SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Getting the type III secretion system (T3SS) genes ready
  4. Imaging the assembly of the T3SS apparatus
  5. Switching from T3SS assembly to the standby mode
  6. Switching from standby mode to the injection of bacterial effectors
  7. Imaging the injection mode
  8. Post-injection events and conclusion
  9. Acknowledgements
  10. References

The type III secretion system (T3SS) is a sophisticated molecular machinery of Gram-negative bacteria used to ‘inject’ (translocate) bacterial proteins (effectors) into eukaryotic cells. For this, the T3SS has to assemble into a multiprotein complex, which is constituted of distinct parts; a basal body spanning the two bacterial membranes connected with a cytoplasmic bulb, an attached needle structure resembling a molecular syringe, and a distal needle tip structure that re-organizes into a ‘translocon’, which is a protein complex that inserts into the host cellular membrane. Upon engaging with eukaryotic cells, the T3SSs perform ‘single-step’ translocation of bacterial effector proteins across three membranes (two bacterial and one eukaryotic). Since the formulation of the major concepts of the T3SS about 15 years ago, imaging has been a major ingredient for elucidating the T3SS structure and function. Direct observation of molecular events in the context of cells will remain a key feature for better understanding of T3SS structure, regulation and function. In this review we describe how light and electron microscopy have been used to shed light on the processes of maturation and activity of the T3SS. Furthermore, we highlight recent imaging innovations with the potential to provide better insight into the T3SS structure and function.


Getting the type III secretion system (T3SS) genes ready

  1. Top of page
  2. Summary
  3. Getting the type III secretion system (T3SS) genes ready
  4. Imaging the assembly of the T3SS apparatus
  5. Switching from T3SS assembly to the standby mode
  6. Switching from standby mode to the injection of bacterial effectors
  7. Imaging the injection mode
  8. Post-injection events and conclusion
  9. Acknowledgements
  10. References

The assembly of the type III secretion system (T3SS) is tightly regulated at multiple levels, starting with its gene expression. Typically, the T3SS and factors with related functions are encoded by > 30 genes which are frequently co-regulated. This has been studied in detail, mainly using genetic and biochemical approaches (see dedicated reviews like Brutinel and Yahr, 2008). Little has been reported on the direct imaging of T3SS gene expression, particularly in real time. Nevertheless, a number of studies have exploited light to investigate T3SS processes in vivo (Lee and Camilli, 2000). For example, bioluminescence imaging demonstrated the induction of T3SS effector protein expression upon host cellular contact in Yersinia (Pettersson et al., 1996). Further, gene expression of T3SS constituents has been analysed during the internalization process of Salmonella into macrophages (Cirillo et al., 1998). To achieve this, fusions with GFP were constructed throughout the entire gene cluster in the Salmonella SPI2 T3SS. These GFP fusions demonstrated that SPI-2 genes encoding structural, regulatory and secreted proteins are preferentially expressed in the intracellular environment of the host macrophage (Cirillo et al., 1998). The usage of both bioluminescence and fluorescence reporter systems is very valuable for the exploration of the induction of the T3SS constituents, particularly during in vivo infections.

Imaging the assembly of the T3SS apparatus

  1. Top of page
  2. Summary
  3. Getting the type III secretion system (T3SS) genes ready
  4. Imaging the assembly of the T3SS apparatus
  5. Switching from T3SS assembly to the standby mode
  6. Switching from standby mode to the injection of bacterial effectors
  7. Imaging the injection mode
  8. Post-injection events and conclusion
  9. Acknowledgements
  10. References

The assembly of the T3SS apparatus is dependent upon hierarchical, T3SS-mediated, delivery of its components to the assembly site. Thus, the fully assembled apparatus can be viewed as a structure that occurs in a tightly regulated series of events (Fig. 1). Electron Microscopy (EM) studies have been pivotal to decipher the details of T3SS assembly. This allowed the depiction of multiple T3SS structures emphasizing the morphological similarities among the various T3SS containing bacteria (e.g. Kubori et al., 1998; Blocker et al., 2001; Hoiczyk and Blobel, 2001; Sekiya et al., 2001). Taken together, T3SS assembly appears to be a conserved, highly organized and inducible process. The assembled T3SS is capable of translocating effector proteins into targeted host cells through a hollow needle complex with an inner diameter of about 25 Ångström (Å) (Cornelis, 2006). Comparing EM micrographs of structures isolated from wild-type bacteria and mutants in genes encoding T3SS components has been the main source of experimental evidence during these studies (Sukhan et al., 2001). Subsequent high-resolution imaging on intact T3SS structures via cryo-EM and computational single-particle analysis revealed the overall organization during its assembly (Blocker et al., 2001; Marlovits et al., 2004; Sani et al., 2007a). Recently, cryo-EM studies combined with scanning transmission EM showed the 12-fold geometric structure at 21–25 Å resolution of the Shigella basal body, its symmetry, and it demonstrated that the needle complex is connected with the basal body via a molecular socket (Fig. 1) (Hodgkinson et al., 2009). Furthermore, Spreter et al. combined structural information obtained by X-ray crystallography on fragments of the inner and outer T3SS membrane ring constituents EscC (from EPEC) and PrgH (from Salmonella), computational approaches and EM data to develop ring models of the T3SS. This revealed their modular structures, and the authors propose that these proteins contain a ‘ring-building’ motif (Spreter et al., 2009). Reports like this demonstrate how much single-particle analyses of electron micrographs have been profiting from data derived by other methods, particularly by X-ray crystallography (reviewed in Moraes et al., 2008). For example, structural data of T3SS constituents have been used to determine the symmetry of the T3SS (Andre et al., 2007).

image

Figure 1. Imaging the type III secretion system (T3SS). A. Schematics of the T3SS based on data from single-particle analysis of purified needle complexes that were imaged by EM. The major structural building blocks are depicted within the image. The needles are about 60 nm long, and the inner and outer ring (including the connector) measures 30 nm. The central channel of the needle measures about 2–3 nm. B. Fluorescence microscopy allows analysis of some T3SS features. Here, the EspA filaments of the T3SS of enterohaemorrhagic Escherichia coli (EHEC) have been imaged. EHEC EDL933 were used to infect EPH-4 epithelial cells. Bacteria were stained with anti-O157 (green), cells with DAPI (blue) and the T3SS EspA extensions, projecting from the infected bacteria and connecting them to the host cells, were stained with anti-EspA antibody (red). The EspA filaments can grow up to 700–800 nm. Scale bar: 5 μm. C. The Shigella flexneri T3SS needle complexes were isolated, and imaged by low-dose electron microscopy. It is of note that some needle complexes are standing upright, and others are lying on the grids. Such electron micrographs have been used to reconstruct the overall structure of the needle complex and symmetry, including its key building blocks that are shown in (A). The image has been kindly provided by Dr A. Blocker. Scale bar: 120 nm. D. Reconstruction of S. flexneri T3SS needle complexes (data set EMD-1617) as determined in Dr Ariel Blocker's laboratory. The data set is displayed using Chimera. Scale bar: 10 nm. Figure 1C and 1D were kindly provided by Dr. Ariel Blocker.

Download figure to PowerPoint

Cryo-EM imaging has also revealed the structure of the core body of the type IV secretion system revealing a 14-fold symmetry at 14 Å resolution that differs largely from the T3SS (Fronzes et al., 2009). Even though this approach allows very good particle preservation of samples close to the native state and yields very high resolution of the observed objects in the range of several angstroms, a major challenge and limitation results from the difficulties in generating unbiased models for the particle reconstruction. To achieve this, several thousands of identical particles of a biological object need to be imaged. One challenge is the determination of the specific particle symmetry of the object that may even result in loss of asymmetric information. Therefore, it would be interesting to complement cryo-EM studies on the T3SS with cryo-tomography studies that feature a lower resolution (around 4 nm), however, resolve single objects without the need of averaging and highlight asymmetric features of the imaged particles within their overall structural organization (Grunewald et al., 2003).

A major question regarding the process of T3SS assembly is how the system determines how much to deliver of each of the components, and how it stops the delivering of a given component and initiates the delivery of the next component. For instance, why the length of the needle, the outmost projecting part of the needle complex (or hook in the case of the closely related flagella), is constant? This problem was investigated using EM analysis the T3SSs of Yersinia, Salmonella and Shigella, leading to the formulation of three different hypotheses of needle length control. A ‘switch or measuring-cup model’ has been proposed mainly based on studies on flagella assembly (Makishima et al., 2001). Direct length measurements via EM of the length of shed needles from Yersinia suggested that it is directly controlled via a mechanism evoking a molecular ruler (Journet et al., 2003). Finally, a ‘socket’ model has been proposed based on electron cryo-microscopy taking into account the connection between the T3SS needle attachment to the basal body (Marlovits et al., 2006). For detailed description of these three hypotheses we refer the readers to recent reviews on this subject (Cornelis, 2006; Cornelis et al., 2006; Waters et al., 2007).

The EM analysis approach has limitations. Apart from the points made above concerning cryo-microscopy and single-particle analysis, sample preparation and the construction of ultra-resolution images are technically challenging. In addition, EM cannot be use to study the dynamic of T3SS assembly in live cells, even though future developments correlating EM and light microscopy may change this (see below). The needle length is 50 and 80 nm and therefore imaging its dynamics by light microscopy appears impossible due to the Abbe spatial resolution limit. We predict, however, that the development of new light microscopy technologies like STED, PALM or STORM, which lower the resolution limit to under 50 nm (see Rust et al., 2006; Shroff et al., 2008; Hell, 2009 for details on these methodologies), will have a tremendous impact on the study of the T3SS and microbial cell biology at large. Main challenges using these super-resolution technologies include the establishment of protocols that will allow the imaging of multiple factors in one sample, and to monitor dynamic events in real time in living samples. Very recently, PALM has been used to monitor the distribution of bacterial components involved in the chemotaxis signalling events at a resolution of 25 nm in fixed samples (Greenfield et al., 2009). This showed that the self-assembly of the constituents results in periodic structures at the bacterial membranes without cytoskeletal support (Greenfield et al., 2009). Furthermore, the dynamic interaction of bacterial chemotaxis signalling pathway constituents CheY, CheZ and CheA have been revealed by dynamic fluorescence imaging showing their relocalization upon signalling induction (Khan et al., 2000; Sourjik and Berg, 2000).

Of note, the EPEC model has already given interesting insight into the assembly process, using light microscopy without breaking the resolution limit. The T3SSs of the EPEC are fitted with the EspA filaments that are attached to the distal end of the T3SS and that were visualized by fluorescence microscopy (Fig. 1) (Knutton et al., 1998). It has been shown that novel subunits are integrated into the growing structure at its tip (Crepin et al., 2005). An EspA filament can grow to several hundred nanometres in length, and it is connected with the basal body via the stiff EscF needle (Sekiya et al., 2001). Importantly, these studies can also be performed in the context of host cellular contact revealing the re-arrangements of the EspA filaments during protein translocation (Crepin et al., 2005; Berger et al., 2009). The EspA filaments shrink and re-localize during the formation of pedestals. Hence, it will be exciting to use dynamic fluorescence imaging to study the underlying molecular mechanism of these events.

Light microscopy can also be performed in conjunction with electron microscopy via correlative light electron microscopy (CLEM) (Sartori et al., 2007; Plitzko et al., 2009; Razi and Tooze, 2009). For this technique the specimen are first grown and analysed by light microscopy on coverslips that contain a grid to allow the exact positioning of individual cells or bacteria. Subsequently, the samples are prepared for electron microscopy with the aim to obtain ultra-resolution on the identical specimen. Such approaches have a tremendous potential to study the T3SS assembly and its interaction with host cells (see below).

Switching from T3SS assembly to the standby mode

  1. Top of page
  2. Summary
  3. Getting the type III secretion system (T3SS) genes ready
  4. Imaging the assembly of the T3SS apparatus
  5. Switching from T3SS assembly to the standby mode
  6. Switching from standby mode to the injection of bacterial effectors
  7. Imaging the injection mode
  8. Post-injection events and conclusion
  9. Acknowledgements
  10. References

Upon assembly completion the T3SS switches into a state that can be termed as ‘standby’ mode. In this mode the T3SS is fully assembled and ready for engagement with the host cell membrane and effector injection, but effectors are not secreted to the medium (or only at minimal rates), although translocon components are frequently released, albeit at rates lower than recorded upon engagements with the host cells (Menard et al., 1994; Wolff et al., 1998). Importantly, mutants in genes encoding the members of the YopN/InvE/SepL-SepD/MxiC protein family exhibit uncontrolled, constitutive secretion of effector proteins (O'Connell et al., 2004; Deng et al., 2005; Wang et al., 2008a; Botteaux et al., 2009). In some cases the constitutive secretion of effectors is associated with partial or complete inhibition in secretion/release of translocon proteins (O'Connell et al., 2004; Deng et al., 2005; Wang et al., 2008a). These reports support the hypothesis that specific proteins are required for either switching from the assembly mode to the standby mode and/or for the maintenance of the standby mode. Whereas the process of switching to the standby mode is still elusive and it is not clear how to image it, most of the T3SS static image analysis has been carried out at this standby state.

Switching from standby mode to the injection of bacterial effectors

  1. Top of page
  2. Summary
  3. Getting the type III secretion system (T3SS) genes ready
  4. Imaging the assembly of the T3SS apparatus
  5. Switching from T3SS assembly to the standby mode
  6. Switching from standby mode to the injection of bacterial effectors
  7. Imaging the injection mode
  8. Post-injection events and conclusion
  9. Acknowledgements
  10. References

Upon infection of cells, polar secretion of the effectors into the host cell, but not into the medium was shown (Vallis et al., 1999; Sundin et al., 2004; Jaumouille et al., 2008; Mills et al., 2008). This implies that only the injectors that engage with the host cell membrane are activated upon bacteria–host cell contact, whereas the non-engaged injectors remain in the standby mode (Vallis et al., 1999; Sundin et al., 2004). The native trigger that activates the T3SS to inject effectors appears to be the contact with the host cell membrane, but the T3SS can also be artificially activated by treatment with specific compounds or by applying specific experimental conditions. For instance, the Shigella T3SS can be triggered by treatment with Congo Red, fetal bovine serum or cholesterol-containing vesicles (van der Goot et al., 2004), and that of Yersinia by depleting the culture medium of Ca2+ ions.

The needle ‘tip complex’ has been implicated in sensing these inducers leading to the T3SS activation (Veenendaal et al., 2007; Mueller et al., 2008). Ultra-resolution studies by scanning and transmission EM have been crucial in revealing the presence of these tip complexes (Mueller et al., 2005; Broz et al., 2007; Sani et al., 2007b; Veenendaal et al., 2007). Comparing EM images of needle tips from various bacteria revealed different tip structures. Purification of Shigella needle complexes in the presence of a cross-linker revealed two large globular structures on each side of the T3SS needle (Sani et al., 2007b), hence further studies will be required to determine the exact composition of these structures. In comparison with these globular complexes, the tip of Yersinia appears to be smaller (Mueller et al., 2005). The precise structure of this tip complex is still disputed, and it has been suggested that it consists either of one constituent (LcrV in the case of Yersinia) or of at least two constituents (IpaD and IpaB in the case of Shigella) (Broz et al., 2007; Olive et al., 2007; Veenendaal et al., 2007). The presence of translocon constituents within the tip complex is intriguing because it underlines its potential active role in switching to the injection mode. To reveal the molecular details of these processes it will be important to image how the needle complex is connected with the translocon complex upon engaging with host cellular membranes or upon contact with model membranes. So far, the number of reports that have attempted the direct visualization of the translocon complex remains rather limited. One EM study on Pseudomonas has investigated the morphology of the translocon complex after insertion into model membranes (Schoehn et al., 2003). The obtained electron micrographs revealed their ring-like structures but could not detect precise stoichiometries. Combination with other imaging technologies could improve this. For example, atomic force microscopy has proposed a pore-shaped structure of the translocon complex after insertion into model membranes (Ide et al., 2001).

The reporter systems developed for effector injection (see next part below) will also be useful to shed light on how the T3SS senses specific triggers, and how it switches from a standby mode to active injection. A major advantage will be the option of dynamic measurements. For example, direct visualization of effector secretion could be performed with bacteria grown in microfluidic systems that can track changes in the bacterial environment to allow a precise timing of secretion induction (Balagadde et al., 2005).

During the triggering of the T3SS, the needle tip complex is structurally modified, presumably mimicking its interaction with host cellular membranes. The direct interaction of the translocon components with cholesterol highlights the involvement of lipid rafts and lipid sensing in triggering the injection mode (Lafont et al., 2002; van der Goot et al., 2004; Hayward et al., 2005). These re-arrangements at the needle tips are under investigation combining ultra-resolution, biochemical, structural and microscopic approaches (Broz et al., 2007; Olive et al., 2007; Veenendaal et al., 2007).

The localization of the T3SS at specific sites on the bacterial surface might also affect the secretion and invasion process. For instance, injection of effectors by Shigella appears to be mediated by T3SSs located at the bacterial poles at the initial step of host cell interaction within the first 5 min (Jaumouille et al., 2008). This has been imaged using the GFP or tetracystein-4Cys-labelled effector/translocator protein IpaC accumulating within the bacterium at the pole that is in contact with the host cell during the onset of effector injection (Jaumouille et al., 2008). Interestingly, the T3SS apparatuses were shown to be distributed all over the bacterial surface, and their embedding into the peptidoglycan certainly restrains their lateral mobility (Blocker et al., 1999). Therefore, the development of biosensors highlighting the activity state of individual T3SS needles with regard to their exact localization (at a pole or not) should give better information on the dynamic activity state of the T3SS. Such studies could also be combined with the dynamic recruitment of cargo to the type III system (e.g. as described for chemotaxis signalling in Sourjik et al., 2007). Alternatively, a large number of fluorescent biosensors have been developed for eukaryotic signalling pathways, for example to monitor the activity state and localization of important messenger molecules, such as small GTPases or kinases (Wang et al., 2008b).

Imaging the injection mode

  1. Top of page
  2. Summary
  3. Getting the type III secretion system (T3SS) genes ready
  4. Imaging the assembly of the T3SS apparatus
  5. Switching from T3SS assembly to the standby mode
  6. Switching from standby mode to the injection of bacterial effectors
  7. Imaging the injection mode
  8. Post-injection events and conclusion
  9. Acknowledgements
  10. References

The cytoplasmic face of the T3SS selects proteins carrying specific translocation signals for injection. Although the nature of the T3SS translocation signals is still controversial, its localization to the protein's N-terminus is very well established (Ghosh, 2004). This allows manipulation of the effector C-termini without interfering with the translocation signals and translocation processes. Using antibodies against effectors or epitope-tagged effectors is the most direct approach to detect translocation, but this method is limited in sensitivity and difficult to quantify. Thus, several innovative reporter systems have been developed to improve the detection and imaging of the effector protein translocation process. This approach was pioneered by Sory and Cornelis (1994) using the catalytic domain of the Bordetella pertusis adenylate cyclase toxin, CyaA, as a reporter gene. This technology is dependent on the specific unleashing of the CyaA activity upon its translocation into host cells and calmodulin binding. The read-out for effector–CyaA entry into cells is the increase in cAMP concentration in the host cell extracts, which is proportional to the amount of injected effector–CyaA protein fusion. Later, Day et al. (2003) developed a method involving a fusion with the phosphorylatable Elk peptide fused to the nuclear localization signal (NLS) from the large T antigen of simian virus 40. The NLS sequence directs the fusion protein to the cell nucleus where the Elk tag is phosphorylated. The translocated protein can be detected by Western blotting with phospho-specific Elk peptide antibodies.

Recently, Charpentier and Oswald (2004) developed a system based on using the catalytic domain of the TEM-1 β-lactamase (BlaM) as reporter gene (see Fig. 2 for details on the experimental principles). This technology was initially used to detect effector translocation into tissue culture cells (Charpentier et al., 2004) and more recently it was used to address the question of which cells are targeted in vivo by the bacteria for injection (Marketon et al., 2005; Geddes et al., 2007). Mills et al. (2008) applied the same blaM reporter gene to study the dynamic of the T3SS activity (Fig. 2). In this study the infection was performed in a 96-well plate and carried out in a fluorimetric plate reader equipped with temperature control allowing real-time reading of the kinetics of fluorescent shift from green to blue, which represent the rates of reduction in substrate concentration and product accumulation. Since the BlaM β-lactamase (TEM-1) follows a simple Michaelis–Menten enzymatic kinetics model and its catalytic constants (Kcat and Km) to CCF2 were determined (Zlokarnik, 2000), the rate of the enzymatic activity in the cells can be used to make quantitative estimations as to the average rate of protein injection, and concentrations in the injected cells (Mills et al., 2008).

image

Figure 2. The β-lactamase-based, protein translocation analysis. A. The β-lactamase substrate: CCF2 (or CCF4) consists of a cephalosporin core linking 7-hydroxycoumarin to fluorescein (Zlokarnik et al., 1998). Excitation of the coumarone at 405 nm results in fluorescence resonance energy transfer (FRET) to the fluorescein moiety, which emits at 535 nm (green). Cleavage of the substrates by β-lactamase results in the disruption of FRET and excitation of the coumarin at 409 nm, which results in emission at 450 nm (blue). B. Schematic of the assay design. Bacteria expressing effectors fused to β-lactamase are used to infect cells, following staining with CCF2, or as shown here, the infected cells were pre-lauded with CCF2 (green) and upon injection of the effector–β-lactamase into the cells, they shifts from green to blue mission. C. Typical results of a translocation assay. Cells seeded in 96-well plate and loaded with CCF2 were infected with EPEC expressing effector fused to the β-lactamase. Each well was inoculated with different EPEC strain, each containing different effector fused to the β-lactamase reported gene. The entire infection process was carried out in this plate reader. The rates of β-lactamase activity (product acomulation), reflecting the amount of the effector–β-lactamase proteins in the infected cells, was determined simultaneously for all the effectors (Tir, EspF, EspZ, Map, EspH and EspG).

Download figure to PowerPoint

The above described assay determines the average translocation efficiency using large population of bacterial and host cells, which is an advantage when using uniform populations. However, the infecting bacterial populations are not necessarily uniform. For instance, it has been recently reported that only ∼30% of the infecting Salmonella express the SPI-1 T3SS under conditions used for invasion of tissue cultured cells (Ackermann et al., 2008). Thus, some questions can only be addressed by single-cell analysis. Schlumberger et al. (2005) developed such an assay based on the specific and strong interaction between effectors and their cogent chaperons. Translocation of Salmonella typhimurium SipA was imaged using host cells expressing their respective specific chaperon InvB fused to GFP. Entry of the effectors into the host cells was tracked by the rapid recruitment of GFP–InvB to the site of bacterial attachment to result in the accumulation of the GFP–InvB (Schlumberger et al., 2005). A modified approach via indirect immunofluorescence on fixed samples was used to determine the translocation of specific effectors (SipA and SptP) by a single bacterium and suggested the hierarchical effector protein transport by the S. typhimurium SPI-1 T3SS (Winnen et al., 2008).

A different approach for single-cell translocation analysis was developed for Shigella with the aim to directly label the injected effectors fluorescently (Enninga et al., 2005). Labelling of the effectors with the bulky fluorescent proteins, like GFP, generally blocks the translocation of the labelled effectors and clogs the T3SS as a whole. To circumvent this, effectors are generated tagged at their C-terminus or internally with a small tetracysteine motif (4Cys) that are subsequently used to complement bacterial mutant strains for these effectors (Griffin et al., 1998; Enninga et al., 2005). Then, the bacteria are labelled with a fluorescein-based biarsenical dye (FlAsH) allowing fluorescence labelling of the intrabacterial effectors. Upon host cell contact the FlAsH-labelled effectors are injected into the cells, and thus the rate of the reduction of the fluorescent signal within the injecting bacterium was correlated with the rate of effector injection. This analysis requires a rapid fluorescence microscopy set up for three-dimensional image acquisition over time. With this approach, the secretion of the two Shigella translocator/effectors IpaB and IpaC was found to be very rapid and complete with 50% of both intrabacterially stored effectors secreted in about 4 min. Recently this method was significantly improved by Van Engelenburg and Palmer (2008) and applied to determine real-time effector translocation by Salmonella and effector localization after translocation.

Post-injection events and conclusion

  1. Top of page
  2. Summary
  3. Getting the type III secretion system (T3SS) genes ready
  4. Imaging the assembly of the T3SS apparatus
  5. Switching from T3SS assembly to the standby mode
  6. Switching from standby mode to the injection of bacterial effectors
  7. Imaging the injection mode
  8. Post-injection events and conclusion
  9. Acknowledgements
  10. References

What is happening with the T3SS after it has injected its load into the host cells? Can the structures be re-activated? Are they re-cycled after usage? It has been shown that multiple pathogens require the T3SS for a number of subsequent steps during the infection process. For example, Shigella requires active type III secretion for cellular invasion, but requires its components also for cellular spread (Page et al., 1999). Also, some pathogens, like Salmonella, express multiple T3SSs that are involved in specific steps of bacterial infection (Cirillo et al., 1998). Therefore, it will be interesting to investigate the dynamic regulation, and the activity of the T3SS during the infection process. Experimentally, this is challenging because it requires read-outs that can detect the subsequent steps of infection and that can correlate this with the T3SS activity. We are convinced that a combination of technological improvements could potentially allow such studies. Particularly, the usage of multiphoton microscopy or rapid spinning disc confocal microscopy will be at the centre of such studies because they allow the bridging of events at the organ and at the cellular level (Frischknecht et al., 2006; Melican and Richter-Dahlfors, 2009). Additionally, the development of reporter assays with microscopic read-outs will be important to maximize the parameters that can be measured simultaneously concerning the T3SS. The combination of multiple assays described in this review, for example the induction of gene expression of T3SS constituents and the tracking of T3SS activity, will bring interesting insights into the precise co-ordination of this process. The rapid technological developments and the creative design of experimental approaches will be at the core to achieve these aims.

Acknowledgements

  1. Top of page
  2. Summary
  3. Getting the type III secretion system (T3SS) genes ready
  4. Imaging the assembly of the T3SS apparatus
  5. Switching from T3SS assembly to the standby mode
  6. Switching from standby mode to the injection of bacterial effectors
  7. Imaging the injection mode
  8. Post-injection events and conclusion
  9. Acknowledgements
  10. References

Work in the laboratories of Jost Enninga and Ilan Rosenshine is supported by the Institut Pasteur and the Fondation Schlumberger pour la Education et la Recherche (IE) and a grant from the Israeli Academy of Science and Humanities (IR). We thank Pina Colarusso and Agathe Subtil for critical reading of the manuscript, Anna Sartori for input concerning the EM parts of this review, and N.Y. Shpigel, L. Golan and A. Blocker for providing Fig. 1B and C. We apologize in advance for limited citation. Many excellent reports were not cited due to space constrains.

References

  1. Top of page
  2. Summary
  3. Getting the type III secretion system (T3SS) genes ready
  4. Imaging the assembly of the T3SS apparatus
  5. Switching from T3SS assembly to the standby mode
  6. Switching from standby mode to the injection of bacterial effectors
  7. Imaging the injection mode
  8. Post-injection events and conclusion
  9. Acknowledgements
  10. References