This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
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 arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Cotter, S. E.
Right arrow Articles by St. Geme, J. W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Cotter, S. E.
Right arrow Articles by St. Geme, J. W., III

 Previous Article  |  Next Article 

Journal of Bacteriology, August 2006, p. 5400-5407, Vol. 188, No. 15
0021-9193/06/$08.00+0     doi:10.1128/JB.00164-06
Copyright © 2006, American Society for Microbiology. All Rights Reserved.

Trimeric Autotransporters Require Trimerization of the Passenger Domain for Stability and Adhesive Activity

Shane E. Cotter,1 Neeraj K. Surana,1 Susan Grass,2 and Joseph W. St. Geme III2*

Edward Mallinckrodt Department of Pediatrics and Department of Molecular Microbiology, Washington University School of Medicine, 660 S. Euclid Ave., St. Louis, Missouri 63110,1 Departments of Pediatrics and Molecular Genetics and Microbiology, Duke University Medical Center, Children's Health Center, Durham, North Carolina 277102

Received 30 January 2006/ Accepted 12 May 2006


arrow
ABSTRACT
 
In recent years, structural studies have identified a number of bacterial, viral, and eukaryotic adhesive proteins that have a trimeric architecture. The prototype examples in bacteria are the Haemophilus influenzae Hia adhesin and the Yersinia enterocolitica YadA adhesin. Both Hia and YadA are members of the trimeric-autotransporter subfamily and are characterized by an internal passenger domain that harbors adhesive activity and a short C-terminal translocator domain that inserts into the outer membrane and facilitates delivery of the passenger domain to the bacterial surface. In this study, we examined the relationship between trimerization of the Hia and YadA passenger domains and the capacity for adhesive activity. We found that subunit-subunit interactions and stable trimerization are essential for native folding and stability and ultimately for full-level adhesive activity. These results raise the possibility that disruption of the trimeric architecture of trimeric autotransporters, and possibly other trimeric adhesins, may be an effective strategy to eliminate adhesive activity.


arrow
INTRODUCTION
 
Microbial adherence to host tissues is an essential early step in the pathogeneses of infectious diseases (18). High-affinity adherence is important in allowing microorganisms to overcome normal host protective mechanisms associated with mechanical force, such as peristalsis, mucociliary clearance, and coughing. Typically, the process of adherence involves a specific interaction between a microbial surface protein, called an adhesin, and a complementary host cell receptor.

Autotransporter proteins are a large family of extracellular proteins that are present in a number of gram-negative pathogenic bacteria (16). These proteins are synthesized as precursor proteins with three functional domains, an N-terminal signal peptide, an internal passenger domain, and a C-terminal pore-forming translocator domain (17, 19, 27). The C-terminal translocator domain is embedded in the outer membrane and facilitates delivery of the internal passenger domain to the bacterial surface.

The trimeric autotransporters represent a subfamily of autotransporter proteins and are defined by the presence of a very short C-terminal translocator domain that forms highly stable trimers in the outer membrane (30, 39). Thus far, all characterized members of the trimeric-autotransporter subfamily have been found to possess adhesive activity, in most cases mediating bacterial adherence to eukaryotic cells (4, 6-8, 31, 35) or extracellular matrix proteins (25, 41) and in some cases resulting in binding of circulating factors, such as immunoglobulins or complement components (1, 13, 33, 34, 42).

The prototype members of the trimeric autotransporter subfamily are the Haemophilus influenzae Hia adhesin and the Yersinia enterocolitica YadA protein. Hia mediates high-affinity adherence to respiratory epithelial cells (23), and YadA is a virulence factor capable of mediating adherence to host cells and extracellular matrix proteins and involved in serum resistance (14). In experiments with Hia, the C-terminal 76 amino acids were capable of presenting a functional heterologous passenger domain on the bacterial surface (39). Similarly, in studies of YadA, the C-terminal 70 amino acids were sufficient for translocating an N-terminal FLAG epitope across the outer membrane (30). Further biochemical analysis established that the C termini of both Hia and YadA form heat-resistant, sodium dodecyl sulfate-resistant trimers in the outer membrane and that Hia requires formic acid denaturation for dissociation (30, 39). Consideration of this information in combination with secondary-structure predictions has led to the proposal that the C-terminal translocator domains of both Hia and YadA form 12-stranded pore-forming ß-barrels containing four strands from each of three subunits (30, 39). The passenger domains of all three subunits are believed to be translocated to the bacterial cell surface through the trimeric pore (9).

Recent crystal structures of the Hia primary binding domain (HiaBD1) and the YadA collagen-binding domain have established that the passenger domains of these proteins are capable of trimerization independent of the translocator domain (28, 43). HiaBD1 is an intricately folded trimer with a large hydrophobic core and multiple subunit-subunit interactions. The adhesive activity of this domain maps to an acidic pocket that is formed by a single monomer and is present on all three faces of the trimer (three pockets per trimer). The YadA collagen-binding domain is a novel left-handed parallel ß-roll with a trimeric architecture and potential to interact with three separate collagen fibers.

In the present study, we examined the relationship between trimerization and the capacity for adhesive activity in the Hia and YadA passenger domains. We found that subunit-subunit interactions and stable trimer formation are essential for native folding and stability and ultimately for full-level adhesive activity. This study provides important insights into the structure and function of trimeric autotransporters and suggests that disruption of the trimeric architecture of these proteins may be an effective strategy to eliminate biological activity.


arrow
MATERIALS AND METHODS
 
Bacterial strains, plasmids, and culture conditions. E. coli strains DH5{alpha} (Life Technologies), BL21(DE3), and XL-1 Blue have been described previously (32). Escherichia coli strain UT5600 is an OmpT– OmpP– derivative of strain RW193 (12).

Plasmid vectors used in this study include pACYC184 (New England BioLabs), pT7-7 (40), and pHAT10 (Clontech). The plasmid pMW10 (also referred to as pYadA) contains the yadA gene from Yersinia enterocolitica strain O:8 and was a gift from Virginia Miller (Washington University, St. Louis, MO). The plasmid pHMW8-7 (also referred to as pHia) contains the hia gene from H. influenzae strain 11 in pT7-7 (4). The plasmid pHMW8-7{Delta}BS2 (also referred to as pHia{Delta}BS2) is a derivative of pHMW8-7 with a deletion of the coding sequence for Hia residues 114 to 127, effectively disrupting HiaBD2 (23). The plasmid pNS1 is a derivative of pHAT10 that contains hia upstream sequence and coding sequence for the Hia signal peptide, the HAT epitope, and Hia residues 977 to 1098 (encompassing the Hia translocator domain) (10). The HAT epitope is a 19-amino-acid polyhistidine peptide that has affinity for nickel and is the target for commercially available antibodies (Clontech, Mountain View, CA).

E. coli strains were grown on Luria-Bertani (LB) agar or in LB broth and were stored at –80°C in LB broth with 30% glycerol. Plasmids were selected with 100 µg/ml of ampicillin and 30 µg/ml of chloramphenicol, as appropriate.

Recombinant DNA methods. DNA ligations, restriction endonuclease digestions, gel electrophoresis, and PCR were performed according to standard techniques (32). Plasmids were introduced into E. coli by electroporation (2). Nucleotide sequencing was performed using an ABI automotated sequencer and the Big Dye Terminator Premix-20 kit version 3.1 (Applied Biosystems/Perkin-Elmer).

Construction of plasmids used in this study. Derivatives of pHMW8-7{Delta}BS2 containing point mutations at residues involved in subunit-subunit interactions were generated using the QuikChange site-directed mutagenesis kit (Stratagene). All constructs were examined by nucleotide sequencing to ensure that the intended mutations were present and that the coding sequence was otherwise wild type.

The plasmid pACYC-HiaC' encodes the Hia signal sequence and then the HAT epitope and Hia residues 977 to 1098, allowing expression of a HAT-tagged outer membrane protein consisting of Hia residues 977 to 1098. This plasmid was generated by digesting pNS1 with PvuII and BamHI, purifying the 0.6-kb fragment, and then ligating the 0.6-kb fragment into EcoRI-BamHI-digested pACYC184. The plasmid pACYC-YadA/C' encodes the Hia signal sequence and then the HAT epitope and YadA residues 325 to 422, allowing expression of a HAT-tagged outer membrane protein consisting of YadA residues 325 to 422. In order to generate this plasmid, pACYC-HiaC' was digested with SacI and EcoRI, releasing the fragment that encodes Hia977-1098. The remaining plasmid backbone was ligated to a 0.3-kb SacI-EcoRI fragment that was amplified from pMW10 and encodes YadA325-422.

Cell fractionation and protein analysis. Whole-cell sonicates were prepared by resuspending bacterial pellets in 10 mM HEPES (pH 7.4) and sonicating them to clarity. Outer membrane fractions were recovered on the basis of Sarkosyl insolubility as described by Carlone et al. (5). Unless otherwise noted, whole-cell sonicates and outer membrane fractions were treated with 95% formic acid to disrupt trimers, as described previously (38). Proteins were resolved by sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis. Western blots were performed with a guinea pig polyclonal antiserum raised against Hia residues 50 to 252 or a rabbit polyclonal antiserum raised against the HAT epitope (Clontech). An anti-guinea pig or an anti-rabbit immunoglobulin G antiserum conjugated to horseradish peroxidase (Sigma) was used as the secondary antibody. Proteins were visualized by incubation with a chemiluminescent substrate solution (Pierce) and exposure to film.

Flow cytometry. Flow cytometry was performed using a guinea pig polyclonal antiserum raised against Hia residues 221 to 658 and a Cy2-conjugated secondary antibody as described previously (23). The antiserum against Hia221-658 reacts specifically with the Hia passenger domain.

Quantitative adherence assays. Chang epithelial cells were maintained in minimal essential medium supplemented with nonessential amino acids and 10% (vol/vol) fetal calf serum and were cultivated at 37°C in 7% CO2. Adherence assays were performed as described previously (36, 37). Briefly, bacteria were inoculated into LB broth containing the appropriate antibiotics and allowed to grow to a density of ~2 x 109 CFU per ml. Approximately 1.5 x 107 CFU was inoculated onto Chang cell monolayers in 24-well tissue culture plates, and the plates were gently centrifuged at 165 x g for 5 min and then incubated at 37°C in 5% CO2 for 25 min. Subsequently, the monolayers were rinsed four times with phosphate-buffered saline to remove nonadherent bacteria, treated with trypsin/EDTA, and resuspended in LB broth. Suspensions were plated on LB agar to yield the number of adherent CFU per monolayer. Adherence was calculated by dividing the number of adherent CFU per monolayer by the number of inoculated CFU per monolayer.


arrow
RESULTS
 
Point mutations that disrupt HiaBD1 subunit-subunit interactions result in increased susceptibility to degradation and loss of adhesive activity. In recent work, we demonstrated that each of the three acidic binding pockets in the HiaBD1 trimeric structure is formed by a single subunit (43). In the present study, we hypothesized that trimerization of HiaBD1 is necessary for proper folding of each subunit and is essential for adhesive activity. To examine this hypothesis, we began by introducing point mutations at subunit-subunit interfaces in the HiaBD1 hydrophobic core, replacing noncharged or hydrophobic residues with charged residues and intending to disrupt subunit-subunit interactions (Fig. 1A). Using a plasmid encoding Hia{Delta}BS2 (an Hia derivative that lacks a functional secondary binding domain), we introduced mutations that changed the alanine at position 645 to aspartic acid (Hia{Delta}BS2/A645D), the leucine at position 650 to aspartic acid (Hia{Delta}BS2/L650D), or the phenylalanine at position 681 to glutamic acid (Hia{Delta}BS2/F681E) and then transformed the mutant constructs into E. coli DH5{alpha}. As shown in Fig. 1B, the mutant proteins were not detectable in outer membrane preparations, suggesting proteolytic degradation due to global misfolding.


Figure 1
View larger version (39K):
[in this window]
[in a new window]
  		FIG. 1. Effects of point mutations at subunit-subunit interfaces on HiaBD1 stability and adhesive activity in E. coli DH5{alpha}. (A) Stereo ribbon diagram of the HiaBD1 trimer highlighting the locations of the three point mutations engineered to disrupt subunit-subunit interactions. Mutations A645D and L650D disrupt an internal {alpha}-helix, and mutation F681E lies deep in the hydrophobic core created by interdigitating ß-strands from all three subunits. (B) Hia levels in outer membrane preparations, as assessed by Western analysis with an antiserum against Hia and a Cy2-conjugated secondary antibody. Samples were loaded as follows: lane 1, DH5{alpha}; lane 2, DH5{alpha}/pHia{Delta}BS2; lane 3, DH5{alpha}/pHia{Delta}BS2/A645D; lane 4, DH5{alpha}/pHia{Delta}BS2/L650D; lane 5, DH5{alpha}/pHia{Delta}BS2/F681E.

In an effort to circumvent proteolysis and examine the abilities of Hia{Delta}BS2/A645D, Hia{Delta}BS2/L650D, and Hia{Delta}BS2/F681E to mediate bacterial adherence, we expressed these proteins in E. coli UT5600, a laboratory strain that lacks the OmpT and OmpP outer membrane proteases responsible for degrading misfolded outer membrane proteins (12). As controls, we expressed these proteins in E. coli RW193, the parent of UT5600 with functional OmpT and OmpP proteases. Western analysis and flow cytometry revealed appreciable levels of the mutant proteins in the outer membrane and on the surface of UT5600, comparable to levels of Hia{Delta}BS2 (Fig. 2A and B). In contrast, the mutant proteins were absent from the bacterial outer membrane and the bacterial surface of E. coli RW193 (Fig. 2A and B) (indicating degradation by OmpT and/or OmpP). Interestingly, UT5600/pHia{Delta}BS2/A645D, UT5600/pHia{Delta}BS2/L650D, and UT5600/pHia{Delta}BS2/F681E were all nonadherent in assays with Chang epithelial cells (Fig. 2C), similar to E. coli harboring empty vector, suggesting that Hia lacking normal subunit-subunit interactions is misfolded and lacks adhesive activity.


Figure 2
View larger version (30K):
[in this window]
[in a new window]
  		FIG. 2. Effects of point mutations at subunit-subunit interfaces on HiaBD1 stability and adhesive activity in E. coli RW193 and E. coli UT5600 (OmpT– OmpP–). (A) Hia levels in outer membrane preparations from RW193 and UT5600, as assessed by Western analysis with an antiserum against Hia. Samples were loaded as follows: lanes 1, E. coli without plasmid; lanes 2, E. coli/pHia{Delta}BS2; lanes 3, E. coli/pHia{Delta}BS2/A645D; lanes 4, E. coli/pHia{Delta}BS2/L650D; lanes 5, E. coli/pHia{Delta}BS2/F681E. (B) Hia levels on the bacterial surfaces of RW193 and UT5600, as assessed by flow cytometry with an antiserum against Hia. E. coli without plasmid is indicated by the black line, and E. coli expressing the designated recombinant protein is indicated by the shaded curve. (C) Adherence to Chang epithelial cells by UT5600 derivatives. Adherence was calculated by determining the percentage of the inoculum that was adherent after a 30-min assay. The bars represent means plus standard errors of three measurements from a representative experiment. E. coli expressing vector alone (–) was used as a negative control.

Trimer formation is required for normal folding, stability, and adhesive activity of Hia and YadA. To extend our observations with point mutations at subunit-subunit interfaces, we sought more direct evidence that trimerization of the Hia passenger domain is required for normal folding, stability, and adhesive activity. With this goal in mind, we exploited the fact that the Hia C-terminal translocator domain undergoes trimerization. We reasoned that coexpression of full-length Hia with a construct encoding only the Hia C-terminal translocator domain (HiaC') would lead to the generation of heterotrimers containing a trimeric translocator domain and either one or two full-length passenger domains (in addition to homotrimeric full-length Hia and homotrimeric HiaC').

In anticipation of the possibility that heterotrimers might be misfolded and prone to degradation, we coexpressed full-length Hia and HiaC' in both E. coli RW193 and E. coli UT5600 (OmpT– OmpP–). As controls, we expressed full-length Hia with pACYC184 (empty vector) in strains RW193 and UT5600. As shown in Fig. 3A, Western analysis of outer membranes prepared from RW193/pHia+pACYC-HiaC' (left, lane 3) revealed no detectable full-length Hia, presumably reflecting minimal formation of trimeric full-length Hia and efficient degradation of heterotrimers containing only one or two full-length Hia passenger domains. In contrast, examination of outer membranes prepared from UT5600/pHia+pACYC-HiaC' (right, lane 3) demonstrated high levels of full-length Hia, comparable to levels in UT5600 expressing Hia alone. To confirm the presence of heterotrimers in UT5600/pHia+pACYC-HiaC', we examined outer membranes preparations that had not been treated with formic acid. As shown in Fig. 3B, in the absence of formic acid denaturation, a small quantity of trimeric full-length Hia was present, but the predominant species containing full-length Hia were heterotrimers containing one or two full-length molecules (1x full-length Hia/2x HiaC' or 2x full-length Hia/1x HiaC').


Figure 3
View larger version (26K):
[in this window]
[in a new window]
  		FIG. 3. Coexpression of HiaC' and full-length Hia. (A) Full-length Hia (arrow) and HiaC' (arrowhead) in outer membrane preparations from E. coli RW193 and E. coli UT5600 (an OmpT– OmpP– derivative of RW193) as visualized by Western analysis with an antibody specific to the Hia passenger domain (reactive with full-length Hia) and an anti-HAT antibody (reactive with HiaC'). After formic acid denaturation, monomeric Hia protein runs at ~115 kDa and monomeric HiaC' protein migrates at ~13 kDa. Samples were loaded as follows: lanes 1, E. coli without plasmid; lanes 2, E. coli/pHia+pACYC184; lanes 3, E. coli/pHia+pACYC-HiaC'. (B) Hia heterotrimers in the outer membrane of E. coli strain UT5600/pHia+pACYC-HiaC'. Outer membrane fractions were recovered from UT5600, UT5600/pHia+pACYC184, and UT5600/pHia+pACYC-HiaC' and suspended in Laemmli buffer, omitting formic acid denaturation. Subsequently, samples were boiled at 95°C for 10 min, resolved on a 4% to 15% SDS-polyacrylamide gel, and examined by Western analysis with an antiserum against Hia50-252. Samples were loaded as follows: lane 1, UT5600; lane 2, UT5600/pHia+pACYC184; lane 3, UT5600/pHia+pACYC-HiaC'. The diamond corresponds to 3x full-length Hia (~345 kDa), the closed circle corresponds to 2x full-length Hia/1x HiaC' (~245 kDa), and the open arrowhead corresponds to 1x full-length Hia/2x HiaC' (~140 kDa). (C) Hia heterotrimers on the bacterial surface of E. coli UT5600 expressing Hia plus HiaC', as assessed by flow cytometry with an antiserum against Hia221-658 and a Cy2-conjugated secondary antibody. UT5600 (OmpT– OmpP–) is depicted with a shaded curve (negative control), UT5600/pHia+pACYC184 is depicted with a solid gray line (positive control), and UT5600/pHia+pACYC-HiaC' is depicted with a dotted black line. (D) Adherence to Chang epithelial cells by E. coli UT5600 expressing Hia or Hia plus HiaC'. Adherence was calculated by determining the percentage of the inoculum that was adherent after a 30-min assay. The bars represent means plus standard errors of three measurements from a representative experiment. UT5600 alone (–) was used as a negative control. The plasmid pACYC184 is abbreviated as pACYC.

To verify that Hia passenger domains present in heterotrimers were delivered to the bacterial cell surface, we performed flow cytometry with an antiserum against the Hia passenger domain. Consistent with the results of Western analysis of outer membrane preparations, surface-localized Hia was abundant in UT5600/pHia+pACYC-HiaC' (Fig. 3C). Interestingly, UT5600/pHia+pACYC-HiaC' exhibited slightly enhanced immunoreactivity compared to UT5600 expressing Hia alone, potentially reflecting increased exposure of epitopes recognized by the Hia antibody in a misfolded passenger domain.

To assess the adhesive activities of Hia passenger domains present on the bacterial surface in heterotrimers, we examined UT5600/pHia+pACYC-HiaC' for adherence to Chang epithelial cells, using UT5600/pHia+pACYC184 (expressing wild-type Hia alone) as a control. As shown in Fig. 3D, adherence by UT5600/pHia+pACYC-HiaC' was reduced ~60% compared to adherence by UT5600/pHia+pACYC184, suggesting that heterotrimers containing only one or two full-length Hia molecules possess minimal adhesive activity. In control experiments, we compared E. coli expressing wild-type Hia encoded by the low-copy-number plasmid pACYC184 and E. coli expressing wild-type Hia encoded by pHia and found marked differences in levels of Hia and comparable levels of adherence (data not shown), arguing that small differences in levels of full-length Hia between UT5600/pHia and UT5600/pHia+pACYC-HiaC' do not account for the major differences in adherence.

To examine whether the results with Hia are applicable to the trimeric-autotransporter subfamily more generally, we coexpressed full-length YadA and the YadA C-terminal translocator domain (YadA/C'). As shown in Fig. 4A, full-length YadA was absent in outer membranes from RW193/pYadA+pACYC-YadA/C' (left, lane 3) but was present at high levels in outer membranes from UT5600/pYadA+pACYC-YadA/C' (right, lane 3). Furthermore, when outer membranes from UT5600/pYadA+pACYC-YadA/C' were assessed by Western analysis after omission of formic acid denaturation, a banding pattern indicative of heterotrimers was observed (data not shown), similar to earlier studies by Roggenkamp et al. (30). Finally, adherence by UT5600/pYadA+pACYC-YadA/C' was reduced ~70% compared to adherence by UT5600/pYadA+pACYC184 (Fig. 4B), supporting the conclusion that heterotrimers containing one or two full-length YadA molecules harbor minimal adhesive activity.


Figure 4
View larger version (23K):
[in this window]
[in a new window]
  		FIG. 4. Coexpression of YadA/C' and full-length YadA. (A) Full-length YadA (arrow) and YadA/C' (arrowhead) in outer membrane preparations from E. coli RW193 and E. coli UT5600 (an OmpT– OmpP– derivative of RW193), as visualized by Western analysis with an anti-HAT antibody. After formic acid denaturation, monomeric YadA protein migrates at ~45 kDa, and monomeric YadA/C' protein migrates at ~11 kDa. Samples were loaded as follows: lanes 1, E. coli without plasmid; lanes 2, E. coli/pYadA+pACYC184; lanes 3, E. coli/pYadA+pACYC-YadA/C'. (B) Adherence to Chang epithelial cells by E. coli UT5600 expressing YadA or YadA plus YadA/C'. Adherence was calculated by determining the percentage of the inoculum that was adherent after a 30-min assay. The bars represent means plus standard errors of three measurements from a representative experiment. UT5600 alone (–) was used as a negative control. The plasmid pACYC184 is abbreviated as pACYC.

Together, these results provide strong evidence that trimerization of the passenger domain is necessary for full adhesive function of both Hia and YadA. In the absence of trimerization, the passenger domain is misfolded, eliminating adhesive activity and conferring susceptibility to degradation by OmpT and/or OmpP.

The Hia and YadA C-terminal translocator domains do not inhibit each other. Given the structural similarities between the Hia and the YadA C-terminal translocators, we wondered whether the C termini of these heterologous proteins might be capable of stably interacting with each other, thereby abrogating adhesive activity. To address this possibility, we coexpressed the YadA C-terminal translocator domain with full-length Hia and the Hia C-terminal translocator domain with full-length YadA in E. coli DH5{alpha} (Fig. 5). YadA/C' had a minimal effect on Hia protein levels (Fig. 5A, lane 4) and no significant effect on Hia-mediated adherence (data not shown). Similarly, HiaC' had no appreciable effect on YadA protein levels (Fig. 5B, lane 3) and no significant effect on YadA-mediated adhesive activity (data not shown).


Figure 5
View larger version (30K):
[in this window]
[in a new window]
  		FIG. 5. Interchangeability of HiaC' and YadA/C'. (A) Hia, Hia plus HiaC', and Hia plus YadA/C' protein levels in outer membrane preparations, as assessed by Western analysis with an antiserum against the Hia passenger domain (reactive with full-length Hia) (arrow) and an anti-HAT antibody that detects HiaC' and YadA/C' (arrowhead). Samples were loaded as follows: lane 1, DH5{alpha}; lane 2, DH5{alpha}/pHia+pACYC184; lane 3, DH5{alpha}/pHia+pACYC-HiaC'; lane 4, DH5{alpha}/pHia+pACYC-YadA/C'. (B) YadA alone, YadA plus HiaC', and YadA plus YadA/C' protein levels in outer membrane preparations as assessed by Western analysis with an anti-HAT antibody that detects full-length YadA (arrow), HiaC' (arrowhead), and YadA/C' (arrowhead). Samples were loaded as follows: lane 1, DH5{alpha}; lane 2, DH5{alpha}/pYadA+pACYC184; lane 3, DH5{alpha}/pYadA+pACYC-HiaC'; lane 4, DH5{alpha}/pYadA+pACYC-YadA/C'.


arrow
DISCUSSION
 
In the present study, we examined whether subunit-subunit interactions and stable trimerization of the Hia and YadA passenger domains are important for Hia and YadA adhesive activities. Based on analysis of the HiaBD1 crystal structure, we were able to target specific residues in the hydrophobic core involved in subunit-subunit interactions, remote from the binding pocket. Insertion of charged residues in place of A645, L650, or F681 in the hydrophobic core disrupted overall protein conformation, resulting in a loss of adhesive activity. In addition, we were able to generate strains expressing Hia and YadA molecules with a trimeric translocator domain and zero, one, two, or three full-length passenger domains, exploiting the fact that the Hia and YadA C-terminal translocator domains undergo trimerization. These strains had decreased quantities of surface-localized trimeric passenger domains and markedly reduced adhesive activities. Together, these data argue that subunit-subunit interactions and trimer formation are essential for normal folding and full-level adhesive activity of individual Hia and YadA passenger domains and potentially other trimeric autotransporters.

In considering our results with the A645D, L650D, and F681E mutations in Hia, it is notable that similar mutations in YadA have been examined. In particular, El Tahir et al. (15) identified a repeating NSVAIG-S motif in the head domain of YadA and replaced the small hydrophobic residues in this motif with either aspartic acid or glutamic acid. Based on the crystal structure of the YadA collagen binding domain, the NSVAIG-S motif forms the core of the YadA trimer, mediating hydrophobic interactions between ß-rolls (28). Mutations in the NSVAIG-S motif disrupted subunit-subunit interactions and resulted in anomalous migration on an SDS-polyacrylamide gel electrophoresis gel and complete or almost complete loss of collagen binding (15), analogous to our observations with Hia.

In our experiments examining E. coli UT5600 expressing either Hia or YadA heterotrimers, we observed that bacterial adherence was decreased but not eliminated compared to control strains. Regarding the low-level adherence by these strains, it is possible that the misfolded monomeric and dimeric passenger domains present on the bacterial surface in heterotrimers retain some adhesive activity. Arguing against this possibility, UT5600/Hia{Delta}BS2/A645D, UT5600/Hia{Delta}BS2/L650D, and UT5600/Hia{Delta}BS2/F681E (with mutations at subunit-subunit interfaces in Hia) all expressed misfolded passenger domains on the bacterial surface and were completely nonadherent, comparable to background. We postulate that the residual adherence by UT5600/pHia+pACYC-HiaC' and UT5600/pYadA+pACYC-YadA/C' is explained instead by the presence of small, but appreciable, quantities of wild-type homotrimeric protein on the bacterial surface, as demonstrated by Western analysis of outer membrane preparations. In support of this explanation, mutation of A645, L650, or F681 completely abrogated Hia-mediated adherence in a background devoid of wild-type homotrimeric protein.

In considering the stringent requirement for trimerization of Hia and YadA for full-scale adhesive activity, it is notable that a trimeric architecture with three identical faces provides the potential for three identical binding pockets and a multivalent interaction with the host cell surface (9). Compared with a monovalent interaction, multivalency results in increased avidity and a more stable interaction with higher affinity (9). This increase in avidity may be important in allowing organisms to overcome mechanical forces in the host, including mucociliary clearance and coughing in the respiratory tract and peristalsis in the intestinal tract. In this context, it is interesting that all characterized trimeric autotransporters have been found to possess adhesive activity, in the majority of cases mediating bacterial adherence to epithelial cells (4, 6-8, 31, 35) or to extracellular matrix proteins (25, 41).

The presence of a trimeric architecture and the capacity for multivalent high-affinity adherence has been observed in a number of viral and eukaryotic adhesive proteins, as well. One example is the adenovirus type 2 fiber, a homotrimer that contains three identical binding pockets and interacts with three separate host cell receptor molecules, referred to as coxsackievirus and adenovirus receptors. Based on studies using surface plasmon resonance to compare trivalent and monovalent interactions between the type 2 fiber and coxsackievirus and adenovirus receptors, the trivalent interaction results in a 25-fold increase in affinity (24).

Beyond the potential for multivalent high-affinity binding, homotrimeric viral proteins also exhibit resistance to proteases and detergents, with examples including the bacteriophage T4 gp9 and gp11 tail components, the bacteriophage P22 tailspike endorhamnosidase, and the adenovirus type 2 fiber (11, 21, 22, 26). The stability of these homotrimeric viral proteins is presumed to be critical for maintaining function as external virion proteins in the presence of proteases and denaturing conditions in nature. Along these lines, it is noteworthy that the three subunits in HiaBD1 are extensively intertwined, resulting in resistance to degradation by trypsin and to denaturation by SDS (43). We speculate that these properties may be applicable to all trimeric autotransporters.

Our results demonstrating the importance of trimerization for Hia and YadA adhesive activities raise the possibility that disruption of the trimeric architecture of Hia and YadA, and perhaps all trimeric autotransporters, may be a productive approach to eliminate adhesive activity. Using the Hia structure as a model, one approach would be to identify small molecules that interact with the hydrophobic core of the trimer and interfere with subunit-subunit interactions during the process of trimerization, analogous to the effect of point mutations at A645, L650, and F681. Additional crystal structures will be necessary to assess the likelihood that a given small molecule is capable of interfering with trimerization of multiple distinct trimeric autotransporters.

Coexpression of YadA/C' with full-length Hia had minimal effect on Hia protein levels and no effect on Hia adhesive activity, and coexpression of HiaC' with full-length YadA had no effect on YadA protein levels or YadA-mediated adherence. Together, these findings suggest that the YadA and Hia C-terminal translocator domains are unable to interact stably with each other. This specificity is somewhat surprising, considering the structural similarity between the C-terminal ends of Hia and YadA, with each predicted to form four transmembrane ß-strands. However, it is possible that this specificity evolved to protect against hetero-oligomerization in bacteria that express more than one member of this subfamily, thereby preventing loss of passenger domain function (7, 29).

A prevailing hypothesis regarding autotransporter proteins is that the energy required for passenger domain translocation across the outer membrane is derived from the native folding of the passenger domain on the bacterial surface (20). Indeed, we have speculated previously that passenger domain translocation in trimeric autotransporters occurs in concert with trimerization (43). However, our findings in the present study argue against this hypothesis, at least for the trimeric-autotransporter subfamily. In particular, we were able to detect Hia and YadA heterotrimers with only one or two passenger domains on the bacterial surface, provided that the OmpT and OmpP outer membrane proteases were absent.

In conclusion, we have shown that trimerization of the Hia and YadA passenger domains is necessary for full-scale adhesive function. These are the first studies to define the relationship between trimer formation and function for members of the trimeric-autotransporter family. We speculate that our findings apply to all members of the growing trimeric-autotransporter subfamily, raising the possibility that disruption of trimerization may be a broadly effective strategy to abrogate the activities of trimeric autotransporters.


arrow
ACKNOWLEDGMENTS
 
This work was supported by NIH grant RO1-AI44167 to J.W.S., by NIH training grant T32-HL07873 to S.E.C., and by NIH training grant T32-AI07172 to N.K.S. S.E.C. and N.K.S. are members of the Medical Scientist Training Program at Washington University School of Medicine.


arrow
FOOTNOTES
 
* Corresponding author. Mailing address: Department of Pediatrics, Duke University Medical Center, Children's Health Center, Room T901, DUMC 3352, Durham, NC 27710. Phone: (919) 681-4080. Fax: (919) 681-2714. E-mail: j.stgeme{at}duke.edu. Back


arrow
REFERENCES
 
    1
  1. Aebi, C., E. R. Lafontaine, L. D. Cope, J. L. Latimer, S. L. Lumbley, G. H. McCracken, Jr., and E. J. Hansen. 1998. Phenotypic effect of isogenic uspA1 and uspA2 mutations on Moraxella catarrhalis 035E. Infect. Immun. 66:3113-3119.[Abstract/Free Full Text]
    2. 2
  2. Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A. Smith, and K. Struhl. 1994. Current protocols in molecular biology. John Wiley & Sons, New York, N.Y.
    3. 3
  3. Barenkamp, S. J. 1996. Immunization with high-molecular-weight adhesion proteins of nontypeable Haemophilus influenzae modifies experimental otitis media in chinchillas. Infect. Immun. 64:1246-1251.[Abstract]
    4. 4
  4. Barenkamp, S. J., and J. W. St. Geme III. 1996. Identification of a second family of high-molecular-weight adhesion proteins expressed by non-typeable Haemophilus influenzae. Mol. Microbiol. 19:1215-1223.[Medline]
    5. 5
  5. Carlone, G. M., M. L. Thomas, H. S. Rumschlag, and F. O. Sottnek. 1986. Rapid microprocedure for isolating detergent-insoluble outer membrane proteins from Haemophilus species. J. Clin. Microbiol. 24:330-332.[Abstract/Free Full Text]
    6. 6
  6. Cole, L. E., T. H. Kawula, K. L. Toffer, and C. Elkins. 2002. The Haemophilus ducreyi serum resistance antigen DsrA confers attachment to human keratinocytes. Infect. Immun. 70:6158-6165.[Abstract/Free Full Text]
    7. 7
  7. Comanducci, M., S. Bambini, B. Brunelli, J. Adu-Bobie, B. Arico, B. Capecchi, M. M. Giuliani, V. Masignani, L. Santini, S. Savino, D. M. Granoff, D. A. Caugant, M. Pizza, R. Rappuoli, and M. Mora. 2002. NadA, a novel vaccine candidate of Neisseria meningitidis. J. Exp. Med. 195:1445-1454.[Abstract/Free Full Text]
    8. 8
  8. Cope, L. D., E. R. Lafontaine, C. A. Slaughter, C. A. Hasemann, Jr., C. Aebi, F. W. Henderson, G. H. McCracken, Jr., and E. J. Hansen. 1999. Characterization of the Moraxella catarrhalis uspA1 and uspA2 genes and their encoded products. J. Bacteriol. 181:4026-4034.[Abstract/Free Full Text]
    9. 9
  9. Cotter, S. E., N. K. Surana, and J. W. St Geme III. 2005. Trimeric autotransporters: a distinct subfamily of autotransporter proteins. Trends Microbiol. 13:199-205.[CrossRef][Medline]
    10. 10
  10. Cotter, S. E., H. J. Yeo, T. Juehne, and J. W. St Geme III. 2005. Architecture and adhesive activity of the Haemophilus influenzae Hsf adhesin. J. Bacteriol. 187:4656-4664.[Abstract/Free Full Text]
    11. 11
  11. Danner, M., A. Fuchs, S. Miller, and R. Seckler. 1993. Folding and assembly of phage P22 tailspike endorhamnosidase lacking the N-terminal, head-binding domain. Eur. J. Biochem. 215:653-661.[Medline]
    12. 12
  12. Elish, M. E., J. R. Pierce, and C. F. Earhart. 1988. Biochemical analysis of spontaneous fepA mutants of Escherichia coli. J. Gen. Microbiol. 134:1355-1364.[Abstract/Free Full Text]
    13. 13
  13. Elkins, C., K. J. Morrow, Jr., and B. Olsen. 2000. Serum resistance in Haemophilus ducreyi requires outer membrane protein DsrA. Infect. Immun. 68:1608-1619.[Abstract/Free Full Text]
    14. 14
  14. El Tahir, Y., and M. Skurnik. 2001. YadA, the multifaceted Yersinia adhesin. Int. J. Med. Microbiol. 291:209-218.[CrossRef][Medline]
    15. 15
  15. El Tahir, Y., P. Kuusela, and M. Skurnik. 2000. Functional mapping of the Yersinia enterocolitica adhesin YadA. Identification Of eight NSVAIG-S motifs in the amino-terminal half of the protein involved in collagen binding. Mol. Microbiol. 37:192-206.[CrossRef][Medline]
    16. 16
  16. Henderson, I. R., and J. P. Nataro. 2001. Virulence functions of autotransporter proteins. Infect. Immun. 69:1231-1243.[Free Full Text]
    17. 17
  17. Henderson, I. R., F. Navarro-Garcia, and J. P. Nataro. 1998. The great escape: structure and function of the autotransporter proteins. Trends Microbiol. 6:370-378.[CrossRef][Medline]
    18. 18
  18. Hultgren, S. J., S. Abraham, M. Caparon, P. Falk, J. W. St. Geme III, and S. Normark. 1993. Pilus and nonpilus bacterial adhesins: assembly and function in cell recognition. Cell 73:887-901.[CrossRef][Medline]
    19. 19
  19. Jacob-Dubuisson, F., R. Fernandez, and L. Coutte. 2004. Protein secretion through autotransporter and two-partner pathways. Biochim. Biophys. Acta 1694:235-257.[Medline]
    20. 20
  20. Klauser, T., J. Pohlner, and T. F. Meyer. 1992. Selective extracellular release of cholera toxin B subunit by Escherichia coli: dissection of Neisseria IgA beta-mediated outer membrane transport. EMBO J. 11:2327-2335.[Medline]
    21. 21
  21. Kostyuchenko, V. A., G. A. Navruzbekov, L. P. Kurochkina, S. V. Strelkov, V. V. Mesyanzhinov, and M. G. Rossmann. 1999. The structure of bacteriophage T4 gene product 9: the trigger for tail contraction. Struct. Fold Des. 7:1213-1222.[Medline]
    22. 22
  22. Kurochkina, L. P., P. G. Leiman, S. Y. Venyaminov, and V. V. Mesyanzhinov. 2001. Expression and properties of bacteriophage T4 gene product 11. Biochemistry 66:141-146.
    23. 23
  23. Laarmann, S., D. Cutter, T. Juehne, S. J. Barenkamp, and J. W. St Geme. 2002. The Haemophilus influenzae Hia autotransporter harbours two adhesive pockets that reside in the passenger domain and recognize the same host cell receptor. Mol. Microbiol. 46:731-743.[CrossRef][Medline]
    24. 24
  24. Lortat-Jacob, H., E. Chouin, S. Cusack, and M. J. van Raaij. 2001. Kinetic analysis of adenovirus fiber binding to its receptor reveals an avidity mechanism for trimeric receptor-ligand interactions. J. Biol. Chem. 276:9009-9015.[Abstract/Free Full Text]
    25. 25
  25. McMichael, J. C., M. J. Fiske, R. A. Fredenburg, D. N. Chakravarti, K. R. VanDerMeid, V. Barniak, J. Caplan, E. Bortell, S. Baker, R. Arumugham, and D. Chen. 1998. Isolation and characterization of two proteins from Moraxella catarrhalis that bear a common epitope. Infect. Immun. 66:4374-4381.[Abstract/Free Full Text]
    26. 26
  26. Mitraki, A., A. Barge, J. Chroboczek, J. P. Andrieu, J. Gagnon, and R. W. Ruigrok. 1999. Unfolding studies of human adenovirus type 2 fibre trimers. Evidence for a stable domain. Eur. J. Biochem. 264:599-606.[Medline]
    27. 27
  27. Newman, C. L., and C. Stathopoulos. 2004. Autotransporter and two-partner secretion: delivery of large-size virulence factors by gram-negative bacterial pathogens. Crit. Rev. Microbiol. 30:275-286.[CrossRef][Medline]
    28. 28
  28. Nummelin, H., M. C. Merckel, J. C. Leo, H. Lankinen, M. Skurnik, and A. Goldman. 2004. The Yersinia adhesin YadA collagen-binding domain structure is a novel left-handed parallel beta-roll. EMBO J. 23:701-711.[CrossRef][Medline]
    29. 29
  29. Peak, I. R., Y. Srikhanta, M. Dieckelmann, E. R. Moxon, and M. P. Jennings. 2000. Identification and characterisation of a novel conserved outer membrane protein from Neisseria meningitidis. FEMS Immunol. Med. Microbiol. 28:329-334.[CrossRef][Medline]
    30. 30
  30. Roggenkamp, A., N. Ackermann, C. A. Jacobi, K. Truelzsch, H. Hoffmann, and J. Heesemann. 2003. Molecular analysis of transport and oligomerization of the Yersinia enterocolitica adhesin YadA. J. Bacteriol. 185:3735-3744.[Abstract/Free Full Text]
    31. 31
  31. Roggenkamp, A., H. R. Neuberger, A. Flugel, T. Schmoll, and J. Heesemann. 1995. Substitution of two histidine residues in YadA protein of Yersinia enterocolitica abrogates collagen binding, cell adherence and mouse virulence. Mol. Microbiol. 16:1207-1219.[Medline]
    32. 32
  32. Sambrook, J., and D. W. Russel. 2001. Molecular cloning: a laboratory manual, 3rd ed. Cold Spring Harbor Press, Cold Spring Harbor, N.Y.
    33. 33
  33. Sandt, C. H., and C. W. Hill. 2000. Four different genes responsible for nonimmune immunoglobulin-binding activities within a single strain of Escherichia coli. Infect. Immun. 68:2205-2214.[Abstract/Free Full Text]
    34. 34
  34. Sandt, C. H., and C. W. Hill. 2001. Nonimmune binding of human immunoglobulin A (IgA) and IgG Fc by distinct sequence segments of the EibF cell surface protein of Escherichia coli. Infect. Immun. 69:7293-7303.[Abstract/Free Full Text]
    35. 35
  35. St. Geme, J. W., III, D. Cutter, and S. J. Barenkamp. 1996. Characterization of the genetic locus encoding Haemophilus influenzae type b surface fibrils. J. Bacteriol. 178:6281-6287.[Abstract/Free Full Text]
    36. 36
  36. St. Geme, J. W., III, and S. Falkow. 1990. Haemophilus influenzae adheres to and enters cultured human epithelial cells. Infect. Immun. 58:4036-4044.[Abstract/Free Full Text]
    37. 37
  37. St. Geme, J. W., III, S. Falkow, and S. J. Barenkamp. 1993. High-molecular-weight proteins of nontypeable Haemophilus influenzae mediate attachment to human epithelial cells. Proc. Natl. Acad. Sci. USA 90:2875-2879.[Abstract/Free Full Text]
    38. 38
  38. St. Geme, J. W., III, V. V. Kumar, D. Cutter, and S. J. Barenkamp. 1998. Prevalence and distribution of the hmw and hia genes and the HMW and Hia adhesins among genetically diverse strains of nontypeable Haemophilus influenzae. Infect. Immun. 66:364-368.[Abstract/Free Full Text]
    39. 39
  39. Surana, N. K., D. Cutter, S. J. Barenkamp, and J. W. St Geme, III. 2004. The Haemophilus influenzae Hia autotransporter contains an unusually short trimeric translocator domain. J. Biol. Chem. 279:14679-14685.[Abstract/Free Full Text]
    40. 40
  40. Tabor, S., and C. C. Richardson. 1985. A bacteriophage T7 RNA polymerase/promoter system for controlled exclusive expression of specific genes. Proc. Natl. Acad. Sci. USA 82:1074-1078.[Abstract/Free Full Text]
    41. 41
  41. Tamm, A., A. M. Tarkkanen, T. K. Korhonen, P. Kuusela, P. Toivanen, and M. Skurnik. 1993. Hydrophobic domains affect the collagen-binding specificity and surface polymerization as well as the virulence potential of the YadA protein of Yersinia enterocolitica. Mol. Microbiol. 10:995-1011.[Medline]
    42. 42
  42. Visser, L. G., P. S. Hiemstra, M. T. van den Barselaar, P. A. Ballieux, and R. van Furth. 1996. Role of YadA in resistance to killing of Yersinia enterocolitica by antimicrobial polypeptides of human granulocytes. Infect. Immun. 64:1653-1658.[Abstract]
    43. 43
  43. Yeo, H. J., S. E. Cotter, S. Laarmann, T. Juehne, J. W. St Geme, and G. Waksman. 2004. Structural basis for host recognition by the Haemophilus influenzae Hia autotransporter. EMBO J. 23:1245-1256.[CrossRef][Medline]

Journal of Bacteriology, August 2006, p. 5400-5407, Vol. 188, No. 15
0021-9193/06/$08.00+0     doi:10.1128/JB.00164-06
Copyright © 2006, American Society for Microbiology. All Rights Reserved.



This article has been cited by other articles:

  • Winter, L. E., Barenkamp, S. J. (2009). Antibodies Specific for the Hia Adhesion Proteins of Nontypeable Haemophilus influenzae Mediate Opsonophagocytic Activity. CVI 16: 1040-1046 [Abstract] [Full Text]  
  • Leduc, I., Olsen, B., Elkins, C. (2009). Localization of the Domains of the Haemophilus ducreyi Trimeric Autotransporter DsrA Involved in Serum Resistance and Binding to the Extracellular Matrix Proteins Fibronectin and Vitronectin. Infect. Immun. 77: 657-666 [Abstract] [Full Text]  
  • Brooks, M. J., Sedillo, J. L., Wagner, N., Laurence, C. A., Wang, W., Attia, A. S., Hansen, E. J., Gray-Owen, S. D. (2008). Modular Arrangement of Allelic Variants Explains the Divergence in Moraxella catarrhalis UspA Protein Function. Infect. Immun. 76: 5330-5340 [Abstract] [Full Text]  
  • Ackermann, N., Tiller, M., Anding, G., Roggenkamp, A., Heesemann, J. (2008). Contribution of Trimeric Autotransporter C-Terminal Domains of Oligomeric Coiled-Coil Adhesin (Oca) Family Members YadA, UspA1, EibA, and Hia to Translocation of the YadA Passenger Domain and Virulence of Yersinia enterocolitica. J. Bacteriol. 190: 5031-5043 [Abstract] [Full Text]  
  • Sheets, A. J., Grass, S. A., Miller, S. E., St. Geme, J. W. III (2008). Identification of a Novel Trimeric Autotransporter Adhesin in the Cryptic Genospecies of Haemophilus. J. Bacteriol. 190: 4313-4320 [Abstract] [Full Text]  
  • Yu, C., Ruiz, T., Lenox, C., Mintz, K. P. (2008). Functional Mapping of an Oligomeric Autotransporter Adhesin of Aggregatibacter actinomycetemcomitans. J. Bacteriol. 190: 3098-3109 [Abstract] [Full Text]  
  • Tang, G., Ruiz, T., Barrantes-Reynolds, R., Mintz, K. P. (2007). Molecular heterogeneity of EmaA, an oligomeric autotransporter adhesin of Aggregatibacter (Actinobacillus) actinomycetemcomitans. Microbiology 153: 2447-2457 [Abstract] [Full Text]  

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
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 arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Cotter, S. E.
Right arrow Articles by St. Geme, J. W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Cotter, S. E.
Right arrow Articles by St. Geme, J. W., III

Copyright © 2009 by the American Society for Microbiology.   For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»