Ai

Ba

b

c

a

ARRAA

KBISV

1

agwtsfvcvf

htmitw[

0d

Vaccine 29 (2011) 3583–3595

Contents lists available at ScienceDirect

Vaccine

journa l homepage: www.e lsev ier .com/ locate /vacc ine

comprehensive analysis of Bordetella pertussis surface proteome anddentification of new immunogenic proteins

urcu E. Tefona, Sandra Maaßb, Erkan Özcengizc, Dörte Becherb, Michael Heckerb, Gülay Özcengiza,∗

Department of Biological Sciences, Middle East Technical University, 06531 Ankara, TurkeyInstitut für Mikrobiologie, Ernst-Moritz Arndt-Universität Greifswald, 17487 Greifswald, GermanyVaccine Biologicals Research Company, Ankara, Turkey

r t i c l e i n f o

rticle history:eceived 29 August 2010eceived in revised form 23 February 2011ccepted 25 February 2011vailable online 11 March 2011

eywords:

a b s t r a c t

Whooping cough, caused by the gram negative pathogen Bordetella pertussis, is a worldwide acute respi-ratory disease that predominantly involves infants. In the present study, surface proteins of B. pertussisTohama I and Saadet strains were identified by using 2DE followed by MALDI-TOF-MS/MS analysisand also geLC–MS/MS. With these approaches it was possible to identify 45 and 226 proteins, respec-tively. When surface proteins of the strains were separated by 2DE and analyzed by Western blotting fortheir reactivity, a total of 27 immunogenic spots which correspond to 11 different gene products weredetermined. Glutamine-binding periplasmic protein, leu/ile/val-binding protein, one putative exported

ordetella pertussismmunoproteomicsurfaceomeaccine candidates

protein, and iron-superoxide dismutase (Fe-SOD) were found as immunogenic for the first time in Bor-detella. Of a total of 226 proteins identified, 16 were differentially expressed in B. pertussis Saadet andTohama I strains. Five proteins were expressed only in Saadet (adhesin, chaperone protein DnaJ, fim-brial protein FimX, putative secreted protein Bsp22 and putative universal stress protein), and two (ABCtransporter substrate-binding protein and a putative binding protein-dependent transport periplasmic

I.

protein) only in Tohama

. Introduction

Bordetella pertussis is a gram-negative, strictly human pathogennd etiologic agent of whooping cough (pertussis), a highly conta-ious, acute respiratory illness [1]. Although, pertussis is relativelyell-controlled by vaccination programs, there are reports about

he global resurgence of B. pertussis, even in the countries with long-tanding pertussis immunization programs [2–6]. Apart from otheractors like increased awareness, improved diagnostics, decreasedaccination coverage and pathogen adaptation, one of the majorauses of the resurgence is, adolescents and adults, with decreasedaccine-induced immunity. They become the sources of the diseaseor unvaccinated or incompletely vaccinated infants [6–8].

Outer membrane proteins (OMPs) of gram-negative bacteriaave various functions and play a fundamental role in the interac-ion between the bacterial cell and its environment. In pathogenic

icroorganisms they also act as virulence factors and are involved

n adhesion, invasion of host cells, and proliferation [9,10]. Forhese reasons, understanding the features of OMPs will pave theay for discovering new antimicrobial drug and vaccine candidates

11,12]. Vaccines based on surface-exposed and secreted proteins

∗ Corresponding author. Tel.: +90 3122105170; fax: +90 3122107976.E-mail address: ozcengiz@metu.edu.tr (G. Özcengiz).

264-410X/$ – see front matter © 2011 Elsevier Ltd. All rights reserved.oi:10.1016/j.vaccine.2011.02.086

© 2011 Elsevier Ltd. All rights reserved.

are already commercially available and others are in development[13,14].

Because of their hydrophobicity, OMPs have traditionally beenhard to purify and solubilize for global analysis on two dimen-sional gels electrophoresis (2DE), but adding strong denaturing andchaothrophic agents to OMP extracts improves solubiliziation ofmembrane proteins for 2D electrophoretic analysis [11,15]. Despitethe advancements of improved solubilization techniques, the beststrategy for the analysis of membrane fractions is the combina-tion of SDS–PAGE with liquid chromatography (LC)–MS/MS. Thismethod has been termed as geLC–MS/MS [16–18].

In 2007, Vidakovics et al. [19] demonstrated differential pro-tein expression of B. pertussis under iron-limitation and iron-excessconditions. A comparative proteome analysis of enriched mem-brane proteins of three vaccine strains and a clinical isolate ofthis pathogen was reported, too [20]. In another study, the biofilmproteome profile of the microorganism was demonstrated by com-bining 2DE with Fourier Transform InfraRed spectroscopy [21].In 2009, our group identified cytoplasmic immunogenic proteinsof Tohama I and Saadet strains [22]. The present study reports

surface proteome and immunoproteome analysis of B. pertussisTohama I and Saadet strains by 2DE coupled with MALDI-TOF-MS/MS and 1DE coupled with LC–MS/MS. Our results providenew insights in the surface-associated immunogenic proteins of B.pertussis.

3 cine 2

2

2

aiI6122aP

2

daf(Rv28Ttw5iaRsb(wv

li(tb

2

mc5aiifi1aww

2

fTs

584 B.E. Tefon et al. / Vac

. Materials and methods

.1. Bacterial strains and preparation of outer membrane proteins

The strains of B. pertussis used in this study were Tohama Ind Saadet, the latter being a local isolate. The method of isolat-ng OMPs was modified from Wright et al. [23]. B. pertussis Tohamaand Saadet strains were grown in Morse-Bray medium [24] for0 h at 37 ◦C and then collected by centrifugation at 7000 × g for0 min. The pellet was treated with extraction buffer (3 M urea,8 mM DTT 0.0% w/v CHAPS) for 10 min and then centrifuged at6,000 × g for 15 min. The supernatant was recentrifuged at 4 ◦Ct 26,000 × g for 1 h and used as outer membrane protein source.rotein concentration was determined by Bradford technique [25].

.2. 1D and 2D gel electrophoresis

IPG strips were passively rehydrated by applying 400 �l of rehy-ration buffer (6 M urea, 2 M thiourea, 2% w/v CHAPS, 28 mM DTTnd 0.5% v/v ampholyte 3–10) containing 250 �g protein sampleor 14 h. IEF was performed with commercially available IPG strips17 cm, pH 3–10, Bio-Rad) and the Protean IEF Cell (Bio-Rad, USA).ehydrated strips were taken to the IEF process. The followingoltage profile was used for IEF: 1 h 100 V; 1 h 300 V; 1 h 600 V;h 1000 V; 2 h 3000 V; 2 h 5000 V followed by a linear increase to000 V. The final phase of 8000 V was terminated after 50,000 Vh.he IPG strips were equilibrated for 15 min each in 5 mL of solu-ion 1 (6 M urea, 50 mM Tris–HCl (pH 8.8), 30% v/v glycerin, 2%/v SDS, 50 mg DTT) and then in 5 mL of solution 2 (6 M urea,

0 mM Tris–HCl (pH 8.8), 30% v/v glycerin, 2% w/v SDS, 225 mgodacetamide) [17]. The isolated proteins were separated in 12%crylamide/bis-acrylamide gels with a Bio-Rad Cell system (Bio-ad, USA), applying approximately 25 mA per gel. To visualize theeparated proteins, each gel was stained with colloidal Coomassielue [26]. Coomassie stained gels were digitized using a scannerHP Scanjet 4070 Photosmart scanner, USA). Spot pattern analysesere accomplished using the 2D image analysis software Delta2D

ersion 3.4 (Decodon, Germany).For 1-D gel separation, an aliquot of 30 �g protein was solubi-

ized with one volume of SDS–PAGE sample buffer and separatedn 12% acrylamide/bis-acrylamide gel with a Bio-Rad Cell systemBio-Rad, USA), applying approximately 16 mA per gel. To visualizehe separated proteins, gels were stained with colloidal Coomassielue.

.3. Preparation of antisera against B. pertussis

Tohama and Saadet strains were grown on Cohen-Wheeler agaredia for 48 h. The cells were suspended in 0.85% saline solution to

ontain ca. 4 × 1010 bacteria/ml. The suspension was inactivated at6 ◦C for 30 min. Inactivated bacterial cells were used as the antigennd anti-B. pertussis polyclonal antibodies were raised by immuniz-ng mice. For each strain, 10 mice received two subcutaneous (sc)njections of 0.5 ml per animal at two week intervals between therst and second injections. Their sera were collected and pooled4 days after the second injection. In Western blot analysis, thentisera Th (sc) and Sa (sc) which were obtained with inactivatedhole cells of B. pertussis Tohama I and Saadet strains, respectively,ere used.

.4. Western blotting of 2DE gels

Proteins from the identical, but non-stained gels were trans-erred to a NC membrane for 1 h at 400 mA in transfer buffer (25 mMris, 192 mM glycine, 2% w/v SDS and 20% v/v methanol) by usingemi dry blotter (Cleaver Scientific Ltd.). After transfer, the mem-

9 (2011) 3583–3595

brane was blocked for 2 h with 10% skim milk in TBS solution(20 mM Tris and 5 M NaCl) at 37 ◦C. After rinsing for 10 min withTBS-Tween 20, the membrane was incubated with primary anti-body, mouse anti-B. pertussis Th (sc) or Sa (sc), at a dilution of 1:300in 0.05% Tween-20 in TBS (TTBS) for 10 min and incubated withrabbit anti-mouse IgG-alkaline phosphatase (Sigma), at a dilutionof 1:15,000 in TTBS containing 5% skim milk for 1 h. The membranewas then washed with TBS for 10 min and developed with substrate(AP Conjugate Substrate Kit, Bio-Rad) until optimum color wasdeveloped. The immunoreactivity of each spot that gave positivesignal in Western blot analysis was verified through their excisionfrom 2D gels followed by dot-blotting.

2.5. Protein identification

MALDI-TOF-MS/MS was performed in Greifswald for identi-fication of 2D spots is described in Eymann et al. [27]. Proteinspots were excised from stained 2D gels, destained and digestedwith trypsin (Promega, Madison, WI, USA). For the extraction ofpeptides, the gel pieces were covered with 60 �l of 0.1% triflu-oroacetic acid in 50% CH3CN and incubated for 30 min at 40 ◦C.Peptide solutions were mixed with an equal volume of saturated�-cyano-3-hydroxycinnamic acid solution in 50% acetonitrile–0.1%trifluoroacetic acid (v/v) and applied to a sample plate for MALDI-TOF-MS. Mass analyses were carried out on the Proteome-Analyzer4800 (Applied Biosystems). The three most abundant peptides ineach MS spectrum were chosen for MS/MS experiment. The result-ing sequence data were included for the database search to increasethe reliability of protein identification. Mass accuracy was usuallyin the range between 10 and 30 ppm.

One lane of 1D-SDS gel was cut into 12 equidistant pieces. In-gel tryptic digestion as well as peptide elution for LC–MS/MS wasperformed for each of the gel pieces as described by Eymann etal. [27]. The nano-LC–MS/MS analysis of peptides derived fromtryptic in-gel digestion was performed on a linear trap quadrupole(LTQ) Oribtrap (Thermo Fisher Scientific, Waltham, MA) equippedwith a nanoACQUITY UPLC (Waters, Milford, MA). Peptides wereloaded onto a trapping column (nanoAcquity Symmetry UPLC col-umn, C18, 5 �m, 180 �m by 20 mm; Waters) at a flow rate of10 �l/min and washed for 3 min with 99% buffer A. Peptides werethen eluted and separated via an analytical column (nanoAcquityBEH130 UPLC column, C18, 1.7 �m, 100 �m by 100 mm; Waters)with a 80 min gradient (from buffer A (0.1% acetic acid) to bufferB (0.1% acetic acid, acetonitrile). The mass spectrometric analysisstarted with a full survey scan in the Orbitrap (m/z 300–2000, res-olution of 60,000) followed by collision-induced dissociation andacquisition of MS/MS spectra of the five most abundant precur-sor ions in the LTQ. Precursors were dynamically excluded for 30 s,and unassigned charge states as well as singly charged ions wererejected. Proteins were identified via an automated database searchusing the SEQUEST software (Bioworks v.3.2, Thermo Electron).The search results were imported to Scaffold 2.02.01 (ProteomeSoftware) used to validate MS/MS-based peptide and protein iden-tifications. Peptide and protein identifications were accepted if theycould be established at greater than 99.9% probablility and con-tained at least two identified peptides.

2.6. Bioinformatic analysis

Amino acid sequences for B. pertussis proteins were obtainedfrom Sanger Institute organism’s genome project web site

(http://www.sanger.ac.uk/Projects/B pertussis/). To predict epi-topes of the identified immunogenic proteins of B. pertussis, theartficial network based B-cell epitope prediction server ABCpred[28] was used. The CELLO version 2.5 [29] was used for pre-diction of subcellular localization of the identified proteins. The

cine 29 (2011) 3583–3595 3585

Sei(

2

ofbnSS0

R

s

3

3

rtblbGtsMt(

saOtb(T

toFppopBwaT

3

aapSe

Fig. 1. (A) Dual channel 2-D imaging of B. pertussis strains Tohama I (green) and

B.E. Tefon et al. / Vac

ignalP webserver (http://www.cbs.dtu.dk/services/signalp/) wasmployed for signal peptide prediction. Functional categories of thedentified proteins were determined using KEGG BRITE databasehttp://www.genome.jp/kegg/brite.html).

.7. Relative abundance of surface proteins

Relative spectal counts (RSC) were calculated for quantitationf abundance differences of the proteins identified by LC–MS/MSrom two strains [30,31]. For each protein, log2 ratio of abundanceetween Sample 1 and Sample 2 constituted an RSC value (Eq. (1)).1 and n2 designate spectral counts for the protein in Sample 1 andample 2, t1 and t2 are total spectral count (sampling depth) foramples 1 and 2; and f is the correction factor set to 1.25 instead of.5, as proposed by Old [31].

SC = log2

[(n2 + f )(n1 + f )

]+ log2

[(t1 − n1 + f )

(t2 − −n2 + f )

](1)

The changes in RSC values greater than 2-fold were accepted asignificant.

. Results

.1. Surface proteome

Surface proteins of B. pertussis Tahoma I and Saadet strains wereesolved by 2DE to obtain an overview of protein distribution. Twoechnical replicates of 2DE gels for each of the three independentiological samples were run for each strain. After staining with col-

oidal CBB G250, nearly 170 spots could be detected on the gelsy the 2D image analysis software Delta2D version 3.4 (Decodon,ermany). There was no detectable difference between the pro-

ein patterns obtained from two strains (Fig. 1A). A total of 141pots were cut from each replicate for each strain and analyzed byALDI-TOF-MS analysis. A total of 125 spots which corresponded

o 45 different gene products could be identified for each strainTable 1). Of these gene products, 19 occurred in multiple spots.

1D-SDS–PAGE followed by LC–MS/MS was next used to identifyurface proteins of B. pertussis strains. Surface proteins were sep-rated by 1DE and proteins were visualized by CBB G250 staining.ne gel lane was cut into 12 equidistant pieces. After in-gel tryp-

ic digestion as well as peptide elution, peptides were analyzedy LC–MS/MS, resulting in identification of 226 proteins in totalTable 1), covering also those already identified via 2DE MALDI-OF-MS analysis.

Of 226 proteins, 16 were differentially expressed in B. per-ussis Saadet and Tohama I strains. 5 proteins were expressednly in Saadet (adhesin, chaperone protein DnaJ, fimbrial proteinimX, putative secreted protein Bsp22 and putative universal stressrotein), and 2 (ABC transporter substrate-binding protein and autative binding protein-dependent transport periplasmic protein)nly in Tohama I. Also, 6 proteins, OmpQ, PT subunit S1 and S2,utative outer protein D-BopD, putative uncharacterized proteincr4 and serotype 3 fimbrial subunit were more abundant in Saadethile 3 proteins, a probable extracellular solute binding protein

nd two putative exported proteins expressed at a higher level inohama I.

.2. Identification of immunoreactive proteins

Western blotting of the 2D gels using antisera designated

s Th (sc) and Sa (sc) as primary antibody and anti-mouse IgGs secondary antibody revealed a total of 27 immunoreactiverotein spots. The surface immunoproteomes of Tohama I andaadet strains were nearly identical. The corresponding spots werexcised, digested and analyzed using MALDI-TOF-MS. These spots

Saadet (red). (B) Fused 2-D Western blot analysis of the surface proteins of B. pertussisstrains Tohama and Saadet. (For interpretation of the references to color in this figurelegend, the reader is referred to the web version of the article.)

corresponded to 11 different gene products (Fig. 1B, Table 2).Of these, 6 proteins [60 kDa chaperonin (Hsp 60), serum resis-tance protein (BrkA), pertactin (PRN), Hsp 10, putative peptidylcis–trans isomerase and ATP synthase subunit beta] have alreadybeen shown in total soluble immunoproteome of B. pertussis [23]while 5 proteins, namely glutamine-binding periplasmic protein,leu/ile/val-binding protein, one putative exported protein, serotype2 fimbrial subunit (FIM2) and iron-superoxide dismutase (Fe-SOD)were detected by immunoproteomics for the first time in thispathogen. The immunogenic ones appeared to undergo chargemodification included PRN, BrkA, Fe-SOD, Hsp10 and leu/ile/val-

binding protein while FIM2 appeared as a mass variant antigen.The accuracy scores of the epitopes of these immunogenic proteinspredicted by the server ABCpred (Table 2) were within a meaningfulrange.

3586B.E.Tefon

etal./V

accine29 (2011) 3583–3595

Table 1Surface proteins of B. pertussis Tohama I and Saadet strains identified by geLC–MS/MS.

Protein no.a Protein name Gene locus MW(kDa)

Number ofisoforms b

Subcellularlocalizationc

Signalpeptide

Function Spectral countsd

Th Sa Rsce

1 10 kDa chaperonin BPP0869 10.25 3(C) C − Folding, sorting and degradation 16 12 −0.29482 2,3,4,5-Tetrahydropyridine-2,6-

dicarboxylateN-succinyltransferase

BB2183 29.24 C − Amino acid metabolism 15 14 −0.0052

3 3-oxoacyl-[acyl-carrier-protein]synthase II

BP2439 43.33 C − Lipid metabolism 10 13 0.4282

4 50S ribosomal protein L9 BB1917 16.34 C − Translation 4 5 0.33835 60 kDa chaperonin BP3495 57.46 C − Folding, sorting and degradation 131 104 −0.24706 ABC transport protein, periplasmic

componentBP2616 35.01 P + Unknown 8 8 0.0866

7 ABC transportersubstrate-binding protein

BAV1159 57.24 P + Membrane transport 14 0 −3.5250

8 ABC transporter substrate-bindingprotein

BAV1080 40.47 P + Membrane transport 16 32 1.0367

9 ABC transporter substrate-bindingprotein

BAV3033 41.86 P + Membrane transport 20 30 0.6451

10 ABC transporter substrate-bindingprotein

BAV1088 43.38 P + Membrane transport 7 13 0.8763

11 ABC transport protein,solute-binding component

BP2692 55.69 P + Membrane transport 7 5 −0.3144

12 Acetylornithine aminotransferase 2 BB4951 42.78 C − Amino acid metabolism 20 20 0.086713 Adenosylhomocysteinase BB0198 51.51 C − Amino acid metabolism 23 28 0.358214 Adenylosuccinate synthetase BB3165 46.82 C − Nucleotide metabolism 15 8 −0.727715 Adhesin BP2667 263.68 OM + Unknown 0 23 4.369216 Alkyl hydroperoxide reductase BP3552 20.15 C − Enzyme 12 16 0.468017 Amino acid-binding periplasmic

proteinBP0558 36.122 P, OM + Membrane transport 20 21 0.1532

18 Aminomethyltransferase BP0195 39.33 C − Energy metabolism 13 9 −0.389519 Antioxidant protein BP0965 23.75 C, P − Unknown 12 18 0.626720 Arginine biosynthesis bifunctional

protein ArgJBB4426 42.66 C − Amino acid metabolism 34 28 −0.1836

21 Argininosuccinate synthase BB1986 49.35 C − Amino acid metabolism 3 7 1.044222 Aromatic-amino-acid

aminotransferaseBP1795 43.07 C − Amino acid metabolism 9 12 0.4576

23 Aspartate-semialdehydedehydrogenase

BP1484 40.31 C − Amino acid metabolism 16 10 −0.5312

24 Aspartokinase BP1913 45.28 C − Amino acid metabolism 12 13 0.191825 ATP synthase gamma chain BB4606 33.32 C − Energy metabolism 11 6 −0.671226 ATP synthase subunit alpha BB4607 55.44 C − Energy metabolism 55 26 −0.964427 ATP synthase subunit beta BP3288 50.48 C − Energy metabolism 32 21 −0.495028 ATP-dependent Clp protease

proteolytic subunitBB2254 23.75 C − Enzyme families 8 6 −0.2653

29 Autotransporter subtilisin-likeprotease (SphB1)

BP0216 99.63 P, OM,E

− Membrane transport, secretionsystem

21 77 1.9125

30 Azurin BB3856 15.95 2 (C) P + Membrane transport 857 316 −1.458431 Bifunctional hemolysin-adenylate

cyclaseBB0324 177.03 E − Bacterial toxinss 0 3 1.8526

32 Bifunctional protein GlmU BB4817 48.36 C − Carbohydrate metabolism 8 14 0.8091

B.E.Tefonet

al./Vaccine

29 (2011) 3583–35953587

Table 1 (Continued)

Protein no.a Protein name Gene locus MW(kDa)

Number ofisoforms b

Subcellularlocalizationc

Signalpeptide

Function Spectral countsd

Th Sa Rsce

33 Branched-chain aminoacid-binding protein

BP1948 44.34 P + Membrane transport 16 16 0.0867

34 Carbonic anhydrase BP3425 23.64 C − Energy metabolism 10 10 0.086635 Chaperone protein ClpB BP1198 96.31 C − Folding, sorting and degradation 7 6 −0.100136 Chaperone protein DnaJ BB3933 40.23 C − Folding, sorting and degradation 0 9 3.123937 Chaperone protein DnaK BB3934 69.66 C − Folding, sorting and degradation 127 116 −0.043738 Chaperone protein FimB/FhaD BP1881 26.41 P + Folding, sorting and degradation 26 21 −0.206739 Chaperone protein HtpG BP0074 71.13 C − Folding, sorting and degradation 12 14 0.289840 Chaperone SurA BB4101 56.8 P + Folding, sorting and degradation 66 38 −0.695441 Cytochrome c oxidase polypeptide

IIBP3744 42.86 CM, P + Energy metabolism 14 18 0.4235

42 D-methionine ABC transporter,substrate-binding protein

BAV2852 28.54 P + Membrane transport 13 15 0.2765

43 Dihydrodipicolinate synthase BP1570 31.07 C − Amino acid metabolism 6 6 0.086544 Dihydrolipoyl dehydrogenase BAV1205 49.97 C − Carbohydrate metabolism 10 5 −0.762445 DNA polymerase iii, beta chain BP0490 41.23 C − Nucleotide metabolism 6 7 0.273246 DNA-directed RNA polymerase

alpha chainBAV0060 36.27 C − Nucleotide metabolism 10 13 0.4282

47 Electron transfer flavoproteinalpha-subunit

BP0962 31 C − Unknown 11 7 −0.4845

48 Electron transfer flavoproteinbeta-subunit

BP0961 26.8 C, P − Unknown 17 26 0.6669

49 Elongation factor Ts BB2606 30.88 C − Translation 34 43 0.416850 Elongation factor Tu BPP0007 42.89 C − Translation 62 41 −0.499251 Enolase BAV1166 45.95 C − Carbohydrate metabolism 36 24 −0.476552 Filamentous hemagglutinin BP1879 367.49 OM + Unknown 576 354 −0.655053 Fimbrial protein FimX BP2674 21.44 P + Unknown 0 6 2.623754 Fructose-bisphosphate aldolase BP1519 38.89 C − Carbohydrate metabolism 21 16 −0.081555 Glutamate dehydrogenase BP1857 46.6 P − Energy metabolism 6 8 0.038456 Glutamine ABC transporter,

glutamine-binding proteinBAV1960 26.66 P + Membrane transport 40 30 −0.0157

57 Glutamine-binding periplasmicprotein

BP1852 27.73 P + Membrane transport 43 49 0.0717

58 Glutathione reductase BP2120 49.96 C − Amino acid metabolism 7 11 0.057759 Glyceraldehyde-3-phosphate

dehydrogenaseBP1000 36.66 C − Carbohydrate metabolism 5 11 1.1586

60 Glycerol-3-phosphate-bindingperiplasmic protein

BP1281 47.78 P + Membrane transport 89 77 −0.0208

61 High-affinity branched-chainamino acid ABC transporter,

BAV1895 39.96 P + Membrane transport 25 26 0.0409

62 Histidinol dehydrogenase BB4854 46.66 C − Amino acid metabolism 3 7 1.144263 Imidazole glycerol phosphate

synthase subunit HisFBB4859 28.8 C − Amino acid metabolism 5 15 1.1671

64 Indole-3-glycerol phosphatesynthase

BP3261 28.84 C − Amino acid metabolism 12 15 0.0817

S65 Inorganic pyrophosphatase BP2533 20.02 C − Energy metabolism 20 16 −0.01566 Inosine-5′-monophosphate

dehydrogenaseBP2625 51.14 C − Nucleotide metabolism 8 7 −0.0787

67 Isocitrate dehydrogenase [NADP] BP2488 45.57 C − Carbohydrate metabolism 24 19 −0.032668 Ketol-acid reductoisomerase BAV2671 36.64 C − Amino acid metabolism 5 9 0.0010

3588B.E.Tefon

etal./V

accine29 (2011) 3583–3595

Table 1 (Continued)

Protein no.a Protein name Gene locus MW(kDa)

Number ofisoforms b

Subcellularlocalizationc

Signalpeptide

Function Spectral countsd

Th Sa Rsce

69 L-amino acid ABC transporter,substrate-binding protein

BAV3058 36.62 P + Membrane transport 139 140 0.0986

70 Leu/ile/val-binding protein BP1285 39.59 5 (C) P + Membrane transport 166 94 −0.738771 Leu/ile/val-binding protein BP1277 39.38 P + Membrane transport 26 26 0.086872 Malate synthase G BP3680 78.55 C + Carbohydrate metabolism 44 41 −0.012673 Molybdate-binding periplasmic

proteinBP3095 26.63 P + Membrane transport 33 13 −1.1824

74 OmpQ BP3405 39.94 OM + Unknown 1 11 2.233375 Orotate phosphoribosyltransferase BB4533 23.3 C + Nucleotide metabolism 10 7 −0.061576 Outer membrane porin protein

BP0840BP0840 41.12 OM + Unknown 91 109 0.0484

77 Outer membrane protein BAV1743 20.03 C, P + Unknown 12 14 0.089878 Outer membrane protein A BP0943 20.06 P + Unknown 1 5 1.161279 Penicillin-binding protein BAV0135 45.57 P + Glycan biosynthesis and

metabolism2 5 1.1305

80 Peptide deformylase 1 BB0247 19.91 C − Unknown 9 9 0.086681 Peptidoglycan-associated

lipoproteinBAV2916 17.79 P, OM + Unknown 10 9 −0.0479

82 Peptidyl-prolyl cis-trans isomeraseB

BP1906 18.84 2 (C) C, P − Folding, sorting and degradation 28 34 0.0572

83 Periplasmic solute-binding protein BP3674 34.41 P + Membrane transport 13 3 −1.160984 Pertactin BP1054 93.33 3 (C) OM, E + Membrane transport, secretion

system6 12 0.0577

85 Pertussis toxin subunit 1 BP3783 29.95 OM, E + Bacterial toxiins 0 14 3.398186 Pertussis toxin subunit 1

homologBB4890 29.97 OM, E + Bacterial toxiins 4 44 3.3022

87 Pertussis toxin subunit 2 BP3784 24.48 OM, E + Bacterial toxiins 1 10 2.210288 Pertussis toxin subunit 3 BP3787 24.47 CM, P, E + Bacterial toxiins 4 13 1.128989 Phosphate ABC transporter,

phosphate-binding periplasmicprotein

BAV0939 36.63 P + Membrane transport, signaltransduction

116 60 −0.0605

90 Phosphate-binding periplasmicprotein

BP1071 36.67 P + Membrane transport, signaltransduction

97 49 −0.0896

91 Phospho-2-dehydro-3-deoxyheptonate aldolase,Phe-sensitive

BP2908 38.88 C − Amino acid metabolism 40 22 −0.0439

92 Phosphoglycerate kinase BB1382 40.06 C − Carbohydrate metabolism 12 10 −0.049993 Polyribonucleotide

nucleotidyltransferaseBP0795 77.72 C − Nucleotide metabolism 20 15 −0.0013

94 Porphobilinogen deaminase BB2085 33.39 C − Metabolism of cofactors andvitamins

5 12 1.1720

95 Probable class IV aminotransferase BP0103 32.27 C − Amino acid metabolism 14 6 −0.087896 Probable extracellular

solute-binding proteinBP0121 47.73 P + Membrane transport 107 10 −3.3986

97 Probable periplasmicsolute-binding protein

BP0128 36.64 P + Membrane transport 7 2 −1.1584

98 Probable short-chaindehydrogenase

BP2770 26.69 C, CM − Unknown 8 13 0.0110

99 Probable surface antigen BP1427 86.69 OM + Unknown 18 7 −1.138100 Probable tonB-dependent receptor

BfrDBP0856 81.16 OM + Unknown 55 38 −0.0355

101 Probable zinc-bindingdehydrogenase

BP0800 33.37 C − Unknown 5 6 0.0009

B.E.Tefonet

al./Vaccine

29 (2011) 3583–35953589

Table 1 (Continued)

Protein no.a Protein name Gene locus MW(kDa)

Number ofisoforms b

Subcellularlocalizationc

Signalpeptide

Function Spectral countsd

Th Sa Rsce

102 Protein GrpE BB3936 19.9 C, P − Folding, sorting and degradation 13 5 −1.1041103 Protein TolB BB4237 47.73 2 (M) OM + Unknown 47 69 0.0335104 Protein-export protein SecB BPP0292 19.96 C − Folding, sorting and degradation,

membrane transport, secretionsystem

5 10 0.0355

105 Putative ABC transport protein,substrate-binding component

BP2418 42.21 P + Membrane transport 6 6 0.0865

106 Putative ABC transportsolute-binding protein

BP2747 40.05 2 (C–M) P + Membrane transport 282 160 −0.0486

107 Putative ABC transporterperiplasmic amino acid-bindingprotein

BP3831 36.66 2 (C) P, OM + Membrane transport 94 173 0.0754

108 Putative ABC transporter substratebinding protein

BP0301 41.15 P + Membrane transport 36 37 0.0253

109 Putative amino acid-bindingperiplasmic protein

BP1532 28.87 P + Membrane transport 7 16 1.1525

110 Putative amino-acid ABCtransporter, periplasmica.a.-binding protein

BP1364 28.86 P + Membrane transport 70 37 −0.0169

111 Putative antioxidant protein BP3551 18.86 P − Unknown 7 10 0.0346112 Putative bacterioferritin

comigratory proteinBP1307 19.95 P + Unknown 52 12 −1.1279

113 Putativebinding-protein-dependenttransport periplasmic protein

BP2396 57.78 P + Membrane transport 10 0 −3.3854

114 Putativebinding-protein-dependenttransport protein

BP3237 58.87 P + Membrane transport 4 15 1.1188

115 Putative carboxy-terminalprocessing protease

BP0609 51.15 P + Enzyme families 11 10 −0.0365

116 Putative cell surface protein BP2219 35.53 P + Unknown 9 20 1.1407117 Putative cyclase BP3130 35.58 C − Unknown 27 23 −0.0343118 Putative DNA-binding protein BP1616 18.8 C, P − Replication and repair 27 22 −0.0953119 Putative exported protein BAV3166 28.82 P + Membrane transport 5 3 −0.0703120 Putative exported protein BAV2471 33.32 P + Membrane transport 26 21 −0.0067121 Putative exported protein BAV0735 34.47 P + Unknown 17 14 −0.073122 Putative exported protein BAV2755 34.42 P + Unknown 16 25 0.0942123 Putative exported protein BAV1140 42.28 P + Membrane transport 7 13 0.0763124 Putative exported protein BP0250 34.41 P + Unknown 29 4 −2.2449125 Putative exported protein BP0205 19.95 P + Unknown 24 4 −2.2832126 Putative exported protein BP2936 37.74 C − Unknown 44 16 −1.1101127 Putative exported protein BP0698 22.26 P + Unknown 60 32 −0.0999128 Putative exported protein BP1480 29.91 P + Membrane transport 36 20 −0.0262129 Putative exported protein BP3827 35.59 P + Unknown 55 32 −0.0761130 Putative exported protein BP2818 28.82 P + Membrane transport 53 31 −0.0678131 Putative exported protein BP0454 33.33 4 (M) P + Unknown 16 9 −0.0658132 Putative exported protein BP3867 35.53 P + Signal transduction 66 39 −0.0589133 Putative exported protein BP2068 33.39 4 (M) P + Membrane transport 14 9 −0.0876134 Putative exported protein BP0334 33.33 C, CM, P + Unknown 18 13 −0.0482

3590B.E.Tefon

etal./V

accine29 (2011) 3583–3595

Table 1 (Continued)

Protein no.a Protein name Gene locus MW(kDa)

Number ofisoforms b

Subcellularlocalizationc

Signalpeptide

Function Spectral countsd

Th Sa Rsce

135 Putative exported protein BP3568 40.05 P + Membrane transport 5 4 −0.0652136 Putative exported protein BP0561 33.33 P + Unknown 6 5 −0.0278137 Putative exported protein BP1292 29.99 C, CM, P + Unknown 155 138 −0.0814138 Putative exported protein BP1506 40.07 C, CM, P + Membrane transport 8 7 −0.0787139 Putative exported protein BP2802 34.48 P + Unknown 55 50 −0.0482140 Putative exported protein BP1900 34.46 P + Unknown 9 9 0.0866141 Putative exported protein BP0385 28.8 P + Membrane transport 11 12 0.000142 Putative exported protein BP3575 43.36 4 (C) P + Membrane transport 38 43 0.0609143 Putative exported protein BP1887 35.52 P + Membrane transport 44 50 0.0679144 Putative exported protein BPP3617 34.43 P + Unknown 6 7 0.0732145 Putative exported protein BP0461 37.72 2 (C–M) P + Membrane transport 18 21 0.0962146 Putative exported protein BP1838 20.02 P + Unknown 59 86 0.0269147 Putative exported protein BP0664 35.54 CM + Unknown 12 19 0.0000148 Putative exported protein BP0562 17.76 P + Unknown 27 44 0.0699149 Putative exported protein BPP1941 29 2 (C–M) P + Unknown 45 74 0.0952150 Putative exported protein BPP4198 33.35 C, CM, P + Unknown 38 89 1.1987151 Putative exported protein BPP3542 34.48 P + Unknown 1 4 1.1095152 Putative exported protein BP3732 21.18 P + Unknown 1 5 1.1612153 Putative exported protein BP0782 39.92 P + Membrane transport 15 44 1.1701154 Putative exported protein BP3481 37.77 P + Membrane transport 5 17 1.1349155 Putative exported solute binding

proteinBP2963 40.09 2 (C) P + Membrane transport 142 110 −0.0832

156 Putative extracellularsolute-binding protein

BP3159 57.73 P + Membrane transport 23 42 0.0254

157 Putative extracellularsolute-binding protein

BP1529 36.64 C, P + Membrane transport 12 31 1.1738

158 Putative extracellularsolute-binding protein

BP3862 57.75 P + Membrane transport 5 17 1.1349

159 Putative glutamine-bindingperiplasmic protein

BP1852 27.73 P + Membrane transport 61 40 −0.0108

160 Putative glutathione S-transferase BP3659 24.41 C, CM, P − Amino acid metabolism 3 3 0.0865161 Putative glutathione transferase BP1300 26.61 C − Amino acid metabolism 5 7 0.0875162 Putative hemin binding protein BP0345 29.98 C, CM, P + Membrane transport 5 3 −0.0703163 Putative iron binding protein BP1605 37.78 P + Membrane transport 31 67 1.1758164 Putative lipoprotein BP1569 41.17 OM + Unknown 5 6 0.0009165 Putative lipoprotein BP2992 16.61 E + Unknown 10 14 0.0263166 Putative lipoprotein BP2072 21.1 P, OM + Unknown 1 5 1.1612167 Putative membrane protein BP3012 37.73 P + Unknown 56 15 −1.1381168 Putative membrane protein BP1056 25.53 P + Unknown 36 23 −0.035169 Putative orotidine 5′-phosphate

decarboxylaseBP3490 29.94 C − Nucleotide metablosim 7 14 0.0743

170 Putative outer membrane ligandbinding protein-BipA

BP1112 137.77 OM − Unknown 107 48 −1.1606

171 Putative outer membrane protein BP3755 23.31 OM + Unknown 17 8 −0.0955172 Putative outer protein D (BopD) BPP2223 31.15 OM − Unknown 1 25 3.3357173 Putative oxidoreductase BP2454 26.64 C + Carbohydrate metablosim 7 8 0.0518174 Putative penicillin-binding protein BP0102 44.46 P + Glycan biosynthesis and

metabolism8 10 0.0694

175 Putative peptidase BP0906 71.15 P + Unknown 41 16 −1.1105

B.E.Tefonet

al./Vaccine

29 (2011) 3583–35953591

Table 1 (Continued)

Protein no.a Protein name Gene locus MW(kDa)

Number ofisoforms b

Subcellularlocalizationc

Signalpeptide

Function Spectral countsd

Th Sa Rsce

176 Putative peptidase BP1721 30.03 P + Unknown 12 17 0.0495177 Putative peptidoglycan-associated

lipoproteinBP3342 17.77 OM + Unknown 8 6 −0.0653

178 Putative peptidyl-prolyl cis-transisomerase

BP3561 28.84 P + Folding, sorting and degradation 168 77 −1.1434

179 Putative periplasmic protein BP3341 24.48 P + Unknown 6 9 0.0867180 Putative periplasmic solute

binding proteinBP3080 33.39 C, P + Membrane transport 5 3 −0.0703

181 Putative periplasmicsolute-binding protein

BP1487 40 P + Unknown 111 153 0.0551

182 Putative periplasmicsubstrate-binding protein

BP2055 36.69 P + Membrane transport 32 8 −1.1639

183 Putative periplasmicsubstrate-binding transportprotein

BP2352 34.48 P + Membrane transport 7 8 0.0518

184 Putative polyamine transportprotein

BP2348 40.09 P + Membrane transport 11 13 0.0052

185 Putative regulatory lipoprotein BP2271 47.71 P + Membrane transport 6 17 1.1206186 Putative secreted protein (Bsp22) BP2256 22.22 P,E − Unknown 0 10 3.3584187 Putative sigma factor regulatory

proteinBP2435 39.99 C, CM, P + Unknown 15 7 −0.0930

188 Putative solute-bindingperiplasmic protein

BP3572 36.68 P + Membrane transport 16 22 0.0185

189 Putative TonB-dependent receptor BP2922 79.95 OM + Unknown 11 8 −0.0193190 Putative uncharacterized protein BP3819 26.69 P + Unknown 20 9 −0.0674191 Putative uncharacterized protein BP3128 68.87 C, OM + Unknown 25 18 −0.0621192 Putative uncharacterized protein BP2909 51.19 C − Enzyme families 5 4 −0.0652193 Putative uncharacterized protein BP2532 24.47 C − Unknown 8 10 0.0694194 Putative uncharacterized protein BP2953 23.36 P, E − Unknown 7 9 0.0001195 Putative uncharacterized protein BP0499 13.3 C, CM − Secretion system 4 7 0.0392196 Putative uncharacterized protein BP0479 35.54 P + Unknown 11 18 0.0401197 Putative uncharacterized protein BP1843 15.56 3 (C) OM − Unknown 8 15 0.0009198 Putative uncharacterized protein BP3013 26.63 C − Unknown 3 7 1.1442199 Putative uncharacterized protein BP2213 16.69 C, P − Translation 7 15 1.1662200 Putative uncharacterized protein BP1320 29.94 C, OM − Unknown 0 2 1.1654201 Putative uncharacterized protein

Bcr4BPP2226 18.82 C − Unknown 1 10 2.2102

202 Putative universal stress protein BP0410 16.62 C + Folding, sorting and degradation 0 8 2.2756

3592B.E.Tefon

etal./V

accine29 (2011) 3583–3595

Table 1 (Continued)

Protein no.a Protein name Gene locus MW(kDa)

Number ofisoforms b

Subcellularlocalizationc

Signalpeptide

Function Spectral countsd

Th Sa Rsce

203 Putative UTP-glucose-1-phosphateuridylyltransferase

BP3403 32.22 C − Carbohydrate metabolism 1 4 1.1095

204 Putative zinc protease BP2497 101.19 P, OM + Enzyme families 73 43 −0.0656205 Ribosome recycling factor BB2608 20.09 C − Translation 7 10 0.0346206 S-adenosylmethionine synthetase BB0195 41.18 C − Amino acid metabolism 7 10 0.0346207 Serine hydroxymethyltransferase 2 BB4348 44.49 C − Energy metabolism 7 5 −0.0144208 Serine protease BP2434 52.21 P + Unknown 14 15 0.0785209 Serotype 2 fimbrial subunit BP1119 21.13 3 (M) E + Membrane transport, secretion

system, cell motility48 128 1.1956

210 Serotype 3 fimbrial subunit BP1568 21.18 E + Unknown 1 76 5.5034211 Serum resistance protein BP3494 103.31 3 (C–M) OM, E + Membrane transport, secretion

system246 741 1.1842

212 Stringent starvation protein BAV3332 23.37 C − Unknown 7 17 1.134213 Succinate dehydrogenase

flavoprotein subunit (Fragment)BAV1184 64.46 C − Carbohydrate metabolism 6 18 1.1978

214 Succinate dehydrogenaseiron-sulfur protein

BAV1185 27.73 C − Carbohydrate metabolism 1 6 1.1756

215 Succinyl-CoA ligase [ADP-forming]subunit alpha

BP2540 30.06 C − Carbohydrate metabolism 7 15 1.1662

216 Succinyl-CoA synthetase, betachain

BP2541 40.02 C − Carbohydrate metabolism 29 19 −0.0943

217 Superoxide dismutase [Cu-Zn] BP2067 17.7 P + Carbohydrate metabolism 10 4 −1.1142218 Superoxide dismutase [Fe] BP2761 21.16 2 (C) E − Unknown 27 42 0.0043219 Thiol:disulfide interchange protein

DsbABB4940 22.29 P + Folding, sorting and degradation 72 17 −1.1289

220 Tracheal colonization factor BP1201 66.67 P, OM,E

+ Membrane transport, secretionsystem

51 100 1.1516

221 Transcription antiterminationprotein NusG

BAV0008 20.05 C − Unknown 8 15 0.0009

222 Trigger factor BB2253 47.77 C − Unknown 19 41 1.1522223 Triosephosphate isomerase BPP3426 25.59 2 (C–M) C − Carbohydrate metabolism 17 37 1.1583224 Tryptophan synthase alpha chain BB3773 29.92 C, CM − Amino acid metabolism 11 11 0.0866225 Vag8 BP2315 94.49 OM, E − Unknown 72 147 1.12226 Virulence factors putative positive

transcription regulator BvgABB2994 22.23 C − Signal transduction 7 6 −0.0001

a Protein numbers also correspond to the numbers shown on the gel (Fig. 1A).b C, charge modification; M, mass modification; C–M, both charge and mass modification.c Predicted location of proteins by CELLO 2.5. C, cytoplasm; OM, outer membrane; CM, cell membrane; P, periplasm and E, extracellular.d Spectral counrs of proteins of strains. Th, Tohama I and Sa, Saadet.e RSC values greater than 2-fold were accepted as significant which are shown in bold characters.

B.E. Tefon et al. / Vaccine 29 (2011) 3583–3595 3593

Table 2Immunogenic proteins detected in surface proteome of B. pertussis Tohama I and Saadet strains.

Protein no. Protein name/function Gene locus Mass (kDa) pI Epitope prediction Accuracy

1 10 kDa chaperonin BPP0869 10.26 5.39 AVGPGKKTEDGKILPV 0.875 60 kDa chaperonin BP3495 57.44 5.09 TGLKGDTADQNAGIKL 0.90

27 ATP synthase subunit beta BP3288 50.49 4.94 VVDIQFPRDNMPKIYE 0.9457 Glutamine-binding periplasmic protein BPP3010 26.84 8.77 YSLAEDPKTHVWSLQR 0.9370 Leu/ile/val-binding protein BP1285 39.38 6.67 YSLAEDPKTHVWSLQR 0.9584 Pertactin BP1054 93.81 9.23 TLTGGADAQGDIVATE 0.95

129 Putative exported protein BP3827 35.38 8.61 ANLPYDPVKDFAPVTI 0.92178 Putative peptidyl-prolyl cis-trans isomerase BP3561 28.94 8.93 PITQKSLDEFVKLVVS 0.93

3

aesdaias

4

mt[abpuB[c(cccc[kwp

209 Serotype 2 fimbrial subunit BP1119211 Serum resistance protein BP3494218 Superoxide dismutase BP2761

.3. Functional classes, protein localization and signal peptides

When functional classes of a total of 226 identified proteins werennotated, they fell into 6 categories: (i) general metabolism andnzyme families (27%), (ii) membrane transport, secretion system,ignal transduction, and cell motility (29%), (iii) folding, sorting andegradation (6%), (iv) bacterial toxins (2%), (v) replication, repairnd translation (3%), and (vi) yet unknown ones (33%) (Fig. 2). Local-zation prediction analysis revealed that ca. 60% of these proteinsre located at periplasm and/or outer membrane which was alsoupported by signal peptide analysis (Table 1).

. Discussion

Besides protecting bacteria against harsh envionments, outerembrane proteins play crucial roles such as signal transduc-

ion, solute and protein efflux/influx and bacterial pathogenesis9,10]. The expression of many virulence factors (autotransporters,dhesins and toxins in particular) of Bordetella species is controlledy the two-component regulatory system BvgAS [2,32]. BvgS is alasma membrane-bound sensor kinase which responds to stim-lation by phosphorylating the cytoplasmic DNA-binding proteinvgA. When phosphorylated, BvgA is able to promote transcription33,34]. Main autotransporters of this genus include PRN, tra-heal colonization factor (TcfA), BrkA, subtilisin like serine proteaseSphB1), and Vag8. Pertussis toxin (PT) which is another primaryomponent of pertussis vaccines and responsible for pertussis asso-iated lymphocytosis, adenylate cyclase (CyaA) with adenylateyclase/hemolysin activiy and dermonecrotic toxin (DNT) as a typi-

al A–B toxin are the most well-known toxins of this microorganism35]. In the present study, we were able to identify all these well-nown virulence factors of B. pertussis in both strains except for DNThich remains in cytoplasm. PT subunits S1 and S2 were among theroteins more abundant in Saadet than in Tohama I.

Fig. 2. Functional classes of the surface proteins identified from B. pertussis.

21.93 8.39 EASAITTYVGFSVVYP 0.90103.31 6.62 YSLAEDPKTHVWSLQR 0.95

21.28 6.05 AYYIDYRNARPKYLEN 0.92

The adhesion of B. pertussis to the host cell is provided by fil-amentous hemagglutinin (FHA) which is a highly immunogeniccomponent of acellular pertussis vaccines, and the autotrans-porters PRN, FIM2 and/or FIM3. Some adhesins exert their effectssynergically or function only in the absence of another adhesin[32,35]. An unknown and high molecular weight adhesin (236 kDa)was identified in outer membrane of Saadet, but not in Tohama I.BLAST search for this protein revealed the most significant databasematch to B. parapertussis adhesin, FhaS (88% identity). FhaS proteinwas characterized only in B. bronchiseptica [36]. Comparison of thepredicted FhaS proteins showed that fhaS gene of B. bronchisep-tica, but not those of B. pertussis or B. parapertussishu codes for aprotein that is nearly identical to FHA. However, FhaS could nei-ther mediate adherence of B. bronchiseptica to epithelial cell linesin vitro, nor was required for colonization in vivo. The allele dis-tribution suggested that FHA and FhaS perform distinct functionsduring the Bordetella pathogenesis, B. parapertussishu most likelyacquired its fhaS allele from B. pertussis horizontally, and fhaS maycontribute to host-species specificity. In B. pertussis, fim2 and fim3encode major fimbrial subunits FIM2 and FIM3, respectively [35].Expression of these two genes is regulated by mutations in theirpromoter regions, which result in serotype switching [37]. B. per-tussis genome also harbors a fimX gene, the product of which hasnot been identified yet [38]. Our study constitutes the first reporton identification of FimX protein in surfaceosome of B. pertussis.To note, this protein was found in Saadet strain, but not in com-monly used Tohama I. The expression of 22 kDa fimbrial subunitFIM3, on the other hand, was negligible in Tahoma I, as deter-mined by LC–MS/MS. This finding accords well with the resultsof an earlier study in which FIM3 could not be detected amongthe purified fimbria proteins both in SDS–PAGE and Western blotanalysis using anti-FIM3 monoclonal antibodies [39]. In anotherstudy, transcript abundance determined by significance analysis ofmicroarrays between the ancestral and 12-times passaged strainsof a clinical isolate revealed significant changes in transcript levelsincluding that of fim3 while the ancestral and passaged descendant(224 passages on plates) of Tohama I, as a laboratory strain, showedvery few differences in transcript abundance [40].

Another virulence factor of Bordetella which is under two-component regulatory system BvgAS is type III secretion system(T3SS) required for long-term colonization during infection [41].Molecules secreted via T3SS can be grouped as effectors whichare exotoxins exerting their functions when they are translocatedinto the host cells and translocons that take place in pore for-mation on the host membrane for the effectors to pass through[42]. In B. pertussis, B. parapertussis, and B. bronchiseptica, the T3SS

gene cluster has been found and designated the bsc locus com-prising 30 ORF that encode the T3SS machinery, BscN ATPase,type III-secreted proteins, and putative chaperones [43]. A highlyimmunogenic putative secreted protein Bsp22 is the most abun-dant T3SS effector protein exported from B. bronchiseptica [44]. It is

3 cine 2

aTo(pTOnoa

wiOroi

efepairFiepP

icrahsatrvspsl[ptbTrtuBaslstWi

cbmw

594 B.E. Tefon et al. / Vac

cell-surface exposed component of the T3SS system, required for3SS-mediated cytotoxicity against host cells and for the functionf the translocation pore complex [45]. Putative outer protein DBopD) is one of the T3SS translocator proteins which make a com-lex with BopB to form pores on the host plasma membrane [46].he putative uncharacterized protein Bcr4 is also encoded by anRF located in T3SS locus [47]. In our work, as known T3SS compo-ents, Bsp22 and BopD were demonstrated well in the surfaceomef Saadet strain. Tohama I did not express Bsp22 while having BopDnd Bcr4 only in negligible amounts.

Another protein with a higher expression level in Saadet strainas OmpQ, a porin-like protein known to play an important role

n allowing access to an essential nutrient. Although the lack ofmpQ did not affect the B. pertussis’s survival in vitro or in vivo, its

ole during infection, perhaps in the colonization of the human hostr in the establishment of a carrier state cannot be ruled out sincet is a Bvg-regulated protein [48].

The immunogenic component Fe-SOD was predicted as anxtracellular protein in both strains. SODs are crucial enzymesor pathogenic bacteria for detoxification of endogenous andxogenous reactive oxygen species produced during the infectionrocess. A Fe-SOD-deficient mutant of B. pertussis had a reducedbility to express CyaA and PRN proteins of pathogenesis, moreovert also had decreased abilities to colonize and persist in the murineespiratory infection model [49]. A cytosolic SOD belonging to thee–Mn SOD family was found to be immunogenic in Mycobacteriummmunogenum [50]. The immunogenic activity of Fe-SOD which isxcreted into the growth medium by the epimastigote form of therotozoan parasite Trypanosoma cruzi and promastigote form ofhytomonas is noteworthy [51,52].

Tohama I, as the B. pertussis strain chosen for genome sequenc-ng and a vaccine strain as well, was originally isolated from aase of whooping cough in Japan in the 1950 s. The sequencingevealed extensive large-scale genome rearrangements and thushigh level of genome plasticity [53]. Brinig et al. [40] found a

igh level of conservation of gene content among 137 B. pertussistrains with different geographical, temporal, and epidemiologicalssociations, using comparative genomic hybridization. The inser-ion element IS481 appeared to provide targets for homologousecombination, giving the capacity to the pathogen to generateariation by rearranging its chromosome and altering gene expres-ion. These researchers also compared genome-wide expressionrofiles of different strains and found significant changes in tran-cript abundance, even in the same strain after as few as 12aboratory passages. By using subtractive hybridization, Caro et al.54] identified genetic regions specific to recent and old isolates of B.ertussis, and isolates of B. parapertussis and B. bronchiseptica. Sincehe genomes of more recently collected isolates of B. pertussis har-or approximately 46 kb of additional genetic material compared toohama I genome, they suggested that Tohama I strain may not beepresentative of the B. pertussis and emhasized the importance ofhe analysis of clinical isolates for the development of new molec-lar diagnostics. Fennely et al. [55] provided the first evidence that. pertussis uses the T3SS for colonization and survival in the host,nd possibly to target the innate immune system and found expres-ion of Bsp22 effector in 15 of 16 clinical isolates as well as 2ow-passage ATCC strains, but not in common laboratory-adaptedtrains like Tohama I and Wellcome 28. They therefore proposedhat the absence of a functional T3SS in B. pertussis Tohama I and

ellcome 28 may be a consequence of long-term laboratory culturen the absence of eukaryotic cell contact.

B. pertussis Saadet had been isolated as a Phase I strain from aase of whooping cough in 1948 in Turkey and extensively usedoth in research and cellular pertussis vaccine manufacture forany years in the country as a highly immunogenic strain. Saadetas given preference over other strains including Tohama I as a

9 (2011) 3583–3595

more protective vaccine strain, most likely owing to the stableexpression of important virulence factors such as adhesin, DnaJ,FimX, Bsp22 and a putative universal stress protein as well asoverexpression of OmpQ, PT subunits S1 and S2, BopD, Bcr4, FIM2and FIM3, as the present study revealed. On the other hand, thisstrain constitutes an example for the lack of one-to-one relation-ship between virulence gene expression and long-term laboratorymaintenance since it is also a high-passage and laboratory-adaptedstrain like Tahoma I.

Label-free geLC–MS/MS involves the comparison of multi-ple unprocessed LC–MS datasets based on the relative peptidepeak intensities and simultaneously increases throughput, facil-itating biomarker discovery [56]. It combines the robust natureof SDS–PAGE and the resolving power of LC–MS/MS. More pro-teins are derived from gel bands (1DE) than from gel spots (2DE)and this leads to a better separation and relative higher quali-ties of tandem mass spectra in a single LC–MS/MS run [19]. Inaddition, it can resolve low abundance proteins and large andhydrophobic ones typically not accessible via 2DE. As expected,GeLC–MS/MS detected more surface proteins than 2DE–MS/MS.226 proteins identified by geLC–MS/MS included each of the 45proteins identified by 2DE–MS/MS in this work, 21 out of 54 sur-faceome components identified by 2DE–MS/MS from 3 vaccinestrains and 1 clinical isolate of B. pertussis as reported by Botteroet al. [20], and 24 out of 49 identified proteins from biofilm-grownTohama I analyzed by Serra et al. by combining 2-DE and Fouriertransform infrared (FT-IR) spectroscopy [21].

Compared with genomics and transcriptomics, proteomics hasthe advantage of defining (i) final active level of proteins, (ii) dif-ferentially expressed ones, not solely at gene level, and (iii) thosedifferentially located or secreted to outside of the cell [57,58].While shotgun proteomics provides better protein coverage, theadvantage of gel-based proteomics is its capacity of analyzing post-translational protein modifications that play an important role inbacterial pathogenesis Accordingly, 2DE–MS/MS approach used inthis study revealed that 19 out of 45 identified proteins occur inmultiple spots in the surface proteome of both strains of B. pertussis,after excluding all possible artefacts (Table 1).

Immunoreactive proteins identified in this study as well as inour previous work [22] were scored for their potential as vaccinecandidates as outlined by Khan et al. [57]. Our studies have alreadybeen extended to demonstrate the humoral and cell-mediatedimmune responses to selected and heterologusly expressed pro-tein targets from B. pertussis and induction of protective immunityagainst lethal challenge in mice models.

Acknowledgements

This research was granted in part by TUBITAK-TBAG (projectno.107T444) and Middle East Technical University Research FundMETU-09-11-DPT.2002K120510. We would like to acknowledgeDr. Dirk Albrecht for his helps in MALDI-TOF-MS analysis. We thankOrhan Özcan, Volkan Yıldırım and Knutt Büttner for their fruitfulcomments.

References

[1] Cherry JD, Heininger U. Pertussis and other Bordetella infections. In: FeiginRD, Cherry JD, Demmler GJ, Kaplan S, editors. Textbook of pediatric infectiousdiseases. 5th ed. Philadelphia: The W. B. Saunders Co; 2004. p. 1588–608.

[2] Cotter PA, Miller JF. Bordetella. In: Groisman EA, editor. Principles of bacterialpathogenesis. London: Academic Press Ltd; 2001. p. 619–74.

[3] Crowcroft NS, Britto J. Whooping cough – a continuing problem. BMJ2002;324:1537–8.

[4] de Carvalho AP, Pereira EM. Acellular pertussis vaccine for adolescents. J Pediatr(Rio J) 2006;82:15–24.

[5] Berbers GA, de Greeff SC, Mooi FR. Improving pertussis vaccination. Hum Vac-cine 2009;5:497–503.

cine 2

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

B.E. Tefon et al. / Vac

[6] Mooi FR. Bordetella pertussis and vaccination: the persistence of a geneticallymonomorphic pathogen. Infect Genet Evol 2010;10:36–49.

[7] Grimprel E, Baron S, Levy-Bruhl D, Garnier JM, Njamkepo E, Guiso N, et al.Influence of vaccination coverage on pertussis transmission in France. Lancet1999;354:1699–700.

[8] Gilberg S, Njamkepo E, Du Châtelet IP, Partouche H, Gueirard P, Ghasarossian C,et al. Evidence of Bordetella pertussis infection in adults presenting with persis-tent cough in a french area with very high whole-cell vaccine coverage. J InfectDis 2002;186:415–8.

[9] Koebnik R, Locher KP, Van Gelder P. Structure and function of bacterial outermembrane proteins: barrels in a nutshell. Mol Microbiol 2000;37:239–53.

10] Lin J, Huang S, Zhang Q. Outer membrane proteins: key players for bacterialadaptation in host niches. Microb Infect 2002;4:325–31.

11] Kurupati P, Teh BK, Kumarasinghe G, Poh CL. Identification of vaccine can-didate antigens of an ESBL producing Klebsiella pneumoniae clinical strain byimmunoproteome analysis. Proteomics 2006;6:836–44.

12] Rodriguez-Ortega MJ, Norais N, Bensi G, Liberatori S, Capo S, Mora M, et al. Char-acterization and identification of vaccine candidate proteins through analysisof the group A Streptococcus surface proteome. Nat Biotechnol 2006;24:191–7.

13] Pizza M, Scarlato V, Masignani V, Giuliani MM, Arico B, Comanducci M, et al.Identification of vaccine candidates against serogroup B meningococcus bywhole-genome sequencing. Science 2000;287:1816–20.

14] Maione D, Margarit I, Rinaudo CD, Masignani V, Mora M, Scarselli M, et al.Identification of a universal Group B streptococcus vaccine by multiple genomescreen. Science 2005;309:148–50.

15] Poetsch A, Wolters D. Bacterial membrane proteomics. Proteomics2008;8:4100–22.

16] Wilm M, Shevchenko A, Houthaeve T, Breit S, Schweigerer L, Fotsis T, et al. Fem-tomole sequencing of proteins from polyacrylamide gels by nano-electrospraymass spectrometry. Nature 1996;379:466–9.

17] Gorg A, Weiss W, Dunn MJ. Current two-dimensional electrophoresis technol-ogy for proteomics. Proteomics 2004;4:3665–85.

18] Ye Y, Mar EC, Tong S, Sammons S, Fang S, Anderson LJ, et al. Application ofproteomics methods for pathogen discovery. J Virol Methods 2010;163:87–95.

19] Vidakovics ML, Paba J, Lamberti Y, Ricart CA, Sousa MV, Rodriguez ME. Pro-filing the Bordetella pertussis proteome during iron starvation. J Proteome Res2007;6:2518–28.

20] Bottero D, Gaillard ME, Fingermann M, Weltman G, Fernández J, Sisti F, et al.Pulsed-field gel electrophoresis, pertactin, pertussis toxin S1 subunit polymor-phisms, and surfaceome analysis of vaccine and clinical Bordetella pertussisstrains. Clin Vaccine Immunol 2007;14:1490–8.

21] Serra DO, Lücking G, Weiland F, Schulz S, Görg A, Yantorno OM, et al. Pro-teome approaches combined with Fourier transform infrared spectroscopyrevealed a distinctive biofilm physiology in Bordetella pertussis. Proteomics2008;8:4995–5010.

22] Altindis E, Tefon BE, Yildirim V, Ozcengiz E, Becher D, Hecker M, et al.Immunoproteomic analysis of Bordetella pertussis and identification of newimmunogenic proteins. Vaccine 2009;27:542–8.

23] Wright A, Wait R, Begum S, Crossett B, Nagy J, Brown K. Proteomic analysis ofcell surface proteins from Clostridium difficile. Proteomics 2005;5:2443–52.

24] Morse SI, Bray KK. The occurrence and properties of leukocytosis andlymphocytosis-stimulating material in the supernatant fluids of Bordetella per-tussis cultures. J Exp Med 1969;129:523–50.

25] Ramagli LS, Rodrigez LV. Quantitation of microgram amounts of protein in two-dimensional polyacrylamide gel electrophoresis sample buffer. Electrophoresis1985;6:559–63.

26] Neuhoff V, Arold N, Taube D, Ehrhardt W. Improved staining of proteins inpolyacrylamide gels including isoelectric focusing gels with clear backgroundat nanogram sensitivity using Coomassie Brilliant Blue G-250 and R-250. Elec-trophoresis 1988;9:255–62.

27] Eymann C, Dreisbach A, Albrecht D, Bernhardt J, Becher D, Gentner S, et al.A comprehensive proteome map of growing Bacillus subtilis cells. Proteomics2004;4:2849–76.

28] Saha S, Raghava GP. Prediction of continuous B-cell epitopes in an antigen usingrecurrent neural network. Proteins 2006;65:40–8.

29] Yu CS, Chen YC, Lu CH, Hwang JK. Prediction of protein subcellular localization.Proteins 2006;64:643–51.

30] Beissbarth T, Hyde L, Smyth GK, Job C, Boon WM, Tan SS, et al. Statistical mod-eling of sequencing errors in SAGE libraries. Bioinformatics 2004;20:i31–9.

31] Old WM, Meyer-Arendt K, Aveline-Wolf L, Pierce KG, Mendoza A, Sevinsky JR,et al. Comparison of label-free methods for quantifying human proteins byshotgun proteomics. Mol Cell Proteomics 2005;4:1487–502.

32] Mattoo S, Cherry JD. Molecular pathogenesis, epidemiology, and clinical man-ifestations of respiratory infections due to Bordetella pertussis and otherBordetella subspecies. Clin Microbiol Rev 2005;18:326–82.

[

[

9 (2011) 3583–3595 3595

33] Uhl M, Mller J. utophosphorylation and phosphotransfer in the Borde-tella pertussis BvgAS signal transduction cascade. Proc Natl Acad Sci USA1994;91:1163–7.

34] Steffen P, Goyard S, Ullmann A. Phosphorylated BvgA is sufficient for transcrip-tional activation of virulence-regulated genes in Bordetella pertussis. EMBO J1996;15:102–9.

35] Locht C. Molecular aspects of Bordetella pertussis pathogenesis. Int Microbiol1999;2:137–44.

36] Julio SM, Cotter PA. Characterization of the filamentous hemagglutinin-likeprotein FhaS in Bordetella bronchiseptica. Infect Immun 2005;73:4960–71.

37] Willems R, Paul A, van der Heide HG, ter Avest AR, Mooi FR. Fimbrial phase vari-ation in Bordetella pertussis: a novel mechanism for transcriptional regulation.EMBO J 1990;9:2803–9.

38] Mooi FR, van der Heide HG, ter Avest AR, Welinder KG, Livey I, van der ZeijstBAM, et al. Characterization of fimbrial subunits from Bordetella species. MicrobPathog 1987;2:473–84.

39] Atakan Ablay P, Özcengiz E. Rapid purification of Bordetella pertussis fimbria 2,3 proteins and their immunogenicities. Mikrobiyol Bul 2007;41:11–20.

40] Brinig MM, Cummings CA, Sanden GN, Stefanelli P, Lawrence A, Relman DA.Significant gene order and expression differences in Bordetella pertussis despitelimited gene content variation. J Bacteriol 2006;188:2375–82.

41] Yuk MH, Harvill ET, Miller JF. The BvgAS virulence control system regu-lates type III secretion in Bordetella bronchiseptica. Mol Microbiol 1998;28:945–59.

42] Abe A, Nagamatsu K, Watanabe M. The Bordetella type III secretion sys-tem: its application to vaccine development. Microbiol Immunol 2008;52:128–33.

43] Fauconnier A, Veithen A, Gueirard P, Antoine R, Wacheul LL, Locht C, et al.Characterization of the type III secretion locus of Bordetella pertussis. Int J MedMicrobiol 2001;290:693–705.

44] Yuk MH, Harvill ET, Cotter PA, Miller JF. Modulation of host immune responses,induction of apoptosis and inhibition of NF-kappaB activation by the Bordetellatype III secretion system. Mol Microbiol 2000;35:991–1004.

45] Medhekar B, Shrivastava R, Mattoo S, Gingery M, Miller JF. Bordetella Bsp22forms a filamentous type III secretion system tip complex and is immunopro-tective in vitro and in vivo. Mol Microbiol 2009;71:492–504.

46] Nogawa H, Kuwae A, Matsuzawa T, Abe A. The type III secreted protein BopDin Bordetella bronchiseptica is complexed with BopB for pore formation on thehost plasma membrane. J Bacteriol 2004;186:3806–13.

47] Brinig MM, Register KB, Ackermann MR, Relman DA. Genomic features of Bor-detella parapertussis clades with distinct host species specificity. Genome Biol2006;7:R81.

48] Finn TM, Li Z, Kocsis E. Identification of a Bordetella pertussis bvg-regulatedporin-like protein. J Bacteriol 1995;177:805–9.

49] Khelef N, DeShazer D, Friedman RL, Guiso N. In vivo and in vitro analysisof Bordetella pertussis catalase and Fe-superoxide dismutase mutants. FEMSMicrobiol Lett 1996;142:231–5.

50] Gupta MK, Subramanian V, Yadav JS. Immunoproteomic identification ofsecretory and subcellular protein antigens and functional evaluation of thesecretome fraction of Mycobacterium immunogenum, a newly recognizedspecies of the Mycobacterium chelonae–Mycobacterium abscessus group. J Pro-teome Res 2009;8:2319–30.

51] Marin C, Hitos AB, Rodriguez-Gonzalez I, Dollet M, Sanchez-Moreno M. Phy-tomonas iron superoxide dismutase: a possible molecular marker. FEMSMicrobiol Lett 2004;234:69–74.

52] Villagran ME, Marin C, Rodriguez-Gonzalez I, De Diego JA, Sanchez-Moreno M.Use of an iron superoxide dismutase excreted by Trypanosoma cruzi in the diag-nosis of Chagas disease: seroprevalence in rural zones of the state of Queretaro,Mexico. Am J Trop Med Hyg 2005;73:510–6.

53] Parkhill J, Sebaihia M, Preston A, Murphy LD, Thomson N, Harris DE, et al. Com-parative analysis of the genome sequences of Bordetella pertussis, Bordetellaparapertussis and Bordetella bronchiseptica. Nat Genet 2003;35:32–40.

54] Caro V, Bouchez V, Guiso N. Is the Sequenced Bordetella pertussis strain TohamaI representative of the species? J Clin Microbiol 2008;46:2125–8.

55] Fennelly NK, Sisti F, Higgins SC, Ross PJ, van der Heide H, Mooi FR, et al. Borde-tella pertussis expresses a functional type III secretion system that subvertsprotective innate and adaptive immune responses. Infect Immun 2008;76:1257–66.

56] Nesatyy VJ, Suter MJ. Proteomics for the analysis of environmental stress

responses in organisms. Environ Sci Technol 2007;41:6891–900.

57] Khan AS, Mujer CV, Alefantis TG, Connolly JP, Mayr UB, Walcher P, et al.Proteomics and bioinformatics strategies to design countermeasures againstinfectious threat agents. J Chem Inf Model 2006;46:111–5.

58] Wu HJ, Wang AH, Jennings MP. Discovery of virulence factors of pathogenicbacteria. Curr Opin Chem Biol 2008;12:93–101.