Biologicals (1999) 27, 71

Biologicals(1999)27,71–76Article No.biol.1999.0182,available online at onReview of the Biology of Bordetella pertussisR.PartonDivision of Infection and Immunity,Joseph Black Building,University of Glasgow,Glasgow G128QQ,U.K. Abstract.Bordetella pertussis produces a complex array of adhesins,aggressins and toxins that are presumed to be important in the colonisation of its human host and in ensuring its survival and propagation. The organism also has highly sophisticated mechanisms for regulating virulence factor expression,in response to environmental signals or by reversible mutations.Despite the rapidly increasing knowledge of these aspects of the biology of B.pertussis,our understanding of the pathogenesis of whooping cough is still far from clear.In defining the role of individual factors,reliance has to be placed on in vitro assays or animal models of the human infection,particularly in the mouse,where different conditions may prevail. Some clues to pathogenic mechanisms may be provided by considering other bordetellae,especially B.parapertussis,B.bronchiseptica and B.avium,their similar,but not identical,range of virulence factors and the common features of the diseases caused by these species in their respective hosts.The bordetellae are usually defined as obligate,non-invasive parasites of the respiratory tracts of warm-blooded animals,including birds,with a predilection for the respiratory ciliated epithelium.This definition has been challenged by a number of recent observations.For example,the ability of Bordetella spp.to regulate virulence factor expression in response to external signals strongly suggests that they have alternative habitats where such regulation would be an advantage.These habitats may be intracellular, since it has been shown that B.pertussis,B.parapertussis and B.bronchiseptica can invade and survive within host cells,or they may be in other sites within the same or different hosts.Recent DNA fingerprinting studies of B.pertussis have revealed hitherto unsuspected heterogeneity amongst isolates which could be reflected in antigenic differences between strains.Some of these new perspectives on Bordetella pathogenicity may have implications for pertussis vaccine development.©1999The International Association for BiologicalsThe genus BordetellaIn reviewing the biology of Bordetella pertussis,it is worthwhilefirst to consider the genus Bordetella because some useful comparisons can be made between the species.Six species have now been described(Table1).Bordetella pertussis,as far as we know,is restricted to man and to the human respir-atory tract.Bordetella parapertussis is no longer considered to be a strict parasite of man since its isolation from healthy and pneumonic sheep in New Zealand and Scotland.1,2Recent genetic studies have shown that the human and sheep strains constitute distinct populations and thus there is no evidence of transmission of B.parapertussis between man and sheep.3Bordetella bronchiseptica is an important respiratory tract pathogen of a wide range of animals and birds and,despite the oppor-tunities for animal to human transmission,is sel-dom isolated from man.In man it appears to be more of a respiratory tract commensal and opportunist pathogen,sometimes associated with cases of septi-caemia in compromised patients.4Bordetella avium is an agent of respiratory disease in turkey poults and other birds.5From DNA hybridization studies, DNA base ratio determinations and other genotypic and phenotypic evidence,two new species have been proposed recently.Bordetella hinzii is the name given to a B.avium-like group of organisms from the respiratory tracts of turkeys and chickens and four human isolates have been reported.6 Bordetella holmesii describes a group of isolates from human blood cultures.7,8The pathogenic potential of these two species is unclear.Both appear to be opportunists in man and have been isolated mainly from compromised patients.No alternative host has been reported for B.holmesii. The bordetellae have been described as obligate, non-invasive parasites of the respiratory tracts of warm-blooded animals,including birds,with a pre-dilection for the respiratory ciliatedepithelium.1045–1056/99/020071+06$30.00/0 1999The International Association for BiologicalsThis definition has been challenged by a number of recent observations.For example,the ability of Bordetella spp.to regulate virulence factor expres-sion in response to external signals(see below) strongly suggests that they have habitats in addition to the surface of the respiratory tract, where such regulation would take place.These habitats may be intracellular,since it has been shown that B.pertussis and other species can invade and survive within host cells(see below),or they may be in other sites within the same or di#erent hosts.As mentioned above,some species have been isolated from human blood.These have been mainly from compromised hosts and perhaps an increase in the number of such isolations is inevitable.Alternatively,an environmental reser-voir may exist for some species although there is no direct evidence for this.However,B.bronchiseptica and B.avium have the remarkable ability to grow, from a low inoculum of washed cells,in lakewater or even phosphate-bu#ered saline at10 C or37 C and to survive for many weeks under these condi-tions(9).B.pertussis and Bordetella parapertussis do not have this ability.Table2shows the distribution of the known virulence factors amongst the four well-characterized species.10Such information may help to define the role of individual factors in Bordetella pathogenicity since not all species have all of the virulence factors and yet the diseases produced in their respective hosts have certain common features.These include:age-related susceptibility to infection;adherence of organisms to ciliated epi-thelium and loss of ciliated cells form the respirat-ory tract;reduced weight gain of the host;excessive mucus production;and some form of cough.A pre-disposition to secondary infections is also common and this increases the mortality in untreated cases. The evidence from Table2suggests that tracheal cytotoxin(TCT),heat-labile toxin(HLT)and fimbriae are likely to be important in pathogenesis as they are produced by all four species.The role of pertussis toxin(PT)in whooping coughOne particular problem,highlighted by Table2, concerns PT,which is unique to B.pertussis and is the pre-eminent component of acellular pertussis vaccines.PT clearly is an important virulence fac-tor and protective antigen.Transposon-insertion mutants of B.pertussis have greatly reduced viru-lence for mice11,12and PT has a remarkably diverse range of biological activities which have been dem-onstrated in vitro and in vivo.10,13Because of these potent activities,PT is presumed to have a centralTable1.Host specificities of Bordetella speciesSpecies Host Site of isolation(man)B.pertussis man respiratory tractB.parapertussis man,sheep respiratory tractB.bronchiseptica animals,birds,(man)respiratory tract,bloodB.avium birdsB.hinzii birds,(man)respiratory tract,bloodB.holmesii?,(man)bloodTable2.Virulence factors of Bordetella speciesVirulence factor B.pertussis B.parapertussis B.bronchiseptica B.avium Tracheal cytotoxin++++ Heat-labile toxin++++ Fimbriae(agglutinogens2&3)++++ Filamentous haemagglutinin+++ Adenylate cyclase toxin+++ Pertactin+++? BrkA+?? Tracheal colonization factor+ ? Pertussis toxin+72R.Partonrole in the pathogenesis of pertussis and,therefore, in causing the typical symptoms of whooping cough. However,some recent studies have shown that the clinical characteristics of pertussis and parapertus-sis can be very similar,even though B.parapertussis does not produce PT.14,15Paroxysmal coughing, whooping and vomiting were all features of para-pertussis infection,with frequencies similar to,or approaching,those in pertussis cases.The main di#erence was that leukocytosis,known to be due to PT,was not a feature in parapertussis.In some cases of whooping cough,both B.pertussis and B. parapertussis have been isolated but such cases seem to have been excluded in the aforementioned studies,by the absence of leukocytosis.Thesefind-ings suggest that B.parapertussis is capable of causing typical whooping cough,even though it is more often associated with milder disease or asymptomatic infection,and that PT does not have a major role in this process.15Suchfindings are very di#icult to reconcile with our studies with a coughing rat model of pertussis.16–18Of all the experimental animals,only primates and rats have been reported to develop the paroxysmal coughing characteristic of whooping cough when infected with B.pertussis.In this model,rats are infected intrabronchially with Bordetella strains embedded infine agarose beads, to prevent rapid clearance from the lungs,and coughing paroxysms are recorded around10days later.Under these conditions,the wild-type or Phase I B.pertussis organisms produced a signifi-cant number of coughing paroxysms,as did an HLT-deficient mutant,whereas a PT mutant,a Phase IV(avirulent)strain of B.pertussis and a B. parapertussis strain,none of which produces PT,did not cause significant coughing.16This suggests that PT does have a role in cough production,at least in rats.Virulence factors and protective antigensA number of virulence factors of B.pertussis and potential protective antigens have been identified and some are at di#erent stages of evaluation as acellular vaccine candidates(Table3).PT,filamentous haemagglutinin(FHA),fimbrial anti-gens(agglutinogens2and3)and pertactin are all components of exisiting acellular pertussis vac-cines.All four are protective in mice.Toxoided PT is protective in man as a monocomponent vaccine and the other three give some additional protection when included in multicomponent vaccines,19 although they have not been tested as mono-component vaccines.Adenylate cyclase toxin(ACT) is an important virulence factor in mouse models of infection.11,12It is also a good protective antigen in mice,as shown by Guiso and co-workers20,21and also in our own laboratory,with purified native and recombinant ACT(Hormozi,Westrop,Parton, Coote,unpublished observations).It might,there-fore,be a candidate antigen for acellular pertussis vaccines.Heat-labile toxin(HLT)is a potent toxin causing dermonecrosis,spleen atrophy,loss of weight gain and lethality in mice but its role in disease has not been defined.HLT-deficient mutants were unaltered in their ability to cause a lethal infection in mice.22HLT is an intracellular com-ponent of B.pertussis and has not been shown to be protective.Tracheal cytotoxin(TCT)is an unusual, low molecular weight toxin derived from the cell wall peptidoglycan and causes ciliostasis andTable3.Virulence factors and protective antigens of Bordetella pertussisVirulence factor Probable role in pathogenicity Protectivein mouse*Protectivein manPertussis toxin adhesion,invasion,interference withimmune e#ector cells ++Filamentous haemagglutinin adhesion,invasion+(+) Fimbriae(agglutinogens2&3)adhesion,invasion+(+) Pertactin adhesion,invasion+(+) Adenylate cyclase toxin interference with immune e#ector cells+? Heat-labile toxin local inflammatory e#ects Tracheal cytotoxin ciliostasis?? BrkA adhesion,invasion,serum resistance?? Tracheal colonisation factor adhesion,invasion?? *Against either intracerebral or pulmonary challenge.73Review of the biology of Bordetella pertussisdestruction of ciliated cells in tracheal organ cul-tures.23Thus,it could contribute to the mucus accumulation,coughing and predisposition to secondary infections seen in pertussis.There are no reports on the immunogenicity of TCT.BrkA24and tracheal colonisation factor25are newly described adhesins of B.pertussis.They are of particular interest because of their structural homology with pertactin and are,therefore,potential vaccine anti-gens.It is noteworthy that the majority of the virulence factors in Table3have been shown to mediate adhesion of B.pertussis to mammalian cells.FHA even has multiple binding specificities.26 It is not clear why the pathogen should need so many adhesins but it is tempting to speculate that they are required for attachment to di#erent cell types,perhaps at di#erent stages of infection,or for colonising habitats in addition to the surface of the human respiratory tract.Antigenic variation and virulence regulation Variation in B.pertussis is potentially important because it can a#ect vaccine potency and vaccine e#icacy.Di#erent types of variation are possible. Serotype variation,involving the gain or loss of expression of thefimbrial antigens,known as agglutinogens2and3,can result from reversible frameshift mutations in thefimbrial gene pro-moters.27There is some evidence that protection of children with whole-cell vaccines is,to some extent,serotype specific.28,29Similar mutations in the virulence regulatory bvg locus30produces avirulent mutants which are unable to express any of the virulence factors listed in Table3,except TCT.This process is called phase variation. Another type of variation,known as antigenic modulation,is a freely-reversible,phenotypic change a#ecting the whole population.It can be induced by changes in growth conditions in vitro but the end result is the same as that of phase variation in that most of the well-characterized virulence factors are no longer expressed.Such strains(Bvg )would clearly be unsuitable for vaccine production.Expression of the virulence factors is controlled by the bvg locus which encodes two proteins,BvgA and BvgS.31,32BvgS is an environment-sensing, transmembrane protein that,in response to signals presumably provided by the usual growth condi-tions,activates the cytoplasmic,DNA-binding pro-tein BvgA by phosphorylation.This activation results in the expression of the so-called virulence-activated genes(vags)and the repression of other genes(virulence-repressed genes;vrgs).There are two classes of vags activated within a few minutes of receiving the appropriate signal,and late vags activated after a few hours and at higher concen-trations of BvgA.31The early vags include the genes for the adhesins FHA andfimbriae.The late vags include the genes for the toxins PT and ACT but also for the adhesin pertactin and it may be that these components are required at a later stage of the infection process.Under other growth conditions (modulating conditions),the process is reversed,the vags are expressed and the vags repressed.Thus,B. pertussis has very sophisticated mechanisms for antigenic variation but the significance of bvg regu-lation is unclear.Any one of a number of in vitro conditions,such as growth at25 C or in the pres-ence of high concentrations of nicotinic acid or magnesium sulfate,will cause modulation but it is not known when,or even if,this change occurs in vivo and what the‘‘real’’signals are.Another mys-tery surrounds the role of the vrg products of which more than twenty have been identified.33Surpris-ingly,at least one of these,Vrg6,appears to be required for full virulence of B.pertussis in the mouse lung.34Invasions and intracellular survivalIt has recently become apparent that some Bordetella species,including B.pertussis,can invade and survive within various epithelial cell lines and phagocytes,including human macrophages.35–37With B.pertussis,only the Bvg+ strains have this ability and several of the well-characterized virulence factors have been impli-cated in the process.However,there is little or no multiplication of B.pertussis in this niche and survival appears to depend on the numbers of bacteria taken up.36Afinding which may be important in intracellular survival is that B. pertussis can induce macrophage apoptosis,due to ACT,but the bacteria need not be intracellular to do this.38The significance of the intracellular state in pertussis is not yet clear but it could be important in the protracted course of the disease or in the establishment of a chronic or quiescent stage,although there is little evidence for this. The importance of cell-mediated immunity in B.pertussis infections,demonstrated in animal models,may be due in part to the destruction of infected host cells.3974R.PartonDNAfingerprinting and strain differencesThe results of recent epidemiological investigations of pertussis have possible implications for vaccine e#icacy and for acellular vaccine development.B. pertussis is generally regarded as a fairly homo-geneous species showing only limited genetic diversity and,apart from serotype variation,limited antigenic diversity.However,DNAfingerprinting using pulsed-field gel electrophoresis of genome fragments have revealed strain di#erences by the presence of numerous DNA types amongst B. pertussis isolates.40–42Moreover,it appears that certain strains predominate during epidemics,even though many di#erent DNA types are present in the population.This raises the question as to whether antigenic or other di#erences occur between strains,that give some strains an advantage in combatting host defence mechanisms or in overcom-ing immunity induced by previous epidemics.43In a Dutch study,with afingerprinting method based on DNA polymorphisms associated with a novel inser-tion sequence,van der Zee et al.43found evidence that vaccination had resulted in the selection of strains which di#ered in DNA type from the vaccine strains and suggested that this could have been due to antigenic di#erences between the strains. References1.Chen W,Alley MR,Manktelow BW et al.Pneumoniain lambs inoculated with Bordetella parapertussis: bronchoalveolar lavage and ultrastructural studies.Vet Pathol1988;25:297–303.2.Porter JF,Connor K,Donachie W.Isolation andcharacterization of Bordetella parapertussis-like bacteria from ovine lungs.Microbiology1994;140: 255–261.3.van der Zee A,Groenendijk H,Peeters M et al.Thedi#erentiation of B.parapertussis and B.bronchi-septica from man and animals as determined by DNA polymorphism mediated by two di#erent insertion sequence elements suggests their phylogenetic relationship.Int J Syst Bacteriol1996;46:640–647.4.Woolfrey BF,Moody JA.Human infections associatedwith Bordetella bronchiseptica.Clin Microbiol Rev 1991;4:243–255.5.Kersters K,Hinz KH,Hertle A et al.Bordetella aviumsp.nov.,isolated from the respiratory tracts of turkeys and other birds.Int J Syst Bacteriol1984;34: 56–70.6.Vandamme P,Hommez J,Vancanneyt M et al.Bordetella hinzii sp.nov.,isolated from poultry and humans.Int J Syst Bacteriol1995;45:37–45.7.Weyant RS,Hollis DG,Weaver RE et al.Bordetellaholmesii sp.nov.,a new gram-negative species associ-ated with septicemia.J Clin Microbiol1995;33:1–7.8.Funke G,Hess T,von Graevnitz A et al.Characteris-tics of Bordetella hinzii from a cysticfibrosis patient over a3-year period.J Clin Microbiol1996;34: 966–969.9.Porter JF,Wardlaw AC.Long-term survival ofBordetella bronchiseptica in lakewater and in bu#ered saline without added nutrients.FEMS Microbiol Lett 1993;110:33–36.10.Parton R.Bordetella.In:Collier L(ed).Topley andWilson’s Principles of Microbiology and Microbial Infections.London:Edward Arnold,1998;vol2:901–918.11.Goodwin M,Weiss AA.Adenylate cyclase toxin iscritical for colonization and pertussis toxin is critical for lethal infection by Bordetella pertussis in infant mice.Infect Immun1990;58:3445–3447.12.Weiss AA,Hewlett EL,Myers GA et al.Pertussistoxin and extracytoplasmic adenylate cyclase as viru-lence factors of Bordetella pertussis.J Infect Dis1984;150:219–222.13.Wardlaw AC,Parton R(eds).Pathogenesis andimmunity in pertussis.Chichester:John Wiley and Sons,1988;482pp.14.Heininger U,Stehr K,SchmittGrohe S et al.Clinicalcharacteristics of illness caused by Bordetella para-pertussis compared with illness caused by Bordetella pertussis.Pediatr Infect Dis J1994;13:306–309. 15.Wirsing von Konig CH,Finger H.Role of pertussistoxin in causing symptoms of Bordetella parapertussis infection.Eur J Clin Microbiol Infect Dis1994;13: 455–458.16.Parton R,Hall E,Wardlaw AC.Responses toBordetella pertussis mutant strains and to vaccina-tion in the coughing rat model of pertussis.J Med Microbiol1994;40:307–312.17.Wardlaw AC,Hall E,Parton R.Coughing rat model ofpertussis.Biologicals1993;21:27–29.18.Hall E,Parton R,Wardlaw AC.Cough production,leukocytosis and serology of rats infected intra-bronchially with Bordetella pertussis.J Med Microbiol1994;40:205–213.19.Marwick C.Acellular pertussis vaccine hailed forinfants.JAMA1995;274:446–447.20.Guiso N,Szatanik M,Rocancourt M.Protectiveactivity of Bordetella pertussis adenylate cyclase-hemolysin against bacterial-colonization.Microb Pathog1991;11:423–431.21.Betsou F,Sebo P,Guiso N.The C-terminal domain isessential for protective activity of the Bordetella pertussis adenylate cyclase-hemolysin.Infect Immun 1995;63:3309–3315.22.Weiss AA,St Mary Goodwin M.Lethal infection byBordetella pertussis mutants in the infant mouse model.Infect Immun1989;57:3757–3764.23.Cookson BT,Cho HL,Herwaldt LA,Goldman WE.Biological activities and chemical composition of purified tracheal cytotoxin of Bordetella pertussis.Infect Immun1989;57:2223–2229.24.Fernandez RC,Weiss AA.Cloning and sequencing ofa Bordetella pertussis serum resistance locus.InfectImmun1994;62:4727–4738.75Review of the biology of Bordetella pertussis25.Finn TM,Stevens LA.Tracheal colonisation factor:a Bordetella pertussis secreted determinant.MolMicrobiol1995;16:625–643.26.Locht C,Bertin P,Menozzi FD,Renauld G.Thefilamentous haemagglutinin,a multifaceted adhesin produced by virulent Bordetella spp.Mol Microbiol 1993;9:653–660.27.Willems R,Paul A,Van Der Heide H et al.Fimbrialphase variation in Bordetella pertussis:A novel mech-anism for transcriptional regulation.EMBO J1990;9: 2803–2809.28.Preston NW.Pertussis today.In:Wardlaw AC,PartonR(ed).Pathogenesis and Immunity in Pertussis.Chichester:John Wiley and Sons,1988:1–18.ler E.Current epidemiology of whooping cough inthe U.K.Biologicals1993;21:10–11.30.Stibitz S,Miller JF.Co-ordinate regulation of viru-lence in Bordetella pertussis mediated by the vir(bvg) locus.In:Miller VL,Kaper JB,Portnoy DA,Isberg RR,ed.Molecular Genetics of Bacterial Pathogenesis.Washington D.C.:American Society for Microbiology, 1994:407–422.31.Rappuoli R.Pathogenicity mechanisms of Bordetella.In:Dangl JL(ed).Bacterial Pathogenesis of Plants and Animals:Molecular and Cellular Mechanisms.Berlin:Springer Verlag,1994:319–336.32.Akerley BJ,Miller JF.Understanding signal trans-duction during bacterial infection.Trends Microbiol 1996;4:141–146.33.Stenson TH,Peppler MS.Identification of two bvg-repressed surface proteins of Bordetella pertussis.Infect Immun1995;63:3780–3789.34.Beattie DT,Shahin R,Mekalanos JJ.A vir-repressedgene of Bordetella pertussis is required for virulence.Infect Immun1992;60:571–577.35.Ewanowich CA,Sherburne RK,Man S,Peppler MS.Bordetella parapertussis invasion of HeLa229cellsand human respiratory epithelial cells in primary culture.Infect Immun1989;57:1240–1247.36.Friedman RL,Nordensson K,Wilson L et al.Uptakeand intracellular survival of Bordetella pertussis in human macrophages.Infect Immun1992;60:4578–4585.37.Guzman CA,Rohde M,Bock M et al.Invasion andintracellular survival of Bordetella bronchiseptica in mouse dendritic cells.Infect Immun1994;62:5528–5537.38.Khelef N,Guiso N.Induction of macrophage apopto-sis by Bordetella pertussis adenylate cyclase-hemolysin.FEMS Microbiol Lett1995;134:27–32. 39.Redhead K,Watkins J,Barnard A et al.E#ectiveimmunization against Bordetella pertussis respiratory infection in mice is dependent on induction of cell-mediated immunity.Infect Immun1993;61:3190–3198.40.De Moissac YR,Ronald SL,Peppler e ofpulsed-field gel electrophoresis for epidemiological study of Bordetella pertussis in a whooping cough outbreak.J Clin Microbiol1994;32:398–402.41.Syedabubakar SN,Mathews RC,Preston NW et al.Application of pulsed-field gel electrophoresis to the1993epidemic of whooping cough in the UK.Epidemiol Infect1995;115:101–113.42.Beall B,Cassiday PK,Sanden GN.Analysis ofBordetella pertussis isolates from an epidemic by pulsed-field gel electrophoresis.J Clin Microbiol1995;33:3083–3086.43.van der Zee A,Vernooij S,Peeters M et al.Dynamicsof the population structure of Bordetella pertussis as measured by IS1002-associated restriction fragment length polymorphism:comparison of pre-and post-vaccination strains and global distribution.Micro-biology1996;142:3479–3485.76R.Parton。