Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Human Genome-Wide RNAi Screen for Host Factors That Modulate Intracellular Salmonella Growth

  • Joshua M. Thornbrough,

    Affiliation Department of Biology, University of Louisville, Louisville, Kentucky, United States of America

    ⨯
  • Tom Hundley,

    Affiliation Department of Biology, University of Louisville, Louisville, Kentucky, United States of America

    ⨯
  • Raphael Valdivia,

    Affiliation Department of Molecular Genetics and Microbiology, Duke University, Durham, North Carolina, United States of America

    ⨯
  • Micah J. Worley

    micah.worley@louisville.edu

    Affiliations Department of Biology, University of Louisville, Louisville, Kentucky, United States of America, Department of Microbiology and Immunology, University of Louisville, Louisville, Kentucky, United States of America

    ⨯

Abstract

Salmonella enterica is a bacterial pathogen of humans that can proliferate within epithelial cells as well as professional phagocytes of the immune system. While much has been learned about the microbial genes that influence the infectious process through decades of intensive research, relatively little is known about the host factors that affect infection. We performed a genome-wide siRNA screen to identify host genes that Salmonella enterica serovar Typhimurium (S. typhimurium) utilizes to facilitate growth within human epithelial cells. In this screen, with siRNAs targeting every predicted gene in the human genome, we identified 252 new human-host-susceptibility factors (HSFs) for S. typhimurium. We also identified 39 genes whose silencing results in increased intracellular growth of S. typhimurium. The HSFs identified are regulated most centrally by NFκB and associate with each other through an extremely dense network of interactions that center around a group of kinases. Most genes identified were not previously appreciated as playing roles in the intracellular lifecycle of S. enterica. Numerous HSFs identified with interesting characteristics that could play plausible roles in mediating intracellular microbial growth are discussed. Importantly, this study reveals significant overlap between the host network that supports S. typhimurium growth within human epithelial cells and the one that promotes the growth of Mycobacterium tuberculosis within human macrophages. In addition to providing much new information about the molecular mechanisms underlying S. enterica-host cell interplay, all 252 HSFs identified are candidates for new anti-microbial targets for controlling S. enterica infections, and some may provide broad-spectrum anti-microbial activity.

Citation: Thornbrough JM, Hundley T, Valdivia R, Worley MJ (2012) Human Genome-Wide RNAi Screen for Host Factors That Modulate Intracellular Salmonella Growth. PLoS ONE 7(6): e38097. https://doi.org/10.1371/journal.pone.0038097

Editor: Dipshikha Chakravortty, Indian Institute of Science, India

Received: April 2, 2012; Accepted: May 3, 2012; Published: June 11, 2012

Copyright: © 2012 Thornbrough et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Funding: The work was supported by National Institutes of Health grant #U54 AI057157 from the Southeastern Regional Center of Excellence for Emerging Infections and Biodefense. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

S. enterica is a Gram negative, enteric bacterial pathogen that can infect diverse hosts including birds, reptiles and mammals. S. typhimurium causes a self-limiting gastroenteritis in humans whereas the closely related S. enterica serovar Typhi (S. typhi) causes typhoid fever, a frequently fatal systemic disease. S. enterica infection is a major public health problem causing more than one billion new human infections each year that lead to more than three million deaths [1]. The problem is greatly exacerbated by the emergence of multi-drug resistant strains [2]. In addition to public health concerns, S. typhimurium is also studied because it is a model pathogen without parallel for dissecting basic pathogenic processes.

S. enterica promotes its virulence with a class of widespread secretion systems termed type III (TTSS). These sophisticated molecular devices function as ‘molecular syringes’ that span the bacterial envelope and inject proteins into host cell cytosol to subvert various cellular functions. S. enterica possesses two distinct TTSSs encoded by Salmonella pathogenicity islands 1 and 2 (SPI-1 and SPI-2). S. enterica primarily utilizes SPI-1 to invade cells and invoke the inflammatory response [3]–[5], and subsequently SPI-2 to proliferate within cells [6]–[8] and to manipulate the migratory properties of phagocytes [9], [10].

Most studies relating to S. enterica-host cell interactions to date have focused on microbial virulence factors. While much has been learned about microbial factors that promote growth within hosts, very little is known about the host cell factors that S. enterica uses to facilitate its proliferation. The identification of these HSFs could shed much light on the molecular mechanisms underlying the ability of S. enterica to cause disease and could also serve as a large class of novel, anti-microbial targets.

New antibiotics are desperately needed to control infectious diseases caused by S. enterica as well as by related intracellular pathogens. Multi-drug resistant strains of S. typhi are now commonplace [1]. Since 1989, strains of S. typhi resistant to chloramphenicol, ampicillin, and trimethoprim have caused numerous outbreaks [11]. As a result of the widespread dissemination of such strains, chloramphenicol was withdrawn as the first-line drug for typhoid fever and replaced with fluoroquinolones and third generation cephalosporins [12]. However, outbreaks of typhoid fever caused by strains resistant to nalidixic acid and ciprofloxacin have become endemic in the Indian subcontinent and have also been reported in the US and UK among other developed countries, reflecting the emergence of a global problem [13]. The presence of a plasmid-borne integron in ciprofloxacin-resistant S. typhi may soon produce widespread instances of nearly intractable typhoid fever [13]. The gravity of this public health problem is heightened by the fact that there are few druggable S. enterica targets remaining to inhibit. A recent systematic network analysis of S. typhimurium metabolism in the murine model of typhoid fever determined that nearly all S. typhimurium enzymes are non-essential, due to extensive metabolic redundancies [14]. Of those that are essential, nearly all belong to metabolic pathways already inhibited by current antibiotics [14]. There is thus a pressing need to develop a new class of drug targets which S. enterica and related pathogens will be unable to quickly evolve ways to overcome.

Targeting HSFs might prove useful in treating S. enterica infections because S. enterica is primarily intracellular in the systemic phase of disease, as gentamicin does not resolve infections, and mutants sensitive to macrophage killing are avirulent [15]. Second, point mutations, which are the most frequent form of mutation, exemplified by ciprofloxacin resistance being produced by point mutations in gyrA and parC [13], would not likely enable the bacteria to overcome this new class of drug, as resistance would likely require multiple, large genetic changes. Unfortunately, only a few HSFs for S. enterica, most notably AKT1, have been reported [16].

We recently completed a global RNAi screen to identify host factors that are subverted by S. Typhimurium to promote its intracellular proliferation. In all, we found 252 human genes that promote S. typhimurium growth within MCF-7 cells whose transient absence does not affect host cell viability. This study will facilitate the delineation of the molecular mechanisms underlying the ability of S. enterica to persist and replicate within human host cells and provides a large, new class of candidate, anti-microbial targets for controlling S. enterica infections, and potentially those caused by other human pathogens as well.

Results

Human Epithelial Cell Screen to Identify Host Factors Required for Infection by S. typhimurium

To identify human genes that modulate the intracellular growth of S. typhimurium, we developed a GFP fluorescence plate reader-based high throughput RNAi assay with human cells (Fig. 1). We tried intensively to develop a high throughput assay with macrophages but were unsuccessful for a variety of technical reasons, including but not limited to their low transfection efficiency. Ultimately, we resorted to MCF-7 cells, a human epithelial cell line, which can be transfected with >95% efficiency in a high throughput format. For the genome-wide screen, ≈22, 000 predicted genes were targeted with siRNAs arrayed in 384 well plates using a reverse transfection protocol with MCF-7 cells. The screen was performed in a 2×2 format, meaning that each gene was targeted by two independent siRNAs in one well and another two independent siRNAs in a second well on a duplicate plate, reducing the likelihood that off-target effects contributed to the phenotypes. Seventy-two hours following transfection, cells were infected with S. typhimurium expressing a plasmid-encoded copy of the GFP at a multiplicity of infection (MOI) of 100 for thirty minutes. The extracellular bacteria were then killed with gentamicin. Eighteen hours later, the number of host cells present in each well and the GFP intensity of infected cells were determined.

thumbnail
Download:
Figure 1. Global RNAi screen of Salmonella-infected human epithelial cells.

(A) Procedural outline for the screen. (B) The plate layout. Wells that contained siRNAs that result in cell death are in yellow and served as a positive control for transfection. The wells colored red contained anti-AKT1 siRNA, which served as a positive control for the reduction of intracellular S. typhimurium growth. The wells colored black contained non-specific siRNA and the wells colored green contained anti-GFP siRNA, which is unrelated to the GFP expressed by the bacteria. Both the non-specific siRNA and anti-GFP siRNA served as negative controls. Water wells, intended to reduce edge effects are in grey, and the experimental wells are in blue and purple.

https://doi.org/10.1371/journal.pone.0038097.g001

Image analysis was carried out by first identifying cell nuclei, which were stained with Hoechst 33342, and extending a boundary a predetermined radial distance from the nuclear border. This region, called a circ, corresponds to an area equivalent to the cell boundary. Only bacteria within this region were analyzed to exclude any residual extracellular bacteria. Targets that modulated intracellular S. typhimurium growth were considered hits if they reduced or increased GFP intensity by ≥2.12 standard deviations versus the non-silencing controls on each of the duplicate plates. Thus, the combined Z-score using this two-dimensional analysis was three. We did not compare individual wells to the plate mean to avoid missing legitimate hits as the siRNAs were grouped by functional class. The kinase plate for example would be expected to have numerous hits. Each plate contained several key controls. Some wells contained siRNAs that are lethal to eukaryotic cells. We inspected these wells on all plates to ensure that transfection had occurred. Anti-AKT1 siRNA was also present on every plate and served as a positive control for microbial growth reduction. The plate layout is shown in Figure 1b. The screen strongly enriched the bank of siRNAs for the few that either facilitate or deter S. typhimurium intracellular growth in a reproducible fashion in regards to the duplicate plates (Fig. S1).

We ensured that there was no correlation between fluorescence changes and changes in host cell viability by excluding wells that had fewer than 800 cells present. We empirically determined with the first four plates of the screen that there was no longer a correlation between the fluorescence intensity of individual, infected cells and the total number of cells present beyond 800 cells per well (Fig S2). We included gentamicin in our assay, as it is impossible to eliminate the growth of extracellular bacteria without it. We cannot rigorously exclude the possibility that a minority of the genes that appeared to promote microbial growth as measured by a decrease in fluorescence owed their phenotypes to the antibiotic accumulating within cells. However, it is unlikely that very many of them do because in most instances, a cell that had a change in its physiology that resulted in it accumulating gentamicin would be unhealthy.

Analysis of the Host Network that Facilitates S. typhimurium Growth within Human Cells

In all, 252 genes were identified in the primary screen as facilitating intracellular microbial growth, and 39 were identified as normally repressing microbial growth. These genes and their annotations are shown Table S1 and Table S2. The data were further analyzed with IPA (Ingenuity® Systems, www.ingenuity.com). An IPA-generated interaction network including the 252 HSFs is shown in Figure S3. The most central node to the host cell network that promotes S. typhimurium growth is the transcription factor NFκB, which among many other things, controls the inflammatory response of host cells. This is perhaps not surprising as a profound inflammatory response in the intestinal epithelium, elicited by the SPI-1 encoded TTSS is a critical feature of S. typhimurium pathogenesis [5], [17]. A host response by-product of the inflammation is tetrathionate, a new respiratory electron acceptor that enables S. typhimurium to use respiration to compete with fermenting gut microbes, thereby enhancing its transmission [18].

In addition to NFκB, centrally located nodes in the host network that permit S. typhimurium growth are a series of kinases including MapK, AKT1, p38 MAPK and the pI3K complex. The SPI-1 TTSS is known to lead to the activation of the mitogen-activated protein kinases, ERK, JNK and p38 [5], which are also near the center of the network. These pathways induce signaling to the nucleus and cytokine production. Central to the network also is AKT1, which was previously shown to play a large role in mediating intracellular survival of S. typhimurium and S. typhi [16]. Four sub-networks that could plausibly play roles in promoting S. typhimurium growth within epithelial cells are displayed in Figure 2, including ones that function in cellular growth and development, cell death, cell cycle, and carbohydrate metabolism. Many of the molecules at the center of the global network as well as the molecules in the center of the sub-networks were not identified in the screen. It is likely that depleting the cells of these proteins adversely affected their viability. The IPA-generated, overrepresented molecular function categories for the down network are displayed in Figure 3a, with cellular growth (p-value  = 7×10−9), cell cycle (p-value  = 8×10−8), cellular development (p-value  = 5×10−7) and carbohydrate metabolism (p-value  = 5×10−6) being the most prominent. The genes that the various categories are composed of are shown in Table S3. The over-represented molecular function categories from the up network are shown in Figure 3b. Members of the over-represented categories from the up network are listed in Table S4. The categories of cell cycle and cell death were both highly enriched for in both the up and down networks. Also noteworthy, was the identification of numerous components of the proteasome as very strong up hits. Perhaps, depleting cells of proteasome components allows for type III effectors to linger, thereby facilitating enhanced microbial growth.

thumbnail
Download:
Figure 2. Four prominent sub-networks and associated functions that could plausibly play roles in promoting microbial growth.

(A) Cellular development, cellular growth. (B) Cell death. (C) Cell cycle. (D) Carbohydrate metabolism. The shaded molecules are the ones identified in the screen. The others were added by IPA to generate the sub-networks.

https://doi.org/10.1371/journal.pone.0038097.g002

thumbnail
Download:
Figure 3. The over-represented molecular and cellular function categories of the entire up and down networks.

The entire IPA-generated networks that modulate the intracellular growth of S. typhimurium were analyzed for over-represented functional categories. (A) The network that promotes growth. (B). The network that restricts growth. The significance threshold (p = 0.05) is the red, dashed line. The X axis is the –log of the p-values.

https://doi.org/10.1371/journal.pone.0038097.g003

The Identification of new HSFs that could Play Plausible Roles in Promoting S. typhimurium Growth within Human Cells

Some of the more interesting, new HSFs identified, which could play plausible roles in mediating bacterial pathogenesis are listed in Table 1, and the raw microscopy data for these hits is shown in Figure 4. These twenty-three hits fall into the general categories of vesicular trafficking, vacuole acidification, signal transduction, lipid synthesis and metabolism, ubiquitylation, carbohydrate metabolism and transport.

thumbnail
Download:
Table 1. Some of the more interesting hits identified and their characteristics.

https://doi.org/10.1371/journal.pone.0038097.t001

We identified at least five HSFs that regulate vesicular trafficking events. Although its functions remain to be fully elucidated, the SPI-2 TTSS likely primarily serves to direct vesicular trafficking events in a manner that renders the Salmonella-containing vacuole (SCV) more hospitable. This presumably includes avoiding SCV-lysosome fusion and also enlarging the SCV and importing important metabolic molecules such as amino acids and lipids. Rab1b and Sec22a are a Rab GTPase and SNARE respectively. Both function in the early secretory pathway. Rab1b regulates anterograde traffic between the ER and the Golgi apparatus. A network of Rab GTPases controls phagosome maturation in epithelial cells and is modulated by S. typhimurium [19]; however, this network was not known to include Rab1b. Coxiella burnetii is known however to recruit Rab1b to its replicative vacuoles and it is required for their proper biogenesis [20]. Similarly, Legionella pneumophila recruits Rab1b to the vacuole that it resides within (LCV), which is important for promoting trafficking from the ER to the vacuole and bacterial growth [21]. Sec22a has not previously been implicated in microbial pathogenesis; however, Sec22b is recruited to the LCV and its depletion from cells reduces the replication efficiency of the bacteria [21]. The M. tuberculosis secreted protein, PtpA, dephosphorylates the S. typhimurium HSF, VPS33b, a regulator of membrane fusion. This interaction inhibits phagosome-lysosome fusion [22]. ARL17P1 and CPNE5 also regulate vesicular trafficking events, but were not previously appreciated as playing roles in microbial virulence [23].

We identified two components of the vacuolar ATPase that acidify intracellular compartments. Numerous pathogens use acid as a cue to escape the vacuole they are within and/or activate virulence gene expression [24]. Acidification of the SCV is important for triggering S. typhimurium virulence gene expression in J774A macrophages, but not in other macrophage cell lines, and not in HeLa, MDCK or Henle epithelial cell lines [25]. MCF-7 cells, to the best of our knowledge, have never been tested. The effect, if any, of vacuole acidification on S. enterica virulence during murine or human infections remains to be determined.

Microbial pathogens often interdict host cell signal transduction pathways and manipulate lipid synthesis and metabolism in ways that promote their virulence. MTMR3 is a phosphatase that acts on lipids with a phosphoinositol (PtdIns) headgroup, with activity towards PtdIns-3-phosphate (PtdIns(3)P) and PtdIns-3,5-bisphosphate (PtdINS(3,5)P2) [26]. These activities have been implicated in producing enlarged SCVs that are diverted from the endocytic pathway [27]. ITPKC phosphorylates inositol 2, 4, 5-triphosphate to inositol 2, 4, 5, 6-tetraphosphate. This particular lipid has not previously been implicated in promoting microbial growth. HRBL is an ARF GTPase activator that also was not previously appreciated as playing a role in microbial virulence.

We identified at least three more genes involved in lipid synthesis and metabolism. AADACL1 is believed to be involved in cholesterol ester hydrolysis into their component sterols and fatty acids [28]. It has not previously been implicated in microbial pathogenesis; however, S. typhimurium encodes an outer membrane esterase, ApeE, which may be important for virulence as it is not present in Escherichia coli [29]. Additionally, ApeE expression is induced by phosphate starvation [30], a condition that S. typhimurium likely encounters in the SCV as SPI-2 genes can be strongly induced by it [31]. Elovl5 catalyzes the synthesis of monounsaturated and polyunsaturated very long chain fatty acids, which are structural components of sphingolipids [32], [33]. S. typhimurium was shown previously to utilize the SPI-2 TTSS to redirect the Golgi to plasma membrane traffic of a sphingolipid to the SCV [34]. S. typhimurium encodes a protein with remarkable similarity to human glucosyl ceramidase and thus could potentially use host sphingolipids as an energy source [35]. Hac1L is involved in lipid and fatty acid metabolism, and was not previously thought to play in role in virulence.

Three additional HSFs that were not previously implicated in microbial virulence are UBE2L6, KHK and AMDHD2. UBE2L6 mediates the covalent attachment of ubiquitin to other proteins and this HSF thus falls into the emerging theme of microbial pathogens manipulating ubiquitylation processes to promote virulence. KHK and AMDHD2 are involved in carbohydrate metabolism and could potentially provide a source of energy to S. typhimurium.

We identified at least three proteins that play roles in transport. It stands to reason that intra-vacuolar pathogens such as S. typhimurium could benefit from manipulating host proteins involved in transport to obtain host factors that facilitate the growth of the microbe within the normally, inhospitable SCV. FTHL17 is involved in iron transport. The iron withholding mechanisms of host cells are essential for controlling microbial infections. Successful pathogens, such as S. typhimurium have developed numerous, clever strategies for circumventing this innate host defense. One among numerous bacterial counter-measures is the sitABCD iron acquisition operon of S. typhimurium that is induced following invasion of the intestinal epithelium [36]. FTHL17 has never been shown before to support microbial growth; however, it is reasonable to think that it could as it is involved in iron transport. FABP5L3 has likewise not been implicated in microbial pathogenesis previously. However, as it transports fatty acids with a high specificity, which could be useful to S. typhimurium, it is not unreasonable to think that it does play a role in supporting virulence. SLC29A3 belongs to the SLC29A transporter family. It localizes to enodosmes and lysosomes normally where it mediates the influx and efflux of nucleosides [37], [38]. Rickettsial organisms, which are obligate intracellular parasites, are known to scavenge nucleosides from infected host cells [39]; however, S. typhimurium as a free-living organism can synthesize its own nucleosides. But, it is possible that S. typhimurium augments endogenous nucleosides with ones from the host cell for DNA and RNA synthesis and additionally, nucleoside tri-phosphates could of course serve as an energy source.

Four additional genes among many identified in this screen that could plausibly play roles in permitting the growth of S. typhimurium within host cells are STBD1, PSPH, EXTL3 and GolgA1. STBD1 is interesting in that it has a microbial starch-binding domain, which is involved in glycogen metabolism. Glycogen is a polysaccharide that is the principal storage form of glucose in human cells, and could serve as an energy source for the bacteria residing within the SCVs. STBD1 is in fact normally localized to late endosomes and lysosomes where it anchors glycogen to the membranes [40], [41]. PSPH catalyzes the last step in the biosynthesis of serine from carbohydrates. It seems possible that the bacteria scavenge serine from the host cells, considering the presumably stringent nutrient conditions of the SCV. S. typhimurium infection has been shown to up-regulate the expression of the cationic amino acid transporters mCAT1 and mCAT2B, which regulate the availability of arginine within macrophages. Interestingly, during infections they localize in close proximity to the SCV. The intra-vacuolar bacteria then use their own arginine transporter to acquire the amino acid [42]. S. typhimurium encodes a putative serine transport protein, SdaC [43]. Perhaps manipulating host amino acid synthesis and localization will prove to be a common strategy deployed by pathogens. EXTL3 is a probable glycosyltransferase involved in the metabolism of glycan and heparin sulfate, which are well known for their roles in mediating pathogen attachment and subsequent invasion of host cells [44]. Glycans have not however previously been implicated in promoting the intracellular growth of microbes. It is possible that EXTL3 creates useful energy sources for S. typhimurium. GolGA1 is a member of a family of Golgi-targeted coil-coil proteins that bind Rab6a [45].

Comparison of the Host Networks that Promote the Intracellular Growth of S. typhimurium and M. tuberculosis

We compared our results to the only other global RNAi screen performed on human cells infected with bacteria. This screen analyzed THP-1 human macrophage-like cells infected with M. tuberculosis [46]. This study identified 272 human HSFs for M. tuberculosis. An IPA-generated network of these HSFs yielded 501 molecules (data not shown) whereas the S. typhimurium network is composed of 474 molecules. Interestingly, they have 49 molecules in common. The union of the two networks is show in Figure S3, and the members are listed in Table S5. The most over-represented molecular function category for the 49 molecules that the two networks share is autophagy (p-value <4.7×10−11).

Discussion

Although much is known about the microbial virulence factors that S. enterica deploys to invade and subsequently grow within host cells, relatively little is known about the cellular processes that modulate virulence. This genome-wide screen identifies the human genes that are exploited by S. enterica to facilitate intracellular growth. In addition to providing a wealth of new information that will ultimately facilitate the unraveling of many of the molecular mechanisms underlying the ability of S. enterica to persist and proliferate within human cells, all of the 252 factors identified in this study are candidates for a new class of anti-microbial targets.

The host gene network derived from this study that facilitates the intracellular growth of S. typhimurium within human epithelial cells has significant overlap with the one that promotes the growth of M. tuberculosis within human macrophages. This suggests a conservation of basic processes between the two cell types and may point to new drug targets that could have broad-spectrum activity. It is interesting that the shared molecules, for the most part, are the central nodes of both networks, perhaps revealing some of the shared, core, underlying logic through which different pathogens manipulate host cells. The union of the two networks contains at least 18 genes that regulate autophagy. Curtailing the growth of intracellular pathogens with autophagy is an ongoing theme in host-pathogen interplay. Following internalization, S. typhimurium can damage the SCV with its SPI-1 encoded TTSS. This damage is believed to produce an intracellular bacterial population that is targeted by the autophagy system of the host cell [47]. In fact, cells deficient in autophagy permit more intracellular S. typhimurium growth than normal cells, due to the creation of a population of bacteria in the cytosol that can replicate rapidly in this nutrient-rich environment [47]. Kumar et. al reported that the most conserved members of the host survival networks that permit the persistence of diverse strains of M. tuberculosis function through the regulation of autophagy [46]. Group A Streptococcus pyogenes deploys Streptolysin O to lyse phagosomes and gain access to the cytosol, following which autophagy limits its growth [48]. Rickettsiae spp. also gain access to the cytosol and can be targeted by autophagy for destruction [49], [50]. Thus, it is possible that negative regulators of autophagy could serve as anti-microbial targets for numerous pathogens. However, as some of the 18 genes that the S. typhimurium and M. tuberculosis host survival networks share that regulate autophagy can promote it, and as these genes have numerous, and in some instances diverse functions within host cells, further work will be required to clarify the role of these genes in permitting intracellular, microbial growth.

The most over-represented molecular function categories from the down network, namely cell growth and development, cell cycle, and carbohydrate metabolism are consistent with some of the general themes of host-microbe interactions. It was interesting that the categories of cell death and cell cycle were significantly over-represented in both the up and down hits. Microbial pathogens are well known for their abilities to kill cells by triggering apoptotic pathways as well as causing necrotic death and conversely to block apoptosis. Pathogens trigger cell death in order to avoid killing by professional phagocytes, or to escape a cell that cannot support additional microbial growth, or to modulate the immune response. Pathogens can also activate cell survival pathways and interdict apoptotic signaling in order to create a safe, replicative niche [51]–[54]. In the screen described here, it is likely that the genes identified in the up hits promoted cell death whereas the ones in the down hits suppressed it. In addition to apoptotic pathway manipulation, numerous pathogens are known to manipulate the host cell cycle machinery in various ways, sometimes blocking the cell cycle at transition phases and also sometimes forcing progression through the cell cycle [55]. In future work, it will be interesting to learn more about why so many of the up and down hits influence cell cycle and to dissect how S. enterica manipulates these genes to its advantage.

Several HSFs identified warrant further consideration. The hits in the trafficking category may be of further interest as understanding exactly how they subvert the endocytic and exocytic pathways would greatly enhance our understanding of SCV biogenesis, which is intimately tied to virulence. We likely did not identify more host proteins known to play roles in regulating intracellular trafficking as it was observed that the depletion of single ER to Golgi transport proteins rarely decreased L. pneoumophila intra-vacuole replication. Combination knock-downs on the other hand did reduce L. pneoumophila growth, suggesting that membrane traffic funnels into the vacuole from multiple sources [23].

MTMR3 is an intriguing HSF that defies easy classification as it is involved in lipid metabolism, signal transduction and also regulating the endocytic pathway. Saccharomyces cerevisiae cells deficient in PtdIns(3,5)P2 synthesis exhibit grossly enlarged vacuoles analogous to the spacious phagosomes S. typhimurium resides within [56]. The levels of PtdIns(3)P and PtdIns(3,5)P2 are established by the activities of the PtdIns(3)P 5-kinase, Fab1, and the PtdIns(3,5)P2 phosphatase, FIG4, and also by members of the myotubularin lipid phosphatase family. It was previously proposed that the PtdIns(3,5)P2 phsophatase activity of the Salmonella SPI-1 type III effector SopB may play a role analogous to FIG4, preventing the transition of PtdIns(3)P to PtdINS(3,5)P2, thereby resulting in enlarged vesicles and diverting the SCV from the endocytic pathway [27]. Depleting cells of MTMR3 would remove one of the mechanisms through which PtdINS(3,5)P2 is hydrolyzed as well as removing PtdIns(3)P substrates for Fab1. On the other hand, high levels of of PtdIns(3)P in the membrane of SCVs was proposed to stimulate homotypic vesicle fusion with other PtdIns(3)P containing empty vesicles formed during bacterial infection to promote virulence, attributable to the PtdIns(3,4,5)P3 phosphatase activity of SopB. SopB and MTMR3 may have somewhat antagonistic activities, which allow for the optimal manipulation of vesicular fate. Clearly, more work will be required to fully understand how MTMR3 is subverted by S. typhimurium to promote its growth within human cells.

thumbnail
Download:
Figure 4. Raw microscopy data from the screen with some of the more interesting hits.

Host cells are in blue, S. typhimurium is green. Only one representative non-silencing control (upper left corner) and one positive control (anti-AKT1), which is adjacent to the non-silencing control are shown due to space constraints. All of the positive and negative controls appeared similar.

https://doi.org/10.1371/journal.pone.0038097.g004

It is perhaps not surprising that we identified a ubiquitin conjugating enzyme as an HSF considering that S. typhimurium secretes multiple SPI-1 and SPI-2 effectors into infected cells that interfere with ubiquitin-related processes. SopA and SspH2 are E3 ligases, whereas SseL is a deubiquitinase required for macrophage killing through the delayed pathway [52], [53]. The effectors SopA, SopE, SptP and SopB are all ubiquitinated in ways that provide spatiotemporal regulation and diversification of function [57], [58]. Clearly, manipulating ubiquitin with secreted bacterial proteins and with host proteins is a major feature of S. typhimurium pathogenesis. It will be interesting to determine in future work how UBE2L6 fits into the bigger picture.

GolgA1 is another interesting HSF in that it could conceivably be involved in tethering the SCV to the Golgi apparatus, a phenomenon that is critical to S. typhimurium virulence [59]. In infected epithelial cells, SCVs migrate to a perinuclear location where they become surrounded by membranes of the Golgi network. This subcellular localization requires the SPI-2 effector SseG. Both the N- can C-terminal domains of SseG are exposed on the cytosolic face of the SCV. sseG mutants are unable to localize the vacuoles they reside within to the Golgi and are unable to multiply. The Golgi-targeting sequence of SseG appears to be novel as there is no similarity between SseG and either prokaryotic or eukaryotic Golgi targeting sequences [59]. The family of proteins that GolgA1 belongs to was proposed to function in Rab6-regulated membrane tethering events [45]. Although purely speculative, it is interesting to consider that SseG might interact with GolgA1, tethering the SCV to the Golgi apparatus. Alternatively, GolgA1 could play a role in manipulating Golgi structure or perhaps exocytic events in a way that benefit the bacteria within the SCV.

With a genome-wide RNAi-based forward genetic screen, we have identified 291 host factors that either promote or repress the growth of S. typhimurium within human epithelial cells. Additionally, this study identifies 49 molecules that are needed by both S. typhimurium and M. tuberculosis to grow optimally within human cells. In future work, it will be important to determine what the core group of HSFs is that are required for both S typhimurium and S. typhi survival within both epithelial cells and macrophages, and further which ones are required by multiple intracellular pathogens. This may help us understand the shared logic though which all pathogens manipulate host cells, and allow for the rational design of new drugs for treating infectious disease. It will also be interesting to study some of the HSFs identified in detail. The genome-wide results from our human RNAi screen provide us with some new insights into the Salmonella intracellular lifestyle and serve as an important starting point for delineating the molecular mechanisms through which intracellular pathogens manipulate host cells.

Materials and Methods

Bacterial and Eukaryotic Cell Growth Conditions

S. typhimurium 14028s carrying a derivative of pACYC184 which expresses the GFP was grown in Luria Bertani broth (LB; Sigma-Aldrich, St. Louis, MO) for 18 hours at 37°C with agitation. It was then diluted 1∶33 in fresh LB and subcultured for 3.5 hours at 37°C with agitation to induce SPI-1 expression prior to infection of MCF-7 cells (ATCC, Manassas, VA). MCF-7 cells were grown in Iscove’s Modified Dulbecco’s Medium (IMDM; Invitrogen, Carlsbad, CA) supplemented with 7.5% fetal bovine serum (FBS; Sigma-Aldrich) without antibiotics. MCF-7 cells were maintained in a humidified tissue culture incubator at 37°C in 5% CO2 and subcultured every four days at 8×104 cells/cm2.

Infections and Gentamicin Protection Assay

MCF-7 cells were infected with S. typhimurium at an MOI of 100, following which cells were centrifuged at 230 relative centrifugal force for 10 minutes at room temperature. Infected cells were incubated at 37°C in 5% CO2 for 30 minutes to allow for invasion and then washed two times. After washing, the cells were incubated in IMDM supplemented with gentamicin (Gibco, Carlsbad, CA) at a concentration of 100 µg/mL for one hour at 37°C in 5% CO2 to prevent further invasion and to eliminate extracellular bacteria. Following the gentamicin kill, cells were washed two times and incubated in IMDM supplemented with gentamicin at a concentration of 10 µg/mL and incubated at 37°C in 5% CO2 for 18 hours before preparing plates for imaging.

High-throughput siRNA Screen of Human Genome

A siRNA library was obtained from the Duke University RNAi screening facility targeting the human genome (Human whole-genome siRNA library v1.0; Qiagen, Valencia, CA) with four individual siRNAs targeting each gene arrayed in two sets, AB and CD, in a 2×2, parallel plate format. MCF-7 cells were seeded at a density of 3000 cells per well in 384 well clear flat bottom plates (Corning 3712; Corning, Lowell, MA). Reverse transfections were performed using 0.05 µL Dharmafect 4 (Dharmacon, Lafayette, CO) and 20 nM siRNA perwell in a total volume of 50 µL. Transfection efficiency was optimized utilizing AllStars Hs Cell Death Control siRNA (Qiagen) and was >95% under these conditions. Seventy-tw0 hours after transfection the cells were infected with S. typhimurium expressing the GFP that were induced for SPI-1 expression. Infected MCF-7 cells were fixed eighteen hours after infection with paraformaldehyde (Sigma-Aldrich) in 1X PBS (Invitrogen) for 1 hour. Nuclei were stained with Hoechst 33342 (Sigma-Aldrich) and wells were imaged and analyzed with the Cellomics ArrayScan VTI High Content imaging system (Thermo Scientific, Pittsburgh, PA). Only infected cells were analyzed to exclude any residual extracellular bacteria. The average intensity of all circs within a well was measured. These were then compared with the mean circ spot average intensity of the negative controls. We compared the experimental wells to the wells containing non-specific siRNA, which proved to be a more conservative approach than comparing them to the anti-GFP wells. The anti-GFP siRNA was unrelated to the GFP expressed by the bacteria.

Bioinformatics

The gene sets identified in our screen were manually curated with Uniprot (http://www.uniprot.org/). Other bioinformatics analyses were performed with IPA (Ingenuity® Systems, www.ingenuity.com). A data set containing gene identifiers was uploaded into the application. Molecules with IDs that could be mapped were overlaid onto a global molecular network developed from information contained in the Ingenuity Knowledge Base. Networks of network eligible molecules were then algorithmically generated based on their connectivity, and merged. Functional analysis identified the molecular and cellular functions that were most significant to the entire network that was generated. Right-tailed Fisher’s exact test was used to calculate a p-value determining the probability that each molecular function assigned was due to chance alone. The network generated is a graphical representation of the molecular relationships between molecules. Molecules are represented as nodes, and the biological relationship between two nodes is represented as an edge (line). All edges are supported by at least one reference from the literature, from a textbook, or from canonical information stored in the Ingenuity Knowledge Base.

Supporting Information

Figure S1.

Distribution of the standard deviations of the average infected cell fluorescence intensity from each well in the screen. (A) The standard deviations of all experimental wells from the mean of the non-specific siRNA negative controls are plotted. The cutoffs for accepting hits were plus 3 or minus 3, which are indicated with arrows. Thus, there was a strong enrichment in the screen. (B) The standard deviations of the hits are plotted. The means for the negative controls should be around 0, 0.

https://doi.org/10.1371/journal.pone.0038097.s001

(TIF)

Figure S2.

Changes in fluorescence were not attributable to changes in host cell viability. (A) We observed a linear relationship between average cell fluorescence intensity and cell numbers up to 800 cells/well. (B) Above 800 cells/well there was no longer a linear relationship between cell numbers and the average fluorescence intensity of infected cells. The lack of linearity above 800 cells/well reveals that increasing the viable cells beyond this point does not increase microbial growth in individual, infected cells, decoupling host cell viability from microbial growth.

https://doi.org/10.1371/journal.pone.0038097.s002

(TIF)

Figure S3.

The network generated from the 252 HSFs identified in this study. Molecules are represented as nodes and lines represent interactions between them. The molecule labels in blue are the union of the S. typhimurium and M. tuberculosis networks.

https://doi.org/10.1371/journal.pone.0038097.s003

(TIF)

Table S1.

The uniprot derived gene annotations and the raw numerical data for the down hits.

https://doi.org/10.1371/journal.pone.0038097.s004

(XLSX)

Table S2.

The uniprot derived gene annotations and the raw numerical data for the up hits.

https://doi.org/10.1371/journal.pone.0038097.s005

(XLSX)

Table S3.

The members of the over-represented molecular function categories for the down hits are listed.

https://doi.org/10.1371/journal.pone.0038097.s006

(XLSX)

Table S4.

The members of the over-represented molecular function categories for the up hits are listed.

https://doi.org/10.1371/journal.pone.0038097.s007

(XLSX)

Table S5.

The molecules that the union of the S . typhimurium and M . tuberculosis host survival networks are composed of are listed.

https://doi.org/10.1371/journal.pone.0038097.s008

(XLSX)

Acknowledgments

We are indebted to all members of the Valdivia laboratory and the Duke RNAi screening facility for their hospitality and guidance. In particular, we are grateful to Nicholas Barrows, James Pearson and Mariano Garcia-Blanco. We thank Amy Thornbrough for proofreading the manuscript.

Author Contributions

Conceived and designed the experiments: JMT RV MJW. Performed the experiments: JMT TH MJW. Analyzed the data: JMT MJW. Contributed reagents/materials/analysis tools: JMT RV MJW. Wrote the paper: MJW.

References

  1. 1. Pang T, Bhutta ZA, Finlay BB, Altwegg M (1995) Typhoid fever and other salmonellosis: a continuing challenge. Trends Microbiol 3: 253–255.
  2. 2. Boyle EC, Bishop JL, Grassl GA, Finlay BB (2007) Salmonella: from pathogenesis to therapeutics. J Bacteriol 189: 1489–1495.
  3. 3. Groisman EA, Ochman H (1993) Cognate gene clusters govern invasion of host epithelial cells by Salmonella typhimurium and Shigella flexneri. Embo J 12: 3779–3787.
  4. 4. Galyov EE, Wood MW, Rosqvist R, Mullan PB, Watson PR, et al. (1997) A secreted effector protein of Salmonella dublin is translocated into eukaryotic cells and mediates inflammation and fluid secretion in infected ileal mucosa. Mol Microbiol 25: 903–912.
  5. 5. Hobbie S, Chen LM, Davis RJ, Galan JE (1997) Involvement of mitogen-activated protein kinase pathways in the nuclear responses and cytokine production induced by Salmonella typhimurium in cultured intestinal epithelial cells. J Immunol 159: 5550–5559.
  6. 6. Cirillo DM, Valdivia RH, Monack DM, Falkow S (1998) Macrophage-dependent induction of the Salmonella pathogenicity island 2 type III secretion system and its role in intracellular survival. Mol Microbiol 30: 175–188.
  7. 7. Hensel M, Shea JE, Waterman SR, Mundy R, Nikolaus T, et al. (1998) Genes encoding putative effector proteins of the type III secretion system of Salmonella pathogenicity island 2 are required for bacterial virulence and proliferation in macrophages. Mol Microbiol 30: 163–174.
  8. 8. Ochman H, Soncini FC, Solomon F, Groisman EA (1996) Identification of a pathogenicity island required for Salmonella survival in host cells. Proceedings of the National Academy of Sciences of the United States of America 93: 7800–7804.
  9. 9. Worley MJ, Nieman GS, Geddes K, Heffron F (2006) Salmonella typhimurium disseminates within its host by manipulating the motility of infected cells. Proc Natl Acad Sci U S A 103: 17915–17920.
  10. 10. McLaughlin LM, Govoni GR, Gerke C, Gopinath S, Peng K, et al. (2009) The Salmonella SPI2 effector SseI mediates long-term systemic infection by modulating host cell migration. PLoS Pathog 5: e1000671.
  11. 11. Rowe B, Ward LR, Threlfall EJ (1997) Multidrug-resistant Salmonella typhi: a worldwide epidemic. Clin Infect Dis 24: Suppl 1S106–109.
  12. 12. Bhan MK, Bahl R, Bhatnagar S (2005) Typhoid and paratyphoid fever. Lancet 366: 749–762.
  13. 13. Gaind R, Paglietti B, Murgia M, Dawar R, Uzzau S, et al. (2006) Molecular characterization of ciprofloxacin-resistant Salmonella enterica serovar Typhi and Paratyphi A causing enteric fever in India. J Antimicrob Chemother 58: 1139–1144.
  14. 14. Becker D, Selbach M, Rollenhagen C, Ballmaier M, Meyer TF, et al. (2006) Robust Salmonella metabolism limits possibilities for new antimicrobials. Nature 440: 303–307.
  15. 15. Fields PI, Swanson RV, Haidaris CG, Heffron F (1986) Mutants of Salmonella typhimurium that cannot survive within the macrophage are avirulent. Proc Natl Acad Sci U S A 83: 5189–5193.
  16. 16. Kuijl C, Savage ND, Marsman M, Tuin AW, Janssen L, et al. (2007) Intracellular bacterial growth is controlled by a kinase network around PKB/AKT1. Nature 450: 725–730.
  17. 17. Keestra AM, Winter MG, Klein-Douwel D, Xavier MN, Winter SE, et al. (2011) A Salmonella virulence factor activates the NOD1/NOD2 signaling pathway. MBio 2:
  18. 18. Winter SE, Thiennimitr P, Winter MG, Butler BP, Huseby DL, et al. (2010) Gut inflammation provides a respiratory electron acceptor for Salmonella. Nature 467: 426–429.
  19. 19. Smith AC, Heo WD, Braun V, Jiang X, Macrae C, et al. (2007) A network of Rab GTPases controls phagosome maturation and is modulated by Salmonella enterica serovar Typhimurium. J Cell Biol 176: 263–268.
  20. 20. Campoy EM, Zoppino FC, Colombo MI (2011) The early secretory pathway contributes to the growth of the Coxiella-replicative niche. Infect Immun 79: 402–413.
  21. 21. Kagan JC, Stein MP, Pypaert M, Roy CR (2004) Legionella subvert the functions of Rab1 and Sec22b to create a replicative organelle. J Exp Med 199: 1201–1211.
  22. 22. Bach H, Papavinasasundaram KG, Wong D, Hmama Z, Av-Gay Y (2008) Mycobacterium tuberculosis virulence is mediated by PtpA dephosphorylation of human vacuolar protein sorting 33B. Cell Host Microbe 3: 316–322.
  23. 23. Dorer MS, Kirton D, Bader JS, Isberg RR (2006) RNA interference analysis of Legionella in Drosophila cells: exploitation of early secretory apparatus dynamics. PLoS Pathog 2: e34.
  24. 24. Ray K, Marteyn B, Sansonetti PJ, Tang CM (2009) Life on the inside: the intracellular lifestyle of cytosolic bacteria. Nature reviews Microbiology 7: 333–340.
  25. 25. Steele-Mortimer O, St-Louis M, Olivier M, Finlay BB (2000) Vacuole acidification is not required for survival of Salmonella enterica serovar typhimurium within cultured macrophages and epithelial cells. Infect Immun 68: 5401–5404.
  26. 26. Walker DM, Urbe S, Dove SK, Tenza D, Raposo G, et al. (2001) Characterization of MTMR3. an inositol lipid 3-phosphatase with novel substrate specificity. Curr Biol 11: 1600–1605.
  27. 27. Hernandez LD, Hueffer K, Wenk MR, Galan JE (2004) Salmonella modulates vesicular traffic by altering phosphoinositide metabolism. Science 304: 1805–1807.
  28. 28. Okazaki H, Igarashi M, Nishi M, Sekiya M, Tajima M, et al. (2008) Identification of neutral cholesterol ester hydrolase, a key enzyme removing cholesterol from macrophages. J Biol Chem 283: 33357–33364.
  29. 29. Carinato ME, Collin-Osdoby P, Yang X, Knox TM, Conlin CA, et al. (1998) The apeE gene of Salmonella typhimurium encodes an outer membrane esterase not present in Escherichia coli. J Bacteriol 180: 3517–3521.
  30. 30. Conlin CA, Tan SL, Hu H, Segar T (2001) The apeE gene of Salmonella enterica serovar Typhimurium is induced by phosphate limitation and regulated by phoBR. J Bacteriol 183: 1784–1786.
  31. 31. Deiwick J, Nikolaus T, Erdogan S, Hensel M (1999) Environmental regulation of Salmonella pathogenicity island 2 gene expression. Mol Microbiol 31: 1759–1773.
  32. 32. Leonard AE, Bobik EG, Dorado J, Kroeger PE, Chuang LT, et al. (2000) Cloning of a human cDNA encoding a novel enzyme involved in the elongation of long-chain polyunsaturated fatty acids. Biochem J 350 Pt 3: 765–770.
  33. 33. Ohno Y, Suto S, Yamanaka M, Mizutani Y, Mitsutake S, et al. (2010) ELOVL1 production of C24 acyl-CoAs is linked to C24 sphingolipid synthesis. Proc Natl Acad Sci U S A 107: 18439–18444.
  34. 34. Kuhle V, Abrahams GL, Hensel M (2006) Intracellular Salmonella enterica redirect exocytic transport processes in a Salmonella pathogenicity island 2-dependent manner. Traffic 7: 716–730.
  35. 35. Worley MJ, Ching KH, Heffron F (2000) Salmonella SsrB activates a global regulon of horizontally acquired genes. Mol Microbiol 36: 749–761.
  36. 36. Janakiraman A, Slauch JM (2000) The putative iron transport system SitABCD encoded on SPI1 is required for full virulence of Salmonella typhimurium. Mol Microbiol 35: 1146–1155.
  37. 37. Schroder B, Wrocklage C, Pan C, Jager R, Kosters B, et al. (2007) Integral and associated lysosomal membrane proteins. Traffic 8: 1676–1686.
  38. 38. Baldwin SA, Yao SY, Hyde RJ, Ng AM, Foppolo S, et al. (2005) Functional characterization of novel human and mouse equilibrative nucleoside transporters (hENT3 and mENT3) located in intracellular membranes. J Biol Chem 280: 15880–15887.
  39. 39. Audia JP, Winkler HH (2006) Study of the five Rickettsia prowazekii proteins annotated as ATP/ADP translocases (Tlc): Only Tlc1 transports ATP/ADP, while Tlc4 and Tlc5 transport other ribonucleotides. J Bacteriol 188: 6261–6268.
  40. 40. Janecek S (2002) A motif of a microbial starch-binding domain found in human genethonin. Bioinformatics 18: 1534–1537.
  41. 41. Jiang S, Heller B, Tagliabracci VS, Zhai L, Irimia JM, et al. (2010) Starch binding domain-containing protein 1/genethonin 1 is a novel participant in glycogen metabolism. J Biol Chem 285: 34960–34971.
  42. 42. Das P, Lahiri A, Lahiri A, Sen M, Iyer N, et al. (2010) Cationic amino acid transporters and Salmonella Typhimurium ArgT collectively regulate arginine availability towards intracellular Salmonella growth. PLoS ONE 5: e15466.
  43. 43. McClelland M, Sanderson KE, Spieth J, Clifton SW, Latreille P, et al. (2001) Complete genome sequence of Salmonella enterica serovar Typhimurium LT2. Nature 413: 852–856.
  44. 44. Chen Y, Gotte M, Liu J, Park PW (2008) Microbial subversion of heparan sulfate proteoglycans. Mol Cells 26: 415–426.
  45. 45. Barr FA (1999) A novel Rab6-interacting domain defines a family of Golgi-targeted coiled-coil proteins. Curr Biol 9: 381–384.
  46. 46. Kumar D, Nath L, Kamal MA, Varshney A, Jain A, et al. (2010) Genome-wide analysis of the host intracellular network that regulates survival of Mycobacterium tuberculosis. Cell 140: 731–743.
  47. 47. Birmingham CL, Smith AC, Bakowski MA, Yoshimori T, Brumell JH (2006) Autophagy controls Salmonella infection in response to damage to the Salmonella-containing vacuole. J Biol Chem 281: 11374–11383.
  48. 48. Nakagawa I, Amano A, Mizushima N, Yamamoto A, Yamaguchi H, et al. (2004) Autophagy defends cells against invading group A Streptococcus. Science 306: 1037–1040.
  49. 49. Walker DH, Popov VL, Crocquet-Valdes PA, Welsh CJ, Feng HM (1997) Cytokine-induced, nitric oxide-dependent, intracellular antirickettsial activity of mouse endothelial cells. Lab Invest 76: 129–138.
  50. 50. Rikihisa Y (1984) Glycogen autophagosomes in polymorphonuclear leukocytes induced by rickettsiae. Anat Rec 208: 319–327.
  51. 51. Weinrauch Y, Zychlinsky A (1999) The induction of apoptosis by bacterial pathogens. Annu Rev Microbiol 53: 155–187.
  52. 52. Rytkonen A, Poh J, Garmendia J, Boyle C, Thompson A, et al. (2007) SseL, a Salmonella deubiquitinase required for macrophage killing and virulence. Proc Natl Acad Sci U S A 104: 3502–3507.
  53. 53. van der Velden AW, Lindgren SW, Worley MJ, Heffron F (2000) Salmonella pathogenicity island 1-independent induction of apoptosis in infected macrophages by Salmonella enterica serotype typhimurium. Infect Immun 68: 5702–5709.
  54. 54. Faherty CS, Maurelli AT (2008) Staying alive: bacterial inhibition of apoptosis during infection. Trends Microbiol 16: 173–180.
  55. 55. Oswald E, Nougayrede JP, Taieb F, Sugai M (2005) Bacterial toxins that modulate host cell-cycle progression. Curr Opin Microbiol 8: 83–91.
  56. 56. Rudge SA, Anderson DM, Emr SD (2004) Vacuole size control: regulation of PtdIns(3,5)P2 levels by the vacuole-associated Vac14-Fig4 complex, a PtdIns(3,5)P2-specific phosphatase. Mol Biol Cell 15: 24–36.
  57. 57. Patel JC, Hueffer K, Lam TT, Galan JE (2009) Diversification of a Salmonella virulence protein function by ubiquitin-dependent differential localization. Cell 137: 283–294.
  58. 58. Thomas M, Holden DW (2009) Ubiquitination - a bacterial effector's ticket to ride. Cell Host Microbe 5: 309–311.
  59. 59. Salcedo SP, Holden DW (2003) SseG, a virulence protein that targets Salmonella to the Golgi network. EMBO J 22: 5003–5014.