Structural, Bioinformatic, and In Vivo Analyses of Two Treponema pallidum Lipoproteins Reveal a...

doi:10.1016/j.jmb.2012.01.015 J. Mol. Biol. (2012) 416, 678–696

Contents lists available at www.sciencedirect.com

Journal of Molecular Biologyj ourna l homepage: ht tp : / /ees .e lsev ie r.com. jmb

Structural, Bioinformatic, and In Vivo Analyses of TwoTreponema pallidum Lipoproteins Reveal a UniqueTRAP Transporter

Ranjit K. Deka 1†, Chad A. Brautigam2†, Martin Goldberg 1,Peter Schuck 3, Diana R. Tomchick 2 and Michael V. Norgard 1⁎1Department of Microbiology, The University of Texas Southwestern Medical Center at Dallas,5323 Harry Hines Boulevard, Dallas, TX 75390, USA2Department of Biochemistry, The University of Texas Southwestern Medical Center at Dallas,5323 Harry Hines Boulevard, Dallas, TX 75390, USA3Dynamics of Molecular Assembly Section, Laboratory of Cellular Imaging and Macromolecular Biophysics,National Institute of Biomedical Imaging and Bioengineering, National Institutes of Health, Bethesda,MD 20892, USA

Received 12 October 2011;received in revised form15 December 2011;accepted 11 January 2012Available online27 January 2012

Edited by R. Huber

Keywords:TRAP transporter;syphilis;Treponema pallidum;TPR motif;protein interactions

*Corresponding author. E-mail add† R.K.D. and C.A.B. contributed eAbbreviations used: AUC, analyti

PDB, Protein Data Bank; PEG, polyeSEC, size-exclusion chromatographytransporter; TPR, tetratricopeptide r

0022-2836/$ - see front matter © 2012 E

Treponema pallidum, the bacterial agent of syphilis, is predicted to encodeone tripartite ATP-independent periplasmic transporter (TRAP-T). TRAP-Ts typically employ a periplasmic substrate-binding protein (SBP) to deliverthe cognate ligand to the transmembrane symporter. Herein, we demon-strate that the genes encoding the putative TRAP-T components from T.pallidum, tp0957 (the SBP), and tp0958 (the symporter), are in an operon withan uncharacterized third gene, tp0956. We determined the crystal structureof recombinant Tp0956; the protein is trimeric and perforated by a pore. Partof Tp0956 forms an assembly similar to those of “tetratricopeptide repeat”(TPR) motifs. The crystal structure of recombinant Tp0957 was alsodetermined; like the SBPs of other TRAP-Ts, there are two lobes separatedby a cleft. In these other SBPs, the cleft binds a negatively charged ligand.However, the cleft of Tp0957 has a strikingly hydrophobic chemicalcomposition, indicating that its ligand may be substantially different andlikely hydrophobic. Analytical ultracentrifugation of the recombinantversions of Tp0956 and Tp0957 established that these proteins associateavidly. This unprecedented interaction was confirmed for the nativemolecules using in vivo cross-linking experiments. Finally, bioinformaticanalyses suggested that this transporter exemplifies a new subfamily ofTPATs (TPR-protein-associated TRAP-Ts) that require the action of a TPR-containing accessory protein for the periplasmic transport of a potentiallyhydrophobic ligand(s).

© 2012 Elsevier Ltd. All rights reserved.

ress: [email protected] to this work.cal ultracentrifugation; β-OG, n-octyl-β-D-glucoside; HA, hydrophobic antechamber;thylene glycol; RT-PCR, reverse transcription PCR; SBP, substrate-binding protein;; SV, sedimentation velocity; TRAP-T, tripartite ATP-independent periplasmicepeat.

lsevier Ltd. All rights reserved.

mailto:[email protected]

http://dx.doi.org/10.1016/j.jmb.2012.01.015

http://www.sciencedirect.com/science/journal/00222836

679A TPR-associated TRAP-T (TPAT)

Introduction

Treponema pallidum, the etiologic agent of syphilis,is an obligate human pathogen that metabolicallyrelies heavily on its human host.1 The spirocheteencodes only 1067 genes (cf. the genome ofEscherichia coli K-12, which has over 4000 genes). T.pallidum lacks genes of the biosynthetic pathways forfatty acids, nucleotides, most of the amino acids, andmany other vital metabolites. Natural selection hasthus honed T. pallidum to be an organism with anextremely parsimonious genome, likely at the cost ofits ability to be free living. From a more practicalperspective, the need for T. pallidum to acquire somany of its vital nutrients from its human host likelyhas accounted for the historic inability to continu-ously cultivate the organism in vitro.2 Indeed, muchof our lack of understanding of this enigmatichuman pathogen derives directly from this majorinvestigative obstacle.Bacteria use many means to import needed

metabolites from their extracellular milieu intotheir cytoplasms. The primary ABC transportersand secondary symporters are examples.3,4 ABCtransporters utilize a substrate-binding protein(SBP) in the periplasm to secure the ligand anddeliver it to a transmembrane permease, which usesthe energy from ATP hydrolysis to transport theligand against a concentration gradient into thecytoplasm. Canonical symporters do not require theservices of an SBP; rather, a periplasmic ligand bindsdirectly to the transmembrane symporter. Anabundant ion (usually either H+ or Na+) alsobinds, and the favorable energetics of cotransportingthe ion are used to allow delivery of the ligand to thecytoplasm, again typically against a concentrationgradient.Recently, a different family of solute transporters

has been described—the tripartite ATP-independentperiplasmic transporters (TRAP-Ts).5 The TRAP-Tsincorporate features of both the primary and thesecondary transporters. The dctPQM operon ofRhodobacter capsulatus is the prototypical TRAP-T;it is responsible for the import of C4-dicarboxylates.

6

Like the ABC transporters, the DctPQM system andall other TRAP-Ts employ an SBP (called “P” herein)to capture the ligand7 and deliver it to thetransmembrane parts of the transporter (called“Q” and “M” herein).6 ATP hydrolysis does notenergize these transporters. Rather, like secondarytransporters, the energy stored in an electrochemicalgradient across the plasma membrane facilitatesligand transport.8 The molecular details of how thetransport is achieved are not yet fully elucidated,though studies on the SiaPQM TRAP-T indicate therequirement for the P component, an electricgradient, and the cotransport of at least two Na+

per ligand transported.8 The ligands for TRAP-Tsare typically acidic small molecules.9 A related, but

distinct, family of transporters that import tricar-boxylic acids has also been described.10

The genome of T. pallidum is predicted to encode asingle TRAP-T,1,5 apparently organized as anoperon. The genes are tp0957 (the P component)and tp0958 (putative fused Q andM components). Athird gene within this putative operon is tp0956,which belongs to the Cluster of Orthologous Groups5660,11 a family of putative integral membraneproteins. This protein is also flagged as a member ofthe RskA superfamily;11 in Mycobacterium tubercu-losis, this protein binds to σ factor K, repressing itsactivity.12 Alternatively, because it is similar in sizeto the P protein, it had been speculated that Tp0956might serve as a second SBP for this TRAP-T.5

Having a second SBP would likely have importantfunctional consequences, given the very limitedgenome of T. pallidum.In this study, we first set out to characterize this

putative second SBP, Tp0956. However, T. pallidumis not cultivatable in vitro, thereby precludinggenetic manipulation of the organism. Thus, wetook advantage of a structure-to-function approachfor this protein that has been successful for other T.pallidum proteins of unknown functions.13–16 First,reverse transcription and amplification of RNA fromrabbit-tissue-derived T. pallidum confirmed thattp0956, tp0957, and tp0958 constitute a singletranscriptional unit. We then determined the crystalstructure of Tp0956. Notably, we found that theprotein has no structural homology to known Pproteins, making it unlikely that it serves as anadditional SBP for this putative TRAP-T. Theprotein is a mostly α-helical irregular cylinder witha pore running through its center. It featuressubstructures similar to tetratricopeptide repeat(TPR) motifs, which are hallmarks of proteins thatare involved in protein–protein interactions.17,18

Because of the unusual nature of this TRAP operon,we also pursued and obtained a crystal structure forTp0957, the putative P component. The overall foldof Tp0957 is similar to P components from otherTRAP-Ts, but there are notable differences, partic-ularly in the ligand-binding clefts of the proteins;that of Tp0957 is more hydrophobic than otherstructurally characterized P proteins. Althoughthere is some evidence of a small-molecule ligandbound to the protein, Tp0957 adopts a conformationmore similar to that of an unliganded TRAP-T SBP.The existence of TPR motifs in Tp0956 led us to seekits binding partner(s). Using analytical ultracentri-fugation (AUC), we found that trimeric Tp0956interacts with Tp0957. This observation was con-firmed in vivo with formaldehyde cross-linking.Finally, we showed that proteins homologous toTp0956 are found in other TRAP-T operons. Thus,the T. pallidum TRAP-T may serve as a paradigm fora hitherto unknown class of TRAP-Ts [which we callTPATs (TPR-protein-associated TRAP-Ts); see

680 A TPR-associated TRAP-T (TPAT)

below] that apparently require a TPR-containingaccessory protein.

Results

Analyses of tp0956, tp0957, and tp0958

Genomic analysis of T. pallidum suggested thattp0956, tp0957, and tp0958 are oriented in the samedirection and thus may be cotranscribed1 (Fig. 1).Furthermore, bacterial genes belonging to a trans-porter often are cotranscribed, and bioinformaticanalyses predicted tp0957 and tp0958 as componentsof a TRAP-T. Indeed, the TRAP-T operon (SiaPQM)from Haemophilus influenzae is organized similarly19

(Fig. 1). Therefore, we performed reverse transcrip-tion PCR (RT-PCR) to examine the transcriptionallinkage of these genes. As shown in Fig. 1, the three-gene cluster from T. pallidum is transcriptionallylinked. One potential caveat is that the RT-PCRproduct observed between the tp0954 and tp0956pair (Fig. 1; Table S1) may be due to the presence ofnoncoding mRNA that is transcribed in the oppositedirection of this operon.

Fig. 1. Transcriptional linkage of the putative trepone-mal TRAP operon. (a) The genomic organizations of TRAPsystems fromH. influenzae (upper) and T. pallidum (lower).The genes for the P components are colored blue, those forthe Q components are yellow, and those for the Mcomponents are gray. These latter two components arefused in the examples shown, but not in all TRAP-Ts. (b)Evidence of transcriptional linkage of Tp0956 and Tp0957.RT-PCR was performed on T. pallidum RNA using primerpairs (small arrows) listed in Table S1, specific for eitherthe intergenic regions of listed gene pairs or the entiretp0955 gene (pseudogene). The lanes are as follows: M,DNA molecular weight markers; C, PCR reaction withindicated primer pair that served as positive control usingT. pallidum genomic DNA as template in place of cDNA;“−”, PCR reaction with indicated primer pair using RNAas template (lacking reverse transcription), which servedas a negative control for DNA contamination; “+”, RT-PCR products with indicated primer pairs.

In accord with its putative function as a permease,Tp0958 (a likely fusion of the M and Q components)is quite hydrophobic (average hydropathy=1.084)and is predicted to be an all-α-helical transmem-brane protein. Signal sequences at their N-terminiidentify both Tp0956 and Tp0957 as putativelipoproteins20 (see Table S2). As such, their N-terminal signal sequences would be removed,leaving a tri-acylated cysteine residue at theirrespective N-termini that would anchor the proteinsto the periplasmic side of either the inner membraneor the outer membrane of the bacterium.A CDD (Conserved Domain Database) search11

using the sequence of Tp0956 places it in the Clusterof Orthologous Groups 5660, a family of integralmembrane proteins. This fact raises the question: isthe Tp0956 an integral membrane protein or alipoprotein? We purified recombinant forms ofboth Tp0956 and Tp0957 almost identically (seeMaterials and Methods). In these proteins, the N-terminal cysteines had been removed. This expedientwas designed to prevent the acylation reactionmentioned above and should render the proteinswater soluble. Indeed, they could be purified withonly a low concentration (2 mM; well below thecritical micelle concentration) of the nonionic deter-gent n-octyl-β-D-glucoside (β-OG). We often includethis additive in preparations of recombinantlysolubilized treponemal lipoproteins,13,14,15,16 as itappears to improve their stability in solution.Additionally, β-OG can have a positive effect onprotein crystallization.21 The water solubility andprobable lipidation of both proteins suggest thatthey exist in the periplasm, not in the cell membrane,as implied for Tp0956. The CDD search also placedTp0956 in the RskA superfamily; as noted above,RskA is an anti-σ factor in M. tuberculosis. It seemsunlikely that Tp0956 could play a role in transcrip-tion, given its probable periplasmic localization.

The crystal structure of Tp0956

In view of the possibility that Tp0956 may be asecond SBP for a TRAP-T,we set out to determine thethree-dimensional structure of Tp0956 (Fig. 2; Table1). The crystal structure of Tp0956 was refined usingX-ray diffraction data to 2.3 Å resolution. The model(Fig. 2a and b) has 13 α-helices, 1 short 310 helix, andno β-strands. The structure is very different from thebilobed structures (see below) of known P compo-nents of TRAP-Ts, making it unlikely that Tp0956fulfills this proposed role. Residues 31–301 (in thenumbering of the mature, processed form of theprotein) are visible in electron-density maps; the first30 residues of this protein construct are presumedpresent but were disordered in the crystals. The α-helices, designated “A” through “M”, have anantiparallel arrangement that forms an irregularring (Fig. 2a–c). The ring surrounds a pore, which

Fig. 2. The crystal structure of Tp0956. (a) A topological diagram of Tp0956. The cylinders are α-helices and are labeledwith their respective alphabetical designations. The protein forms a ring with a pore running through it, but in thisdiagram, the ring has been opened and “flattened”, with the helices most proximal to the pore shown behind the othersand colored pink. All other helices are colored red. This color coding holds for (b) and (c) as well. (b) A “side view” of theprotein, with the C, lateral, and N faces denoted. (c) A surface representation of the protein, viewed from the C face. (d) Across-section through the protein, which has been colored according to its electrostatic potential. The scale for thecoloration is also shown. An area of negative potential on the lateral surface of the protein is circled.


perforates the entire protein (Fig. 2c). Hereafter, werefer to the “N”, “C”, and “lateral” faces of Tp0956,which are defined in Fig. 2b. The “N” and “C” facesare so called because they respectively house theobserved N-terminus and C-terminus of Tp0956.

Characteristics of the pore

Tp0956's pore runs from the N to the C faces of theprotein, a total of ∼43 Å (Fig. 3a). It has an innerportion (∼20 Å in length, average diameter of∼5.8 Å, and minimum diameter of 4.8 Å; see Fig.S1) and two vestibules (the “N” and “C” vestibules,depending on their respective proximity to the Nand C faces). Some soluble protein enzymes havepores that are related to their respective functions;the archaeal 20 S proteasome is a multimericassembly that has narrow (∼13 Å) annuli,23 andthe DNA topoisomerase of E. coli has a large (∼28 Å)

hole. 24 However, there is no evidence of anenzymatic activity for Tp0956. Although Tp0956 isa soluble protein (see above), the dimensions of itschannel invite comparisons to membrane-spanningproteins. For example, the minimum pore diametersof aquaporin (AQP1), glyceroporins (GlpF), and anion channel (KcsA) range from 2.6 to 3.7 Å,25–27 andthose of bacterial porin outer-membrane proteinscan be as small as 5 Å.28 The pore of Tp0956accommodates water molecules and a sulfate ion(from the precipitation buffer). There is a hydro-phobic antechamber (HA) in the N vestibule (Fig.3c); it measures approximately 10 Å×5 Å. Theidentities of the side chains lining the HA are shownin Table S3. Electrostatic calculations show (Fig. 2d)that the C vestibule is mostly negatively charged,that the N vestibule is mostly positive, and that thepore is mixed. No particular class of amino aciddominates the lining of the pore.

image of Fig. 2

Table 1. Data collection, phasing, and refinement statistics

Crystal

Tp0956, Sea peak Tp0956, native Tp0957, native

Space group C2 I23 P21Energy (eV) 12,687.7 12,682.7 12,675.0Resolution range (Å) 34.1–2.30 (2.34–2.30) 40.2–2.30 (2.34–2.30) 37.8–1.40 (1.42–1.40)Unique reflections 86,064 (3818)b 153,999 (769) 125,807 (6229)Multiplicity 3.0 (2.3) 19.2 (14.3) 4.1 (3.6)Data completeness (%) 91.8 (81.9) 99.9 (99.7) 99.8 (99.7)Rmerge (%)b,c 6.3 (46.4) 7.7 (89.9) 3.7 (56.6)I/σ(I) 25.1 (2.10) 42.7 (2.89) 33.7 (2.03)Wilson B-value (Å2) 44.6 52.4 18.0Anomalous scatterers Selenium, 24 out of 24 possible sites None NoneFigure of merit (90.5–2.30 Å) 0.13 N/A N/AFigure of merit post-DM (90.5–2.3 Å) 0.69 N/A N/A

Resolution range (Å) 29.9–2.30 (2.48–2.30) 37.8–1.40 (1.41–1.40)No. of reflections Rwork/Rfree 15,250/758 (2868/146) 125,752/6322 (3800/198)Data completeness (%) 99.5 (100.0) 99.7 (97.0)Atoms (non-hydrogen protein/solvent/ions) 2153/61/20 5634/431/0Rwork (%) 19.7 (29.9) 17.8 (30.0)Rfree (%) 24.2 (41.1) 20.1 (36.2)r.m.s.d. bond length (Å) 0.008 0.009r.m.s.d. bond angle (°) 1.0 1.2Mean B-value (Å2) (protein/solvent/ions) 49.2/48.9/108.0 21.9/29.1/38.5Ramachandran plot (%)

(favored/additional/disallowed)d98.9/1.1/0.0 99.0/1.0/0.0

Maximum-likelihood coordinate error 0.31 0.36Missing residues A: 1–30 A: 7–8, 65–66, 326–328

B: 7–8, 325–328a Bijvoet pairs were kept separate for data processing.b Data for the outermost shell are given in parentheses.c Rmerge=100∑h∑i|Ih,i− ⟨Ih⟩|/∑h∑iIh,i, where the outer sum (h) is over the unique reflections, and the inner sum (i) is over the set of

independent observations of each unique reflection.58d As defined by the validation suite MolProbity.22


Comparisons to other structures

DALI29 was used to find two structural homologsof Tp0956. They are DrR162B from Deinococcus

Fig. 3. Pore characteristics of Tp0956. (a) A ribbon represenpore. The N, middle, and C portions of the pore are colored bluthe pore than that in (a) is shown, and the HA is depicsemitransparently for clarity, and the α-helices that contribucalculated with CAVER 2.0.22

radiodurans [Protein Data Bank (PDB) ID: 3GW4, noassociated publication; Fig. 4a] and TTC0263 (PDBID: 2PL2) from Thermus thermophilus strain HB27.30

Their structural homologies (both r.m.s.d. values are

tation of the protein with a surface representation of thee, green, and pink, respectively. (b) HA. A different view ofted as a yellow surface. Secondary structure is shownte residues to the HA are labeled. The HA surface was

image of Fig. 3

Fig. 4. The cTPR motifs of Tp0956. (a) A superpositionof Tp0956 (red) and DrR162B (blue). (b) The positions ofthe cTPR motifs. The motifs are colored gold, while theremainder of the protein is red. (c) The contributions of thecTPR motifs to the pore of Tp0956. A surface representa-tion is shown, with the protein colored as in (b).

Fig. 5. The trimer of Tp0956. The cTPRs are colored asin Fig. 4b, but the remainders are colored differently todistinguish the monomers. The triangular symbolrepresents the crystallographic 3-fold axis that relatesthe monomers.


1.8 Å) apply to roughly half (148 and 149,respectively) of the Cα atoms in Tp0956. Neitherhomologous protein forms a pore nor do theyappear to be involved in a TRAP-T operon.

These structural comparisons reveal that Tp0956contains four helical hairpins that are structurallysimilar to “TPR” motifs. Both DrR162B andTTC0263 contain these motifs, which occur astandem, 34-amino-acid repeats in the proteins thatharbor them. The TPR-like portions of Tp0956encompass the α-helical pairs αC/αD, αG/αH,αI/αJ, and αK/αL (Fig. 4b); thus, these motifsform part of the pore and also part of the lateralsurface of the protein. To our knowledge, Tp0956 isthe only known protein having TPR-like motifs thatform part of a pore.TPR motifs exist in many proteins.17 Their folds

comprise an antiparallel α-helical hairpin containingtwo α-helices and a short turn that connects them.These motifs are most often found as tandem repeatsof three or more in proteins, where they stack toform a superhelical arrangement that features aconcave surface and a convex surface. The concavesurface of the TPR-like superhelix in Tp0956 formspart of the protein's pore (Fig. 4c). The superhelicalstack is often “capped” by a C-terminal α-helix withan unrelated sequence; helix αM serves this functionin Tp0956. The tandem arrays of TPR motifs mostlyfunction as protein-interaction domains17 in gluco-corticoid receptors31 and in the anaphase-promot-ing complex.32 To date, the transporter associationof Tp0956 (and others identified below) is uniqueamong TPR-motif-bearing proteins.The TPR motifs in Tp0956 are cryptic. That is, they

failed to be detected by a hidden-Markov algorithmdesigned to reveal their presence,33 and they do not

image of Fig. 4

image of Fig. 5


adhere to the canonical 34-residue length. Never-theless, the structural homology between the TPRsof DrR162B and Tp0956 is pronounced (Fig. 4a).Sequence analysis of Tp0956 in its TPR-like regionsreveals that the protein does exhibit TPR-likesequence signatures (Fig. S2), but the interhelicalregion is longer than usual in all cases except one.These cryptic TPRs of Tp0956 are termed “cTPRmotifs” below.

0 5 10 15 200

0.2

0.4

0.6

0.8

1

sedimentation coefficient (S)

norm

aliz

ed c

(s)

(arb

itrar

y un

its)

TDod.

M

(a) (b

0 5 10 150

0.2

0.4

0.6

0.8

1


norm

aliz

ed c

(s)

(arb

itrar

y un

its)

Tp0956Tp0957Mixture

(c) (d

Fig. 6. The solution behavior of Tp0956 and Tp0957. (a)distributions are shown; they have been normalized suchassignments of the oligomeric states are shown: monomer (M)concentrations used for this experiment. (b) The oligomerizatio(D) aremarked. Probable oligomers are alsomarked. (c) HydrodAgain, three normalized distributions are shown. Here, the diwould have a maximum value of 1. (d) The Lamm-equation fitThe concentration of Tp0956was held constant at 0.7 μM,whilethe 3.5-μMexperiment is shown. The x-axis of the graph is the daxis in the upper part is the absorbance. The y-axis in the loweabsorbance units. The filled circles are individual data points, asuch that early scans are shown in violet and that late scans aparameters (see Materials and Methods) was used to fit to all spoint and every third scan used in the analysis are shown. Ad

The quaternary structure of Tp0956

Although there is only one copy of Tp0956 in theasymmetric unit, applying crystallographic symme-try to the protein reveals the presence of multiplepotential oligomeric forms. There are two identicaldodecamers in the unit cell, and these compriseeight identical trimers (Fig. 5). Physicochemicalcharacterization of the interfaces in the crystal

0 5 10 150

1

2

3


c(s)

(fr

inge

s/S

)

M

D

Oligomers

)

)

radius (cm)

6.2 6.4 6.6 6.8 7.0

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.020.00

-0.02

sign

al (

AU

)re

sidu

als

The solution oligomerization state of Tp0956. Three c(s)that the maximum value in each distribution is 1. Our, trimer (T), and dodecamer (Dod.). The inset shows threen status of Tp0957. The peaks for monomer (M) and dimerynamic evidence for the interaction of Tp0956 and Tp0957.stributions were normalized such that the peak of interestto the SV data for wild-type Tp0956 and Tp0957 mixtures.that of Tp0957was 0.7, 1.4, 2.1, 3.5, 7, 10.5, and 14 μM.Onlyistance from the center of rotation in centimeters, and the y-r part is the residual between the data and the fits, also innd the lines are the fits to these data; both are color codedre shown in red. The fits are global, that is, a single set ofeven data sets. For the sake of clarity, only every third dataditional concentrations are shown in Fig. S6.

image of Fig. 6


structure using PISA34 demonstrated that the moststable oligomer present in the crystals is thecrystallographic trimer shown in Fig. 5. To assessthe oligomeric state of Tp0956 in solution, wesubjected it to sedimentation velocity (SV) AUC.As seen in Fig. 6a, there are two dominant forms ofthe protein under our solution conditions. Thevirtually concentration independent sedimentationcoefficients and the diffusional spreads of the clearlyseparate boundaries denote the presence of slowlyinterconverting species with apparent molar massesof 98,000 and 380,000 g/mol, within error consistentwith trimeric and dodecameric forms of Tp0956(theoretical values of 101,028 and 404,112 g/mol,respectively). There is evidence of monomericTp0956 (Fig. 6a), but only at a low concentration(≤1.6 μM). The dodecameric form only becomesprevalent at high concentrations (N20 μM). In thepresence of a low concentration (2 mM; well belowthe critical micelle concentration) of β-OG, thetrimeric form is predominant, and the tendency toform dodecamers is substantially reduced (Fig. S3).It therefore appears that the detergent disrupts therelatively weak interactions that drive dodecamerformation, but it cannot disrupt the formation of thetrimer at this concentration. We therefore concludethat the trimeric form of Tp0956 is the most relevantin solution. The cTPR motifs are arrayed on theoutside of this assembly (Fig. 5). Importantly, all ofthe membrane-anchoring N-termini (i.e., the “N”faces) are located on the same face of this trimer, that

Fig. 7. The topology, structure, and surface features of Tp09as cyan arrows, while α-helices are depicted as blue cylindershelices are shown to scale, but the connecting regions and 310 hwill be used throughout the text. Lobe I is outlined in yellowmarked with “N” and “C”, respectively. (b) A ribbon-style drand lobes are color coded as in (a). (c) The surface electrostatic fon the surface of Tp0957. A key to the coloring scale is shown

is, facing away from the viewer in Fig. 5. This datumconfirms the topological feasibility of this assembly.In the dodecamer (data not shown), the N faces allpoint to the center of the “ball” of Tp09056s, a veryunlikely arrangement for a membrane-tetheredoligomer.

The crystal structure of Tp0957

The existence of a cTPR-motif-containing proteinin the operon of this putative treponemal TRAP-Tsuggested that this transporter may act differentlythan other TRAP-Ts. Further, no crystallographical-ly characterized P components had a sequence morethan 25% identical with the putative treponemal P,Tp0957. To explore the impact that these differenceshave on the structure of the P component, wedetermined the crystal structure of Tp0957. X-raydiffraction data to 1.4 Åwere used for determinationand model refinement (Table 1). The protein isbilobed (Fig. 7a and b); here, we call the two lobes“Lobe I” (residues 9–136 and 226–263) and “Lobe II”(residues 137–225 and 264–320). The protein chaincrosses from one lobe to the other at three points, themost prominent being a 40-residue α-helix (α11) thatforms a backbone-like feature in the protein. Lobe Ihas a central, five-stranded β-sheet that is sur-rounded by helices. Four of the strands in the β-sheet are parallel. Like Lobe I, Lobe II also has acentral β-sheet in which only one strand is antipar-allel; however, this sheet has six strands. This second

57. (a) A topology diagram of Tp0957. β-Strands are shown, and 310 helices are shown as ovals. The β-strands and α-elices are not. All elements are labeled with the names that; Lobe II, in pink. The amino- and carboxyl-termini are

awing of the structure of Tp0957. The secondary structureeatures of Tp0957. The electrostatic potential is color coded. The orientation of Tp0957 is identical with that in (b).


β-sheet is also surrounded by helices. The proteinhas no outstanding surface electrostatic features(Fig. 7c). Between the lobes is an opening (or “cleft”)in the protein (Figs. 7c and 8a).There are two copies of Tp0957 in the asymmetric

unit of these crystals. These molecules have verysimilar structures; the r.m.s.d. of all comparable Cα

atoms is 0.5 Å. The contents of this asymmetric unitraise the question of the oligomeric status of Tp0957.Most characterized P proteins are monomeric, buttwo show evidence of oligomerization.36,37 We used

Fig. 8. Features of the Tp0957 cleft. (a) Cavities proximal toshown in green, and Cavity C is shown in red; the three cavitiprotein; the mottled appearance of the surface here demonstraopening. The secondary structure of Tp0957 is shown semitransthat drawn in Fig. 7. The surfaces were calculated using CAVERunliganded form of SiaP, in the analogous orientation to that shlobe moves down (right side), enveloping the ligand. (c) Superpshown as smoothed traces through their respective Cα atomUnassigned electron density in Tp0957's cleft. A 2mFo−DFc mapside chains are shown and labeled. The secondary structure of

SV experiments to establish Tp0957's oligomericstatus in solution, and we found that Tp0957 ismonomeric under our solution conditions (Fig. 6b).Further, a physicochemical analysis34 of the in-teractions of the monomers found none that wouldlead to the formation of oligomers.The fold of Tp0957 is typical of a P component of a

TRAP-T. This group of SBPs belongs to superfamily7 of the extracellular transport proteins.38 Thus far,the structures of P components suggest that theproteins use a “Venus flytrap”motion to capture the

the Tp0957 cleft. Cavity A is shown in cream, Cavity B ises are also labeled. All three fuse near to the surface of thetes that all three of the calculated paths contribute to thisparently for clarity. The orientation is slightly different from2.0.22 (b) Observed closure of SiaP.35 The left side shows theown for Tp0957 in Fig. 7b. Upon ligand binding, the upperosition of Tp0957 and SiaP (open form). Both structures ares; Tp0957 is colored blue, and SiaP is colored green. (d)contoured at the 1-σ level is shown for the density. NearbyTp0957 is shown semitransparently for clarity.

image of Fig. 8


ligand in a cleft between the two lobes.9 Thismechanism has been convincingly shown for SiaP,which is part of a TRAP-T for sialic acid.19 Thatprotein has been crystallized in both an “open”(unliganded) form and a “closed” (liganded)form.35,39 The open-to-closed transition may becharacterized35 as an ∼30° rigid-body rotation ofLobe I closer to Lobe II about three hinge points (Fig.8b). The structure of Tp0957 is much closer to the“open” form of SiaP (Fig. 8c); according to DaliLite,compared to Tp0957, 296 comparable Cα positions of“open” SiaP have an r.m.s.d. of 2.8 Å, while 297 Cα

atoms of “closed” SiaP have an r.m.s.d. of 3.6 Å.40

The unique cleft of Tp0957

As mentioned above, Tp0957 has a cleft betweenthe two lobes. This opening leads to three differentcavities (Figs. 7c and 8a). We call them Cavity A(19 Å deep by 5 Å wide), Cavity B (11 Å×7 Å), andCavity C (10 Å×4 Å). The residues whose sidechains line these cavities are shown in Table S3.They are predominantly hydrophobic. While Cavi-ties B and C have analogs in other P structures,Cavity A does not.In comparison to other P proteins, the ligand-

binding pocket of Tp0957 is more hydrophobic.Indeed, almost all of the residues making polarcontacts to ligands in other P proteins are hydro-phobic at the equivalent positions in Tp0957.Notably, an arginine known to contact a carboxylatemoiety in the ligands of structurally characterized Pproteins9,41,42 (cf. R147 of SiaP) is not present inTp0957 (it is an alanine), and no compensatorypositively charged residue is observed near to thisposition. The lack of polar side chains in this areaimplies that ligand recognition by Tp0957 is basedon shape complementarity alone; matching patternsof opposite charge or hydrogen bond donors/acceptors on the ligand and protein are not likelyto be binding determinants. This finding couldimply a lack of specificity for ligands.Further, the shapes of the clefts of “open” SiaP and

Tp0957 differ significantly. In its open form, the cleftof SiaP is deep and spans the protein (Fig. 8b, left),whereas the cleft of Tp0957 resembles a muchsmaller pocket leading to the three aforementionedcavities (Figs. 7c and 8a). In the comparison of thetwo clefts, that of Tp0957 appears to have beenpartially covered with protein residues from a loopbetween β6 and α7 (residues 161–164), formingCavity B. For Tp0957 to undergo the rigid-bodyrotation observed in SiaP, this loop would have to besignificantly rearranged. These data suggest that thecognate ligand of Tp0957 differs in its chemicalcharacter from the small organic acid ligands ofmost P proteins and that the ligand-capture mech-anism of Tp0957 may deviate from those of other Pand SBP proteins.

There is no clear evidence for a bound ligand inthe electron-density maps for Tp0957. There is asmall patch of density proximal to the indole ring ofW24 (Fig. 8d), as if the unidentified chemical moietywere stacked on the ring. This density does notclosely correspond to the structures of any buffer,salt, precipitant, or cryoprotectant present. It mayrepresent a small molecule that copurified with theprotein or a contaminant in the precipitation orcryoprotection solutions. Despite this ambiguity, theposition of the density feature roughly correspondsto that adopted by the ligands in other P-proteinstructures. Besides the side chain of W24, the featureis proximal to the side chains of residues I17, P19,A136, L163, and I228, making it likely that themoiety bound here is hydrophobic.

The interaction of Tp0956 and Tp0957

Given that cTPR motifs exist inTp0956, we set outto identify its binding partner. Our first attemptwas with Tp0957, as the transcription of the twoproteins is linked (Fig. 1). SV experiments werethus used to query a possible interaction betweenTp0956 and Tp0957 in solution. We studied amixture of the proteins in which the Tp0956 trimer(0.7 μM) was saturated with a 24-fold molar excessof Tp0957. As is evident in Fig. 6c, all of the Tp0956was in a complex that sedimented faster thanTp0956 alone. The 9.3-S complex has an estimatedmolar mass of 217,000 g/mol, consistent with asingle trimer of Tp0956 interacting with threemonomers of Tp0957, forming a 1:3 complex(theoretical molar mass=214,560 g/mol). A mixtureof Tp0956 and Tp0957 also formed a co-elutingcomplex in size-exclusion chromatography (SEC)experiments (data not shown). As a negativecontrol to demonstrate specificity, we also sedi-mented Tp0956 with the treponemal lipoproteinTp0655, the SBP of an ABC-type polyaminetransporter,16 and there was no evidence of aninteraction between the two (Fig. S4). Further,Tde1020, the Treponema denticola homolog ofTp0957, does not interact well with Tp0956,demonstrating species specificity in this interaction(Fig. S4).We then used SV to examine in more detail the

strength of the interaction between Tp0956 andTp0957, exploiting the exquisite sensitivity of thetime-average sedimentation velocity of the reactionboundary to the complex size and the fractionaldissociation of its components. We analyzed datafrom mixtures of Tp0956 and Tp0957 at a range ofconcentrations with a model directly fitted to the SVdata (Fig. 5d and Fig. S6) given the reaction kinetics,sedimentation coefficients, and three macroscopicequilibrium association constants associated withan A+B+B+B↔AB+B+B↔ABB+B↔ABBB reac-tion scheme (see Materials and Methods; here,


A≡Tp0956 trimer and B≡Tp0957). Because all ofthe binding sites on trimeric Tp0956 are identical,the model includes the Kd of the first bindingevent [Kd(1)] and cooperativity factors (α-factors)for adding the second (α2) and third (α3)molecules of B (an αN1 suggests positive coop-erativity and vice versa). Excellent global fits to theraw data were achieved (Fig. 6d and Fig. S6),demonstrating that the first binding event isstrong {Kd(1)=KA(1)

−1=60 nM [50, 80]; the num-bers in square brackets indicate the 68.3% errorinterval} and that a slight positive cooperativitymay be associated with the second event (α2=1.5[1, 3]). However, there is a small but statisticallysignificant (on a 99.3% confidence interval) nega-tive cooperativity (α3=0.34 [0.23, 0.51]) for bind-ing the third Tp0957 to the Tp0956 trimer.

In vivo evidence for interaction between nativeTp0956 and Tp0957

The strong, specific interaction between the tworecombinant proteins in vitro prompted us todemonstrate an association between native Tp0956and Tp0957 in vivo. We therefore used in vivoformaldehyde cross-linking of T. pallidum to pre-serve putative Tp0956–Tp0957 complexes. Formal-dehyde is a short cross-linker that covalently linksprimary amines within ∼2 Å of each other (reflect-ing protein–protein interactions as they occur invivo). The link is reversible and used to preserve invivo protein–protein complexes. In this study, wechemically cross-linked treponemes with 1% form-aldehyde. Lysed cells were resolved in duplicate ontwo identical SDS polyacrylamide gels, electro-blotted, and then individually probed with eitheran anti-Tp0956 or an anti-Tp0957 antibody (Fig. 9).As a negative control, immunodetection was alsoperformed using pre-immune serum collected from

2, formaldehyde cross-linked treponemes; and lane 3, boiled tryellow arrows represent the putative positions of the cross-liarrows denote the positions of monomeric Tp0956 and Tp0957standards (M; values are in kilodaltons) are indicated in the im

the same rats later immunized with antigens (datanot shown). In the presence of cross-linker, bothblots featured a strong band at a position consistentwith a 250-kDa protein or complex (lane 2). Themolecules in this band were recognized by bothTp0956 and Tp0957 antibodies, providing evidencefor the occurrence of an in vivo complex. The blotswere then stripped and re-probed with the negativecontrol antibody raised against Tp47 (the mostabundant periplasmic lipoprotein in T. pallidum).Nonspecific cross-linked intermediates of nativeTp47 were not observed with either native Tp0956or Tp0957 (data not shown). The broader profiles ofcross-linked species may be caused by inefficientdissociation of the complex in the presence of SDS.As shown in Fig. 9, the large complexes formedupon formaldehyde treatment were no longerdetected after boiling; only Tp0956 and Tp0957monomers were observed (lane 3). For a number ofreasons, including the unequal electrotransfer ofsmall versus large complex proteins, the signals onthe blot do not give an accurate representation of theamount of target proteins (especially in the case ofcross-linked complexes of ∼250 kDa). Nevertheless,this result strongly supports the conclusion that thelarge entities (Fig. 9, lane 2) are reversible complexesof Tp0956 and Tp0957. Whereas the majority ofTp0956 protein is found at the apparent molecularmass of ∼250 kDa, Tp0956-specific bands corre-sponding to molecular masses of 35, 75, and 120 kDaarose, indicating partial dissociation of nativecomplexes (Tp0956–Tp0957) into Tp0956 trimer,dimer, and monomer in the presence of SDS uponincubation at 65 °C. The smear observed in the cross-linked sample developed using anti-Tp0957 anti-body also warrants comment. Some of the broaderbands might correspond to Tp0957 cross-linked toas yet unknown physiological partner(s) or possiblyto its operonic partner, Tp0958 (however, we do not

Fig. 9. In vivo formaldehydecross-linkage of Tp0956 andTp0957. Samples were resolved intwo identical gels and then electro-transferred to two separate mem-branes. The membranes wereindividually probed with either ananti-Tp0956 (α-Tp0956; left panel)or an anti-Tp0957 (α-Tp0957; rightpanel) antibody and visualized byan enhanced chemiluminescent de-tection using an anti-goat IgG-HRP.Lane 1 shows T. pallidum notexposed to the formaldehyde; lane

eponemes to break the cross-links prior to SDS PAGE. Thenked hetero-hexameric complex, while the red and green, respectively. The positions of pre-stained molecular massage of the blot.

image of Fig. 9


possess Tp0958 antibodies to substantiate thissecond possibility). A control experiment in whichtreponemes were incubated under cross-linkingconditions without the addition of formaldehydewas conducted. In this control, only uncomplexedTp0956 and Tp0957 were observed (lane 1). Based onthese results, we contend that there is a bona fideinteraction between native Tp0956 and Tp0957 in vivo.

Tp0956 and Tp0957 homologs in othermicroorganisms

A position-specific iterated BLAST search43 ofgenomic databases for proteins similar to Tp0956returned only 38 proteins that may reliably beconsidered homologous (see Table 2). For compar-ison, there are 756 P-protein homologs currentlycollected in a database of known TRAP systemproteins.45 Homologous proteins are clearly presentin the genus Treponema but are also found in other

Table 2. Putative TPAT T components and their genomic neig

Organism Name ClassLocu

Tp0956-lik

T. pallidum Spirochaetes tpTreponema paraluiscuniculi Spirochaetes tpccT. denticola Spirochaetes tdeTreponema phagedenis Spirochaetes hmpref9Treponema vincentii Spirochaetes trevi00Treponema primitia ZAS-2 Spirochaetes trepVerrucomicrobiae bacterium DG1235 Verrucomicrobiae vdg12Treponema azotonutricum ZAS-9 Spirochaetes treaTreponema brennaborense DSM 12168 Spirochaetes trebAlcanivorax borkumensis SK2 γ-Proteobacteria aboSyntrophobacter fumaroxidans MPOB δ-Proteobacteria sfumSpirochaeta thermophila DSM 6578 Spirochaetes spithDRAnaeromyxobacter dehalogenans 2CP-C δ-Proteobacteria adehAlcanivorax sp. DG881 γ-Proteobacteria adg88Anaeromyxobacter sp. Fw109-5 δ-Proteobacteria anae1Anaeromyxobacter sp. K δ-Proteobacteria anaeA. dehalogenans 2CP-1 δ-Proteobacteria A2cpDesulfotalea psychrophila LSv54 δ-Proteobacteria dpSpirochaeta smaragdinae DSM 11293 Spirochaetes spirSideroxydans lithotrophicus ES-1 β-Proteobacteria slitDechloromonas aromatica RCB β-Proteobacteria daroDesulfuromonas acetoxidans DSM 684 δ-Proteobacteria dace

daceDesulfobacterium autotrophicum HRM2 δ-Proteobacteria hrm2γ-Proteobacterium HTCC5015 γ-Proteobacteria gp5Desulfatibacillum alkenivorans AK-01 δ-Proteobacteria dalk

dalkReinekea sp. MED297 γ-Proteobacteria med29Desulfococcus oleovorans Hxd3 δ-Proteobacteria dole

doleγ-Proteobacterium NOR5-3 γ-Proteobacteria nor5Methylomonas methanica MC09 γ-Proteobacteria metmThiobacillus denitrificans ATCC 25259 β-Proteobacteria tbdLeptothrix cholodnii SP-6 β-Proteobacteria lchoS. thermophila DSM 6192 Spirochaetes sthermB. marinus SJ δ-Proteobacteria bmsPlesiocystis pacifica SIR-1 δ-Proteobacteria ppsirLeptospira biflexa serovar Patoc

strain “Patoc 1 (Paris)”Spirochaetes lepb

a As defined by PSI-BLAST.43,44

spirochetes, as well as in β-, γ-, and δ-proteobac-teria. No eukaryotic homologs were found. Thetreponemes are the only human pathogens thatwere found to harbor Tp0956 homologs. OneTp0956-homolog-containing organism, Bacteriovoraxmarinus, parasitizes other bacteria.46,47 The remain-ing proteins were from diverse free-living oligotro-phic bacteria. Remarkably, roughly one-third of the35 species harboring Tp0956-like proteins appear tometabolize saturated or aromatic hydrocarbons.We examined the genome maps for all organisms

whose genome contains a Tp0956 homolog. Strik-ingly, 92% of the Tp0956 homologs appear to becomponents of TRAP-Ts (Table 2). Of those that donot, the Tp0956-like gene is directly adjacent to aputative P protein of a TRAP system 67% of thetime; that is, the Q and M components are missing,perhaps encoded elsewhere in the genome. Thesecorrelations strongly suggest that the Tp0956 ho-mologs are carrying out an essential function for

hbors

s tag ofe protein (T) E-valuea

Locus tag of nearbyDctP-like protein

Presence of DctMand/or DctQ

0956 3e-99 tp0957 Yesa_0956 3e-99 tpcca_0957 Yes1021 7e-92 tde1020 Yes554_2429 1e-87 hmpref9554_2430 Yes01_2396 1e-87 trevi0001_2397 Yesr_1682 9e-85 trepr_1683 Yes35_4783 2e-84 vdg1235_2234 Noz_0940 3e-83 treaz_0941 Yesr_1914 2e-80 trebr_1913 Yes_0687 2e-75 abo_0688 Yes_2673 3e-75 sfum_2672 YesAFT_1602 3e-74 spithDRAFT_1603 Yes_3980 3e-74 adeh_3979 Yes1_3113 4e-72 adg881_421 Yes09_0439 9e-70 anae109_0440 YesK_4091 4e-69 anaeK_4090 Yes1_4123 6e-69 a2cp1_4122 Yes0443 3e-68 dp0444 Yess_2251 1e-67 spirs_2250 Yes_0818 5e-66 slit_0817 Yes_3578 7e-65 daro_3577 Yes_1925 9e-63 dace_1924 Yes_2628 4e-33 N/A No_33310 2e-62 hrm2_33300 Yes015_96 2e-61 gp5015_80 Yes_2688 4e-59 N/A No_2187 2e-52 N/A Yes7_13632 3e-57 med297_13637 Yes_3154 2e-56 dole_3157 Yes_3077 8e-54 dole_3078, dole_3074 Yes3_2500 2e-56 nor53_2704 Yese_0487 2e-55 metme_0486 Yes_0465 2e-55 tbd_0466 Yes_3613 3e-53 lcho_3612 Yes_c20400 6e-50 stherm_c20410 Yes_0699 2e-44 bms_0700 Yes1_38751 1e-38 ppsir1_38741 Yesi_I1477 4e-32 lepbi_I1476 Yes


their respective TRAP-T. That function probablyinvolves binding to the P component. Indeed, apreliminary study shows that the Tp0956- andTp0957-like proteins (Tde1020 and Tde1021, respec-tively) from T. denticola interact (Fig. S5). Given thestructural and functional characterizations delineat-ed above, the Tp0956 homologs likely perform anauxiliary function in the transport of the targetligand. We name these cTPR-containing proteins the“T components”.Do the TRAP-Ts with T components have other

features that distinguish them from canonical TRAPsystems? To begin to answer this question, wecompared the sequences of 218 P proteins, includingall of those that neighbor T components. The resultingphylogenetic tree (Fig. 10 and Fig. S6) demonstrates

Fig. 10. Phylogenetic comparison of 218 P-proteinsequences. Only a portion of the entire phylogenetic tree isshown. The locus tag protein names are provided. The scale,given in substitutions per site, is shown at the bottom.

that the Ps associated with T-containing TRAP-Ts(PTs) form a distinct clade; that is, they are evolution-arily more related to each other than to other Ps (Fig.10). Further, as noted above, most P proteins have anarginine residue at the position equivalent to A159 ofTp0957.9 The positively charged guanidinium moietyof this residue is critical for the binding of thecarboxylate moiety of the small carboxylic acids thatare normally bound by P proteins. This arginine isfound in only 3 of the 36 PTs that we havecharacterized to date. The two most likely explana-tions for this phenomenon are either (1) the PTs bind tochemically different compounds or (2) they bind tochemically similar compounds in amanner that differsfrom most P proteins. Regardless, we conclude thatthe transport systems harboring a T component are adistinct subfamily of TRAP-Ts that we call the TPATs.We suggest that the operon take on the prefix “tat”,standing for “TPR-protein-associated transporter”.Thus, the treponemal operon would be calledtatTPQ/M.

Discussion

This investigation has uncovered a new subfamilyof TRAP-Ts, the TPATs (Fig. 10). The existence of astrong, specific interaction in vitro (Fig. 6c, seeabove) and in vivo (Fig. 9) between the T and PTcomponents suggests that the mechanism of thesetransporters apparently is assisted by this associa-tion. In ABC and TRAP-Ts, obviously, protein–protein interactions between the SBP and thetransmembrane component(s) are functionallyvital.3,8,48,49 In Halomonas elongata, the cytoplasmic,tetrameric, ATP-binding universal stress proteinTeaD of the teaABCD modulates the activity of anectoine TRAP-T (TeaA/TeaB/TeaC), presumablyby affecting translation of its components or theactivity of the permease.50 However, the associationof an SBP with another periplasmic protein isunprecedented.What purpose is served by the binding of the T

and PT components of the TPATs? The full answer tothis question is not known at this time. It is possiblethat T may have a role in delivering a ligand to PT orin receiving a ligand from PT. Interestingly, if theligand does interact with T, this association mayrepresent an additional layer of specificity for theTPAT; that is, ligands must have chemical charac-teristics that are compatible with association withboth T and PT components. Additionally, if Tinteracts with the ligand, it implies that the proteinmay be acting as a chaperone to escort the ligandthrough a periplasmic environment that is otherwisehostile to its stability or solubility. Recently, a similarprotein-mediated mechanism for the transport ofhydrophobic compounds was proposed for an ABC-type transporter.51

image of Fig. 10


What is the ligand for this TPAT system in T.pallidum? Three lines of evidence suggest that theligand is hydrophobic: (1) the existence of HA of theT component (Fig. 3b; Table S3), (2) the hydrophobicnature of the PT component's cleft and associatedcavities (Fig. 8a; Table S3), and (3) the preponder-ance of T-containing organisms that utilize hydro-phobic compounds (Table 2). While theseoligotrophs can metabolize essentially insolublepetroleum hydrocarbons, treponemes probably ac-quired or retained the TPATs during their evolutionto import required hydrophobic ligands that theyencounter in humans.This study represents only the beginning of the

characterization of the TPATs. It will be necessaryto discern the ligands of these transporters; suchresearch currently remains infeasible for T. palli-dum, owing to the organism's need to be cultivatedin rabbit testicles. The nature of the ligand mayexplain the need for the T component and mayalso provide new seminal clues concerning what“nutrients” are needed ultimately to achieve the invitro culture of T. pallidum. Also, knowledge onwhether the ligand interacts with and passesthrough the T component's pore is imperative fora full understanding of the system. Finally,additional biophysical, biochemical, and biologicalstudies of the interactions among the T, PT, Q, andM (or Q/M) components are warranted to bettercomprehend the mechanism(s) of this new sub-family of bacterial transporters. Further examina-tions of this system might be better accomplishedin a different bacterium, such as T. denticola, whichcan replicate in culture. Additionally, informationregarding ligand variability in the TPAT Pcomponents should be available from a morecomprehensive study of the TPAT proteins fromdiverse bacterial species (Table 2).

Materials and Methods

Protein expression and purification

To produce a non-lipidated, recombinant derivative ofTp0957 in E. coli, we PCR amplified the DNA fragmentencoding amino acid residues 7–328 (cloned without theposttranslationally modified N-terminal Cys plus fiveother hydrophobic residues) of Tp0957 using end-specificprimers. Note that the recombinant versions of bothTp0956 and Tp0957 are numbered in this report such thatthe processed, modified cysteine residue is residue 1. ThePCR product was subcloned into the pGEM-T Easy vector(Promega) and transformed into E. coli XL1-Blue cells(Stratagene). The plasmids isolated from colonies thattested positive by restriction digestion were verified byDNA sequencing. The inserted fragment was excised bydigestion with NotI and BamHI and then ligated into thecorresponding restriction sites of the expression vectorpIVEX2.4d vector (5 Prime). The resultant construct

encoded a fusion protein with a His6 tag at its N-terminus.Ligation mixtures were transformed into E. coli XL1-Bluecells, and plasmids isolated from colonies were verified byDNA sequencing. A verified plasmid was then cotrans-formed with pGroESL (TAKARA) into E. coli BL21 AI(Invitrogen) cells for soluble protein expression.E. coli BL21 AI cells were grown at 37 °C in LB medium

containing 0.1% (w/v) glucose, 100 μg/mL of ampicillin,and 30 μg/mL chloramphenicol until the cell densityreached anA600 of 0.5. The culturewas then induced for 3 hwith 0.2% (w/v) L-arabinose. Cells derived from 1 L ofculture were harvested by centrifugation and lysed atroom temperature with gentle rocking for 30 min using50 mL of B-PER II (Thermo Scientific). The resultingsuspension was centrifuged at 25,000 g for 15 min toremove cell debris. Tp0957 was isolated from the super-natant by affinity chromatography using Ni-NTAAgarose(Qiagen). The protein was then subjected to SEC using aHiLoad 16/60 Superdex 200 prep grade column (GEHealthcare) equilibrated with buffer A [20 mM Hepes,0.1MNaCl (pH 7.5), and 2mM β-OG]. Peak fractions wereanalyzed by SDS PAGE. Fractions containing purifiedrTp0957 were pooled and stored at 4 °C in buffer A. After2 weeks, mass spectrometry and N-terminal sequencingdetermined that the tag of rTp0957 was specificallycleaved, leaving two residual amino acid residues (Gly–Arg) at the N-terminus followed by residues 7–328 ofTp0957. The mixture was further purified by SEC usingbuffer A. This purified, cleaved Tp0957 was used forcrystallization.For the production of a non-lipidated, recombinant

derivative of Tp0956 in E. coli, a DNA fragment encodingresidues 2–302 (cloned without the posttranslationallymodified N-terminal Cys) was amplified by PCR fromtreponemal genomic DNA using primer pairs encoding thepredicted 5′- and 3′-termini. The PCR product wassubcloned into BsaI/XbaI digested pE-SUMOpro3-basedbacterial expression vector (LifeSensors) in-frame with anN-terminalHis-SUMO tag. The construct was confirmed byDNA sequencing. The recombinant protein was over-expressed in E. coli BL21(DE3) cells. The selenomethio-nine-substituted protein was overexpressed in the E. colimethionine auxotroph B834(DE3). Protein expression wasinduced using an overnight express autoinduction system 2(Novagen). Cells harvested from 1 L of induced culturewere lysed at room temperature for 30minusing 50mLofB-PER II (Thermo Scientific) with gentle rocking. Lysed cellswere centrifuged at 25,000 g for 15min to remove cell debris,and soluble recombinant Tp0956 was immobilized on a 4-mLNi2+ affinity column (Qiagen). The immobilized proteinwas then washed with 10 column volumes of 20 mM Tris–HCl, 20mMNaCl, and 20mMimidazole (pH8.5) (buffer B),followedby 10 columnvolumes of 20mMTris–HCl and 1MNaCl (pH 8.5) and 5 column volumes of buffer B. Theprotein was eluted with 3 column volumes of 20 mM Tris–HCl, 200 mM NaCl, and 200 mM imidazole (pH 8.5). Theeluted protein was then subjected to SEC using a HiLoad16/60 Superdex 200 prep grade column (GE Healthcare)equilibrated with buffer A. The SEC was performed at 4 °Cusing an FPLC system (GEHealthcare). The His-SUMO tagwas removed by the SUMO-specific protease 2 (Life-Sensors), and the digestion mixture was further purifiedby SEC as above. The selenomethionine variant of Tp0956was purified in the presence of 2 mM DTT.


Protein concentrations were determined spectrophoto-metrically from their extinction coefficients calculatedusing the ProtParam utility of ExPASy‡.

RNA isolation and RT-PCR

Treponemal RNA extraction and RT-PCRmethods weredescribed previously.13,14 The multi-gene operon wasexamined by RT-PCR using RNA isolated from T. pallidumthat had been extracted from rabbit tissue. Intergenicregions were amplified to verify that the genes arecotranscribed in one polycistronic mRNA. cDNA andcDNA negative control were used as templates. The latterensured that no chromosomal DNA was carried out overto cDNA preparation. The PCR amplification was carriedout with GoTaq DNA polymerase (Promega) with astandard protocol and the primers listed in Table S1.

Analytical ultracentrifugation

All AUC experiments were performed in an Optima XL-I centrifuge (Beckman Coulter, Fullerton, CA). Thesamples, which had been incubated overnight at 4 °C,were placed in centrifugation cells equipped with dual-sectored Epon centerpieces, and sapphire windows, whichwere inserted in an An50-Ti rotor. The samples werecentrifuged at speeds between 45,000 and 50,000 rpm at20 °C after they had equilibrated for several hours. Bothabsorbance and interference optics were used to acquirethe concentration profiles.The c(s) analyses were carried out using SEDFIT and

SEDPHAT.52 All sedimentation coefficients were con-verted to s20,w-values. These distributions typically wereregularized using the maximum entropy approach with aconfidence level of 0.7. Alternatively, to evaluate thebinding equilibria, we globally analyzed Tp0956–Tp0957mixture data53,54 by direct modeling with a set of Lammequations describing the coupled sedimentation/diffu-sion/reaction process for the three-site scheme withmacroscopic equilibrium constants:

KA 1ð Þ = AB½ �A½ � B½ �

KA 2ð Þ = ABB½ �AB½ � B½ �

KA 3ð Þ = ABBB½ �ABB½ � B½ �

where A≡Tp0956 trimer and B≡Tp0957 monomer.Because the binding sites for B on A are identical,statistical factors govern that KA(2)/KA(1)=1/3 andKA(3)/KA(1)=1/9 in the absence of cooperativity. Thus,the equilibrium was treated as a system of equations with

KA 2ð Þ = 13a2KA 1ð Þ and

KA 3ð Þ = 19a3KA 1ð Þ

‡http://us.expasy.org

In the nonlinear regression, in addition to KA(1), thefactors α2 and α3 were allowed to refine to account forcooperativity. The kinetic component of all of the reactionswas fixed such that the koff for reaction 1 was 10−3 s−1, andno kinetic cooperativity was taken into account. We usedHYDROPRO55 to estimate the sedimentation coefficients ofthe complexes using the appropriately edited complexcrystal structures (C.A.B., R.K.D., andM.V.N. (unpublisheddata)). Because the program overestimated this value forTp0956, we compensated by lowering the estimatedcomplex sedimentation coefficients by about 3%. Thesevalues were also fixed during the analyses. Thus, the onlyrefined quantities were the positions of the menisci, theexact concentrations of soluble Tp0956 and Tp0957 presentin the centrifugation cells, and the association constantsparameters log(KA(1)), log((1/3)·α2), and log((1/9)·α3).

In vivo formaldehyde cross-linking

Protein cross-linking was performed according to themethod of Skare et al.56 Approximately 1×108 freshlyharvested T. pallidum (prepared as described above forRNA extraction) in phosphate-buffered saline was incu-bated with 1% (v/v) methanol-free formaldehyde (Ther-mo Scientific) for 2 h at 25 °C with gentle rocking on anutating mixer. To stop the cross-linking reaction, weadded 1.25 M glycine to a final concentration of 125 mM(5 min at 25 °C). A control reaction was performedsimilarly, but without adding formaldehyde. Both cross-linked and uncross-linked treponemes were harvested bycentrifugation at 16,000 g and rinsed twice with ice-coldphosphate-buffered saline to remove excess unreactedcross-linkers, then frozen at −70 °C until needed foranalysis.

Antibodies, SDS PAGE, electroblotting,and immunodetection

Antisera-recognizing Tp0956 and Tp0957 wereobtained from rats by injecting purified recombinantantigens. Antibodies from the antisera were then affinitypurified using an antibody purification kit (ThermoScientific). The specificity of the purified antibodies wasdocumented against recombinant test and control anti-gens (data not shown).For SDS PAGE analysis, ∼1×108 treponemes were

lysed in 50 μL of 2× SDS Laemmli sample buffercontaining 500 mM β-mercaptoethanol. Cross-linkedsamples were either incubated at 65 °C for 5 min (tomaintain the cross-links) or boiled for 20 min (to break thechemical cross-links) prior to SDS PAGE analysis on a 4–15% (w/v) polyacrylamide TGX precast gel (Bio-Rad). Thecontrol sample was boiled for 20min. Identical numbers ofcells (∼5×107 per lane or 25 μL per lane) were loadedwithin each experiment. After gel electrophoresis, theproteins were electrotransferred onto a nitrocellulosemembrane and probed with antigen-specific antibodiesusing a standard Western blot protocol. The primaryantibodies were detected using anti-goat IgG-HRP anti-bodies (Jackson ImmunoResearch) and visualized by anenhanced chemiluminescent reaction system (ThermoScientific) and exposure to a Fuji luminescent imageanalyzer LAS-3000.

http://us.expasy.org

§www.phylosoft.org


Crystallization and X-ray diffraction data collection

All crystalswere obtained using the hanging-drop vapordiffusion method. Crystals of selenomethionine-substitut-ed Tp0956 were grown at 20 °C from solutions containing4 μL protein (40 mg/mL in buffer A containing 2 mMdithiothreitol) and 4 μL reservoir solution [30% (v/v)polyethylene glycol (PEG) 400, 0.1 M N,N-bis(2-hydro-xyethyl)glycine (pH 9.0), and 0.2 M MgCl2] that wereequilibrated over 0.5 mL of reservoir solution. Cryopro-tection was performed by transferring the crystals first to astabilization solution of 30% PEG 400, 0.1 M N,N-bis(2-hydroxyethyl)glycine (pH 9.0), 0.2 M MgCl2, and 0.1 MNaCl, then serially transferring them to similar solutionsof increasing [PEG 400]; the final [PEG 400] was 40%. Thecrystals were flash-cooled in liquid nitrogen. Nativecrystals of Tp0956 were obtained and treated similarly,except as noted below. A 4-μL drop of the protein solution(13 mg/mL in buffer A) was mixed with 4 μL of a differentreservoir solution [1.26 M ammonium sulfate, 0.1 M Ches(pH 9.5), and 0.2 M NaCl]. The stabilization solution was1.36M ammonium sulfate, 5% (v/v) ethylene glycol, 0.1 MChes (pH 9.0), and 0.3 M NaCl, and the final cryostabil-ization solution was 1.36 M ammonium sulfate, 15% (v/v)ethylene glycol, 0.1 M Ches (pH 9.0), and 0.3 M NaCl. AllX-ray diffraction data in this report (Table 1) wereobtained at 100 K at beamline 19-ID at the StructuralBiology Center at the Advanced Photon Source inArgonne National Laboratories. Tp0956 crystals exhibitedthe symmetry of space group I23 and contained onemolecule of Tp0956 per asymmetric unit. Data wereindexed, integrated, and scaled using the HKL-3000program package.57 Data from selenomethionine-substituted crystals were indexed in the lower symmetryspace group C2.Tp0957 was crystallized at 20 °C. Drops comprised 4 μL

protein solution (12mg/mL in buffer A) and 4 μL reservoirsolution [0.2 M KSCN and 20% (w/v) PEG 3350].Parallelepiped-shaped crystals appeared within 4 daysand varied widely in size. The crystal used for datacollection was approximately 0.1 mm×0.1 mm×0.25 mm.Crystals were stabilized by transferring them to a solutionconsisting of 0.1 M Hepes (pH 7.5), 0.2 M KSCN, 2 mM β-OG, 20% PEG 3350, and 5% (v/v) ethylene glycol.Stabilized crystals were serially transferred into similarsolutions that contained increasing concentrations ofethylene glycol; the final [ethylene glycol] was 25%. Afterabout 5 min in this final solution, the crystals were flash-cooled in liquid nitrogen.The crystals, which had the symmetry of space

group P21, were diffracted as noted above for Tp0956.The X-ray diffraction data were indexed, integrated, andscaled using the same program and procedures as usedfor Tp0956.

Phase determination, structure refinement, and analysis

Phases for Tp0956 were obtained from a single-wavelength anomalous diffraction experiment using sele-nomethionine-substituted protein (Table 1). SHELXD58

was used to locate the seleniumatoms. Phaseswere refinedwith MLPHARE59 and were further improved by densitymodification with DM.60 An initial model containing 82%of all Tp0956 residues was generated by alternating cycles

of Resolve61 and ARP/wARP.62 Additional residues weremanually modeled in O.63 Refinement (Table 1) wasperformed with the data collected on a native crystalusing PHENIX.64

The Tp0957 structure was determined using molecularreplacement. The search model was that of Tp0957 thathad been obtained from a co-crystal structure of theTp0956–Tp0957 complex (C.A.B., R.K.D., and M.V.N,unpublished results). Two molecules of Tp0957 werelocated in the asymmetric unit of the Tp0957 crystalsusing Phaser.65 The model was subjected to Cartesiansimulated-annealing, positional, and individual B-factorrefinement in PHENIX. At this point, missing parts of themolecules could be built using Coot.66 The modelrefinement was finalized using PHENIX; riding hydro-gens were added to the model in the final stages ofrefinement. Electrostatic calculations were performedwith APBS,67 using a protein dielectric constant of 1and monovalent ion concentrations of 0.15 M. Allstructure figures were made using the PyMOL version1.2 (Schrödinger, LLC). For both proteins, MolProbitywas used to verify the structures.68

Bioinformatics

The amino acid sequence of Tp0956 was subjected toPSI-BLAST.43,44 Five iterations were performed. Wediscarded “hits” with E-values above 1e-32 because suchproteins were deemed to be non-orthologous (other TPR-containing proteins). The sequences of 218 P proteins wereselected for comparison. The selection criteria were asfollows:

1. All P proteins with Tp0956-like genetic neighborswere included (n=36).

2. All P proteins with known crystal structures wereincluded (n=8).

3. The remaining proteins (n=174) were selected fromthe TRAP database.45 One protein was includedfrom each organism present. In cases where morethan one P protein was present in an organism, onewas chosen at random.

PROMALS3D69 was used to align the sequences; thestructures of the proteins from criterion 2 above wereused to provide structural constraints for the algorithm(the structure of Tp0957 was not included in this group).The unedited alignment was converted to PHYLIPformat, and the phylogenetic tree was calculated usingPhyML;70 the LG model of amino acid substitution71 wasemployed. The tree was viewed and output usingArchaeopteryx (C. M. Zmasek§).

Accession numbers

Coordinates and structure factors for the TatT and TatPTstructures have been deposited in the PDB; the accessionnumbers are 3U64 and 3U65, respectively.

http://www.phylosoft.org


Acknowledgements

We thank Dr. Zhiming Ouyang for technicalassistance with the RT-PCR analyses, the scientistsin the University of Texas Southwestern ProteinChemistry Core for protein sequence and massanalyses, and Dr. Lisa Kinch for helpful commentson the bioinformatics. This research was supportedby a National Institutes of Health grant (AI056305)and a Welch Foundation grant (I-0940) to M.V.N.This work was also supported in part by theIntramural Research Program of the NationalInstitute of Biomedical Imaging and Bioengineering(P.S.). Results shown in this report are derived fromwork performed at Argonne National Laboratory,Structural Biology Center at the Advanced PhotonSource. Argonne is operated by UChicago Argonne,LLC, for the U.S. Department of Energy, Office ofBiological and Environmental Research, undercontract DE-AC02-06CH11357.

Supplementary Data

Supplementary data to this article can be foundonline at doi:10.1016/j.jmb.2012.01.015

References

1. Fraser, C. M., Norris, S. J., Weinstock, G. M.,White, O.,Sutton, G. G., Dodson, R. et al. (1998). Completegenome sequence of Treponema pallidum, the syphilisspirochete. Science, 281, 375–388.

2. Norris, S. J. (1993). Polypeptides of Treponema palli-dum: progress towared understanding their structural,functional, and immunologic roles. Microbiol. Rev. 57,750–779.

3. Davidson, A. L. & Maloney, P. C. (2007). ABCtransporters: how small machines do a big job. TrendsMicrobiol. 15, 448–455.

4. Abramson, J. & Wright, E. M. (2009). Structure andfunction of Na+-symporters with inverted repeats.Curr. Opin. Struct. Biol. 19, 425–432.

5. Kelly, D. J. & Thomas, G. H. (2001). The tripartite ATP-independent periplasmic (TRAP) transporters ofbacteria and archaea. FEMS Microbiol. Rev. 25,405–424.

6. Forward, J. A., Behrendt, M. C., Wyborn, N. R., Cross,R. & Kelly, D. J. (1997). TRAP transporters: a newfamily of periplasmic solute transport systemsencoded by the dctPQM genes of Rhodobacter capsula-tus and by homologs in diverse gram-negativebacteria. J. Bacteriol. 179, 5482–5493.

7. Shaw, J. G., Hamblin, M. J. & Kelly, D. J. (1991).Purification, characterization and nucleotide sequenceof the periplasmic C4-dicarboxylate-binding protein(DctP) from Rhodobacter capsulatus. Mol. Microbiol. 5,3055–3062.

8. Mulligan, C., Geertsma, E. R., Severi, E., Kelly, D. J.,Poolman, B. & Thomas, G. H. (2009). The substrate-

binding protein imposes directionality on an electro-chemical sodium gradient-driven TRAP transporter.Proc. Natl Acad. Sci. USA, 106, 1778–1783.

9. Fischer, M., Zhang, Q. Y., Hubbard, R. E. & Thomas,G. H. (2010). Caught in a TRAP: substrate-bindingproteins in secondary transport. Trends Microbiol. 18,471–478.

10. Winnen, B., Hvorup, R. N. & Saier, M. H. (2003). Thetripartite tricarboxylate transporter (TTT) family. Res.Microbiol. 154, 457–465.

11. Marchler-Bauer, A., Lu, S., Anderson, J. B., Chitsaz, F.,Derbyshire, M. K., DeWeese-Scott, C. et al. (2011).CDD: a Conserved Domain Database for the func-tional annotation of proteins. Nucleic Acids Res. 39,D225–D229.

12. Saïd-Salim, B., Mostowy, S., Kristof, A. S. & Behr,M. A. (2006). Mutations in Mycobacterium tuberculosisRv0444c, the gene encoding anti-SigK, explain highlevel expression of MPB70 and MPB83 in Mycobacte-rium bovis. Mol. Microbiol. 62, 1251–1263.

13. Deka, R. K., Brautigam, C. A., Tomson, F. L.,Lumpkins, S. B., Tomchick, D. R., Machius, M. &Norgard, M. V. (2007). Crystal structure of the Tp34(TP0971) lipoprotein of Treponema pallidum: implica-tions of its metal-bound state and affinity for humanlactoferrin. J. Biol. Chem. 282, 5944–5958.

14. Deka, R. K., Brautigam, C. A., Yang, X. F., Blevins,J. S., Machius, M., Tomchick, D. R. & Norgard, M. V.(2006). The PnrA (Tp0319; TmpC) lipoprotein repre-sents a new family of bacterial purine nucleosidereceptor encoded within an ATP-binding cassette(ABC)-like operon in Treponema pallidum. J. Biol.Chem. 281, 8072–8081.

15. Deka, R. K., Machius, M., Norgard, M. V. & Tomchick,D. R. (2002). Crystal structure of the 47-kDa lipopro-tein of Treponema pallidum reveals a novel penicillin-binding protein. J. Biol. Chem. 277, 41857–41864.

16. Machius, M., Brautigam, C. A., Tomchick, D. R.,Ward, P., Otwinowski, Z., Blevins, J. S. et al. (2007).Structural and biochemical basis for polyaminebinding to the Tp0655 lipoprotein of Treponemapallidum: putative role for Tp0655 (TpPotD) as apolyamine receptor. J. Mol. Biol. 373, 681–694.

17. D'Andrea, L. D. & Regan, L. (2003). TPR proteins: theversatile helix. Trends Biochem. Sci. 28, 655–662.

18. Sampathkumar, P., Roach, C., Michels, P. A. M. &Hol, W. G. J. (2008). Structural insights into therecognition of peroxisomal targeting signal 1 byTrypanosoma brucei peroxin 5. J. Mol. Biol. 381,867–880.

19. Severi, E., Randle, G., Kivlin, P., Whitfield, K., Young,R., Moxon, R. et al. (2005). Sialic acid transport inHaemophilus influenzae is essential for lipopolysaccha-ride sialylation and serum resistance and is dependenton a novel tripartite ATP-independent periplasmictransporter. Mol. Microbiol. 58, 1173–1185.

20. Setubal, J. C., Reis, M., Matsunaga, J. & Haake, D. A.(2006). Lipoprotein computational prediction in spir-ochaetal genomes. Microbiology, 152, 113–121.

21. McPherson, A., Koszelak, S., Axelrod, H., Day, J.,Williams, R., Robinson, L. et al. (1986). An experimentregarding crystallization of soluble proteins in thepresence of β-octyl glucoside. J. Biol. Chem. 261,1969–1975.

http://dx.doi.org/10.1016/j.jmb.2012.01.015


22. Petřek, M., Otyepka, M., Banáš, P., Košinová, P., Koča,J. & Damborský, J. (2006). CAVER: a new tool toexplore routes from protein clefts, pockets andcavities. BMC Bioinformatics, 7, 316.

23. Löwe, J., Stock, D., Jap, B., Zwickl, P., Baumeister, W.& Huber, R. (1995). Crystal structure of the 20Sproteasome from the archaeon T. acidophilum at 3.4 Åresolution. Science, 268, 533–539.

24. Lima, C. D., Wang, J. C. & Mondragón, A. (1994).Three-dimensional structure of the 67K N-terminalfragment of E. coli DNA topoisomerase I. Nature, 367,138–146.

25. Tajkhorshid, E., Nollert, P., Jensen, M. O., Miercke,L. J. W., O'Connell, J., Stroud, R. M. & Schulten, K.(2002). Control of the selectivity of the aquaporinwater channel family by global orientational tuning.Science, 296, 525–530.

26. Sui, H., Han, B. G., Lee, J. K., Walian, P. & Jap, B. K.(2001). Structural basis of water-specific transportthrough the AQP1 water channel. Nature, 414,872–878.

27. Fu, D., Libson, A., Miercke, L. J. W., Weitzman, C.,Nollert, P., Krucinski, J. & Stroud, R. M. (2000).Structure of a glycerol-conducting channel and thebasis for its selectivity. Science, 290, 481–486.

28. Zeth, K., Diederichs, K., Welte, W. & Engelhardt, H.(2000). Crystal structure of Omp32, the anion-selectiveporin form Comamonas acidovorans, in complex with aperiplasmic peptide at 2.1 Å resolution. Structure, 8,981–992.

29. Holm, L. & Rosenstrōm, P. (2010). Dali server:conservation mapping in 3D. Nucleic Acids Res. 38,W545–W549.

30. Lim, H., Kim, K., Han, D., Oh, J. & Kim, Y. (2007).Crystal structure of TTC0263, a thermophilic TPRprotein from Thermus thermophilusHB27.Mol. Cell, 24,27–36.

31. Smith, D. F. (2004). Tetratricopeptide repeat cocha-perones in steroid receptor complexes. Cell StressChaperones, 9, 109–121.

32. Vodermaier, H. C., Gieffers, C., Maurer-Stroh, S.,Eisenhaber, F. & Peters, J. M. (2003). TPR subunitsof the anaphase-promoting complex mediate bind-ing to the activator protein CDH1. Curr. Biol. 13,1459–1468.

33. Biegert, A., Mayer, C., Remmert, M., Soding, J. &Lupas, A. N. (2006). The MPI Bioinformatics Toolkitfor protein sequence analysis. Nucleic Acids Res. 34,W335–W339.

34. Krissinel, E. & Henrick, K. (2007). Inference ofmacromolecular assemblies from crystalline state.J. Mol. Biol. 372, 774–797.

35. Müller, A., Severi, E., Mulligan, C., Watts, A. G., Kelly,D. J., Wilson, K. S. et al. (2006). Conservation ofstructure and mechanism in primary and secondarytransporters exemplified by SiaP, a sialic acid bindingvirulence factor from Haemophilus influenzae. J. Biol.Chem. 281, 22212–22222.

36. Cuneo, M. J., Changela, A., Miklos, A. E., Beese, L. S.,Krueger, J. K. & Hellinga, H. W. (2008). Structuralanalysis of a periplasmic binding protein in thetripartite ATP-independent transporter family revealsa tetrameric assembly that may have a role in ligandtransport. J. Biol. Chem. 283, 32812–32820.

37. Gonin, S., Arnoux, P., Pierru, B., Lavergne, J., Alonso,B., Sabaty, M. & Pignol, D. (2007). Crystal structures ofan extracytoplasmic solute receptor from a TRAPtransporter in its open and closed forms reveal a helix-swapped dimer requiring a cation for α-keto acidbinding. BMC Struct. Biol. 7, 11.

38. Tam, R. & Saier, M. H. (1993). Structural, functional,and evolutionary relationships among extracellularsolute-binding receptors of bacteria.Microbiol. Rev. 57,320–346.

39. Johnston, J. W., Coussens, N. P., Allen, S., Houtman,J. C. D., Turner, K. H., Zaleski, A. et al. (2008).Characterization of the N-acetyl-5-neuraminic acid-binding site of the extracytoplasmic solute receptor(SiaP) of nontypeable Haemophilus influenzae strain2019. J. Biol. Chem. 283, 855–865.

40. Holm, L. & Park, J. (2000). DaliLite workbench forprotein structure comparison. Bioinformatics, 16,566–567.

41. Akiyama, N., Takeda, K. & Miki, K. (2009). Crystalstructure of a periplasmic substrate-binding protein incomplex with calcium lactate. J. Mol. Biol. 392,559–565.

42. Kuhlmann, S. I., van Scheltinga, A. C. T., Bienert, R.,Kunte, H. J. & Ziegler, C. (2008). 1.55 Å structure of theectoine binding protein TeaA of the osmoregulatedTRAP-transporter TeaABC from Halomonas elongata.Biochemistry, 47, 9457–9485.

43. Altschul, S. F., Gish, W., Miller, W., Myers, E. W. &Lipman, D. J. (1990). Basic local alignment search tool.J. Mol. Biol. 215, 403.

44. Altschul, S. F., Madden, T. L., Schäffer, A. A., Zhang,J., Zhang, Z., Miller, W. & Lipman, D. J. (1997).Gapped BLAST and PSI-BLAST: a new generation ofprotein database search programs. Nucleic Acids Res.25, 3389–3402.

45. Mulligan, C., Kelly, D. J. & Thomas, G. H. (2007).Tripartite ATP-independent periplasmic trans-porters: application of a relational database forgenome-wide analysis of transporter gene frequen-cy and organization. J. Mol. Microbiol. Biotechnol. 12,218–226.

46. Stolp, H. & Starr, M. P. (1965). Bacteriolysis. Ann. Rev.Microbiol. 19, 79–104.

47. Baer, M. L., Ravel, J., Piñeiro, S. A., Guether-Borg, D. &Williams, H. N. (2004). Reclassification of salt-waterBdellovibrio sp. as Bacteriovorax marinus sp. nov. andBacteriovorax litoralis sp. nov. Int. J. Syst. Evol. Microbiol.54, 1011–1016.

48. Oldham, M. L., Khare, D., Quiocho, F. A., Davidson,A. L. & Chen, J. (2007). Crystal structure of a catalyticintermediate of the maltose transporter. Nature, 450,515–520.

49. Oldham, M. L. & Chen, J. (2011). Crystal structure ofthe maltose transporter in a pretranslocation interme-diate state. Science, 332, 1202–1205.

50. Schweikhard, E. S., Kuhlmann, S. I., Kunte, H. J.,Grammann, K. & Ziegler, C. M. (2010). Structure andfunction of the universal stress protein TeaD and itsrole in regulating the ectoine transporter TeaABC ofHalomonas elongata DSM 2581T. Biochemistry, 49,2194–2204.

51. Malinverni, J. C. & Silhavy, T. J. (2009). An ABCtransport system that maintains lipid asymmetry in


the Gram-negative outer membrane. Proc. Natl Acad.Sci. USA, 106, 8009–8014.

52. Schuck, P. (2000). Size distribution analysis ofmacromolecules by sedimentation velocity ultracen-trifugation and Lamm equation modeling. Biophys. J.78, 1606–1619.

53. Brautigam, C. A. (2011). Using Lamm-Equationmodeling of sedimentation velocity data to determinethe kinetic and thermodynamic properties of macro-molecular interactions. Methods, 54, 4–15.

54. Dam, J., Velikovsky, C. A., Mariuzza, R. A.,Urbanke, C. & Schuck, P. (2005). Sedimentationvelocity analysis of heterogeneous protein–proteininteractions: Lamm equation modeling and sedi-mentation coefficient distributions c(s). Biophys. J. 89,619–634.

55. de la Torre, J. G., Huertas, M. L. & Carrasco, B. (2000).Calculation of hydrodynamic properties of globularproteins from their atomic-level structures. Biophys. J.78, 719–730.

56. Skare, J. T., Ahmer, B. M. M., Seachord, C. L.,Darveau, R. P. & Postle, K. (1993). Energy transduc-tion between membranes: TonB, a cytoplasmic mem-brane protein, can be chemically cross-linked in vivo tothe outer membrane receptor FepA. J. Biol. Chem. 268,16302–16308.

57. Minor, W., Cymborowski, M., Otwinowski, Z. &Chruszcz, M. (2006). HKL-3000: the integration ofdata reduction and structure solution—from diffrac-tion images to an initial model in minutes. ActaCrystallogr., Sect. D: Biol. Crystallogr. 62, 859–866.

58. Schneider, T. R. & Sheldrick, G. M. (2002). Substruc-ture solution with SHELXD. Acta Crystallogr., Sect. D:Biol. Crystallogr. 58, 1772–1779.

59. Otwinowski, Z. (1991). Maximum likelihood refine-ment of heavy atom parameters. Proceedings of theCCP4 Daresbury Study Weekend, 80–86.

60. Cowtan, K. & Main, P. (1998). Miscellaneous algo-rithms for density modification. Acta Crystallogr., Sect.D: Biol. Crystallogr. 54, 487–493.

61. Terwilliger, T. C. (2003). Automated main-chainmodel building by template matching and iterative

fragment extension. Acta Crystallogr., Sect. D: Biol.Crystallogr. 59, 38–44.

62. Langer, G., Cohen, S. X., Lamzin, V. S. & Perrakis, A.(2008). Automated macromolecular model buildingfor X-ray crystallography usng ARP/wARP version 7.Nat. Protocols. 3, 1171–1179.

63. Jones, T. A., Zou, J. Y. & Cowan, S. W. (1991).Improved methods for building protein models inelectron density maps and the location of errors inthese models. Acta Crystallogr., Sect. A: Found. Crystal-logr. 47, 110–119.

64. Adams, P. D., Afonine, P. V., Bunkóczi, G., Chen, V.B., Davis, I. W., Echols, N. et al. (2010). PHENIX: acomprehensive Python-based system for macromo-lecular structure determination. Acta Crystallogr., Sect.D: Biol. Crystallogr. 66, 213–221.

65. Read, R. J. (2001). Pushing the boundaries ofmolecular replacement with maximum likelihood.Acta Crystallogr., Sect. D: Biol. Crystallogr. 57,1373–1382.

66. Emsley, P. & Cowtan, K. (2004). Coot: model-buildingtools for molecular graphics. Acta Crystallogr., Sect. D:Biol. Crystallogr. 60, 2126–2132.

67. Baker, N. A., Sept, D., Joseph, S., Holst, M. J. &McCammon, J. A. (2001). Electrostatics of nanosys-tems: application to microtubules and the ribosome.Proc. Natl Acad. Sci. USA, 98, 10037–10041.

68. Davis, I. W., Leaver-Fay, A., Chen, V. B., Block, J. N.,Kapral, G. J., Wang, X. et al. (2007). MolProbity: all-atom contacts and structure validation for proteinsand nucleic acids. Nucleic Acids Res. 35, W375–W383.

69. Pei, J., Kim, B. H. & Grishin, N. V. (2008).PROMALS3D: a tool for mulotiple sequence andstructure alignment. Nucleic Acids Res. 36, 2295–2300.

70. Guindon, S., Dufayard, J. F., Lefort, V., Anisimova,M., Hordijk, W. & Gascuel, O. (2010). New algorithmsand methods to estimate maximum-likelihood phy-logenies: assessing the performance of PhyML 3.0.Syst. Biol. 59, 307–321.

71. Le, S. Q. & Gascuel, O. (2008). An improved generalamino acid replacement matrix. Mol. Biol. Evol. 25,1307–1320.

Structural, Bioinformatic, and In Vivo Analyses of Two Treponema pallidum Lipoproteins Reveal a...

Documents

Transcript of Structural, Bioinformatic, and In Vivo Analyses of Two Treponema pallidum Lipoproteins Reveal a...