Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Grosskinsky, U. Articles by Linke, D. Search for Related Content PubMed PubMed Citation Articles by Grosskinsky, U. Articles by Linke, D.

 Previous Article  |  Next Article 

Journal of Bacteriology, December 2007, p. 9011-9019, Vol. 189, No. 24
0021-9193/07/$08.00+0     doi:10.1128/JB.00985-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

A Conserved Glycine Residue of Trimeric Autotransporter Domains Plays a Key Role in Yersinia Adhesin A Autotransport

Ulrike Grosskinsky,^1, Monika Schütz,^1, Michaela Fritz,¹ Yvonne Schmid,¹ Marina C. Lamparter,² Pawel Szczesny,^2,3 Andrei N. Lupas,^2* Ingo B. Autenrieth,¹ and Dirk Linke²

Institut für Medizinische Mikrobiologie und Hygiene, Universitätsklinikum Tübingen, Tübingen, Germany,¹ Max Planck Institute for Developmental Biology, Department Protein Evolution, Tübingen, Germany,² Department of Bioinformatics, Institute of Biochemistry and Biophysics, Polish Academy of Sciences, Warsaw, Poland³

Received 21 June 2007/ Accepted 25 September 2007

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The Yersinia adhesin A (YadA) is a trimeric autotransporter adhesin of enteric yersiniae. It consists of three major domains: a head mediating adherence to host cells, a stalk involved in serum resistance, and an anchor that forms a membrane pore and is responsible for the autotransport function. The anchor contains a glycine residue, nearly invariant throughout trimeric autotransporter adhesins, that faces the pore lumen. To address the role of this glycine, we replaced it with polar amino acids of increasing side chain size and expressed wild-type and mutant YadA in Escherichia coli. The mutations did not impair the YadA-mediated adhesion to collagen and to host cells or the host cell cytokine production, but they decreased the expression levels and stability of YadA trimers with increasing side chain size. Likewise, autoagglutination and resistance to serum were decreased in these mutants. We found that the periplasmic protease DegP is involved in the degradation of YadA and that in an E. coli degP deletion strain, mutant versions of YadA were expressed almost to wild-type levels. We conclude that the conserved glycine residue affects both the export and the stability of YadA and consequently some of its putative functions in pathogenesis.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Yersinia enterocolitica and Yersinia pseudotuberculosis are the causative agents of a number of enteric diseases ranging from enterocolitis, acute enteritis, and mesenteric lymphadenitis to autoimmune disorders (2). Infection typically occurs after ingestion of contaminated food or water. The pathogenicity of Yersinia is determined by a number of virulence factors, including the secreted Yersinia outer proteins (Yops) and the outer membrane proteins invasin and Yersinia adhesin A (YadA) (7). YadA is a trimeric autotransporter adhesin (TAA) (26), which has a variety of functions in Yersinia pathogenesis. Other medically important members of this protein family include Moraxella catarrhalis UspA1 and A2 (5), Haemophilus influenzae Hia and Hsf (8, 39), Neisseria meningitidis NadA (3), and Bartonella henselae BadA (34). Electron microscopy studies revealed that YadA forms a lollipop-shaped structure with a head-stalk-anchor architecture on the bacterial surface (16).

The N-terminal head domain is a left-handed beta-roll (30) and is responsible for binding to host cell matrix proteins such as collagen, fibronectin, and laminin and for autoagglutination (12). It is connected to a coiled-coil stalk (25), which serves as a spacer between the head domain and the bacterial outer membrane. This spacing is important for the formation of functional type III secretion needles (29); moreover, the stalk domain confers serum resistance (36), presumably by binding complement factor H (4).

The head and stalk are exported to the bacterial cell surface via the C-terminal transmembrane domain of YadA. This process, called autotransport, is poorly understood. Monomeric autotransporters, such as Neisseria NalP (40), form a 12-stranded beta-barrel at their C-terminal end, through which the passenger domain passes, closing the pore with a helix after exit (31). Trimeric autotransporters form a structurally very similar barrel by trimerization (27, 43), and somehow, all three head and stalk domains need to pass through this narrow pore in order to form the functional trimer at the cell surface, closing it with the C-terminal end of their stalk after exit. Whether periplasmic factors are necessary for the export process is not known, but it is assumed that the folding and trimerization of stalk and head can take place only after completion of the transport process. In the case of the monomeric autotransporter Hbp, the periplasmic chaperone/protease DegP (also named HtrA) (32) degrades the protein when the autotransport is blocked by disulfide bridges (23).

While head and stalk domains are diverse and appear in different combinations, the membrane anchor domains are homologous and display the properties of autotransporters in all TAAs (reviewed in reference 26). Sequence alignments of the membrane anchor regions of trimeric autotransporters revealed a nearly invariant glycine residue (corresponding to G389 of YadA O:8) within this pore-forming domain, which faces the pore lumen (16).

In the present study we examined the role of this glycine (G389 of YadA) in the translocation process and the effects of mutations at G389 on the subsequent YadA-mediated interaction with the host cell by replacing G389 with amino acids of increasing side chain size.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Modeling of the YadA membrane anchor. A homology model of the YadA membrane anchor was built in MODELLER (13), using the crystal structure of the membrane anchor of Haemophilus influenzae Hia (27) as a template. Mutations were introduced into the Hia and YadA models using the program SwissPDB viewer (15). The program VMD (18) was used to visualize the available space for the mutated residues in the two models, in combination with the rendering software POV-Ray (www.povray.org). The dimensions of the cavity facing the conserved glycine were computed by the CASTp server (sts.bioengr.uic.edu/castp/) (10).

P1 transduction. To obtain the strain Escherichia coli BL21(DE3)Omp8 degP, we performed P1 transduction of E. coli BL21(DE3)Omp8 (33). An overnight culture of MC4100 degP::cam (19) was infected with P1 phage in Luria-Bertani (LB) medium (5 mM CaCl₂, 0.2% glucose). After lysis, surviving bacteria were killed with chloroform, and culture supernatants were collected. The resulting P1 lysate was used to infect BL21(DE3)Omp8 cells, and infection was stopped after 20 min by adding 0.1 M sodium citrate buffer (pH 5.5). Cells were harvested by centrifugation and were allowed to recover in LB medium for 1 h. Selection was done on LB agar plates with 33 µg/ml chloramphenicol. Successful deletion of degP was shown by Western blot analysis with an anti-DegP antibody (data not shown). All resistant colonies tested showed a significant increase in mutant YadA expression levels at 27°C compared to BL21(DE3)Omp8.

Bacterial strains and growth conditions. If not otherwise stated, bacterial strains were grown at 27°C in LB broth with 100 µg/ml ampicillin, because the degP deletion strain does not grow well at 37°C. Overnight cultures were resuspended in fresh medium and diluted to an optical density at 600 nm of 0.1. After 2 h of growth, anhydrotetracycline (AHTC) was added to a final concentration of 200 ng/ml. Growth curves of E. coli BL21(DE3)Omp8 and E. coli BL21(DE3)Omp8 degP were comparable during this time interval (data not shown).

DNA constructs. Exchanges of single amino acids in the Y. enterocolitica yadA gene were performed with the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA) using the plasmid pYadAO:8 (37) and were verified by DNA sequencing. The primer pairs used are available upon request. Using the restriction sites SalI and XbaI, the mutated yadA gene was subcloned into the AHTC-inducible expression vector pASK-IBA2 (IBA GmbH, Göttingen, Germany) to obtain the full-length constructs. Mutant versions of the YadA membrane anchor domain were cloned into pASK-IBA2 as described previously (43), using the respective full-length constructs of YadA as templates for PCR.

Purification of YadA and generation of YadA antibodies. YadA protein was purified using E. coli HB101 with the plasmid pYadAO:8 induced with IPTG (isopropyl-β-D-thiogalactopyranoside) (37). Cells were disrupted by French press treatment and incubation with N-lauroylsarcosinate. After centrifugation (20,000 rpm, 90 min; Sorvall SS34), the pellet (containing the bacterial membrane fraction and YadA) was resuspended in water. Approximately 1 mg of protein was subjected to 8% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). YadA could be identified as distinct bands with a molecular mass of 200 kDa (trimer) or 50 kDa (monomer). The trimeric YadA band was excised from a preparative Coomassie blue-stained gel and electroeluted with Laemmli buffer containing 1 M NaCl. The resulting eluate ( 500 µg/ml YadA protein) was used to immunize a rabbit (in accordance with NIH animal welfare assurance no. A5235-01). At 30 days after the first immunization, the rabbit was boosted. Serum was tested for specificity in Western blots and with immunofluorescence studies. For further experiments the immunoglobulin G fraction of the polyclonal serum was purified with protein A-Sepharose columns (Sigma).

Sample preparation for Western blot analysis. Bacterial pellets were lysed in SDS sample buffer (5 x 10⁷ E. coli cells were loaded per lane) and incubated for 10 min at 95°C prior to loading. For the preparation of unheated samples, alkaline lysis of approximately 1.5 x 10⁹ E. coli bacteria was performed using 180 µl of buffer 1 of the peqGOLD plasmid miniprep kit (PEQLAB-Biotechnologie GmbH, Erlangen, Germany) and 20 µl 1 M NaOH. After adjustment to pH 7.0, 1 µl of DNase I incubation buffer (10x) and 2 µl DNase (10 U/ml; Roche Diagnostics GmbH, Mannheim, Germany) were added and left for 20 min before SDS sample buffer was added.

Western blot analysis. After SDS-PAGE, proteins were transferred onto nitrocellulose membranes. The membranes were blocked overnight with phosphate-buffered saline (PBS)-5% milk powder at 4°C. For detection of YadA, a polyclonal rabbit antibody (diluted 1:1,000) and a peroxidase-conjugated secondary anti-rabbit antibody (diluted 1:10,000; Dianova) were added and left for 4 h and 1 h, respectively. Detection of bound antibodies was carried out using the ECL detection kit (Amersham Biosciences).

Stability assays for the isolated YadA membrane anchor. YadA membrane anchor and mutant versions were isolated using membrane fractionation, phase separation, and cation exchange chromatography as described previously (43). For mutated side chain residues larger than serine, no stable trimer could be isolated from outer membrane fractions of BL21(DE3)Omp8. To assay the stability of the protein, samples were diluted to 1 mg/ml and mixed 1:1 with 2x SDS sample buffer. Samples were then heated for the times indicated and applied to an SDS-polyacrylamide gel.

Immunofluorescence microscopy. Bacteria (2 x 10⁷) were spotted on microscopy slides, air dried at 37°C, and fixed for 10 min with 4% paraformaldehyde (PFA). After two washing steps with PBS, the bacteria were incubated with a polyclonal rabbit antibody (immunoglobulin G fraction) directed against YadAO:8 (diluted 1:200) and a Cy2-conjugated secondary anti-rabbit antibody (diluted 1:200; Dianova) in 1% bovine serum albumin in PBS at room temperature for 1 h or 45 min, respectively. Bacterial DNA was stained with 0.01 mg/ml 4',6-diamidino-2-phenylindole (DAPI), and fluorescence images were obtained with a DMRE fluorescence microscope (Leica, Wetzlar, Germany).

Flow cytometry. Bacteria were induced with AHTC to a final concentration of 200 ng/ml. After 15 min of induction, 3 x 10⁸ bacteria were harvested by centrifugation. Cells were washed once with PBS, fixed with 4% PFA, washed again, and resuspended in PBS. Samples were stained with a polyclonal antibody directed against YadA (1:200) and Cy2-conjugated secondary antibody (1:100; Dianova) for 1 h at room temperature and then washed twice with PBS. Surface localization of YadA was measured by flow cytometry on a FACSCalibur (Becton Dickinson), and data were analyzed with WinMDI (J. Trotter) software.

Adhesion to collagen. Coverslips were coated overnight at 4°C with 10 µg/ml human collagen type I (Calbiochem/Merck, Darmstadt, Germany) in PBS. After washing with PBS, 2 x 10⁷ bacteria were centrifuged onto the coverslips at 300 rpm for 5 min. After 1 h of incubation at 37°C, the coverslips were washed three times with PBS and then fixed with 4% PFA. Bacteria were stained with DAPI and counted in three randomly selected fields of view obtained at a magnification of x100 under a fluorescence microscope. Wild-type adhesion was set to 100%.

Adhesion to HeLa cells. HeLa cells (2 x 10⁵) per well were seeded and grown overnight in 24-well cell culture plates (Nunc Life Technologies, Wiesbaden, Germany) on coverslips. Cell monolayers were washed with PBS and incubated for 2 h with fresh RPMI 1640 before the addition of bacteria. Bacteria were resuspended in fresh RPMI 1640, and HeLa cells were infected at a multiplicity of infection of 50. The bacteria were centrifuged onto the cells (300 rpm, 5 min) and incubated at 37°C for 1 h. After one washing step with PBS, the cells with the bacteria were fixed with 4% PFA, stained with fuchsin, and counted under a microscope. Wild-type adhesion was set to 100%.

Cell culture and IL-8 ELISA. Human HeLa cervical epithelial cells (ATCC CCL-2.1) were cultivated in RPMI 1640 medium supplemented with 2 mM glutamine and 10% fetal calf serum. Infection experiments and interleukin-8 (IL-8) enzyme-linked immunosorbent assay (ELISA) were carried out as described previously (37), using a multiplicity of infection of 100. IL-8 concentrations were calculated using recombinant human IL-8 (BD Biosciences Pharmingen) as a standard.

Autoaggregation. E. coli was cultivated as described above. After 2 h of induction with AHTC, bacteria were incubated at room temperature without shaking. Rapid clearance of the medium and aggregated bacteria at the bottom of the tubes were recorded with a digital camera.

Serum resistance. Bacteria were grown at 27°C and induced for 2 h with AHTC. Normal human serum (NHS) from healthy donors was collected, aliquoted, and stored at –80°C. Bacteria (1 x 10⁸) were incubated at 37°C for 15 min in 500 µl 2% NHS in PBS. As a control, heat-inactivated NHS (HIS) was used. Complement function was stopped by addition of 500 µl brain heart infusion broth (1), and tubes were kept on ice. The serum bactericidal effect was calculated as the survival percentage, taking the bacterial counts obtained with bacteria incubated in HIS as 100%. The killing experiment was repeated for each strain at least three times, starting from independent cultures.

Statistics. Representative experiments were performed in triplicate. Differences between mean values were analyzed using Student's t test. A P value of 0.05 was considered statistically significant.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
TAAs reveal a nearly invariant glycine residue in the pore-forming membrane anchor domain. Sequence alignments of YadA and related proteins show a nearly invariant glycine residue (corresponding to G389 of YadA) in the second beta-strand of the pore-forming membrane anchor domain (16) (Fig. 1A). This residue, which faces the channel lumen, is present in more than 95% of TAAs, being replaced by alanine, serine, threonine, or asparagine in only 23 of almost 500 sequences in the nonredundant database at NCBI (four representative sequences are shown at the bottom of Fig. 1A). These mutations seem to have arisen independently from each other, as they appear in very different microbial species whose closest relatives retain the glycine. They show no statistical correlation with residues at other positions of the anchor domain and appear to be compatible with the structural requirements of the beta-barrel. In the Hia anchor structure, the relevant glycine faces an empty space of about 90 ų, large enough to accommodate A, S, C, D, P, N, T, V, E, or Q (in ascending order of side chain volume) without alterations in the conformations of adjacent residues; the same situation is observed in a model of the YadA membrane anchor (Fig. 1B). Histidine is at the limit of the pocket size but can be accommodated with only minor local structural adjustments. The necessity for glycine at this position can thus not be explained from the structure of the anchor domain. This led us to surmise that its primary role is in the autotransport process.

View larger version (77K): [in this window] [in a new window]

FIG. 1. (A) Sequence alignment of the membrane anchor of TAAs. Hia (GI:21536216) and Hsf (GI:15602579) are from Haemophilus influenzae, DsrA (GI:7188575) is from Haemophilus ducreyi, VompA to -D (GI:51949816, GI:51949817, GI:51949813, and GI:51949812, respectively) are from Bartonella quintana, BadA (GI:119890727) is from Bartonella henselae, NadA (GI:83616350) is from Neisseria meningitidis, STEC (GI:73853322) is from Escherichia coli, UspA1 and UspA2 (GI:5453178 and GI:26284393, respectively) are from Moraxella catarrhalis, XadA (GI:7542317) is from Xanthomonas campestris, SadA (GI:10945148) is from Salmonella enterica serovar Typhimurium, YadA (GI:32470319) is from Yersinia enterocolitica O:8, TaaHs (GI:53728655) is from Haemophilus somnus, TaaAs (GI:50084797) is from an Acinetobacter sp., TaaOa (GI:83944938) is from Oceanicaulis alexandrii, and TaaEs (GI:85709253) is from an Erythrobacter sp. The secondary structure assignment is taken from the membrane anchor crystal structure of Hia (27) and is in good agreement with earlier bioinformatic predictions (16). The coiled-coil region (in green) has core positions marked in bold. Residues of the β-strands facing the membrane are highlighted in red. The conserved glycine residue in β-sheet 2 is highlighted in blue and its mutations in violet. (B) Schematic drawing of slices through the models of Hia and YadA membrane anchors, approximately at the level of glycine residues (highlighted in blue). For comparison of the histidine side chain size with the internal cavity around glycine residues, one of the glycines in each protein was mutated to histidine (in violet). Loops connecting β-sheet with coiled coil are highlighted in cyan. Differences in slices are the consequence of different residues building each protein and the local optimization procedures of the modeling software.

Mutation of G389 leads to reduced YadA protein expression, destabilized YadA trimers, and increased degradation of YadA by the periplasmic protease DegP. To investigate the role of the conserved glycine residue G389 in YadA of Y. enterocolitica serotype O:8 (GI:55583865, designated YadAwt), we used site-directed mutagenesis to generate G389A, G389S, G389T, G389N, and G389H versions of the pNIV136-derived plasmid pYadAO:8 (37). These mutants include all four naturally occurring substitutions; we also included histidine in order to explore the upper size limit of the pocket (Fig. 1). Amino acid side chain size increases in the order G < A < S < T < N < H and polarity in the order A < G < S = T < N = H. Expression of YadA from this plasmid resulted in low yields of YadA even for the wild-type construct (data not shown) but was significantly improved by subcloning the coding sequence into the expression vector pASK-IBA2. Expression tests were performed in E. coli BL21(DE3)Omp8, a strain designed for the expression of outer membrane proteins (33). Using Western blot analysis of whole-cell lysates with anti-YadA antibodies, we found that the quantity of expressed full-length YadA was decreased in the serine and alanine mutants and barely detectable in the threonine, asparagine, and histidine mutants (Fig. 2). We decided to modify the expression strain, based on the fact that the periplasmic stress response and especially the periplasmic protease DegP detects and degrades misfolded outer membrane proteins (28). To this end, we transduced the degP deletion from MC4100 degP::cam (19) to BL21(DE3)Omp8 using P1 phage. In E. coli BL21(DE3)Omp8 degP, protein expression was restored almost to wild-type levels for all mutant versions of YadA (Fig. 2A, lower panel). Wild-type YadA (YadAwt) and truncated versions thereof have been demonstrated to form stable trimers which cannot be disrupted by heating in SDS sample buffer (36, 43). This results in a band of approximately 200 kDa for YadA and several bands of intermediate size in Western blots of whole-cell extracts. Comparable denaturation ladders are observed for the homologous B. henselae BadA (35), suggesting that heat stability and incomplete denaturation may be a general feature of TAAs. YadAwt and YadA G389A show such denaturation behavior, and a complete shift from trimer to monomer was observed for YadA G389S in heated samples (Fig. 2A) and for YadA G389T to YadA G389H even in unheated samples (Fig. 2B).

View larger version (40K): [in this window] [in a new window]

FIG. 2. (A and B) Expression of full-length YadAwt and mutant YadA in wild-type (wt) and mutant E. coli (Ec). Whole-cell lysates were made of bacteria grown for 2 h with 200 µg/liter AHTC. Samples were either heated (A) or not heated (B) prior to SDS-PAGE, and YadA was detected by Western blotting. YadA trimer bands are observed at 200 kDa and monomer bands at 50 kDa. (C) Coomassie blue-stained SDS-PAGE of the YadA membrane anchor domain (amino acids 335 to 422). Proteins were solubilized and purified from E. coli outer membrane fractions and heated for the times indicated prior to SDS-PAGE. Gels were stained by colloidal Coomassie blue, because the membrane anchor domain is not detected by the YadA antiserum.

To investigate whether the stability of the isolated membrane anchor domain would be similarly affected by mutations of G389 as in the full-length protein, the membrane anchor domain (amino acids 335 to 422) was expressed and isolated by solubilization with detergents as described previously (43). The membrane anchor could be isolated for the wild-type protein and the YadA alanine and serine mutants but not for the threonine, asparagine, and histidine mutants. In order to compare the stabilities of the YadAwt, YadA G389A, and YadA G389S anchors, we analyzed their denaturation in a heating time course by SDS-PAGE (Fig. 2C). The serine mutant denatured completely after 1 min of heating and the alanine mutant after about 5 min, while most of the wild-type protein remained trimeric even after 1 hour. Some intermediate bands appeared, while the monomer bands did not stain well in SDS-PAGE (data not shown).

Together, replacement of G389 resulted in a reduction of expressed full-length protein with a side chain size larger than serine. Protein levels were restored when the proteins were expressed in an E. coli strain deficient in the periplasmic chaperone protease DegP. Additionally, the heat stability of YadA trimers decreased for side chain sizes larger than alanine. In unheated samples YadAwt and mutant proteins YadA G389A and YadA G389S still revealed trimer formation. The decrease in trimer stability could also be observed with purified membrane anchor domain proteins.

Surface display of YadA is affected by the mutation of G389 and is partially restored in the degP background. To evaluate how the decreased protein stability influences the surface display of the YadA mutants, we performed immunofluorescence microscopy and flow cytometric analysis with a rabbit anti-YadA antibody. Expression of YadAwt in E. coli yielded a typical ring-like staining pattern. Up to a side chain size of serine, the fluorescence signal was comparable to the wild-type situation; for larger side chains, the staining signal was irregular and weak (Fig. 3, upper panels). When YadA was expressed in the E. coli degP mutant strain, the surface display of the mutants was partially restored. We observed fluorescence intensities comparable to that of the wild type for all YadA mutant versions except the histidine mutant. Overall, the fluorescence seemed irregularly distributed on the cell surface (Fig. 3, lower panels). Thus, YadA surface display is obviously influenced by the mutations compared to YadAwt in E. coli, but differences between YadA mutants and between E. coli and E. coli degP strains were difficult to quantify. Therefore we quantified the surface display of YadA by flow cytometry analysis. For this purpose we used the same antibodies as for the immunofluorescence samples. The mean fluorescence intensities of 20,000 single bacteria were recorded for each sample in triplicates of independent cultures. We could show that in E. coli the surface display of YadA mutants is reduced in the order G > A > S > T > N > H (Fig. 4). For the G389H mutant, surface display is 50% of the wild-type level, measured as mean fluorescence of 20,000 bacteria. In E. coli degP, the surface display of the mutant YadA versions is partially restored (to 70% of the wild-type levels for the G389H mutant [Fig. 4, lower panels]). We did not observe periplasmic inclusion bodies for any mutant (electron microscopy data not shown) in both expression strains, suggesting that YadA is either autoexported or degraded by DegP but does not accumulate in the periplasm.

View larger version (41K): [in this window] [in a new window]

FIG. 3. Immunofluorescence microscopy of E. coli expressing wild-type and mutant YadA. The bacteria were grown at 27°C for 15 min with 200 µg/liter AHTC, and YadA on the bacterial surface was detected with rabbit anti-YadA antibodies and Cy-2 secondary antibodies. The upper panel shows YadA surface display in wild-type E. coli (Ec) and the lower panel in E. coli degP. Note the differences in fluorescence signal intensities between YadAwt and mutant YadA when expressed in E. coli or E. coli degP. Bacteria containing the empty vector were used as controls.

View larger version (24K): [in this window] [in a new window]

FIG. 4. Immunostaining and flow cytometry analysis of YadA surface exposure. (A) Mean fluorescence of bacteria expressing YadAwt before (bold black line, no filling) and 10 min after (thin line, gray filling) induction. (B) Bacteria expressing YadAwt (thin line, gray filling) compared to the corresponding histograms of mutated YadA versions (bold black line, no filling). The mean fluorescence of bacteria was set to 100% for YadAwt.

YadA G389 mutants display different autoagglutination properties. YadA promotes autoagglutination of yersiniae. To analyze the impact of G389 mutations on autoagglutination, bacteria were grown in standard LB medium, induced, and observed for aggregation. In our assay, YadAwt and the G389A and G389S mutants settled out of suspension rapidly, but even after 2 h, clarification of bacterial cultures expressing YadA G389T, G389N, or G389H could not be observed (Fig. 5). In some of the experiments, YadA G389A and G389S seemed to be even more efficient in mediating autoagglutination. The results suggest that the ability to promote autoagglutination of YadA expressing bacteria correlates with YadA surface display in a dose-dependent manner.

View larger version (60K): [in this window] [in a new window]

FIG. 5. Autoaggregation of wild-type and mutant YadA. After 2 h of induction, cultures were allowed to settle at room temperature for 2 h. Clearance of the culture is a qualitative measure of autoaggregation. Ec, E. coli.

YadA G389 mutants display the same properties as YadAwt regarding host cell adhesion and proinflammatory host cell responses. In order to address whether and how the disturbed autotransport of YadA may affect functions of YadA relevant for pathogenicity, the interaction of YadA with collagen and host cells was studied. We examined adhesion of E. coli expressing YadAwt and mutant YadA to collagen type I and to epithelial HeLa cells, using bacteria carrying the empty expression vector as a control (Fig. 6). We found that all YadA mutants displayed comparable adhesion to collagen type I and HeLa cells, in spite of the differences in protein levels, surface display, and protein stability shown above. Upon engagement of host cells by YadA-expressing E. coli, host cells have been demonstrated to activate NF-B, which subsequently gives rise to a proinflammatory host cell response including production of, e.g., IL-8 (37). IL-8 secretion by HeLa cells induced by mutant YadA in E. coli reached the same levels as with YadAwt (Fig. 7). From these results we conclude that obviously a low threshold amount of properly folded YadA on the bacterial cell surface is required and sufficient for adhesion to collagen and thus to host cells. This in turn is sufficient to trigger the full IL-8 response by host cells.

View larger version (20K): [in this window] [in a new window]

FIG. 6. Adhesion to collagen type I and HeLa cells mediated by wild-type and mutant YadA expressed in E. coli BL21(DE3)Omp8. Bacteria adherent to collagen type I (A) or HeLa cells (B) were counted, and the adherence of the wild type was set to 100%. The values are means ± standard deviations and are representative of three independent experiments. Bacteria with empty vector were used as controls.

View larger version (16K): [in this window] [in a new window]

FIG. 7. YadA-induced IL-8 secretion of HeLa cells mediated by wild-type and mutant YadA expressed in E. coli BL21(DE3)Omp8 upon infection. Bacteria with empty vector were used as controls. IL-8 levels were determined by ELISA from culture supernatants. The values are means and standard deviations from four independent experiments.

Mutation of G389 impairs YadA-mediated serum resistance. Resistance against killing by serum complement is one of the hallmarks of pathogenic yersiniae and is mediated by YadA and Ail in Y. enterocolitica (1, 4). To compare the serum resistance capabilities of the YadA mutants, E. coli expressing YadAwt or mutant YadA was incubated with 2% human serum, and the viability of the bacteria was determined after 15 min of incubation. Figure 8 shows that YadAwt and the alanine and serine variants protected E. coli cells from complement killing compared to the vector control, while the threonine to histidine variants did not confer serum resistance. No obvious difference was observed between the wild-type and the degP backgrounds, except that the degP strain is generally more susceptible to complement killing even for the empty vector control. The serum resistance drops dramatically when the side chain size is greater then serine for both E. coli strains; this result clearly correlates with the results obtained for autoagglutination (Fig. 5).

View larger version (16K): [in this window] [in a new window]

FIG. 8. Serum resistance of E. coli (Ec) mediated by wild-type and mutant YadA expressed in E. coli BL21(DE3)Omp8. Bacteria with empty vector were used as controls. Equal amounts of bacteria were incubated with NHS or HIS and plated on LB agar. The bacterial survival was calculated as a percentage, taking the bacterial counts obtained with bacteria incubated in HIS as 100%. The data are means ± standard deviations for at least three independent experiments.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
YadA is a prototypical TAA, located on the cell surface and anchored in the outer membrane. It has a highly modular structure with a general head-stalk-anchor architecture, in which different parts account for the various biological functions (6, 12, 36). While the head and stalk domains are diverse among members of the TAA family and vary in length and combinations, the pore-forming membrane anchors are highly conserved and display the properties of autotransporters as shown for YadA, H. influenzae Hia, and N. meningitidis NhhA (reviewed in reference 26). In order to reach the cell surface, TAAs need to pass several transport and quality control steps (Fig. 9). The proteins are expressed in the cytosol and exported by the Sec complex, where their signal peptide is cleaved. In the periplasm, the proteins must stay unfolded, as no cellular machinery is known that could export folded, oligomeric proteins of this size through the outer membrane. This unfolded state is presumably maintained by yet-unidentified chaperones. TAAs then form the trimeric pore that mediates the export of the passenger domains. An involvement of the Omp85 complex, which is responsible for proper insertion of many (possibly all) outer membrane proteins (14, 41), is possible, as in monomeric autotransporters, involvement of Omp85 has been shown (22). The still-unfolded passenger domains then pass the pore, and the only driving force for the autoexport is probably the gain in enthalpy upon folding of the passenger domains into the extremely stable trimer structure on the cell surface.

View larger version (34K): [in this window] [in a new window]

FIG. 9. Autotransport mechanism: after translation (1) the signal peptide is cleaved (2) and the polypeptide is exported via the Sec machinery (3). Insertion via C-terminal signals (4) is followed by the formation of the barrel domain (5). This is the prerequisite for autotransport (6) and folding of the extracellular adhesin domains (7). If autotransport is slowed or blocked, the protein is degraded by DegP (8). OM, outer membrane; IM inner membrane.

In our experiments, we have replaced a nearly invariant glycine residue in the pore region of YadA. We observed a decrease in stability of the transmembrane domain alone and of the complete protein and an increase in degradation by the periplasmic protease DegP, which is part of the periplasmic stress response and is responsible for the degradation of misfolded porins (32). Degradation by DegP has also been shown for monomeric autotransporters that were blocked by artificially introduced disulfide bonds (23). The loss of stability and rate of degradation were clearly correlated with the size of the side chain that we introduced at position 389 and less with its polarity; thus, the major transition in stability was observed between serine and threonine, which corresponds to a change in side chain size but not polarity. In a DegP-deficient strain, mutant YadA reached the cell surface even with this histidine mutation, showing that G389 affects primarily the rate of export, which determines the ratio of folded to degraded protein in the kinetic partitioning process of the quality control machinery. Even though displayed at reduced levels compared to the wild type, all mutant forms of YadA still conferred adhesion to extracellular matrix (ECM) proteins and triggered intracellular signaling, as judged by monitoring host cell cytokine production. Enteropathogenic yersiniae express two major outer membrane adhesins, monomeric invasin and YadA (sometimes referred to as "trimeric invasin" in earlier studies), both of which contribute to pathogenesis. Whereas invasin appears to bind directly to β1 integrin receptors on the host cell surface (21, 42), YadA binds indirectly through various ECM components (9, 11, 24). In general, binding to β1 integrin receptors leads to integrin clustering and to the onset of a signaling cascade that activates NF-B and, e.g., IL-8 production. Prerequisite for robust and persistent integrin signaling is a high substrate density on either the host cell surface (i.e., ECM proteins) or the bacterial surface (i.e., adhesins). Activation via the monomeric invasin requires the engagement of several receptors at the same time to induce integrin clustering and thus signaling. Therefore, it is strictly dose dependent and requires high concentrations of invasin (17, 20, 38). In contrast, YadA as a trimeric protein provides three ECM binding pockets with each complex, analogous to the Haemophilus adhesin Hia (44); moreover, it is capable of binding to various ligands. ECM proteins are organized in a complex network and may additionally multiply the YadA binding capacity. Therefore, even a low quantity of YadA protein could provide enough binding motifs for ECM proteins, and consequently, even YadA G389H would be present in sufficient quantity on the bacterial surface to trigger adhesion and induction of IL-8 production. This is in fact what we found for E. coli expressing mutant YadA proteins with impaired surface display. Although all YadA mutants mediated adhesion and IL-8 production, some conferred reduced protection against complement factors and showed a reduced ability to autoaggregate. Presumably, both deficiencies are due to the lower levels of YadA displayed in the mutants, although we cannot rule out secondary detrimental effects caused by the cellular stress response which is induced when mutant proteins accumulate in the periplasm. Autoaggregation is mediated by YadA-YadA interactions, and stable aggregates, which settle out of suspension, can arise only if there is an appropriate quantity of protein on the bacterial surface. The exact molecular mechanism of how YadA confers complement resistance still remains to be elucidated, but it appears to be located primarily in the C-terminal part of the stalk (36) and is likely to be coupled to surface protein levels as well.

In conclusion, our data show that autotransport is possible in YadA mutants lacking the highly conserved G389, but that the mutants have a reduced stability and export rate. The severity of this deficiency correlates better with the size than with the polarity of the side chain introduced at position 389. The mutant proteins may accumulate in the periplasm and may activate the periplasmic stress response, triggering degradation by the chaperone protease DegP. At reduced levels of surface display, the mutant proteins still mediate adhesion and IL-8-production but are deficient in autoaggregation and serum resistance. We have thus characterized a set of mutant proteins in vitro that display different levels of YadA-mediated effects. As YadA-deficient Y. enterocolitica strains are avirulent in conventional mouse strains, YadA obviously has a decisive role in virulence. Therefore, future studies in our laboratory will focus on understanding the actual role of YadA in pathogenicity in vivo. The implications of YadAwt and G389 mutants for adhesion, invasion, signaling, and survival in or without the context of invasin will be examined in Y. enterocolitica with the experimental mouse infection model.

 
We thank Alexander Diemand for help with modeling software and Heinz Schwarz for YadA antiserum production. E. coli BL21(DE3)Omp8 was a gift from Ralf Koebnik, CNRS Montpellier, and E. coli MC4100 degP::cam was a gift from Thomas Silhavy, Princeton University. MS analysis was kindly performed by Stefan Stevanovic and Florian Altenberend, University of Tübingen. Anti-DegP antiserum was a gift from Joanna Skórko-Glonek, University of Gdansk.

This work was supported by SFB766 of the DFG (German Science Foundation).

 
* Corresponding author. Mailing address: Max Planck Institut für Entwicklungsbiologie, Abteilung Proteinevolution, Spemannstr. 35, 72076 Tübingen, Germany. Phone: 49-7071-601341. Fax: 49-7071-601349. E-mail: andrei.lupas{at}tuebingen.mpg.de

Published ahead of print on 5 October 2007.

These authors contributed equally to this study.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1
1. Biedzka-Sarek, M., R. Venho, and M. Skurnik. 2005. Role of YadA, Ail, and lipopolysaccharide in serum resistance of Yersinia enterocolitica serotype O:3. Infect. Immun. 73:2232-2244.[Abstract/Free Full Text]
2
2. Bottone, E. J. 1997. Yersinia enterocolitica: the charisma continues. Clin. Microbiol. Rev. 10:257-276.[Abstract]
3
3. Capecchi, B., J. Adu-Bobie, F. Di Marcello, L. Ciucchi, V. Masignani, A. Taddei, R. Rappuoli, M. Pizza, and B. Arico. 2005. Neisseria meningitidis NadA is a new invasin which promotes bacterial adhesion to and penetration into human epithelial cells. Mol. Microbiol. 55:687-698.[CrossRef][Medline]
4
4. China, B., M. P. Sory, B. T. N'Guyen, B. M. de, and G. R. Cornelis. 1993. Role of the YadA protein in prevention of opsonization of Yersinia enterocolitica by C3b molecules. Infect. Immun. 61:3129-3136.[Abstract/Free Full Text]
5
5. 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]
6
6. Cornelis, G. R. 2000. Molecular and cell biology aspects of plague. Proc. Natl. Acad. Sci. USA 97:8778-8783.[Abstract/Free Full Text]
7
7. Cornelis, G. R. 1994. Yersinia pathogenicity factors. Curr. Top. Microbiol. Immunol. 192:243-263.[Medline]
8
8. Cotter, S. E., H. J. Yeo, T. Juehne, and J. W. St Geme, 3rd. 2005. Architecture and adhesive activity of the Haemophilus influenzae Hsf adhesin. J. Bacteriol. 187:4656-4664.[Abstract/Free Full Text]
9
9. Dersch, P., and R. R. Isberg. 1999. A region of the Yersinia pseudotuberculosis invasin protein enhances integrin-mediated uptake into mammalian cells and promotes self-association. EMBO J. 18:1199-1213.[CrossRef][Medline]
10
10. Dundas, J., Z. Ouyang, J. Tseng, A. Binkowski, Y. Turpaz, and J. Liang. 2006. CASTp: computed atlas of surface topography of proteins with structural and topographical mapping of functionally annotated residues. Nucleic Acids Res. 34:W116-W118.[Abstract/Free Full Text]
11
11. Eitel, J., and P. Dersch. 2002. The YadA protein of Yersinia pseudotuberculosis mediates high-efficiency uptake into human cells under environmental conditions in which invasin is repressed. Infect. Immun. 70:4880-4891.[Abstract/Free Full Text]
12
12. el Tahir, Y., and M. Skurnik. 2001. YadA, the multifaceted Yersinia adhesin. Int. J. Med. Microbiol. 291:209-218.[CrossRef][Medline]
13
13. Fiser, A., and A. Sali. 2003. Modeller: generation and refinement of homology-based protein structure models. Methods Enzymol. 374:461-491.[Medline]
14
14. Gentle, I., K. Gabriel, P. Beech, R. Waller, and T. Lithgow. 2004. The Omp85 family of proteins is essential for outer membrane biogenesis in mitochondria and bacteria. J. Cell Biol. 164:19-24.[Abstract/Free Full Text]
15
15. Guex, N., and M. C. Peitsch. 1997. SWISS-MODEL and the Swiss-PdbViewer: an environment for comparative protein modeling. Electrophoresis 18:2714-2723.[CrossRef][Medline]
16
16. Hoiczyk, E., A. Roggenkamp, M. Reichenbecher, A. Lupas, and J. Heesemann. 2000. Structure and sequence analysis of Yersinia YadA and Moraxella UspAs reveal a novel class of adhesins. EMBO J. 19:5989-5999.[CrossRef][Medline]
17
17. Hudson, K. J., J. B. Bliska, and A. H. Bouton. 2005. Distinct mechanisms of integrin binding by Yersinia pseudotuberculosis adhesins determine the phagocytic response of host macrophages. Cell. Microbiol. 7:1474-1489.[CrossRef][Medline]
18
18. Humphrey, W., A. Dalke, and K. Schulten. 1996. VMD: visual molecular dynamics. J. Mol. Graph. 14:33-38.[CrossRef][Medline]
19
19. Isaac, D. D., J. S. Pinkner, S. J. Hultgren, and T. J. Silhavy. 2005. The extracytoplasmic adaptor protein CpxP is degraded with substrate by DegP. Proc. Natl. Acad. Sci. USA 102:17775-17779.[Abstract/Free Full Text]
20
20. Isberg, R. R., and P. Barnes. 2001. Subversion of integrins by enteropathogenic Yersinia. J. Cell Sci. 114:21-28.[Abstract]
21
21. Isberg, R. R., Z. Hamburger, and P. Dersch. 2000. Signaling and invasin-promoted uptake via integrin receptors. Microbes Infect. 2:793-801.[CrossRef][Medline]
22
22. Jain, S., and M. B. Goldberg. 2007. Requirement for YaeT in the outer membrane assembly of autotransporter proteins. J. Bacteriol. 189:5393-5398.[Abstract/Free Full Text]
23
23. Jong, W. S., C. M. ten Hagen-Jongman, T. den Blaauwen, D. Jan Slotboom, J. R. Tame, D. Wickstrom, J. W. de Gier, B. R. Otto, and J. Luirink. 2007. Limited tolerance towards folded elements during secretion of the autotransporter Hbp. Mol. Microbiol. 63:1524-1536.[CrossRef][Medline]
24
24. Kapperud, G., E. Namork, M. Skurnik, and T. Nesbakken. 1987. Plasmid-mediated surface fibrillae of Yersinia pseudotuberculosis and Yersinia enterocolitica: relationship to the outer membrane protein YOP1 and possible importance for pathogenesis. Infect. Immun. 55:2247-2254.[Abstract/Free Full Text]
25
25. Koretke, K. K., P. Szczesny, M. Gruber, and A. N. Lupas. 2006. Model structure of the prototypical non-fimbrial adhesin YadA of Yersinia enterocolitica. J. Struct. Biol. 155:154-161.[CrossRef][Medline]
26
26. Linke, D., T. Riess, I. B. Autenrieth, A. Lupas, and V. A. Kempf. 2006. Trimeric autotransporter adhesins: variable structure, common function. Trends Microbiol. 14:264-270.[CrossRef][Medline]
27
27. Meng, G., N. K. Surana, J. W. St. Geme III, and G. Waksman. 2006. Structure of the outer membrane translocator domain of the Haemophilus influenzae Hia trimeric autotransporter. EMBO J. 25:2297-2304.[CrossRef][Medline]
28
28. Misra, R., M. Castillo-Keller, and M. Deng. 2000. Overexpression of protease-deficient DegP(S210A) rescues the lethal phenotype of Escherichia coli OmpF assembly mutants in a degP background. J. Bacteriol. 182:4882-4888.[Abstract/Free Full Text]
29
29. Mota, L. J., L. Journet, I. Sorg, C. Agrain, and G. R. Cornelis. 2005. Bacterial injectisomes: needle length does matter. Science 307:1278.[Abstract/Free Full Text]
30
30. 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]
31
31. Oomen, C. J., P. van Ulsen, P. van Gelder, M. Feijen, J. Tommassen, and P. Gros. 2004. Structure of the translocator domain of a bacterial autotransporter. EMBO J. 23:1257-1266.[CrossRef][Medline]
32
32. Pallen, M. J., and B. W. Wren. 1997. The HtrA family of serine proteases. Mol. Microbiol. 26:209-221.[CrossRef][Medline]
33
33. Prilipov, A., P. S. Phale, P. van Gelder, J. P. Rosenbusch, and R. Koebnik. 1998. Coupling site-directed mutagenesis with high-level expression: large scale production of mutant porins from E. coli. FEMS Microbiol. Lett. 163:65-72.[CrossRef][Medline]
34
34. Riess, T., S. G. Andersson, A. Lupas, M. Schaller, A. Schafer, P. Kyme, J. Martin, J. H. Walzlein, U. Ehehalt, H. Lindroos, M. Schirle, A. Nordheim, I. B. Autenrieth, and V. A. Kempf. 2004. Bartonella adhesin a mediates a proangiogenic host cell response. J. Exp. Med. 200:1267-1278.[Abstract/Free Full Text]
35
35. Riess, T., G. Raddatz, D. Linke, A. Schafer, and V. A. Kempf. 2007. Analysis of Bartonella adhesin A expression reveals differences between various B. henselae strains. Infect. Immun. 75:35-43.[Abstract/Free Full Text]
36
36. 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]
37
37. Schmid, Y., G. A. Grassl, O. T. Buhler, M. Skurnik, I. B. Autenrieth, and E. Bohn. 2004. Yersinia enterocolitica adhesin A induces production of interleukin-8 in epithelial cells. Infect. Immun. 72:6780-6789.[Abstract/Free Full Text]
38
38. Schulte, R., S. Kerneis, S. Klinke, H. Bartels, S. Preger, J. P. Kraehenbuhl, E. Pringault, and I. B. Autenrieth. 2000. Translocation of Yersinia entrocolitica across reconstituted intestinal epithelial monolayers is triggered by Yersinia invasin binding to beta1 integrins apically expressed on M-like cells. Cell. Microbiol. 2:173-185.[CrossRef][Medline]
39
39. St. Geme, J. W., III, and D. Cutter. 2000. The Haemophilus influenzae Hia adhesin is an autotransporter protein that remains uncleaved at the C terminus and fully cell associated. J. Bacteriol. 182:6005-6013.[Abstract/Free Full Text]
40
40. van Ulsen, P., L. van Alphen, J. ten Hove, F. Fransen, P. van der Ley, and J. Tommassen. 2003. A neisserial autotransporter NalP modulating the processing of other autotransporters. Mol. Microbiol. 50:1017-1030.[CrossRef][Medline]
41
41. Voulhoux, R., and J. Tommassen. 2004. Omp85, an evolutionarily conserved bacterial protein involved in outer-membrane-protein assembly. Res. Microbiol. 155:129-135.[Medline]
42
42. Wiedemann, A., S. Linder, G. Grassl, M. Albert, I. Autenrieth, and M. Aepfelbacher. 2001. Yersinia enterocolitica invasin triggers phagocytosis via beta1 integrins, CDC42Hs and WASp in macrophages. Cell. Microbiol. 3:693-702.[CrossRef][Medline]
43
43. Wollmann, P., K. Zeth, A. N. Lupas, and D. Linke. 2006. Purification of the YadA membrane anchor for secondary structure analysis and crystallization. Int. J. Biol. Macromol. 39:3-9.[CrossRef][Medline]
44
44. Yeo, H. J., S. E. Cotter, S. Laarmann, T. Juehne, J. W. St. Geme III, 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, December 2007, p. 9011-9019, Vol. 189, No. 24
0021-9193/07/$08.00+0     doi:10.1128/JB.00985-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

This article has been cited by other articles:

• 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]  

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Grosskinsky, U. Articles by Linke, D. Search for Related Content PubMed PubMed Citation Articles by Grosskinsky, U. Articles by Linke, D.