Department of Pathology and Laboratory Medicine, Medical University of South Carolina, 165 Ashley Avenue, Suite 309, PO Box 250908, Charleston, SC 29425, USA
Correspondence
Joan C. Olson
olsonj{at}musc.edu
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The complex effects of ExoS on eukaryotic cell function are believed to relate to its bifunctional mode of action. ExoS includes a GTPase-activating function within its amino terminus (Goehring et al., 1999) and an ADPRT activity within its carboxy terminus (Knight et al., 1996
). The GTPase-activating protein activity of ExoS has been linked to cytoskeletal alterations (Pederson et al., 1999
), while the ADPRT activity of ExoS is required for the effects of bacterially translocated ExoS on human epithelial cell DNA synthesis, re-adherence and long-term cytoskeletal alterations (Fraylick et al., 2001
). When internalized by bacterial TTS, ExoS ADPRT activity targets multiple cellular LMMG proteins (McGuffie et al., 1998
; Fraylick et al., 2002a
, b
) and requires a eukaryotic cofactor, 14-3-3, for catalytic activity (Coburn et al., 1991
; Fu et al., 1993
; Masters et al., 1999
; Henriksson et al., 2000b
). ADP-ribosylation of cellular targets, Ras and RalA, has been found to interfere with their respective signal transduction pathways (Vincent et al., 1999
; Ganesan et al., 1999
; Henriksson et al., 2000a
; Fraylick et al., 2002b
), thereby linking a cellular mechanism to the effects of ExoS on eukaryotic cell function. Recent studies have drawn attention to possible modulatory influences the eukaryotic cell can have on ExoS function. Notably, ExoS ADPRT activity appears to be enhanced within the membrane compartment of epithelial cells when internalized by TTS (Fraylick et al., 2001
; Riese & Barbieri, 2002
). In addition, an E381A mutant form of ExoS, which is unable to ADP-ribosylate Ras in vitro, was found to modify Ras when translocated into cells by the TTS process (Fraylick et al., 2001
).
The examination of multiple human epithelial cell lines has identified differences in cellular sensitivities to bacterially translocated ExoS. Normal, confluent, polarized epithelial monolayers were found to be highly resistant to the effects of bacterially translocated ExoS (Fleiszig et al., 1997b; McGuffie et al., 1999
). However, when normal epithelial cell monolayers were disrupted or subconfluent, cells became sensitive to ExoS (McGuffie et al., 1999
). In examining the effects of ExoS-producing bacteria on epithelial cell lines derived from different tissues, it was noted that tumour-derived cell lines were highly sensitive to the effects of ExoS on cell growth (McGuffie et al., 1999
). The detection of cellular differences in sensitivity to bacterially translocated ExoS, along with the finding that alterations in cellular physiology can affect sensitivity to ExoS, support the notion that eukaryotic cell factors can influence the toxicity of ExoS.
To clarify how the eukaryotic cell might contribute to the toxicity of bacterially translocated ExoS, we examined how different cell lines compared in their responses to ExoS-producing bacteria. Our studies found murine fibroblasts to be more resistant to the effects of ExoS than human epithelial cells. The increased resistance of fibroblastic cells to ExoS was associated with less efficient ADP-ribosylation of cellular substrates by ExoS and occurred independently of detectable differences in the efficiency of the TTS-mediated translocation of ExoS. In examination of rodent, simian and human cell lines, distinct patterns of ExoS ADPRT substrate modification were detected, with human cell lines showing both more-efficient and more-extensive LMMG protein substrate modification. These studies support the notion that, once internalized, host cell properties can influence the substrate specificity of ExoS ADPRT activity, which corresponds with alterations in the severity of effects of ExoS on cell function.
![]() |
METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Eukaryotic cell culture.
CHO-K1, Cos-1, Detroit 532, HT-1080, HT-29, LNCaP, NIH-3T3, Swiss-3T3, TCMK-1, T24 and Vero cells used in this study were obtained from the American Type Culture Collection (ATCC; Manassas, VA, USA). RHEC cells were provided by Carwile LeRoy, and WI-38 and WI-38 SV40-transformed cells were kindly provided by Daohong Zhou, both of the Medical University of South Carolina, Charleston, SC, USA. Characteristics of cell lines used in this study are shown in Table 1. Cell lines were cultured, as described below, in their respective medium containing 10 % (v/v) fetal bovine serum, and 100 U penicillin ml-1 and 100 µg streptomycin ml-1 (PS) at 37 °C in 5 % CO2/95 % air. All cell culture media and components were purchased from Gibco-BRL.
|
Fibroblastic cell lines.
Detroit 532 cells (ATCC CCL 54) were grown in RPMI medium and maintained as described previously (Olson et al., 1997). Cos-1 cells (ATCC CRL 1650), NIH-3T3 cells (ATCC CRL 1658), Swiss-3T3 cells (ATCC CCL 92), WI-38 and SV40-transformed WI-38 (WI-38t) were cultured in DMEM and split 1 : 6 and passaged twice a week.
Co-culture of eukaryotic cells with bacteria.
For bacterial/eukaryotic cell co-culture studies, eukaryotic cells were detached with 0·05 % (w/v) trypsin/0·53 mM EDTA (trypsin-EDTA), seeded at 1x105 cells ml-1 in 6-, 24- or 48-well plates or culture dishes (Costar) and cultured for either 24 or 48 h (depending on the growth rate of the cell line) to 6080 % confluency, prior to co-culture with bacteria. Strain 388 or 388S bacteria were diluted, based on culture OD590, to 1x107 c.f.u. ml-1 in medium specific for each eukaryotic cell line and containing 0·6 % (w/v) BSA. An m.o.i. range of 10100 was used in these studies, unless otherwise indicated, and bacteria were co-cultured with eukaryotic cells for 26 h. Control cells received the respective medium with no bacteria. Effects of bacteria on eukaryotic cell function were analysed as described below.
(i) Quantification of DNA synthesis.
Following a 4 h co-culture period, bacteria were removed and cells were washed with culture medium containing 10 % fetal bovine serum, 200 µg gentamicin ml-1 and 100 µg ciprofloxacin ml-1 (GC medium) to inhibit further bacterial growth. Cells were pulsed for 20 h in GC medium containing 1 µCi [methyl-3H]thymidine ml-1 (25 Ci mmol-1; Amersham). DNA synthesis was quantified as described previously (Olson et al., 1997) and is reported as the percentage of [3H]thymidine incorporated relative to that of non-bacterial treated control cells.
(ii) Analysis of cell viability.
Cell viability was examined following a 3, 4·5 and 6 h co-culture period, after which bacteria were removed, and total cell viability was determined by trypan blue exclusion immediately after removal of bacteria and following an extended 20 h culture period in GC medium.
(iii) Examination of cell morphology.
To compare long-term effects of ExoS on cellular morphology, cells were co-cultured with bacteria for 4 h, bacteria were removed, then cells were washed with PBS and cultured in GC medium for an additional 20 h. Morphological alterations were examined by phase-contrast microscopy.
(iv) Analysis of eukaryotic cell adherence.
The effects of ExoS-producing bacteria on eukaryotic cell adherence to matrix were assessed after a 26 h co-culture period by quantifying adherent and non-adherent cells, using trypan blue exclusion, 20 or 40 h after bacteria were removed. To assess the effects of ExoS on re-adherence, cells were detached with trypsin-EDTA following co-culture with bacteria, washed, resuspended in GC medium, replated and allowed to re-adhere for 20 h (HT-29, NIH-3T3, Swiss-3T3 cells) or 48 h (LNCaP cells). (LNCaP cells were previously recognized to require a longer adherence time (Horoszewicz et al., 1980)). The number and viability of non-adherent and trypsinized, adherent cells were determined after this time based on trypan blue exclusion. The percentage of viable and non-viable, adherent or non-adherent cells was calculated relative to total cell number.
Bacterial association with eukaryotic cells.
To compare the adherence of bacteria to different eukaryotic cell lines, cells were seeded in 24-well plates, as described above, and co-cultured with bacteria at an m.o.i. of approximately 200 : 1 for 4·5 h (HT-29, NIH-3T3 and Swiss-3T3 cells) or 3 h (LNCaP cells). (The limited time of exposure of LNCaP cells related to their increased sensitivity to bacterial effects on cell viability.) Bacteria were removed and 1 ml HEPES buffered saline (HBS: 137 mM NaCl; 4 mM KCl; 10 mM HEPES, pH 7·4; 11 mM glucose) was added to wells. Cells were detached by scraping, transferred to microcentrifuge tubes and washed three times with HBS, centrifuging cells at 400 g for 4 min at 4 °C between washes. Eukaryotic cells were lysed in PBS containing 0·25 % (v/v) Triton X-100 for 30 min on ice and lysates were diluted 1 : 1000 and plated on Luria broth (LB) agar plates. Bacterial colonies were counted 16 h later and the number of bacteria associated per eukaryotic cell was determined. Data were analysed for statistical significance using one-way factorial ANOVA (http://members.aol.com/johnp71/anova1sm.html).
ADP-ribosylation of cellular substrates by bacterially translocated ExoS
(i) In vivo ADP-ribosylation of LMMG protein substrates.
In vivo ADP-ribosylation of LMMG protein substrates by bacterially translocated ExoS was assessed based on altered protein mobility by SDS-PAGE following co-culture with bacteria for 4 or 6 h, as described previously (McGuffie et al., 1998; Fraylick et al., 2001
, 2002a
). To examine Ras modification, cells were lysed in TBS-TDS (10 mM Tris, pH 7·4; 140 mM NaCl; 1 %, v/v, Triton X-100; 0·5 %, w/v, sodium deoxycholate; 0·1 %, w/v, SDS) and cellular Ras was immunoprecipitated using Y13-259 anti-pan Ras antibody, as described previously (McGuffie et al., 1998
). Precipitates were resolved by 12·5 % SDS-PAGE (Laemmli, 1970
), transferred to PVDF membranes (Millipore) (Towbin et al., 1979
) and Ras was detected using 100 ng LA045 pan-Ras antibody ml-1 or 100 ng LA069 H-Ras specific antibody ml-1 (Quality Biotech), followed by horseradish peroxidase-conjugated anti-mouse immunoglobulin G (Jackson Immuno Research) and visualized by enhanced chemiluminescence (ECL) (Amersham). RalA, Rab5, Rab7, Rab8, Rab11 and Rac1 modification was detected following lysis of cells with Laemmli electrophoresis sample buffer and lysates were resolved by 12 % SDS-PAGE and transferred to PVDF membranes, as described above. Individual LMMG proteins were detected using the following antibodies: RalA, 50 ng mouse anti-RalA mAb ml-1 (Transduction Laboratories); Rab5, 10 ng rabbit anti-Rab5B (A-20) polyclonal antibody ml-1 (Santa Cruz Biotechnology); Rab7, 0·7 µg goat anti-Rab7 (C-19) polyclonal antibody ml-1 (Santa Cruz); Rab8, 250 ng mouse anti-Rab8 mAb ml-1 (Transduction Laboratories); Rab11, 250 ng mouse anti-Rab11 mAb ml-1 (Transduction Laboratories); and Rac1, 250 ng mouse anti-Rac1 mAb ml-1 (Transduction Laboratories). Immunoblots were developed with horseradish peroxidase-conjugated anti-mouse immunoglobulin G (Sigma) or anti-rabbit immunoglobulin G (Transduction Laboratories), as appropriate, and visualized by ECL.
(ii) In vitro ADP-ribosylation of fibroblastic RalA, Rab proteins or Rac1.
In vitro ADP-ribosylation of fibroblastic RalA, Rab proteins or Rac1 was performed, as described previously (Fraylick et al., 2002a), using 0·2 µM purified ExoS (McGuffie et al., 1998
), 10 mM NAD (Sigma) and 40 µl NIH-3T3 or Swiss-3T3 fibroblast cell lysate (as a source of the endogenous proteins and 14-3-3 cofactor) in a 100 µl reaction volume, which was incubated for 30 min at room temperature. Reactions were stopped with 4x times; Laemmli sample buffer and heating at 95 °C for 5 min. Fibroblast cell lysates were prepared by incubating cells on ice for 30 min in lysis buffer containing 50 mM Tris, 0·1 % Triton X-100, 10 µg aprotinin ml-1, 10 µg leupeptin ml-1, 1 mM EDTA and 100 µM PMSF. Lysates were cleared of nuclei and unbroken cells by centrifugation at 16 000 g for 5 min prior to use in reactions. Following the ADPRT reaction, proteins were resolved by 12 % SDS-PAGE and immunoblotted for the specific proteins, as described above.
Time-course analysis of ExoS internalization.
Substrate modification and the internalization of ExoS ADPRT activity was assessed in fibroblastic and epithelial cells seeded in 100 mm dishes and co-cultured with bacteria for 3, 4·5 or 6 h. Following removal of bacteria, cell monolayers were washed twice with PBS and cells were recovered from washes by centrifugation at 400 g for 10 min. Adherent cells were detached with trypsin-EDTA and combined with cells recovered from washes. The treatment of cells with trypsin in these studies minimized the likelihood that extracellularly bound ExoS would remain in cellular fractions. The cell pellet was resuspended in 100 µl cell-permeabilization solution (0·2 %, w/v, saponin; 100 mM NaCl; 250 mM sucrose; 5 mM EDTA; 1 mM PMSF; 10 µg leupeptin ml-1; 10 µg aprotinin ml-1 (Sigma); 100 µg ciprofloxacin ml-1) and lysed on ice for 30 min. Lysates were centrifuged at 16 000 g for 20 min at 4 °C. The supernatant (cytosolic fraction) was removed and the pellet was resuspended in 100 µl cell-permeabilization solution containing 1 % (v/v) Triton X-100 and vortexed for 15 s, every 5 min, four times to solubilize membrane proteins. The solubilized pellets were centrifuged at 16 000 g for 10 min and the supernatant (Triton X-100 membrane-soluble fraction) was removed. To analyse substrate modification in the cytosolic and membrane fractions, a 50 µl aliquot of each sample was mixed with 4x times; Laemmli sample buffer and an equivalent volume of each fraction was resolved by SDS-PAGE and immunoblotted for RalA, Rab proteins, or Rac1, as described above. ExoS ADPRT activity in remaining portions of cytosolic and membrane fractions was quantified, as described previously (Ferguson et al., 2001), and related to total protein concentrations in the respective sample determined using the Sigma Diagnostics Lowry-based protein assay. Data were analysed for statistical significance using one-way factorial ANOVA, as described above.
![]() |
RESULTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
In examining effects of bacterially translocated ExoS on DNA synthesis, HT-29 and LNCaP epithelial cells were found to be more sensitive to effects of ExoS on DNA synthesis, showing a 51·6±0·5 and 88·6±1·0 % inhibition of DNA synthesis, respectively, relative to that of strain 388S. This compared to a 36·0±1·3 and 20·3±0·7 % inhibition of DNA synthesis caused by strain 388 in NIH-3T3 and Swiss-3T3 murine fibroblasts, respectively, relative to strain 388
S. In determining how effects of ExoS on cell viability related to effects on DNA synthesis, cell viability was assayed, based on trypan blue exclusion, under the same co-culture conditions and at the same time point as DNA synthesis analyses. In these studies, LNCaP and NIH-3T3 cells exhibited a greater loss of viability (95·2±0·4 and 44·6±0·9 %, respectively). This compared to a minimal 6·1±0·9 and 6·8±0·2 % loss of viability caused by bacterially translocated ExoS in HT-29 and Swiss-3T3 cells, respectively. These studies found that human epithelial cells were more sensitive than murine fibroblasts to effects of ExoS on DNA synthesis, and that effects on DNA synthesis did not necessarily correlate with effects of ExoS on cell viability.
Previous studies from our laboratory found permanent, long-term effects of strain 388 on HT-29 epithelial cell rounding to be directly associated with ExoS ADPRT activity (Olson et al., 1999; Fraylick et al., 2001
). This differed from transient effects on cell morphology caused by the non-ExoS-producing strain 388
S, which were less severe and reversible upon removal of bacteria. In comparing the effects of ExoS on the morphology of these four different cell lines, severe cell rounding persisted in HT-29 and LNCaP epithelial cells 20 h after exposure to strain 388, while long-term rounding was limited in NIH-3T3 cells and absent from Swiss-3T3 fibroblastic cells (Fig. 1
).
|
|
Comparison of bacterial association with cellular sensitivity to ExoS
Contact between bacteria and eukaryotic cells is required for the type III translocation of effector proteins (Hueck, 1998). To assess whether the differential sensitivity among the cell lines to ExoS might relate to efficiency of bacterial adherence, bacterial association with the different cell lines was quantified after a 3 or 4·5 h exposure to bacteria. Bacterial adherence to LNCaP cells was assayed after a shorter 3 h period, since this cell line begins to lose viability after a 4 h co-culture period (refer to previous section). As shown in Fig. 3
, a mean of two bacteria associated with each HT-29 epithelial cell, with five to six bacteria associating with each LNCaP cell. In analysis of the NIH-3T3 and Swiss-3T3 fibroblastic cells, a mean of 20 bacteria were associated per cell, which was statistically greater (P<0·006) than the number adhering to epithelial cells. Comparisons of the size of epithelial and fibroblastic cells as cultured monolayers found fibroblasts to have about 2·5-times greater surface area than epithelial cells, making the association of bacteria per surface area still slightly greater for fibroblastic than epithelial cells. Based on these analyses, and consistent with previous reports (Fleiszig et al., 1997a
; McGuffie et al., 1999
), we were unable to relate the efficiency of bacterial association with cells to differences in cellular sensitivity to the effects of ExoS. Also evident in Fig. 3
is the lack of significant difference in the association of strain 388 or 388
S bacteria with the different cell lines, indicating no influence of ExoS production on bacterial adherence in these studies.
|
While Ras appeared to be modified similarly in the human epithelial and murine fibroblasts, based on SDS-PAGE analyses, a difference was detected in the modification of the other LMMG proteins between the two cell types (Fig. 4a). Two modified forms of RalA were detected in HT-29 and LNCaP cells, but only one modified form of RalA was detected in NIH-3T3 or Swiss-3T3 fibroblastic cell lines. Comparisons of the ADP-ribosylation of Rab proteins by bacterially translocated ExoS found Rab5, Rab7, Rab8 and Rab11 to be modified in the human epithelial cell lines, but no modification of Rab proteins was detected in murine fibroblasts. [Representative results of Rab5 and Rab8 are shown in Fig. 4(a)
.] Also, Rac1 modification was detected in human epithelial, but not murine fibroblastic, cells. The different number of modifications observed in individual LMMG proteins was previously shown to relate to the sequential addition of ADP-ribose moieties (Fraylick et al., 2002a
, b
).
|
Time-course analysis of bacterial translocation and substrate modification by ExoS ADPRT activity
To further explore whether the limited substrate specificity of fibroblastic cells related to less efficient bacterial translocation of ExoS into cells, LMMG protein substrate modification was monitored relative to the internalization of ExoS ADPRT activity into the eukaryotic cell in a time-dependent study. In these experiments, cytosolic and Triton X-100-solubilized membrane fractions of cells were analysed for ExoS ADPRT activity and RalA, Rab5, Rab8 and Rac1 modification following a 3, 4·5 or 6 h exposure to bacteria. Our studies examined ExoS ADPRT activity, rather than the ExoS protein, because of our ability to quantify ExoS ADPRT activity precisely and because of previous difficulties in detecting ExoS protein within cells by immunoblot analyses (Fraylick et al., 2001).
Consistent with the studies shown in Fig. 4(a), patterns of LMMG protein substrate modification in human epithelial and murine fibroblastic cells remained distinctly different, even with increasing time of exposure to bacteria (Fig. 5
a). The difference in substrate modification was specifically recognized in the single modification of RalA and the absence of modification of Rab5, Rab8 or Rac1 in murine fibroblastic cells, compared with the two modifications of RalA, the two modifications of Rab5 and the single modification of Rab8 and Rac1 detected in human epithelial cells. Analysis of all cell lines was similar in that substrate modification was more efficient in the Triton X-100 membrane fraction when compared with the cytosolic fraction. Also, while variations in levels of individual LMMG proteins were detected between the cytosolic and membrane compartments, the relative distribution of these proteins was generally consistent in all cell lines. For example, there was a general preferential localization of RalA to the membrane fraction, Rab proteins were somewhat equally distributed in both the cytosolic and membrane fractions, and Rac1 preferentially localized to the cytosolic fraction. The lack of prominent differences in the cellular localization of LMMG proteins is supportive of this not being a factor in differences observed in patterns of ExoS substrate modification in human epithelial and murine fibroblastic cells.
|
Patterns of ExoS substrate modification in different cell lines
To explore the specificity of the ExoS ADPRT substrate-modification pattern relative to cell type or animal origin, we compared substrate modification by bacterially translocated ExoS in fibroblasts, epithelial and endothelial cell lines from rodents, monkeys and humans (Table 1; Fig. 6
). Rodent-derived CHO-K1 and TCMK-1 epithelial cells and RHEC endothelial cells exhibited the same pattern of substrate modification as murine NIH-3T3 and Swiss-3T3 cells; that is, showing more limited modification of RalA and no modification of Rab proteins or Rac1. Simian-derived Cos-1 fibroblasts and Vero epithelial cells showed a pattern of substrate modification similar to that of human HT-29 and LNCaP cell lines, although modification was consistently less efficient. Human-derived HT-1080 fibrosarcoma and T24 transitional cell carcinoma cells also displayed the same pattern of substrate modification as LNCaP and HT-29 cells. Since previous studies found tumour-derived cells to be highly sensitive to the effects of ExoS (McGuffie et al., 1999
), we examined whether oncogenic transformation or viral transformation of cell lines could influence substrate specificity. When substrate modification in normal and SV40-transformed human WI-38 fibroblasts (WI-38t) was compared, the pattern of substrate modification of both cell lines was similar to that of other human cell lines, indicating that viral transformation did not alter ExoS substrate modification. In addition, substrate modification by bacterially translocated ExoS in WI-38t cells did not differ from that detected in HT-29, LNCaP, HT-1080 and T24 cells, indicating that neither viral nor oncogenic transformation influenced the substrate specificity of ExoS. The one exception to the pattern of substrate modification observed for human cells was the Detroit 532 fibroblastic cell line, which exhibited a more limited pattern of substrate modification, similar to that of rodent cell lines.
|
![]() |
DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
We have used the TTS-mediated translocation of ExoS by P. aeruginosa to explore how the host cell might influence the translocation and cellular effect of type III effector proteins. Previous studies found that macrophage and epithelial cell lines responded differently to the P. aeruginosa TTS process, with cell morphology being affected in epithelial cell lines and cell viability being affected in macrophages (Coburn & Frank, 1999). ExoS is the most extensively studied of the P. aeruginosa type III effector proteins and comparisons of different epithelial cell lines identified differences in cell sensitivity to bacterially translocated ExoS (McGuffie et al., 1999
). Notable in these comparisons was that epithelial cell sensitivity to ExoS appeared to parallel the opportunistic nature of P. aeruginosa infections, implying a relationship between the TTS process and cells targeted by P. aeruginosa. Since studies supported the notion that the host cell can influence the outcome of TTS-mediated translocation events, we reasoned that we might be able to gain insight into these eukaryotic cell influences by comparing the cellular mechanism of ExoS in cells that respond differently to bacterially translocated ExoS.
We had previously observed differences in the sensitivities of human carcinoma and mouse fibroblastic cells to the effects of bacterially translocated ExoS. Based on these observations, we used HT-29 and LNCaP epithelial cells and murine NIH-3T3 and Swiss-3T3 fibroblasts to explore how the host cell might contribute to differences in the toxicity of ExoS. Previously defined effects of ExoS on epithelial cell function, which include inhibition of DNA synthesis, long-term cell rounding and loss of re-adherence (Olson et al., 1999), were used as criteria for comparisons of the toxicity of ExoS. Two strains of bacteria, 388 and 388
S, were used to distinguish the effects of ExoS from those of other bacterial factors. Studies performed in parallel using strain 388
ExoT confirmed that the cellular effects being monitored related to ExoS, and not ExoT, which is also expressed by strain 388 (data not shown). Our studies identified individual cell line variations in the effects of ExoS on cell function, but overall the human epithelial cell lines examined were more sensitive than the murine fibroblasts to the effects of ExoS on DNA synthesis, morphology and cellular adherence. Of these four cell lines, LNCaP epithelial cells were the most sensitive to the toxic effects of bacterially translocated ExoS and Swiss-3T3 fibroblasts were the most resistant. We also found that decreased sensitivity to ExoS did not relate to decreased bacterial association.
In defining the mechanism for the differential effects of ExoS on cell function, our studies focused on ExoS ADPRT activity, rather than GTPase activating protein activity, since previous studies linked toxic effects of bacterially translocated ExoS on human epithelial cell morphology and DNA synthesis to ExoS ADPRT activity (Fraylick et al., 2001). ExoS ADPRT activity was functionally evaluated based on the efficiency of ADP-ribosylation of cellular substrates. Bacterially translocated ExoS has been found to ADP-ribosylate multiple LMMG proteins in HT-29 epithelial cells, with Ras and RalA modification detected first, followed by the modification of Rab proteins, then Rac1 (Fraylick et al., 2002a
). A different pattern of LMMG protein substrate modification by bacterially translocated ExoS was detected in murine fibroblasts. Although no differences in Ras modification were evident in the two cell types, on the basis of SDS-PAGE, differences were detected by two-dimensional electrophoresis, with one modification of Ras being detected in murine fibroblasts (E. A. Rucks, unpublished results), while two or three sites of Ras modification have been detected in human epithelial cells (Vincent et al., 1999
; Fraylick et al., 2001
). Differences in RalA modification were also detected, with only one modification of RalA detected in murine fibroblasts by SDS-PAGE, while two modifications were evident in human epithelial cells. In analyses of Rab proteins, Rab5, Rab7, Rab8 and Rab11 were all modified in human epithelial cells, while no modification of Rab proteins was detected in murine fibroblastic cell lines. Consistent with previous reports identifying LMMG protein targets of ExoS ADPRT activity in NIH-3T3 fibroblast membranes (Coburn et al., 1989
), when murine fibroblastic cell extracts were used as a source of endogenous LMMG proteins in in vitro ADPRT reactions using purified ExoS, RalA, Rab proteins and Rac1 were found to be modified by ExoS as efficiently as that observed in human epithelial cells. These results support the notion that the decreased efficiency of RalA, Rab and Rac1 modification in murine fibroblasts did not relate to intrinsic protein properties, but rather to in situ cellular effects.
The decreased efficiency of LMMG protein modification in murine fibroblasts suggested that the TTS process functioned differently in murine fibroblastic and human epithelial cells relative to either the efficiency of ExoS internalization or the action of ExoS once it was internalized. In comparing the efficiency of ExoS internalization in the two cell types, ExoS was found to be translocated as (or slightly more) efficiently in murine fibroblasts when compared to human epithelial cells, supporting that view that the TTS apparatus was functional in murine fibroblasts. Consistent with the translocation of ExoS through the membrane into cytosol via the TTS apparatus, levels of ExoS ADPRT activity and the efficiency of substrate modification were higher in membrane than cytoplasmic fractions of all cell lines, as has been recognized previously (Fraylick et al., 2001; Riese & Barbieri, 2002
). Our analyses of ExoS internalization relied on quantification of ExoS ADPRT activity, rather than ExoS protein, for we have had difficulty in detecting ADPRT-active ExoS within cells by immunoblot analyses using antibodies against ExoS (Fraylick et al., 2001
). The more limited modification of LMMG protein substrates in murine fibroblastic cells, in association with efficient ExoS internalization, indicated that cellular factors, beyond membrane translocation, were influencing the ADP-ribosylation of substrates by ExoS within the cell.
Differences in substrate modification between human epithelial and murine fibroblasts prompted us to examine how the pattern of ExoS substrate modification related to cell type, the animal origin of the cells, or the state of viral or oncogenic cellular transformation. In comparing ExoS substrate modification in murine, rat, hamster, simian or human cells, of epithelial, endothelial and fibroblastic cell types, two basic patterns of substrate modification were identified. One pattern showed more restricted substrate ADP-ribosylation, detecting limited modification of RalA and no modification of Rab proteins or Rac1. This pattern was common to rodent cell lines and independent of cell type. The second pattern showed more extensive LMMG protein substrate ADP-ribosylation, with bacterially translocated ExoS modifying RalA more efficiently and also modifying Rab proteins and Rac1. This pattern was apparent in human and simian cell lines, but modification in simian cell lines appeared less efficient. Notably, one human cell line, Detroit 532 fibroblasts, was found to have the more limited ExoS substrate-modification pattern seen in rodent cells. While the mechanistic link between the restricted ExoS ADPRT substrate pattern observed in Detroit fibroblasts and rodent cells is not known, a possible clue may be hidden in the trisomy 21 karyotype of Detroit fibroblasts, which has been linked to alterations in cell signalling pathways involving integrins and protein kinase C (Govoni et al., 1996; Jongewaard et al., 2002
). In assessing the role of viral or oncogenic transformation in ADPRT substrate modification, no alterations in patterns were detected in any of the cell lines in association with cellular transformation.
The cellular mechanism for the different patterns of ADPRT substrate modification is not yet understood. Factors that could alter the substrate specificity of bacterially translocated ExoS include differences in the subcellular targeting or localization of ExoS within cells and/or differences in co-localization of the ExoS eukaryotic cofactor, 14-3-3 protein. Relative to the functional co-localization of the 14-3-3 cofactor, based on our in vitro ExoS ADPRT reactions, we know that the cofactor is present and functional in murine fibroblastic cell extracts, which exhibit the more restricted cellular substrate-modification pattern. It has also been found that 14-3-3 protein co-localizes to endosomes with Rab4 (Bette-Bobillo et al., 1998), although Rab4 does not appear to be a substrate of ExoS when internalized by the TTS process in either epithelial cells (Fraylick et al., 2002a
) or fibroblasts (E. A. Rucks, unpublished observation). These data imply that co-localization of 14-3-3 co-factor by itself is not sufficient for ExoS substrate ADP-ribosylation. An alternative, more likely explanation for altered patterns of ExoS ADPRT substrate modification, based on currently available data, is that differences in the targeting of ExoS within the cell affect its ADPRT substrate specificity. Consistent with this hypothesis, both an N-terminal processing site and a membrane localization domain have been identified within the first 78 aa of ExoS (Pederson et al., 2000
). These sites appear to influence the subcellular localization of ExoS and may also prove to influence the substrate targeting specificity of ExoS ADPRT activity.
In relating substrate patterns of ADP-ribosylation by bacterially translocated ExoS to effects of ExoS on cell function, ExoS has previously been found to affect the function of several of its LMMG protein substrates. The ADP-ribosylation of Ras by ExoS is known to interfere with guanine nucleotide exchange factor-catalysed GDP to GTP exchange, the interaction of Ras with its downstream effector, Raf-1, and activation of the MAP-kinase, Erk-2 (Vincent et al., 1999; Ganesan et al., 1999
; Henriksson et al., 2000a
). In addition, the modification of Ras has been found to correlate with ExoS-dependent inhibition of DNA synthesis (McGuffie et al., 1998
). RalA is involved in signalling processes that affect cell proliferation, vesicular transport and cytoskeletal structure (Feig et al., 1996
; Takai et al., 2001
; Moskalenko et al., 2002
; Sugihara et al., 2002
). Recent studies have shown that the ADP-ribosylation of RalA by ExoS can also affect its function, as recognized by interference of RalA binding to the downstream effector, Ral binding protein 1 (Fraylick et al., 2002b
). Rab proteins control distinct pathways in endocytosis and vesicular trafficking, which affect membrane recycling (Rodman & Wandinger-Ness, 2000
). In vitro analysis found that the ADP-ribosylation of Rab5 by ExoS interfered with its ability to bind early endosomal autoantigen 1 (EEA1), which retarded endosomeendosome fusion (Barbieri et al., 2001
), which, in turn, could influence the regeneration of adherence processes. Together, these data implicate the potential for ExoS-associated alterations of Ras, Ral and Rab protein function to contribute together to the cellular effects of bacterially translocated ExoS.
Consistent with a functional relationship between substrate modification and effects of bacterially translocated ExoS on cell function, LMMG protein ADP-ribosylation was more extensive in human HT-29 and LNCaP epithelial cells, in association with their greater sensitivity to effects of ExoS on cell growth, morphology and re-adherence, when compared with murine fibroblasts. However, while a direct relationship was observed between efficiency of substrate modification and increased sensitivity to the effects of bacterially translocated ExoS, an absolute link between patterns of substrate modification and specific cellular effects of ExoS was not evident, based on analyses used in our studies. For example, effects on DNA synthesis, cell rounding and re-adherence, which were associated with human epithelial cell responsiveness to ExoS, were also detectable in NIH-3T3 cells, which had more limited ExoS substrate modification; albeit the responses of NIH-3T3 cells were less severe. We suspect that an inability to observe a direct relationship between functional effects of ExoS and specific patterns of substrate modification reflects the functional interrelationship of the LMMG proteins targeted by ExoS within the cell. The functions of Ras, Ral, Rac and Rab proteins are all linked to each other through a hierarchy of cellular network interactions. It therefore might be predicted that the more limited pattern of Ras and Ral modification could equate to cellular effects associated with more extensive patterns of substrate modification if given more time or upon more prolonged exposure to ExoS. Such a doseresponse functional relationship of ExoS was evident in cellular re-adherence studies in which NIH-3T3 cells required a longer exposure to ExoS-producing bacteria than HT-29 cells to obtain a similar loss of adherence. While the complexity of cellular signalling events might mask distinct relationships between ExoS substrate modification and effects on cell function, what is clearly evident in comparisons of different cell lines is that the host cell can influence the efficiency of the intracellular targeting of ExoS ADPRT activity.
We conclude from these studies that the cellular response to bacterially translocated ExoS is dictated to varying degrees by the host cell targeted by P. aeruginosa. ADPRT substrate modification by bacterially translocated ExoS is more restricted in rodent cells than in human and simian cells, yet the type III translocation process appears equally efficient in these cells. It becomes apparent from these studies that the host cell is not a passive recipient during the bacteria-driven TTS process, but, rather, cell properties are able to influence the activity and targeting of type III effector proteins.
![]() |
ACKNOWLEDGEMENTS |
---|
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Bette-Bobillo, P., Giro, P., Sainte-Marie, J. & Vidal, M. (1998). Exoenzyme S from P. aeruginosa ADP-ribosylates rab4 and inhibits transferrin recycling in SLO-permeabilized reticulocytes. Biochem Biophys Res Commun 244, 336341.[CrossRef][Medline]
Coburn, J. & Frank, D. W. (1999). Macrophages and epithelial cells respond differently to the Pseudomonas aeruginosa type III secretion system. Infect Immun 67, 31513154.
Coburn, J., Wyatt, R. T., Iglewski, B. H. & Gill, D. M. (1989). Several GTP-binding proteins, including p21c-H-ras, are preferred substrates of Pseudomonas aeruginosa exoenzyme S. J Biol Chem 264, 90049008.
Coburn, J., Kane, A. V., Feig, L. & Gill, D. M. (1991). Pseudomonas aeruginosa exoenzyme S requires a eukaryotic protein for ADP-ribosyltransferase activity. J Biol Chem 266, 64386446.
Cowell, B., Chen, D., Frank, D., Vallis, A. & Fleiszig, S. (2000). ExoT of cytotoxic Pseudomonas aeruginosa prevents uptake by corneal epithelial cells. Infect Immun 68, 403406.
Feig, L. A., Urano, T. & Cantor, S. (1996). Evidence for a Ras/Ral signaling cascade. Trends Biochem Sci 21, 438441.[CrossRef][Medline]
Feltman, H., Schulert, G., Khan, S., Jain, M., Peterson, L. & Hauser, A. R. (2001). Prevalence of type III secretion genes in clinical and environmental isolates of Pseudomonas aeruginosa. Microbiology 147, 26592669.
Ferguson, M. W., Maxwell, J. A., Vincent, T. S., Silva, J. & Olson, J. C. (2001). Comparison of the exoS gene and protein expression in soil and clinical isolates of Pseudomonas aeruginosa. Infect Immun 69, 21982210.
Fleiszig, S., Wiener-Kronish, J., Miyazaki, H., Vallas, V., Mostov, K., Kanada, D., Sawa, T., Yen, T. & Frank, D. (1997a). Pseudomonas aeruginosa-mediated cytotoxicity and invasion correlate with distinct genotypes at the loci encoding exoenzyme S. Infect Immun 65, 579589.[Abstract]
Fleiszig, S., Evans, D., Do, N., Vallas, V., Shin, S. & Mostov, K. (1997b). Epithelial cell polarity affects susceptibility to Pseudomonas aeruginosa invasion and cytotoxicity. Infect Immun 65, 28612867.[Abstract]
Fraylick, J. E., LaRocque, J. R., Vincent, T. S. & Olson, J. C. (2001). The independent and coordinate effects of the ADP-ribosyltransferase and GTPase-activating activities of exoenzyme S on HT-29 epithelial cell function. Infect Immun 69, 53185328.
Fraylick, J. E., Rucks, E. A., Greene, D. M., Vincent, T. S. & Olson, J. C. (2002a). Eukaryotic cell determination of ExoS ADP-ribosyltransferase substrate specificity. Biochem Biophys Res Commun 291, 91100.[CrossRef][Medline]
Fraylick, J. E., Barbieri, J. T., Riese, M. J., Vincent, T. S. & Olson, J. C. (2002b). ADP-ribosylation and functional effects of Pseudomonas ExoS on cellular Ral. Biochemistry 41, 96809687.[CrossRef][Medline]
Frithz-Lindsten, E., Du, Y., Rosqvist, R. & Forsberg, A. (1997). Intracellular targeting of exoenzyme S of Pseudomonas aeruginosa via type III-dependent translocation induces phagocytosis resistance, cytotoxicity and disruption of actin microfilaments. Mol Microbiol 25, 11251139.[Medline]
Fu, H., Coburn, J. & Collier, R. J. (1993). The eukaryotic host factor that activates exoenzyme S of Pseudomonas aeruginosa is a member of the 14-3-3 protein family. Proc Natl Acad Sci U S A 90, 23202324.[Abstract]
Ganesan, A. K., Vincent, T. S., Olson, J. C. & Barbieri, J. T. (1999). Pseudomonas aeruginosa exoenzyme S disrupts Ras-mediated signal transduction by inhibiting guanine nucleotide exchange factor-catalyzed nucleotide exchange. J Biol Chem 274, 2182321829.
Goehring, U.-M., Schmidt, G., Pederson, K. J., Aktories, K. & Barbieri, J. T. (1999). The N-terminal domain of Pseudomonas aeruginosa exoenzyme S is a GTPase-activating protein for Rho GTPases. J Biol Chem 274, 3636936372.
Govoni, S., Bergamaschi, S., Gasparini, L. & 8 other authors (1996). Fibroblasts of patients affected by Down's syndrome oversecrete amyloid precursor protein and are hyporesponsive to protein kinase C stimulation. Neurology 47, 10691075.[Abstract]
Henriksson, M. L., Troller, U. & Hallberg, B. (2000b). 14-3-3 proteins are required for the inhibition of Ras by exoenzyme S. Biochem J 349, 697701.[CrossRef][Medline]
Henriksson, M. L., Rosqvist, R., Telepnev, M., Wolf-Watz, H. & Hallberg, B. (2000a). Ras effector pathway activation by epidermal growth factor is inhibited in vivo by exoenzyme S ADP-ribosylation of Ras. Biochem J 347, 217222.[CrossRef][Medline]
Horoszewicz, J. S., Leong, S. S., Chu, T. M. & 8 other authors (1980). The LNCaP cell line a new model for studies in human prostatic carcinoma. In Models for Prostate Cancer, pp. 115132. Edited by G. P. Murphy. New York: Alan R. Liss.
Hueck, C. J. (1998). Type III protein secretion systems in bacterial pathogens of animals and plants. Microbiol Mol Biol Rev 62, 379433.
Iglewski, B. H., Sadoff, J., Bjorn, M. J. & Maxwell, E. S. (1978). Pseudomonas aeruginosa exoenzyme S: an adenosine diphosphate ribosyltransferase distinct from toxin A. Proc Natl Acad Sci U S A 75, 32113225.[Abstract]
Jongewaard, I. N., Lauer, R. M., Behrendt, D. A., Patil, S. & Klewer, S. E. (2002). Beta 1 integrin activation mediates adhesive differences between trisomy 21 and non-trisomic fibroblasts on type IV collagen. Am J Med Genet 109, 298305.[CrossRef][Medline]
Knight, D. A., Fink-Barbancon, V., Kulich, S. M. & Barbieri, J. T. (1996). Functional domains of Pseudomonas aeruginosa exoenzyme S. Infect Immun 63, 33043309.
Kulich, S. M., Frank, D. W. & Barbieri, J. T. (1995). Expression of recombinant exoenzyme S of Pseudomonas aeruginosa. Infect Immun 63, 18.[Abstract]
Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680685.[Medline]
Masters, S. C., Pederson, K. J., Zhang, L., Barbieri, J. T. & Fu, H. (1999). Interaction of 14-3-3 with a nonphosphorylated protein ligand, exoenzyme S of Pseudomonas aeruginosa. Biochemistry 38, 52165221.[CrossRef][Medline]
McGuffie, E. M., Frank, D. W., Vincent, T. S. & Olson, J. C. (1998). Modification of Ras in eukaryotic cells by Psuedomonas aeruginosa exoenzyme S. Infect Immun 66, 26072613.
McGuffie, E. M., Fraylick, J. E., Hazen-Martin, D. J., Vincent, T. S. & Olson, J. C. (1999). Differential sensitivity of human epithelial cells to Pseudomonas aeruginosa exoenzyme S. Infect Immun 67, 34943503.
Moskalenko, S., Henry, D. O., Rosse, C., Mirey, G., Camonis, J. H. & White, M. A. (2002). The exocyst is a Ral effector complex. Nat Cell Biol 4, 6672.[CrossRef][Medline]
Olson, J. C., McGuffie, E. M. & Frank, D. W. (1997). Effects of differential expression of the 49-kilodalton exoenzyme S by Pseudomonas aeruginosa on cultured eukaryotic cells. Infect Immun 65, 248256.[Abstract]
Olson, J. C., Fraylick, J. E., McGuffie, E. M., Dolan, K. M., Yahr, T. L., Frank, D. W. & Vincent, T. S. (1999). Interruption of multiple cellular processes in HT-29 epithelial cells by Pseudomonas aeruginosa exoenzyme S. Infect Immun 67, 28472854.
Pederson, K. J., Vallis, A. J., Aktories, K., Frank, D. W. & Barbieri, J. T. (1999). The amino-terminal domain of Pseudomonas aeruginosa ExoS disrupts actin filaments via small-molecular weight GTP-binding proteins. Mol Microbiol 32, 393401.[CrossRef][Medline]
Pederson, K. J., Pal, S., Vallis, A. J., Frank, D. W. & Barbieri, J. T. (2000). Intracellular localization and processing of Pseudomonas aeruginosa ExoS in eukaryotic cells. Mol Microbiol 37, 287299.[CrossRef][Medline]
Riese, M. J. & Barbieri, J. T. (2002). Membrane localization contributes to the in vivo ADP-ribosylation of Ras by Pseudomonas aeruginosa ExoS. Infect Immun 70, 22302232.
Rodman, J. S. & Wandinger-Ness, A. (2000). Rab GTPases coordinate endocytosis. J Cell Sci 113, 183192.
Roy-Burman, A., Savel, R. H., Racine, S., Swanson, B. L., Revadigar, N. S., Fujimoto, J., Sawa, T., Frank, D. W. & Wiener-Kronish, J. P. (2001). Type III protein secretion is associated with death in lower respiratory and systemic Pseudomonas aeruginosa infections. J Infect Dis 183, 17671774.[CrossRef][Medline]
Sugihara, K., Asano, S., Tanaka, K., Iwamatsu, A., Okawa, K. & Ohta, Y. (2002). The exocyst complex binds the small GTPase RalA to mediate filopodia formation. Nat Cell Biol 4, 7378.[CrossRef][Medline]
Takai, Y., Sasaki, T. & Matozaki, T. (2001). Small GTP-binding proteins. Physiol Rev 81, 153208.
Towbin, H., Staehelin, T. & Gordon, J. (1979). Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc Natl Acad Sci U S A 76, 43504354.[Abstract]
Vincent, T. S., Fraylick, J. E., McGuffie, E. M. & Olson, J. C. (1999). ADP-ribosylation of oncogenic Ras proteins by Pseudomonas aeruginosa exoenzyme S in vivo. Mol Microbiol 32, 10541064.[CrossRef][Medline]
Yahr, T. L., Barbieri, J. T. & Frank, D. W. (1996a). Genetic relationship between the 53- and 49-kilodalton forms of exoenzyme S from Pseudomonas aeruginosa. J Bacteriol 178, 14121419.[Abstract]
Yahr, T. L., Goranson, J. & Frank, D. W. (1996b). Exoenzyme S of Pseudomonas aeruginosa is secreted by a type III pathway. Mol Microbiol 22, 9911003.[Medline]
Received 10 September 2002;
revised 21 October 2002;
accepted 21 October 2002.