A variety of pathogenic bacteria use type III secretion pathways to translocate virulence proteins into host eukaryotic cells. YopE is an important virulence factor that is translocated into mammalian cells via a plasmid-encoded type III system in Yersinia spp. YopE action in mammalian cells promotes the disruption of actin filaments, cell rounding and blockage of phagocytosis. It was reported recently that two proteins with sequence similarity to YopE, SptP of Salmonella typhimurium and ExoS of Pseudomonas aeruginosa, function as GTPase-activating proteins (GAPs) for Rho GTPases. YopE contains an ‘arginine finger’ motif that is present in SptP, ExoS and other Rho GAPs and is essential for catalysis by this class of proteins. We show here that a GST–YopE fusion protein stimulated in vitro GTP hydrolysis by the Rho family members Cdc42, RhoA and Rac1, but not by Ras. Conversion of the essential arginine in the arginine finger motif to alanine (R144A) eliminated the in vitro GAP activity of GST–YopE. Infection assays carried out with a Yersinia pseudotuberculosis strain producing YopER144A demonstrated that GAP function was essential for the disruption of actin filaments, cell rounding and inhibition of phagocytosis by YopE in HeLa cells. Furthermore, the GAP function of YopE was important for Y. pseudotuberculosis pathogenesis in a mouse infection assay. Transfection of HeLa cells with a vector that produces a constitutively active form of RhoA (RhoA-V14) prevented the disruption of actin filaments and cell rounding by YopE. Production of an activated form of Rac1 (Rac1-V12), but not RhoA-V14, in HeLa cells interfered with YopE antiphagocytic activity. These results demonstrate that YopE functions as a RhoGAP to downregulate multiple Rho GTPases, leading to the disruption of actin filaments and inhibition of bacterial uptake into host cells.
Yersinia pseudotuberculosis and Yersinia enterocolitica are enteric pathogens of humans and animals and are capable of causing acute ileitis, mesenteric lymphadenitis and septicaemia (Brubaker, 1991; Bottone, 1997). After ingestion by a host, the enteropathogenic Yersinia transit to the terminal ileum and initiate infection by penetrating intestinal lymphoid follicles (Peyer's patches). The bacteria then colonize mesenteric lymph nodes and may spread to deeper tissues such as the spleen. The ability of enteropathogenic Yersinia spp. to replicate extracellularly and avoid phagocytosis and killing is dependent upon a type III protein secretion system encoded on an ≈ 70 kb virulence plasmid (reviewed by Galán and Bliska, 1996; Cornelis and Wolf-Watz, 1997; Cornelis et al., 1998; Hueck, 1998). This type III secretion system is activated upon close contact of the bacterium with the host cell. Tight interaction is mediated by bacterial adhesins, such as invasin, which bind with high affinity to β1-integrin receptors on the surface of host cells (Isberg and Leong, 1990; Isberg and Tran Van Nhieu, 1994). Activation of the type III secretion system permits a number of bacterial virulence proteins called Yops to be translocated from Yersinia directly into the cytosol of the host cell (for reviews, see Straley et al., 1993; Forsberg et al., 1994; Galán and Bliska, 1996; Aepfelbacher et al., 1999). Within minutes after infection, cultured mammalian cells lose actin filaments, round up and eventually detach from the extracellular matrix (Rosqvist et al., 1990; 1991). In the absence of the virulence plasmid, chromosomally encoded invasin promotes efficient uptake of enteropathogenic Yersinia into host cells, including macrophages and epithelial cells (Isberg, 1989; Isberg and Tran Van Nhieu, 1994; Fällman et al., 1995). It is believed that the engagement of β1-integrin receptors by invasin leads to downstream signalling events, which trigger cytoskeletal rearrangements promoting bacterial internalization. However, when enteropathogenic Yersinia harbour the virulence plasmid, bacterial uptake is antagonized by the action of several translocated Yops (Forsberg et al., 1994).
Recently, it was reported that SptP and ExoS function as GTPase-activating proteins (GAPs) for the Rho subfamily of small GTPases (Fu and Galán, 1999; Goehring et al., 1999). Rho GTPases, which include Rho, Rac and Cdc42, are key regulators of actin dynamics and function as molecular switches in various signal transduction pathways (reviewed by Van Aelst and D'Souza-Schorey, 1997; Hall, 1998; Schoenwaelder and Burridge, 1999). Rho GTPases cycle between an active GTP-bound state and an inactive GDP-bound state. GAPs stimulate the intrinsic GTPase activity of small G proteins and facilitate the conversion of the GTP-bound form of the protein to the GDP-bound form. GAPs share a common motif, which includes an invariant arginine residue. This motif is called an ‘arginine finger’ and is essential for efficient catalysis (Scheffzek et al., 1998). Sequence alignment of YopE with SptP and ExoS revealed the presence of a conserved arginine finger motif, and mutation of the invariant arginine within this motif eliminated the GAP activity of SptP and ExoS (Fu and Galán, 1999; Goehring et al., 1999). These results prompted us to examine whether YopE also displays GAP activity. In this report, we show that YopE is a GAP for Rho family GTPases and that GAP activity is essential for the known biological effects of YopE in cultured cells infected with Y. pseudotuberculosis. Furthermore, we show that YopE GAP activity is important for Yersinia virulence in a mouse infection model. In addition, we demonstrate that constitutively active Rho GTPases can block specific YopE functions.
YopE is a GTPase-activating protein for Rho GTPases
The ability of YopE to function as a RhoGAP was tested using a filter-binding assay (see Experimental procedures). Purified GST–YopE was incubated with recombinant GST–RhoA, GST–Rac1 or GST–Cdc42 that had been loaded with [γ-32P]-GTP. Aliquots of the reactions were taken at various times and filtered onto nitrocellulose membranes. The amount of radioactive nucleotide remaining bound to the GTPase was determined by scintillation counting. As shown in Fig. 1, GST–YopE significantly stimulated the intrinsic GTPase activity of GST–RhoA (Fig. 1A), GST–Rac1 (Fig. 1B) and GST–Cdc42 (Fig. 1C). However, even at a 10-fold higher concentration, GST–YopE had no measurable effect on the GTPase activity of GST–Ras (Fig. 1A). These data indicate that YopE is a GAP for the Rho family of GTPases.
Arginine residue 144 in YopE is essential for GAP activity
To determine whether the arginine finger motif in YopE is important for GAP activity, the corresponding arginine residue in YopE, arginine 144, was changed to an alanine residue, and the effect of this mutation (R144A) on the GAP activity of GST–YopE was examined. As shown in Fig. 1D, this mutation eliminated the ability of GST–YopE to stimulate GTP hydrolysis by GST–Rac1. However, this mutation did not appear to cause a major conformational change in YopE, as YopER144A was still capable of binding to Rho GTPases in vitro (data not shown). In addition, YopER144A was translocated into HeLa cells at wild-type levels (see below). These results are consistent with the idea that YopE functions in a manner similar to that of other GAP proteins.
GAP activity is required for YopE to disrupt actin filaments
We next investigated whether the GAP activity was required for the disruption of actin filaments and cell rounding by YopE. Hela cells were infected with YP17, a Y. pseudotuberculosis yopEyopH mutant, containing an empty expression vector (pVec), a vector expressing wild-type yopE (pYopE) or a vector expressing catalytically inactive yopER144A (pYopER144A). A yopH null background was used in these studies in order to focus solely on the effects of YopE. The parent of this strain, YPIII, does not contain an intact yopT gene (Persson et al., 1999; J. Zitzler, unpublished observation) and therefore does not produce YopT. Infections were carried out at a multiplicity of infection (MOI) of 50:1 for 2 h. The infected cells were then fixed and stained with rhodamine–phalloidin to detect F-actin and examined by immunofluorescence microscopy. Cells infected with YP17/pVec contained numerous actin filaments (Fig. 2A), whereas cells infected with YP17/pYopE rounded up and lost actin filaments (Fig. 2B). Loss of actin filaments was evident as early as 45 min after infection (data not shown). Cells infected with YP17/pYopER144A did not round up and contained numerous actin filaments (Fig. 2C), suggesting that GAP function is required for YopE to trigger morphological changes.
YopE and YopER144A are translocated into HeLa cells at similar levels
To conclude that the GAP activity of YopE was required for actin filament disruption and cell rounding, it was important to show that YopER144A was translocated into HeLa cells at wild-type levels. Several assays were performed to address this issue. First, a detergent solubility assay was conducted. HeLa cells were infected for 2 h with one of several different Y. pseudotuberculosis strains and then exposed to detergent conditions that selectively solubilize HeLa cell membranes but not bacterial membranes (see Experimental procedures). The resulting lysates were centrifuged to separate the soluble and insoluble cell fractions. Samples of each fraction were analysed by immunoblotting with a polyclonal antibody to YopE to determine the amounts of translocated Yop protein (soluble fraction) versus non-translocated Yop protein (insoluble fraction). Approximately equal amounts of YopE were detected in the fractions derived from cells infected with YP17/pYopE (Fig. 3, lane 4). More YopER144A was detected in the soluble fraction than in the insoluble fraction of cells infected with YP17/pYopER144A, suggesting that the mutant Yop protein might be translocated at slightly higher levels than the wild type (Fig. 3, lane 5). As a control, HeLa cells were also infected with YP19/pYopE, a yopEyopHyopB mutant harbouring pYopE (lane 2). YP19 does not produce a functional YopB protein and is defective for translocation of Yops (Boland et al., 1996; Håkansson et al., 1996; Palmer et al., 1999). As shown in Fig. 3, lane 2, YopE was not detected in the soluble fraction when cells were infected with the yopB mutant. In addition, YopE was not detected in the soluble fraction when mock infections were performed in the absence of HeLa cells (data not shown). These results indicate that YopER144A is translocated at levels similar to or higher than wild-type YopE.
In order to confirm these results and to determine the intracellular localization of YopER-144A, an immunofluorescence microscopy assay was performed. Infected HeLa cells were processed for indirect immunofluorescence microscopy using a rabbit polyclonal anti-YopE antibody followed by fluorescein isothiocyanate (FITC)-conjugated goat anti-rabbit IgG to detect YopE. The cells were also stained with rhodamine–phalloidin to detect F-actin. In HeLa cells infected with a wild-type strain (YP126) or YP17/pYopE, wild-type YopE protein was detected in the cytoplasmic compartments of rounded cells (Fig. 4A and E). In cells infected with YP17/pYopER144A, the YopER144A protein was localized to the HeLa cytoplasm but, in this case, the cells remained spread (Fig. 4G). Some labelling of bacteria was also evident under these conditions (Fig. 4G). Labelling of F-actin with rhodamine–phalloidin demonstrated that the cells containing wild-type YopE lacked actin filaments, whereas the cells containing YopER144A contained actin filaments (Fig. 4B, F and H). Therefore, we conclude that GAP activity is required for actin filament disruption and cell rounding in HeLa cells by YopE.
Inhibition of phagocytosis by YopE requires GAP activity
A double-label immunofluorescence assay (see Experimental procedures) was used to determine whether GAP activity was required for YopE to inhibit integrin-mediated uptake of Y. pseudotuberculosis into HeLa cells. The yopEyopH mutant YP17/pVec was internalized ninefold more efficiently than the yopH mutant YP15/pVec (Fig. 5). YP17/pYopE was taken up at levels similar to YP15/pVec. On the other hand, YP17/pYopER144A was taken up at levels similar to YP17/pVec. Thus, the GAP function of YopE is required for its antiphagocytic activity.
The GAP activity of YopE is important for Y. pseudotuberculosis virulence
Y. pseudotuberculosis yopE mutants are rapidly cleared from mouse Peyer's patches and fail to colonize the spleen after oral infection (Holmström et al., 1995); thus, these mutants are considered avirulent (Forsberg and Wolf-Watz, 1988). To determine whether the GAP activity of YopE is important for Yersinia virulence, we orally infected mice with a wild-type Y. pseudotuberculosis strain (IP2666/pIB1), a yopE mutant (IP6/pVec), the yopE mutant carrying pYopE (IP6/pYopE) or the yopE mutant harbouring pYopER144A (IP6/pYopER144A). Mice were sacrificed 4 days after infection, and the colony-forming units (cfu) per spleen were determined (see Experimental procedures) (Fig. 6). Spleens from mice infected with wild-type Y. pseudotuberculosis had a mean of 4.5 logs of bacteria per organ, whereas the spleens from mice infected with the yopE mutant had a mean bacterial load that was below the limits of detection. Introduction of pYopE into the yopE mutant restored virulence to some degree, although complementation was not complete. This may result from the lower levels of expression of yopE from pYopE versus the native gene on the virulence plasmid (see Fig. 3). Although pYopE partially complemented the yopE mutant for virulence, pYopER144A did not. The mean number of bacteria per spleen in mice infected with the yopE mutant harbouring pYopE was at least three logs greater than that seen in infections with the yopE mutant carrying pYopER144A. These results indicated that the GAP activity of YopE is important for Yersinia pathogenicity.
Constitutively active RhoA protects HeLa cells from YopE-mediated cell rounding and disruption of actin filaments
The preceding results indicated that the RhoGAP activity of YopE is essential for YopE function, both in vivo and in vitro. Because GAPs function to downregulate GTPases by stimulating the conversion of the active GTP-bound form of the GTPase to the inactive GDP-bound form (Scheffzek et al., 1998), it was reasoned that constitutively active Rho GTPases may be able to counteract YopE-mediated events. To test this hypothesis, HeLa cells were transfected with vectors producing constitutively active mutants of RhoA and Rac1 that are insensitive to GAP downregulation (RhoA-V14 and Rac1-V12 respectively). First, we examined the effects of expressing Rho-V14 on actin filament disruption and cell rounding mediated by YopE. RhoA activity is important for focal adhesion and stress fibre formation in cells (Hall, 1998); therefore, it was predicted that RhoA-V14 would block the ability of YopE to cause cell rounding and actin filament disruption. HeLa cells were transfected with either an empty vector (pEFBosHA) or a vector that produces HA epitope-tagged RhoA-V14 (pEFBosHARhoA-V14). Twenty-four hours later, the cells were infected with YP17/pVec or YP17/pYopE. The samples were processed for immunofluorescence microscopy using a monoclonal anti-HA antibody, followed by FITC-conjugated secondary antibody to identify transfected cells. Cells were also co-stained with a polyclonal anti-YopE antibody, followed by a TRITC-conjugated secondary antibody to detect translocated YopE. Examination of the samples by phase-contrast and epifluorescence microscopy indicated that infection of non-transfected cells with YP17/pYopE resulted in cell rounding, as expected (Fig. 7A). However, RhoA-V14-transfected cells infected with YP17/pYopE remained spread out (compare Fig. 7G and H). YopE was detected in RhoA-V14-transfected cells, indicating that transfection per se did not prevent translocation (compare Fig. 7H and I). To confirm that RhoA-V14 prevented YopE-mediated disruption of actin filaments, YP17/pYopE-infected cells were stained with rhodamine–phalloidin (Fig. 8). Actin filaments were detected in transfected cells, whereas no actin filaments were observed in non-transfected cells (compare Fig. 8A and B).
Constitutively active Rac1 counteracts the antiphagocytic activity of YopE in HeLa cells
The effect of constitutively active RhoA and Rac1 proteins on the antiphagocytic activity of YopE in HeLa cells was also examined. The invasin protein mediates efficient uptake of Y. pseudotuberculosis into epithelial cells. Invasin binds to β1-integrins on the surface of the cells to promote bacterial entry (Isberg, 1989; Isberg and Tran Van Nhieu, 1994). Although Rac1 has been shown to be required for Fc receptor-mediated phagocytosis (Caron and Hall, 1998; Massol et al., 1998), RhoA has been suggested to promote β1-integrin-dependent uptake of Y. pseudotuberculosis into epithelial cells (Mecsas et al., 1998). A triple-label immunofluorescence assay enabled the identification of transfected cells and differentiation between intracellular and extracellular cell-associated bacteria (see Experimental procedures). Transfection of HeLa cells with a vector that produces Rac1-V12 (pCGTRac1-V12) increased cellular uptake of YP17/pYopE about ninefold over that seen in cells transfected with the empty vector pCGT (Fig. 9A). Bacteria expressing YopE were internalized as readily as bacteria deficient for the expression of YopE. However, transfection of HeLa cells with pEFBosRhoA-V14 did not counteract the inhibitory effect of YopE on uptake (Fig. 9A). Even though transfection of cells with RhoA-V14 resulted in an overall decrease in bacterial uptake, uptake of YP17/pYopE was ninefold less efficient than uptake of YP17/pVec (Fig. 9A). A similar absolute difference in uptake was observed in cells transfected with the empty vector pEFBosHA (Fig. 9A). The numbers of total bacteria (inside and outside) associated with transfected HeLa cells are presented in Fig. 9B. On average, there were similar numbers of bacteria associated with each type of transfected cell. Thus, it appears that the production of Rac1-V12 did not lead to an artificial increase in bacterial uptake. These results support the idea that downregulation of RhoA by YopE is important for disruption of actin filaments and cell rounding, whereas downregulation of Rac1, but not RhoA, by YopE is important for inhibition of β1-integrin-mediated bacterial uptake.
We show here that YopE functions as a GAP for the Rho subfamily of small GTPases and that GAP activity is essential for the known biological activities of this effector protein.
A GST–YopE fusion protein stimulated the intrinsic GTPase activity of RhoA, Rac1 and Cdc42, but had no effect on Ras GTP hydrolysis. Based on sequence alignment with other GAPs, including SptP and ExoS, YopE was predicted to contain an arginine finger motif (Fu and Galán, 1999). Recent crystal structure analysis has demonstrated that Rho and Ras GTPase-activating proteins act by inserting an arginine finger into the GTPase to stabilize the intermediate state of GTP hydrolysis (Rittinger et al., 1997; Scheffzek et al., 1997). The arginine finger contains an invariant arginine residue that is essential for efficient catalysis (Scheffzek et al., 1998). Arginine 144 of YopE was predicted to be this invariant residue (Fu and Galán, 1999). Mutation of arginine 144 to alanine (YopER144A) effectively eliminated the GAP activity of YopE, confirming the importance of this residue in catalysis and supporting the idea that YopE contains an arginine finger motif. Our results suggest that YopE acts in a manner that is mechanistically similar to other GAP proteins.
The GAP activity of YopE was shown to be necessary for YopE function. Y. pseudotuberculosis expressing YopER144A did not disrupt actin filaments or induce cell rounding in HeLa cells. The R144A mutation also eliminated the ability of YopE to inhibit β1-integrin-dependent uptake of Y. pseudotuberculosis into HeLa cells. The effect of this mutation on YopE function was not a result of protein instability or misfolding. A detergent fractionation assay and an immunofluorescence microscopy assay confirmed that YopER144A was translocated into HeLa cells at levels similar to that of the wild-type protein. In addition, YopER144A retained the ability to bind Rho GTPases in vitro (data not shown). We also showed that YopE GAP activity is important for Yersinia virulence. Y. pseudotuberculosis expressing YopER144A failed to colonize the spleens of orally challenged mice. Thus, YopE GAP activity is required not only for the function of YopE as examined in vitro, but also for YopE function during an animal infection.
GAPs function to downregulate GTPases by stimulating the conversion of the active GTP-bound form of the GTPase to the inactive GDP-bound form (Scheffzek et al., 1998). If the biological effects of YopE result from the downregulation of Rho GTPases, then cells should resist YopE effects if they contain constitutively active Rho GTPases that are insensitive to YopE GAP activity. Indeed, we found that HeLa cells producing the constitutively active RhoA-V14 protein were protected from YopE-mediated disruption of actin filaments and cell rounding. Also, production of Rac1-V12 in HeLa cells blocked the antiphagocytic effect of YopE. The finding that Rac1-V12, but not RhoA-V14, could overcome the inhibitory effect of YopE on bacterial internalization mediated by β1-integrins was surprising. There is significant evidence that RhoA function is important for signalling through β1-integrins (Schoenwaelder and Burridge, 1999). In addition, RhoA activity is required for phagocytosis mediated by β2-integrin (Caron and Hall, 1998). However, recent studies suggest that Rac1 but not RhoA is necessary for invasin-mediated uptake of Y. pseudotuberculosis into host cells (M. A. Alrutz, unpublished observation), and our results support this conclusion.
The Rho subfamily of GTPases are key regulators of actin cytoskeleton dynamics (reviewed by Hall, 1998; Schoenwaelder and Burridge, 1999). In addition, it is becoming clear that these GTPases play important roles in signalling pathways that control gene transcription, cell proliferation and membrane trafficking (Van Aelst and D'Souza-Schorey, 1997). Thus, it is not surprising that a variety of bacterial toxins and virulence factors target these GTPases (Aktories, 1997; Boquet, 1999). Here, we show that the Yersinia virulence factor YopE acts as a RhoGAP to downregulate the GTPase activity of Cdc42, RhoA and Rac1. The finding that YopE targets all three GTPases, at least in vitro, may indicate that each of these Rho GTPases plays a key role in regulating actin cytoskeletal rearrangements and GTPase-mediated signalling events during Yersinia infection.
Cell culture and transfections
HeLa cells were cultured routinely in DMEM (Gibco BRL) supplemented with 10% heat-inactivated fetal calf serum (FCS; Gibco BRL) and 1 mM sodium pyruvate in a 5% CO2 humidified incubator at 37°C.
For transfections, HeLa cells (3 × 104) were seeded onto sterile glass coverslips placed in a 24-well tissue culture plate. Approximately 24 h later, cells were transiently transfected with 0.25 µg of the indicated plasmid using Fugene 6 (Boehringer Mannheim) as specified by the manufacturer. Infections with Y. pseudotuberculosis were performed 24 h after transfection.
Y. pseudotuberculosis strains and infections
The YPIII-derived Y. pseudotuberculosis strains YP126 (wild type), YP15 (yopH), YP17 (yopEyopH) and YP19 (yopByopEyopH) have been described previously (Black and Bliska, 1997; Palmer et al., 1999). IP6 is a yopE mutant derived from the wild-type Y. pseudotuberculosis strain IP2666(pIB1) (Bliska et al., 1991). Inactivation of yopE was accomplished by the insertion of a kanamycin resistance gene cassette into the unique Pst1 restriction site of the yopE gene (Forsberg and Wolf-Watz, 1988). Expression plasmids pYopE, pYopER144A and pMMB67EH (pVec) were introduced into Y. pseudotuberculosis by conjugation as described previously (Bliska and Black, 1995)
For infection assays, bacteria were grown overnight at 26°C with shaking in Luria–Bertani (LB) broth. Media was supplemented with 100 µg ml−1 ampicillin (LBAmp) for plasmid-bearing strains. The next day, bacteria were subcultured into fresh media supplemented with 2.5 mM CaCl2 to an OD600 of 0.1. Cultures were shaken at 37°C for 2 h. Bacteria were centrifuged and resuspended in warm Hanks' balanced salt solution (HBSS) to an OD600 of 1.0 (≈1 × 109 cfu ml−1). Unless otherwise specified, infections were performed by adding bacteria (MOI = 50:1) to the cells in tissue culture media and then incubating the plates for 2 h at 37°C in a 5% CO2 incubator.
The yopE open reading frame (ORF) was amplified from Y. pseudotuberculosis using polymerase chain reaction (PCR) and the primers yopE3 (5′-CGGATCCCATATGAAAATATCATCATTTATTTC-3′) and yopE2 (5′-GGGATCCCCATATCACATCAATGACA-3′). The PCR product was digested with BamHI and inserted into the BamHI site of pGEX2T (Pharmacia) to produce pGEX2T-YopE. Sequencing revealed that nucleotide 345 differed from the published sequence, but this was a conservative change that did not alter codon usage. pGEX2T-YopER144A was constructed by PCR using the primers yopE3 and yopER144A (5′-CATCAGCCCTTGGCATTGAGTGATACTGCCAGCAAGAGG-3′). The PCR product was digested with DraIII and StyI, and the ≈ 250 bp fragment containing the yopER-144A mutation was isolated and used to replace the corresponding fragment in pGEX2T-YopE. The presence of the R-144A mutation was confirmed by sequencing. The plasmids pYopE and pYopER144A were constructed by digesting pYopH, a derivative of pMMB67EH (Bliska and Black, 1995), with NdeI and EcoRI to remove the yopH coding region. Into this vector, an NdeI–EcoRI restriction fragment from pGEX2T-YopE or pGEX2T-YopER144A was inserted to place yopE or yopER144A under the control of the tac promoter and the LacIq repressor. The level of YopE protein produced from these plasmids in the absence of inducer is sufficient to complement a yopE mutant (e.g. see Fig. 5). Protein expression from these plasmids is increased after induction with IPTG.
pGEX4T3-Rac1 and pGEX4T3-Cdc42 were obtained from Dr X. Bustelo (SUNY, Stony Brook, NY, USA). pEFBosHA, pEFBosHARhoA-V14 and pGEX2T-RhoA were obtained from Dr M. Frohman (SUNY, Stony Brook, NY, USA). pCGT and pCGT-Rac1-V12 were obtained from Dr D. Bar-Sagi (SUNY, Stony Brook, NY, USA).
The rabbit anti-GST-YopE antiserum was provided by Dr J. Clemens (University of Michigan, USA). The polyclonal anti-YopE antibody was affinity purified from this serum using standard techniques (Harlow and Lane, 1988). SB349 is a commercially produced (Cocalico Biologicals) polyclonal anti-Yersinia antibody obtained by injecting rabbits with heat-killed Y. pseudotuberculosis. Mouse monoclonal anti-T7 tag antibody was purchased from Novagen. Rat monoclonal anti-HA (clone 3F10) antibody was purchased from Boehringer Mannheim. Anti-rabbit IgG conjugated to horseradish peroxidase (HRP) was purchased from Sigma. FITC-conjugated goat anti-rabbit IgG, lissamine–rhodamine (LRSC)-conjugated goat anti-rabbit IgG, FITC-conjugated goat anti-rat IgG and FITC-conjugated F(ab′)2 goat anti-mouse IgG were purchased from Jackson ImmunoResearch Laboratories. Alexa Fluor 350-conjugated goat anti-rabbit IgG was purchased from Molecular Probes.
Purification of GST fusion proteins
Overnight cultures of Escherichia coli containing pGEX2T-YopE were diluted 50-fold into 500 ml of fresh LBAmp and grown at 37°C to an OD600 of approximately 0.5. IPTG was then added to a final concentration of 0.1 mM, and cultures were incubated for an additional 2 h to induce protein expression. Bacteria were harvested by centrifugation, and pellets were quick frozen in liquid nitrogen and stored at −70°C until needed. Bacterial pellets were thawed and resuspended in 10 ml of TBS/NP40 buffer [50 mM Tris, pH 8.0, 1% NP40, 5 mM EDTA, 1 mM dithiothreitol (DTT), 1 mM phenylmethylsulphonyl fluoride (PMSF)]. Cells were lysed in a French press (1200 psi), and the lysates were clarified by centrifugation at 10 000 g for 15 min at 4°C. GST–YopE was affinity purified from cleared lysates by incubation with a 1 ml packed volume of washed glutathione-SepharoseB beads (Pharmacia) for 2–3 h at 4°C with rotation. Beads were washed five times with 10 ml of TBS/NP40 buffer and then five times with 10 ml of TBS buffer. GST–YopE was eluted from the beads with 50 mM Tris, pH 8.0, containing 10 mM reduced glutathione. The eluted protein was dialysed at 4°C against 2 l of 50 mM Tris, pH 8.0, for at least 3 h with three buffer changes. Rho GTPase fusion proteins were purified in the same fashion with the following modifications. All buffers used during the purification were modified by the addition of 2 mM MgCl2, and EDTA was omitted. Dialysis buffer contained 10 mM Tris, pH 7.6, 2 mM MgCl2 and 0.1 mM DTT. The GST–Ras protein was provided by B. Hall (SUNY, Stony Brook, NY, USA). Protein concentrations were determined using the Bio-Rad protein assay, and purity was examined by SDS–PAGE. Proteins were aliquoted, quick frozen in liquid N2 and stored at −70°C.
GAP assays were performed essentially as described previously (Self and Hall, 1995). GST–Cdc42, GST–Rac1 or GST–RhoA (0.9 µg) were preloaded with 10 µCi of [γ-32P]-GTP (6000 Ci mmol−1, 10 µCi µl−1; NEN Life Science Products) in 20 µl of 20 mM Tris, pH 7.6, 5 mM EDTA, 1 mM DTT for 15 min at room temperature. Samples were placed on ice, and 1 µl of 0.4 M MgCl2 (19 mM final concentration) was added to stop the reaction. Three microlitres (0.13 µg) of the preloaded protein (to give a final concentration of 90 nM) was diluted in buffer (20 mM Tris, pH 7.6, 0.1 mM DTT, 1 mM GTP, 1 mg ml−1 BSA) to give a final volume of 30 µl. A 5 µl sample was removed (time 0) and diluted into 1 ml of ice-cold assay buffer (50 mM Tris, pH 7.6, 50 mM NaCl, 5 mM MgCl2). In some reactions, GST–YopE was added at this time. Reactions were incubated at room temperature and, at 5, 10 and 15 min thereafter, 5 µl samples were removed and diluted into 1 ml of cold assay buffer. Samples were filtered through nitrocellulose filters (Millipore) prewetted with cold assay buffer to trap the GTPase protein and its bound 32P-labelled nucleotide. Filters were washed three times with 3 ml of cold assay buffer to remove hydrolysed [32P]-Pi and then air dried. The amount of radioactivity remaining bound to the filter was determined by scintillation counting.
Detergent solubility assay
HeLa cells (2 × 106) were seeded into 100 mm tissue culture dishes on the day before the assay. Cells were infected with bacteria as described above. The dishes containing infected cells were washed three times with ice-cold PBS containing 1 mM Na3VO4 and 10 mM NaF. To each dish, 0.5 ml of cold 1% Triton X-100 buffer (10 mM Tris, pH 7.6, 150 mM NaCl, 10% glycerol, 1% Triton X-100, 1 mM Na3VO4, 10 mM NaF, 200 µM AEBSF, 20 µM leupeptin and 1 µM pepstatin) was added, and the dishes were incubated for 15 min on ice with occasional rocking. Cells were scraped into microcentrifuge tubes and centrifuged in a Sorvall microfuge for 10 min at 12 000 r.p.m. at 4°C. Supernatants (soluble fractions) were transferred to new tubes, and protein concentrations were determined using the Bio-Rad protein assay. Pellets (insoluble fractions) were carefully washed in lysis buffer, resuspended in 100 µl of Laemmli sample buffer and boiled. Samples of the fractions containing equivalent amounts of protein (or sample volume for insoluble fractions) were separated by SDS–PAGE under reducing conditions and transferred electrophoretically onto nitrocellulose (Schleicher and Schuell). Immunoblotting was performed using a polyclonal anti-YopE antibody diluted 1:2000 as described previously (Black and Bliska, 1997).
Mouse infection assay
Six- to 8-week-old female Balb/c mice were obtained from Taconic Farms. Bacteria were grown overnight in brain–heart infusion (BHI) broth at 26°C (Difco). The BHI was supplemented with 100 µg ml−1 ampicillin for the strains containing pVec, pYopE or pYopER144A. Bacteria were pelleted by centrifugation and resuspended to 1.8 × 1011 cfu ml−1 in 10% NaHCO3. For each strain tested, five mice were fed 25 µl of the bacterial suspension (4.5 × 109). On day 4 after infection, mice were sacrificed by CO2 inhalation, and spleens were removed and homogenized in 4.5 ml of PBS. Serial 10-fold dilutions were plated on LB agar plates and incubated at room temperature for 2 days to determine cfu per organ. To assess the stability of pVec, pYopE and pYopER144A plasmids during infection, bacterial colonies were picked from LB plates and patched onto LBAmp plates. Plasmid stability was 100% for the recovered colonies. The procedures were approved by the Institutional Animal Care and Use Committee (project 99–0894) at SUNY Stony Brook.
Bacterial uptake assay
HeLa cells were seeded onto glass coverslips at 1 × 105 cells per well in a 24-well tissue culture plate approximately 20 h before infection. Transient transfections were carried out as described above. Cells were infected with bacteria at a calculated MOI of 25:1. After a brief centrifugation step (5 min at 100 g), the plates were incubated for 30 min at 37°C in a 5% CO2 incubator. A double-label immunofluorescence assay (Heesemann and Laufs, 1985) was used to differentiate between extracellular and intracellular cell-associated bacteria. Coverslips containing infected cells were washed with PBS and fixed in 2% paraformaldehyde for 15 min. The coverslips were washed and then incubated with the polyclonal anti-Yersinia antibody SB349 (diluted 1:1000) for 40 min to stain extracellular bacteria. Coverslips were washed and incubated for 40 min with Alexa Fluor 350-conjugated goat anti-rabbit IgG diluted 1:250. After washing, cells were permeabilized with 0.2% Triton X-100 for 10 min. Coverslips were washed and incubated with SB349 (1:1000) for 40 min to label both extracellular and intracellular bacteria. Samples were then washed and incubated for 40 min with LRSC-conjugated goat anti-rabbit IgG (1:300). For the identification of transiently transfected cells, the permeabilized cells were incubated with SB349 and anti-HA (0.2 µg ml−1) or anti-T7 (0.25 µg ml−1) antibodies for 40 min. The anti-HA and anti-T7 antibodies were detected with FITC-conjugated goat anti-rat IgG (1:250) or FITC-conjugated F(ab′)2 goat anti-mouse IgG (1:250) respectively. All antibodies were diluted in PBS containing 3% BSA, and washes were conducted three times for 5 min with PBS containing 1% BSA. Coverslips were washed well with PBS before mounting and examined by immunofluorescence microscopy as described below. Transfected cells were identified, and the percentage uptake was calculated as the number of [total bacteria (LRSC)–extracellular bacteria (Alexa-Fluor)/total bacteria (LRSC)] × 100.
HeLa cells (1 × 105) were seeded in 1 ml of media onto glass coverslips placed in a 24-well tissue culture plate approximately 20 h before the assay. Transient transfections were carried out as described above. Cells were infected as described above, with the exception that 100 µM IPTG was added to the tissue culture medium during infections to increase protein expression from pYopE and pYopER-144A. All steps were performed at room temperature. Coverslips were washed twice with PBS, fixed with 4% paraformaldehyde for 10 min and then permeabilized with 0.2% Triton X-100 for 10 min. Coverslips were washed twice with PBS containing 1% BSA and then incubated for 1 h with either anti-YopE (0.6 µg ml−1), anti-HA (0.2 µg ml−1) or anti-T7 (0.3 µg ml−1) antibody. Antibodies were diluted in PBS containing 3% BSA. Coverslips were washed and incubated for 1 h with FITC-conjugated goat anti-rat IgG (1:250), FITC-conjugated F(ab′)2 goat anti-mouse IgG (1:250), or lissamine–rhodamine (LRSC)-conjugated goat anti-rabbit IgG (1:300). Coverslips were washed well with PBS before mounting in 10% Airvol (Air Products) in 100 mM Tris, pH 8.5, 25% glycerol and 2.5% DABCO (Sigma) to reduce fading. In some experiments, rhodamine–phalloidin (final concentration 0.2 U ml−1; Molecular Probes) was included during the incubation step with the secondary antibodies to label F-actin. Samples were analysed by epifluorescence microscopy using a Zeiss AxioPlan 2 microscope equipped with a 100× (NA 1.3) PlanNeoFluar objective oil immersion lens. Images were captured with a Diagnostic Instruments Spot camera and processed using Adobe Photoshop 5.5.
We thank Xose Bustelo, Jim Clemens, Mike Frohman, Dafna Bar-Sagi and Brian Hall for generously supplying reagents. We gratefully acknowledge Xose Bustelo and Dafna Bar-Sagi for providing scientific advice, and the members of our laboratory for editing the manuscript. This research was funded by grants from the National Institute of Health (AI35175) and (AI43389) to J.B.B. D.S.B. was supported in part by a postdoctoral fellowship (CO15699) from the New York State Department of Health.
Note added in proof
It has recently been demonstrated by von Pawel-Rammingen et al. that YopE exhibits GAP activity for Rho GTPases (Mol Microbiol 2000, 36: 737–748).