Structure-based functional analysis of effector protein SifA in living cells reveals motifs important for Salmonella intracellular proliferation
Introduction
Salmonella enterica serovar Typhimurium is a Gram-negative intracellular pathogen which causes gastroenteritis in humans and a systemic disease in mice that resembles human typhoid fever. Salmonella virulence relies on the ability to survive in host cells and proliferate within Salmonella-containing vacuole (SCV) (reviewed in LaRock et al., 2015, Liss and Hensel, 2015). The intracellular phase depends on the Salmonella pathogenicity island 2 (SPI2)-encoded type III secretion system (T3SS) which is expressed in response to signals in the SCV environment such as low pH and nutritional limitation (Kuhle and Hensel, 2004). The SPI2-T3SS mediates the translocation of at least 30 effector proteins that support bacterial intra-vacuolar survival and replication by altering the membrane and cytoskeleton of the host cells, manipulation of vesicular trafficking, inhibition of cell death pathways and blocking innate and adaptive immunity (LaRock et al., 2015).
A remarkable phenotype induced by intracellular S. enterica is the massive remodeling of the host endosomal system with formation of Salmonella-induced filaments or SIF as a prominent phenotype (Garcia-del Portillo et al., 1993, Liss and Hensel, 2015). SIF are tubular aggregations of late endosomal/lysosomal vesicles. Presence of lysosomal glycoproteins (lgp) such as lysosome-associated membrane protein 1 (LAMP1) is characteristic for SIF membranes. The Salmonella SPI2-T3SS effector protein SifA is crucial for formation of SIF (Stein et al., 1996) and stability of SCV during intracellular replication (Beuzon et al., 2000). Mutant strains lacking SifA are more frequently found in the cytosol of host cells, leading to hyper-replication in epithelial cells or reduced survival in macrophages. In addition to SifA the formation of SIF requires further SPI2-T3SS effector proteins including SseF, SseG, SopD2 and PipB2 (Brumell et al., 2002, Kuhle et al., 2004). Contributions of further SPI2-T3SS effector proteins SteA, SseJ and GtgE to manipulation of the host endosomal system and the SCV membrane were reported (see Jennings et al., 2017, for recent review).
SifA is the best characterized effector that is crucial for virulence in a murine model of systemic Salmonellosis, formation of SIF by aggregation of late endosomal/lysosomal compartments (Stein et al., 1996), and SCV stability (Beuzon et al., 2000), which all together underscore the importance of this protein to Salmonella pathogenesis. SifA is composed of two distinct domains, each harboring distinct functional motifs. The N-terminal domain has been proposed for secretion/translocation (Miao and Miller, 2000). SifA specifically interacts with the Pleckstrin homology (pH) domain of the SifA kinesin-interacting protein (SKIP, also termed PLEKHM2) by its N-terminal domain (Jackson et al., 2008). The SifA-SKIP interaction of prevents accumulation of kinesin at the SCV and leads to SIF tubule formation (Boucrot et al., 2005, Diacovich et al., 2009, Ohlson et al., 2008). The minimal binding domain (residues 1–140) of SifA has been identified by co-IP analysis. Based on the structure of SifA in complex with the SKIP pH domain (PDB: 3CXB) functionally important residues were predicted and mutation SifA L130D resulted in abrogation of SIF formation in epithelial cells and attenuation of virulence in a murine infection model (Diacovich et al., 2009). Recently, additional interaction of the N-terminal domain of SifA with PLEKHM1 was reported (McEwan et al., 2015). This interaction leads to recruitment of endolysosomal membranes that are essential for SCV growth and proliferation in cells. In infected cells, SifA interferes with positioning of late endosomes/lysosomes by preventing the interaction of SKIP and Rab9 (Jackson et al., 2008). SKIP also indirectly interacts with motor protein kinesin-1 (Dumont et al., 2010). SPI2-T3SS effector protein PipB2 recruits inactive kinesin-1 to the SCV and activation of kinesin-1 depends on SKIP binding to the kinesin-1 light chain (KLC) (Henry et al., 2006).
Recent analysis in the mouse model of virulence (Zhao et al., 2015) demonstrated that absence of the C-terminal domain of SifA causes higher attenuation of virulence than absence of a functional N-terminal domain. The C-terminal domain of SifA is substrate of post-translational modification of its CAAX motif by geranylgeranyltransferase (GGT) (Boucrot et al., 2003). By this prenylation SifA is membrane-anchored (Reinicke et al., 2005). Furthermore, SifA is modified by S-acylation catalyzed by acyltransferases. Unlike prenylation, this post-translational modification is dynamically reversible (Reinicke et al., 2005). Additionally, the C-terminal portion of SifA interacts with prenylated Rab7 which likely leads to dissociation of Rab7 from Rab7-interacting lysosomal protein (RILP) and consequently prevents fusion of lysosomes with the SCV (Harrison et al., 2004). RILP interacts with dynein and by uncoupling Rab7 from RILP, SifA contributes to centripetal extension of SIFs (Harrison et al., 2004, Johansson et al., 2007). The C-terminal domain of SifA shares high structural similarity with SopE, an effector protein of the SPI1-T3SS. SopE has Guanine nucleotide exchange factor (GEF) activity (Rudolph et al., 1999) that modulates the host cell actin cytoskeleton. The SifA C-terminal domain harbors a WxxxE motif that is characteristic for a family GEF-mimicking bacterial effector proteins (reviewed in Orchard and Alto, 2012). It was shown that SifA interacts with RhoA in GDP-bound form, but not with GTP-bound RhoA (Ohlson et al., 2008). However, no GEF activity was detected with purified SifA alone, indicating that other proteins are likely required as well.
Several studies addressed the role of vesicle fusion and membrane tubulation for intracellular lifestyle of Salmonella: Beuzon and colleagues suggested SifA-driven membrane acquisition ensures maintenance of a compartment for bacterial replication (Beuzon et al., 2000). This is supported by the finding that SifA-SKIP subverts Rab9-mediated trafficking, resulting in a detoxified compartment with perinuclear location (Jackson et al., 2008, McGourty et al., 2012). Several groups proposed that the extended membrane network of SIFs provides nutrients for growth of Salmonella in the SCV (McGourty et al., 2012, Ohlson et al., 2008, Schroeder et al., 2011, Stein et al., 1996). Our finding that SifA-mediated redirection of endosomal transport increases access of intracellular Salmonella to extracellular amino acids supports this idea (Popp et al., 2015). We recently demonstrated that SCV and SIFs form a continuum providing access to endocytosed material and by this nutrient acquisition for Salmonella in the SCV (Liss et al., 2017). All of these phenotypes depend on remodeling of the endosomal system by function of SifA. To identify regions involved in translocation and function of SifA, we performed a functional analysis using substitution and deletion mutagenesis strategies based on structural analysis. The SIF network is highly dynamic and prone to fragmentation by chemical fixation (Rajashekar et al., 2014) and both features call for analyses in living cells. We used high-resolution live cell imaging and quantitative volume analyses of the SIF network to determine the contribution of SifA motifs to endosomal remodeling.
Section snippets
Structure-based analyses of SifA
In this work we extended the structure-based prediction of functionally relevant residues in SifA (Fig. 1A). Based on computational analysis including alanine scanning (Fig. S1, Table S3), molecular dynamics simulations (Fig. S2), and multiple sequence alignment (MSA, Fig. S3), we selected single residues and small sequence motifs for mutagenesis and subsequent functional analyses if mutant SifA.
Prediction of critical residues at the SifA-SKIP interface in the N-terminal domain
Mutant SifA alleles SifAE24R, SifAY128R, SifAL130D and SifAM131A were selected for further analysis.
Discussion
In this study, we analyzed the structure-function relationship of bacterial effector SifA with focus on its interaction with known or predicted eukaryotic target proteins. The approach was based on structural information and bioinformatics analysis of interaction surfaces of SifA. Phenotypes of mutations in SifA were quantified by live cell imaging of Salmonella-infected cells performed with high spatial resolution. To our knowledge, this is the first time the volume of SIFs and bacteria inside
Bacterial strains and growth conditions
Salmonella enterica serovar Typhimurium (S. Typhimurium) NCTC 12023 was used as wild-type strains and mutant strains are isogenic to this strain. Strain P2D6 is defective in the Salmonella Pathogenicity Island 2 (SPI2)-encoded type III secretion system (T3SS) due to ssaV mutation (Shea et al., 1996) and MvP503 (ΔsifA::FRT) is defective in sifA. If indicated, strains harbored plasmids for expression of WT sifA or various mutant alleles as listed in. For live cell imaging, strains harboring
Acknowledgements
We are grateful to Monika Nietschke and Ursula Krehe for technical support, and Dr. Xu Xin for initial experimental work of this project. This work was supported by the Deutsche Forschungsgemeinschaft grant HE1964/18-2 within Priority Program SPP 1580 and SFB 944, Z project, and the Niedersächsisches Ministerium für Wissenschaft und Kultur. HS acknowledges support by the Deutsche Forschungsgemeinschaft (SFB 796, project A2).
References (50)
- et al.
Salmonella typhimurium SifA effector protein requires its membrane-anchoring C-terminal hexapeptide for its biological function
J. Biol. Chem.
(2003) - et al.
Interaction between the SifA virulence factor and its host target SKIP is essential for Salmonella pathogenesis
J. Biol. Chem.
(2009) - et al.
Advances and pitfalls of protein structural alignment
Curr. Opin. Struct. Biol.
(2009) - et al.
Heterogeneity of intracellular replication of bacterial pathogens
Curr. Opin. Microbiol.
(2013) - et al.
Salmonella SPI-2 type III secretion system sffectors: molecular mechanisms and physiological consequences
Cell Host Microbe
(2017) - et al.
Structure of Shigella IpgB2 in complex with human RhoA: implications for the mechanism of bacterial guanine nucleotide exchange factor mimicry
J. Biol. Chem.
(2010) - et al.
Salmonella enterica remodels the host cell endosomal system for efficient intravacuolar nutrition
Cell Host Microbe
(2017) - et al.
PLEKHM1 regulates Salmonella-containing vacuole biogenesis and infection
Cell Host Microbe
(2015) - et al.
Structure and function of Salmonella SifA indicate that its interactions with SKIP, SseJ, and RhoA family GTPases induce endosomal tubulation
Cell Host Microbe
(2008) - et al.
Dynamic modification of microtubule-dependent transport by effector proteins of intracellular Salmonella enterica
Eur. J. Cell. Biol.
(2011)