Pseudomonas syringae HrpP Is a Type III Secretion Substrate Specificity Switch Domain Protein That Is Translocated into Plant Cells but Functions Atypically for a Substrate-Switching Protein

Pseudomonas syringae delivers virulence effector proteins into plant cells via an Hrp1 type III secretion system (T3SS). P. syringae pv. tomato DC3000 HrpP has a C-terminal, putative T3SS substrate specificity switch domain, like Yersinia YscP. A ΔhrpP DC3000 mutant could not cause disease in tomato or elicit a hypersensitive response (HR) in tobacco, but the HR could be restored by expression of HrpP in trans. Though HrpP is a relatively divergent protein in the T3SS of different P. syringae pathovars, hrpP from P. syringae pv. syringae 61 and P. syringae pv. phaseolicola 1448A restored HR elicitation and pathogenicity to DC3000 ΔhrpP. HrpP was translocated into Nicotiana benthamiana cells via the DC3000 T3SS when expressed from its native promoter, but it was not secreted in culture. N- and C-terminal truncations of HrpP were tested for their ability to be translocated and to restore HR elicitation activity to the ΔhrpP mutant. No N-terminal truncation completely abolished translocation, implying that HrpP has an atypical T3SS translocation signal. Deleting more than 20 amino acids from the C terminus abolished the ability to restore HR elicitation. HrpP fused to green fluorescent protein was no longer translocated but could restore HR elicitation activity to the ΔhrpP mutant, suggesting that translocation is not essential for the function of HrpP. No T3SS substrates were detectably secreted by DC3000 ΔhrpP except the pilin subunit HrpA, which unexpectedly was secreted poorly. HrpP may function somewhat differently than YscP because the P. syringae T3SS pilus likely varies in length due to differing plant cell walls.Many proteobacterial pathogens use a type III secretion system (T3SS) as their primary mechanism to overcome and infect eukaryotic hosts. T3SSs are complex macromolecular machines that span both the bacterial cell envelope and host cell barriers to deliver proteins, commonly termed effectors, from the bacterial cytoplasm into the host cytoplasm (13, 19). After delivery into the host, effector proteins manipulate host cell function and suppress host defenses, allowing bacterial proliferation and disease development (6, 20). Bacteria that rely on T3SS to cause disease include plant pathogens such as Pseudomonas syringae, Ralstonia solanacearum, Erwinia and Xanthomonas species and animal pathogens in the genera Yersinia, Salmonella, Shigella, Escherichia, and Pseudomonas. While the repertoire of effectors delivered by a given T3SS is unique, the T3SS machinery is more universal (13). T3SS includes a core set of eight conserved proteins. These proteins, which are also conserved in bacterial flagellar biogenesis machines, make up the multiringed base structure, or basal body, that spans the bacterial membranes and cell wall. T3SS machines are also comprised of less-conserved and unique proteins that vary between systems. These include regulatory proteins that orchestrate construction of the machine and the extracellular components that function to translocate effectors across host barriers.The extracellular portion of the T3SS is comprised of the pilus or needle appendage (in plant or animal pathogens, respectively), which acts as a conduit for effector delivery, and the translocon complex, which creates the pore in the host cell membrane. These substructures vary between different T3SSs; presumably these external structures have adapted to allow different bacteria to infect different types of host cells. For Yersinia enterocolitica to infect macrophage cells, the T3SS needle must be a particular length (∼58 nm) to bridge the lipopolysaccharides extending from the bacterial outer membrane and reach the host cell membrane (35). Several other animal pathogens have T3SS needles of a defined length (48). Enteropathogenic Escherichia coli also has an additional extension beyond the needle called the EspA filament that functions to span the mucous layer found outside enterocyte cells (13). In plant pathogens, however, the extracellular gap between a bacterium and a plant cell includes a thick plant cell wall that is variable in width between plant species. Consequently, plant pathogenic Pseudomonas syringae has a pilus that can measure over 1 μm in vitro (25).Another major difference between the T3SS machineries of animal and plant pathogens is their translocon complexes. In animal pathogens, these are typically comprised of three essential proteins, but there is growing evidence that plant pathogen translocons employ diverse, functionally redundant components (28). There is growing interest in understanding the regulatory players that orchestrate the construction of diverse machinery. It is hypothesized that the assembly of the T3SS must involve several tightly regulated steps that allow secretion of the required components, followed by that of effectors upon completion. Of particular interest here is the control of pilus/needle subunit secretion, which is necessary when the pilus/needle is being constructed but would presumably compete with translocon and effector secretion after the T3SS is complete.We study the model plant pathogen P. syringae pv. tomato (Pto) DC3000, the causal agent of bacterial speck of tomato and Arabidopsis thaliana (8). DC3000 has a T3SS that delivers ca. 28 effectors and is essential for pathogenesis (11, 12, 30, 43). The P. syringae T3SS is encoded by hrp and hrc genes (hypersensitive response and pathogenicity/conserved), which are located in a pathogenicity island on the chromosome (4). hrc genes encode the conserved core components present in every T3SS. hrp genes encode T3SS components that are divergent or unique to P. syringae and enterobacterial plant pathogens, which also possess Hrp1 class T3SS (13). In contrast, plant pathogenic Ralstonia and Xanthomonas spp. have Hrp2 class T3SS, as indicated by several different Hrp proteins and distinct regulatory systems.To better understand the T3SS machinery, we previously conducted a survey of the hrp genes of P. syringae pv. syringae (Psy) 61 to complete the inventory of all those encoding proteins capable of traveling the T3SS into plant cells when expressed from a constitutive promoter (39). We hypothesized that these proteins might aid in pilus or translocon construction or regulate the construction process. HrpP was one protein found to be a T3SS substrate and important for secretion and translocation of the model effector AvrPto. Importantly, HrpP is related to a well-studied protein from Yersinia enterocolitica, YscP, which is a T3SS-secreted protein and a regulator responsible for switching the T3SS from secreting needle subunits to secreting effector proteins (15, 38, 47). It has also been shown that secretion of YscP into the culture medium is not essential for the switch function and that there may be two type III secretion signals embedded in YscP (2).The phenotype of a yscP mutant is unregulated secretion of the needle subunit, no secretion of effectors, and production of needles of indeterminate length. The switching phenotype requires a domain at the C terminus of YscP called the type III secretion substrate specificity switch (T3S4) domain, which is a conserved feature unifying its homologs (1). YscP has been proposed to act as a molecular ruler because the length of the YscP protein is directly correlated with the length of the Ysc needle (26). According to this model, when the needle has reached its proper length, YscP signals to the T3SS machinery to stop secreting needle subunits and begin secreting effector proteins. However, other functional models have been hypothesized for homologs of YscP. A recent study of the Salmonella enterica serovar Typhimurium YscP homolog InvJ showed that an invJ mutant lacked an inner rod. When the inner rod protein PrgJ was overexpressed, the length of the needle decreased relative to that of the wild type, leading the researchers to conclude that InvJ controls the inner rod, which in turn controls needle length (33). Recent evidence in Yersinia has lent more support to this model. YscP was found to negatively control secretion of YscI, the inner rod protein (51). Also, certain YscI mutations affected needle assembly but not effector secretion, implying that YscI may be a key player in substrate switching. Little is known about HrpB, the inner rod homolog in P. syringae (22), other than that the protein can be translocated into plant cells and is essential for T3SS function (39).Other models for length control/substrate switching have been proposed, such as the “C-ring cup model” in flagella, which was based on the observation that certain mutations in proteins that make up the inner membrane C ring of the basal body lead to shorter hooks (the flagellar equivalent of the needle), thus suggesting that C-ring capacity controls hook length (32). A more recent, flagellar “molecular-clock” model suggests that because overexpression of hook subunits leads to longer hooks and hook polymerization-defective mutants make shorter hooks, hook polymerization initiates a countdown, and the timing, in cooperation with the YscP homolog FliK, determines final hook length (34).HrpP is considered a member of the YscP/FliK family due mostly to the presence of a T3S4 domain at its C terminus. HrpP is also proline rich (10.6%), which is considered a characteristic of the family. The most striking feature of HrpP is its small size; the protein is 189 amino acids, compared with YscP from Y. enterocolitica, which is 453 amino acids and 8.4% proline. We were intrigued by how HrpP functions in P. syringae to regulate a pilus that can measure several hundred nanometers in length. Also, unlike animal pathogen needles and flagellar hooks, the pilus of P. syringae is predicted to be indeterminate in length, based on the fact that plant cell walls vary in width between species (40).We hypothesized that HrpP would be a main player in regulating pilus construction in P. syringae by allowing the system to make the transition between secretion of pilus subunits and secretion of translocon or effector proteins, though perhaps by a novel mechanism. In this study, we more precisely define the role of HrpP in the P. syringae T3SS. We show that HrpP is a T3SS substrate in DC3000, is translocated into plant cells at levels equivalent to those of effectors, and is essential for the function of the T3SS. Though it is highly translocated and variable, we found that HrpP from different P. syringae pathovars could complement the DC3000 hrpP mutant. Analysis of truncations of HrpP and an impassible HrpP-green fluorescent protein (GFP) fusion suggests that it has structural similarities to YscP, but surprisingly, HrpP was found to be required for full secretion of the pilus subunit HrpA as well as for translocation of HrpB.