Limiting Too Much of a Good Thing: a Negative Feedback Mechanism Prevents Unregulated Translocation of Type III Effector Proteins

Type III secretion systems (T3SS) are highly specialized structures termed injectisomes which function by translocating bacterial effector proteins via hollow needles directly into eukaryotic target cells. The translocated effectors alter target cell physiology to promote the pathogenic and symbiotic lifestyles of many gram-negative bacterial species. Orchestrating T3SS activity requires multiple levels of regulatory control to ensure appropriate timing of gene expression, needle complex assembly, pore formation in the target cell membrane, and translocation of the effectors into target cells. Most of the work thus far has focused on regulatory events leading up to and including translocation of the effectors. Although fewer studies have examined regulatory events that occur following translocation of the effectors into target cells, recent evidence indicates that translocation in Yersinia pseudotuberculosis is a regulated event that is controlled by the translocated YopE effector (1). In this issue of the Journal of Bacteriology, Cisz et al. demonstrate that a similar negative feedback mechanism operates in Pseudomonas aeruginosa (3). Using an elegant microscopy-based assay, they provide evidence that the autolimiting effect on translocation prevents subsequent bacteria from deploying the T3SS against the same target cell. In a separate line of investigation, they propose that host contact-mediated activation of translocation involves a shift in the substrate specificity of the injectisome and provide evidence that the signal associated with target cell contact is not the low-Ca²⁺ environment of the host cell cytoplasm.

Arming and subsequent activation of the T3SS.

The P. aeruginosa T3SS elicits cytotoxicity toward eukaryotic target cells through the activity of four translocated effector proteins (ExoS, ExoT, ExoU, and ExoY) (2, 13, 18). Translocation of the effectors is dependent upon ∼36 genes that encode proteins comprising the needle complex, translocators, chaperones, and regulators (5). These genes are transcribed at a basal level in the absence of inducing signals, allowing for assembly of one or two needle complexes per cell (11). As a result, P. aeruginosa is preloaded with small amounts of the effector proteins and is poised to rapidly deploy the T3SS.

The T3SS remains in standby mode until activation signals trigger secretion and/or translocation of the effector and regulatory proteins and up-regulation of T3SS gene expression. The primary activating signals are growth under Ca²⁺-limiting conditions or contact of P. aeruginosa with target cells (17). Activation by the low-Ca²⁺ signal was previously thought to trigger secretion of the effector proteins as well as the translocators (PcrV, PopB, and PopD) and a negative regulator of T3SS gene expression (ExsE) (9, 16). Secretion of ExsE into the growth medium or its translocation into target cells results in a concomitant decrease in the bacterial intracellular concentration of ExsE, thereby relieving the block on T3SS gene expression (16). The study by Cisz et al. presents evidence that whereas secretion of ExsE and the effector proteins is dependent upon the low-Ca²⁺ signal, secretion of the translocator proteins (PcrV, PopB, and PopD) occurs irrespective of Ca²⁺ levels (3). Thus, rather than generally activating the P. aeruginosa injectisome, low Ca²⁺ triggers a switch in the substrate specificity of the injectisome. Interestingly, low Ca²⁺ was recently shown to induce substrate switching of the enteropathogenic Escherichia coli injectisome as well (4). These important findings establish a hierarchy in secretion whereby translocator proteins are secreted first, allowing for rapid assembly of the translocation pore, followed by the regulated delivery of the effector proteins. It is worth noting, however, that translocator proteins secreted from one cell are unable to cross-complement translocator mutants in trans (3, 12). What then might be the purpose of constitutively secreting translocator proteins that are apparently unable to be recruited from the extracellular milieu and assembled into a functional pore complex? It is possible that formation of the translocation pore requires a high local concentration of the translocators and that these concentrations can be achieved only by constitutively secreting the translocators in the vicinity of the needle tip. An alternative explanation consistent with the data presented by Cisz et al. is that the translocators are stably associated with the needle complex near the cell surface. Rather than being continuously secreted, however, the translocators may be released into the culture medium by the shear forces associated with growth of P. aeruginosa under shaking conditions. Constitutive secretion of the translocators would then permit continuous regeneration of the needle complexes. A final possibility is that constitutive release of the translocators reflects an immune avoidance strategy. It is well established that antibody responses to one of the translocators (PcrV) prevent assembly of the translocation channel (6, 7) and are protective in experimental infection models (14). Constitutive shedding of PcrV might effectively dampen the anti-PcrV antibody response through formation of neutralizing antigen-antibody complexes.

Limiting further rounds of translocation.

The unregulated secretion of translocators stands in contrast to the stringent regulation of effector protein secretion. Not only does the latter require a specific signal to initiate translocation, but it has now been shown by Cisz et al. in P. aeruginosa and by Aili et al. in Y. pseudotuberculosis that the translocation process is itself regulated by translocated effector proteins (1, 3). Through a poorly understood mechanism, the ExoS effector of P. aeruginosa and the YopE effector of Y. pseudotuberculosis limit translocation of additional effectors and prevent subsequent bacteria from intoxicating that same target cell. The ExoS and YopE effectors both possess highly related GTPase-activating protein domains that are sufficient to inhibit translocation activity. In addition, ExoS has an ADP-ribosyltransferase domain that is lacking from YopE. In an interesting twist, Cisz et al. find that either enzymatic activity of ExoS is sufficient to inhibit P. aeruginosa translocation, suggesting that redundant mechanisms are in place to control translocation activity (3). This regulatory mechanism may serve to redirect other pseudomonads to target cells that remain to be intoxicated, limit the release of antigenic proteins, and aid in the conservation of energetic resources. An alternative host-centric interpretation is that the inhibition of translocation activity represents a host defense mechanism that is activated by translocated ExoS or YopE.

Homing in on the host contact signal.

Although the physiological relevance of the low-Ca²⁺ signal is unclear, it is thought to simulate contact of P. aeruginosa with target cells. In the highly related T3SS of Yersinia pestis it has been proposed that structural alterations induced by low Ca²⁺ and target cell contact are propagated through the needle complex and serve to activate the secretory system (15). One hypothesis accounting for the relationship between low Ca²⁺ and host cell contact proposes that the injectisome senses the low Ca²⁺ concentration of the target cell cytoplasm (8, 10). Cisz et al., however, present evidence that this is not the case for P. aeruginosa (3). In those experiments, target cells were pretreated with an ionophore that specifically elevates the host intracellular Ca²⁺ levels. This pretreatment had no effect on the ability of bacteria to activate the T3SS, indicating that the target cell-associated signal functions independently of intracellular Ca²⁺ levels. What, then, is the target cell signal? While an answer to this question is not immediately forthcoming, Cisz et al. find that the activating properties of target cells are not dependent upon active protein synthesis or on actin polymerization. Pretreatment of target cells with pore-forming toxins, however, completely eliminates activation of the T3SS, indicating a requirement for cell membrane integrity. These combined findings provide much-needed direction toward understanding the basis of Ca²⁺- and host contact-mediated regulation of the P. aeruginosa T3SS. Important questions for the future include the mechanism by which low Ca²⁺ induces substrate switching, whether this mechanism also triggers substrate switching upon target cell contact, and the mechanism by which ExoS inhibits translocation activity.