Tracking bacterial effector protein delivery into host cells

Abstract

Bacterial Type IV secretion systems (T4SSs) are a functionally heterogeneous group of nanomachines that can deliver substrates into a wide range of target cells. The Helicobacter pylori Cag T4SS has an important role in the pathogenesis of gastric cancer. CagA, the only effector protein known to be secreted by the H. pylori Cag T4SS, enters human gastric cells and causes alterations in intracellular signaling that are linked to cancer pathogenesis. Understanding the molecular mechanisms by which CagA is delivered into gastric cells has been hindered by the lack of robust methods for monitoring this process. A publication in this issue of Molecular Microbiology describes a split luciferase assay for monitoring T4SS-mediated translocation of CagA into host cells. Use of this translocation reporter allowed quantification of CagA translocation in real-time assays, thereby facilitating analysis of the kinetics of CagA delivery. This system also allowed the tracking of several types of CagA fusion proteins and confirmed that protein unfolding is important for secretion by the Cag T4SS. This commentary discusses T4SS-dependent delivery of H. pylori CagA into host cells and use of the split luciferase system for monitoring bacterial protein secretion and delivery into target cells.

Abbreviated Summary

Bacteria can use several types of secretion systems to deliver effector proteins into host cells through contact-dependent processes. This commentary discusses type IV secretion system-dependent translocation of Helicobacter pylori CagA into gastric epithelial cells and the use of a split luciferase system for monitoring bacterial protein secretion and delivery into target cells.

Bacteria utilize an assortment of secretion systems for exporting proteins across the bacterial cell envelope (1). Protein-injection machines are a specialized group of secretion systems that can deliver bacterial effector proteins directly into eukaryotic cells or other bacteria through contact-dependent processes (2). Three main categories of protein-injection machines, designated as type III, type IV, and type VI secretion systems, are recognized (2).

Type IV secretion systems (T4SSs) are a functionally heterogeneous group of nanomachines (3–5). In addition to delivering effector proteins into recipient target cells, T4SSs can mediate interbacterial DNA transfer (conjugation), DNA uptake into bacteria, DNA release into the external environment, and DNA delivery into eukaryotic cells. T4SSs can translocate effector proteins into a wide range of eukaryotic cells (mammalian, plant, and protozoa) or other bacteria. Most components of T4SSs in Gram-negative bacteria are organized into an inner membrane complex and an outer membrane core complex (3–5), and in some cases, pilus structures are present (6, 7).

Prototype T4SSs, exemplified by conjugation systems and the VirB/VirD4 T4SSs of Agrobacterium tumefaciens and Xanthomonas citri, contain 12 components (designated VirB1–11 and VirD4) (3–5). In contrast to these “minimized” systems, “expanded” T4SSs, such as those present in Helicobacter pylori and Legionella pneumophila, are larger in size and contain additional components (8–10). The number of different effector proteins secreted by the T4SSs of individual bacterial species ranges from one to several hundred (11). T4SS-mediated secretion of effector proteins has an important role in the pathogenesis of human diseases caused by Helicobacter pylori, Legionella pneumophila, Bordetella pertussis, Coxiella, Brucella, Bartonella, Ehrlichia, Anaplasma, and Rickettsia (3).

The H. pylori Cag Type IV secretion system is of interest because of the frequency with which it contributes to human illness (9, 10). About half of all humans worldwide are persistently colonized with H. pylori. H. pylori colonization of the stomach is a strong risk factor for development of peptic ulcer disease and gastric cancer. H. pylori strains can be broadly classified into two groups based on whether they contain a chromosomal region known as the cag pathogenicity island (cag PAI). The risk of gastric cancer or peptic ulcer disease is significantly higher in individuals colonized with cag PAI-positive H. pylori strains than in individuals colonized with cag PAI-negative strains or H. pylori-negative individuals (12).

One gene in the cag PAI encodes a secreted effector protein (CagA), which enters host cells and interacts with multiple different cytoplasmic proteins, resulting in alterations of intracellular signaling (13–15). Other genes in the cag PAI encode components of a T4SS (Cag T4SS) required for delivery of CagA into host cells (9, 10). CagA is the only protein known to be secreted by the Cag T4SS. Adherence of H. pylori to gastric epithelial cells triggers the translocation of CagA from adherent bacteria into the epithelial cells. In contrast, CagA is not released into the extracellular space when H. pylori are grown in broth cultures in the absence of host cells. In addition to its role in delivery of CagA into gastric cells, the Cag T4SS facilitates the entry of non-protein molecules (including LPS metabolites, peptidoglycan and DNA) into host cells through processes that are poorly understood (9, 10). In experimental animal models of H. pylori infection and gastric disease, the development of premalignant conditions, gastric cancer and gastric ulceration is dependent on the presence of an active Cag T4SS and CagA (9, 10). Expression of CagA in transgenic mice promotes the development of adenocarcinomas in the stomach and small intestine, as well as hematologic malignancies (16).

Seventeen genes in the H. pylori cag PAI are required for delivery of CagA into gastric cells (9, 10, 17). The proteins encoded by about half of these genes exhibit sequence relatedness to components of T4SSs in other bacterial species and the remainder are found only in H. pylori. Recent cryo-electron tomography and single particle cryo-EM studies have provided insights into the structural organization of the Cag T4SS (18–21). The Cag T4SS outer membrane core complex is a large mushroom-shaped structure, about 40 nm in width and 25 nm in height, that contains multiple copies of 5 different proteins (20–22). The dimensions of the Cag T4SS outer membrane core complex are similar to those of the corresponding subassembly in the Legionella Dot/Icm T4SS (23–26). The corresponding core complexes in prototype T4SSs contain multiple copies of only 3 different proteins and are much smaller in size (7, 27, 28). The inner membrane complex of the H. pylori Cag T4SS is about 36 nm in width and 27 nm in height and is predicted to contain at least seven different proteins (19).

Two decades ago, the discovery that H. pylori CagA is an effector protein delivered into gastric cells through a T4SS-mediated process relied on the detection of tyrosine-phosphorylated CagA within host cells (a modification mediated by host cell kinases) and changes in the shape of these cells (known as a “hummingbird phenotype”) (29). Subsequently, detection of tyrosine-phosphorylated CagA by Western blotting has remained the main method used for detecting CagA delivery into host cells. Efforts to develop alternate methods for tracking CagA delivery into host cells have been thwarted by various obstacles. One publication reported that β-lactamase (TEM-1) fused to CagA could be used as a quantitative translocation reporter system (30). This approach offered some advantages compared to the detection of tyrosine-phosphorylated CagA, but there were several limitations. The levels of the CagA-TEM-1 fusion protein produced by H. pylori were relatively low compared to levels of wild-type CagA; it was difficult to detect tyrosine phosphorylation of the CagA-TEM-1 fusion protein within gastric cells; cell shape changes in response to the CagA-TEM-1 fusion protein were not detected; and experiments with the CagA-TEM-1 reporter did not readily allow monitoring of CagA translocation in real time (30). The lack of good tools has hampered efforts to understand H. pylori Cag T4SS activity and delivery of CagA into host cells.

In the current issue of the Molecular Microbiology, Lettl et al. report the use of a split luciferase-based translocation reporter system for monitoring CagA delivery into host cells (31). The assay described in this study employed an H. pylori strain producing CagA with an 11-residue NanoLuc-HiBiT tag at the amino-terminus. AGS gastric epithelial cells were engineered to stably produce the remainder of the luciferase protein (NanoLuc-LgBiT). The NanoLuc (NLuc) luciferase used for these assays is a small highly stable form of the enzyme that uses furimazine as a substrate and generates a very bright luminescent signal through an ATP-independent process. Upon co-culture of H. pylori with the gastric cells, HiBiT-tagged CagA entered the gastric cells, allowing reconstitution of luciferase activity. The translocated HiBiT-tagged CagA underwent tyrosine phosphorylation within the gastric cells and caused changes in the shape of the cells, similar to what was observed with unmodified CagA (31).

The split luciferase translocation reporter assay allowed quantification of HiBiT-CagA entry into host cells in real-time assays, thereby facilitating the first analysis of the kinetics of CagA translocation (31). Furimazine was added to the host cells at the beginning of the experiment, and luminescence was then monitored at serial time points. An initial burst of luminescence was detectable within 10 minutes after the addition of H. pylori to the gastric cells and increased over the next hour until a plateau was reached. The observed plateau might be attributable to downregulation of T4SS activity at this stage of bacteria-host cell interaction. CagA is produced at relatively high levels in H. pylori compared to the relatively low levels of effector protein production in several other bacterial species. The current study showed that only a small proportion of CagA within H. pylori (≤12.5%) was translocated into the gastric cells (31). Experiments with protein synthesis inhibitors have suggested that only newly synthesized CagA is in a translocation-competent state (30). The production of CagA at levels considerably higher than what can be delivered into host cells suggests that non-translocated CagA might have a useful function.

Most investigators have presumed that CagA must be unfolded in order to transit through the T4SS, but there has been relatively little experimental evidence. The availability of the new translocation reporter assay facilitated investigation of this topic. The investigators generated H. pylori strains producing multiple types of HiBiT-tagged CagA fused to heterologous passenger proteins with varying folding characteristics (31). A HiBit-CagA-GFP fusion was not translocated into host cells, consistent with expectations that the compact structure of GFP would preclude unfolding (32). Experiments were also conducted with H. pylori producing HiBiT-tagged CagA fused to dihydrofolate reductase (DHFR), which can be stabilized in a folded conformation by methotrexate. Translocation of HiBiT-DHFR-CagA into host cells was detected under baseline conditions but was blocked by addition of methotrexate. Experiments with a strain producing two forms of CagA (HiBiT-CagA and DHFR-CagA) showed that DHFR-CagA inhibited translocation of HiBiT-CagA in the presence of methotrexate, consistent with a model in which DHFR-CagA competes with HiBiT-CagA for entry into the T4SS channel and jams the translocation machinery (32). Unexpectedly, a ubiquitin-CagA fusion was translocated into host cells, suggesting that ubiquitin can be translocated in a folded state or that H. pylori can unfold the ubiquitin domain and then translocate the unfolded fusion protein. In contrast to this result, ubiquitin folding impedes T4SS-mediated translocation of fusion proteins across membranes in other bacterial species (33, 34). Overall, the experiments in the current paper provide support for the view that unfolding is required for CagA secretion and translocation (31).

Multiple types of reporters are available to monitor the secretion of bacterial proteins and the delivery of bacterial effector proteins into host cells (35). The readouts include enzymatic activity, fluorescence, and luminescence. Each of the available reporter systems has advantages and disadvantages, depending on the intended use. For example, the fusion of large heterologous reporter proteins to effector proteins can potentially result in loss of effector protein functionality or can interfere with effector protein secretion. The relatively large size of the TEM-1 reporter fused to CagA in a previous study (30) likely accounted for some limitations of that translocation reporter (discussed above). The small size of the 11-residue split luciferase (HiBiT) tag used in the paper by Lettl et al. did not interfere with CagA secretion and it allowed functional properties of CagA to be preserved (31). On the other hand, the split luciferase assay requires transfection of target cells with a LgBiT expression construct and isolation of cells stably producing the LgBiT protein, whereas the TEM-1 assay can be conveniently applied to multiple cell types without a need for these steps. Therefore, these can be viewed as two complementary assays.

In previous studies the split luciferase system has been used to monitor Sec-dependent protein secretion (36, 37) as well as the secretion of effector proteins by the T3SS encoding by the Salmonella pathogenicity island 1 and T3SS-dependent delivery of effector proteins into host cells (38). The successful use of the split luciferase system to study both T3SS- and T4SS-mediated secretion of proteins suggests that this approach will be a valuable addition to the toolbox of translocation reporter assays for studying the delivery of effector proteins into target cells by contact-dependent secretion systems. Similarly, this approach should be generally useful to monitor bacterial protein secretion and the entry of secreted bacterial proteins into target cells.

Figure 1.

Split luciferase assay for tracking bacterial effector protein delivery into host cells. Bacteria are engineered to produce the effector protein of interest fused to an 11-amino-acid peptide (NanoLuc HiBiT tag, yellow triangle). Target cells produce the remainder of the luciferase protein (NanoLuc-LgBiT). Upon translocation of the effector protein into host cells, the two proteins interact, and in the presence of added furimazine substate, a luminescent signal is generated.