• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Jun 16, 2009; 106(24): 9661–9666.
Published online May 28, 2009. doi:  10.1073/pnas.0812900106
PMCID: PMC2701029

IpaB–IpgC interaction defines binding motif for type III secretion translocator


The delivery of virulence factors into host cells through type III secretion systems is essential for enterobacterial pathogenesis. Molecular chaperones bind specifically to virulence factors in the bacterial cytosol before secretion. Invasion plasmid gene C (IpgC) is a chaperone that binds 2 essential virulence factors of Shigella: invasion plasmid antigens (Ipa) B and C. Here, we report the crystal structure of IpgC alone and in complex with the chaperone binding domain (CBD) of IpaB. The chaperone captures the CBD in an extended conformation that is stabilized by conserved residues lining the cleft. Analysis of the cocrystal structure reveals a sequence motif that is functional in the IpaB translocator class from different bacteria as determined by isothermal titration calorimetry. Our results show how translocators are chaperoned and may allow the design of inhibitors of enterobacterial diseases.

Keywords: asymmetric homodimer, chaperone, microbiology, pathogens, tetratricopeptide repeat

Bacillary dysentery caused by Shigella species presents a global human health threat. The Shigella host invasion involves penetration into epithelial cells, intracellular multiplication, and spreading to adjacent cells (1), altogether eliciting strong inflammation (2). The process of host invasion depends on delivery of virulence factors by a type III secretion (TTS) system (TTSS) encoded on a 220-kb plasmid (3). Specifically a 31-kb “entry region” of the virulence plasmid is sufficient to mediate Shigella infection (4). The entry region comprising mxi-spa [membrane expression of invasion plasmid antigens (Ipas)-surface presentation of antigens] operon encodes the TTS apparatus. In Shigella the genes ipgC, ipaB, ipaC, and ipaD present on the ipa operon of the entry region are essential for host cell entry (5, 6). Both IpaB (62 kDa) and IpaC (39 kDa) are essential for invasion of epithelial cells, membrane lysis of the phagocytic vacuole, contact hemolysis, and macrophage cell death (5, 7, 8). The small cytosolic chaperone invasion plasmid gene C (IpgC) (18 kDa) binds independently to IpaB and IpaC and is crucial for Shigella invasion (8). The stabilization of IpaB and IpaC in the cytoplasm of Shigella prevents their premature association and degradation and depends on their association with IpgC (8). Besides stabilizing IpaB and IpaC, IpgC coactivates transcription of effectors regulated by TTSS activity (9).

Chaperones associated with TTSSs possess low molecular mass, have acidic pI, and usually are located adjacent, in the operon, to the gene of the virulence factor they bind (10). The requirement of cytoplasmic molecular chaperones for some virulence factors is characteristic of TTSSs. These chaperones have little sequence similarity but they can be clustered into 3 classes according to the substrate they bind (11). Class I chaperones bind to either one (class IA) or multiple (class IB) effectors. To date, many of crystal structures solved from various TTSS chaperones belong to this class. Despite the lack of sequence identity, the structures reveal a dimeric organization with very similar α/β folds. Interestingly, the structures with the chaperone binding domains (CBDs) of their cognate effectors show mostly nonglobular substrate organization around the surface of the chaperone (1214). Flagellar chaperones grouped as class III differ from other TTS chaperones structurally and in binding to the C termini of their cognate substrates. However, the surface stabilization of secondary structure elements of the substrate may reflect a mode of recognition similar to class I chaperones (15). Class II chaperones are unique because they bind to at least 2 tranlocators that might associate to the eukaryotic membrane to form a pore that allows translocation of other effectors (16). Sequence analysis of class II chaperones predict the presence of tetratricopeptide repeat (TPR) (17) motifs that are known to be involved in protein–protein interactions.

The crystal structure of class II chaperone from Yersinia, SycD, was recently determined (18). The truncated SycD structure corroborates the presence of tandem TPRs (17). However, although SycD has been shown to dimerize in solution (19), the structural arrangement in the crystal reveals 3 possible dimers (18). Moreover, how the substrates associate with the class II chaperone is unclear.

In this study, we investigated the crystal structure and function of IpgC from Shigella flexneri. We identified the crucial role of N-terminal 21 amino acids in the dimerization of IpgC and defined the N termini of the CBD in IpaB. The structure of IpgC with the CBD allowed us to identify a sequence motif in substrates recognized by this class of chaperones. This sequence motif was extended for other translocators of the IpaB class.


Crystallization of a Functional IpgC.

To understand the molecular basis of IpgC function, we determined the crystal structure of apo IpgC by multiwavelength anomalous diffraction (MAD) using selenomethionine (SeMet)-substituted protein (Table S1). We deleted 4 residues from the C-terminal end (IpgCct), of the total 155 aa to obtain good-quality crystals. Both C-terminal deletion (ipgCct) or WT (ipgCwt) ipgC complemented a nonpolar ipgC deletion mutant (ΔipgC) (8). ΔipgC complemented with ipgCwt or ipgCct but not with empty vector produced and secreted in exponentially growing cells as analyzed by immunoblotting (Fig. 1A). Furthermore, HeLa cell invasion (Fig. 1B) and macrophage cytotoxicity (Fig. 1C) were similar in S. flexneri WT strain and ΔipgC complemented with either ipgCct or ipgCwt. These data demonstrate that ipgCct is fully functional.

Fig. 1.
C-terminal truncated ipgC (ipgCct) is biologically active. (A) Expression and secretion of IpaB and IpgC were analyzed by immunoblot. DnaK was used as an indicator for bacterial membrane integrity and as a loading control. IpaB was detected in bacterial ...

IpgC Is an All α-Helical Protein with TPR Motifs.

The crystallographic asymmetric unit contains 2 copies of IpgC (Fig. 2A). We traced the model inside the electron density of residues 9–151 for each copy independently. The protein shows an all-α structure with 8 helices per molecule labeled H1–H8. Helices H2–H7 fold into 3 TPR motifs, and H8 defines the first helix of a fourth motif. Together, helices H2–H8 adopt a superhelical scaffold creating a concave surface, henceforth referred to as cleft (Fig. 2A). The structure of the 2 copies, designated subunit (SU) A and B, are similar in the TPR moiety (residues 36–151) with a rmsd of 0.7 Å.

Fig. 2.
IpgC forms a dimer in the crystal. (A) Ribbon representation of the apo IpgC dimer found in the asymmetric unit with SUs colored brown and light blue. Helices of either SU A or B are labeled as described. (B) Hydrophobic interaction between SU A and B ...

Structural studies published recently on class I chaperones (reviewed in ref. 20) show that oligomerization allows efficient substrate binding. For class II chaperones that bind to at least 2 substrates and are characterized by the presence of TPR motifs, no such structural data on substrate binding exists. In IpgC, which belongs to class II, ≈16% of the calculated accessible surface area of the 2 molecules found in the asymmetric unit (2,764/17,580 Å2) is involved in dimer contacts. These asymmetric contacts involve the amphipatic H1(A) that is stabilized nearly throughout its entire length by a hydrophobic interface provided by H1, H3, H4, and H5 of chain B (Fig. 2 A and B). Additionally, 3 hydrogen bonds anchor the N- and C- terminal ends of H1(A) (Glu-9, Ala-18, and Ser-21 in Fig. 2C) to the second TPR (H4, H5) of chain B. Several other interactions at the interface between both chains formed by the first (H3) and the second TPR (H4, H5) of SU A and the first TPR (H2, H3) of SU B (Fig. 2 B and C) complement the extensive intermolecular contact.

TPR motifs are used by various macromolecular complexes for protein–protein interactions and comprise degenerate 34-aa antiparallel α-helical repeats (21). An alignment of the sequence of IpgC homologues from several Gram-negative pathogens validates the preference for a small amino acid at positions 8, 20, and 27 of each TPR motif (22) (Fig. S1A). Many conserved residues are located on the concave side of the TPR moiety (Fig. S1B) and form a putative peptide binding cleft each composed of the first helices of the TPR, i.e., H2, H4, H6, and H8 (Fig. 2D).

Helix H2 forms the border of the cleft on 1 side and contains a tyrosine ladder formed by residues 40, 44, and 47 with their hydroxyl groups pointing into the cleft. Opposite to H2, H8 exposes residues Lys-138, Lys-142, and Tyr-146 toward the cleft (Fig. 2D). The cleft at its narrowest end, between H2 and H8, is 10.5 Å wide. This narrow-end groove can hardly accommodate an entire helix as proposed (22). Interestingly, H1(A), H5(B), H6(B), and H7(B), on the convex side of the cleft on monomer B, form a shallow groove that could have functional significance (Fig. S2A).

TTSS-associated chaperones usually have an acidic isoelectric point. Interestingly, there is a patch with negative electrostatic potential at 1 end of the cleft (Fig. S2B) contributed by 8 (at positions 27, 33, 34, 37, 38, 71, 103, and 136), mostly conserved, aspartates. A similar “aspartic acid-array” in a TPR containing protein, YrrB (23) from Bacillus subtilis, is proposed to also mediate distinct substrate interaction.

Asymmetric Dimerization: The Role of H1.

The consequence of the asymmetrically-organized interactions mentioned before is that H1 and the loop connecting it to H2 show different arrangement in both SUs. The superimposition of the TPR domains of both SUs clearly show that the N-terminal 32 aa face away from each other (Fig. 2D). In fact, the connecting loop, whose termini are defined by the conserved residues Gly-22 and Pro-32, provides the flexibility required for the 2 H1 conformations. We confirmed that IpgC is a dimer in solution by multiangle laser light scattering (MALLS) (Fig. S3A). Moreover, purified N-terminal 21-aa truncated IpgC (IpgCΔ21) forms aggregates of high molecular mass (Fig. S4A) and cannot complement ΔIpgC in HeLa cell invasion assay (Fig. S3B). This shows that the interactions mediated by H1 are required for a functional dimer. Additionally, we verified that the dimer arrangement found in the crystal is the same as in solution by introducing a dimer-disrupting double mutation: Ala94Glu/Val95Gln (IpgCdm). Because of the asymmetric nature of the dimer, these amino acids are either involved in hydrophobic contacts with the interface formed by H2 and H3 of the adjacent SU or with the hydrophobic surface of H1 (Fig. S3C). Substitution of nonpolar by polar or even negatively-charged residues may disrupt dimerization of IpgC. Indeed the double mutant, IpgCdm was detected as monomer in solution analyzed by MALLS (Fig. S3D). ipgCdm complemented ΔipgC showed abrogated IpaB production and secretion (Fig. 2E) and inefficient HeLa cell invasion (Fig. 2F), corroborating the physiological requirement of dimerization. Therefore, because of different dimer assemblies in crystal packing observed for an amino terminal-truncated SycD (18), the Yersinia homologue of IpgC could be caused by lack of H1 (Fig. S5). Indeed, it has been shown that SycD (18) needs to dimerize to secrete its substrates. Furthermore, SicA, the Salmonella homologue, also dimerizes under physiological conditions (Fig. S4B). These data show that dimer formation is essential for class II chaperone function and depends on the presence of N-terminal 20 aa.

IpaB Is Bound in an Extended Conformation.

To understand the chaperone function of IpgC, we solved the crystal structure of IpgC complexed with the CBD of IpaB encompassing residues 51–72 (Table S2). To map the IpgC binding region, we performed proteolytic digestion on copurified complex of IpaB/IpgC (Fig. 3A). Thermolysin treatment of purified complex yielded a stable core of IpaB comprising residues 51–507, identified by N-terminal sequencing and mass spectrometry. This information combined with the yeast 2-hybrid analysis (24) defines the CBD in IpaB to be from residues 51 to 72. The crystallographic asymmetric unit contains a dimer similar to the apo structure with 1 IpaB peptide bound to each SU. The corresponding SU of apo and IpaB51–72 complexed IpgC are superimposable in the TPR region (rmsd 0.5 Å). One difference, however, is that the SUs in the complex are tilted by ≈15° and shifted by 5 Å relative to each other. It is unclear whether this difference is the result exclusively of substrate binding or is influenced by crystal packing.

Fig. 3.
Defining CBD in IpaB and its association with IpgC. (A) Limited proteolysis on purified IpaB/IpgC complex. Proteolysis experiments were conducted with thermolysin on ice and at 25 °C. The enzyme/substrate ratio was 1:100 (by weight). Aliquots ...

The cleft is the substrate binding surface and supports an extended conformation of the interacting peptide (Fig. 3B). The conformation of the IpaB peptide is defined by the electron density (2FoFc) in the cleft of the IpgC monomer A (from Leu-63 to Ser-72) and B (from Ser-60 to Ser-72) (Fig. 3C). Both peptides have an antiparallel orientation to H2 with a solvent-accessible surface area of 605 Å2 buried between the peptide and SU B. The location and orientation of the IpaB peptide are related to that of human Hop in complex with C-terminal peptides of chaperones Hsp70 and Hsp90 (25), the only other available structure of TPR-containing complex. Unlike this eukaryotic complex, where the bound peptide's free C terminus is stabilized by a conserved 2-carboxylate clamp, the residues following the CBD fold into a stable domain, as evidenced by restricted access to thermolysin in IpaB/IpgC complex beyond Ile-51 on IpaB (Fig. 3A).

Extensive Interactions Mediate Substrate-Chaperone Binding.

The cocrystal structure shows that there are 3 types of interactions that allow the binding between the IpaB peptide and IpgC (Fig. 4 A and B). First, there are 3 salt bridges: (i) between the side chain of amino acids Lys-68 in IpaB51–72 and Asp-71 in IpgC, (ii) Glu-66 in IpaB51–72 and Lys-142 in SU A, and (iii) Lys-71 in IpaB51–72 and Asp-37 in SU B. Second, there are strong, but not sequence-specific, hydrogen bonds between the carbonyl of Ile-62, Pro-65, Lys-68 in the IpaB peptide and the amide of Gln-112, hydroxyl of Tyr-47 and Tyr-40 in IpgC. Third, there are hydrophobic and van der Waals interactions between IpaB51–72 and 3 discrete pockets on the surface of IpgC (Fig. 4B). The pockets, designated P1, P2, and P3, interact with side chains of conserved residues Pro-65, Leu-67 and Pro-70 in the IpaB peptide. P1 is formed by conserved residues Tyr-47, Ala-78, Gln-81, Tyr-93, His-109, and partially by Ile-82. The predominantly hydrophobic central pocket, P2, is formed exclusively by conserved residues Tyr-44, Tyr-47, Phe-59, Met-74, Gly-75 and Ala-78 from H2 and H4. Interestingly, protein binding is also mediated through pockets positioned similarly to P1 and P2 in the eukaryotic protein Hop (25). The P3 pocket is formed by conserved residues Tyr-40 and Tyr-44 of the tyrosine ladder supported by Ser-41, partially by Asp-45 from H2 and Asp-37 in SU A. The spatial distribution of P1, P2, and P3 and the residues forming these pockets are conserved in SycD (Table S3). Indeed, mutations in the predicted pockets of LcrH, IpgC homologue from Yersinia pseudotuberculosis, affect binding and/or secretion of its substrates (26). Together, these observations indicate that the spatial distribution of P1, P2, and P3 is a key feature in TTSS class II chaperones.

Fig. 4.
Interaction between the CBD of IpaB and IpgC and substrate motif for class II chaperones. (A) Diagram representation of IpgC SU B and a yellow ribbon model for the CBD. The residues involved in intermolecular interactions are shown. The H-bonds and 1 ...

A Conserved Chaperone Binding Motif.

Based on the IpaB/IpgC structure and considering the sequences of homologous substrates, we identified a chaperone binding motif in TTSS translocon proteins of the IpaB class. This motif (Fig. 4C), in agreement with other observations (20), is located within the 100 N-terminal residues of TPR-chaperone substrates. The motif consists of a conserved Pro or Val occupying P1, while conserved Leu and Pro occupy P2 and P3, respectively. An amino acid, preferentially with a negative charge, is sandwiched between P1 and P2. A basic residue and a nonpolar amino acid make a bridge between P2 and P3 in the substrate. IpaC, another IpgC substrate, does not contain this motif.

The peptide PELKAP, representing the chaperone binding sequence motif in IpaB (residues 65–70), bound to IpgC with a Kd of 625 ± 11 μM as measured by isothermal titration calorimetry (ITC) (Fig. 4D). The interaction of peptides containing an unfavorable amino acid in any of the pocket-interacting positions (RELKAP, PENKAP or PELKAD) with IpgC was too weak to allow for reliable quantification (Fig. S6). Even the disruption of the salt bridge when substituting the Lys in PELAAP reduced the affinity for IpgC that was too low to quantify (Fig. S6). The IpaB51–72 peptide, used in the cocrystallization, bound to IpgC with a Kd of 72 ± 8 μM (Fig. S7). Conversely, IpgCpb (S41M/Y44G/G75Q/A78M) in which the peptide-binding site is disrupted showed no binding to the peptide PELKAP (Fig. 4D) and cannot complement ΔipgC in secretion and HeLa cell invasion assays (Fig. 4 E and F). These results clearly demonstrate that the interaction indeed occurs at the binding site and not elsewhere on the protein. Notably, peptides encoding the chaperone binding motif of YopB (VQLPAP) and YopD (PELIKP) bound to SycD with a Kd of 455 ± 19 μM and 1,370 ± 38 μM, respectively (Fig. S7).


Here, we report the crystal structures of a virulence-associated class II chaperone, IpgC alone and in complex with 1 of its substrates, IpaB. IpgC reveals 8 helices and folds into an all-α-helical structure with consecutive TPR motifs forming a cleft-like scaffold. The chaperone IpgC forms a dimer in both the crystals and solution. A notable feature of the IpgC is the asymmetric organization of the SUs in the dimer in both the apo and complexed structure. The TPR moieties of the SUs overlap with a rmsd of 0.7 Å but exhibit different arrangement of H1. H1 involvement in breaking the symmetry of the dimer could explain its crucial role in the functional dimerization of IpgC. Deletion of H1 led to formation of soluble aggregates in heterologously-expressed IpgC and in vivo rendered Shigella noninvasive. Further, dimerization for a functional chaperone was demonstrated in vivo for a Yersinia homologue (18). We also show dimerization of SicA from Salmonella. This emphasizes dimerization property of this class of chaperones for proper function. Interestingly, Birket et al. (27) found a 1:1 stoichiometry for the IpaB/IpgC complex, even though experiments with the IpgCdm indicate that the chaperone needs to dimerize. It is possible that the heterodimers observed by Birket et al. could be caused by binding of full-length IpaB to IpgC. It will be interesting to address whether changes in oligomerization have functional significance. Thus, the architecture and specificity of interactions between the SUs found in the crystal and dimer identified in solution indicates that this is the biologically active form.

We show in this class of chaperones that the major cleft is the substrate binding surface. The binding interface in the cocrystal structure between IpaB and IpgC is substantial, with solvent-accessible surface area of 605 Å2 buried between IpaB peptide and SU B. Interestingly, the cleft facilitates association with the interacting polypeptide by providing an amphipathic surface. The surface provided by the cleft, which is 10 Å wide at its narrowest end, can barely accommodate a helical peptide. Indeed, the chaperone binds the IpaB CBD in an extended conformation and is stabilized by conserved residues lining the cleft. The IpaB binds in an antiparallel orientation to the IpgC helix 2 (H2) in the cleft unlike a helical peptide modeled in parallel orientation in the peptide-binding groove of PP5 (22) and SycD (17). Given that this class of chaperones binds to at least 2 different molecules, a shallow groove, as a consequence of asymmetric dimerization formed on the opposite side of the major cleft is of interest in mediating substrate binding.

Based on the IpaB/IpgC structure we identified a sequence motif in translocator substrates bound by class II chaperones. This motif is required to fit into the 3 pockets in the IpgC cleft and is well conserved among many translocators. Moreover, the pockets formed by highly-conserved residues could be mapped in the SycD structure as well. The interactions along the cleft define the orientation and impart specificity to the substrate binding to the chaperone. The micromolar range of the affinities exhibited are in the range of those determined for other chaperone–substrate interactions measured by surface plasmon resonance (28) and consistent with the chaperone function of binding and releasing proteins. This information allows predictions and manipulation for this class of chaperone–substrate interaction. Because IpgC is essential for bacterial virulence and the substrate binding surface of class II chaperones is distinct from substrate binding TPRs in eukaryotes, identifying a ligand that disrupts the interaction between IpgC and its substrate could contribute to the development of specific and potent antibacterial agents.


Cloning and Purification of ipgC.

ipgC was amplified from S. flexneri (M90T) by standard PCR method and cloned into the expression vector pET28a (+). The C-terminus amino acids DIKE were mutagenized by using a QuikChange site-directed mutagenesis kit (Stratagene). For functional assays the WT and mutant ipgC were cloned into pBAD/Myc-His (A) (Invitrogen) vector.

Standard expression conditions were applied for obtaining IpgC from BL21(DE3) RIL-harboring plasmid containing ipgC. For the SeMet labeling, the protein was expressed in Escherichia coli B834(DE3).

His-tagged IpgC was purified from the soluble fraction by using a HisTrap HP column (GE Healthcare). His-tag cleaved product was purified by size-exclusion chromatography (Superdex 200) in 20 mM Hepes (pH 7.4), 150 mM NaCl. The protein was finally concentrated to 12–15 mg/mL.

Crystal Structure Determination.

Initial phases for the IpgC apo-structure were obtained by the MAD method. The software package autoSHARP (29) was used to manage the following programs: SHELXC/D (30) found the position of 6 selenium atoms in the SeMet labeled crystal. Phases were calculated by SHARP (31), and improved by solvent flattening in SOLOMON (32). An initial helical model was built with ARP/wARP helix building module (33) and subsequently remodeled to the IpgC sequence in Coot (34) using the selenium positions as sequence markers. Cycles of refinement using CNS version 1.2 (35) against data collected from the native crystal and manual rebuilding with Coot led to the final model including residues 9–151 for both copies in the asymmetric unit, 86 water molecules, 7 glycerol molecules, 2 sulfate anions, and 1 sodium cation. Finally, REFMAC version 5.5.0044 (36) was used to perform TLS refinement, defining 2 groups for each IpgC molecule including residues 9–32 and 33–151.

The crystal structure of the IpaB51–72/IpgC complex was solved by molecular replacement with Phaser (37) using as template residues 40–150 of copy A of the apo-structure and cycles of manual building and refinement using Coot and CNS. After building the final protein model (residues 8–151 of copy A and 10–151 of copy B) a peptide density was visible in the difference map close to the cleft of IpgC. Modeling of the amino acids with their side chains was done using as sequence markers the 2 lysine residues that form salt bridges with negatively-charged aspartates. The complete model includes IpaB peptide residues 63–72 bound to the copy A, residues 60–72 bound to the copy B, 64 waters, and 3 glycerol molecules. A final TLS refinement was performed with REFMAC, partitioning the IpgC molecule in 2 segments like in the apo-structure and adding 2 other TLS groups corresponding to the 2 IpaB peptides.


Titration experiments were carried out with a VP-ITC calorimeter from MicroCal. IpgC titrations were performed by injecting consecutive 12-μL aliquots of 9 mM peptide solution into 1.4-mL IpgC solution (487 μM) in the chamber. IpgCpb titration was performed by injecting 12-μL aliquots of 4.5 mM peptide into 189 μM protein. For YopB/SycD titrations were performed by injecting consecutive 12-μL aliquots of 9 mM YopB peptide solution into 1.4 mL of SycD solution (316 μM) in the chamber. For YopD/SycD titrations were performed by injecting consecutive 12-μL aliquots of 4.5 mM YopD peptide solution into 1.4 mL of SycD solution (228 μM) in the chamber. The heats of dilution of the peptides, PELKAP (−0.06 kcal/mol), PELKAD (−4.51 kcal/mol), IpaB51–72 (−5.87 to −4.39 kcal/mol), and YopD (−0.045 kcal/mol) have been subtracted from the reported heat measured at each injection. Peptides were dissolved in and the proteins dialyzed against 20 mM Hepes (pH 7.5), 150 mM NaCl. Two independent titration experiments were performed at 25 °C. Binding stoichiometry, enthalpy, and equilibrium association constants were determined by fitting the corrected data to 1 set of sites model equation using the evaluation software provided by the manufacturer.

Secretion Assay.

Protein expression and secretion was analyzed from pellets and supernatants respectively of cultures grown with specific inducers as described (8). Briefly, cultures of exponentially growing Shigellae were induced at OD600 = 0.3–0.4 with 0.1% (final concentration) arabinose and incubated for another 2 h. After harvesting crude bacterial extracts were obtained from the pellets, and proteins of filtered (0.2-μm pore size) culture supernatants were precipitated with 10% trichloroacetic acid. Immunoblotting procedures were carried out with antibodies as applicable.

Virulence Assay.

To test for epithelial cell invasion, the number of intracellular bacteria in infected HeLa cells was determined by using a gentamicin protection assay as described (38). Cultures of Shigellae were grown and induced as described above. Infections of HeLa cells were performed by using a multiplicity of infection of 100. Briefly, HeLa cells infected for 20 min were incubated in the presence of gentamicin (100 μg/mL) for additional 2 h. Intracellular bacteria were determined after lysing the infected cells, plating dilutions of the lysates, and counting the cfu. In the assays described above, the standard error was calculated based on at least 3 independent determinations.

Lactate Dehydrogenase Release Assay.

Cytotoxicity assay was performed as described (39). Murine macrophages were grown in 96-well plates and infected in serum-free medium for 2 h. Gentamicin (100 μg/mL) was added at 30-min postinfection. Cytotoxicity was quantified by measuring the release of lactate dehydrogenase enzyme from infected cells with the CytoTox 96 kit (Promega) following the manufacturer's instructions.

See SI Text for more details.

Supplementary Material

Supporting Information:


We thank P. Jungblut and M. Schmid for help with mass spectrometry analysis; N. Jahnke, S. Keller, P. Burkhardt, and D. Fasshauer for help with ITC; A. Grabbe, L. Senerovic, and B. Eilers for help with the functional assays; U. Mueller, S. Monaco, and H. Bartunik for assistance in using beamlines; and B. Raupach, Y. Weinrauch, J. de Diego, K. Metzler, and L. Senerovic for useful comments and critical reading of the manuscript.


The authors declare no conflict of interest.

This article is a PNAS Direct Submission. S.J.H. is a guest editor invited by the Editorial Board.

Data deposition: The atomic coordinates have been deposited in the Protein Data Bank, www.pdb.org (PDB ID codes 3GYZ and 3GZ1).

This article contains supporting information online at www.pnas.org/cgi/content/full/0812900106/DCSupplemental.


1. Formal SB, Hale TL, Sansonetti PJ. Invasive enteric pathogens. Rev Infect Dis. 1983;4(Suppl 5):S702–S707. [PubMed]
2. Labrec EH, Schneider H, Magnani TJ, Formal SB. Epithelial cell penetration as an essential step in the pathogenesis of bacillary dysentery. J Bacteriol. 1964;88:1503–1518. [PMC free article] [PubMed]
3. Sansonetti PJ, Kopecko DJ, Formal SB. Involvement of a plasmid in the invasive ability of Shigella flexneri. Infect Immun. 1982;35:852–860. [PMC free article] [PubMed]
4. Maurelli AT, Baudry B, d'Hauteville H, Hale TL, Sansonetti PJ. Cloning of plasmid DNA sequences involved in invasion of HeLa cells by Shigella flexneri. Infect Immun. 1985;49:164–171. [PMC free article] [PubMed]
5. Menard R, Sansonetti PJ, Parsot C. Nonpolar mutagenesis of the ipa genes defines IpaB, IpaC, and IpaD as effectors of Shigella flexneri entry into epithelial cells. J Bacteriol. 1993;175:5899–5906. [PMC free article] [PubMed]
6. Menard R, Prevost MC, Gounon P, Sansonetti P, Dehio C. The secreted Ipa complex of Shigella flexneri promotes entry into mammalian cells. Proc Natl Acad Sci USA. 1996;93:1254–1258. [PMC free article] [PubMed]
7. Zychlinsky A, et al. IpaB mediates macrophage apoptosis induced by Shigella flexneri. Mol Microbiol. 1994;11:619–627. [PubMed]
8. Menard R, Sansonetti P, Parsot C. The secretion of the Shigella flexneri Ipa invasins is activated by epithelial cells and controlled by IpaB and IpaD. EMBO J. 1994;13:5293–5302. [PMC free article] [PubMed]
9. Parsot C, et al. A secreted anti-activator, OspD1, and its chaperone, Spa15, are involved in the control of transcription by the type III secretion apparatus activity in Shigella flexneri. Mol Microbiol. 2005;56:1627–1635. [PubMed]
10. Feldman MF, Cornelis GR. The multitalented type III chaperones: All you can do with 15 kDa. FEMS Microbiol Lett. 2003;219:151–158. [PubMed]
11. Parsot C, Hamiaux C, Page AL. The various and varying roles of specific chaperones in type III secretion systems. Curr Opin Microbiol. 2003;6:7–14. [PubMed]
12. Stebbins CE, Galan JE. Maintenance of an unfolded polypeptide by a cognate chaperone in bacterial type III secretion. Nature. 2001;414:77–81. [PubMed]
13. Birtalan SC, Phillips RM, Ghosh P. Three-dimensional secretion signals in chaperone-effector complexes of bacterial pathogens. Mol Cell. 2002;9:971–980. [PubMed]
14. Lilic M, Vujanac M, Stebbins CE. A common structural motif in the binding of virulence factors to bacterial secretion chaperones. Mol Cell. 2006;21:653–664. [PubMed]
15. Evdokimov AG, et al. Similar modes of polypeptide recognition by export chaperones in flagellar biosynthesis and type III secretion. Nat Struct Biol. 2003;10:789–793. [PubMed]
16. Tardy F, et al. Yersinia enterocolitica type III secretion-translocation system: Channel formation by secreted Yops. EMBO J. 1999;18:6793–6799. [PMC free article] [PubMed]
17. Pallen MJ, Francis MS, Futterer K. Tetratricopeptide-like repeats in type III-secretion chaperones and regulators. FEMS Microbiol Lett. 2003;223:53–60. [PubMed]
18. Büttner CR, Sorg I, Cornelis GR, Heinz DW, Niemann HH. Structure of the Yersinia enterocolitica type III secretion translocator chaperone SycD. J Mol Biol. 2008;375:997–1012. [PubMed]
19. Schmid A, et al. Yersinia enterocolitica type III secretion chaperone SycD: Recombinant expression, purification and characterization of a homodimer. Protein Expression Purif. 2006;49:176–182. [PubMed]
20. Cornelis GR. The type III secretion injectisome. Nat Rev Microbiol. 2006;4:811–825. [PubMed]
21. Lamb JR, Tugendreich S, Hieter P. Tetratrico peptide repeat interactions: to TPR or not to TPR? Trends Biochem Sci. 1995;20:257–259. [PubMed]
22. Das AK, Cohen PW, Barford D. The structure of the tetratricopeptide repeats of protein phosphatase 5: Implications for TPR-mediated protein–protein interactions. EMBO J. 1998;17:1192–1199. [PMC free article] [PubMed]
23. Han D, Oh J, Kim K, Lim H, Kim Y. Crystal structure of YrrB: A TPR protein with an unusual peptide-binding site. Biochem Biophys Res Commun. 2007;360:784–790. [PubMed]
24. Page AL, Fromont-Racine M, Sansonetti P, Legrain P, Parsot C. Characterization of the interaction partners of secreted proteins and chaperones of Shigella flexneri. Mol Microbiol. 2001;42:1133–1145. [PubMed]
25. Scheufler C, et al. Structure of TPR domain–peptide complexes: Critical elements in the assembly of the Hsp70–Hsp90 multichaperone machine. Cell. 2000;101:199–210. [PubMed]
26. Edqvist PJ, et al. Tetratricopeptide repeats in the type III secretion chaperone, LcrH: Their role in substrate binding and secretion. Mol Microbiol. 2006;59:31–44. [PubMed]
27. Birket SE, et al. Preparation and characterization of translocator/chaperone complexes and their component proteins from Shigella flexneri. Biochemistry. 2007;46:8128–8137. [PubMed]
28. Swietnicki W, et al. Novel protein–protein interactions of the Yersinia pestis type III secretion system elucidated with a matrix analysis by surface plasmon resonance and mass spectrometry. J Biol Chem. 2004;279:38693–38700. [PubMed]
29. Vonrhein C, Blanc E, Roversi P, Bricogne G. Automated structure solution with autoSHARP. Methods Mol Biol. 2007;364:215–230. [PubMed]
30. Schneider TR, Sheldrick GM. Substructure solution with SHELXD. Acta Crystallogr D. 2002;58:1772–1779. [PubMed]
31. de La Fortelle E, Bricogne G. Maximum-likelihood heavy-atom parameter refinement for multiple isomorphous replacement and multiwavelength anomalous diffraction methods. Methods Enzymol. 1997;276:472–494.
32. Abrahams JP, Leslie AG. Methods used in the structure determination of bovine mitochondrial F1 ATPase. Acta Crystallogr D. 1996;52:30–42. [PubMed]
33. Lamzin VS, Perrakis A, Wilson KS. In: Int. Tables for Crystallography. Vol. F: Crystallography of Biological Macromolecules. Rossmann MG, Arnold E, editors. The Netherlands: Kluwer Academic Publishers, Dordrecht; 2001. pp. 720–722.
34. Emsley P, Cowtan K. Coot: Model-building tools for molecular graphics. Acta Crystallogr D. 2004;60:2126–2132. [PubMed]
35. Brunger AT, et al. Crystallography and NMR system: A new software suite for macromolecular structure determination. Acta Crystallogr D. 1998;54:905–921. [PubMed]
36. Murshudov GN, Vagin AA, Dodson EJ. Refinement of macromolecular structures by the maximum-likelihood method. Acta Crystallogr D. 1997;53:240–255. [PubMed]
37. McCoy A, et al. Phaser crystallographic software. J Appl Crystallogr. 2007;40:658–674. [PMC free article] [PubMed]
38. Mounier J, Vasselon T, Hellio R, Lesourd M, Sansonetti PJ. Shigella flexneri enters human colonic Caco-2 epithelial cells through the basolateral pole. Infect Immun. 1992;60:237–248. [PMC free article] [PubMed]
39. Hersh D, et al. The Salmonella invasin SipB induces macrophage apoptosis by binding to caspase-1. Proc Natl Acad Sci USA. 1999;96:2396–2401. [PMC free article] [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...