Defining the phosphinositide-binding effector arsenal of Legionella pneumophila: identification, structural features, and functional roles

Abstract

Microbial pathogens have evolved diverse virulence factors to manipulate host membrane-bound compartments. Phosphoinositide lipids, although minor components of cellular membranes, play essential roles in membrane trafficking events and are critical for defining membrane identity. They act as molecular beacons, recruiting and activating protein complexes to coordinate cellular processes. To exploit these pathways, pathogens employ their own phosphoinositide-binding proteins that rewire host trafficking and signaling for bacterial benefit. The Gram-negative facultative, intracellular pathogen, Legionella pneumophila, employs the largest known arsenal of bacterial effecters to establish a replicative niche within alveolar macrophages. In its natural environment, Legionella resides in freshwater and soil where it interacts with and infects a variety of amoeba species. Conserved cellular pathways between amoebae and macrophages enable infection in humans following inhalation of contaminated aerosolized water droplets. In immunocompromised individuals, Legionella causes a severe pneumonia-like illness known as Legionnaires’ disease. ☐ Upon entry into macrophages via phagocytosis, Legionella resides within a nascent phagosome that would typically undergo sequential fusion with endocytic compartments and ultimately be degraded by lysosomes in a process known as phagosome maturation. Instead, Legionella diverts the phagosome from the endolysosomal pathway and remodels it into a compartment resembling the endoplasmic reticulum, known as the Legionella-containing vacuole (LCV). This process is driven by the coordinated efforts of a large artillery of effector proteins that are translocated into the host cytosol through a specialized secretion system. Functional characterization of these effectors has been hindered by their functional redundancy and remote homology to known protein domains. An emerging feature of many effectors is their ability to recognize and bind host phosphoinositides; however, given the large number of uncharacterized effectors, we hypothesized that the full extent of this targeting remains undiscovered. In this dissertation, we uncover an army of phosphoinositide-binding effectors and ascribe functional roles for several newly characterized binders. ☐ In chapter 2, we systematically identified phosphoinositide-binding Legionella effector proteins using a three-pronged biochemical approach: lipid bead pulldowns, localization studies with phosphoinositide-biosensors, and protein-lipid overlay assays. Through this approach we successfully identified eighteen novel phosphoinositide-binding effectors, with a striking majority showing specificity for PI(3)P. Many of these effectors contained uncharacterized alpha-helical regions, which we confirmed as phosphoinositide-binding domains. We leveraged these conserved structural features to identify additional phosphoinositide-binding effectors through a bioinformatic approach. This search revealed another set of candidates, and we validated twelve as novel phosphoinositide-binding effectors. Expanding our search across bacterial pathogens, revealed structurally conserved helical folds in multiple pathogens, including Coxiella burnetti and Burkholderia pseudomallei. We confirmed phosphoinositide-binding for two Coxiella and one Burkholderia effector. These findings suggest that targeting phosphoinositides is a conserved strategy among intracellular pathogens to manipulate host membranes. ☐ In chapter 3, we delineate the compartment specific localization for four of the PI(3)P-binding effectors discovered in chapter 2: Lpg0405, Lpg1602, Lpg1851, and Lpg2546. We found each effector preferentially localizes to a subset of PI(3)P-positive membrane compartments, indicating their localization is driven by additional membrane features. Lpg1851 localized to endocytic vesicles, while Lpg2546 primarily associated with late endosomes and contacted autophagosomes. In contrast, Lpg0405 and Lpg1602 localized specifically to autophagosomes. Together, these results reveal the membrane-targeting preferences of newly identified PI(3)P-binding effectors and suggest they can recognize additional components beyond phosphoinositide identity. ☐ Chapters 4 and 5, describe the functional characterization of two phosphoinositide-binding effectors. In chapter 4, we examined the PI(3)P-binding effector, Lpg2385. We found this effector binds to PI(3)P with high affinity with a non-C-terminal domain. Lpg2385 associates with endocytic compartments and interacts with the small GTPase Rab5, therefore, the effector has a dual-key mechanism governing its membrane localization. We developed a highly sensitive and specific Lpg2385 antibody to determine Lpg2385 is translocated during early infection, and it localizes to unknown structures near the LCV, implicating a role in LCV establishment or endocytic evasion. In chapter 5, we investigate Lpg2587, an SGNH-hydrolase domain-containing effector that binds PI(3)P via a C-terminal domain. During ectopic expression, Lpg2587 dynamically localizes between endocytic vesicles and the endoplasmic reticulum. We identified 14-3-3 as a host interaction partner and observed Lpg2587 disrupts mitochondria morphology through its catalytic activity. We further determined Lpg2587 is translocated during the intermediate stages of infection and localizes to the LCV, likely via its PI(3)P-binding region. Lpg2587’s targeting of host membranes and timing of translocation suggest it aids in lipid acquisition to support the LCV membrane. While both Lpg2385 and Lpg2587 target PI(3)P, they appear to have distinct roles during infection, underscoring the diversity of host cell processes phosphoinositide-binding effectors influence. ☐ Collectively, this work reveals that Legionella uses phosphoinositide-binding as a widespread strategy to selectively target host membranes. We discovered phosphoinositide-binding effectors likely employ a dual-key mechanism for compartment recognition and feature diverse domain architectures that co-opt distinct host processes. Remarkably, structurally similar folds are found in effectors from other pathogens and confer conserved phosphoinositide-binding capacity. Therefore, this work indicates virulence effectors employ a shared mechanism for binding phosphoinositides and thus, represent an Achilles heel that can be leveraged to reveal the molecular mechanisms used by intracellular bacteria to thrive within hosts.