of pertussis toxin. Infect. Immun. 63,3227^3230.

[176] Baker, S.M., Masi, A., Liu, D.F., Novitsky, B.K. and Deich, R.A.(1995) Pertussis toxin export genes are regulated by the ptx pro-moter and may be required for e¤cient translation of ptx mRNAin Bordetella pertussis. Infect. Immun. 63, 3920^3926.

[177] Hausman, S.Z., Cherry, J.D., Heininger, U., Wirsing von Konig,C.H. and Burns, D.L. (1996) Analysis of proteins encoded by theptx and ptl genes of Bordetella bronchiseptica and Bordetella para-pertussis. Infect. Immun. 64, 4020^4026.

[178] Welch, R.A. (1991) Pore-forming cytolysins of Gram-negative bac-teria. Mol. Microbiol. 5, 521^528.

[179] Khelef, N., Sakamoto, H. and Guiso, N. (1992) Both adenylatecyclase and hemolytic activities are required by Bordetella pertussisto initiate infection. Microb. Pathogen. 12, 227^235.

[180] Glaser, P., Ladant, D., Sezer, O., Pichot, F., Ullmann, A. and Dan-chin, A. (1988) The calmodulin-sensitive adenylate cyclase of Borde-tella pertussis : cloning and expression in Escherichia coli. Mol. Mi-crobiol. 2, 19^30.

[181] Gentile, F., Knipling, L.G., Sackett, D.L. and Wol¡, J. (1990) In-vasive adenylyl cyclase of Bordetella pertussis. Physical, catalytic,and toxic properties. J. Biol. Chem. 265, 10686^10692.

[182] Leusch, M.S., Paulaitis, S. and Friedman, R.L. (1990) Adenylatecyclase toxin of Bordetella pertussis : production, puri¢cation, andpartial characterization. Infect. Immun. 58, 3621^3626.

[183] Rogel, A., Schultz, J.E., Brownlie, R.M., Coote, J.G., Parton, R.and Hanski, E. (1989) Bordetella pertussis adenylate cyclase: puri¢-cation and characterization of the toxic form of the enzyme. EMBOJ. 8, 2755^2760.

[184] Ladant, D., Michelson, S., Sarfati, R., Gilles, A.M., Predeleanu, R.and Barzu, O. (1989) Characterization of the calmodulin-bindingand of the catalytic domains of Bordetella pertussis adenylate cy-clase. J. Biol. Chem. 264, 4015^4020.

[185] Heveker, N. and Ladant, D. (1997) Characterisation of mutantBordetella pertussis adenylate cyclase toxins with reduced a¤nityfor calmodulin. Eur. J. Biochem. 243, 643^649.

[186] Confer, D.L. and Eaton, J.W. (1982) Pathogen-host jujitsu: phag-ocyte impotence caused by internalized bacterial adenylate cyclase.Trans. Assoc. Am. Phys. 95, 1^7.

[187] Greenlee, D.V., Andreasen, T.J. and Storm, D.R. (1982) Calcium-independent stimulation of Bordetella pertussis adenylate cyclasetoxin by calmodulin. Biochemistry 21, 2759^2764.

[188] Sakamoto, H., Bellalou, J., Sebo, P. and Ladant, D. (1992) Borde-tella pertussis adenylate cyclase toxin. Structural and functional in-dependence of the catalytic and hemolytic activities. J. Biol. Chem.267, 13598^13602.

[189] Boehm, D.F., Welch, R.A. and Snyder, I.S. (1990) Domains of Es-cherichia coli haemolysin (HlyA) involved in binding of calcium anderythrocyte membranes. Infect. Immun. 58, 1959^1966.

[190] Ludwig, A., Jarchau, T., Benz, R. and Goebel, W. (1988) Mutationsa¡ecting activity and transport of haemolysin in Escherichia coli.Mol. Gen. Genet. 214, 553^561.

[191] Rose, T., Sebo, P., Bellalou, J. and Ladant, D. (1995) Interaction ofcalcium with Bordetella pertussis adenylate cyclase toxin. Character-ization of multiple calcium-binding sites and calcium-induced con-formational changes. J. Biol. Chem. 270, 26370^26376.

[192] Goyard, S., Sebo, P., D'Andria, O., Ladant, D. and Ullmann, A.(1993) Bordetella pertussis adenylate cyclase: a toxin with multipletalents. Int. J. Med. Microbiol. Virol. Parasitol. Infect. Dis. 278,326^333.

[193] Glaser, P., Sakamoto, H., Bellalou, J., Ullmann, A. and Danchin,A. (1988) Secretion of cyclolysin, the calmodulin-sensitive adenylatecyclase-haemolysin bifunctional protein of Bordetella pertussis.EMBO J. 7, 3997^4004.

[194] Barry, E.M., Weiss, A.A., Ehrmann, I.E., Gray, M.C., Hewlett,E.L. and Goodwin, M.S. (1991) Bordetella pertussis adenylate cy-clase toxin and hemolytic activities require a second gene, cyaC, foractivation. J. Bacteriol. 173, 720^726.

[195] Hackett, M., Guo, L., Shabanowitz, J., Hunt, D.F. and Hewlett,E.L. (1994) Internal lysine palmitoylation in adenylate cyclase toxinfrom Bordetella pertussis. Science 266, 433^435.

[196] Gordon, V.M., Young, W.J., Lechler, S.M., Gray, M.C., Leppla,S.H. and Hewlett, E.L. (1989) Adenylate cyclase toxins from Bacil-lus anthracis and Bordetella pertussis. Di¡erent processes for inter-action with and entry into target cells. J. Biol. Chem. 264, 14792^14796.

[197] Rogel, A. and Hanski, E. (1992) Distinct steps in the penetration ofadenylate cyclase toxin of Bordetella pertussis into sheep erythro-cytes. Translocation of the toxin across the membrane. J. Biol.Chem. 267, 22599^22605.

[198] Arciniega, J.L., Hewlett, E.L., Johnson, F.D., Deforest, A., Wassi-lak, S.G., Onorato, I.M., Manclark, C.R. and Burns, D.L. (1991)Human serologic response to envelope-associated proteins and ade-nylate cyclase toxin of Bordetella pertussis. J. Infect. Dis. 163, 135^142.

[199] Guiso, N., Szatanik, M. and Rocancourt, M. (1991) Protective ac-tivity of Bordetella adenylate cyclase-hemolysin against bacterialcolonization. Microb. Pathogen. 11, 423^431.

[200] Hewlett, E.L., Manclark, C.R. and Wol¡, J. (1977) Adenylyl cyclasein Bordetella pertussis vaccines. J. Infect. Dis. 136, 5216^5219.

[201] Guiso, N., Rocancourt, M., Szatanik, M. and Alonso, J.M. (1989)Bordetella adenylate cyclase is a virulence associated factor and animmunoprotective antigen. Microb. Pathogen. 7, 373^380.

[202] Novotny, P., Chubb, A.P., Cownley, K. and Montaraz, J.A. (1985)Adenylate cyclase activity of a 68,000-molecular-weight protein iso-lated from the outer membrane of Bordetella bronchiseptica. Infect.Immun. 50, 199^206.

[203] Friedman, R.L., Nordensson, K., Wilson, L., Akporiaye, E.T. andYocum, D.E. (1992) Uptake and intracellular survival of Borde-tella pertussis in human macrophages. Infect. Immun. 60, 4578^4585.

[204] Masure, H.R. (1992) Modulation of adenylate cyclase toxin produc-tion as Bordetella pertussis enters human macrophages. Proc. Natl.Acad. Sci. USA 89, 6521^6525.

[205] Karimova, G., Fayolle, C., Gmira, S., Ullmann, A., Leclerc, C. andLadant, D. (1998) Charge-dependent translocation of Bordetella per-tussis adenylate cyclase toxin into eukaryotic cells : implications forthe in vivo delivery of CD8(+) T cell epitopes into antigen present-ing cells. Proc. Natl. Acad. Sci. USA 95, 12532^12537.

[206] Dadaglio, G., Moukrim, Z., Lo-Man, R., Sheshko, V., Sebo, P. andLeclerc, C. (2000) Induction of a polarized Th1 response by inser-tion of multiple copies of a viral T-cell epitope into adenylate cy-clase of Bordetella pertussis. Infect. Immun. 68, 3867^3872.

[207] Gueirard, P., Druilhe, A., Pretolani, M. and Guiso, N. (1998) Roleof adenylate cyclase toxin/haemolysin in alveolar macrophage apo-ptosis during Bordetella pertussis infection in vivo. Infect. Immun.66, 1718^1725.

[208] Khelef, N., Zychlinsky, A. and Guiso, N. (1993) Bordetella pertussisinduces apoptosis in macrophages: role of adenylate cyclase-hemo-lysin. Infect. Immun. 61, 4064^4071.

[209] Khelef, N. and Guiso, N. (1995) Induction of macrophage apoptosisby Bordetella pertussis adenylate cyclase-hemolysin. FEMS Micro-biol. Lett. 134, 27^32.

FEMSRE 716 1-5-01


[210] Bordet, J. and Gengou, O. (1909) L'endotoxine coquelucheuse. Ann.Inst. Pasteur (Paris) 23, 415^419.

[211] Iida, T. and Okonogi, T. (1971) Lienotoxicity of Bordetella pertussisin mice. J. Med. Microbiol. 4, 51^61.

[212] Livey, I. and Wardlaw, A.C. (1984) Production and properties ofBordetella pertussis heat-labile toxin. J. Med. Microbiol. 17, 91^103.

[213] Endoh, M., Amitani, M. and Nakase, Y. (1986) E¡ect of puri¢edheat-labile toxin of Bordetella bronchiseptica on the peripheral bloodvessels in guinea pigs or suckling mice. Microbiol. Immunol. 30,1327^1330.

[214] Endoh, M., Nagai, M., Ueda, T., Yoshida, Y. and Nakase, Y.(1988) Cytopathic e¡ect of heat-labile toxin of Bordetella paraper-tussis on aortic smooth muscle cells from pigs or guinea pigs. Micro-biol. Immun. 32, 423^428.

[215] Gentry-Weeks, C.R., Keith, J.M. and Thompson, J. (1988) Toxicityof Bordetella avium beta-cystathionase toward MC3t3-El osteogeniccells. J. Biol. Chem. 268, 7298^7314.

[216] Cowell, J.L., Hewlett, E.L. and Manclark, C.R. (1979) Intracellularlocalisation of the dermonecrotic toxin of Bordetella pertussis. In-fect. Immun. 25, 896^901.

[217] Horiguchi, Y., Nakai, T. and Kume, K. (1990) Simpli¢ed procedurefor puri¢cation of Bordetella bronchiseptica dermonecrotic toxin.FEMS Microbiol. Lett. 66, 39^43.

[218] Zhang, Y.L. and Sekura, R.D. (1991) Puri¢cation and character-ization of the heat-labile toxin of Bordetella pertussis. Infect. Im-mun. 59, 3754^3759.

[219] Walker, K.E. and Weiss, A.A. (1994) Characterization of the der-monecrotic toxin in members of the Genus Bordetella. Infect. Im-mun. 62, 3817^3828.

[220] Falbo, V., Pace, T., Picci, L., Pizzi, E. and Caprioli, A. (1993)Isolation and nucleotide sequence of the gene encoding cytotoxinnecrotising factor I in Escherichia coli. Infect. Immun. 61, 4909^4916.

[221] Hoshijima, M., Kondo, J., Kikuchi, A., Yamamoto, K. and Takai,Y. (1990) Puri¢cation and characterisation from bovine membranesof a GTP-binding protein with a Mr of 21 000, ADP-ribosylated byan ADP-ribosyltransferase contaminated in botulinum toxin typeC1-identi¢cation as the rhoA gene product. Mol. Brain Res. 7, 9^16.

[222] Oswald, E., Sugai, M., Labigne, A., Wu, H.C., Fiorentini, C., Bo-quet, P. and O'Brien, A.D. (1994) Cytotoxic necrotising factor type2 produced by virulent Escherichia coli modi¢es the small GTP-binding proteins Rho involved in assembly of actin stress ¢bres.Proc. Natl. Acad. Sci. USA 91, 3814^3818.

[223] Rozengurt, E., Higgins, T., Chater, N., Lax, A.J. and Staddon, J.M.(1990) Pasteurella multocida toxin: potent mitogen for cultured ¢-broblasts. Proc. Natl. Acad. Sci. USA 87, 123^127.

[224] Schmidt, G., Goehring, U.M., Schirmer, J., Lerm, M. and Aktories,K. (1999) Identi¢cation of the C-terminal part of Bordetella dermo-necrotic toxin as a transglutaminase for rho GTPases. J. Biol.Chem. 275, 31875^31881.

[225] Cookson, B.T., Cho, H.L., Herwaldt, L.A. and Goldman, W.E.(1989) Biological activities and chemical composition of puri¢edtracheal cytotoxin of Bordetella pertussis. Infect. Immun. 57,2223^2229.

[226] Cookson, B.T. and Goldman, W.E. (1987) Tracheal cytotoxin: aconserved virulence determinant of all Bordetella species. J. Cell.Biochem. (Suppl.) 11B, 124 (abstr. H208).

[227] Sinha, R.K. and Rosenthal, R.S. (1980) Release of soluble peptido-glycan from growing gonococci: demonstration of anhydro-muram-yl-containing fragments. Infect. Immun. 29, 914^925.

[228] Luker, K.E., Tyler, A.N., Marshall, G.R. and Goldman, W.E.(1995) Tracheal cytotoxin structural requirements for respiratoryepithelial damage in pertussis. Mol. Microbiol. 16, 733^743.

[229] Heiss, L.N., Moser, S.A., Unanue, E.R. and Goldman, W.E. (1993)Interleukin-1 is linked to the respiratory epithelial cytopathology ofpertussis. Infect. Immun. 61, 3123^3128.

[230] Heiss, L.N., Lancaster, J.J., Corbett, J.A. and Goldman, W.E.

(1994) Epithelial autotoxicity of nitric oxide: role in the respiratorycytopathology of pertussis. Proc. Natl. Acad. Sci. USA 91, 267^270.

[231] Cundell, D.R., Kanthakumar, K., Taylor, G.W., Goldman, W.E.,Flak, T., Cole, P.J. and Wilson, R. (1994) E¡ect of tracheal cyto-toxin from Bordetella pertussis on human neutrophil function invitro. Infect. Immun. 62, 639^643.

[232] Amano, K., Fukushi, K. and Watanabe, M. (1990) Biochemical andimmunological comparison of lipopolysaccharides from Bordetellaspecies. J. Gen. Microbiol. 136, 481^487.

[233] Watanabe, M., Takimoto, H., Kumazawa, Y. and Amano, K.(1990) Biological properties of lipopolysaccharides from Bordetellaspecies. J. Gen. Microbiol. 136, 489^493.

[234] Chaby, R. and Caro¡, M. (1988) in: Pathogenesis and Immunity inPertussis (Wardlaw, A.C. and Parton, R., Eds.), pp. p247^271. Wi-ley, New York.

[235] Peppler, M.S. (1984) Two physically and serologically distinct lipo-polysaccharide pro¢les in strains of Bordetella pertussis and theirphenotype variants. Infect. Immun. 43, 224^232.

[236] Caro¡, M., Chaby, R., Karibian, D., Perry, J., Deprun, C. andSzabo, L. (1990) Variations in the carbohydrate regions of Borde-tella pertussis lipopolysaccharides : electrophoretic, serological, andstructural features. J. Bacteriol. 172, 1121^1128.

[237] van den Akker, W.M.R. (1998) Lipopolysaccharide expression with-in the genus Bordetella ^ in£uence of temperature and phase varia-tion. Microbiology 144, 1527^1535.

[238] Porter, J.F., Connor, K., van der Zee, A., Reubsaet, F., Ibsen, P.,Heron, I., Chaby, R., Le Blay, K. and Donachie, W. (1995) Char-acterisation of ovine Bordetella parapertussis isolates by analysis ofspeci¢c endotoxin (lipopolysaccharide) epitopes, ¢lamentous haem-agglutinin production, cellular fatty acid composition and antibioticsensitivity. FEMS Microbiol. Lett. 132, 195^201.

[239] van der Zee, A., Groenendijk, H., Peeters, M. and Mooi, F.R.(1996) The di¡erentiation of Bordetella parapertussis and Bordetellabronchiseptica from humans and animals as determined by DNApolymorphism mediated by two di¡erent insertion sequence ele-ments suggests their phylogenetic relationship. Int. J. Syst. Bacteriol.46, 640^647.

[240] Di Fabio, J.L., Caro¡, M., Karibian, D., Richards, J.C. and Perry,M.B. (1992) Characterization of the common antigenic lipopolysac-charide O-chains produced by Bordetella bronchiseptica and Borde-tella parapertussis. FEMS Microbiol. Lett. 76, 275^281.

[241] Banemann, A., Deppisch, H. and Gross, R. (1998) The lipopolysac-charide of Bordetella bronchiseptica acts as a protective shieldagainst antimicrobial peptides. Infect. Immun. 66, 5607^5612.

[242] Leblay, K., Gueirard, P., Guiso, N. and Chaby, R. (1997) Antigenicpolymorphism of the lipopolysaccharides from human and animalisolates of Bordetella bronchiseptica. Microbiology 143, 1433^1441.

[243] Allen, A. and Maskell, D. (1996) The identi¢cation, cloning andmutagenesis of a genetic locus required for lipopolysaccharide bio-synthesis in Bordetella pertussis. Mol. Microbiol. 19, 37^52.

[244] Finn, T.M., Shahin, R. and Mekalanos, J.J. (1991) Characterizationof vir-activated TnphoA gene fusions in Bordetella pertussis. Infect.Immun. 59, 3273^3279.

[245] Finn, T.M. and Stevens, L.A. (1995) Tracheal colonization factor: aBordetella pertussis secreted virulence determinant. Mol. Microbiol.16, 625^634.

[246] Fernandez, R.C. and Weiss, A.A. (1994) Cloning and sequencing ofa Bordetella pertussis serum resistance locus. Infect. Immun. 62,4727^4738.

[247] Rambow, A.A., Fernandez, R.C. and Weiss, A.A. (1998) Character-ization of brkA expression in Bordetella bronchiseptica. Infect. Im-mun. 66, 3978^3980.

[248] Fernandez, R.C. and Weiss, A.A. (1998) Serum resistance in bvg-regulated mutants of Bordetella pertussis. FEMS Microbiol. Lett.163, 57^63.

[249] Bromberg, K., Tannis, G. and Steiner, P. (1991) Detection of Bor-detella pertussis associated with the alveolar macrophages of chil-

FEMSRE 716 1-5-01


dren with human immunode¢ciency virus infection. Infect. Immun.59, 4715^4719.

[250] Ewanowich, C.A., Melton, A.R., Weiss, A.A., Sherburne, R.K. andPeppler, M.S. (1989) Invasion of HeLa 229 cells by virulent Borde-tella pertussis. Infect. Immun. 57, 2698^2704.

[251] Petersen, J.W., Ibsen, P.H., Haslov, K. and Heron, I. (1992) Pro-liferative responses and gamma interferon and tumor necrosis factorproduction by lymphocytes isolated from tracheobroncheal lymphnodes and spleen of mice aerosol infected with Bordetella pertussis.Infect. Immun. 60, 4563^4570.

[252] Redhead, K., Watkins, J., Barnard, A. and Mills, K.H. (1993) Ef-fective immunization against Bordetella pertussis respiratory infec-tion in mice is dependent on induction of cell-mediated immunity.Infect. Immun. 61, 3190^3198.

[253] Mahon, B.P., Ryan, M.S., Gri¤n, F. and Mills, K.H.G. (1996)Interleukin-12 is produced by live or killed Bordetella pertussisand enhances the e¤cacy of an acellular pertussis vaccine by pro-moting induction of Th1 cells. Infect. Immun. 64, 5295^5301.

[254] Mills, K.H.G., Ryan, M., Ryan, E. and Mahon, B.P. (1998) Amurine model in which protection correlates with pertussis vaccinee¤cacy in children reveals complementary roles for humoral andcell-mediated immunity in protection against Bordetella pertussis.Infect. Immun. 66, 594^602.

[255] Barnard, A., Mahon, B.P., Watkins, J., Redhead, K. and Mills,K.H.G. (1996) Th1/Th2 cell dichotomy in acquired immunity toBordetella pertussis : variables in the in vivo priming and in vitrocytokine detection techniques a¡ect the classi¢cation of T-cell sub-sets as Th1, Th2 or Th0. Immunology 87, 372^380.

[256] Ausiello, C.M., Urbani, F., La Sala, A., Lande, R., Piscitelli, A. andCassone, A. (1997) Acellular vaccines induce cell-mediated immun-ity to Bordetella pertussis antigens in infants undergoing primaryvaccination against pertussis. Dev. Biol. Stand. 89, 315^320.

[257] Zepp, F., Knuf, M., Habermehl, P., Schmitt, H.J., Rebsch, C.,Schmidtke, P., Clemens, R. and Slaoui, M. (1996) Pertussis-speci¢ccell-mediated immunity in infants after vaccination with a tricom-ponent acellular pertussis vaccine. Infect. Immun. 64, 4078^4084.

[258] Zepp, F., Knuf, M., Habermehl, P., Schmitt, H.J., Meyer, C., Clem-ens, R. and Slaoui, M. (1997) Cell-mediated immunity after pertus-sis vaccination and after natural infection. Dev. Biol. Stand. 89,307^314.

[259] de Melker, H.E., Conyn-van Spaendonck, M.A.E., Ru«mke, H.C.,van Wijngaarden, J.K., Mooi, F.R. and Schellekens, J.F.P. (1997)Pertussis in the Netherlands: an outbreak despite high levels ofimmunisation with whole-cell vaccine. Emerg. Infect. Dis. 3, 175^178.

[260] Mooi, F.R., van Oirschot, H., Heuvelman, K., van der Heide,H.G.J., Gaastra, W. and Willems, R.J.L. (1998) Polymorphism inthe Bordetella pertussis virulence factors P.69/pertactin and pertussistoxin in the Netherlands: temporal trends and evidence for vaccine-driven evolution. Infect. Immun. 66, 670^675.

[261] Peppoloni, S., Pizza, M., de Magistris, M.T., Bartoloni, A. andRappuoli, R. (1995) Acellular pertussis vaccine composed of genet-ically inactivated pertussis toxin. Physiol. Chem. Phys. Med. NMR27, 355^361.

[262] Sanger Centre Website (2000) http://www.sanger.ac.uk/Projects/B_pertussis/ [19 March 2000, last date accessed].

FEMSRE 716 1-5-01


the virulence factors of bordetella pertussis: a matter of control

Documents