Advertisement
JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M412885200 on February 22, 2005

J. Biol. Chem., Vol. 280, Issue 17, 17339-17345, April 29, 2005
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
280/17/17339    most recent
M412885200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Otto, B. R.
Right arrow Articles by Tame, J. R. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Otto, B. R.
Right arrow Articles by Tame, J. R. H.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Crystal Structure of Hemoglobin Protease, a Heme Binding Autotransporter Protein from Pathogenic Escherichia coli*

Ben R. Otto{ddagger}, Robert Sijbrandi{ddagger}, Joen Luirink{ddagger}, Bauke Oudega{ddagger}, Jonathan G. Heddle§, Kenji Mizutani§, Sam-Yong Park§, and Jeremy R. H. Tame§

From the {ddagger}Department of Molecular Microbiology, Faculty of Earth and Life Sciences, Vrije Universiteit de Boelelaan 1085, 1081 HV Amsterdam, The Netherlands and the §Protein Design Laboratory, Yokohama City University, Suehiro-cho 1-7-29, Tsurumi, Yokohama 230-0045, Japan

Received for publication, November 15, 2004 , and in revised form, February 17, 2005.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The acquisition of iron is essential for the survival of pathogenic bacteria, which have consequently evolved a wide variety of uptake systems to extract iron and heme from host proteins such as hemoglobin. Hemoglobin protease (Hbp) was discovered as a factor involved in the symbiosis of pathogenic Escherichia coli and Bacteroides fragilis, which cause intra-abdominal abscesses. Released from E. coli, this serine protease autotransporter degrades hemoglobin and delivers heme to both bacterial species. The crystal structure of the complete passenger domain of Hbp (110 kDa) is presented, which is the first structure from this class of serine proteases and the largest parallel {beta}-helical structure yet solved.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Autotransporters (ATs)1 are virulence proteins produced by a variety of pathogenic bacteria, including several types of Escherichia coli associated with severe, even fatal, diarrhea and urinary tract infections. An increasing number of these proteins have been discovered over the last decade, and a large body of research is now devoted to these proteins and their different roles in various diseases. ATs have evolved a unique export mechanism, the type V pathway (1, 2). Initially the protein is synthesized with an N-terminal leader peptide directing the protein to the periplasm via the signal recognition particle (SRP) (3). Once across the inner membrane, the C-terminal domain forms a {beta}-barrel structure within the outer membrane, allowing the passenger domain to escape to the external medium, where it is released proteolytically from the cell (4–6). Recently the crystal structure of a translocator (membrane) domain of an AT has been solved (7). AT passenger domains are generally more than 100 kDa in size and vary widely in sequence and function but include a group of closely related serine proteases, the SPATE family (serine protease autotransporters of the Enterobacteriaceae). These proteins are found only in pathogenic bacteria, but their precise roles remain unknown (8, 9). The SPATE family includes EspC from enteropathogenic E. coli (10), EspP from enterohemorrhagic E. coli (11), plasmid-encoded toxin (Pet) from enteroaggregative E. coli (12), Sat from uropathogenic E. coli (13), Tsh and Vat from avian pathogenic E. coli (14, 15), and Pic and SepA from Shigella flexneri (16, 17). These AT proteins are associated with a number of serious, possibly life-threatening diseases affecting millions world-wide. Tsh and Vat are essential virulence factors in avian diseases that cost the poultry industry millions of dollars annually. SPATE proteins do not appear to have conserved or even related functions (8, 9). Vat has vacuolating activity, and Pet enters eukaryotic cells to disturb the cytoskeletal system. As essential proteins in pathogenesis, SPATE proteins are key drug targets and potential routes to vaccines. Knowledge of the molecular structure is a first step toward understanding their mechanisms and hopefully how they may be inactivated. We have solved the crystal structure of the passenger domain of heme binding hemoglobin protease (Hbp), an AT involved in bacterial synergy in the development of peritonitis.

Peritonitis is an inflammation of the peritoneal cavity and may develop into a life-threatening condition through the formation of persistent abscesses, which leak pathogenic bacteria into the general circulation (18, 19). This common complication of invasive surgery then results in septic shock and multiorgan failure with mortality rates well in excess of 50%. These abscesses are difficult to treat with antimicrobial therapy, and usually require surgical intervention. It is frequently found that pathogenic E. coli and the strictly anaerobic Bacteroides fragilis occur together in intra-abdominal abscesses, the combination surviving better than either species alone (20, 21). B. fragilis is a common member of the normal human gut flora, but it is also a significant opportunistic pathogen (18, 19). It promotes fibrin deposition by the host, which localizes the bacteria and prevents systemic spread, but also provides an environment where the bacteria can escape attack by the immune system. In return, E. coli helps to supply both species with heme scavenged from hemoglobin (Hb). Hbp was discovered as the protease secreted by E. coli, which degrades Hb and binds heme (21, 22). The hbp gene is located on a large episome, pColV-K30, which also encodes an aerobactin-dependent iron uptake system (23). Hbp is highly expressed by E. coli during peritonitis and plays a key role in bacterial synergy with B. fragilis. Adding Hbp and heme to cultures of B. fragilis was shown to stimulate growth, and mice immunized with Hbp showed no abscess formation when subsequently challenged with mixed E. coli/B. fragilis cultures. Hbp therefore appears to be a useful target in the treatment and prevention of peritonitis. We have determined the crystal structure of Hbp with the longer term view of producing inhibitors, vaccines, or other treatments for peritonitis and other diseases caused by SPATE proteins. From this first molecular model of an AT, we have constructed homology models of other ATs to observe both common and unique features of these proteins.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Protein Purification and Crystallization—The cloning of Hbp has been described previously (22). E. coli strain DH5{alpha} harboring plasmid pACY-CHbp was grown to an A600 of 0.6 in MA medium, and the supernatant was collected and concentrated. Following 1000-fold concentration by ultrafiltration, the protein was purified by gel filtration in 10 mM Tris-Cl, pH 8.0, 300 mM sodium chloride. Apo-Hbp was crystallized by the published method (24). 4 mg/ml protein was crystallized using the hanging drop method using 0.1 M sodium acetate, pH 4.6, 200 mM ammonium sulfate, 10% glycerol, and 15–20% polyethylene glycol 6000.

Data Collection and Processing—Multi-wavelength x-ray data were collected at beam line 45XU-PX, SPring8, Harima, Japan. Native data were collected at beam line BL5 of the Photon Factory, Tsukuba, to a resolution of 2.2 Å. All data were indexed and integrated using HKL2000 and scaled with the program SCALEPACK (25). The crystals are in space group P6122 with unit cell dimensions of a = b = 114.86 Å, c = 437.05 Å, with one molecule per asymmetric unit and a solvent content of 65%.

Structure Determination and Refinement—The structure of Hbp was solved by MAD (multiple-wavelength anomalous dispersion) using selenomethionine-containing protein. Phase determination and refinement were performed using the programs SOLVE and RESOLVE (26, 27). 12 Se sites were located out of 14 expected. The 3.0-Å resolution electron density map obtained from RESOLVE was of sufficient quality to permit unambiguous chain tracing without further refinement of the heavy atom sites. After model building using the programs O (28) and Turbo (29), initial refinement was carried out with XPLOR (30) and final refinement with REFMAC (31, 32) using the 2.2-Å resolution native data. Solvent molecules were placed at positions where spherical electron-density peaks were found above 1.5 {sigma} in the 2Fo – Fc map and above 3.0 {sigma} in the Fo – Fc map and where stereochemically reasonable hydrogen bonds were allowed. The final model extends from the N terminus (residue 1 of the mature protein) to residue 1048, the C-terminal residue of the passenger domain. The Ramachandran plot has 86% of the residues in the most favorable regions and none in disallowed regions. A summary of the data collection and refinement statistics is given in Table I. The final model and structure factors have been deposited in the Protein Data Bank (PDB, 1WXR [PDB] ).


View this table:
[in this window]
[in a new window]
  		TABLE I
Data collection and refinement statistics

 

   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
The overall structure of the protein is shown in Fig. 1. The most striking feature is the right-handed {beta}-helical C-terminal tail, which extends from residue 260 onwards, forming three parallel {beta} sheets. This is highly similar to pertactin (PDB 1DAB [PDB] ) (33) and the N terminus of filamentous hemagglutinin (PDB 1RWR [PDB] ) (34), both adhesins from Bordetella pertussis. Pertactin is the only AT structure refined to date, one of only a dozen or so right-handed parallel {beta} helix protein structures so far solved. The first was the pectate lyase PelC (PDB 1AIR [PDB] ), and a number of different pectate lyase structures are now known (35). The pectate lyase superfamily members are a functionally eclectic mix, including P22 tailspike protein (PDB 1TYU [PDB] ), chondroitinase B (PDB 1DGB [PDB] ), and insect cysteine-rich antifreeze protein (PDB 1EZG [PDB] ). All share a similar pattern of repeats of about 25 amino acid residues, which form one turn of the right-handed helix. Each turn creates three {beta} strands that form extended {beta} sheets with neighboring turns. The core of this region is generally but not exclusively hydrophobic. The leucine-rich repeat family is another group of protein structures formed from repeated sequence motifs, but these include an {alpha} helix in each turn and are found to be highly curved into horseshoe shapes. The pectate lyase superfamily forms a much straighter superhelix. The shape of Hbp was previously determined to low resolution using a combination of small-angle x-ray scattering and analytical ultracentrifugation (36). The crystal structure shows a slightly straighter structure than the model from small-angle x-ray scattering, possibly due to greater flexibility in solution.



View larger version (42K):
[in this window]
[in a new window]
  		FIG. 1.
A stereo ribbon diagram showing the overall structure of the entire passenger region of Hbp. Helices and strands are colored red for the protease domain (domain 1, residues 1–256) and green for the chitinase b-like domain (domain 2, residues 481–556). The {beta} strands in the parallel {beta} helix are colored dark blue. Residues 950–1048 (the C terminus of the mature protein) are colored light blue. This region is required for transport of the passenger domain through the membrane pore created by the C-terminal region of the complete protein.

 
The Hbp model includes 24 turns of right-handed {beta} helix, compared with 16 in pertactin, the longest parallel {beta}-helix known to date. Hbp is highly reminiscent of pertactin, but less regular and slightly kinked in the center. It also has a large N-terminal globular domain and a smaller domain of 75 residues (481–556) about half-way along the {beta}-helix, as well as decorative loops where the polypeptide chain departs from the helical stem. The extra surface features make the length of Hbp (1048 residues) almost twice that of pertactin (539 residues). The extended hydrophobic core is largely but not exclusively filled with aliphatic side chains (Fig. 2, top and bottom). The N-terminal region close to the serine protease domain has a higher proportion of buried aromatic residues. Very few polar residues point into the hydrophobic core, but Ser780 is one exception (Fig. 2, bottom). Other SPATE proteins generally have threonine at this position. Unlike the pectate lyases there are no regular "stacks" of identical residues packed in layers within the core, but a row of asparagine residues is found forming an extended hydrogen bonding chain on the surface of the {beta} helix (Fig. 3). Overall the electron density for the model is very clear, but model building became increasingly difficult toward the C terminus. LAFIRE, a model fitting and rebuilding program using local correlation of electron density,2 was essential in assembling the last 100 residues of the model. Details of the program may be found at: altair.sci.hokudai.ac.jp/g6/Research/Lafire_English.html. There are several breaks in the final electron density maps in this region where short loops cannot be modeled. The C-terminal region from residue 950 to the end is a conserved domain found among AT proteins believed to assist folding after passage through the C-terminal pore domain (37).



View larger version (74K):
[in this window]
[in a new window]
  		FIG. 2.
Top, a slice through the {beta} helix showing the hydrophobic core. The backbone is shown as a green ribbon, and hydrophobic side chains are shown in pink. Bottom, the 2mFo – DFc electron density map, shown in stereo and contoured at 1.3{sigma}, showing a cross-section through the {beta} helix.

 



View larger version (93K):
[in this window]
[in a new window]
  		FIG. 3.
An orthogonal view to Fig. 2 (bottom) of the final 2mFo – DFc electron density map, at the same contour level, showing a stack of asparagine residues on the surface of the {beta} helix (colored red), which are hydrogen bonding through their side chains. These asparagines are not conserved among AT sequences, which have a variety of hydrophilic side chains at equivalent positions.

 

Residues 1–256 of the model form a clearly separate domain (domain 1 in Fig. 1), including six {beta} strands rolled into a barrel-like structure, with several long {beta} hairpins over its surface. The N-terminal glycine residue is deep within the structure, close to the bottom of the barrel. Folding of the domain can clearly only occur after the signal peptide has been removed. The glycine N atom hydrogen bonds to Asp206, which also interacts with Arg140 and the main-chain of Gly144. An {alpha}-helix (Leu245-Asp255) connects this domain to the {beta} helical stem. Searching the PDB for structural homologues with DALI (38, 39) revealed that the N-terminal domain is highly similar to bovine trypsin (PDB 5PTP [PDB] ), which gave a Z-score of 17.5 (scores above 2.0 are considered significant). The 162 residues matched showed 20% sequence identity to Hbp and could be fitted with root mean square deviation of 2.5 Å over the main chain. The C{alpha} traces are overlaid in Fig. 4, top. Other proteases gave rather poorer scores. Overlaying trypsin on the model of Hbp gave an excellent match of the catalytic triad (Ser195, His57, and Asp102) of trypsin with Ser207, His73, and Asp101 of Hbp (Fig. 4, bottom). Comparing the active sites, Hbp misses a loop (Ser93 to Asp101 in trypsin) just prior to the essential aspartate. In trypsin this loop includes Leu99, which contacts Trp215, one of the residues that forms antiparallel-{beta} sheet interactions with the substrate polypeptide. Trp215 in trypsin is changed to alanine (Ala229) in Hbp, making the active site of Hbp much more open. At the bottom of the specificity pocket, which accommodates the residue preceding the scissile bond, trypsin has an aspartate (Asp189), which leads to the preference for Lys and Arg residues at the S1 position. Chymotrypsin and elastase have serine in this position, but Hbp has tyrosine (Tyr201). Among the trypsin/chymotrypsin family, Pro225 is very highly conserved from bacteria to man, but Hbp misses a loop just before this residue, and the nearest equivalent is Gly237. Hbp is also unusual in having an asparagine residue at the position equivalent to glycine 226 in trypsin. Thus, although the fold of the N-terminal domain is close to trypsin, it also has unique features not seen in other serine proteases. The active site suggests broad substrate specificity with a preference for hydrophobic side chains in the specificity pocket. The very open active site presumably helps Hbp to attack folded globular proteins such as Hb, but more work is required to determine its functional properties as a protease. Low resolution electron density maps calculated using data collected with crystals soaked with two serine protease inhibitors, tosylphenylalanine chloromethyl ketone and tosyllysine chloromethyl ketone, show peaks of density close to the active site serine, suggesting the protease is still active at pH 4.6 (data not shown).



View larger version (76K):
[in this window]
[in a new window]
  		FIG. 4.
Top, a stereo overlay of the C{alpha} trace of bovine trypsin (blue) and the serine protease domain of Hbp (red). The C{alpha} of the catalytic serine (Ser207) is shown as a green sphere. Trypsin (PDB 5PTP [PDB] ) was matched to HBP by least-squares fitting residues 16–19 (1–4), 21–24 (15–18), 26–32 (49–55), 42–60 (58–76), 63–69 (78–84), 79–93 (85–99), 101–114 (100–113), 118–127 (114–123), 129–132 (128–131), 134–146 (136–148), 147–152 (156–161), 153–165 (163–175), 176–183 (177–184), 184–187 (186–189), 188–202 (200–214), 224–231 (236–243), and 234–245 (244–255). These 648 main-chain atoms selected by DALI have a root mean square deviation of 2.5 Å, and a maximum deviation of 8.2 Å. Bottom, the 2mFo – DFc electron density map contoured at 1.3 {sigma} over the catalytic site of the serine protease domain, showing the catalytic serine (Ser207) and a water molecule hydrogen bonded to it. Tyr201, the residue in the specificity pocket, is seen on the lower left.

 
Sequence alignments of SPATE family members shows the sequence GDSGSPL to be highly conserved between Gly205 and Leu211 (Fig. 5). vat, an AT gene found in a pathogenicity island of avian pathogenic E. coli Ec222, has a slightly altered sequence, ATSGSPL, which may be related to its vacuolating activity (15), but it maintains both the active site serine (Ser207) and Ser209, which hydrogen bonds to the carbonyl oxygen of Asp206 to make the turn carrying Ser207. The N-terminal Gly1 lies close to the active site serine and hydrogen bonds to the side chain of Asp206. This strongly suggests the enzyme can only become active after the leader peptide is removed. All SPATE proteins tested show proteolytic activity, but with different substrate specificities, suggesting different functions (9). Replacement of the active site serine of Pet (a SPATE from enteroaggregative E. coli) with isoleucine did not prevent release of the passenger domain from the C-terminal pore, but did eliminate all cytotoxic effects (40). Comparing the sequences of various ATs shows that Tyr201 in the specificity pocket of Hbp is preserved in Pic and SepA, whereas Pet and Sat have leucine (Fig. 5). The sequence alignment also suggests that EspC has glutamate at this position, which presumably would affect substrate preference strongly. Using the N domain structure of Hbp, a homology model of Pet protease domain was built using SwissModel (41). From this a second homology model of Sat protease domain was also constructed. The weaker sequence conservation between Hbp and Sat prevented a model of the latter being built more directly. These homology models show no unexpectedly loose packing or unfavorable geometry, indicating that each protein has identical connectivity, with minor differences in loop regions, in the N-terminal protease domain.



View larger version (71K):
[in this window]
[in a new window]
  		FIG. 5.
A sequence alignment of Hbp with other members of the SPATE family, covering the N-terminal protease domain. Conservative mutations are shown in red, and residues common to all of the six compared sequences are shown in white on red. The UNIPROT reference numbers for the sequences are O88093 [GenBank] (Hbp), Q8CWC7 (Pic), Q8VSL2 (SepA), P77070 (EspC), Q6KD43 (Sat), and O68900 [GenBank] (Pet). Secondary structure elements of Hbp are shown over the alignment. The conserved regions show strong correlation to these structural elements, with the exception of the catalytic region GDSGS. The side-chain four residues upstream from this sequence sits in the specificity pocket (Tyr201 in the case of Hbp).

 
The N-terminal domain makes extensive contact with the {beta}-helical stem of Hbp, but given its globular shape the contact region with the roughly cylindrical {beta} helix is long and narrow. Many N domain contacts are found between Asp108 and Val114. Arg106 makes buried hydrogen bonds to the carbonyl oxygen atoms of Asn301 and Gly303. The main hydrophobic contact is provided by Leu110, which is buried at the interface. Arg106 and Lys109 are very strongly conserved among ATs, the latter forming a salt bridge with Asp367. Arg441 and Asp49, both well conserved, make another bond between the N-terminal protease domain and the {beta}-helix. These contacts, and the preservation of key glycines in the sequences of ATs, suggest that these proteins share a very similar structure over the first half of the protein.

Residues 481–556 of Hbp make a small separate domain (domain 2 in Fig. 1), the main chain returning to the {beta} stem close to the point of departure. DALI found only one notable match, to the chitin binding domain of chitinase b (PDB 1E15 [PDB] ). This domain consists of only 49 residues folded into three antiparallel {beta} strands connected by long loops. The chitin binding domain fitted the 481–556 loop of Hbp with a Z-score of 3.6 over 44 residues, with 16% sequence identity. Not surprisingly, the surface aromatic groups of chitin binding domain presumed to bind chitin are altered in Hbp, but the 481–556 loop has a different cluster of aromatic residues, which appear to form a binding pocket of some sort. Tyr510 and Tyr526 point into solvent and are highly exposed (and make crystal contacts with a neighboring molecule). The nearby Trp492, Tyr506, and Phe527 make a hydrophobic pocket open to solvent. The 481–556 domain is quite different to any known heme-binding site, and a mutant in which this domain is deleted shows identical heme binding to the wild-type protein (Fig. 6). Possibly its function is docking to extracellular matrix proteins or the bacterial receptor proteins, but these have yet to be identified. Using residues 250–615 from the Hbp model, a homology model of the same region could be built (again using SwissModel) for Pic. Key residues such as Glu538 and Arg554, which hold domain 2 against the {beta}-helix by a salt bridge, are preserved in Pic. Sequence alignment of ATs shows that Pet, Sat, and EspC among others do not have this domain.



View larger version (88K):
[in this window]
[in a new window]
  		FIG. 6.
Heme binding by Hbp. PCR was used to create an Hbp mutant (called "Gdel-2") missing domain 2 by linking Ala481 to Asn556 with a glycine residue. Purified native Hbp and Hbp(Gdel-2) (20 pmol) were incubated with different concentrations of heme and subsequently subjected to nondenaturing gel electrophoresis as described previously (22). Heme containing proteins were visualized by tetramethyl benzidine staining. The upper panel shows the tetramethyl benzidine-stained gel; lanes 1–3, Hbp incubated with 0, 50, or 100 pmol of heme, respectively. Heme binding, shown by the dark band, is unchanged in the mutant. Lanes 4–6, Hbp(Gdel-2) incubated with the different amounts of heme. The lower panel shows the Coomassie Blue-stained gel of Hbp and Hbp(Gdel-2). The lower molecular weight of the mutant is apparent.

 
Residues 608–644 of Hbp form a long discursive loop projecting from the {beta}-helix, which also forms close contacts with the N domain. Thr616-Ala626 forms an {alpha} helix, but this probably requires hydrophobic contact with the N domain residues Tyr9 and Met158 to form. The 608–644 loop has no hydrophobic core of its own to suggest it is a stable independent fold. Because other ATs show no homology to Hbp in this region, it may be the heme binding site. Residue 629 is histidine, which is the most common residue bonding to heme groups, but the structure does not readily suggest how heme might bind. Repeated attempts to co-crystallize Hbp with heme or soak the crystals were unsuccessful, and mutagenesis is now underway to test the function of this region. The charge distribution over the molecular surface shows a concentration of negative charge on one face of the molecule, centered around the interface between the {beta} helix and domain 1 (Fig. 7). Further mutagenesis is required to determine whether this negatively charged patch is involved in heme transport.



View larger version (74K):
[in this window]
[in a new window]
  		FIG. 7.
Surface charge of Hbp. The molecular surface of the protein, colored by electrostatic charge (red, negative; blue, positive). Color saturation corresponds to an electron energy of ±20 kBT.

 
Sequence similarity among ATs shows an increase again toward the C-terminal region of the passenger proteins. Gly-cines 868 and 875 in turns of Hbp are completely preserved, as are Met886 and Trp891. The {beta}-helix structure is broken at the conserved Ala979/Pro980, which mark a change in the direction of the polypeptide chain. Breaks in the electron density prevented modeling residues 913–918, 937–941, and 1024–1031. Despite the sequence conservation over the last 140 residues, these gaps in the Hbp model prevented making homology models of this region. The pattern of preserved residues matches very well interactions seen in the Hbp structure, however, and the C-terminal region of other ATs presumably shows a very similar fold. Key residues include Trp1015, which is buried and forms a hydrogen bond through its indole side chain to make a turn in the protein backbone.

Hbp has been shown to cross the inner membrane by the signal recognition particle (SRP)-dependent pathway (3), but the precise mechanism of transport across the outer membrane of the cell remains unclear. Several different proposals have been made, including the suggestion that autotransporters form a multimeric pore in the outer membrane (42). Recently the structure of the pore-forming domain of NalP has been solved, showing that it is in fact a monomer, a 12-stranded {beta}-barrel blocked by an N-terminal {alpha} helix (7). The pore diameter, roughly 10–12 Å, is insufficient to allow passage of folded proteins and suggests that only extended polypeptides can pass through. This appears incompatible with the idea that the passenger domain is threaded through from its C-terminal end (the hairpin model), because the pore would then have to accommodate two stretches of polypeptide simultaneously. An alternative model is that the N terminus of the passenger domain is threaded through the pore, but, because other protein domains fused to the pore domain are also transported, there is no evidence of sequence-specific interactions that target the N terminus to the pore. Oomen and colleagues (7) have suggested that the passenger domain may instead pass though the pore of the Omp85 complex, and that the name "autotransporter" is in fact a misnomer. From the crystal structure of Hbp described here, it is clear that the C-terminal domain (residues 950–1048) of the mature protein is more flexible and mobile than the N-terminal protease domain. This is consistent with the prediction that the C-terminal domain plays a key role in passage across the outer membrane. On the other hand, because the N-terminal glycine of the mature protein is buried within the structure, the protease domain cannot fold properly until the signal peptide is removed, and this may provide a mechanism for keeping the protein in a readily transported form.

Tsh (temperature-sensitive hemagglutination), a SPATE found in avian pathogenic E. coli, has only two amino acid changes compared with Hbp (14, 43, 44), but Hbp showed no hemagglutination activity in tests (22). Recently this result has been confirmed for Tsh (45). The two mutations, Lys157 -> Gln and Thr790 -> Ala, are both on the protein surface. Lys157 is roughly 18 Å from the active site, but replacement with glutamine seems unlikely to alter the functional behavior of the protein greatly. Hbp and Tsh are therefore essentially identical, suggesting that Tsh too is a hemoglobin protease. If so, then the name "Tsh" seems redundant, and we suggest the name "Hbp" be used exclusively in the future. Tsh has been shown to bind red blood cells, hemoglobin, fibronectin, and collagen IV (45). It remains to be tested which of these are bound by domain 2.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Hbp was first purified from E. coli EB1, a strain isolated from an intra-abdominal wound infection (22). Without Hbp, E. coli and B. fragilis are unable to form abscesses and are more readily cleared from the site of infection by the immune system, presumably due to iron limitation. Although a great deal is known about iron-uptake by pathogens through siderophores, very few bacterial heme uptake systems have been characterized in molecular detail. The best understood is the HasA protein of Serratia marcescens for which a crystal structure has been refined to 1.9 Å (46). Hbp has a quite different fold, and is over five times larger. Much of the biology involving Hbp remains to be elucidated. No surface protein from either E. coli or B. fragilis has yet been identified that can interact with Hbp and receive heme from it, yet these must presumably exist given the known effects of the protein on these bacteria under conditions of iron limitation. B. fragilis up-regulates expression of a number of proteins under iron-limiting conditions (47), and these are currently being examined as possible receptors for Hbp.

One functional role of Hbp is to remove heme from Hb and supply it to bacteria growing within the body. The crystal structure shows a novel trypsin-like serine protease domain that is well conserved among the SPATE family. The overall {beta}-helix architecture is also common to the ATs, but different loops and domains carried by this stable fold appear to give the proteins distinctive functional properties. Hbp does not have a sequence motif or structural domain, which resembles a known heme binding site. Further heme binding studies with mutants of Hbp and other ATs are currently underway to determine how heme interacts with Hbp.

Homology models of other ATs based on the Hbp structure described here will hopefully be useful in designing new experiments to test their functional roles and find ways to block their activity. Hbp may be a convenient vehicle for presenting peptides for antibody and vaccine production, by inserting them in place of domain 2. Domain 2 itself may well allow us to produce a clinically useful vaccine to reduce substantially the occurrence of severe peritonitis.


   FOOTNOTES
 
* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The atomic coordinates and structure factors (code 1WXR) have been deposited in the Protein Data Bank, Research Collaboratory for Structural Bioinformatics, Rutgers University, New Brunswick, NJ (http://www.rcsb.org/). Back

To whom correspondence should be addressed. Tel.: 81-(0)45-508-7228; Fax: 81-(0)45-508-7366; E-mail: jtame{at}tsurumi.yokohamacu.ac.jp.

1 The abbreviations used are: AT, autotransporter; Hb, hemoglobin; Hbp, hemoglobin protease. Back

2 M. Yao, unpublished program. Back


   ACKNOWLEDGMENTS
 
We Drs. N. Kamiya, Y. Kawano, and H. Nakajima of SPring8 and Prof. S. Wakatsuki and Drs. M. Suzuki, N. Matsunaga, and N. Igarashi of the Photon Factory for help with data collection.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Henderson, I. R., Cappello, R., and Nataro, J. P. (2000) Trends Microbiol. 8, 529–532[CrossRef][Medline] [Order article via Infotrieve]
  2. Henderson, I. R., Navarro-Garcia, F., Desvaux, M., Fernandez, R. C., and Ala'Aldeen, D. (2004) Microbiol. Mol. Biol. Rev. 68, 692–744[Abstract/Free Full Text]
  3. Sijbrandi, R., Urbanus, M. L., ten Hagen-Jongman, C. M., Bernstein, H. D., Oudega, B., Otto, B. R., and Luirink, J. (2003) J. Biol. Chem. 278, 4654–4659[Abstract/Free Full Text]
  4. Klauser, T., Pohlner, J., and Meyer, T. F. (1993) Bioessays 15, 799–805[CrossRef][Medline] [Order article via Infotrieve]
  5. Pohlner, J., Halter, R., Beyreuther, K., and Meyer, T. F. (1987) Nature 325, 458–462[CrossRef][Medline] [Order article via Infotrieve]
  6. Jose, J., Jahnig, F., and Meyer, T. F. (1995) Mol. Microbiol. 18, 378–380[CrossRef][Medline] [Order article via Infotrieve]
  7. Oomen, C. J., Van Ulsen, P., Van Gelder, P., Feijen, M., Tommassen, J., and Gros, P. (2004) EMBO J. 23, 1257–1266[CrossRef][Medline] [Order article via Infotrieve]
  8. Henderson, I. R., and Nataro, J. P. (2001) Infect. Immun. 69, 1231–1243[Free Full Text]
  9. Dutta, P. R., Cappello, R., Navarro-Garcia, F., and Nataro, J. P. (2002) Infect. Immun. 70, 7105–7113[Abstract/Free Full Text]
  10. Stein, M., Kenny, B., Stein, M. A., and Finlay, B. B. (1996) J. Bacteriol. 178, 6546–6554[Abstract/Free Full Text]
  11. Brunder, W., Schmidt, H., and Karch, H. (1997) Mol. Microbiol. 24, 767–778[CrossRef][Medline] [Order article via Infotrieve]
  12. Eslava, C., Navarro-Garcia, F., Czeczulin, J. R., Henderson, I. R., Cravioto, A., and Nataro, J. P. (1998) Infect. Immun. 66, 3155–3163[Abstract/Free Full Text]
  13. Guyer, D. M., Henderson, I. R., Nataro, J. P., and Mobley, H. L. (2000) Mol. Microbiol. 38, 53–66[CrossRef][Medline] [Order article via Infotrieve]
  14. Provence, D. L., and Curtiss, R., 3rd. (1994) Infect. Immun. 62, 1369–1380[Abstract/Free Full Text]
  15. Parreira, V. R., and Gyles, C. L. (2003) Infect. Immun. 71, 5087–5096[Abstract/Free Full Text]
  16. Henderson, I. R., Czeczulin, J., Eslava, C., Noriega, F., and Nataro, J. P. (1999) Infect. Immun. 67, 5587–5596[Abstract/Free Full Text]
  17. Benjelloun-Touimi, Z., Si Tahar, M., Montecucco, C., Sansonetti, P. J., and Parsot, C. (1998) Microbiology 144, 1815–1822[Abstract/Free Full Text]
  18. Farthmann, E. H., and Schoffel, U. (1998) Infection 26, 329–334[Medline] [Order article via Infotrieve]
  19. Aldridge, K. E. (1995) Am. J. Surg. 169, 2S–7S[CrossRef][Medline] [Order article via Infotrieve]
  20. Rotstein, O. D., Kao, J., and Houston, K. (1989) J. Med. Microbiol. 29, 269–276[Abstract/Free Full Text]
  21. Otto, B. R., van Dooren, S. J., Dozois, C. M., Luirink, J., and Oudega, B. (2002) Infect. Immun. 70, 5–10[Abstract/Free Full Text]
  22. Otto, B. R., van Dooren, S. J., Nuijens, J. H., Luirink, J., and Oudega, B. (1998) J. Exp. Med. 188, 1091–1103[Abstract/Free Full Text]
  23. Williams, P. H. (1979) Infect. Immun. 26, 925–932[Abstract/Free Full Text]
  24. Tame, J. R. H., van Dooren, S. J., Oudega, B., and Otto, B. R. (2002) Acta Crystallogr. D. Biol. Crystallogr. 58, 843–845[CrossRef][Medline] [Order article via Infotrieve]
  25. Otwinowski, Z., and Minor, W. (1997) Methods Enzymol. 276, 307–326[CrossRef]
  26. Terwilliger, T. C. (2003) Methods Enzymol. 374, 22–37[Medline] [Order article via Infotrieve]
  27. Terwilliger, T. C., and Berendzen, J. (1999) Acta Crystallogr. D. Biol. Crystallogr. 55, 849–861[CrossRef][Medline] [Order article via Infotrieve]
  28. Jones, T. A., Zou, J. Y., Cowan, S. W., and Kjeldgaard, M. (1991) Acta Crystallogr. A 47, 110–119[CrossRef]
  29. Roussel, A., and Cambillau, C. (1989) TURBO-FRODO, Silicon Graphics, Mountain View, CA
  30. Brunger, A. T. (1996) X-PLOR version 3.85, Yale University Press, New Haven, CT
  31. CCP4 (1994) Acta Crystallogr. D Biol. Crystallogr. 50, 760–763[CrossRef][Medline] [Order article via Infotrieve]
  32. Murshudov, G. N., Vagin, A. A., Lebedev, A., Wilson, K. S., and Dodson, E. J. (1999) Acta Crystallogr. D Biol. Crystallogr. 55, 247–255[CrossRef][Medline] [Order article via Infotrieve]
  33. Emsley, P., Charles, I. G., Fairweather, N. F., and Isaacs, N. W. (1996) Nature 381, 90–92[CrossRef][Medline] [Order article via Infotrieve]
  34. Clantin, B., Hodak, H., Willery, E., Locht, C., Jacob-Dubuisson, F., and Villeret, V. (2004) Proc. Natl. Acad. Sci. U. S. A. 101, 6194–6199[Abstract/Free Full Text]
  35. Jenkins, J., and Pickersgill, R. (2001) Prog. Biophys. Mol. Biol. 77, 111–175[CrossRef][Medline] [Order article via Infotrieve]
  36. Scott, D. J., Grossmann, J. G., Tame, J. R., Byron, O., Wilson, K. S., and Otto, B. R. (2002) J. Mol. Biol. 315, 1179–1187[CrossRef][Medline] [Order article via Infotrieve]
  37. Oliver, D. C., Huang, G., Nodel, E., Pleasance, S., and Fernandez, R. C. (2003) Mol. Microbiol. 47, 1367–1383[CrossRef][Medline] [Order article via Infotrieve]
  38. Holm, L., and Sander, C. (1995) Trends Biochem. Sci. 20, 478–480[CrossRef][Medline] [Order article via Infotrieve]
  39. Holm, L., and Sander, C. (1993) J. Mol. Biol. 233, 123–138[CrossRef][Medline] [Order article via Infotrieve]
  40. Navarro-Garcia, F., Canizalez-Roman, A., Luna, J., Sears, C., and Nataro, J. P. (2001) Infect. Immun. 69, 1053–1060[Abstract/Free Full Text]
  41. Schwede, T., Kopp, J., Guex, N., and Peitsch, M. C. (2003) Nucleic Acids Research 31, 3381–3385[Abstract/Free Full Text]
  42. Veiga, E., Sugawara, E., Nikaido, H., de Lorenzo, V., and Fernandez, L. A. (2002) EMBO J. 21, 2122–2131[CrossRef][Medline] [Order article via Infotrieve]
  43. Stathopoulos, C., Provence, D. L., and Curtiss, R., 3rd. (1999) Infect. Immun. 67, 772–781[Abstract/Free Full Text]
  44. Dozois, C. M., Dho-Moulin, M., Bree, A., Fairbrother, J. M., Desautels, C., and Curtiss, R., 3rd. (2000) Infect. Immun. 68, 4145–4154[Abstract/Free Full Text]
  45. Kostakioti, M., and Stathopoulos, C. (2004) Infect. Immun. 72, 5548–5554[Abstract/Free Full Text]
  46. Arnoux, P., Haser, R., Izadi, N., Lecroisey, A., Delepierre, M., Wandersman, C., and Czjzek, M. (1999) Nat. Struct. Biol. 6, 516–520[CrossRef][Medline] [Order article via Infotrieve]
  47. Otto, B. R., Verweij-van Vught, A. M., van Doorn, J., and Maclaren, D. M. (1988) Microb. Pathog. 4, 279–287[CrossRef][Medline] [Order article via Infotrieve]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 J. Biol. Chem.Home page
M.-E. Charbonneau, J. Janvore, and M. Mourez
Autoprocessing of the Escherichia coli AIDA-I Autotransporter: A NEW MECHANISM INVOLVING ACIDIC RESIDUES IN THE JUNCTION REGION
J. Biol. Chem., June 19, 2009; 284(25): 17340 - 17351.
[Abstract] [Full Text] [PDF]

Home page
 J. Bacteriol.Home page
M.-E. Charbonneau and M. Mourez
Functional Organization of the Autotransporter Adhesin Involved in Diffuse Adherence
J. Bacteriol., December 15, 2007; 189(24): 9020 - 9029.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
H.-J. Yeo, T. Yokoyama, K. Walkiewicz, Y. Kim, S. Grass, and J. W. St. Geme III
The Structure of the Haemophilus influenzae HMW1 Pro-piece Reveals a Structural Domain Essential for Bacterial Two-partner Secretion
J. Biol. Chem., October 19, 2007; 282(42): 31076 - 31084.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
K. A. Gangwer, D. J. Mushrush, D. L. Stauff, B. Spiller, M. S. McClain, T. L. Cover, and D. B. Lacy
Crystal structure of the Helicobacter pylori vacuolating toxin p55 domain
PNAS, October 9, 2007; 104(41): 16293 - 16298.
[Abstract] [Full Text] [PDF]

Home page
 J. Bacteriol.Home page
M.-E. Charbonneau, F. Berthiaume, and M. Mourez
Proteolytic Processing Is Not Essential for Multiple Functions of the Escherichia coli Autotransporter Adhesin Involved in Diffuse Adherence (AIDA-I)
J. Bacteriol., December 15, 2006; 188(24): 8504 - 8512.
[Abstract] [Full Text] [PDF]

Home page
 Infect. Immun.Home page
M. Kostakioti and C. Stathopoulos
Role of the {alpha}-Helical Linker of the C-Terminal Translocator in the Biogenesis of the Serine Protease Subfamily of Autotransporters
Infect. Immun., September 1, 2006; 74(9): 4961 - 4969.
[Abstract] [Full Text] [PDF]

Home page
 J. Bacteriol.Home page
D. S. H. Kim, Y. Chao, and M. H. Saier Jr
Protein-translocating trimeric autotransporters of gram-negative bacteria.
J. Bacteriol., August 1, 2006; 188(16): 5655 - 5667.
[Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
J. H. Peterson, R. L. Szabady, and H. D. Bernstein
An Unusual Signal Peptide Extension Inhibits the Binding of Bacterial Presecretory Proteins to the Signal Recognition Particle, Trigger Factor, and the SecYEG Complex
J. Biol. Chem., April 7, 2006; 281(14): 9038 - 9048.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
M. Junker, C. C. Schuster, A. V. McDonnell, K. A. Sorg, M. C. Finn, B. Berger, and P. L. Clark
Pertactin beta-helix folding mechanism suggests common themes for the secretion and folding of autotransporter proteins
PNAS, March 28, 2006; 103(13): 4918 - 4923.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
R. Simkovsky and J. King
An elongated spine of buried core residues necessary for in vivo folding of the parallel beta-helix of P22 tailspike adhesin
PNAS, March 7, 2006; 103(10): 3575 - 3580.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
280/17/17339    most recent
M412885200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Otto, B. R.
Right arrow Articles by Tame, J. R. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Otto, B. R.
Right arrow Articles by Tame, J. R. H.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 All ASBMB Journals   Molecular and Cellular Proteomics 
 Journal of Lipid Research   ASBMB Today 
Copyright © 2005 by the American Society for Biochemistry and Molecular Biology.
Advertisement
spacer
Advertisement
Advertisement