Bifunctional Role of the Treponema pallidum …Treponema pallidum, the causative agent of syphilis, is a highly invasive pathogenic spirochete capable of attaching to host cells, invading

INFECTION AND IMMUNITY, Mar. 2011, p. 1386–1398 Vol. 79, No. 30019-9567/11/$12.00 doi:10.1128/IAI.01083-10Copyright © 2011, American Society for Microbiology. All Rights Reserved.

Bifunctional Role of the Treponema pallidum Extracellular MatrixBinding Adhesin Tp0751�

Simon Houston, Rebecca Hof, Teresa Francescutti,† Aaron Hawkes,Martin J. Boulanger, and Caroline E. Cameron*

Department of Biochemistry and Microbiology, University of Victoria, Victoria, British Columbia V8W 3P6, Canada

Received 7 October 2010/Returned for modification 9 October 2010/Accepted 1 December 2010

Treponema pallidum, the causative agent of syphilis, is a highly invasive pathogenic spirochete capable ofattaching to host cells, invading the tissue barrier, and undergoing rapid widespread dissemination via thecirculatory system. The T. pallidum adhesin Tp0751 was previously shown to bind laminin, the most abundantcomponent of the basement membrane, suggesting a role for this adhesin in host tissue colonization andbacterial dissemination. We hypothesized that similar to that of other invasive pathogens, the interaction of T.pallidum with host coagulation proteins, such as fibrinogen, may also be crucial for dissemination via thecirculatory system. To test this prediction, we used enzyme-linked immunosorbent assay (ELISA) methodologyto demonstrate specific binding of soluble recombinant Tp0751 to human fibrinogen. Click-chemistry-basedpalmitoylation profiling of heterologously expressed Tp0751 confirmed the presence of a lipid attachment sitewithin this adhesin. Analysis of the Tp0751 primary sequence revealed the presence of a C-terminal putativeHEXXH metalloprotease motif, and in vitro degradation assays confirmed that recombinant Tp0751 purifiedfrom both insect and Escherichia coli expression systems degrades human fibrinogen and laminin. Theproteolytic activity of Tp0751 was abolished by the presence of the metalloprotease inhibitor 1,10-phenanthro-line. Further, inductively coupled plasma-mass spectrometry showed that Tp0751 binds zinc and calcium.Collectively, these results indicate that Tp0751 is a zinc-dependent, membrane-associated protease thatexhibits metalloprotease-like characteristics. However, site-directed mutagenesis of the HEXXH motif toHQXXH did not abolish the proteolytic activity of Tp0751, indicating that further mutagenesis studies arerequired to elucidate the critical active site residues associated with this protein. This study represents the firstpublished description of a T. pallidum protease capable of degrading host components and thus provides novelinsight into the mechanism of T. pallidum dissemination.

The spirochete bacterium Treponema pallidum subsp. palli-dum is the causative agent of syphilis, a chronic multistagesexually transmitted disease. New cases of syphilis now affectmore than 15 million people each year, resulting in a 20%increase in global syphilis cases over the past 3 years. Themajority of these infections occur within developing nations(26), although over the course of the past decade rapidly in-creasing annual syphilis infection rates have been observed inNorth America (38), Europe (61), and Australia (37). Congen-ital syphilis remains a major global pediatric health concern,with 50% of infected fetuses dying in utero or shortly afterbirth. Also, it is now well established that T. pallidum infectionresults in an increased risk of acquiring and transmitting HIV(53).

Treponema pallidum is a highly invasive pathogen capable ofattaching to host cells, invading the tissue barrier and under-going rapid dissemination via the circulatory system. In hu-mans, the primary T. pallidum localized infection manifestsitself as a chancre; however, the bacteria disseminate rapidly,resulting in widespread infection (40). In vitro, the bacteria are

capable of penetrating intercellular junctions of endothelialcell monolayers (71, 72) and mouse abdominal wall tissuebarriers (62). In rabbits, T. pallidum is capable of entering thecirculatory system minutes after intratesticular inoculation (17,60). The ability to disseminate so widely and rapidly is likely tobe a crucial mechanism of pathogenesis for this bacterium;however, investigations focusing on the identification of pro-teins involved in T. pallidum pathogenesis have been hindereddue to the fact that the microbe is an obligate human pathogenthat cannot be continuously cultured in vitro. As a conse-quence, T. pallidum virulence factors have yet to be identified.Complete genome sequencing of the 1.14-Mbp Nichols strainidentified 17% of the open reading frames (ORFs) as codingfor hypothetical proteins, with an additional 28% of ORFscategorized as those of unknown function (24). These findings,in conjunction with an absence of identified potential classicalbacterial virulence factors from the remaining 55% of anno-tated ORFs, further underscore the current lack of under-standing regarding proteins critical for the virulence andpathogenesis of this important enigmatic human pathogen andthe difficulty in identifying suitable targets for vaccine design.

Adhesin-mediated microbial adherence is a key mechanismin the colonization, establishment, and dissemination of patho-genic bacteria (52, 55). Although it is now 13 years since thepublication of the Nichols strain complete genome sequence(24) and over 30 years since initial binding studies demon-strated that T. pallidum is capable of adhering to mammaliancells (3, 21, 30) and components of the extracellular matrix

* Corresponding author. Mailing address: Department of Biochem-istry and Microbiology, University of Victoria, P.O. Box 3055 STNCSC, Victoria, BC V8W 3P6, Canada. Phone: (250) 853-3189. Fax:(250) 721-8855. E-mail: [email protected].

† Present address: Department of Medical Microbiology and Immu-nology, University of Alberta, 116 St. and 85th Ave., Edmonton, Al-berta T6G 2R3, Canada.

� Published ahead of print on 13 December 2010.

1386

on March 10, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

(ECM) (23, 70), data regarding the molecular basis of theseimportant pathogenic interactions have only recently begun toemerge. Previously, our laboratory identified the T. pallidumadhesin Tp0751 and demonstrated that this adhesin binds spe-cifically to a variety of laminin isoforms, which are abundantglycoprotein components of the basement membrane that un-derlie endothelial cell layers, in a dose-dependent manner (8,9). Furthermore, the presence of Tp0751-specific antibodies inserum from both natural and experimental T. pallidum infec-tions was also demonstrated (8), indicating that the protein isexpressed during the course of infection. Heterologous expres-sion of Tp0751 on the surface of the culturable nonadherentspirochete Treponema phagedenis was shown to confer uponthe bacterium the ability to bind laminin, a specific interactionwhich was inhibited by antibodies generated against recombi-nant T. pallidum Tp0751 (11). Although this interaction washypothesized to be central to the infection process (8, 9), theexact role of Tp0751 in T. pallidum tissue invasion and rapidwidespread dissemination remained to be determined.

Expression of proteases in pathogenic bacteria is a mecha-nism commonly employed for the purpose of host componentand tissue proteolysis (50, 68). This process has been linked tothe virulence of many pathogenic bacteria, including Vibriocholerae (27), Pseudomonas aeruginosa (48), and Legionellapneumophila (13, 51), by observations that it promotes bacte-rial invasion and dissemination of infection via the degradationof structural tissue components and plasma proteins. Host-interacting proteases have yet to be identified within T. palli-dum at either an experimental or a bioinformatic level, al-though the highly invasive nature of this bacterium suggeststheir existence.

Fibrinogen (Fg) is an abundant protein synthesized primar-ily in the liver but also present in the blood at approximately 2mg/ml. It is considered a positive acute-phase reactant proteinsince plasma concentrations increase in response to inflamma-tion. It is a 340-kDa glycoprotein dimer comprised of threepairs of nonidentical polypeptide chains, namely, � (�67 kDa),� (�55 kDa), and � chains (�48 kDa) (32). This multifunc-tional protein is a key component of the host coagulationcascade and is also found in the ECM at damaged tissue sites

(64). The primary role of fibrinogen is that of the key structuralclotting protein in blood plasma; as such, fibrinogen is con-verted to fibrin by the action of thrombin. This function isbelieved to be important for bacterial containment; however,many pathogenic bacteria have evolved mechanisms to over-come this host defense strategy.

In the current study we further investigate the binding ca-pability of the T. pallidum adhesin Tp0751 by expressing thefull-length mature protein in a soluble form and by demon-strating the capacity of this protein to bind to human fibrino-gen. We also show that Tp0751 is palmitoylated when ex-pressed in the spirochete Treponema phagedenis, possesseszinc-dependent protease activity, and is capable of degradinghuman fibrinogen and laminin. We therefore propose to assignTp0751 the name “pallilysin,” a name that reflects a combina-tion of the nomenclature used for several bacterial proteasesinvolved in host component degradation and the Treponemadenticola protease dentilisin (36). To our knowledge, this rep-resents the first published report of a T. pallidum proteincapable of degrading host molecules and suggests a novel pro-teolytic mechanism for tissue damage, destruction, and dissem-ination of T. pallidum during the course of infection.

MATERIALS AND METHODS

Bacteria. T. pallidum subsp. pallidum (Nichols strain) was propagated in NewZealand White rabbits as described elsewhere (45). All animal studies wereapproved by the University of Victoria Institutional Review Board and con-ducted in accordance with the Canadian Council on Animal Care (CCAC)guidelines.

Host proteins. Plasminogen-depleted human fibrinogen (Calbiochem) waspurchased from VWR International (Mississauga, Ontario, Canada). Lamininisolated from an Engelbreth-Holm-Swarm murine sarcoma was purchased fromSigma Chemical Co. (Oakville, Ontario, Canada).

Construct cloning and site-directed mutagenesis. For Escherichia coli recom-binant protein expression, Tp0751 (GenBank accession number AAC65720)DNA fragments encoding amino acid residues C24 to P237, F25 to P237, and S78to P237 were PCR amplified from T. pallidum subsp. pallidum (Nichols strain)genomic DNA using the primers listed in Table 1. Each amplicon was cloned intothe T7 promoter destination expression vector pDEST-17 (which had beenmodified to incorporate an N-terminal thrombin site for histidine tag removal),according to the manufacturer’s instructions (Gateway Technology, Invitrogen).The pDEST-17 expression vector allows for the generation of N-terminally fusedhexahistidine-tagged recombinant proteins. For insect recombinant protein ex-

TABLE 1. Primers used to amplify DNA for recombinant protein expression

ORF (type)b Primerorientation Sequencea

Tp0751_C24-P237 (E) Sense 5�-GGGGACAAGTTTGTACAAAAAAGCAGGCTGCTTTCAGCACGGTCACAntisense 5�-GGGGACCACTTTGTACAAGAAAGCTGGGTTTCAGGGCGAAGGAGCACTAG

Tp0751_C24-P237 (I) Sense 5�-CTACTACCATGGCTTGCTTTCAGCACGGTCACGAntisense 5�-CTACTAGCGGCCGCGGGCGAAGGAGCACTAGC

Tp0751_F25-P237 (E) Sense 5�-GGGGACAAGTTTGTACAAAAAAGCAGGCTTTCAGCACGGTCACGAntisense 5�-GGGGACCACTTTGTACAAGAAAGCTGGGTTTCAGGGCGAAGGAGCACTAG

Tp0751_S78-P237 (E) Sense 5�-GGGGACAAGTTTGTACAAAAAAGCAGGCTCACATGGAAACGCCCCGCCAntisense 5�-GGGGACCACTTTGTACAAGAAAGCTGGGTTTCAGGGCGAAGGAGCACTAG

Tp0751_S78-P237 (E) (M) Sense 5�-GGTAACATTTGAATCCCACCAGGTGATACACGTAAGGGAntisense 5�-CCCTTACGTGTATCACCTGGTGGGATTCAAATCTTACC

Tp0453 (E) Sense 5�-CTAGACCATATGGCATCAGTAGATCCGTTGGAntisense 5�-GTCAGCTCGAGTCACGAACTTCCCTTTTTGGAG

LIC12032/KatE (E) Sense 5�-CTAGACCATATGATGAGTAGAAAAACCCTTACTACAntisense 5�-GTCAGAAGCTTTTAAACTCGACTCAGAGCAAG

a Restriction sites are highlighted in bold. Base changes resulting in the site-directed mutation are underlined.b E, E. coli expressed; I, insect expressed; M, mutated.

VOL. 79, 2011 BIFUNCTIONAL ROLE OF T. PALLIDUM Tp0751 1387


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

pression, Tp0751 DNA fragments encoding amino acid residues C24 to P237were PCR amplified using the primers listed in Table 1. The amplicon was thencloned into the NotI/NcoI sites of a pAcGP67B baculovirus transfer vector (BDBiosciences, Mississauga, Ontario, Canada) modified to incorporate a C-termi-nal hexahistidine tag and thrombin cleavage site. This vector also included theGP67 secretion sequence which allows for the export of recombinant Tp0751protein into the insect culture medium. For Tp0453 (GenBank accession numberAAC65443) expression, a DNA fragment encoding amino acid residues A32 toS287 was PCR amplified from T. pallidum subsp. pallidum (Nichols strain)genomic DNA using the primers listed in Table 1. The amplicon was cloned intothe NdeI/XhoI sites of the pET28a expression vector (Novagen) according to themanufacturer’s instructions. The negative control, katE (LIC12032) (GenBankaccession number AAS70607), was cloned as previously described (20). Forsite-directed mutagenesis, the pDEST-17 Tp0751_S78-P237 construct describedabove was subjected to replacement of E199 with Q199 using the primers listedin Table 1 and the QuikChange site-directed mutagenesis kit (Stratagene; pur-chased from VWR International), according to the manufacturer’s instructions.The sequence and reading frames of all the expression constructs were verifiedas correct by DNA sequencing.

Recombinant protein expression and purification. For E. coli recombinantprotein expression, the pDEST-17 expression constructs of tp0751 were trans-formed into the E. coli expression strain BL21-AI (Invitrogen). Transformantswere cultured in LB medium containing 100 �g/ml ampicillin at 37°C until theA600 reached 1.7 to 2.0. Bacterial cultures were induced with 0.2% L-arabinosefor 16 h at 16°C, harvested by centrifugation, resuspended in buffer (20 mMHEPES [pH 7.5], 500 mM NaCl, 20 mM imidazole, 1% glycerol), flash frozen inliquid nitrogen, and stored at �80°C. After thawing, cell pellets were lysed byadding CHAPS {3-[(3-cholamidopropyl)-dimethylammonio]-1-propanesulfon-ate} buffer to a final concentration of 5 mg/ml (Sigma) and subjected to mildsonication and centrifugation at 13,000 � g for 30 min at 4°C. The tp0453construct was transformed into the E. coli expression strain BL21 Star(DE3)(Invitrogen), and transformants were cultured in LB medium broth containing 50�g/ml kanamycin at 37°C until the A600 reached 1.0. Cultures were induced with0.4 mM IPTG (isopropyl-�-D-thiogalactopyranoside) for 16 h at 16°C. Fast pro-tein liquid chromatography (FPLC) purification of soluble histidine-tagged re-combinant proteins was performed using immobilized metal ion affinity chroma-tography (IMAC). Briefly, proteins were injected onto 1-ml HisTrap FF affinitycolumns (GE Healthcare, Baie D’Urfe, Quebec City, Canada) prepacked withprecharged nickel Sepharose 6 fast flow resin using an AKTA Prime Plus FPLCsystem (GE Healthcare). The columns were washed with cold buffer (20 mMimidazole, 20 mM HEPES [pH 7.5], 500 mM NaCl, 1% glycerol), and the boundproteins were eluted with a gradient of cold elution buffer (0 to 500 mM imid-azole, 20 mM HEPES [pH 7.5], 500 mM NaCl, 1% glycerol). Fractions weremonitored measuring the optical density at 280 nm (OD280) using the AKTAprime plus spectrophotometer. The purity of eluted proteins was determined bysodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) andCoomassie brilliant blue staining. LIC12032 (KatE) was expressed as previouslydescribed (20) and purified as described above. Histidine tag removal fromrecombinant Tp0751 was accomplished by cleavage with thrombin (Sigma) ac-cording to the manufacturer’s instructions. Fractions containing the recombinantproteins (estimated purity of 95%) were pooled, concentrated using 10-kDa-molecular-mass-cutoff Millipore Amicon Ultra centrifugal filter units (FisherScientific, Ottawa, Ontario, Canada), and loaded onto a preparative gel filtrationcolumn (HiLoad 16/60 Superdex 75; GE Healthcare) previously equilibrated in20 mM HEPES [pH 7.5], 150 mM NaCl, and 1% glycerol. Recombinant proteinswere eluted in cold elution buffer (20 mM HEPES [pH 7.5], 150 mM NaCl, and1% glycerol). Peak protein fractions were analyzed by SDS-PAGE, concentratedas described above, quantitated by measuring the absorbance at 280 nm and bythe Thermo Scientific BCA protein assay kit (Fisher Scientific), flash frozen inliquid nitrogen, and stored at �80°C. Insoluble recombinant histidine-taggedTp0751 LBR protein was expressed and purified as previously described (8).Insect recombinant protein expression was performed as previously described(16).

Fibrinogen-binding adherence assays. To test for the adherence of recombi-nant His-tagged proteins (Tp0751 and the negative-control Tp0453) to fibrino-gen, enzyme-linked immunosorbent assay (ELISA)-based assays were performedas previously described (8). The plates were read at 600 nm with a BioTek ELISAplate reader (Fisher Scientific). Statistical analyses were performed using theStudent two-tailed t test.

Lipoprotein determination. The 17-octadecynoic acid (17-ODYA) labelingand detection of palmitoylated Tp0751 protein from Treponema phagedenistransformed with the tp0751 gene from T. pallidum subsp. pallidum Nichols strain(11) were based on the method developed by Martin and Cravatt (46). Briefly,

cultured cells were metabolically labeled by adding 17-ODYA (Cayman Chem-icals, Ann Arbor, MI) to tp0751-transformed T. phagedenis cultures grown toearly exponential phase (1.0 � 108 cells/ml). Treponema phagedenis culturestransformed with tp0751 were grown in the absence of 17-ODYA as a negativecontrol. Bacteria were grown until late exponential phase (7.0 � 108 cells/ml),washed with Tris-buffered saline (TBS), lysed using sonication, and separatedinto soluble (supernatant) and insoluble (membrane) fractions by centrifugation.The insoluble pellet was resuspended in 0.5 ml phosphate-buffered saline(PBS)-1% Triton X-100, and rotated at 4°C for 1 h. Samples were desalted withNAP-5 columns (GE Healthcare) to remove excess probe and other probe-incorporated metabolites, and the protein concentration was determined by UV(280 nm) absorbance. Click chemistry was used to add the reporter group,biotin-azide, to the lysate. Briefly, 1.5 mg/ml lysate per culture was reacted with100 �M biotin-azide (Invitrogen) for 1 h in the presence of 1 mM Tris phosphine(TCEP; Sigma), 100 �M Tris ([1-benzyl-1H-1,2,3 triazol-4-yl] methyl) amine(TBTA; Sigma) in 20% dimethyl sulfoxide (DMSO)-80% t-butanol, and 1 mMCuSO4. Precipitated protein was washed twice with cold methanol, and proteinpellets were sonicated in 1.2% SDS-PBS until solubilized. Samples were boiledfor 5 min, diluted in PBS, and frozen at �80°C overnight. After the samples werethawed on ice, streptavidin-agarose beads (Thermo Scientific, Waltham, MA)were added to the samples and rotated for 90 min at room temperature. Sampleswere washed 1 time with 10 ml PBS-0.2% SDS, 3 times with 10 ml PBS, and 3times with 10 ml 50 mM Tris (pH 6.8). Biotin-labeled lipoproteins bound tostreptavidin beads were separated by SDS-PAGE using 12% precast Tris-glycinepolyacrylamide gels (Invitrogen) and stained with GelCode blue stain (ThermoScientific), and bands corresponding to the approximate size of Tp0751 wereexcised and analyzed by mass spectrometry as described below.

LC-MS/MS. Protein bands were reduced in 10 mM dithiothreitol, alkylated in100 mM iodoacetamide, and subjected to in-gel tryptic digestion for 5 h at 37°Cand lyophilization to dryness. The protein samples were rehydrated in 2% ace-tonitrile-0.1% formic acid, concentrated, and desalted. Samples were subjectedto liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis us-ing a Waters NanoAcquity UPLC coupled to an Applied Biosystems (Carlsbad,CA)/MDS Sciex QSTAR Pulsar I mass spectrometer. Mass spectrum analyseswere acquired by collecting a 1-s time-of-flight (TOF) MS survey scan (400 to1,600 m/z) followed by two 2.5-s product ion scans. LC-MS/MS data weresearched by centroiding with Analyst QS 1.1 Mascot script 1.6 b24 (MDS Sciex,Concord, Ontario, Canada) to create the Mascot generic format file (mgf) fordatabase searching. The mgf files were searched against the Uniprot-Swissprot20090225 (410,518 sequences; 148,080,998 residues) “all species” or “Eubacteriaspecies only” and “Treponema pallidum (Nichols strain)” databases with thefollowing parameters: fixed modifications included carbamidomethyl (C), deami-dation (NQ), oxidation (M), and propionamide (C); NHS-biotin was set as avariable modification; peptide mass tolerance was 0.3 amu; fragment masstolerance was 0.15 amu; and enzyme specificity was set for trypsin.

In vitro substrate degradation assays. The E. coli-expressed recombinant pro-teins Tp0751_C24-P237, Tp0751_S78-P237 (wild-type and mutant E199Q), andKatE (LIC12032) and the insect-expressed Tp0751_C24-P237 recombinant pro-tein (30 �g) were mixed with 60 �g of plasminogen-free human fibrinogen orlaminin in protease activation buffer (25 mM Tris [pH 7.5], 25 mM CaCl2). Thereaction mixtures were incubated at 37°C for 0 to 24 h, and samples wereremoved at various time points. Samples from each time point (10 �l) weremixed with 10 �l of sample buffer and heated at 95°C for 10 min, and thethree fibrinogen chains, (�, �, and �) or three laminin chains (A, B1, and B2)were separated and analyzed for degradation by electrophoresis in 15% and 10%Tris-glycine polyacrylamide gels, respectively, at a constant voltage of 200 V for1 h. Gels were stained in Coomassie brilliant blue R-250 solution and destainedin 5% (vol/vol) acetic acid, 20% (vol/vol) methanol, and 75% (vol/vol) H2O. Fordensitometry analysis of fibrinogen and laminin degradation, the freely availableImageJ software from the National Institutes of Health was used (http://rsb.info.nih.gov/ij). The following parameters were used for analysis: image type was setto grayscale/inverted, background noise subtracted with rolling ball radius wasset to 50, scale/unit of length was set to pixels, and measurements were set toarea, mean gray value, and integrated density. Relative intensities of SDS-PAGEbands were calculated as the mean gray value at a t of 24 h/mean gray value ata t of 0 h.

Fibrinogen degradation FRET assays. A fluorescence resonance energy trans-fer (FRET)-based assay was developed to analyze the fibrinogenolytic activity ofTp0751. Plasminogen-free human fibrinogen was labeled with fluorescein iso-thiocyanate (FITC) as previously described (78). The FITC-labeled fibrinogenassay for measuring Tp0751 protease activity was based on the method of Twin-ing (73). Briefly, the recombinant proteins (Tp0751_C24-P237 and the negative-control KatE) and the positive-control protease, trypsin (Promega, Madison,

1388 HOUSTON ET AL. INFECT. IMMUN.


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

WI), were diluted in activation buffer (25 mM Tris [pH 7.5], 25 mM CaCl2) togive a final concentration of 10 �g/ml. One hundred microliters of the proteinsolutions was added in triplicate to sterile Corning Costar 96-well plates (1.0 �gper well) (Fisher Scientific). Blank sample wells consisted of activation buffer inthe absence of recombinant proteins. FITC-labeled fibrinogen (10 �g) was addedto all sample wells, mixed gently, and incubated at 37°C in the dark for 0 to 48 h.Fluorescence was measured hourly using a standard fluorescein excitation/emis-sion filter set at 485 nm/538 nm and a BioTek Synergy HT plate reader (FisherScientific). The mean blank fluorescence readings were subtracted from themean sample fluorescence readings for each protein, and the increase in relativefluorescence units (RFU) was plotted against time. Statistical analyses wereperformed using Student’s two-tailed t test.

Inductively coupled plasma-mass spectrometry analysis. Metal cofactorspresent in Tp0751 protein samples were analyzed at Cantest Ltd. (Burnaby,British Columbia, Canada) using inductively coupled plasma-mass spectrometry(ICP-MS). Briefly, Tp0751_S78-P237 protein and purification buffer (20 mMHEPES [pH 7.5], 150 mM NaCl, 1% glycerol) were dispensed into individual15-ml polypropylene tubes. Both samples were reacted with 200 �l of nitric acidand heated in a 95°C hot water bath for 1 h. The digested solutions were madeup to a final volume of 5 ml with deionized water. The samples were analyzed byconventional ICP-MS at a dilution of �5. The metal elements Zn, Ca, Mn, Mg,Fe, and Co were fully quantified against a certified standard using single-pointcalibration. Sample analysis and operation of the ICP-MS were done accordingto Cantest’s in-house standard operating procedures.

Protease inhibition assays. To investigate the effect of various broad-spectrumprotease inhibitors on the proteolytic activity of Tp0751, recombinant proteins inprotease activation buffer (25 mM Tris [pH 7.5], 25 mM CaCl2) were incubatedat 37°C for 1 h in the presence and absence of 10 mM EDTA, 1 mM and 10 mM1,10-phenanthroline, 1 mM phenylmethylsulfonyl fluoride (PMSF), 1 mM ben-zamidine, or 200 �M trans-epoxysuccinyl-L-leucylamido-(4-guanidino) butane(E64) (all inhibitors purchased from Sigma). Inhibitor stock solutions wereprepared immediately before use and filter sterilized using 0.2 �m syringe filters(Sarstedt Inc., Montreal, Quebec City, Canada). The proteolytic activity of thesamples was analyzed using the SDS-PAGE-based in vitro substrate degradationand FRET-based degradation assays as detailed above.

RESULTS

Soluble expression of recombinant Tp0751 protein. In orderto investigate the potential role of the adhesin Tp0751 in T.pallidum dissemination, several Tp0751 constructs were clonedand tested for soluble protein expression (Table 2). Solubleexpression was achieved only when C-terminal amino acids, inparticular amino acid residues P231 to P237 (PASAPSP), wereincluded in the construct design. Truncation of N-terminalresidues M1 to R77 did not adversely affect soluble proteinexpression. Electrophoresis of purified Tp0751 proteins by

SDS-PAGE followed by Coomassie staining (Fig. 1A and B)revealed the presence of multiple bands, and Western blotanalysis using Tp0751-specific polyclonal antibodies (Fig. 1D)similarly detected multiple immunoreactive protein bands. E.coli (Fig. 1A)- and insect (Fig. 1B)-expressed Tp0751_C24-P237 molecular masses ranged from approximately 32 kDa toapproximately 15 kDa. In-gel tryptic digestion and analysis ofthe resulting peptides using matrix-assisted laser desorptionionization (MALDI)–TOF MS and LC-MS/MS confirmed thatall bands detected by GelCode blue staining were Tp0751(data not shown). The MS sequence coverage data indicatedthat the lower-mass bands represent Tp0751 protein moleculeswhich had been reduced in mass due to cleavage of aminoacids from the protein termini (data not shown). The calcu-lated molecular mass of His-tagged recombinant Tp0751_C24-P237 is 25.8 kDa; however, when analyzed using SDS-PAGE,the main Tp0751 protein band migrated in the gel with amobility similar to that of a protein corresponding to approx-imately 32 kDa (Fig. 1). The slower-than-predicted anomalousmigration of the protein may be due to the high proline con-tent in the N terminus of the protein (8.4% from residueC24-P237 and 16.9% from residue C24-T100), as anomalousSDS-PAGE migration has been observed with other proline-rich proteins (33).

Recombinant Tp0751 binds specifically to human fibrino-gen. We hypothesized that, similar to the situation with otherinvasive bacteria, the interaction of T. pallidum with coagula-tion proteins is important for the rapid widespread dissemina-tion of this pathogen via the circulatory system. Therefore, theability of Tp0751 to bind specifically to human fibrinogen wasinvestigated using an ELISA-based assay which compared theattachments of the adhesin and a negative-control recombi-nant protein, Tp0453, to immobilized plasminogen-free humanfibrinogen and the positive-control host protein, laminin.Tp0453 is predicted to be located in the outer membrane of T.pallidum (31) and has previously been shown to lack significantbinding to laminin (8). The soluble recombinant proteinsTp0751_C24-P237 (data not shown) and Tp0751_F25-P237(Fig. 2) exhibited statistically significant levels of binding toboth the positive-control host protein, laminin (P � 0.0001),and to plasminogen-free human fibrinogen (P � 0.0045) com-pared to the levels of laminin and fibrinogen binding exhibitedby Tp0453. These results suggest that Tp0751 is an importantfactor mediating interactions with multiple host proteins indifferent host environments during the course of infection.

Heterologously expressed Tp0751 is S-palmitoylated. Previ-ous in silico analysis of Tp0751 indicated the presence of aputative signal peptide II (SpII) cleavage site, suggesting thatthe protein may be lipidated within T. pallidum (66). In orderto confirm that Tp0751 is a lipoprotein, we adapted the mam-malian protein palmitoylation global profiling methodology ofMartin and Cravatt (46) for use in the prokaryote T. phagede-nis. For these experiments, we used Tp0751-expressing T.phagedenis (11) to enable the lipidation state of the Tp0751adhesin to be determined in a culturable treponeme. Tp0751-expressing T. phagedenis cultures were metabolically labeled bygrowing cells in the presence of the palmitate analog 17-octa-decynoic acid (17-ODYA), while negative-control cells weregrown in the absence of the analog. The reporter group, biotin-azide, was reacted with the insoluble membrane fractions, and

TABLE 2. Recombinant expression of Tp0751 constructs

Constructa Expressionvector(s)

Amino acidsexpressed

Estimatedexpression profile

Tp0751_R54-S185 pDEST-17 R54–S185 100% insolubleTp0751_R54-V200 pDEST-17,

pET28a,pET32

R54–V200 100% insoluble

Tp0751_R54-G215 pDEST-17 R54–G215 100% insolubleTp0751_R54-W220 pDEST-17 R54–W220 100% insolubleTp0751_R54-T225 pDEST-17 R54–T225 95% insolubleTp0751_R54-G230 pDEST-17 R54–G230 95% insolubleTp0751_R54-P237 pDEST-17 R54–P237 40% solubleTp0751_S78-P237 pDEST-17 S78–P237 40% solubleTp0751_S78_P237 (M) pDEST-17 S78–P237 40% solubleTp0751_F25-P237 pDEST-17 F25–P237 50% solubleTp0751_C24-P237 pDEST-17 C24–P237 50% solubleTp0751_C24-P237 (I) pAcGP67B C24–P237 50% soluble

a All constructs were expressed in E. coli unless otherwise stated. M, mutatedprotein (E199Q); I, insect expressed.



http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

biotin-labeled octadecynoylated proteins resulting from theCu(I)-catalyzed azide-alkyne cycloaddition reaction (clickchemistry) were purified on streptavidin beads, separated bySDS-PAGE, excised, and analyzed by LC-MS/MS. Followingelectrophoresis, approximately 15 GelCode blue-stained bandscorresponding to proteins ranging from �20 kDa to �60 kDawere detected in Tp0751-transformed T. phagedenis cells cul-tured in the presence, but not the absence, of 17-ODYA (Fig.

3). To ensure that we could detect palmitoylated Tp0751 ifpresent, regions of the gel within the molecular mass range of20 kDa to 34 kDa were excised from lanes corresponding to thecultures grown in both the presence and absence of 17-ODYAand subjected to LC-MS/MS analysis (Fig. 3). As shown inTable 3, Tp0751 was detected in the purified proteins from T.phagedenis grown in the presence of 17-ODYA but not in cellsgrown in the absence of analog. These results provide directevidence that Tp0751 is expressed and S-palmitoylated intransformed T. phagedenis and thus provide further evidencethat the protein is membrane associated in T. pallidum.

Recombinant Tp0751 degrades human fibrinogen and lami-nin. The invasive nature of T. pallidum suggests the possibleexistence of host-interacting proteases within this spirochete.The detection of multiple Tp0751 bands on SDS-PAGE gels,combined with the fact that the current study and previousstudies within our laboratory (8, 9) have demonstratedTp0751-ECM interactions, prompted the investigation of theprimary sequence of Tp0751 for the presence of potentialprotease motifs. A putative conserved zinc-binding HEXXHmotif (HEVIH), which is found in approximately 50% ofknown metalloendopeptidases, was identified in the C termi-nus at amino acids H198 to H202. In order to determine ifTp0751 is a protease, recombinant proteins were expressedand purified from both a baculovirus-insect cell expressionsystem (Tp0751_C24-P237) and an E. coli expression system(Tp0751_S78-P237 and Tp0751_C24-P237). The purified re-combinant proteins were incubated with either plasminogen-

FIG. 1. Analysis of soluble E. coli- and insect-expressed recombinant Tp0751. Purified recombinant Tp0751_C24-P237 wild-type (WT) andTp0751_S78-P237 E199Q mutant (M) proteins from E. coli (A and C) and insect (B) expression systems were separated by SDS-PAGE and stainedwith GelCode blue (E. coli WT) or Coomassie brilliant blue (insect WT and E. coli M) staining solutions. (D) E. coli Tp0751_C24-P237 WT proteinwas analyzed via Western blot (WB) methodology using Tp0751-specific polyclonal antibodies. Numbers to the left of the lanes indicate the sizes(kDa) of the corresponding molecular mass (MM) markers. Electrophoresis of purified Tp0751_C24-P237 protein consistently revealed thepresence of multiple protein bands, corresponding to peptides with molecular masses ranging from approximately 32 kDa to 15 kDa. The majorTp0751_C24-P237 protein bands (�32 kDa) are indicated by arrows. Excision of each band from the gel, followed by MALDI-TOF andLC-MS/MS analyses, confirmed that all bands detected by GelCode blue staining were derived from Tp0751 (data not shown).

FIG. 2. Soluble Tp0751 binds host components. The ability of sol-uble recombinant Tp0751 to bind plasminogen-free human fibrinogenand laminin was assessed using an ELISA-based assay. Average read-ings (A600) are presented with bars indicating standard error (SE). Theresults are representative of five independent experiments. For statis-tical analyses, attachment to laminin and fibrinogen by Tp0751 wascompared to the attachment of Tp0453 using Student’s two-tailed ttest. Tp0751 exhibited a statistically significant level of binding to boththe positive-control host protein, laminin (P � 0.0001), and to plas-minogen-free human fibrinogen (P � 0.0045) compared to the level oflaminin and fibrinogen binding exhibited by Tp0453.



http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

free human fibrinogen or laminin at 37°C. These proteins werechosen as potential substrates, as we had previously demon-strated their interaction with Tp0751 via binding assays. Sam-ples were removed at defined time intervals, and degradationwas analyzed by SDS-PAGE and subsequent staining withCoomassie blue. E. coli-expressed Tp0751_S78-P237 degradedthe �, �, and � chains of human fibrinogen by 24 h postincu-bation, whereas the corresponding negative controls failed todegrade any of the three fibrinogen chains over 24 h (Fig. 4A).Time course analysis of fibrinogen degradation by the full-length mature Tp0751 protein (C24 to P237) indicated that the

� and � chains were completely degraded by the E. coli-ex-pressed protein within 8 to 16 h of incubation, with almostcomplete degradation of the � chain by 16 to 24 h. Similarly,the insect-expressed protein degraded the � chain by 8 to 16 hand the � chain by 16 to 24 h (Fig. 4B). Removal of theN-terminal histidine tag from the full-length mature proteinresulted in enhanced proteolysis (compared to the rate offibrinogen degradation by the His-tagged Tp0751 protein),with complete degradation of all three fibrinogen chains by 6 to18 h (Fig. 4C). Partial hydrolysis of laminin A, B1, and B2chains by the full-length mature Tp0751 protein (C24 to P237)was also observed following 24 h of incubation (Fig. 4D).Densitometry analysis confirmed the laminin degradation, asthe relative intensity levels (mean intensity of the protein bandat t � 24 h/mean intensity of the protein band at t � 0 h) of thelaminin A chain (400 kDa) and laminin B1/B2 chain (210/210kDa) bands decreased by 62% and 23%, respectively, following24 h of incubation with Tp0751. Densitometry (data notshown) and SDS-PAGE analyses (Fig. 4A to D) indicated thatfibrinogen and laminin remained stable over the course of 24 hin the negative controls (fibrinogen/laminin incubated in bufferwithout a recombinant protein, and fibrinogen/laminin incu-bated in the presence of the recombinant Leptospira proteinLIC12032/KatE). Interestingly, Tp0751 protease constructspurified from both insect cells (data not shown) and E. coli(Fig. 4E) also underwent further degradation over the courseof the 24 h of incubation.

A FRET-based assay was developed to further confirm thefibrinogenolytic activity of Tp0751. FITC quenching occurs inheavily labeled FITC proteins due to the close proximity of thelabels. However, upon digestion of the intact molecule by pro-teases, the FITC-labeled protein becomes dequenched, result-ing in a measurable increase in fluorescence, indicative ofproteolysis. FITC-labeled fibrinogen was incubated with E.coli-expressed Tp0751_C24-P237, and fibrinogen degradationwas detected by measuring the increase in fluorescence inten-sities over 48 h. The relative fluorescence units (RFU) for thepositive control (trypsin) increased from 13.7 RFU at 0 hpostincubation to 10,150 RFU at 3 h postincubation, repre-senting a 743-fold increase in relative fluorescence, and 48% ofthe total fluorescence detected at 48 h postincubation (21,091RFU). As the graph for trypsin is approximately linear from 0to 3 h, the maximum specific activity of trypsin, calculated asthe relative increase in fluorescence units per minute, is 56.3RFU min�1. As expected, these results indicate rapid degra-dation of human fibrinogen by trypsin immediately followingincubation of the protease with substrate, followed by slower

FIG. 3. Heterologously expressed Tp0751 is lipid modified.Tp0751-expressing T. phagedenis cultures were metabolically labeledby growing cells in the presence of the palmitate analog 17-ODYA.Following click chemistry, octadecynoylated proteins were purifiedusing streptavidin beads, separated by SDS-PAGE and stained withGelCode blue reagent. Protein bands from cultures grown in the pres-ence ( 17-ODYA) or absence (�17-ODYA; negative control) of thepalmitate analog, corresponding to proteins ranging from �20 kDa to�34 kDa, were excised from the gel lanes (excised areas indicated byrectangles) and analyzed by LC-MS/MS. For comparison, the SDS-PAGE migration mobility of purified recombinant Tp0751_C24-P237is shown (rTp0751). Numbers to the left of the lanes indicate the sizes(kDa) of the corresponding molecular mass (MM) markers.

TABLE 3. Palmitoylated proteins identified by LC-MS/MS from 17-ODYA labeled and non-labeled tp0751-transformed T. phagedenisd

Protein/species 17-ODYA Uniprot no. Mass(Da)

No. ofobservedpeptidesa

MASCOTscoreb

Seq. cov.(%)c

Tp0751/T. pallidum 083732 Y751_ TREPA 25,810 3 101 15No protein detected � NA NA NA NA NA

a No. of observed peptides includes all peptides that differ either by sequence, modification, or charge, with a MASCOT score equal to or greater than the significancethreshold (P � 0.05).

b A MASCOT score of 39 is equal to or greater than the significance threshold (P � 0.05).c Sequence coverage (Seq. cov.) is based on peptides with a unique sequence.d NA, not applicable.



http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

FIG. 4. Recombinant insect-expressed and E. coli-expressed Tp0751 proteins degrade human fibrinogen and laminin and undergo self-degradation. (A) Recombinant Tp0751_S78-P237 protein purified from E. coli and (B) recombinant Tp0751_C24-P237 proteins purified from E.coli and insect cells were incubated with plasminogen-free human fibrinogen. Two negative controls were included: fibrinogen incubated in bufferwithout a recombinant protein (Fg only) and fibrinogen incubated in the presence of the recombinant Leptospira protein LIC12032/KatE (Fg plusKatE, negative control). (C) Following cleavage of the N-terminal histidine tag from full-length mature recombinant Tp0751, the cleaved andnoncleaved proteins were incubated with fibrinogen. (D) Recombinant Tp0751_C24-P237 protein purified from E. coli was incubated with laminin(LMN). Two negative controls were included: laminin incubated in buffer without a recombinant protein (LMN only) and laminin incubated inthe presence of the recombinant Leptospira protein LIC12032/KatE (LMN plus KatE, negative control). (E) Recombinant Tp0751 protease was



http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

trypsin degradation over the next 45 h, presumably due to theloss of trypsin activity (Fig. 4F). In contrast, the RFU forTp0751 increased slowly during the first 16 h, from zero fluo-rescence above background to 2,365 RFU, representing 8.8%of the total fluorescence detected at 48 h postincubation(27,005 RFU). However, the RFU for Tp0751 increased from2,365 at 16 h postincubation to 24,821 RFU at 35 h postincu-bation, representing a 10.5-fold increase in relative fluores-cence and 92% of the total fluorescence detected at 48 hpostincubation (27,005 RFU) (Fig. 4F). The maximum specificactivity of Tp0751, calculated as the relative increase in fluo-rescence units per minute between 22 and 28 h postincubation,is 37.1 RFU min�1. These results are in agreement with theSDS-PAGE analysis (Fig. 4B) and confirm that, after an initiallag period of approximately 16 h, the Tp0751 protease initiatesrapid degradation of human fibrinogen to a level that exceedsthe degradation potential of trypsin. As shown in Fig. 4F,recombinant Tp0751_C24-P237 exhibited a statistically signif-icant level of fibrinogenolysis at 20 h postincubation (P �0.0021) and beyond, compared to the level of fibrinogen deg-radation exhibited by the negative-control recombinant pro-tein KatE. These results demonstrate that, in addition to thehost component binding capability of Tp0751, the adhesin isalso capable of degrading these host proteins.

Tp0751 protease binds zinc and is inhibited by metallopro-tease inhibitors. Quantitative inductively coupled plasma-massspectrometry (ICP-MS) was used to investigate Tp0751 diva-lent metal ion binding. Analyses indicated that Tp0751 bindszinc and calcium, but not iron, magnesium, cobalt, or manga-nese (Table 4). In order to determine the catalytic type of theHEXXH motif-containing Tp0751 protease, various proteaseinhibitors were incubated with the recombinant proteins andthe effect on fibrinogenolysis was analyzed using SDS-PAGEand the FRET-based assays. The metal chelator EDTA (10mM) completely inhibited the fibrinogenolytic activity of in-sect-expressed Tp0751_C24-P237, as all three fibrinogenchains were found to be stable, when analyzed by SDS-PAGEand Coomassie blue staining, for up to 24 h after incubation(Fig. 5A). Fibrinogenolysis was not inhibited when E. coli-expressed Tp0751_S78-P237 and Tp0751_C24-P237 were incu-bated with the serine protease inhibitor PMSF (1 mM), theserine protease inhibitor benzamidine (1 mM), or the cysteineprotease inhibitor E64 (200 �M) (Fig. 5B to D). However,incubation of recombinant Tp0751 with the zinc chelator 1,10-phenanthroline (1 to 10 mM) abolished fibrinogenolytic activ-ity (Fig. 5B to D). Fibrinogen was degraded by recombinant

Tp0751 to similar extents in both the presence (1 to 100 �M)and absence of zinc. This may be explained by the fact that zincis ubiquitous in nature and may have been incorporated intothe protease active site during E. coli expression and purifica-tion, likely originating from the media and buffers used duringthe purification. Zinc concentrations greater than 100 �M in-hibited the fibrinogenolytic activity of Tp0751 (Fig. 5E), whichis consistent with the finding that high zinc concentrationsinhibit metalloprotease activity due to the formation of a zincmonohydroxide complex in the active site (41) or the presenceof two zinc ions in the active site (12, 34). Interestingly, whencalcium was present at 0 to 1 mM concentrations, only the �and � chains were degraded, whereas increasing calcium con-centrations above 10 mM enhanced proteolysis as seen by thedegradation of all three fibrinogen chains (Fig. 5E). Site-di-rected mutagenesis studies indicated that the Tp0751 E199Q mu-tant degrades fibrinogen at a rate comparable to that of thewild-type protein (data not shown). Similar to that of the wild-type Tp0751 protein, the fibrinogenolytic activity of the Tp0751E199Q mutant was inhibited by the metalloprotease inhibitor1,10-phenanthroline, but not by serine or cysteine protease inhib-itors (data not shown).

DISCUSSION

It has been demonstrated that bacterial adhesins can pro-mote antiphagocytosis (6, 57), initial establishment and main-tenance of infection, and bacterial dissemination by interferingwith critical sites in fibrinogen chains which are normally re-quired for essential host defense responses to bacterial infec-tions, including platelet and thrombin interaction sites (18, 56,58). Although T. pallidum and T. pallidum proteins have been

degraded further over the course of the 24-h fibrinogen degradation assays. The major Tp0751 band (�32 kDa) is indicated by an arrow. In allexperiments, samples were removed over a 24-h time period at the indicated intervals and loaded onto SDS-PAGE gels. Degradation of thefibrinogen �, �, and � chains and the laminin A (400-kDa), B1 (210-kDa), and B2 (200-kDa) chains and degradation of the Tp0751 protease weredetected by staining with Coomassie blue. Numbers to the left of the lanes indicate the sizes (kDa) of the corresponding molecular mass (MM)markers. (F) A FRET-based assay was performed to further confirm the fibrinogenolytic activity of Tp0751. FITC-labeled plasminogen-free humanfibrinogen was incubated for 48 h with either recombinant E. coli-expressed Tp0751_C24-P237, the positive-control protein trypsin, or thenegative-control protein KatE. The degree of fibrinogen degradation was measured every hour by detecting the increase in relative fluorescenceunits (RFU) over 48 h using standard fluorescein excitation/emission filters. The increase in RFU for the three proteins is shown. Averagefluorescence intensity readings from triplicate measurements are presented with bars indicating standard error (SE). The results are representativeof four independent experiments. For statistical analyses, fibrinogen degradation by Tp0751 was compared to that of KatE using Student’stwo-tailed t test. The recombinant Tp0751_C24-P237 protein exhibited a statistically significant level of fibrinogenolysis at 20 h postincubation (P �0.0021) and beyond, compared to the level of fibrinogen degradation exhibited by the negative-control recombinant protein KatE.

TABLE 4. ICP-MS quantification of metal ions in purifiedrecombinant Tp0751 protein

Protein ormetal ion

Concn in sample(nM)a

Ratio of �metalion�:�Tp0751�

Tp0751 17.0Zn 7.0 0.4:1.0Ca 6.8 0.4:1.0Fe NDMg NDCo NDMn ND

a ND, not detected.



http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

FIG. 5. Proteolytic activity of Tp0751 is abolished by metalloprotease inhibitors. (A) Insect-expressed Tp0751_C24-P237 was preincubated at37°C with or without the metalloprotease inhibitor EDTA. E. coli-expressed Tp0751_C24-P237 (B) and Tp0751_S78-P237 (C) were preincubatedat 37°C with or without the serine protease inhibitors PMSF and benzamidine, the cysteine protease inhibitor E64, and the metalloproteaseinhibitor 1,10-phenanthroline. The proteins were incubated with fibrinogen for 24 h, removed at the indicated intervals, and loaded ontoSDS-PAGE gels, and degradation of the fibrinogen �, �, and � chains was detected by staining with Coomassie blue. The negative controls includedfibrinogen incubated in buffer without a recombinant protein (Fg only) and fibrinogen incubated in the presence of the recombinant Leptospiraprotein, LIC12032/KatE (Fg KatE). (D) E. coli-expressed Tp0751_C24-P237 was incubated with or without E64 and 1,10 phenanthroline.FITC-labeled fibrinogen was incubated for 48 h with the Tp0751_C24-P237 proteins and the negative-control protein KatE. The degree offibrinogen degradation was measured every hour by detecting the increase in relative fluorescence units (RFU) using standard fluoresceinexcitation/emission filters. Average fluorescence intensity readings from triplicate measurements are presented with bars indicating standard error(SE). The results are representative of three independent experiments. For statistical analyses, fibrinogen degradation by Tp0751 in the presence(1 mM) and degradation in the absence of 1,10-phenanthroline were compared using Student’s two-tailed t test. The Tp0751_C24-P237 proteinnot exposed to the metalloprotease inhibitor exhibited a statistically significant level of fibrinogenolysis at 25 h postincubation (P � 0.0281) andbeyond, compared to the level of fibrinogen degradation exhibited by the Tp0751 protein exposed to 1,10-phenanthroline. (E) The effect ofincreasing calcium and zinc concentrations on fibrinogenolysis by Tp0751 was analyzed using the same SDS-PAGE assay described above. Numbersto the left of the lanes indicate the sizes (kDa) of the corresponding molecular mass (MM) markers.

1394


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

shown to bind to various host components, including fibronec-tin (10, 22), laminin, collagen I, and collagen IV (8, 9, 23), thisrepresents the first report to identify a T. pallidum proteincapable of binding specifically to human fibrinogen. Fibrinogenexpression is significantly upregulated as a host defense re-sponse to infection and inflammation, with the thrombin-cat-alyzed conversion of fibrinogen to fibrin clots allowing forpathogen containment and localization by providing a struc-tural protective barrier to the surrounding tissues to preventbacterial invasion and dissemination (42). During T. palliduminfection, the bacterium could interfere with this fibrinogenfunction through Tp0751-mediated binding at several stages,including at the initial site of infection during chancre forma-tion, in the bloodstream during dissemination via the circula-tory system, and at sites of damaged tissue.

The ability of bacteria to degrade fibrinogen via the expres-sion of fibrinogenolytic proteases, thereby inhibiting or pre-venting containment through coagulation, is an importantvirulence factor in various pathogenic bacterial species, pro-moting rapid widespread dissemination via the circulatory sys-tem (29, 35, 47, 50). The oral spirochete Treponema denticolaexpresses a chymotrypsin-like protease (CTLP) that is capableof degrading several host proteins, including epithelial junc-tion-associated proteins and fibrinogen, which in turn is be-lieved to promote bacterial colonization, hemostatic interfer-ence, and tissue invasion (2, 19, 36, 74). In the current study,recombinant Tp0751 proteins purified from both insect and E.coli expression systems degraded plasminogen-free human fi-brinogen. We were also able to demonstrate that, as for otherinvasive pathogenic bacteria, Tp0751 is capable of degradingthe abundant ECM component laminin. This attribute mayalso contribute to T. pallidum dissemination, given the fact thatthe bacterium must traverse several basement membranes dur-ing the course of infection. The current study represents thefirst published report to identify a T. pallidum protein that iscapable of degrading human molecules.

Tp0751 possesses a potential metalloprotease zinc-bindingHEXXH motif, binds zinc and calcium, and is inhibited by thezinc-chelating metalloprotease inhibitor 1,10-phenanthrolineand the general metal chelator EDTA. Based upon these find-ings we propose to designate Tp0751 “pallilysin” to match thenomenclature used for several pathogenic bacterial proteasescapable of degrading human proteins. However, the designa-tion of the protease as a true metalloprotease cannot be madeat this time, since site-directed mutagenesis performed in thisstudy indicated that, unlike other HEXXH-containing metal-loproteases (25, 43), mutation of the glutamic acid residuewithin the potential pallilysin metalloprotease motif did notaffect the proteolytic potential. In light of this finding and thefact that 50% of known metalloproteases do not contain anHEXXH motif, it is apparent that further mutagenesis studiesneed to be performed to identify the pallilysin active site res-idues.

Candidate factors proposed to mediate the rapid, wide-spread dissemination of and associated tissue damage and de-struction by T. pallidum have included periplasmic flagella,chemotaxis, inflammation, and the adaptive immune response(40); however, the exact role of these factors in T. pallidumdissemination remains to be determined. The identification ofthe proteolytic activity of pallilysin in the current study indi-

cates that T. pallidum expresses at least one protease that ispotentially involved in promoting bacterial dissemination inthe ECM and via the circulatory system. Together, the bindingand proteolytic results further implicate pallilysin as a criticalfactor in the virulence of T. pallidum by showing that it pro-motes the attachment to and degradation of important hostmolecules, thereby potentially facilitating colonization of anddissemination from various host infection sites. Genome se-quencing indicated that T. pallidum has limited metabolic ca-pabilities, including a lack of amino acid biosynthesis pathways,but expresses an abundance of amino acid, carbohydrate, andcation transporters (24). Hence, binding and degradation offibrinogen, laminin, and potentially other host factors intosmall peptides and amino acids may not only promote dissem-ination through inhibition of coagulation but may also be ben-eficial for the pathogen during infection by providing a richnutrient source.

Gel-based analysis of heterologously expressed proteasesoften results in the appearance of multiple protein bands whichmay be due to degradation by E. coli proteases or autolysis (65,77). The fact that pallilysin formed multiple bands when ana-lyzed via SDS-PAGE following purification from both a lonand ompT protease-deficient E. coli expression strain and aninsect secretory protein expression system suggests that theprotease may be synthesized as a proprotein which undergoesself-lysis, resulting in cleavage of the disordered proline-richN-terminal prodomain and generation of the mature activeprotease. Such an autocatalytic mechanism has been shown tobe essential for the activation of thermolysin-like metallopro-teases (5, 39), which are potent virulence factors of variouspathogenic bacteria involved in extracellular protein degrada-tion. Membrane type 1 matrix metalloprotease (MT1-MMP), akey protease involved in tumor cell invasion, is also activatedvia autocatalysis (63). In the current study, the possibility ofautocatalytic activation of pallilysin is further supported by ourfibrinogen degradation data which demonstrated that maximalproteolytic activity for full-length pallilysin is attained onlyfollowing an 8 to 16-h lag period in vitro, as well as by MSanalysis of the multiple pallilysin bands excised from SDS-polyacrylamide gels and by the fact that purified pallilysincontinues to undergo further degradation during incubationwith the host substrate.

The aforementioned extended proteolytic lag period, com-bined with the finding that near-equimolar concentrations ofpallilysin are required for substrate degradation, indicates thatthe specific fibrinogenolytic activity of the recombinant pro-tease in vitro is initially relatively modest but rapidly increasesafter 16 h. This may be due partially to the recombinant natureof the pallilysin protease that was utilized in this study, since wewere able to demonstrate enhanced substrate proteolysis fol-lowing cleavage of the terminal histidine tag. The full proteo-lytic potential of the native T. pallidum protease awaits furtherinvestigation.

A requirement for bacterial proteins involved in host-patho-gen interactions is surface exposure. The identification of sur-face-exposed outer membrane proteins in T. pallidum hasproven to be technically challenging and thus highly contro-versial due to the fragility arising from the unusual ultrastruc-ture of this bacterium. In addition, several reports have pro-vided evidence for a paucity of T. pallidum integral outer



http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

membrane proteins on the surface of this bacterium (14, 15, 44,59, 76). Although not previously documented for T. pallidum,one potential strategy to achieve surface exposure is the an-chorage of a lipid-modified cysteine residue within the N ter-minus of the protein to the bacterial membrane, as is found inmany bacteria, including the related spirochetes Borrelia burg-dorferi and Leptospira (28). To investigate the possibility thatpallilysin exists as a lipoprotein within T. pallidum, the currentstudy determined the lipidation state of heterologously ex-pressed pallilysin. Previous labeling studies of T. pallidum pro-teins have utilized [3H]palmitate as the lipoprotein label (1, 4,67, 69), and here we used an alternative click-chemistry-basedtechnology originally developed for the global analysis of thepalmitoylation of eukaryotic proteins (46) to demonstrate thatheterologously expressed pallilysin is palmitoylated in the re-lated spirochete T. phagedenis. This work represents the firstreport which has successfully adapted this eukaryotic palmi-toylation profiling technique for the identification of proteinlipidation in bacteria, thus representing a powerful new toolfor studying both human and bacterial protein palmitoylation.The obtained palmitoylation results for pallilysin confirm aprevious in silico analysis that suggested that this protein existsas a lipoprotein within T. pallidum (66). This result, whencombined with the results from a previous study that demon-strated surface exposure of heterologously expressed Tp0751within T. phagedenis (11), suggests the possibility that pallilysinis a surface-exposed protease within T. pallidum that is lipi-dated via palmitoylation of the cysteine residue at position 24and is, through this cellular location, capable of interactingwith host components.

Treponema pallidum has been referred to as the “stealthpathogen” due to its ability to cause lifetime latent infection. Aprevious study, which examined a group of six recombinant T.pallidum proteins for their sensitivities and specificities withsera from patients with syphilis, detected a low antibody re-sponse to Tp0751/pallilysin (75). In this study, ELISA-basedassays showed 25 of the 43 tested syphilis patient sera, at allstages of disease, failed to recognize Tp0751/pallilysin. At leastfour studies have detected antibodies specific for Tp0751/pal-lilysin during the course of natural and experimental syphilis(7, 8, 49, 75), indicating that the protein is expressed duringinfection. Although reverse transcriptase PCR studies haveshown that tp0751 is transcribed in T. pallidum, Western blotanalyses have consistently failed to detect pallilysin expressionin T. pallidum (S. Houston and C. E. Cameron, unpublishedobservations). Combined, these results suggest an inherent lowimmunogenicity and/or a low level of expression of pallilysinduring infection. This feature, together with the host ECMbinding (8, 9) and proteolysis capabilities presented herein,may help explain, at least in part, how this pathogen is able toevade the host immune response during widespread dissemi-nation.

The limited effectiveness of current worldwide public healthprograms in controlling syphilis further underpins the need fordeveloping a novel approach for halting increasing T. palliduminfection rates through an alternative means of syphilis pre-vention. The current study is the first published report to iden-tify and isolate a palmitoylated T. pallidum protein that iscapable of binding and degrading at least two important hu-man proteins, namely, laminin and fibrinogen, that have po-

tential roles in controlling disseminated bacterial infection.Future studies investigating the ability of pallilysin to bind anddegrade a diverse array of human proteins and the mechanismof proteolysis will be essential in determining the contributionof this protein to the rapid widespread dissemination of thishighly invasive human pathogen.

ACKNOWLEDGMENTS

We thank Azad Eshghi for his methodological ideas and technicalassistance regarding lipoprotein labeling and mass spectrometry, re-spectively, and for supplying the Leptospira putative KatE expressionconstruct. We thank Joanna Crawford for performing the insect pro-tein expression.

This work was funded by Public Health Service grant AI-051334from the National Institutes of Health, as well as by awards from theCanada Foundation for Innovation, the Michael Smith Foundation forHealth Research, and the British Columbia Knowledge DevelopmentFund. Support for this research was also provided in part by the BCProteomics Network. C.E.C. is a Canada Research Chair in MolecularPathogenesis and a Michael Smith Foundation for Health ResearchScholar.

REFERENCES

1. Akins, D. R., B. K. Purcell, M. M. Mitra, M. V. Norgard, and J. D. Radolf.1993. Lipid modification of the 17-kilodalton membrane immunogen ofTreponema pallidum determines macrophage activation as well as amphiphi-licity. Infect. Immun. 61:1202–1210.

2. Bamford, C. V., J. C. Fenno, H. F. Jenkinson, and D. Dymock. 2007. Thechymotrypsin-like protease complex of Treponema denticola ATCC 35405mediates fibrinogen adherence and degradation. Infect. Immun. 75:4364–4372.

3. Baseman, J. B., and E. C. Hayes. 1980. Molecular characterization of recep-tor binding proteins and immunogens of virulent Treponema pallidum. J.Exp. Med. 151:573–586.

4. Belisle, J. T., M. E. Brandt, J. D. Radolf, and M. V. Norgard. 1994. Fattyacids of Treponema pallidum and Borrelia burgdorferi lipoproteins. J. Bacte-riol. 176:2151–2157.

5. Bitar, A. P., M. Cao, and H. Marquis. 2008. The metalloprotease of Listeriamonocytogenes is activated by intramolecular autocatalysis. J. Bacteriol. 190:107–111.

6. Boschwitz, J. S., and J. F. Timoney. 1994. Inhibition of C3 deposition onStreptococcus equi subsp. equi by M protein: a mechanism for survival inequine blood. Infect. Immun. 62:3515–3520.

7. Brinkman, M. B., et al. 2006. Reactivity of antibodies from syphilis patientsto a protein array representing the Treponema pallidum proteome. J. Clin.Microbiol. 44:888–891.

8. Cameron, C. E. 2003. Identification of a Treponema pallidum laminin-bind-ing protein. Infect. Immun. 71:2525–2533.

9. Cameron, C. E., N. L. Brouwer, L. M. Tisch, and J. M. Kuroiwa. 2005.Defining the interaction of the Treponema pallidum adhesin Tp0751 withlaminin. Infect. Immun. 73:7485–7494.

10. Cameron, C. E., E. L. Brown, J. M. Kuroiwa, L. M. Schnapp, and N. L.Brouwer. 2004. Treponema pallidum fibronectin-binding proteins. J. Bacte-riol. 186:7019–7022.

11. Cameron, C. E., et al. 2008. Heterologous expression of the Treponemapallidum laminin-binding adhesin Tp0751 in the culturable spirochete Trepo-nema phagedenis. J. Bacteriol. 190:2565–2571.

12. Coffey, A., B. van den Burg, R. Veltman, and T. Abee. 2000. Characteristicsof the biologically active 35-kDa metalloprotease virulence factor from Lis-teria monocytogenes. J. Appl. Microbiol. 88:132–141.

13. Conlan, J. W., A. Baskerville, and L. A. Ashworth. 1986. Separation ofLegionella pneumophila proteases and purification of a protease which pro-duces lesions like those of Legionnaires’ disease in guinea pig lung. J. Gen.Microbiol. 132:1565–1574.

14. Cox, D. L., P. Chang, A. W. McDowall, and J. D. Radolf. 1992. The outermembrane, not a coat of host proteins, limits antigenicity of virulent Trepo-nema pallidum. Infect. Immun. 60:1076–1083.

15. Cox, D. L., et al. 27 September 2010. Surface immunolabeling and consensuscomputational framework to identify candidate rare outer membrane pro-teins of Treponema pallidum. Infect. Immun. doi:10.1128/IAI.00834-10.

16. Crawford, J., et al. 2010. Structural and functional characterization of Spo-roSAG: a SAG2-related surface antigen from Toxoplasma gondii. J. Biol.Chem. 285:12063–12070.

17. Cumberland, M. C., and T. B. Turner. 1949. The rate of multiplication ofTreponema pallidum in normal and immune rabbits. Am. J. Syph. GonorrheaVener. Dis. 33:201–212.



http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

18. Davis, S. L., S. Gurusiddappa, K. W. McCrea, S. Perkins, and M. Hook.2001. SdrG, a fibrinogen-binding bacterial adhesin of the microbial surfacecomponents recognizing adhesive matrix molecules subfamily from Staphy-lococcus epidermidis, targets the thrombin cleavage site in the B� chain.J. Biol. Chem. 276:27799–27805.

19. Ellen, R. P., K. S. Ko, C. M. Lo, D. A. Grove, and K. Ishihara. 2000.Insertional inactivation of the prtP gene of Treponema denticola confirmsdentilisin’s disruption of epithelial junctions. J. Mol. Microbiol. Biotechnol.2:581–586.

20. Eshghi, A., P. A. Cullen, L. Cowen, R. L. Zuerner, and C. E. Cameron. 2009.Global proteome analysis of Leptospira interrogans. J. Proteome Res. 8:4564–4578.

21. Fitzgerald, T. J., J. N. Miller, and J. A. Sykes. 1975. Treponema pallidum(Nichols strain) in tissue cultures: cellular attachment, entry, and survival.Infect. Immun. 11:1133–1140.

22. Fitzgerald, T. J., and L. A. Repesh. 1985. Interactions of fibronectin withTreponema pallidum. Genitourin. Med. 61:147–155.

23. Fitzgerald, T. J., L. A. Repesh, D. R. Blanco, and J. N. Miller. 1984. Attach-ment of Treponema pallidum to fibronectin, laminin, collagen IV, and col-lagen I, and blockage of attachment by immune rabbit IgG. Br. J. Vener. Dis.60:357–363.

24. Fraser, C. M., et al. 1998. Complete genome sequence of Treponema palli-dum, the syphilis spirochete. Science 281:375–388.

25. Fushimi, N., C. E. Ee, T. Nakajima, and E. Ichishima. 1999. Aspzincin, afamily of metalloendopeptidases with a new zinc-binding motif. Identifica-tion of new zinc-binding sites (His128, His132, and Asp164) and three catalyt-ically crucial residues (Glu129, Asp143, and Tyr106) of deuterolysin fromAspergillus oryzae by site-directed mutagenesis. J. Biol. Chem.274(34):24195–24201.

26. Gerbase, A. C., J. T. Rowley, D. H. Heymann, S. F. Berkley, and P. Piot. 1998.Global prevalence and incidence estimates of selected curable STDs. Sex.Transm. Infect. 74(Suppl. 1):S12–S16.

27. Ghosh, A., et al. 2006. Enterotoxigenicity of mature 45-kilodalton and pro-cessed 35-kilodalton forms of hemagglutinin protease purified from a chol-era toxin gene-negative Vibrio cholerae non-O1, non-O139 strain. Infect.Immun. 74:2937–2946.

28. Haake, D. A. 2000. Spirochaetal lipoproteins and pathogenesis. Microbiology146(7):1491–1504.

29. Harris, T. O., D. W. Shelver, J. F. Bohnsack, and C. E. Rubens. 2003. Anovel streptococcal surface protease promotes virulence, resistance toopsonophagocytosis, and cleavage of human fibrinogen. J. Clin. Invest.111:61–70.

30. Hayes, N. S., K. E. Muse, A. M. Collier, and J. B. Baseman. 1977. Parasitismby virulent Treponema pallidum of host cell surfaces. Infect. Immun. 17:174–186.

31. Hazlett, K. R., et al. 2005. TP0453, a concealed outer membrane protein ofTreponema pallidum, enhances membrane permeability. J. Bacteriol. 187:6499–6508.

32. Henschen, A., F. Lottspeich, M. Kehl, and C. Southan. 1983. Covalentstructure of fibrinogen. Ann. N. Y. Acad. Sci. 408:28–43.

33. Ho, H. Y., R. Rohatgi, L. Ma, and M. W. Kirschner. 2001. CR16 forms acomplex with N-WASP in brain and is a novel member of a conservedproline-rich actin-binding protein family. Proc. Natl. Acad. Sci. U. S. A.98:11306–11311.

34. Holland, D. R., A. C. Hausrath, D. Juers, and B. W. Matthews. 1995.Structural analysis of zinc substitutions in the active site of thermolysin.Protein Sci. 4:1955–1965.

35. Imamura, T., et al. 1995. Effect of free and vesicle-bound cysteine protein-ases of Porphyromonas gingivalis on plasma clot formation: implications forbleeding tendency at periodontitis sites. Infect. Immun. 63:4877–4882.

36. Ishihara, K., T. Miura, H. K. Kuramitsu, and K. Okuda. 1996. Character-ization of the Treponema denticola prtP gene encoding a prolyl-phenylala-nine-specific protease (dentilisin). Infect. Immun. 64:5178–5186.

37. Jin, F., et al. 2005. Epidemic syphilis among homosexually active men inSydney. Med. J. Aust. 183:179–183.

38. Kent, M. E., and F. Romanelli. 2008. Reexamining syphilis: an update onepidemiology, clinical manifestations, and management. Ann. Pharmaco-ther. 42:226–236.

39. Kooi, C., C. R. Corbett, and P. A. Sokol. 2005. Functional analysis of theBurkholderia cenocepacia ZmpA metalloprotease. J. Bacteriol. 187:4421–4429.

40. Lafond, R. E., and S. A. Lukehart. 2006. Biological basis for syphilis. Clin.Microbiol. Rev. 19:29–49.

41. Larsen, K. S., and D. S. Auld. 1991. Characterization of an inhibitory metalbinding site in carboxypeptidase A. Biochemistry 30:2613–2618.

42. Levi, M., T. van der Poll, and H. R. Buller. 2004. Bidirectional relationbetween inflammation and coagulation. Circulation 109:2698–2704.

43. Li, L., T. Binz, H. Niemann, and B. R. Singh. 2000. Probing the mechanisticrole of glutamate residue in the zinc-binding motif of type A botulinumneurotoxin light chain. Biochemistry 39(9):2399–2405.

44. Liu, J., et al. 17 September 2010. Cellular architecture of Treponema palli-

dum: novel flagellum, periplasmic cone, and cell envelope as revealed by cryoelectron tomography. J. Mol. Biol. doi:10.1016/j.jmb.2010.09.020.

45. Lukehart, S. A., S. A. Baker-Zander, and S. Sell. 1980. Characterizationof lymphocyte responsiveness in early experimental syphilis. I. In vitroresponse to mitogens and Treponema pallidum antigens. J. Immunol.124:454–460.

46. Martin, B. R., and B. F. Cravatt. 2009. Large-scale profiling of proteinpalmitoylation in mammalian cells. Nat. Methods 6:135–138.

47. Matsuka, Y. V., S. Pillai, S. Gubba, J. M. Musser, and S. B. Olmsted. 1999.Fibrinogen cleavage by the Streptococcus pyogenes extracellular cysteine pro-tease and generation of antibodies that inhibit enzyme proteolytic activity.Infect. Immun. 67:4326–4333.

48. Matsumoto, K. 2004. Role of bacterial proteases in pseudomonal and ser-ratial keratitis. Biol. Chem. 385:1007–1016.

49. McKevitt, M., et al. 2005. Genome scale identification of Treponema palli-dum antigens. Infect. Immun. 73:4445–4450.

50. Miyoshi, S., and S. Shinoda. 2000. Microbial metalloproteases and patho-genesis. Microbes Infect. 2:91–98.

51. Moffat, J. F., P. H. Edelstein, D. P. Regula, Jr., J. D. Cirillo, and L. S.Tompkins. 1994. Effects of an isogenic Zn-metalloprotease-deficient mutantof Legionella pneumophila in a guinea-pig pneumonia model. Mol. Micro-biol. 12:693–705.

52. Norman, M. U., et al. 2008. Molecular mechanisms involved in vascularinteractions of the Lyme disease pathogen in a living host. PLoS Pathog.4:e1000169.

53. Nusbaum, M. R., R. R. Wallace, L. M. Slatt, and E. C. Kondrad. 2004.Sexually transmitted infections and increased risk of co-infection with humanimmunodeficiency virus. J. Am. Osteopath. Assoc. 104:527–535.

54. Reference deleted.55. Patti, J. M., B. L. Allen, M. J. McGavin, and M. Hook. 1994. MSCRAMM-

mediated adherence of microorganisms to host tissues. Annu. Rev. Micro-biol. 48:585–617.

56. Pei, L., M. Palma, M. Nilsson, B. Guss, and J. I. Flock. 1999. Functionalstudies of a fibrinogen binding protein from Staphylococcus epidermidis.Infect. Immun. 67:4525–4530.

57. Poirier, T. P., M. A. Kehoe, E. Whitnack, M. E. Dockter, and E. H. Beachey.1989. Fibrinogen binding and resistance to phagocytosis of Streptococcussanguis expressing cloned M protein of Streptococcus pyogenes. Infect. Im-mun. 57:29–35.

58. Ponnuraj, K., et al. 2003. A “dock, lock, and latch” structural model for astaphylococcal adhesin binding to fibrinogen. Cell 115:217–228.

59. Radolf, J. D., M. V. Norgard, and W. W. Schulz. 1989. Outer membraneultrastructure explains the limited antigenicity of virulent Treponema palli-dum. Proc. Natl. Acad. Sci. U. S. A. 86:2051–2055.

60. Raiziss, G. W., and M. Severac. 1937. Rapidity with which Spirochaeta pallidainvades the bloodstream. Arch. Dermatol. Syphilol. 35:1101–1109.

61. Righarts, A. A., I. Simms, L. Wallace, M. Solomou, and K. A. Fenton. 2004.Syphilis surveillance and epidemiology in the United Kingdom. Euro Sur-veill. 9:21–25.

62. Riviere, G. R., D. D. Thomas, and C. M. Cobb. 1989. In vitro model ofTreponema pallidum invasiveness. Infect. Immun. 57:2267–2271.

63. Rozanov, D. V., and A. Y. Strongin. 2003. Membrane type-1 matrix metal-loproteinase functions as a proprotein self-convertase. Expression of thelatent zymogen in Pichia pastoris, autolytic activation, and the peptide se-quence of the cleavage forms. J. Biol. Chem. 278:8257–8260.

64. Rybarczyk, B. J., S. O. Lawrence, and P. J. Simpson-Haidaris. 2003. Matrix-fibrinogen enhances wound closure by increasing both cell proliferation andmigration. Blood 102:4035–4043.

65. Schlomann, U., et al. 2002. The metalloprotease disintegrin ADAM8. Pro-cessing by autocatalysis is required for proteolytic activity and cell adhesion.J. Biol. Chem. 277:48210–48219.

66. Setubal, J. C., M. Reis, J. Matsunaga, and D. A. Haake. 2006. Lipopro-tein computational prediction in spirochaetal genomes. Microbiology152:113–121.

67. Shevchenko, D. V., et al. 1999. Membrane topology and cellular location ofthe Treponema pallidum glycerophosphodiester phosphodiesterase (GlpQ)ortholog. Infect. Immun. 67:2266–2276.

68. Supuran, C. T., A. Scozzafava, and A. Mastrolorenza. 2001. Bacterial pro-teases: current therapeutic use and future prospects for the development ofnovel antibiotics. Expert Opin. Ther. Pat. 11:221–259.

69. Swancutt, M. A., J. D. Radolf, and M. V. Norgard. 1990. The 34-kilodaltonmembrane immunogen of Treponema pallidum is a lipoprotein. Infect. Im-mun. 58:384–392.

70. Thomas, D. D., J. B. Baseman, and J. F. Alderete. 1985. Fibronectin medi-ates Treponema pallidum cytadherence through recognition of fibronectincell-binding domain. J. Exp. Med. 161:514–525.

71. Thomas, D. D., A. M. Fogelman, J. N. Miller, and M. A. Lovett. 1989.Interactions of Treponema pallidum with endothelial cell monolayers. Eur. J.Epidemiol. 5:15–21.

72. Thomas, D. D., et al. 1988. Treponema pallidum invades intercellular junctions ofendothelial cell monolayers. Proc. Natl. Acad. Sci. U. S. A. 85:3608–3612.



http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

73. Twining, S. S. 1984. Fluorescein isothiocyanate-labeled casein assay forproteolytic enzymes. Anal. Biochem. 143:30–34.

74. Uitto, V. J., D. Grenier, E. C. Chan, and B. C. McBride. 1988. Isolation of achymotrypsinlike enzyme from Treponema denticola. Infect. Immun.56:2717–2722.

75. Van Voorhis, W. C., et al. 2003. Serodiagnosis of syphilis: antibodies torecombinant Tp0453, Tp92, and Gpd proteins are sensitive and specificindicators of infection by Treponema pallidum. J. Clin. Microbiol. 41:3668–3674.

76. Walker, E. M., G. A. Zampighi, D. R. Blanco, J. N. Miller, and M. A.

Lovett. 1989. Demonstration of rare protein in the outer membrane ofTreponema pallidum subsp. pallidum by freeze-fracture analysis. J. Bac-teriol. 171:5005–5011.

77. Witte, V., N. Wolf, T. Diefenthal, G. Reipen, and H. Dargatz. 1994. Heter-ologous expression of the clostripain gene from Clostridium histolyticum inEscherichia coli and Bacillus subtilis: maturation of the clostripain precursoris coupled with self-activation. Microbiology 140(5):1175–1182.

78. Wu, J. H., and S. L. Diamond. 1995. A fluorescence quench and dequenchassay of fibrinogen polymerization, fibrinogenolysis, or fibrinolysis. Anal.Biochem. 224:83–91.

Editor: S. R. Blanke



http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

Documents

Bifunctional Role of the Treponema pallidum …Treponema pallidum, the causative agent of syphilis, is a highly invasive pathogenic spirochete capable of attaching to host cells, invading