A systematic approach to the molecular and cellular basis of Bartonella henselae-triggered invasome formation

Abstract

A systematic approach to the molecular and cellular basis of Bartonella henselae-triggered invasome formation Invasion of host cells is an often employed strategy of pathogenic bacteria to enable replication, persistence and dissemination. Among these pathogens is the zoonotic alpha-proteobacterium Bartonella henselae (Bhe). Human infection by Bhe can result in a variety of clinical symptoms. In immuno-competent patients, the most frequent manifestation is cat-scratch disease, a self-limiting disease resulting in local swellings of lymph nodes and fever. Immuno-compromised patients frequently develop bacillary angiomatosis-peliosis, a clinical condition hallmarked by the formation of vaso-proliferative lesions. Many of the distinct cell biological outcomes of vascular endothelial cells (ECs) infection by Bhe depend on a functional VirB/VirD4 type IV secretion (T4S) system. This macromolecular machinery is known to transport seven effector proteins, BepA to BepG (Bartonella-translocated effector proteins), into infected ECs. These effector proteins trigger all so far described VirB/VirD4-dependent cellular phenotypes of ECs, that are: (i) protection from apoptosis, (ii) activation of the transcription factor NF-kB and stimulation of a pro-inflammatory response and (iii) cell invasion by a unique cellular structure termed the invasome. Invasome-mediated internalization is a multi-step process that occurs with slow kinetics and requires at least 16 h for completion in individual ECs. In contrast, Bhe can alternatively enter ECs within hours as individual bacteria by a VirB/VirD4-independent endocytosis-like route. Started in November 2006, the aim of my thesis was to systematically identify and characterize host cell factors required for Bartonella henselae‐triggered invasome formation and to analyze how the Bartonella effector proteins (Beps) manipulate host cell signaling in order to provoke invasome assembly. The work presented in this thesis is sub-categorized into four results chapters, presented as published, submitted or prepared manuscripts. Research article I represents a reductionistic approach to define the minimal effector set required for Bhe-triggered invasome formation on endothelial cells (ECs). Further, the role of the Rho family of small GTPases was assessed. We identified the effector protein BepG to be sufficient to trigger the formation of invasomes that are morphologically indistinguishable from wild-type promoted structures. Investigation of the Rho GTPases Cdc42, Rac1 and their respective downstream interaction partners signified their importance for invasome establishment; in contrast, RhoA was shown to not contribute to invasome formation. Thus, our findings suggest a dominant role for Rho GTPases in regulating F-actin rearrangements leading to the composition of invasome structures. In research article II, we identify a second, BepG-independent mechanism to promote invasome assembly. We show that the combination of BepC and BepF mediates the establishment of invasomes. Next, we outline similarities and differences in characteristics and signaling cascades promoting BepG- or BepC/BepF-dependent invasome formation. We demonstrate that BepC/BepF in combination interfere with endocytosis-like internalization of inert microspheres and show the requirement for Cdc42 and Rac1-mediated cytoskeletal rearrangements to enable BepC/BepF-promoted invasome establishment. We also demonstrate that cofilin1 is only essential for BepC/BepF but not BepG-triggered invasome formation. This implies the involvement of partially different signaling cascades behind the two functionally redundant mechanisms triggering invasome establishment. Research article III focuses on the characterization of the effector protein BepF. We show that the individual Bid domains BidF1 and BidF2, but not BidF3, are sufficient to promote invasome establishment together with BepC. TEM-assessment of morphological changes induced by BepF or BidF1 indicated that they may hyper-activate Cdc42 and Rac1. In vitro guanine nucleotide exchange factor (GEF) assays confirmed that BidF1 and BidF2 exhibit weak GEF activity against Cdc42. Thus, BepF was demonstrated to act as a Cdc42-GEF protein in host cells, thereby interfering with F-actin dynamics and regulation. In research article IV, we focus on the host cell contribution to invasome formation. We identify five additional proteins essential for invasome formation and map their contribution to the different steps of invasome establishment. We demonstrate that the engagement of both outside-in signaling events via integrin β1, FAK and Src as well as inside-out signal transmission controlled by talin1 significantly contribute to invasome formation. This is the first study demonstrating active integrin outside-in signaling and parallel inside-out activation of these receptors during bacteria invasion of the host cells.