SummaryEnteropathogenic Escherichia coli (EPEC) is a major cause of paediatric diarrhoea and a model for the family of attaching and effacing (A/E) pathogens. A/E pathogens encode a type III secretion system to transfer effector proteins into host cells. The EPEC Tir effector protein acts as a receptor for the bacterial surface protein intimin and is involved in the formation of Cdc42-independent, actin-rich pedestal structures beneath the adhered bacteria. In this paper, we demonstrate that EPEC binding to HeLa cells also induces Tir-independent, cytoskeletal rearrangement evidenced by the early, transient formation of filopodia-like structures at sites of infection. Filopodia formation is dependent on expression of the EPEC Map effector molecule -a protein that targets mitochondria and induces their dysfunction. We show that Map-induced filopodia formation is independent of mitochondrial targeting and is abolished by cellular expression of the Cdc42 inhibitory WASP-CRIB domain, demonstrating that Map has at least two distinct functions in host cells. The transient nature of the filopodia is related to an ability of EPEC to downregulate Map-induced cell signalling that, like pedestal formation, was dependent on both Tir and intimin proteins. The ability of Tir to downregulate filopodia was impaired by disrupting a putative GTPase-activating protein (GAP) motif, suggesting that Tir may possess such a function, with its interaction with intimin triggering this activity. Furthermore, we also found that Map-induced cell signalling inhibits pedestal formation, revealing that the cellular effects of Tir and Map must be co-ordinately regulated during infection. Possible implications of the

Salmonella virulence depends on an ability to invade host cells, which is in turn dependent on a type III protein secretion system encoded in Salmonella pathogenicity island 1 (SPI1). Several protein targets of the SPI1‐encoded secretion system are translocated into host cells, where they subvert cellular processes that contribute to bacterial invasion, actin rearrangement, membrane ruffling and other aspects of virulence. We examined the role of sipA (encoding the translocated protein SipA) and found that a sipA mutant was significantly less invasive in Madin–Darby canine kidney (MDCK) cells than in its parental strain at the earliest stages of infection (5 min). The invasion defect associated with sipA was no longer apparent after 15 min of infection. Confocal microscopy of F‐actin in tetramethyl rhodamine isothiocyanate (TRITC)–phalloidin‐stained MDCK cells revealed no difference in either the frequency or the morphology of membrane ruffles induced by wild‐type and sipA mutant strains of S. typhimurium. Time‐lapse phase‐contrast microscopy of membrane ruffle propagation in live cells confirmed that the sipA mutant induced membrane ruffles as efficiently as the wild‐type bacteria. These studies also revealed that, after ruffle propagation, individual sipA mutant S. typhimurium either invaded more slowly than wild‐type bacteria or failed to invade at all. Furthermore, although wild‐type S. typhimurium typically maintained a position central to the developing membrane ruffle, sipA mutant bacteria frequently moved initially to the periphery of the spreading ruffle and were sometimes observed to detach from it. A wild‐type pattern of invasion was restored to the sipA mutant after the introduction of sipA on a plasmid. Together, these data indicate that loss of sipA significantly decreases the efficiency of S. typhimurium invasion at the early stages of infection without affecting its ability to induce membrane ruffles. It thus appears that the secreted effector protein SipA promotes invasion by a previously unrecognized mechanism separate from the induction of membrane ruffling per se.

SummaryEnteropathogenic Escherichia coli (EPEC) are a major cause of paediatric diarrhoea and a model for the family of attaching and effacing (A/E) pathogens. Enteropathogenic Escherichia coli encode a type III secretion system (TTSS) to transfer effector proteins into host cells, a process which is essential for virulence. In addition to generation of A/E lesions, the TTSS is also implicated in the ability of EPEC to invade cultured cells but the effector proteins responsible for promoting invasion have not been identified. In this paper we confirm the requirement of TTSS in EPEC invasion and demonstrate important roles for the Map and Tir effector molecules. Whereas in trans expression of Tir in the tir mutant restored invasion to wild-type levels, similar complementation of the map mutation by in trans expression of Map results in a hyperinvasive phenotype. The Map effector protein has two distinct functions within host cells, mediating Cdc42-dependent filopodia formation and targeting mitochondria to elicit dysfunction. The former function appears to be related to Map's ability to promote invasion as this was inhibited by interference with Cdc42 signalling. Conversely, Map targeting to mitochondria is not necessary for invasion. Promotion of EPEC invasion by Tir appears to involve interaction with intimin but is independent of pedestal formation, and intimin-Tir interaction is neither necessary nor sufficient for invasion. Comparison of the invasiveness of strains lacking Tir and/or Map with wild-type or mutant strains expressing the effectors in trans provides evidence that Map and Tir stimulate invasion by synergistic mechanisms. This synergism, which is in stark contrast to the antagonistic actions of Map and Tir in regulating filopodia and pedestal formation, further illustrates the complex interplay between EPEC effectors.

The neuropeptide galanin has diverse roles in the central and peripheral nervous systems, by activating the G protein-coupled receptors Gal1, Gal2 and the less studied Gal3 (GalR1–3 gene products). There is a wealth of data on expression of Gal1–3 at the mRNA level, but not at the protein level due to the lack of specificity of currently available antibodies. Here we report the generation of knock-in mice expressing Gal1 or Gal2 receptor fluorescently tagged at the C-terminus with, respectively, mCherry or hrGFP (humanized Renilla green fluorescent protein). In dorsal root ganglia (DRG) neurons expressing the highest levels of Gal1-mCherry, localization to the somatic cell membrane was detected by live-cell fluorescence and immunohistochemistry, and that fluorescence decreased upon addition of galanin. In spinal cord, abundant Gal1-mCherry immunoreactive processes were detected in the superficial layers of the dorsal horn, and highly expressing intrinsic neurons of the lamina III/IV border showed both somatic cell membrane localization and outward transport of receptor from the cell body, detected as puncta within cell processes. In brain, high levels of Gal1-mCherry immunofluorescence were detected within thalamus, hypothalamus and amygdala, with a high density of nerve endings in the external zone of the median eminence, and regions with lesser immunoreactivity included the dorsal raphe nucleus. Gal2-hrGFP mRNA was detected in DRG, but live-cell fluorescence was at the limits of detection, drawing attention to both the much lower mRNA expression than to Gal1 in mice and the previously unrecognized potential for translational control by upstream open reading frames (uORFs).

Dynamic subcellular distributions of signaling system components are critical regulators of cellular signal transduction through their control of molecular interactions. Understanding how signaling activity depends on such distributions and the cellular structures driving them is required for comprehensive insight into signal transduction. In the activation of primary murine T cells by antigen presenting cells (APC) signaling intermediates associate with various subcellular structures, prominently a transient, wide, and actin-associated lamellum extending from an interdigitated T cell:APC interface several micrometers into the T cell. While actin dynamics are well established as general regulators of cellular organization, their role in controlling signaling organization in primary T cell:APC couples and the specific cellular structures driving it is unresolved. Using modest interference with actin dynamics with a low concentration of Jasplakinolide as corroborated by costimulation blockade we show that T cell actin preferentially controls lamellal signaling localization and activity leading downstream to calcium signaling. Lamellal localization repeatedly related to efficient T cell function. This suggests that the transient lamellal actin matrix regulates T cell signaling associations that facilitate T cell activation.