Interplay between predicted inner-rod and gatekeeper in controlling substrate specificity of the type III secretion system
Article first published online: 6 FEB 2013
© 2013 Blackwell Publishing Ltd
Volume 87, Issue 6, pages 1183–1199, March 2013
How to Cite
Cherradi, Y., Schiavolin, L., Moussa, S., Meghraoui, A., Meksem, A., Biskri, L., Azarkan, M., Allaoui, A. and Botteaux, A. (2013), Interplay between predicted inner-rod and gatekeeper in controlling substrate specificity of the type III secretion system. Molecular Microbiology, 87: 1183–1199. doi: 10.1111/mmi.12158
- Issue published online: 11 MAR 2013
- Article first published online: 6 FEB 2013
- Accepted manuscript online: 22 JAN 2013 04:37AM EST
- Manuscript Accepted: 15 JAN 2013
- FRS-FNRS. Grant Number: F.3.4556.11
- European Community. Grant Number: 261742
- Belgian Fonds National de Recherches Industrielles et Agronomiques (FRIA)
- Fonds Defay
- Alice and David Van Buuren foundation
- Top of page
- Experimental procedures
- Supporting Information
The type III secretion apparatus (T3SA) is a multi-protein complex central to the virulence of many Gram-negative pathogens. Currently, the mechanisms controlling the hierarchical addressing of needle subunits, translocators and effectors to the T3SA are still poorly understood. In Shigella, MxiC is known to sequester effectors within the cytoplasm prior to receiving the activation signal from the needle. However, molecules involved in linking the needle and MxiC are unknown. Here, we demonstrate a molecular interaction between MxiC and the predicted inner-rod component MxiI suggesting that this complex plugs the T3SA entry gate. Our results suggest that MxiI–MxiC complex dissociation facilitates the switch in secretion from translocators to effectors. We identified MxiCF206S variant, unable to interact with MxiI, which exhibits a constitutive secretion phenotype although it remains responsive to induction. Moreover, we identified the mxiIQ67A mutant that only secretes translocators, a phenotype that was suppressed by coexpression of the MxiCF206S variant. We demonstrated the interaction between MxiI and MxiC homologues in Yersinia and Salmonella. Lastly, we identified an interaction between MxiC and chaperone IpgC which contributes to understanding how translocators secretion is regulated. In summary, this study suggests the existence of a widely conserved T3S mechanism that regulates effectors secretion.
- Top of page
- Experimental procedures
- Supporting Information
Type 3 secretion systems (T3SSs) are macromolecular complexes of many Gram-negative bacteria and are crucial factors for their virulence capabilities. They are composed of at least 21 distinct proteins forming three major components: a transmembrane region, a cytoplasmic bulb also called C-ring, and an extracellular needle through which bacteria can inject its proteins into the host cells (Cornelis, 2006). This multi-protein complex is also comprised of a membrane-associated ATPase which faces the bacterial cytoplasm thought to be involved in facilitating the entry of export substrates into the T3SA channel (Akeda and Galan, 2005; Lara-Tejero et al., 2011).
In contrast to the T3SA structural components, the injected effectors are highly diversified and possess multiple mechanisms to hijack numerous cellular pathways of the host cell leading to bacterial establishment. For example, some effectors can induce cytoskeleton reorganization to promote bacterial entry (Patel and Galan, 2005; Parsot, 2009), others act as immune modulators to avoid host defences (Espinosa and Alfano, 2004; Ogawa et al., 2008) or phagocytosis by macrophages (Shao, 2008). One of the most important properties of this secretion system is the ability to adapt rapidly to host cell contact in a well-organized manner to achieve effective virulence (Deane et al., 2010; Buttner, 2012). First, T3SA inserts hydrophobic proteins, also called translocators, into the cell membrane to form a pore through which effector proteins are injected into the host cell. In the absence of signal from host cell, the T3SA is inactive and is plugged by the tip complex proteins (Blocker et al., 2008).
Among bacteria that use a T3SA, Shigella causes the most devastating epidemics of bacterial infection in the developing world with up to one million deaths each year, essentially among young children (Kotloff et al., 1999). Shigella uses the T3SA to penetrate enterocytes and to disseminate into the colonic epithelium leading to a partial destruction of the mucosal lining and shigellosis symptoms. Most of the virulence factors of Shigella are encoded by a large virulence plasmid harbouring a 31 kb region that is sufficient to promote bacterial entry into host cells (Buchrieser et al., 2000). The ‘entry’ region is organized in two loci, one encoding Mxi-Spa proteins that form the T3SA and the other encoding early substrates of this apparatus and their cognate chaperones. At 37°C, the needle subunit, MxiH and the predicted inner-rod component MxiI are secreted through the T3SA base, then MxiH polymerizes to form up to 50 nm needle (Magdalena et al., 2002). MxiI is homologue to proteins forming the inner rod in other bacteria including PrgJ of Salmonella, YscI of Yersinia and PscI of Pseudomonas (Blocker et al., 2001; Marlovits et al., 2004; 2006). Once the needle length has reached 50 nm, it is plugged by the tip complex composed of IpaB and IpaD in the absence of T3S induction and before host cell contact. Exposure of IpaD at the tip serves, with IpaB, as sensors of the host cell contact (Olive et al., 2007; Veenendaal et al., 2007). After the cell contact is sensed by this tip, IpaC is secreted to form with IpaB, a translocation pore in the host cell membrane. At this stage, another set of IpaB/IpaC and early effectors accumulate within the bacterial cytoplasm and bind to their respective chaperone, such as IpgC for the translocators. Once Shigella comes into contact with the host cell membrane, T3SA activation by an uncharacterized signal leads to the insertion of IpaB and IpaC into the host membrane and enhances translocators/effectors secretion. The resulting pore, also called the translocon, serves as a gate for secreted effectors to gain access to host cytoplasm. Moreover, the expression of a set of ‘late’ effectors, acting at later stage of infection, is controlled by the T3SA activity (Mavris et al., 2002). In the absence of cell contact, their expression is turned off by OspD1, a protein sequestering the transcriptional activator MxiE that requires the IpgC cofactor for activation of late effectors gene transcription (Parsot et al., 2005). Upon IpaB, IpaC and OspD1 secretion, released IpgC binds to MxiE leading to activation of late effectors genes transcription (Parsot et al., 2005; Pilonieta and Munson, 2008).
In Shigella tip complex mutants (ipaB, ipaD), the translocators, early and late effectors are constitutively secreted (Menard et al., 1994; Parsot et al., 1995). This suggests that the activation signal (cell contact) in these mutants is already transmitted to the T3SA base. Interestingly, similar findings were described for the mxiC mutant although the tip complex was still present suggesting the existence of a mechanism that retains them inside bacteria prior to cell contact. Moreover, the translocators secretion is known to be impaired in response to Congo red (CR) (Botteaux et al., 2009; Martinez-Argudo and Blocker, 2010), an amphipathic dye used to mimic host cell contact (Bahrani et al., 1997). MxiC shares its function with the other members of the YopN family including YopN (Yersinia enterocolitica), InvE (Salmonella enterica), SepL [enterohaemorrhagic (EHEC) and enteropathogenic (EPEC) Escherichia coli], HrpJ (Pseudomonas syringae) and PopN (Pseudomonas aeruginosa) (Pallen et al., 2005). The MxiC crystal structure reveals three distinct ‘X-bundle’ structures made of four, five and four α-helices respectively (Deane et al., 2008). A recent study also highlighted the role of MxiC in the activation signal transmission from T3SA tip to the base whereby mxiHK69A and mxiHR83A mutants are unable to promote effectors and MxiC secretion (Kenjale et al., 2005; Martinez-Argudo and Blocker, 2010). Interestingly, this defect was abrogated in these mutants through mxiC gene inactivation, suggesting that MxiC secretion is a prerequisite for effectors secretion upon T3S induction.
How the activation signal is transmitted and leads to effectors secretion remains a matter of debate. The most highly supported model, called ‘allosteric model’, attempted to highlight the needle role in signal transmission. Some mxiH mutants, such as mxiHK69A, mxiHR83A, mxiHP44A and mxiHQ51A, are known to affect secretion of effectors through a variety of mechanisms (Kenjale et al., 2005). It was suggested that the signal is first sensed by the tip complex and transmitted along the needle, thereby reaching the base and promoting effectors secretion. MxiC is considered as a strong signal receiver candidate but the components facilitating the interaction between MxiC and the needle remain unknown.
In the present study, we identified one of these missing elements as the predicted inner-rod component MxiI. We revealed an important biological insight based on MxiI and MxiC interaction and highlighted its role in sequestering effectors prior to T3S induction. We have also shown that the molecular interaction between MxiI and MxiC homologues is common in other bacterial species, leading us to suggest the existence of a shared mechanism that regulates the switch to effectors secretion. Furthermore, we generated the mxiIQ67A mutant and showed that it exhibits almost the mxiHK69A effector mutant phenotype. Interestingly, effectors secretion was restored in this strain by a suppressive mxiCF206S extragenic mutation. Lastly, we put forward evidence suggesting the role of a newly identified interaction between MxiC and the chaperone IpgC in the regulation of translocators secretion.
- Top of page
- Experimental procedures
- Supporting Information
Mutational analysis revealed the MxiCF206S variant that discriminates between constitutive and inducible secretion states
While wild-type Shigella secretes only low levels of Ipa proteins, mxiC mutant exhibits a constitutive secretion of effectors (Botteaux et al., 2009). Ferracci et al. (2005) showed that residues clustered in the central region of YopN are important for secretion. Such region corresponds to X-bundle 2 of MxiC, which is spatially separated from the SycN/YscB-binding domain of YopN (residues 81–160, X-bundle 1) and from the region homologue to TyeA (residues 271–356, X-bundle 3) (Day and Plano, 1998; Pallen et al., 2005; Joseph and Plano, 2007) (Fig. 1A). In other bacterial species, the structural and sequence alignment analysis point towards highly conserved residues between MxiC and its homologues. In support of this, mutations of conserved residues in the interfaces between YopN domains, such as Y202 to an Asparagine, constitutively block Yop secretion (Ferracci et al., 2005). Interestingly, Deane et al. (2008) suggested that the hydrophobic patch on the surface of MxiC central domain (X-bundle 2) consisting of residues Leu222, Met226, Gly239, Leu242 and Leu245 interacts with other T3S components (Fig. 1A). Thus, to gain further insights into the mechanism by which MxiC regulates secretion, we generated point mutations on the pAB108′ (pSU18-mxiC) of the five residues described above and mutated residue Y185 corresponding to YopNY202. Additionally, we mutated F206 to a Serine as this residue is conserved in YopN (Fig. 1A). We subsequently analysed proteins secreted by the mxiC mutant expressing generated variants (Y185N, F206S, L222A, M226A, G239A, L242A and L245A) under both constitutive and CR inducible secretion conditions (Fig. 1B and C). When bacteria are growing in broth, the T3SA is weakly active and translocators and effectors are stored in the cytoplasm in association with their dedicated chaperones. The activation of the T3SA upon contact with epithelial cells or exposure to the dye Congo red (Parsot et al., 1995; Bahrani et al., 1997) leads to increased transcription of 12 effector-encoding genes scattered on the virulence plasmid, including members of the ipaH family (Demers et al., 1998). While MxiCM226A, MxiCG239A and MxiCL245A were able to restore wild-type secretion, the MxiCL222A, MxiCL242A and MxiCY185N exhibited the mxiC mutant phenotype since they were unable neither to control effectors secretion nor to respond to CR induction (Fig. 1B and C). Interestingly, the mxiCF206S variant presented a new and unexpected phenotype discriminating between constitutive and CR induction states (Fig. 1B and C). The constitutive expression of the seven MxiC variants from the plac promoter was assessed by Western blot analysis using anti-MxiC rabbit polyclonal antibodies (Fig. 1D). Out of the seven generated variants, MxiCL222A and MxiCL242A were not detectable (Fig. 1D) which explain the lack of secretion control by these two variants (Fig. 1B and C).
Next, we focused our study on the newly identified phenotype expressed by the mxiCF206S mutation. In the absence of T3S induction, MxiCF206S gave rise to a constitutive secretion of effectors IpaA and IpaH (Fig. 1E and F). In contrast, upon T3S induction with CR for 10 min, MxiCF206S fully restored wild-type secretion of IpaB, OspD1, as well as effectors IpaA and IpaH (Fig. 1G and H). As MxiC is itself a T3S substrate, compared with native MxiC, we found that secretion of MxiCF206S, although produced at wild-type level, was reduced under CR induction (Fig. 1H). This phenotype could be due to the increased secretion of MxiCF206S seen in the absence of induction (Fig. 1F). Hence, the observed effect on secretion was not due to the lack of proteins production (Fig. 1I).
MxiC folding is not impaired by the F206S single mutation
One possible explanation regarding the phenotype associated with the mxiCF206S mutation could be the misfolding of MxiCF206S. To test this hypothesis, highly purified His-MxiC and His-MxiCF206S (Fig. 2A) were assessed in two fluorescence assays (see Experimental procedures). The usefulness of fluorescence spectroscopy, as a valid tool to monitor possible structural changes in His-MxiCF206S protein upon the single mutation, was based on available MxiC crystallographic data (Deane et al., 2008) and the presence of a single Tryptophan (W202) located in close proximity to residue F206 (Fig. 2B). It is thus reasonable to assume that if some structural modification could occur, the Tryptophan fluorescence will be also affected by its environment changes. As shown in Fig. 2C, the intrinsic fluorescence spectrum of purified His-MxiCF206S is quite superimposable on that of unmodified His-MxiC. In addition, the fluorescence peak occurred at around 330 nm, demonstrating that the aromatic residues are located in an apolar environment as a result of the native and compact conformation adopted by the two protein forms, at least around the mutated domain.
The second method used to assess protein folding was based on the ANS binding assay (See Experimental procedures). The incubation of purified His-MxiC or His-MxiCF206S with a large excess of ANS showed the same behaviour of the two proteins, resulting in an increase in the fluorescence intensity and a blue shifting of its λmax, demonstrating some dye binding to the two proteins (Fig. 2D). On a quantitative point of view, the fluorescence intensity reached in the presence of the two proteins is quite comparable. ANS binding can be partly explained by the fact that the first 70–80 residues of MxiC were reported to be poorly structured and consequently not characterized by a well-defined electron density. Furthermore, it was suggested that MxiC does not possess a globular structure and/or be partially disordered (Deane et al., 2008), probably explaining the ANS binding characteristics. Taking together, our results suggest that the folding of His-MxiCF206S is not affected by the single mutation and adopts the same overall conformation as His-MxiC.
MxiC interacts with the predicted inner-rod protein MxiI
It was reported that some specific mxiH mutants affect MxiC secretion in response to CR induction and thereby inhibiting the secretion of effectors. Hence, the signal leading to effector secretion is thought to be transmitted to MxiC via needle subunits (Martinez-Argudo and Blocker, 2010). Subsequently, we investigated potential direct interaction between MxiC, MxiH and MxiI. To do so, we constructed three plasmids encoding GST-MxiI, GST-MxiH and His-MxiC and performed GST pull-down assay (see Experimental procedure). Bacterial lysates containing His-MxiC were incubated with glutathione-Sepharose beads pre-coated with GST-MxiI, GST-MxiH or GST alone. We revealed an interaction between MxiC and MxiI, while no interaction was detected between MxiC and MxiH or GST alone (Fig. 3A). Besides, the MxiC–MxiI interaction was detected in E. coli suggesting that no additional T3S components are required.
As the molecular weight of GST is 28 kDa and the one of MxiH does not exceed 9 kDa, we cannot exclude the possibility that GST itself may prevent MxiH interaction with MxiC by some steric hindrance effect. To overcome such experimental artefact, we produced His-MxiH and tested it for GST-MxiC interaction. As a positive control, we also incubated GST-MxiC with His-MxiI. The obtained results confirmed that MxiI, but not MxiH, interacts with MxiC (Fig. 3B).
Since the MxiCF206S variant does not control secretion of effectors under non-inducible secretion condition but is still able to respond to CR induction, we assessed MxiCF206S interaction with MxiI. For that purpose, we generated the His-MxiCF206S variant and performed a GST-MxiI pull-down assay. In contrast to unmodified His-MxiC, His-MxiCF206S was not co-eluted with GST-MxiI although produced at wild-type level (Fig. 3C). This finding suggests that the interaction between MxiC and MxiI may sequester effectors prior to T3S induction.
MxiC–MxiI complex dissociation is presumably required to recover effectors secretion in the mxiHK69A background
Needle subunit mxiH mutants (mxiHK69A and mxiHR83A) were defined as ‘effector mutants’ since they secrete translocators upon CR induction, but are unable to secrete MxiC and effector proteins. This defect was suppressed upon mxiC gene inactivation in these mutants (Kenjale et al., 2005; Martinez-Argudo and Blocker, 2010). Such a finding directly corroborates the crucial role of MxiH in MxiC release prior to effectors secretion. The mxiCF206S mutant is unable to control effectors secretion but can still secrete translocators/effectors in response to CR induction. We subsequently investigated whether or not the introduction of the F206S mutation in the mxiHK69A background would restore effectors secretion. The mxiC mxiH double mutant construction did not induce proteins secretion and this was restored to wild-type state by pYC163 plasmid producing native MxiC and MxiH (Fig. 4A). Not surprisingly, the mxiH mxiC mutant expressing MxiCF206S–MxiH (pYC166) or MxiC–MxiHK69A (pYC165) exhibited the single mxiCF206S or mxiHK69A phenotype respectively (Fig. 4A). Upon CR induction, the secretion of early effectors was restored in strain mxiHK69A mxiCF206S (pSL164) (Fig. 4A). As MxiCF206S does not interact with MxiI, our results suggest that MxiC–MxiI complex dissociation is required for switching to effectors secretion. Unexpectedly, MxiCF206S was not secreted in the mxiHK69A background even though it was produced at wild-type level (Fig. 4B). The data presented here suggests that MxiC secretion is not a prerequisite for the switch to effectors secretion.
Exclusive translocators secretion in the mxiIQ67A background
The effect of the needle mutants (mxiHK69A, mxiHR83A, mxiHP44A, mxiHQ51A) in the regulation of secretion was previously reported (Kenjale et al., 2005). As MxiH and MxiI share around 20% sequence identity (Fig. 5A), we investigated whether or not selective mutations of MxiH's conserved residues within MxiI might also affect secretion. While residues K69 and R83 are not conserved in MxiI, P44 and Q51 correspond to MxiI residues P60 and Q67 respectively. To test this hypothesis, we first constructed a low-copy plasmid, pSM6, expressing MxiI that restores the secretion defect of the mxiI mutant (Fig. 5B). To avoid the potential effect of P60 mutation on the structure of MxiI, we replaced residue Q67 by an Alanine (pSM33) and further tested the strain mxiIQ67A for its secretion ability. Proteins analysis of the mxiIQ67A strain revealed that in response to CR induction, secretion of IpaA was reduced, and IcsB, MxiC and OspD1 were dramatically affected, while IpaB secretion remained unaffected (Fig. 5B). The observed reduction in secretion was not associated with the lack of proteins production as shown in Fig. 5C. As a control, we monitored the presence of the cytoplasmic marker DnaK protein in both cells lysate and supernatant fractions. Our results clearly show the absence of any cytoplasmic material in the supernatant fractions (Fig. 5B). We conclude that although secretion of IpaA appeared slightly different (Figs 4B and 5B), the mxiHK69A and mxiIQ67A mutants exhibit almost a similar phenotype.
Mutation mxiCF206S restores effectors secretion in the mxiIQ67A background
To further investigate the phenotype of the mxiIQ67A mutant, we constructed an mxiI mxiC double mutant (see Experimental procedures) that did not secrete proteins. This phenotype was restored to wild-type state by expression of native MxiI and MxiC from plasmids pSM6 and pF13b respectively (Fig. 5D). As expected, in the absence of MxiC, mxiI mxiC/pSM33 (MxiIQ67A) secreted IpaB, IpaA, IcsB and OspD1 (Fig. 5D). Comparatively, wild-type secretion of IpaB and IpaA and a reduced secretion of IcsB and OspD1 were observed in the mxiCF206S mxiIQ67A background (Fig. 5D). Interestingly, MxiCF206S was not secreted in the mxiIQ67A background, although it was produced at wild-type level (Fig. 5D and E). Lastly, using GST pull-down assay, we found that MxiIQ67A interacts with His-MxiC, but not with the His-MxiCF206S variant (Fig. 5F).
The interaction between MxiI and MxiC homologues occurs in Salmonella and Yersinia
As mutants in invE of Salmonella and in yopN of Yersinia exhibit a deregulated secretion phenotype (Forsberg et al., 1991; Kubori and Galan, 2002), we investigated whether, as for MxiC, InvE and YopN may also interact with MxiI counterparts PrgJ and YscI respectively. Hence, we constructed four plasmids producing GST-PrgJ, His-InvE, GST-YopN and His-YscI. After immobilization on gluthatione-Sepharose beads, GST-PrgJ was incubated with E. coli lysate expressing His-InvE. We found that InvE interacts with GST-PrgJ but not with GST alone (Fig. 6A). As YopN requires specific chaperones for its stability, GST-YopN was produced in the Y. enterocolitica background before incubation with His-YscI. We also showed that YscI interacts, either directly or indirectly, with YopN but not with GST alone (Fig. 6B).
Since residue F206 seems crucial for the MxiC–MxiI interaction, we mutated the corresponding residues, Y211 and F223, in His-InvE and in GST-YopN respectively (Fig. 1A). The insolubility of His-InvEY211S meant that we could not assess its interaction with PrgJ (data not shown). In contrast, we found that GST-YopNF223S interacts with YscI (Fig. 6B). Hence, we postulate here, unlike the yopN mutant, yopNF223S strain has to control Yop secretion in the absence of induction by calcium (+ Ca2+). To test such hypothesis, we constructed low-copy plasmid pYC171 (YopN) that fully restored the secretion defect of the yopN mutant (Fig. 6C). Next, we constructed plasmid pYC174 expressing YopNF223S and showed that Yop secretion was blocked in the yopNF223S strain under inducible secretion condition (− Ca2+) (Fig. 6D). As a control, no secretion was observed with the yscN T3S deficient secretion mutant.
Secretion of IpaB and IpaC translocators is delayed in the mxiC mutant
In contrast to wild-type Shigella, the mxiC mutant is not responsive to CR induction and is defective in IpaC secretion. Moreover, it constitutively secretes low levels of IpgD between 2 and 10 min post CR induction (Martinez-Argudo and Blocker, 2010). As MxiC is required in T3S signal transmission, we asked whether this defect could be due to a delay in CR responsiveness. Bahrani et al. (1997) previously established that maximal recovery of IpaC in the supernatant fraction is obtained after 30 min of CR induction. Therefore, we lengthened the induction time from 5 to 30 min and analysed the proteins of the supernatant. Our results showed that IpaB, IpaC and IpaD secretion, by the mxiC mutant, reached wild-type level within 30 min post induction (Fig. 7). The secretion defect of IpaB, IpaC and IpaD at 5 and 10 min post induction is not due to their expression since they were produced at wild-type level (Fig. 7). Compared to IpaB, IpaC and IpaD, MxiC secretion was delayed in the wild-type strain as it started 10 min post induction (Fig. 7). These results suggest that MxiC is involved in the induction of translocators secretion which appears to be delayed rather than impaired in the mxiC mutant. As a control, under similar conditions, secretion of the IcsB effector was almost comparable although a slight increase was detected in the mxiC mutant background (Fig. 7).
IpgC enhances MxiC interaction with ATPase Spa47 and regulates translocators secretion
To investigate the mechanism by which MxiC regulates IpaB and IpaC secretion, we took advantage of the direct interaction that we previously reported (Botteaux et al., 2009) between the ATPase Spa47 and MxiC. The significance of this interaction remained only speculative and needs to be further clarified. As this interaction is hardly seen in E. coli, we assessed it in the Shigella background. We found that the interaction is enhanced suggesting the existence of an additional specific Shigella component that stabilizes the complex (Fig. 8A). Taking into account that MxiC regulates the secretion of IpaB and IpaC, we next assessed Spa47–MxiC interaction in E. coli strain harbouring the pHI1 plasmid encoding the translocator chaperone (His-IpgC). We incubated GST alone or GST-Spa47 with His-IpgC and MxiC-FLAG either independently, simultaneously or sequentially, before analysing the eluted fractions using anti-His and anti-FLAG antibodies (Fig. 8B and C). Hereby, we demonstrated that MxiC efficiently interacts with Spa47 in the Shigella background when Spa47 is already complexed to IpgC (Fig. 8B and C, lane 4). In contrast, the interaction of Spa47 with the preformed MxiC–IpgC complex was undetectable (Fig. 8B and C, lane 3). It is worth noting that the interaction between Spa47 and IpgC is reported here for the first time.
- Top of page
- Experimental procedures
- Supporting Information
We have previously reported that mxiC gene deletion deregulates secretion of early and late effectors in the absence of T3S induction highlighting the crucial role of MxiC in the control of effectors release (Botteaux et al., 2009). However, the molecular mechanism remained uncharacterized. Here, we report for the first time an interaction between the predicted inner-rod component MxiI and the gatekeeper MxiC. We also demonstrated that such interaction is potentially involved in the secretion switch from translocators to effectors. Mutational analysis also reveals that the inability to control effectors secretion by the MxiCF206S variant whose folding was not affected compared with native MxiC, is likely due to its lack of interaction with MxiI. This suggests that in the absence of T3S induction, MxiI–MxiC complex is docked at the T3SA entry gate forming a plug that would specifically prevent effectors secretion. Upon T3S induction, MxiC is probably detached from MxiI which consequently opens the gate for effectors release. For timely secretion, MxiC is likely to receive a signal from the tip complex. Interestingly, several studies have suggested that the needle itself is required for signal transmission from T3SA tip to base but the underlying mechanism remains unknown (Martinez-Argudo and Blocker, 2010). Even though MxiI has never been precisely localized at the Shigella T3SA base, its counterpart PrgJ in Salmonella was shown to be part of the inner-rod and to stabilize the needle structure (Marlovits et al., 2004). In the Salmonella needle structure study, it was suggested that the large plasticity in how subunits are locally packed around the canal could be responsible for signal transmission along the needle (Kimbrough and Miller, 2002). Obviously, such a signal must go down from the tip to base through MxiH and MxiI. In support of this hypothesis, two mutations within mxiH (K69A and R83A) prevent early effectors secretion (Kenjale et al., 2005). Such a mechanism is likely achieved by sequestering MxiC within the bacterial cytoplasm. Indeed, Martinez-Argudo and Blocker (2010) reported that this defect can be reverted by deleting the mxiC gene in the mxiHK69A and mxiHR83A backgrounds. Here, we demonstrate that effectors secretion can be recovered in the mxiHK69A mxiCF206S background, which can be explained at least by the inability of MxiCF206S to interact with MxiI. Besides, K69A mutation may cause potential loss of plasticity between the needle and inner-rod subunits which blocks the activation signal transmission and subsequently affects the dissociation of MxiC–MxiI complex. Alternatively, we cannot exclude that MxiHK69A no longer receives the activation signal from the tip complex.
In the course of this study, we generated the mxiIQ67A strain that exhibits a closely related phenotype to the mxiHK69A effector mutant. This is the first report demonstrating that a single mutation within the inner-rod component deregulated secretion. Effectors secretion was also recovered in the mxiIQ67A background either by deleting the mxiC gene or upon coexpressing MxiCF206S variant. Comparatively, the mutation yscIQ84A in Yersinia corresponding to mxiIQ67A only reduces Yop secretion (Wood et al., 2008).
As mutants in genes encoding mxiC homologues in other pathogens also deregulate effectors secretion, we investigated whether or not equivalent MxiC–MxiI interaction may also take place in Yersinia and Salmonella. We demonstrate here that MxiC–MxiI interaction occurs in both bacteria suggesting the existence of a shared mechanism regulating the switch to effectors secretion. In Yersinia, numerous mutations within highly conserved residues of YopN block Yop secretion by a mechanism that remains poorly understood (Ferracci et al., 2005). Accordingly, we have shown here that YopNF223S mutation also blocks Yop secretion. In contrast to MxiCF206S, YopNF223S still interacts with YscI, suggesting that YscI-YopNF223S may form a complex that sticks at the T3SA entry gate to block Yop secretion. It is also worth noting that, due to proteins complex instability in E. coli, the interaction assay was performed using GST-YopN or GST-YopNF223S produced in the Yersinia background. Thus, the interaction observed between YscI and YopNF223S, in contrast to Shigella, may involve additional T3S partners such as SycN or SycB which are of course missing in Shigella. Comparatively, the MxiCY185N variant, corresponding to the YopNY202N that also prevents Yop secretion in Yersinia, only deregulates effectors secretion in Shigella. Therefore, our findings contribute to unravel the mechanism involved in blocking Yop secretion.
To date, InvE (MxiC counterpart) was not detected in Salmonella secreted fraction supporting the fact that gatekeeper secretion is not a prerequisite for its function. In the mxiI or mxiH ‘effector mutants’, MxiC and MxiCF206S secretion is totally abolished while only MxiCF206S expression restored effectors secretion. Knowing that MxiCF206S lost its ability to interact with MxiIQ67A or MxiI and promotes effectors secretion, the activation signal could be facilitated via MxiC detachment from MxiI rather than MxiC secretion per se. Accordingly, while MxiCF206S secretion was significantly reduced compared to MxiC in the wild-type background, effectors were equally secreted upon T3S induction. This could be explained by the inability of MxiCF206S to interact with MxiI, which would consequently limit its own recruitment to the T3SA entry gate. In this case, instead of being secreted MxiCF206S, might be preferentially taken up by the Spa47–IpgC complex.
The mxiC mutant is known to secrete translocators in response to CR induction within 10 min (Martinez-Argudo and Blocker, 2010). In the present study, we have demonstrated that secretion of translocators released by the mxiC mutant reaches wild-type level at 30 min post CR induction. Both IpaB and IpaC interact with IpgC and the dissociation of these complexes in vitro implies various physical events (Birket et al., 2007). The latter is poorly understood in vivo but is believed to be achieved via the ATPase Spa47 complex. Here, we identify an unexpected role of IpgC in stabilizing Spa47 interaction with MxiC. The absence of a direct interaction between Spa47 and MxiC (Fig. 8, line 3) could be explained by the fact that the conformation adopted by the preformed IpgC–MxiC complex in E. coli is not suitable for such a complex. However, MxiC-FLAG, produced in the Shigella background, interacts with Spa47 which is likely to be in a complex with IpgC-translocators. Thus, it is tempting to speculate that, under physiological conditions, two events might occur: (i) MxiC interacts with the complex Spa47–IpgC-translocators, and (ii) MxiC displaces translocators to promote their secretion upon T3S induction. Such a mechanism of action is supported by the fact that mxiC mutant expressing MxiCF206S, which still interacts with IpgC, remains responsive to CR induction. Comparatively, different links between MxiC and the translocators were previously reported. These include, for example, the interaction between InvE (MxiC) and SicA (IpgC) in Salmonella (Kubori and Galan, 2002), the interaction between CopN (MxiC) and Ssc3 (IpgC) in Chlamydia (Archuleta et al., 2011) and the interaction between the chaperone TyeA (MxiCXB3) and YopD (IpaC) in Yersinia. Collectively, these results suggest the existence of a shared mechanism regulating the hierarchy of translocators secretion.
Altogether, our findings contribute to the emerging general pattern related to the T3S hierarchy (Fig. 9). Briefly, before T3S induction, MxiC is engaged in a complex with MxiI at the T3SA entry gate to prevent effectors secretion. A first pool of IpaB and IpaC is already exposed at the tip of the needle to sense CR or host cell contact, while a second one, bound to IpgC, remains stored in the cytoplasm. Upon T3S induction, an activation signal is allosterically transmitted from the tip complex to MxiH, then to MxiI, and finally to MxiC. At this stage, we may consider that MxiC is partly attached to Spa47 which would facilitate IpaB and/or IpaC detachment from IpgC. Once translocators are secreted, MxiC is released from Spa47 and early effectors secretion follows. In parallel, following the anti-activator OspD1 and MxiC secretion, free IpgC will subsequently bind to the transcriptional activator MxiE to initiate transcription, production and finally secretion of late effectors.
In summary, the present work reports the identification of a novel mechanism involved in effectors sequestration prior to T3S induction. This implies the blockade of the T3SA entry gate upon intimate interaction between the gatekeeper MxiC and MxiI. For the first time, our results depict, a new discrete step, involving MxiC interacting to IpgC to control translocators secretion. Collectively, the present work opens new research avenues for future studies focusing on the real-time localization of MxiC during the major secretion steps. Crystal structure resolution of MxiI (parental versus MxiIQ67A) either alone or in complex with MxiC should aid to understand why the activation signal transmission is blocked in the mxiIQ67A mutant. The first switch in secretion from the needle to translocators is known to be controlled by the Spa40–Spa32 complex and these proteins are conserved among several bacterial T3SSs (Kubori et al., 2000; Edqvist et al., 2003; Botteaux et al., 2008; Lorenz and Buttner, 2011). The switch in substrate specificity in Yersinia involves YscP (Spa32) and YscU (Spa40) who exert their effects on substrate export by controlling YscI secretion (Wood et al., 2008). Interestingly, HrpB2 (MxiI) of the plant pathogen Xanthomonas campestris pv. vesicatoria was shown to interact with the cleaved form of HrcU (Spa40) (Lorenz et al., 2008). More recently, it was suggested that the inner-rod component EscI (MxiI) of EPEC may be involved in substrate secretion switch via its interaction with EscU (Spa40) (Sal-Man et al., 2012). Under our hands, we confirmed the interaction between MxiI and the cleaved forms of Spa40 (data not shown), supporting the broad implication of the rod component in the T3S switches. Thus, it will also be important to address the link between MxiI and the Spa40–Spa32 complex in relation to the first secretion switch from needle to translocators. Answering these queries will certainly improve our knowledge and understanding of the secretion hierarchy mechanisms.
- Top of page
- Experimental procedures
- Supporting Information
Strains and bacterial growth conditions
The list of E. coli, Shigella flexneri and Y. enterocolitica strains used in this study is shown in Table S1. Shigella strains were phenotypically selected on Congo red agar plates (Meitert et al., 1991) and grown in Trypticase Soy Broth (VWR) with the antibiotics at the following concentrations: ampicillin, 100 μg ml−1; kanamycin, 50 μg ml−1; streptomycin, 100 μg ml−1; zeocin, 50 μg ml−1, chloramphenicol, 25 μg ml−1 for E. coli strains and 3 μg ml−1 for Shigella strains. Yersinia strains were grown routinely in BHI (Brain Heart Infusion) Broth or TSA plates at a temperature of 26°C or 37°C respectively.
Construction of the mxiC mxiH and mxiC mxiI double mutants
The suicide vector pMS017 (pGP704-mxiC::ble) (Botteaux et al., 2009) was transferred to the mxiH::aphA3 or mxiI::aphA3 mutant (Blocker et al., 2001) by conjugal mating. Transconjugants were first selected for their resistance to zeocin and streptomycin. Clones in which a double recombination event had exchanged the wild-type mxiC gene by the mutated copy of the pMS017 were identified by screening for sensitivity to ampicillin. The structure of pWR100 derivatives carrying the double mxiH mxiC or mxiI mxiC mutation was confirmed by PCR.
The plasmids and primers used in this study are listed in Tables S1 and S2. Plasmid pAB108′ (pSU18-mxiC), used to complement the mxiC mutant, was constructed by inserting a digested PCR fragment, carrying native mxiC gene, into the EcoRI/SalI sites of the low-copy vector pSU18 (Invitrogen). Plasmid pYC163, encoding both native MxiC and MxiH, was used to complement the mxiC mxiH mutant. This was constructed in two steps. First, plasmid (pSU18-mxiC) was constructed by inserting the mxiC gene obtained by PCR into EcoRI/KpnI digested pSU18. Then, the latter was used to insert a PCR-amplified mxiH gene into the KpnI/SalI sites of pSU18-mxiC. Plasmid pSM6, encoding native MxiI, was used to complement the mxiI mutant. The latter was obtained by cloning mxiI PCR digested DNA fragment into the pSU18 vector.
Recombinant (GST, His or FLAG) proteins, used in the interaction assays, were produced from plasmids listed in Table S1. Plasmids expressing GST fused to MxiI, MxiH, MxiC, IpgC, Spa47, YopN, YopNF223S or PrgJ were constructed by inserting PCR digested DNA fragments into the pGEX4T1 expressing vector. Plasmids expressing His6 fused to MxiC, MxiI, MxiH, InvE and YscI were constructed by inserting PCR digested DNA fragments into the pET30 vector expressing His6 recombinant proteins. Plasmid pYC130 expressing MxiC-FLAG was constructed by inserting a PCR digested DNA fragment into the pUC18K2 plasmid (Menard et al., 1993).
Directed mutagenesis was carried out according to the procedure of the Quick Change Mutagenesis kit (Stratagene). The use of each primer in PCR creates a restriction site (see Table S2) to easily confirm the introduced mutation. Single directed mutagenesis of residues F206S (MxiC) and/or K69A (MxiH) were carried out on plasmids pAB108′ (pSU18-mxiC), pYC163 (pSU18-mxiC-mxiH), pSL30 (pET30-MxiC) and pF13b (pUC19-MxiC) (Botteaux et al., 2009). The resulting plasmids were named pAB130 (pSU18-MxiCF206S), pYC166 (pSU18-MxiCF206S-MxiH), pYC165 (pSU18-MxiC-MxiHK69A), pSL164 (pSU18-MxiCF206S-MxiHK69A), pAB133 (pET30-MxiCF206S) and pAB93 (pUC19-MxiCF206S). Mutations Y185N, M226A, G239A and L245A within MxiC were carried out on pAB108′ (pSU18-mxiC). Mutation Q67A of mxiI was carried out on pSM6 and the resulting plasmid was named pSM33. Plasmid pYC171 encoding native YopN was constructed by inserting a PCR digested DNA fragment encompassing the yopN gene of Y. enterocolitica into the pSU18 EcoRI/SalI digested vector. The latter was used to generate, by directed mutagenesis, plasmid pYC174 encoding YopNF223S.
Crude extracts and culture supernatants of S. flexneri strains were prepared and analysed as previously described (Allaoui et al., 1992). For the induction with Congo red (CR), bacteria were grown until OD600 has reached 2 units, harvested by centrifugation, resuspended in PBS containing 200 μg ml−1 CR, and incubated for 10–30 min at 37°C. Bacteria were centrifuged at 13 000 g for 15 min at 4°C and proteins present in the supernatant were analysed by SDS-PAGE. Western blotting was performed on PVDF membranes (Roche) and developed using chemiluminescence (Perkin Elmer). Immunodetection were carried out as described by Botteaux et al. (2009) using monoclonal antibodies (mAbs) directed against IpaB and His6 motif and a serial of polyclonal antibodies against IpaC, IpaD, IpaA, MxiC, IpaH9.8, OspD1, DnaK, IcsB, GST or the FLAG motif.
Protein production and GST pull-down assays
Escherichia coli BL21 (DE3 Rosetta) were transformed with pGEX4T1 plasmid encoding GST alone or its derivatives encoding GST fusion proteins and grown in 100 ml of LB to OD600 ∼ 0.7 at 37°C. Protein expression was induced with 0.1 mM isopropyl β-d-1-thiogalactopyranoside (IPTG) for 3 h at 37°C. Cells were harvested, resuspended in lysis buffer (40 mM Tris pH 8.0, 500 mM NaCl, 1 mM EDTA) and then lysed by sonication. A final concentration of 1% Triton X-100 was added to the lysate followed by incubation on ice for 20 min. The lysates were then clarified by centrifugation and the supernatants mixed with gluthatione sepharose 4B matrix beads (GE Healthcare) previously equilibrated with PBS buffer. The mixtures were incubated overnight at 4°C in a rotor-shaker with bacterial lysates producing His- or FLAG-tagged recombinant proteins. Beads were washed eight times and proteins eluted by incubating beads 10 min with elution buffer (40 mM Tris pH 8.0, 500 mM NaCl and 50 mM reduced glutathione). The eluted proteins were resolved by SDS-PAGE and analysed by Coomassie blue staining or Western blotting.
Protein folding analysis by fluorescence spectroscopy
Bacteria harbouring plasmids pSL30 and pAB133, expressing His-MxiC and His-MxiCF206S proteins, respectively, were grown overnight at 37°C with shaking (200 r.p.m.). The subculture was then used to inoculate 1 l of LB culture and subsequently grown until the optical density (OD600) of 0.6 was reached. Proteins expression was induced by adding 1 mM IPTG at 37°C during 3 h. Cells were subsequently harvested by centrifugation for 10 min at 8500 r.p.m and the pellets were resuspended in 50 mM Tris-HCl buffer at pH 8.0, containing 500 mM NaCl, 20 mM imidazole and proteases inhibitors cocktail. After lysis by sonication, the unbroken cells were removed with a low speed centrifugation at 8500 r.p.m. for 60 min. The protein mixture was applied to FPLC chromatography on a HisTrap column (5 ml; GE Healthcare) equilibrated with 50 mM Tris-HCl buffer at pH 8.0, containing 500 mM NaCl, 20 mM imidazole. His-MxiC and His-MxiCF206S were eluted with a linear gradient of 0.02–500 mM imidazole in 50 mM Tris-HCl buffer at pH 8.0, containing 500 mM NaCl. The eluted proteins fractions, as revealed by SDS-PAGE, were pooled and subsequently concentrated by ultrafiltration using Centricon® centrifugal concentrator (30 kDa molecular mass cut-off) against 50 mM Tris-HCl buffer at pH 8.0, containing 500 mM NaCl and loaded on a FPLC Superdex200 column (120 ml; GE Healthcare) equilibrated with 50 mM Tris-HCl buffer at pH 8.0, containing 500 mM NaCl. The eluted fractions containing His-MxiC or His-MxiCF206S were pooled and concentrated by ultrafiltration against PBS buffer at pH 7.4 (Fig. 2A). Protein concentration was determined spectrophotometrically using a molar extinction coefficient at 280 nm of 20 400 M−1 cm−1 (Gill and von Hippel, 1989) for purified proteins. A final yield of 3 mg per litre of each purified protein was obtained.
To investigate His-MxiC and His-MxiCF206S folding, the corresponding purified proteins were analysed by fluorescence spectroscopy at a constant temperature of 25.0 ± 0.1°C using a Shimadzu spectrofluorimeter, model RF-5001PC. The fluorescence spectra were obtained with 1 μM final protein concentration. For tryptophan fluorescence measurements, protein solutions were prepared in PBS buffer at pH 7.4. Spectra were recorded between 270 and 500 nm using an excitation wavelength of 285 nm. The excitation and the emission slit widths were 5 nm. The second approach used to assess possible protein misfolding was based on the anilino-8-naphthalenesulphonic acid (ANS, from Sigma-Aldrich, Steinhein, Germany) binding experiments (Azarkan et al., 2006). In aqueous solution and/or in the presence of native or completely unfolded proteins, the hydrophobic dye ANS is characterized by a low quantum yield and an emission λmax at about 520 nm. Partitioning of this probe into the hydrophobic core of a protein, e.g. when it forms a molten globule state, enhances its fluorescence intensity and induces a blue shifting of its maximum wavelength, typically from 520 nm to 470 nm. The molar ratio of protein and ANS used was 1:50. ANS concentration was determined spectrophotometrically using a molar extinction coefficient at 350 nm of 5000 M−1 cm−1. The excitation was set at 370 nm and the emission spectra were measured in the range 400–600 nm using an excitation and an emission slit widths of 5.0 nm. Both, for tryptophan and ANS binding experiments, a solution without protein was used as a blank and the resulting spectrum was used for fluorescence intensity correction. All used solutions were filtered through 0.2 μm.
Yop proteins secretion by Y. enterocolitica
The Yersinia supernatants were prepared as previously described (Heesemann et al., 1986). Briefly, Yersinia was first grown overnight in BHI broth at 26°C with shaking (120 r.p.m.). Next, bacteria were diluted 1:20 in a fresh BHI broth and incubated during 90 min with shaking (120 r.p.m.) at 37°C. After that, the medium was supplemented with EGTA (10 mM) and incubation at 37°C was continued for an additional 90 min. The bacterial cells were then removed by centrifugation (7000 g, at 4°C for 20 min). Proteins from the filtrated culture supernatant were precipitated overnight by addition of ammonium sulphate (40% w/v). After centrifugation (10 000 g at 4°C for 45 min), the resulting pellets were washed and then resuspended in 50 μl of water respectively. Finally, 15 μl of each fraction was resolved by SDS-PAGE and stained with Coomassie blue.
- Top of page
- Experimental procedures
- Supporting Information
The present work was supported by grants from the ‘Fonds National de la Recherche Scientifique – Fonds national Belge de la Recherche Scientifique’ (FRS-FNRS; Convention F.3.4556.11) and from the European Community's Seventh framework programme FP7/2011-2015 under grant agreement No. 261742. Y.C., L.S. and S.M., are/were recipients of a PhD fellowship from the Belgian Fonds National de Recherches Industrielles et Agronomiques (FRIA). L.S. received also a funding from the European Community's Seventh framework programme FP7/2011-2015. A.B., L.B. and A.M. are/were Postodoctoral researchers of the FRS-FNRS. A part of this work was also supported by the Fonds Defay and by the Alice and David Van Buuren foundation. Lastly, we are very grateful to Pr. Guy Cornelis for the gift of the Yersinia strains and to Drs H. Rahmoune, P. Guest and N. El Hajjami for critical reading of the manuscript.
- Top of page
- Experimental procedures
- Supporting Information
- 2005) Chaperone release and unfolding of substrates in type III secretion. Nature 437: 911–915. , and (
- 1992) icsB: a Shigella flexneri virulence gene necessary for the lysis of protrusions during intercellular spread. Mol Microbiol 6: 1605–1616. , , , , and (
- 2011) The Chlamydia effector chlamydial outer protein N (CopN) sequesters tubulin and prevents microtubule assembly. J Biol Chem 286: 33992–33998. , , , , , , et al. (
- 2006) The papaya Kunitz-type trypsin inhibitor is a highly stable beta-sheet glycoprotein. Biochim Biophys Acta 1764: 1063–1072. , , , , and (
- 1997) Secretion of Ipa proteins by Shigella flexneri: inducer molecules and kinetics of activation. Infect Immun 65: 4005–4010. , , and (
- 2007) Preparation and characterization of translocator/chaperone complexes and their component proteins from Shigella flexneri. Biochemistry 46: 8128–8137. , , , , , , et al. (
- 2001) Structure and composition of the Shigella flexneri ‘needle complex’, a part of its type III secreton. Mol Microbiol 39: 652–663. , , , , , , et al. (
- 2008) What's the point of the type III secretion system needle? Proc Natl Acad Sci USA 105: 6507–6513. , , , , , , and (
- 2008) Spa32 interaction with the inner-membrane Spa40 component of the type III secretion system of Shigella flexneri is required for the control of the needle length by a molecular tape measure mechanism. Mol Microbiol 70: 1515–1528. , , , , and (
- 2009) MxiC is secreted by and controls the substrate specificity of the Shigella flexneri type III secretion apparatus. Mol Microbiol 71: 449–460. , , , , and (
- 2000) The virulence plasmid pWR100 and the repertoire of proteins secreted by the type III secretion apparatus of Shigella flexneri. Mol Microbiol 38: 760–771. , , , , , , et al. (
- 2012) Protein export according to schedule: architecture, assembly, and regulation of type III secretion systems from plant- and animal-pathogenic bacteria. Microbiol Mol Biol Rev 76: 262–310. (
- 2001) Regulated secretion of YopN by the type III machinery of Yersinia enterocolitica. J Bacteriol 183: 5293–5301. , , and (
- 2006) The type III secretion injectisome. Nat Rev Microbiol 4: 811–825. (
- 1998) A complex composed of SycN and YscB functions as a specific chaperone for YopN in Yersinia pestis. Mol Microbiol 30: 777–788. , and (
- 2008) Structures of the Shigella flexneri type 3 secretion system protein MxiC reveal conformational variability amongst homologues. J Mol Biol 377: 985–992. , , , , and (
- 2010) Timing is everything: the regulation of type III secretion. Cell Mol Life Sci 67: 1065–1075. , , , and (
- 1998) Induction of type III secretion in Shigella flexneri is associated with differential control of transcription of genes encoding secreted proteins. EMBO J 17: 2894–2903. , , and (
- 2003) YscP and YscU regulate substrate specificity of the Yersinia type III secretion system. J Bacteriol 185: 2259–2266. , , , , , , and (
- 2004) Disabling surveillance: bacterial type III secretion system effectors that suppress innate immunity. Cell Microbiol 6: 1027–1040. , and (
- 2005) Selection and characterization of Yersinia pestis YopN mutants that constitutively block Yop secretion. Mol Microbiol 57: 970–987. , , , and (
- 1991) The surface-located YopN protein is involved in calcium signal transduction in Yersinia pseudotuberculosis. Mol Microbiol 5: 977–986. , , , and (
- 1989) Calculation of protein extinction coefficients from amino acid sequence data. Anal Biochem 182: 319–326. , and (
- 1986) Immunochemical analysis of plasmid-encoded proteins released by enteropathogenic Yersinia sp. grown in calcium-deficient media. Infect Immun 54: 561–567. , , , and (
- 2007) Identification of TyeA residues required to interact with YopN and to regulate Yop secretion. Adv Exp Med Biol 603: 235–245. , and (
- 2005) The needle component of the type III secreton of Shigella regulates the activity of the secretion apparatus. J Biol Chem 280: 42929–42937. , , , , , , and (
- 2002) Assembly of the type III secretion needle complex of Salmonella typhimurium. Microbes Infect 4: 75–82. , and (
- 1999) Global burden of Shigella infections: implications for vaccine development and implementation of control strategies. Bull World Health Organ 77: 651–666. , , , , , , et al. (
- 2002) Salmonella type III secretion-associated protein InvE controls translocation of effector proteins into host cells. J Bacteriol 184: 4699–4708. , and (
- 2000) Molecular characterization and assembly of the needle complex of the Salmonella typhimurium type III protein secretion system. Proc Natl Acad Sci USA 97: 10225–10230. , , , and (
- 2011) A sorting platform determines the order of protein secretion in bacterial type III systems. Science 331: 1188–1191. , , , , and (
- 2011) Secretion of early and late substrates of the type III secretion system from Xanthomonas is controlled by HpaC and the C-terminal domain of HrcU. Mol Microbiol 79: 447–467. , and (
- 2008) HpaC controls substrate specificity of the Xanthomonas type III secretion system. PLoS Pathog 4: e1000094. , , , , , and (
- 2002) Spa32 regulates a switch in substrate specificity of the type III secreton of Shigella flexneri from needle components to Ipa proteins. J Bacteriol 184: 3433–3441. , , , , , , and (
- 2004) Structural insights into the assembly of the type III secretion needle complex. Science 306: 1040–1042. , , , , , and (
- 2006) Assembly of the inner rod determines needle length in the type III secretion injectisome. Nature 441: 637–640. , , , , , and (
- 1988) pACYC184-derived cloning vectors containing the multiple cloning site and lacZ alpha reporter gene of pUC8/9 and pUC18/19 plasmids. Gene 68: 159–162. , (
- 2010) The Shigella T3SS needle transmits a signal for MxiC release, which controls secretion of effectors. Mol Microbiol 78: 1365–1378. , and (
- 2002) Regulation of transcription by the activity of the Shigella flexneri type III secretion apparatus. Mol Microbiol 43: 1543–1553. , , , , , and (
- 1991) Correlation between Congo red binding as virulence marker in Shigella species and Sereny test. Roum Arch Microbiol Immunol 50: 45–52. , , , , , and (
- 1993) Nonpolar mutagenesis of the ipa genes defines IpaB, IpaC, and IpaD as effectors of Shigella flexneri entry into epithelial cells. J Bacteriol 175: 5899–5906. , , and (
- 1994) The secretion of the Shigella flexneri Ipa invasins is activated by epithelial cells and controlled by IpaB and IpaD. EMBO J 13: 5293–5302. , , and (
- 2008) The versatility of Shigella effectors. Nat Rev Microbiol 6: 11–16. , , , , and (
- 2007) Bile salts stimulate recruitment of IpaB to the Shigella flexneri surface, where it colocalizes with IpaD at the tip of the type III secretion needle. Infect Immun 75: 2626–2629. , , , , , and (
- 1999) The secreted IpaB and IpaC invasins and their cytoplasmic chaperone IpgC are required for intercellular dissemination of Shigella flexneri. Cell Microbiol 1: 183–193. , , , and (
- 2005) Bioinformatics analysis of the locus for enterocyte effacement provides novel insights into type-III secretion. BMC Microbiol 5: 9. , , and (
- 2009) Shigella type III secretion effectors: how, where, when, for what purposes? Curr Opin Microbiol 12: 110–116. (
- 1995) Enhanced secretion through the Shigella flexneri Mxi-Spa translocon leads to assembly of extracellular proteins into macromolecular structures. Mol Microbiol 16: 291–300. , , , and (
- 2005) 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 56: 1627–1635. , , , , , , et al. (
- 2005) Manipulation of the host actin cytoskeleton by Salmonella – all in the name of entry. Curr Opin Microbiol 8: 10–15. , and (
- 2008) The Chaperone IpgC copurifies with the virulence regulator MxiE. J Bacteriol 11: 11. , and (
- 2012) EscI: a crucial component of the type III secretion system forms the inner rod structure in enteropathogenic Escherichia coli. Biochem J 442: 119–125. , , and (
- 1998) The Yersinia Yop virulon: LcrV is required for extrusion of the translocators YopB and YopD. J Bacteriol 180: 1207–1214. , , , and (
- 2008) Biochemical functions of Yersinia type III effectors. Curr Opin Microbiol 11: 21–29. (
- 2007) The type III secretion system needle tip complex mediates host cell sensing and translocon insertion. Mol Microbiol 63: 1719–1730. , , , , , and (
- 1994) YscN, the putative energizer of the Yersinia Yop secretion machinery. J Bacteriol 176: 1561–1569. , , , and (
- 2008) YscP and YscU switch the substrate specificity of the Yersinia type III secretion system by regulating export of the inner rod protein YscI. J Bacteriol 190: 4252–4262. , , and (
- Top of page
- Experimental procedures
- Supporting Information
Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.