Characterization of the Filamentous Hemagglutinin-Like Protein FhaS in Bordetella ... · Bordetella...

INFECTION AND IMMUNITY, Aug. 2005, p. 4960–4971 Vol. 73, No. 80019-9567/05/$08.00�0 doi:10.1128/IAI.73.8.4960–4971.2005Copyright © 2005, American Society for Microbiology. All Rights Reserved.

Characterization of the Filamentous Hemagglutinin-LikeProtein FhaS in Bordetella bronchiseptica

Steven M. Julio and Peggy A. Cotter*Department of Molecular, Cellular and Developmental Biology, University of California, Santa Barbara,

Santa Barbara, California 93106-9610

Received 5 January 2005/Returned for modification 24 February 2005/Accepted 23 March 2005

Filamentous hemagglutinin (FHA) is a large (>200 kDa), rod-shaped protein expressed by bordetellae thatis both surface-associated and secreted. FHA mediates bacterial adherence to epithelial cells and macrophagesin vitro and is absolutely required for tracheal colonization in vivo. The recently sequenced Bordetella bron-chiseptica genome revealed the presence of a gene, fhaS, that is nearly identical to fhaB, the FHA structuralgene. We show that although fhaS expression requires the BvgAS virulence control system, it is maximal onlyunder a subset of conditions in which BvgAS is active, suggesting an additional level of regulation. We alsoshow that, like FHA, FhaS undergoes a C-terminal proteolytic processing event and is both surface-associatedand secreted and that export across the outer membrane requires the channel-forming protein FhaC. UnlikeFHA, however, FhaS was unable to mediate adherence of B. bronchiseptica to epithelial cell lines in vitro andwas not required for respiratory tract colonization in vivo. In a coinfection experiment, a �fhaS strain wasout-competed by wild-type B. bronchiseptica, indicating that fhaS is expressed in vivo and that FhaS contributesto bacterial fitness in a manner revealed when the mutant must compete with wild-type bacteria. These datasuggest that FHA and FhaS perform distinct functions during the Bordetella infectious cycle. A survey ofvarious Bordetella strains revealed two distinct fhaS alleles that segregate according to pathogen host range andthat B. parapertussishu most likely acquired its fhaS allele from B. pertussis horizontally, suggesting fhaS maycontribute to host-species specificity.

Members of the gram-negative genus Bordetella cause respi-ratory tract infections. Bordetella pertussis and Bordetella parap-ertussishu strains infect humans, causing the disease known aswhooping cough or pertussis, while Bordetella bronchisepticainfects a broad range of mammals, causing either asymptom-atic infections or acute diseases such as kennel cough in dogsand atrophic rhinitis in pigs (9). Several Bordetella virulencefactors, including those required for adherence to host tissues,induction and/or modulation of the immune response, andcytotoxicity, have been identified (9, 40). With the exception oftracheal cytotoxin, all are positively regulated by the two-com-ponent sensory transduction system BvgAS (39, 57, 62, 69).Under the appropriate growth conditions (37°C and relativelack of MgSO4 or nicotinic acid), BvgAS activates transcriptionof virulence genes (collectively called vag loci) and repressestranscription of genes that are not required for or are detri-mental to virulence (vrg loci) (2, 8, 10, 39). Growth at lowtemperature or in the presence of high concentrations ofMgSO4 or nicotinic acid inactivates BvgAS. Under these con-ditions, virulence genes are not expressed and Bvg-repressedphenotypes are expressed (9). When bordetellae are grown inmedium containing relatively low concentrations of MgSO4 ornicotinic acid, BvgAS controls the expression of a phase (theBvgi phase), characterized by expression of some virulencefactors (e.g., adhesins) but not others (e.g., toxins) and lack ofexpression of vrg loci (11, 39). Studies with both B. bronchisep-tica and B. pertussis have shown that Bvg-mediated control of

gene expression is critical for the development of respiratoryinfection (2, 8, 9, 39).

One of the best-characterized Bordetella virulence factors isfilamentous hemagglutinin (FHA), a long, rod-shaped, highlyimmunogenic surface-associated and secreted protein that is aprimary component of all acellular pertussis vaccines ( 53).FHA is prototypical of a class of proteins exported by thetwo-partner secretion (TPS) pathway (27, 36). It is synthesizedas a �350-kDa precursor, FhaB. Following transport acrossthe cytoplasmic membrane and loss of its unusually long signalsequence, FhaB is translocated across the outer membrane bythe channel-forming protein FhaC (22, 31, 33, 35, 51). At somepoint during this process, FhaB undergoes a C-terminal pro-teolytic cleavage event, resulting in the formation of the ma-ture �250-kDa FHA protein. A serine protease of the auto-transporter family, SphB1, has been implicated in thematuration process; however, its precise role has not beenestablished (13, 14). The mechanism by which some FHAmolecules remain surface localized while others are released isunknown.

FHA function has been investigated extensively in vitro. Itmediates adherence to epithelial cells (5, 12) and has beenshown to bind directly to ciliary membrane glycosphingolipids(59). FHA may modulate host immunity by influencing theexpression of pro-inflammatory cytokines in epithelial cells andmacrophages (1, 29). An Arg-Gly-Asp (RGD) triplet, locatedin the middle of the mature FHA protein, mediates interac-tions with leukocyte response integrin/integrin-associated pro-tein (LRI/IAP) on monocytes/macrophages, resulting in up-regulation of CR3 binding activity (28, 68). The RGD motifappears to also interact with very late antigen 5 (VLA-5) onepithelial cells, stimulating the upregulation of intercellular

* Corresponding author. Mailing address: Department of Molecu-lar, Cellular and Developmental Biology, University of California,Santa Barbara, Santa Barbara, CA 93106-9610. Phone: 805-893-5176.Fax: 805-893-4724. E-mail: [email protected].

4960

on January 12, 2021 by guesthttp://iai.asm

.org/D

ownloaded from

https://crossmark.crossref.org/dialog/?doi=10.1128/IAI.73.8.4960-4971.2005&domain=pdf&date_stamp=2005-08-01

http://iai.asm.org/

adhesion molecule 1 (ICAM-1) via an NF-�B signaling path-way (30). In vivo experiments with B. bronchiseptica and nat-ural-host animal models have shown that FHA is absolutelyrequired for tracheal colonization (12).

Determination of the complete nucleotide sequences of thegenomes of B. pertussis Tohama 1, B. parapertussishu 12822,and B. bronchiseptica RB50 revealed two additional loci, fhaLand fhaS, with the potential to encode FHA-like proteins (48).The fhaL genes from all three subspecies are nearly identicalto each other and potentially encode proteins that are approx-imately 19% similar to their cognate FhaB proteins. The fhaSgenes of B. pertussis and B. parapertussishu are also nearlyidentical to each other and potentially encode proteins with38% similarity to their cognate FhaB proteins. The fhaS geneof B. bronchiseptica, by contrast, is similar to the fhaS genes ofB. pertussis and B. parapertussishu only at its 5� and 3� ends. Itscentral region, spanning 8 kb and constituting a majority of theopen reading frame, is nearly identical (91% identity at thenucleotide level) to B. bronchiseptica fhaB. In this study, wecharacterized the B. bronchiseptica fhaS gene and its product.Our results show that while the FhaS protein appears to beprocessed and localized similarly to FhaB, its expression pat-tern and function(s) differ substantially.

MATERIALS AND METHODS

Bacterial strains, plasmids, and growth media. All strains and plasmids usedin this study are listed in Table 1. Bordetella strains were grown in Stainer-Scholte(SS) or Luria-Bertani (LB) broth as indicated or on Bordet-Gengou agar (Bec-ton Dickinson Microbiology Systems) supplemented with 7.5% defribrinatedsheep blood (Mission Laboratories). Escherichia coli strains were grown in LBbroth or on LB agar. Relevant antibiotics were used at the following concentra-tions: streptomycin, 20 �g/ml; ampicillin, 100 �g/ml; gentamicin, 20 �g/ml.

Plasmid pEG110 was constructed by ligating a 1.9-kb EcoRI-XmaI fragmentcorresponding to bp 2585 to 4465 of the B. bronchiseptica cyaA gene into EcoRI–XmaI-digested pEGZ, a pBR322-based suicide vector containing a promoterlesslacZ gene (39). In a similar manner, a 1.6-kb XhoI-BamHI fragment correspond-ing to bp 1316 to 2927 of the B. bronchiseptica fhaB gene was ligated into pEGZto create plasmid pEG111. Plasmid pSJ22 was constructed by amplifying a323-bp product corresponding to bp 7 to 330 from the B. bronchiseptica fhaCgene by use of primers fhaC.fwd (5�-GATTGAATTCGACGCAACGAACCGTTTCCGG-3�) and fhaC.rev (5�-GATTGGATCCAACGGCGCGGGGTCGAAC-3�) (Integrated DNA Technologies, Coralville, IA). The primers were de-signed such that EcoRI and BamHI sites would be incorporated at the ends ofthe PCR product. The PCR product was digested with EcoRI and BamHI,purified, and ligated to EcoRI–BamHI-digested pEGZ.

DNA methods. Isolation of plasmid DNA and genomic DNA, restrictionendonuclease digestions, agarose gel electrophoresis, and DNA ligations wereperformed using standard methods (52). Restriction enzymes, T4 DNA ligase,Taq polymerase, and high-fidelity Phusion polymerase were purchased fromPromega Corp. (Madison, WI), New England Biolabs (Beverly, MA), or MJResearch (Waltham, MA) and were used according to the manufacturers’ in-structions.

Construction of bacterial strains. To create an in-frame (nonpolar) deletion inthe B. bronchiseptica fhaS gene, an allelic exchange system employing the sacBgene was used as previously described (2, 39). Initially, an allelic exchangeplasmid, pEG7S�fhaS, was constructed as follows. Primers EBPSFHA3x (5�-AATCTAGAACCCGCAACCAGTCCGAGAAGTACG-3�) and EBPSFHA4x(5�-AAGGTACCCGCGTTCATGTTGTCTCCTGATGTCG-3�) were used toamplify a 672-bp fragment corresponding to the 5� region and containing the firstthree codons of the fhaS coding sequence. Primers were designed such that XbaIand KpnI sites are introduced at the 5� and 3� ends of the PCR product,respectively. Primers EBPSFHA1x (5�-AAGGTACCCTGCATCGCTAGCGCGAGG-3�) and EBPSFHA2 (5�-AAGCATGCCGCAGGACATCGACATAGGG-3�) were used to amplify a 679-bp fragment including the last three codons ofthe fhaS coding sequence and downstream sequences. Primers were designedsuch that KpnI and SphI sites were introduced at the 5� and 3� ends of the PCRproduct, respectively. These fragments were purified, ligated into TA cloning

vectors (Invitrogen), digested with the appropriate restriction enzymes, andcloned into BamHI–XbaI-digested pEG7S, a derivative of pEG7 (2), resulting inan in-frame deletion of all but the first three and last three codons of the fhaSgene. Sequence analysis of the cloned DNA (Laragen, Los Angeles, CA) wasperformed to confirm the fidelity of B. bronchiseptica sequences. The resultingplasmid (pEG7S�fhaS) was transformed into E. coli SM10 �pir and then mobi-lized into the RB50 chromosome by conjugation. Cointegrants were selected onmedium containing streptomycin (B. bronchiseptica strains are naturally strepto-mycin resistant) and gentamicin. Recombinants that had excised the plasmidfrom the chromosome were selected on media containing 10% sucrose (the sacBgene encodes for levansucrase, which is toxic to cells grown in the presence ofhigh concentrations of sucrose) (19), and PCR was used to screen for mutantsthat had undergone a recombination event resulting in a chromosomal deletionof the fhaS gene. A �fhaB �fhaS double mutant was constructed by the samemethod by mobilizing pEG7S�fhaS into RBX9.

To construct strains that harbor chromosomal lacZ reporter plasmids or fhaCnull mutations, pEG110, pEG111, or pSJ22 was transformed into SM10 �pir andthen conjugated into the appropriate B. bronchiseptica strain. Cointegrants wereselected on media containing streptomycin and gentamicin. PCR was used toverify integration of the plasmid at the proper chromosomal locus. For strainscontaining pEG111, PCR was used to distinguish between integration events ateither the fhaB or fhaS locus, since the 1.7-kb fhaB fragment in pEG111 is nearlyidentical to sequences present in the fhaS gene.

�-Galactosidase assays. Stationary phase cultures from strains harboring lacZtranscriptional fusions were used to measure -galactosidase activity. Bacteriawere grown overnight in either LB broth or SS media and permeabilized by theaddition of sodium dodecyl sulfate (SDS) and CHCl3, and -galactosidase ac-tivity was determined essentially as described previously (39), except that mea-surements were taken using a Victor3 1420 microplate reader (PerkinElmer LifeSciences, Boston, MA).

Bacterial cell fractionation and immunoblot analysis. Stationary-phase cul-tures were used for all bacterial cell preparations. Whole-cell lysates were pre-pared as described previously (39). Outer membrane preparations were isolatedfrom bacterial cell sonicates as the Triton X-100 insoluble fraction. Briefly,clarified cell sonicates were spun at 16,000 g, the pellet was resuspended in 2%Triton X-100 and spun again, and the pellet was resuspended in an appropriatevolume of 10 mM HEPES (pH 7.5). Secreted proteins were concentrated byaddition of 10% TCA to the culture supernatant and then recovered by centrif-ugation. Proteins were resuspended in an equal mixture of 1 SDS-polyacryl-amide gel electrophoresis (PAGE) sample buffer (52) and saturated Tris solution(to neutralize excess acid).

Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) wasperformed by the method of Laemmli (38) using denaturing 3 to 8% lineargradient gels (acrylamide-bisacrylamide, 29:1). Proteins from whole cells (de-rived from approximately 8.0 108 CFU), outer membranes (derived from 5.0 109 CFU), or supernatant fractions (derived from 1.5 1010 CFU) were runon SDS-PAGE, transferred to nitrocellulose membranes (Schleicher and SchuellBioScience, Dassel, Germany), and incubated with a 1:10,000 dilution of chickenpolyclonal FHA antibody (raised against amino acids [aa] 1048 to 1385 of B.bronchiseptica FHA, which are identical [except for one amino acid difference]with the corresponding sequence in FhaS). Antigen-antibody complexes weredetected by incubating with a 1:20,000 dilution of goat anti-chicken IRdye 800-conjugated secondary antibody (Rockland, Gilbertsville, PA) and visualized us-ing the Odyssey infrared imaging system (Li-Cor Biosciences, Lincoln, NE).

In vitro adherence assay. Adherence to rat lung epithelial (L2) cells wasperformed as described previously (12), with the exception that LB broth wasused as the bacterial growth medium as well as the diluent for addition ofbacteria to cell monolayers.

In vitro competition experiment. Wild-type (RB50G) and �fhaS strains weregrown overnight in either SS or LB and then diluted by coinoculating equivalentnumbers of each (based on measurements of optical density at 600 nm) into freshmedia to an optical density at 600 nm of approximately 0.1. Cultures were thenincubated on a roller at 37°C for 8 h (approximately five cell divisions). Therelative proportion of wild-type to �fhaS bacteria was determined by platingaliquots onto media with and without gentamicin.

In vivo competition experiment. Three- to 4-week-old female Wistar rats(Charles River Laboratories, Wilmington, MA) were coinoculated intranasallywith approximately 250 CFU of the wild-type and 250 CFU of the �fhaS strain(from cultures grown overnight in SS media) essentially as previously described(41). Animals were sacrificed 7, 14, and 28 days postinoculation, and colonizationlevels in the nasal cavity and trachea were determined as described previously(41).

VOL. 73, 2005 FhaS IN B. BRONCHISEPTICA 4961


.org/D

ownloaded from

http://iai.asm.org/

Phylogenetic analysis of fhaS alleles. Total genomic DNA was prepared fromvarious B. bronchiseptica strains, B. pertussis strain Tohama I, and B. parapertus-sishu strain 12822 and used as templates for PCR. Primers PfhaSOUT1 (5�-TGGCGGCCCGCTATATCGA-3�) and CD64 (5�-GGCTGAGCGGGCCGCGCTGG-3�) were used to amplify B. bronchiseptica RB50-specific fhaS sequences, andprimers PfhaSOUT1 and fhaSBpBpp1086.rev (5�-GCCGGCATGCAGCGACAAGGCGTC-3�) were used to amplify fhaS sequences specific to B. pertussis andB. parapertussishu. Cycling conditions were as follows: 94°C for 2 min followed by30 cycles of 94°C for 30 s, 65°C for 30 s, and 72°C for 2 min, with a final extensionstep of 72°C for 10 min.

Statistical analyses. For animal experiments, significant differences betweenthe ratio of mutant to wild-type CFU present in the inocula and the ratio ofmutant to wild-type CFU recovered from host tissues were examined using aStudent’s t test. A P value of �0.05 was considered statistically significant.

RESULTSThe fhaS gene of B. bronchiseptica, but not those of B. per-

tussis or B. parapertussishu, codes for a protein that is nearlyidentical to FHA. The annotated genome sequences of B. per-tussis, B. parapertussishu and B. bronchiseptica indicate the pres-ence of a locus, named fhaS, in all three subspecies that ispredicted to encode a protein with similarity to the well-char-acterized Bordetella virulence factor FHA (48). Comparison ofthe predicted FhaS proteins from the three subspecies, how-ever, revealed that the amino acid sequence predicted for FhaSfrom B. bronchiseptica (FhaSBb) differs substantially from

TABLE 1. Bacterial strains and plasmids

Strain or plasmid Relevant genotype/description Source or reference

StrainsB. bronchiseptica

RB50 Wild type 10RB50G Gentamycin-resistant derivative of RB50 6RBX9 �fhaB 12RBX11 �fhaS This studyRBX20 �fhaB �fhaS This studyRB50::pEG111(fhaB) RB50 containing a chromosomal fhaB::lacZ fusion This studyRB50::pEG111(fhaS) RB50 containing a chromosomal fhaS::lacZ fusion This studyRB50::pEG110 RB50 containing a chromosomal cyaA::lacZ fusion This studyRB50::pTEN34 RB50 containing a chromosomal bipA::lacZ fusion 20RB53::pEG111 (fhaB) Bvg-constitutive containing a chromosomal fhaB::lacZ fusion This studyRB53::pEG111(fhaS) Bvg-constitutive containing a chromosomal fhaS::lacZ fusion This studyRB53::pEG110 Bvg-constitutive containing a chromosomal cyaA::lacZ fusion This studyRB53::pTEN34 Bvg-constitutive containing a chromosomal bipA::lacZ fusion 20RB54::pEG111(fhaB) Bvg� locked containing a chromosomal fhaB::lacZ fusion This studyRB54::pEG111(fhaS) Bvg� locked containing a chromosomal fhaS::lacZ fusion This studyRB54::pEG110 Bvg� locked containing a chromosomal cyaA::lacZ fusion This studyRB54::pTEN34 Bvg� locked containing a chromosomal bipA::lacZ fusion 20RB53i::pEG111(fhaS) Bvgi locked containing a chromosomal fhaS::lacZ fusion This studyRB53i::pTEN34 Bvgi locked containing a chromosomal bipA::lacZ fusion 20RB50::pSJ22 RB50 containing plasmid insertion (null mutation) in fhaC This studyRBX9::pSJ22 �fhaB containing plasmid insertion (null mutation) in fhaC This studyRBX11::pSJ22 �fhaS containing plasmid insertion (null mutation) in fhaC This study548 Pig isolate 64590 Dog isolate 64595 Dog isolate 64625 Rat isolate 64675 Human isolate 64676 Pig isolate 64704 Rabbit isolate 65Cb2 Dog isolate 20JC100 Human isolate 20

B. pertussisTohama l Wild type 37

B. parapertussishu12822 Wild type 26

E. coliDH5 Molecular cloning strain BRL; Gaithersburg, MDSM10�pir Conjugation strain 47

PlasmidspEG110 pBR322-based Bordetella suicide vector containing 1.8 kb cyaA gene

fragment 5� to promoterless lacZ gene; Ampr, GmrThis study

pEG111 pBR322-based Bordetella suicide vector containing 1.7 kb fhaB genefragment 5� to promoterless lacZ gene; Ampr, Gmr

This study

pSJ22 pBR322-based Bordetella suicide vector containing 0.3 kb kb fhaCgene fragment 5� to promoterless lacZ gene; Ampr, Gmr

This study

4962 JULIO AND COTTER INFECT. IMMUN.


.org/D

ownloaded from

http://iai.asm.org/

those predicted for FhaS from B. pertussis and B. parapertus-sishu (FhaSBp and FhaSBpp, respectively). The predicted aminoacid sequences of FhaSBp and FhaSBpp are �90% identicalover their entire length (Fig. 1A). By contrast, FhaSBb is pre-dicted to be significantly larger (3206 aa compared to 2601 and2554 for FhaSBp and FhaSBpp, respectively) and its similaritywith FhaSBp and FhaSBpp is confined primarily to the N and Ctermini: the sequences of the predicted N-terminal 148 aa andC-terminal 933 aa of FhaSBb and FhaSBpp are 97% and 98%identical, respectively. The central region of FhaSBb, however,is only 36% identical to that of FhaSBpp. Comparison ofFhaSBb with FhaBBb, however, revealed striking similarity(Fig. 1B). From aa 149 to aa 2658, which constitutes nearly80% of the predicted protein, FhaSBb is nearly identical toFhaBBb: exclusive of a gap of eight 19-aa repeat segments inFhaSBb, these sequences are 95% identical. This region of nearidentity includes several functional domains previously identi-fied in FHA, including the heparin binding domain (HBD),carbohydrate recognition domain (CRD), RGD, and C-termi-nal maturation site (14, 44, 49, 50, 54). Conversely, the extremeN and C termini of FhaSBb and FhaBBb exhibit little similarity(Fig. 1B). Overall, therefore, the predicted amino acid se-quence of FhaSBb is much more closely related to that ofFhaBBb than it is to those of FhaSBp and FhaSBpp.

Expression of fhaS. Expression of all Bordetella virulencegenes characterized so far is positively regulated by the BvgASphosphorelay and is maximal in cells grown at 37°C in SS broth(Bvg� phase conditions) and minimal in cells grown at roomtemperature in SS broth or at 37°C in SS broth supplementedwith �20 mM MgSO4 or nicotinic acid (Bvg� phase or “mod-

ulating” conditions). Expression of the Bvgi phase gene bipA isboth positively and negatively regulated by BvgAS; it is maxi-mal in cells grown at intermediate temperatures in SS broth orat 37°C in SS broth containing “semimodulating” concentra-tions of MgSO4 or nicotinic acid (Bvgi phase conditions) (71).To determine the expression pattern of fhaS in B. bronchisep-tica, we constructed a strain containing a chromosomal fhaS-lacZ fusion and measured -galactosidase activity in culturesgrown at 37°C � 50 mM MgSO4. fhaS-lacZ expression wasapproximately 2.5-fold greater in cells grown in SS broth thanin cells grown in SS broth � MgSO4. This expression patternwas similar in profile, but not magnitude, to Bvg-activatedgenes such as fhaB and cyaA (Fig. 2). To determine whetherfhaS expression is controlled by BvgAS, we introduced thefhaS-lacZ fusion into Bvg� and Bvg– phase-locked derivativesof RB50. Expression of fhaS-lacZ in the Bvg� phase-lockedstrain grown in SS broth with or without added MgSO4 wassimilar to fhaS-lacZ expression in the wild-type strain grown inSS broth. fhaS-lacZ expression was the same in the Bvg– phase-locked strain grown in SS � MgSO4 as in the wild-type straingrown in SS � MgSO4 (Fig. 2). These results demonstrate thatfhaS is positively regulated by BvgAS.

We discovered serendipitously that fhaS expression is ap-proximately fivefold greater in cells grown in LB broth than incells grown in SS broth (Fig. 2). Since we had previously ob-served that wild-type B. bronchiseptica showed differences incolony morphology and motility when grown on LB agar, con-sistent with modulation to the Bvgi or Bvg– phase (not shown),we hypothesized that LB might be semimodulating. Consistentwith this hypothesis, expression of the gene encoding adenylate

FIG. 1. Similarity of the predicted B. bronchiseptica FhaS protein (FhaSBb) to related Bordetella proteins. (A) Comparison of FhaSBb to thepredicted FhaS protein of B. parapertussis (FhaSBpp) and comparison of FhaSBpp with the predicted FhaS protein of B. pertussis (FhaSBp).(B) Comparison of FhaSBb to full-length B. bronchiseptica FHA (FhaBBb). The entire FhaSBb protein and regions in other proteins similar toFhaSBb are depicted in gray. White and stippled segments denote regions that are not significantly similar to FhaSBb. Thin dashed lines betweentwo proteins denote the boundaries of regions of similarity, and the percentage amino acid identity within that region is shown. The amino acidnumber at the junctions of the various regions and at the end of each protein is indicated. Thick vertical dashed lines represent the predicted Nand C termini of mature proteins. A horizontal dashed line in FhaSBb depicts the absence of eight 19-amino acid repeat segments that are presentin FhaBBb. HBD, heparin binding domain; RGD, Arg-Gly-Asp integrin binding domain; CRD, carbohydrate recognition domain.



.org/D

ownloaded from

http://iai.asm.org/

cyclase, cyaA, (which is minimal under Bvgi phase conditions)(11), was decreased in LB-grown cultures compared with SS-grown cultures, and expression of bipA (which is maximal un-der Bvgi phase conditions) (18) was slightly greater in LB-grown cultures than in SS-grown cultures (Fig. 2). Theseresults suggested that fhaS, like bipA, might be expressed max-imally under Bvgi phase conditions. However, fhaS-lacZ ex-pression was greater in the Bvg� phase-locked strain than inthe Bvgi phase-locked strain, and expression in both of thesestrains was greater when cells were grown in LB than in SS(Fig. 2). fhaS therefore does not appear to be a Bvgi phasegene like bipA (compare fhaS-lacZ expression with bipA-lacZexpression in the various strain backgrounds and media in Fig.2). Instead, fhaS appears to be expressed differently than allpreviously characterized Bvg-regulated genes; it is expressedmaximally only under a subset of Bvg-activating conditions.Inspection of the 246-nucleotide sequence between the end ofthe gene upstream of fhaS (purL) and the fhaS initiation codonrevealed no discernible BvgA binding sites, suggesting Bvg-mediated control of fhaS expression is indirect.

Cellular localization of FhaS. In Bordetella strains grown invitro, mature FHA can be detected in concentrated culturesupernatants, outer membrane fractions, and whole-cell ly-sates, while preprocessed FhaB can be detected only in whole-cell lysates. Because of the extensive similarity between FhaSand FhaB, we hypothesized that FhaS is processed and local-ized in a manner similar to FhaB. To test this hypothesis, weperformed Western blot analysis on proteins derived fromwhole-cell, outer membrane, and supernatant fractions fromwild-type, �fhaB, �fhaS, and �fhaB �fhaS B. bronchisepticastrains grown overnight in either LB or SS. Membranes wereprobed with an anti-FHA antibody generated against aminoacids 1048 to 1385 of B. bronchiseptica FHA, a region sur-rounding the RGD motif in which FhaS and FHA are pre-dicted to differ by only one amino acid. In whole-cell lysates,this antibody detected many polypeptides, with one of �350kDa and one of �260 kDa being most prominent in the wild-type and �fhaS strains grown in either LB or SS (Fig. 3).Because only the highest-molecular-weight polypeptides weredetected when an antibody against the C terminus of FhaB was

FIG. 2. Comparison of fhaS expression patterns with those of known Bvg-regulated genes. An fhaS::lacZ, fhaB::lacZ, bipA::lacZ, or cyaA::lacZtranscriptional fusion was introduced into the chromosome of wild type (WT), Bvg� phase locked (Bvgc), Bvgi phase locked (Bvgi) or Bvg� phaselocked (Bvg�) B. bronchiseptica strains as indicated. Bacteria were grown overnight in Stainer-Scholte (SS) or L broth (LB) in the presence (�)or absence (–) of 50 mM MgSO4. -galactosidase activity is reported in Miller units, and data represent the average of three independentexperiments done in duplicate. Error bars indicate one standard deviation from the mean.



.org/D

ownloaded from

http://iai.asm.org/

used (data not shown), we can conclude that the �350-kDaand �250-kDa polypeptides correspond to unprocessed FhaBand mature FHA, respectively. In whole-cell lysates of thewild-type and �fhaB strains, prominent polypeptides of ap-proximately 230 to 240 kDa were detected when the cells weregrown in LB. Similarly sized polypeptides were only minimallydetectable when these strains were grown in SS (Fig. 3). Theabsence of these polypeptides in lysates of the �fhaB �fhaSdouble mutant strongly suggests that they are FhaS. Cleavagewithin the amino acids that are identical to those in FhaB thatare predicted to represent the C-terminal maturation sitewould result in maturation of the 328-kDa full-length FhaSprotein into a mature protein of �233 kDa. It therefore ap-pears that the 230-to-240-kDa polypeptides present in lysatesof the wild-type and �fhaB strains, but absent in the �fhaS and�fhaS �fhaB strains, represent a processed, mature form of FhaS.

In outer membrane preparations, the anti-FHA antibodydetected distinct polypeptides from 230 to 260 kDa in thewild-type and �fhaS strains grown in SS, corresponding toFHA. Polypeptides with the same mobility were present inthese strains grown in LB; however, several additional promi-nent polypeptides ranging in size from �180 to 240 kDa weredetected in the wild-type and �fhaB strains grown in this me-dium (but not in SS). Because these polypeptides were absentin the �fhaB �fhaS double mutant, they most likely correspondto FhaS. Moreover, these data suggest that, like FHA, severaldistinct forms of FhaS are localized to the outer membrane, oralternatively, there is significant proteolysis of FHA and FhaSin outer membrane preparations.

The anti-FHA antibody also detected several polypeptidesin concentrated culture supernatants. At least five polypeptidesranging from �150 to 260 kDa were detected in supernatantsprepared from the wild-type and �fhaS strains when the cellswere grown in SS (Fig. 3). The top three bands were alsopresent in these strains when grown in LB, indicating that FHAis efficiently secreted from the cell under both growth condi-tions. In addition, polypeptides of �230 to 240 kDa weredetected in the wild-type and �fhaB strains when the cells weregrown in LB (and were only minimally detected from the�fhaB strain when the cells grown in SS), demonstrating thatFhaS is also secreted from the cell, and in amounts similar toFHA when the cells are grown in LB.

Export of FhaS is dependent on FhaC. FhaC is an outermembrane channel-forming protein required for the export ofFhaB to the cell surface (22, 31, 33, 51). Sequences withinFhaB required for FhaC-dependent export are located withinthe N-terminal 300 aa (the so-called “secretion domain”) (7,36). Although the first �150 aa of FhaS and FhaB are notidentical, they are moderately similar and two amino acid mo-tifs (NPGL and NPNG) that have been shown to be critical forexport of FhaB (32) are present in the same relative location inFhaS. Moreover, a search of the B. bronchiseptica genomerevealed no other open reading frames with similarity to FhaC.We therefore hypothesized that FhaS is exported via FhaC.Previous experiments have shown that FhaB/FHA is undetect-able in strains containing loss-of-function mutations in fhaC,presumably because it is degraded in the periplasm (32, 70).Using Western blot analysis, we found that polypeptides cor-responding to FHA and/or FhaS were detectable in whole-celllysates of the wild-type, �fhaB, and �fhaS strains grown in LBbut were undetectable in lysates of strains containing disrup-tion mutations in fhaC (Fig. 4). These results demonstrate thatlike FHA, FhaS is dependent on FhaC, and strongly suggestthat FhaS is exported across the outer membrane by FhaC. B.bronchiseptica, therefore, appears to use the same protein forthe export of both FHA and FhaS.

Evaluation of FhaS-mediated adherence of B. bronchisepticato epithelial cells. The role of B. pertussis FHA as an adhesinis well documented (4, 25, 28, 43, 44, 49, 50, 54, 59–61). Ad-ditionally, it has been demonstrated that FHA is both neces-sary and sufficient to mediate B. bronchiseptica adherence to arat lung epithelial (L2) and other cell lines in vitro (12, 41).FhaS shares a high degree of similarity with FHA in regionsthat have been implicated in mediating adherence to eukary-otic cells or cell surface components, including the CRD(�99% identical), which directs binding to glycolipids and cil-

FIG. 3. FhaS localizes to the outer membrane and is released intothe culture supernatant. Proteins from whole cell (A), outer membrane(B), and supernatant (C) fractions were prepared from the wild-type(WT) and the indicated mutant strains grown in either LB or Stainer-Scholte (SS) broth and subjected to immunoblot analysis using a poly-clonal antibody to B. bronchiseptica FHA, which also recognizes FhaS.The positions of molecular weight markers (in kDa) are shown.



.org/D

ownloaded from

http://iai.asm.org/

iated cells (49), and the HBD (99% identical), which promotesbinding to highly sulfated polysaccharides, structures com-monly found on eukaryotic cell surfaces (23). The integrin-binding RGD motif is also conserved between FhaS and FHA.To test the hypothesis that FhaS can mediate binding of B.bronchiseptica to eukaryotic cells, we utilized L2 cells in a

standard in vitro adherence assay with a �fhaB strain grownovernight in LB in order to induce maximal expression of FhaSin the absence of FHA. Under these growth conditions, FhaSis produced and localized to the cell surface in amounts similarto FHA (Fig. 3). Although wild-type RB50 grown in LB ad-hered efficiently to the L2 cells (Fig. 5A), the �fhaB strain wasseverely defective in adherence (Fig. 5B), indicating that, atleast under these experimental conditions, FhaS was not ableto mediate binding of B. bronchiseptica to L2 cells. To deter-mine whether FhaS contributes to adherence in the presenceof FHA (i.e., that both FHA and FhaS are required for ad-herence of LB-grown bacteria to epithelial cells), we comparedthe binding ability of the �fhaS strain with that of wild-type B.bronchiseptica. The �fhaS strain adhered to L2 cells as effi-ciently as RB50, suggesting that binding of the wild-type strainto L2 cells, even when grown in LB, is mediated primarily byFHA (Fig. 5C). We obtained similar results using a humanbronchial epithelial (BEAS-2B) cell line (data not shown).FhaS, therefore, does not appear to be important for mediat-ing adherence to epithelial cells—at least not in a manner thatwe can detect with these cultured cell assays.

Role of FhaS in respiratory tract infection. To evaluate thecontribution of fhaS to respiratory tract infection, we per-formed both single-infection and coinfection experiments.When inoculated into separate animals, the �fhaS strain didnot differ from RB50 in its ability to colonize the nasal cavitiesand tracheas of Wistar rats at days 14 and 28 post infection

FIG. 4. Export of FhaS is dependent on FhaC. Whole cell lysateswere prepared from LB-grown wild-type (WT) and �fhaS, �fhaB, and�fhaB �fhaS mutant strains in isogenic fhaC� (�) and fhaC� (�)backgrounds. Proteins were subjected to immunoblot analysis andprobed with anti-FHA antibody, which also recognizes FhaS. Thepositions of molecular mass markers (in kDa) are shown.

FIG. 5. FhaS does not mediate adherence of B. bronchiseptica to rat lung epithelial (L2) cells. Bacteria were grown overnight in LB and addedto a monolayer of L2 cells in a standard adherence assay. L2 cells were incubated with wild-type RB50 (A); �fhaB (B); �fhaS (C); or L broth alone(D). Data are representative of similar results obtained from three independent experiments. Magnification is 1,000.



.org/D

ownloaded from

http://iai.asm.org/

(data not shown). However, when coinoculated into the sameanimal with RB50G, a derivative of our wild-type strain ex-pressing a gentamicin resistance gene (6), a defect for the�fhaS strain was revealed. In these experiments, groups of ratswere inoculated intranasally with 5 �l of phosphate-bufferedsaline containing approximately 250 CFU each of wild-typeand mutant bacteria, and the number of wild-type and mutantbacteria recovered from the nasal cavities and tracheas ofinfected animals was determined 7, 14, and 28 days postinocu-lation. The competitive index (CI) was calculated as the ratioof mutant to wild-type bacteria recovered divided by the ratioof mutant to wild-type bacteria administered; thus, a CI of lessthan 1 indicates a virulence defect for the mutant strain. Ondays 7 and 14 postinoculation, although the total CFU recov-ered (shown at the top of each panel in Fig. 6) was comparable

to that from animals infected with wild-type bacteria alone, the�fhaS strain was significantly outcompeted by the wild-typestrain in both the nasal cavity and trachea (Fig. 6). By day 28,however, although in the majority of animals the number ofCFU of the �fhaS mutant recovered was far lower than thatseen with the wild-type strain, in a few cases the number of�fhaS CFU isolated was close to or higher than the number ofwild-type CFU isolated. The �fhaS strain was not outcompetedby RB50G when coinoculated into SS or LB broth and grownfor several generations in vitro (see Materials and Methods).Together, these data indicate that fhaS contributes to bacterialfitness in vivo in a manner revealed when the �fhaS strain mustcompete with wild-type bacteria.

Comparison of fhaS alleles among B. bronchiseptica strains.Consistent with the predicted amino acid sequence comparisondescribed above, nucleotide sequences at the 5� and 3� ends ofthe fhaS loci of B. bronchiseptica RB50, B. pertussis Tohama 1,and B. parapertussishu 12822 are nearly identical, while thosecorresponding to a majority of the central portion of the openreading frame in RB50 differ substantially from those of To-hama 1 and 12822 (Fig. 7, top panel). Sequences similar to thecentral region of fhaS from Tohama 1 and 12822 are notpresent anywhere within the RB50 genome (48). Because B.pertussis and B. parapertussishu strains appear to have divergedindependently from B. bronchiseptica or a B. bronchiseptica-like ancestor (21, 64), these observations suggest that either B.pertussis and B. parapertussishu strains acquired their fhaS al-leles independently after diverging from B. bronchiseptica orthat both subspecies evolved from a B. bronchiseptica(-like)strain that possessed an fhaS allele similar to those present inTohama 1 and 12822. To investigate these possibilities, weused PCR to identify the fhaS alleles present in B. bronchisep-tica strains representing various clades within the B. bronchi-septica cluster (20, 64). Interestingly, although nearly identicalpurL homologs are present 5� to fhaS in RB50, Tohama I, and12822, the intergenic sequences differ substantially in TohamaI compared to RB50 and 12833 (Fig. 7, top panel). Therefore,PCR was performed using a primer designed to hybridize tosequences at the 3� end of purL. Genomic DNA was preparedfrom the various B. bronchiseptica strains, B. pertussis TohamaI, and B. parapertussishu 12822 and used as template DNA forPCR. Using primers 1 and 2 (Fig. 7, top panel), a 1.4-kb DNAfragment was amplified from RB50 and all other B. bronchi-septica strains tested (except strain 675, in which only a slightlylarger faint band could be detected that was determined byDNA sequence analyses to not correspond to fhaSBb), but noPCR product was amplified from Tohama 1 or 12822 (Fig. 7,bottom panel). Conversely, PCR using primers 1 and 3 (Fig. 7,top panel) resulted in the amplification of a 1.6-kb DNA frag-ment only from Tohama 1 and 12822 (Fig. 7, bottom panel).These results suggest that all B. bronchiseptica strains containa common fhaS allele (or no fhaS allele at all) and support thehypothesis that the fhaS allele present in B. pertussis and B.parapertussishu was acquired by these strains independentlyafter they diverged from a B. bronchiseptica(-like) ancestor.

DISCUSSION

FHA is prototypical of proteins secreted by the two-partnersecretion (TPS) pathway (36). Many of these large exoproteins,

FIG. 6. �fhaS strains are out-competed by wild-type B. bronchisep-tcia in vivo. Wistar rats were inoculated intranasally with approxi-mately 250 CFU of both the wild-type (RB50G) and �fhaS strains.Animals were sacrificed on the indicated day postinoculation, and thenumber of wild-type and �fhaS bacteria present in the nasal cavity andtrachea was determined. Each symbol represents the competitive index(CI) from the indicated tissue of a single animal, which was calculatedas the ratio of mutant to wild-type bacteria recovered divided by theratio of mutant to wild-type bacteria inoculated. A CI value of 1 (thindashed line) indicates equal numbers of wild-type and �fhaS bacteriawere recovered. Symbols below the thick dashed line represent caseswhere no �fhaS bacteria were present in at least 500 total coloniesrecovered. The average number of CFU recovered from each tissue �1 standard error of the mean (log10) is shown at the top of each panel.*, P � 0.05.



.org/D

ownloaded from

http://iai.asm.org/

which can be surface-associated, secreted into the extracellularmilieu, or both, have postulated or proven roles in virulence(34, 36). In some cases, two exoproteins are encoded within thesame genome, e.g., the HMW1 and HMW2 adhesins fromHaemophilus influenzae and the LspA1 and LspA2 proteinsfrom Haemophilus ducreyi (55, 67). Although in each pathogenthese proteins share a high degree of similarity to one another,they are thought to play distinct roles in virulence, due todifferences in expression profile, protein functionality, or both(16, 17, 55, 56, 63, 66, 67). In this study, we characterized thefhaS gene from B. bronchiseptica, which encodes a protein thatis nearly identical to that encoded by the fhaB gene. We showthat fhaS is expressed maximally under different growth con-ditions than have been described for all other characterizedBordetella virulence genes. We also show that FhaS is exported(via FhaC), processed, and localized in a manner similar tothat of FhaB. However, despite the extremely high degree ofamino acid similarity between mature FhaS and FHA, FhaSdid not mediate bacterial adherence to cultured epithelial cellsin vitro and was apparently dispensable for tracheal coloniza-

tion in vivo. We hypothesize, therefore, that similarly to whathas been proposed for the HMW and Lsp proteins, FHA andFhaS may play distinct roles in the Bordetella infectious cycle.

Although fhaS expression requires BvgAS, the fact that it ismuch greater in LB than SS suggests that fhaS belongs to aclass of Bordetella genes different from those that have beencharacterized thus far. Since inspection of sequences upstreamof the start of translation failed to reveal a discernible consen-sus or near-consensus BvgA binding site(s), we hypothesizethat BvgAS control of fhaS expression is indirect. Althoughother possibilities exist, the simplest explanation for these datais that BvgAS controls a regulator(s) that controls fhaS expres-sion and that this intermediate regulator(s) is differentiallyactive in LB and SS. Indirect control of gene expression byBvgAS in Bordetella is not unprecedented. For example, BvgRcontrols the expression of vrg loci (45, 46), FrlAB is a masterregulator required for the synthesis and assembly of flagella inthe Bvg� phase of B. bronchiseptica (3), and the regulatorsencoded by the btr locus control expression of the type IIIsecretion system (42). However, fhaS is to our knowledge the

FIG. 7. Distribution of fhaS alleles among B. bronchiseptica strains. (Upper panels) Genetic organization of the fhaS loci in B. bronchiseptica(Bb), B. parapertussis (Bpp), and B. pertussis (Bp). Open reading frames (thick lines) and intergenic DNA (thin lines) are shown with nucleotidenumbering relative to the first codon of fhaS. The intergenic region upstream of fhaS in B. pertussis is hatched to indicate that its sequence issignificantly different from those of B. bronchiseptica and B. parapertussishu. The percentage of identity of nucleotide sequences between variousregions (thin dashed lines) is shown. The entire B. bronchiseptica fhaS gene and regions of nucleotide similarity of the B. parapertussis and B.pertussis fhaS genes to B. bronchiseptica fhaS are shown in dark gray. White segments denote nucleotide sequences in the B. parapertussis and B.pertussis fhaS genes that are similar to each other but dissimilar to B. bronchiseptica fhaS. The purL gene is located upstream of fhaS in all threesubspecies. Highly similar open reading frames are located downstream of fhaS in B. bronchiseptica and B. parapertussis. By contrast, an openreading frame encoding an insertion sequence transposase (IS481) is located immediately downstream of fhaS in B. pertussis. Approximate bindingsites of primers used to amplify fhaS sequences are shown. Primer PfhaSOUT1 (Primer 1) hybridizes to the 3� end of purL in all three subspecies.Primer CD64 (Primer 2) hybridizes to RB50-specific fhaS sequences, and primer fhaSBpBpp1086rev (Primer 3) hybridizes to B. pertussis- and B.parapertussis-specific fhaS sequences. (Lower panels) 0.8% agarose gel electrophoresis of PCR products obtained from Bordetella genomic DNAtemplates. Bb, various B. bronchiseptica strains; Bp, B. pertussis strain Tohama I; Bpp, B. parapertussishu strain 12822. The upper gel shows PCR resultsusing primers 1 and 2; the lower gel shows PCR results using primers 1 and 3. Molecular mass (MM) markers (in kilobases) are shown at the right.



.org/D

ownloaded from

http://iai.asm.org/

first example of a gene that is potentially subject to interme-diate-regulatory control and is expressed in response to a dif-ferent set of environmental conditions than all other knownBvg-regulated genes. An intriguing possibility is that fhaS be-longs to a subset of virulence genes whose coordinate expres-sion is activated in response to specific temporal or spatialenvironmental conditions in vivo that are different from thosethat induce the expression of known Bordetella virulence genes.We are currently attempting to identify the hypothesized fhaSintermediate regulator and other gene(s) under its control inorder to determine whether such a regulatory pathway exists.

In addition to revealing differences in expression pattern,our in vitro analyses also suggest that FhaS functions differ-ently than FHA. FhaS failed to mediate or contribute to bind-ing of B. bronchiseptica to two different epithelial cell lines,under conditions in which FHA was proficient at promotingbacterial adherence. This result was surprising given the highdegree of similarity between FhaS and FHA, especially withinthe domains that have been implicated in binding to cell sur-face components (i.e., the HBD, CRD, and RGD). These datasuggest that the amino acids within FHA that are important forbinding eukaryotic cells are those that are different betweenFhaS and FHA. Interestingly, most of the amino acids that aredifferent between the two mature proteins are localized to theextreme C termini, a region of FHA that has not previouslybeen implicated as a functional domain. Although it is possiblethat FhaS does not function as an adhesin, it is also possiblethat FhaS is an adhesin with a different cellular binding spec-ificity than FHA. Such a scenario would suggest FhaS andFHA recognize different host cell receptors, as has been pro-posed for the HMW1 and HMW2 adhesins, which, despitetheir high degree of similarity, have been shown to mediateadherence to different cell types (55). We are currently per-forming targeted mutagenesis and constructing strains express-ing chimeric FHA and FhaS proteins to investigate the role ofamino acids near the C terminus of FHA and FhaS in host celladherence.

Our rat experiments indicated that FhaS is also not func-tionally redundant with FHA in vivo. Unlike the case for fhaB,which is absolutely required for tracheal colonization (12), ourfhaS deletion strain was indistinguishable from wild-type B.bronchiseptica in its ability to establish respiratory infection inrats. When compared in coinfection experiments, however, the�fhaS strain was outcompeted by the wild-type strain, suggest-ing that fhaS is expressed in vivo and that its role is revealedwhen the mutant must compete with wild-type bacteria. Thus,like cyaA (24), fhaS appears to play a role in the infectiouscycle that is not revealed in single-infection rat colonizationexperiments. For example, fhaS may be involved in the initialestablishment of infection, modulation of the host immuneresponse, and/or transmission. We are in the process of devel-oping sensitive in vivo assays using natural infection models totest each of these possibilities.

Whatever the function of FhaS in the B. bronchiseptica lifecycle, it is apparently not essential, since our PCR-based assayindicated that at least one B. bronchiseptica strain does notappear to contain an fhaS gene. Moreover, the fhaS allelespresent in B. pertussis Tohama I and B. parapertussis 12822,while very similar to each other, are very different from fhaS ofB. bronchiseptica. This allele distribution suggests the possibil-

ity that fhaS may contribute to a phenotype that distinguishesB. bronchiseptica strains from B. pertussis and B. parapertussishu

strains. One such phenotype is host specificity. One can envis-age that the FhaS protein encoded by B. bronchiseptica pro-motes respiratory colonization of a broad range of mammalswhereas those encoded by B. pertussis and B. parapertussishu

allow, or facilitate, infection of humans specifically. Alterna-tively, it may be that expression of the B. bronchiseptica fhaSallele is somehow disadvantageous for human infection. If theB. pertussis and B. parapertussishu alleles play similar roles, itwould seem likely that their expression would be regulatedsimilarly. However, although we did not measure fhaS expres-sion in these subspecies, comparison of the Tohama I and12822 genomes revealed that the nucleotide sequence of theintergenic region upstream of fhaSBp differed significantlyfrom that of fhaSBpp (which was nearly identical to that offhaSBb), suggesting that fhaSBp may be regulated differently.We are currently conducting experiments to compare the ex-pression profiles of fhaSBp, fhaSBpp and fhaSBb and are alsoconstructing strains expressing heterologous fhaS genes to in-vestigate these hypotheses.

The distribution of fhaS alleles among the bordetellae is alsointeresting from an evolutionary perspective. How did B. per-tussis and B. parapertussishu, which diverged independentlyfrom a B. bronchiseptica-like ancestor (15, 21, 64), evolve tocontain nearly identical fhaS alleles that are not present in theB. bronchiseptica genome? While it is of course possible that B.pertussis and B. parapertussishu strains diverged from an ances-tral B. bronchiseptica(-like) strain that did contain an fhaSallele similar to fhaSBp and fhaSBpp, the fact that nucleotidesequences at the 5� and 3� ends of fhaSBpp are more similar tothose of B. bronchiseptica than B. pertussis suggests that B.parapertussishu acquired the central region of fhaS (the regionthat is similar to fhaSBp and different from fhaSBb) by horizon-tal transfer from a B. pertussis strain. The original source of thefhaSBp allele is more mysterious, as a BLAST search of thesequences unique to fhaSBp revealed no significant databasematches. We showed previously that a similar allele distribu-tion and genomic organization exists for the bipA genes of B.pertussis, B. parapertussishu, and B. bronchiseptica (20). bipAencodes a large outer membrane protein, and its similarity tointimin of enteropathogenic and enterohemorrhagic Esche-richia coli and invasin of Yersinia spp. suggests it may play arole in adherence to host tissues (58). Together, these datasuggest a mechanism by which B. bronchiseptica, which caninfect humans but which rarely (if ever) causes acute disease inthis host, may have evolved rapidly into B. parapertussishu viahorizontal acquisition of specific alleles from B. pertussis. Ad-ditional comparison of Bordetella genomes, together with di-rect investigation of the roles of genes with allele distributionpatterns like those of fhaS and bipA, will allow us to test thishypothesis and may shed light on the molecular bases of hostspecificity and mechanisms underlying the establishment ofacute versus chronic infections.

ACKNOWLEDGMENTS

We thank Carol Inatsuka for helpful discussions and technical as-sistance, Robin Hulbert for critical review of the manuscript, andJoseph Mazar for technical assistance.



.org/D

ownloaded from

http://iai.asm.org/

This research was supported by grants from the National Institutesof Health (AI43896), U.S. Department of Agriculture (National Re-search Initiative of the USDA Cooperative State Research, Educationand Extension Service grant 2003-35204-13555), and the University ofCalifornia Superfund Basic Research and Education Program toP.A.C.

REFERENCES

1. Abramson, T., H. Kedem, and D. A. Relman. 2001. Proinflammatory andproapoptotic activities associated with Bordetella pertussis filamentous hem-agglutinin. Infect. Immun. 69:2650–2658.

2. Akerley, B. J., P. A. Cotter, and J. F. Miller. 1995. Ectopic expression of theflagellar regulon alters development of the Bordetella-host interaction. Cell80:611–620.

3. Akerley, B. J., and J. F. Miller. 1993. Flagellin gene transcription in Borde-tella bronchiseptica is regulated by the BvgAS virulence control system. J.Bacteriol. 175:3468–3479.

4. Arico, B., S. Nuti, V. Scarlato, and R. Rappuoli. 1993. Adhesion of Bordetellapertussis to eukaryotic cells requires a time-dependent export and maturationof filamentous hemagglutinin. Proc. Natl. Acad. Sci. USA 90:9204–9208.

5. Arico, B., and R. Rappuoli. 1987. Bordetella parapertussis and Bordetellabronchiseptica contain transcriptionally silent pertussis toxin genes. J. Bac-teriol. 169:2847–2853.

6. Burns, V. C., E. J. Pishko, A. Preston, D. J. Maskell, and E. T. Harvill. 2003.Role of Bordetella O antigen in respiratory tract infection. Infect. Immun.71:86–94.

7. Clantin, B., H. Hodak, E. Willery, C. Locht, F. Jacob-Dubuisson, and V.Villeret. 2004. The crystal structure of filamentous hemagglutinin secretiondomain and its implications for the two-partner secretion pathway. Proc.Natl. Acad. Sci. USA 101:6194–6199.

8. Cotter, P. A., and A. M. Jones. 2003. Phosphorelay control of virulence geneexpression in Bordetella. Trends Microbiol. 11:367–373.

9. Cotter, P. A., and J. F. Miller. 2001. Bordetella, p. 619–674. In E. A.Groisman (ed.), Principles of bacterial pathogenesis. Academic Press, SanDiego, Calif.

10. Cotter, P. A., and J. F. Miller. 1994. BvgAS-mediated signal transduction:analysis of phase-locked regulatory mutants of Bordetella bronchiseptica in arabbit model. Infect. Immun. 62:3381–3390.

11. Cotter, P. A., and J. F. Miller. 1997. A mutation in the Bordetella bronchi-septica bvgS gene results in reduced virulence and increased resistance tostarvation, and identifies a new class of Bvg-regulated antigens. Mol. Micro-biol. 24:671–685.

12. Cotter, P. A., M. H. Yuk, S. Mattoo, B. J. Akerley, J. Boschwitz, D. A.Relman, and J. F. Miller. 1998. Filamentous hemagglutinin of Bordetellabronchiseptica is required for efficient establishment of tracheal colonization.Infect. Immun. 66:5921–5929.

13. Coutte, L., S. Alonso, N. Reveneau, E. Willery, B. Quatannens, C. Locht, andF. Jacob-Dubuisson. 2003. Role of adhesin release for mucosal colonizationby a bacterial pathogen. J. Exp. Med. 197:735–742.

14. Coutte, L., R. Antoine, H. Drobecq, C. Locht, and F. Jacob-Dubuisson. 2001.Subtilisin-like autotransporter serves as maturation protease in a bacterialsecretion pathway. EMBO J. 20:5040–5048.

15. Cummings, C. A., M. M. Brinig, P. W. Lepp, S. van de Pas, and D. A.Relman. 2004. Bordetella species are distinguished by patterns of substantialgene loss and host adaptation. J. Bacteriol. 186:1484–1492.

16. Dawid, S., S. J. Barenkamp, and J. W. St. Geme III. 1999. Variation inexpression of the Haemophilus influenzae HMW adhesins: a prokaryoticsystem reminiscent of eukaryotes. Proc. Natl. Acad. Sci. USA 96:1077–1082.

17. Dawid, S., S. Grass, and J. W. St. Geme III. 2001. Mapping of bindingdomains of nontypeable Haemophilus influenzae HMW1 and HMW2 ad-hesins. Infect. Immun. 69:307–314.

18. Deora, R., H. J. Bootsma, J. F. Miller, and P. A. Cotter. 2001. Diversity in theBordetella virulence regulon: transcriptional control of a Bvg-intermediatephase gene. Mol. Microbiol. 40:669–683.

19. Donnenberg, M. S., and J. B. Kaper. 1991. Construction of an eae deletionmutant of enteropathogenic Escherichia coli by using a positive-selectionsuicide vector. Infect. Immun. 59:4310–4317.

20. Fuchslocher, B., L. L. Millar, and P. A. Cotter. 2003. Comparison of bipAalleles within and across Bordetella species. Infect. Immun. 71:3043–3052.

21. Gerlach, G., F. von Wintzingerode, B. Middendorf, and R. Gross. 2001.Evolutionary trends in the genus Bordetella. Microbes Infect. 3:61–72.

22. Guedin, S., E. Willery, C. Locht, and F. Jacob-Dubuisson. 1998. Evidencethat a globular conformation is not compatible with FhaC-mediated secre-tion of the Bordetella pertussis filamentous haemagglutinin. Mol. Microbiol.29:763–774.

23. Hannah, J. H., F. D. Menozzi, G. Renauld, C. Locht, and M. J. Brennan.1994. Sulfated glycoconjugate receptors for the Bordetella pertussis adhesinfilamentous hemagglutinin (FHA) and mapping of the heparin-binding do-main on FHA. Infect. Immun. 62:5010–5019.

24. Harvill, E. T., P. A. Cotter, M. H. Yuk, and J. F. Miller. 1999. Probing the

function of Bordetella bronchiseptica adenylate cyclase toxin by manipulatinghost immunity. Infect. Immun. 67:1493–1500.

25. Hazenbos, W. L., B. M. van den Berg, J. W. van’t Wout, F. R. Mooi, and R.van Furth. 1994. Virulence factors determine attachment and ingestion ofnonopsonized and opsonized Bordetella pertussis by human monocytes. In-fect. Immun. 62:4818–4824.

26. Heininger, U., K. Stehr, S. Schmitt-Grohe, C. Lorenz, R. Rost, P. D. Chris-tenson, M. Uberall, and J. D. Cherry. 1994. Clinical characteristics of illnesscaused by Bordetella parapertussis compared with illness caused by Bordetellapertussis. Pediatr. Infect. Dis. J. 13:306–309.

27. Henderson, I. R., R. Cappello, and J. P. Nataro. 2000. Autotransporterproteins, evolution and redefining protein secretion. Trends Microbiol.8:529–532.

28. Ishibashi, Y., S. Claus, and D. A. Relman. 1994. Bordetella pertussis filamen-tous hemagglutinin interacts with a leukocyte signal transduction complexand stimulates bacterial adherence to monocyte CR3 (CD11b/CD18). J. Exp.Med. 180:1225–1233.

29. Ishibashi, Y., and A. Nishikawa. 2002. Bordetella pertussis infection of humanrespiratory epithelial cells up-regulates intercellular adhesion molecule-1expression: role of filamentous hemagglutinin and pertussis toxin. Microb.Pathog. 33:115–125.

30. Ishibashi, Y., and A. Nishikawa. 2003. Role of nuclear factor-kappa B in theregulation of intercellular adhesion molecule 1 after infection of humanbronchial epithelial cells by Bordetella pertussis. Microb. Pathog. 35:169–177.

31. Jacob-Dubuisson, F., C. Buisine, N. Mielcarek, E. Clement, F. D. Menozzi,and C. Locht. 1996. Amino-terminal maturation of the Bordetella pertussisfilamentous haemagglutinin. Mol. Microbiol. 19:65–78.

32. Jacob-Dubuisson, F., C. Buisine, E. Willery, G. Renauld-Mongenie, and C.Locht. 1997. Lack of functional complementation between Bordetella pertus-sis filamentous hemagglutinin and Proteus mirabilis HpmA hemolysin secre-tion machineries. J. Bacteriol. 179:775–783.

33. Jacob-Dubuisson, F., C. El-Hamel, N. Saint, S. Guedin, E. Willery, G. Molle,and C. Locht. 1999. Channel formation by FhaC, the outer membraneprotein involved in the secretion of the Bordetella pertussis filamentous hem-agglutinin. J. Biol. Chem. 274:37731–37735.

34. Jacob-Dubuisson, F., R. Fernandez, and L. Coutte. 2004. Protein secretionthrough autotransporter and two-partner pathways. Biochim. Biophys. Acta1694:235–257.

35. Jacob-Dubuisson, F., B. Kehoe, E. Willery, N. Reveneau, C. Locht, and D. A.Relman. 2000. Molecular characterization of Bordetella bronchiseptica fila-mentous haemagglutinin and its secretion machinery. Microbiology 146:1211–1221.

36. Jacob-Dubuisson, F., C. Locht, and R. Antoine. 2001. Two-partner secretionin Gram-negative bacteria: a thrifty, specific pathway for large virulenceproteins. Mol. Microbiol. 40:306–313.

37. Kloos, W. E., N. Mohapatra, W. J. Dobrogosz, J. W. Ezzel, and C. R.Manclark. 1981. Deoxyribonucleotide sequence relationships among Borde-tella species. J. Int. Syst. Bacteriol. 31:173–176.

38. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly ofthe head of bacteriophage T4. Nature 227:680–685.

39. Martinez de Tejada, G., J. F. Miller, and P. A. Cotter. 1996. Comparativeanalysis of the virulence control systems of Bordetella pertussis and Bordetellabronchiseptica. Mol. Microbiol. 22:895–908.

40. Mattoo, S., A. K. Foreman-Wykert, P. A. Cotter, and J. F. Miller. 2001.Mechanisms of Bordetella pathogenesis. Front. Biosci. 6:E168–E186.

41. Mattoo, S., J. F. Miller, and P. A. Cotter. 2000. Role of Bordetella bronchi-septica fimbriae in tracheal colonization and development of a humoralimmune response. Infect. Immun. 68:2024–2033.

42. Mattoo, S., M. H. Yuk, L. L. Huang, and J. F. Miller. 2004. Regulation oftype III secretion in Bordetella. Mol. Microbiol. 52:1201–1214.

43. Menozzi, F. D., P. E. Boucher, G. Riveau, C. Gantiez, and C. Locht. 1994.Surface-associated filamentous hemagglutinin induces autoagglutination ofBordetella pertussis. Infect. Immun. 62:4261–4269.

44. Menozzi, F. D., R. Mutombo, G. Renauld, C. Gantiez, J. H. Hannah, E.Leininger, M. J. Brennan, and C. Locht. 1994. Heparin-inhibitable lectinactivity of the filamentous hemagglutinin adhesin of Bordetella pertussis.Infect. Immun. 62:769–778.

45. Merkel, T. J., C. Barros, and S. Stibitz. 1998. Characterization of the bvgRlocus of Bordetella pertussis. J. Bacteriol. 180:1682–1690.

46. Merkel, T. J., and S. Stibitz. 1995. Identification of a locus required for theregulation of bvg-repressed genes in Bordetella pertussis. J. Bacteriol. 177:2727–2736.

47. Miller, V. L., and J. J. Mekalanos. 1988. A novel suicide vector and its usein construction of insertion mutations: osmoregulation of outer membraneproteins and virulence determinants in Vibrio cholerae requires toxR. J.Bacteriol. 170:2575–2583.

48. Parkhill, J., M. Sebaihia, A. Preston, L. D. Murphy, N. Thomson, D. E.Harris, M. T. Holden, C. M. Churcher, S. D. Bentley, K. L. Mungall, A. M.Cerdeno-Tarraga, L. Temple, K. James, B. Harris, M. A. Quail, M. Acht-man, R. Atkin, S. Baker, D. Basham, N. Bason, I. Cherevach, T. Chilling-worth, M. Collins, A. Cronin, P. Davis, J. Doggett, T. Feltwell, A. Goble, N.Hamlin, H. Hauser, S. Holroyd, K. Jagels, S. Leather, S. Moule, H. Norb-



.org/D

ownloaded from

http://iai.asm.org/

erczak, S. O’Neil, D. Ormond, C. Price, E. Rabbinowitsch, S. Rutter, M.Sanders, D. Saunders, K. Seeger, S. Sharp, M. Simmonds, J. Skelton, R.Squares, S. Squares, K. Stevens, L. Unwin, S. Whitehead, B. G. Barrell, andD. J. Maskell. 2003. Comparative analysis of the genome sequences ofBordetella pertussis, Bordetella parapertussis and Bordetella bronchiseptica.Nat. Genet. 35:32–40.

49. Prasad, S. M., Y. Yin, E. Rodzinski, E. I. Tuomanen, and H. R. Masure.1993. Identification of a carbohydrate recognition domain in filamentoushemagglutinin from Bordetella pertussis. Infect. Immun. 61:2780–2785.

50. Relman, D., E. Tuomanen, S. Falkow, D. T. Golenbock, K. Saukkonen, andS. D. Wright. 1990. Recognition of a bacterial adhesion by an integrin:macrophage CR3 (alpha M beta 2, CD11b/CD18) binds filamentous hem-agglutinin of Bordetella pertussis. Cell 61:1375–1382.

51. Renauld-Mongenie, G., J. Cornette, N. Mielcarek, F. D. Menozzi, and C.Locht. 1996. Distinct roles of the N-terminal and C-terminal precursor do-mains in the biogenesis of the Bordetella pertussis filamentous hemagglutinin.J. Bacteriol. 178:1053–1060.

52. Sambrook, J., T. Maniatis, and E. F. Fritsch. 1989. Molecular cloning: alaboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold SpringHarbor, N.Y.

53. Sato, Y., and H. Sato. 1999. Development of acellular pertussis vaccines.Biologicals 27:61–69.

54. Saukkonen, K., C. Cabellos, M. Burroughs, S. Prasad, and E. Tuomanen.1991. Integrin-mediated localization of Bordetella pertussis within macro-phages: role in pulmonary colonization. J. Exp. Med. 173:1143–1149.

55. St. Geme, J. W., III, S. Falkow, and S. J. Barenkamp. 1993. High-molecular-weight proteins of nontypable Haemophilus influenzae mediate attachment tohuman epithelial cells. Proc. Natl. Acad. Sci. USA 90:2875–2879.

56. St. Geme, J. W., III, and S. Grass. 1998. Secretion of the Haemophilusinfluenzae HMW1 and HMW2 adhesins involves a periplasmic intermediateand requires the HMWB and HMWC proteins. Mol. Microbiol. 27:617–630.

57. Stibitz, S., and J. F. Miller. 1994. Coordinate regulation of virulence inBordetella pertussis mediated by the vir (bvg) locus, p. 407–422. In V. L.Miller, J. B. Kaper, D. A. Portnoy, and R. A. Isberg (ed.), Molecular geneticsof bacterial pathogenesis. ASM Press, Washington, D.C.

58. Stockbauer, K. E., B. Fuchslocher, J. F. Miller, and P. A. Cotter. 2001.Identification and characterization of BipA, a Bordetella Bvg-intermediatephase protein. Mol. Microbiol. 39:65–78.

59. Tuomanen, E., H. Towbin, G. Rosenfelder, D. Braun, G. Larson, G. C.Hansson, and R. Hill. 1988. Receptor analogs and monoclonal antibodies

that inhibit adherence of Bordetella pertussis to human ciliated respiratoryepithelial cells. J. Exp. Med. 168:267–277.

60. Tuomanen, E., and A. Weiss. 1985. Characterization of two adhesins ofBordetella pertussis for human ciliated respiratory-epithelial cells. J. Infect.Dis. 152:118–125.

61. Tuomanen, E., A. Weiss, R. Rich, F. Zak, and O. Zak. 1985. Filamentoushemagglutinin and pertussis toxin promote adherence of Bordetella pertussisto cilia. Dev. Biol. Stand. 61:197–204.

62. Uhl, M. A., and J. F. Miller. 1995. Bordetella pertussis BvgAS virulencecontrol system, p. 333–349. In J. A. Hoch and T. J. Silhavy (ed.), Two-component signal transduction. ASM Press, Washington, D.C.

63. Vakevainen, M., S. Greenberg, and E. J. Hansen. 2003. Inhibition of phago-cytosis by Haemophilus ducreyi requires expression of the LspA1 and LspA2proteins. Infect. Immun. 71:5994–6003.

64. van der Zee, A., F. Mooi, J. Van Embden, and J. Musser. 1997. Molecularevolution and host adaptation of Bordetella spp.: phylogenetic analysis usingmultilocus enzyme electrophoresis and typing with three insertion se-quences. J. Bacteriol. 179:6609–6617.

65. Verghese, M. W., C. D. Smith, and R. Snyderman. 1985. Potential role for aguanine nucleotide regulatory protein in chemoattractant receptor mediatedpolyphosphoinositide metabolism, Ca�� mobilization and cellular responsesby leukocytes. Biochem. Biophys. Res. Commun. 127:450–457.

66. Ward, C. K., J. L. Latimer, J. Nika, M. Vakevainen, J. R. Mock, K. Deng,R. J. Blick, and E. J. Hansen. 2003. Mutations in the lspA1 and lspA2 genesof Haemophilus ducreyi affect the virulence of this pathogen in an animalmodel system. Infect. Immun. 71:2478–2486.

67. Ward, C. K., S. R. Lumbley, J. L. Latimer, L. D. Cope, and E. J. Hansen.1998. Haemophilus ducreyi secretes a filamentous hemagglutinin-like protein.J. Bacteriol. 180:6013–6022.

68. Weingart, C. L., and A. A. Weiss. 2000. Bordetella pertussis virulence factorsaffect phagocytosis by human neutrophils. Infect. Immun. 68:1735–1739.

69. Weiss, A. A., and S. Falkow. 1984. Genetic analysis of phase change inBordetella pertussis. Infect. Immun. 43:263–269.

70. Willems, R. J., C. Geuijen, H. G. van der Heide, G. Renauld, P. Bertin, W. M.van den Akker, C. Locht, and F. R. Mooi. 1994. Mutational analysis of theBordetella pertussis fim/fha gene cluster: identification of a gene with se-quence similarities to haemolysin accessory genes involved in export of FHA.Mol. Microbiol. 11:337–347.

71. Williams, C. L., P. E. Boucher, S. Stibitz, and P. A. Cotter. 2005. BvgAfunctions as both an activator and a repressor to control Bvgi phase expres-sion of bipA in Bordetella pertussis. Mol. Microbiol. 56:175–188.

Editor: V. J. DiRita



.org/D

ownloaded from

http://iai.asm.org/

Characterization of the Filamentous Hemagglutinin-Like Protein FhaS in Bordetella ... · Bordetella...

Documents

Transcript of Characterization of the Filamentous Hemagglutinin-Like Protein FhaS in Bordetella ... · Bordetella...