Extraintestinal Pathogenic Escherichia coli Survives within Neutrophils

Bacterial strains and media.The model pathogen CP9 (O4:K54) is an E. coli blood isolate cultured from a patient with sepsis and has been described previously in detail (19, 28). CP9 possesses many of the characteristics of typical ExPEC strains (17, 19) and is highly virulent in a urinary tract infection model (29), an intraperitoneal infection model (31), and a pneumonia model (24, 25). The survival of six additional ExPEC strains within neutrophils was also evaluated. One strain was an isolate with an unknown serotype that was obtained from a patient with bacteremia in Buffalo, NY (strain 104), and two bacteremic isolates with known serotypes were kindly provided by Alan Cross (strains 470 [O25:K5] and 743 [O6:K2]); in addition, we assessed three strains that have been used as model pathogens for investigations of ExPEC pathogenesis (K1/Y [O7:K1] [39], IA2 [O4:K12] [15], and CFT073 [O6:K2] [40]), as well as K-12 laboratory strain HB101. Bacterial strains were grown in iron-chelated LB broth (5 g/liter yeast extract, 10 g/liter tryptone, 10 g/liter NaCl, 31.24 mg/liter 2,2′-dipyridyl). Pilot data demonstrated that there was no difference between the intracellular survival of stationary-phase bacteria (grown at 37°C in a reciprocal shaking water bath at 70 rpm for 18 h) and the intracellular survival of logarithmic-phase bacteria; therefore, stationary-phase bacteria were used for in vitro assays.

Human neutrophil purification.“Naïve” neutrophils were purified from blood from healthy human volunteers by dextran sedimentation and centrifugation through Ficoll-Paque Plus (Amersham Biosciences, Piscataway, NJ) performed according to the manufacturer's instructions. After purification, neutrophil preparations consisted of 96 to 100% neutrophils and 0 to 4% lymphocytes and eosinophils. Neutrophils were resuspended in 1× phosphate-buffered saline (pH 7.4) (PBS)-20% autologous heat-inactivated human plasma to obtain the desired concentrations.

Neutrophil-mediated bactericidal assay.Neutrophils were isolated as described above, and 5 × 10⁵ neutrophils were mixed with approximately 1 × 10³, 1 × 10⁴, and 1 × 10⁵ CFU of CP9 or 1.0 × 10⁵ CFU of CFT073 or IA2 in 1 ml 1× PBS containing 10% autologous plasma (heated for 30 min at 56°C to inactivate complement), 1% anti-CP9 rabbit polyclonal antibody, and 1% active autologous plasma. Samples were mixed with a Nutator (Becton, Dickinson and Co., Sparks, MD) at 37°C, and after 0, 30, 60, and 90 min, aliquots were removed and plated on LB plates following serial dilution for determination of bacterial titers.

Assessment of neutrophil morphology and enumeration of neutrophils.To assess neutrophil morphology after interaction with E. coli strains, a cytoslide was prepared as described previously (26), stained with Diff-Quik reagents (Baxter, Miami, FL), and examined by light microscopy (Nikon ECLIPSE 80i microscope; Nikon Instruments Inc., Melville, NY). To enumerate neutrophils, a 50-μl aliquot was diluted 1:200 in Isoton II solution (Beckman Coulter, Miami, FL), and the neutrophil counts were determined using a Multisizer 3 Coulter Counter (Beckman Coulter).

Assessment of the survival of ExPEC within human neutrophils.To measure the survival of E. coli within neutrophils, a modification of the well-described cell invasion assay was used (14). The premise of this assay is that bacteria and cells are allowed to interact for a certain time to enable entry of the bacteria into the cells being assessed. Gentamicin is then added, and after a defined time the cells are lysed and the number of bacterial CFU is determined. Since gentamicin does not enter cells, the bacteria counted are considered to be intracellular.

First, we tested the assumption that gentamicin killed all extracellular bacteria (both non-cell-associated bacteria and bacteria associated with the surface of the neutrophils). To accomplish this, neutrophils were preincubated for 10 min at 37°C with 20 μM cytochalasin D (Sigma-Aldrich, St. Louis, MO). Cytochalasin D inhibits actin polymerization, thereby inhibiting phagocytosis (3, 35). Next, 5 × 10⁵ purified human neutrophils, pretreated with cytochalasin D, were incubated with 1 × 10⁷ CFU of CP9, 1% autologous serum, and 1% rabbit anti-CP9 serum in polystyrene tubes for 60 min and mixed with a Nutator at 37°C. Subsequently, gentamicin (final concentration, 200 μg/ml) was added, and the tubes were incubated for an additional 60 min. Next, the neutrophils were lysed, and the number of CP9 CFU was determined. Surprisingly, in repeated assays, approximately 1 × 10³ to 3 × 10³ CFU were consistently found. Together, these experiments suggested that under our assay conditions, either treatment with gentamicin did not kill all extracellular bacteria or cytochalasin D did not completely inhibit phagocytosis. To try to determine which of these possibilities was correct, experiments were first performed to determine if increasing the gentamicin concentration or incubation time eradicated residual extracellular bacteria that survived treatment with 200 μg/ml of gentamicin for 60 min. Data obtained in these experiments confirmed that treatment with gentamicin did not eradicate all extracellular bacteria and that treatment with 200 μg/ml of gentamicin for 30 min was as effective as treatment with higher concentrations for longer incubation times (data not shown). Therefore, in all subsequent experiments 30 min of incubation with 200 μg/ml of gentamicin was utilized. Next, we assessed whether cytochalasin D completely inhibited phagocytosis, as incomplete inhibition would allow some bacteria to be internalized and therefore escape killing by gentamicin. (i) In pilot experiments we determined that incubation of neutrophils with 20 μM cytochalasin D for up to 60 min was as effective as a 10-min incubation with decreasing background titers (data not shown). Therefore, incubation with 20 μM cytochalasin D for 10 min was used in all subsequent experiments. (ii) A total of 1 × 10⁷ CFU of CP9 were incubated for 60 min with neutrophils that were pretreated with cytochalasin D or were not treated. Next, a cytoslide was prepared and examined by light microscopy. Intracellular CP9 cells were readily observed in many of the untreated neutrophils but were never observed in neutrophils pretreated with cytochalasin D. However, the sensitivity of this method is not optimal, and low titers of intracellular bacteria may not be identified. (iii) Finally, using a flow cytometry-based assay that measured uptake of fluorescein isothiocyanate-labeled zymosan particles (23), we found that phagocytosis in neutrophils treated with 20 μM cytochalasin D was mostly but not completely inhibited (Fig. 1). Using our flow cytometric data, we could not precisely determine the degree to which phagocytosis was inhibited by cytochalasin D, but even if phagocytosis was inhibited by 99% (which we estimated), a measurable number of intracellular bacteria may still have been present. Therefore, we could not be sure whether under the conditions of our assay the bacteria enumerated when neutrophils were pretreated with cytochalasin D were intracellular or were extracellular and not killed by gentamicin. The most conservative course was to calculate survival by subtracting this number of bacteria. We felt that it was better to underestimate survival than to overestimate survival. Therefore, we opted to calculate intracellular survival by determining the difference between the bacterial titer obtained for untreated neutrophils and the bacterial titer obtained for cytochalasin D-pretreated neutrophils.

• Open in new tab
• Download powerpoint

FIG. 1.

Flow cytometric analysis to determine the efficacy of cytochalasin D inhibition of phagocytosis. Purified neutrophils that were or were not pretreated with 20 μM cytochalasin D for 10 min at 37°C were or were not mixed with fluorescein isothiocyanate-labeled zymosan particles for 60 min at 37°C and then quenched for 1 min with 0.4% trypan blue, and samples were subjected to flow cytometric analysis. (A) Quenched neutrophils not pretreated with cytochalasin D and not exposed to zymosan particles. (B) Quenched neutrophils not pretreated with cytochalasin D and exposed to zymosan particles. (C) Quenched neutrophils pretreated with cytochalasin D and not exposed to zymosan particles. (D) Quenched neutrophils pretreated with cytochalasin D and exposed to zymosan particles. cyto. D, cytochalasin D.

Standard assay used to measure survival of E. coli within neutrophils.As a result of these findings, assays to measure the survival of E. coli within neutrophils were performed as follows. A 1-ml mixture consisting of (i) 5 × 10⁵ purified human neutrophils (resuspended in 500 μl of 1× PBS-20% autologous human plasma heated at 56°C for 30 min to inactivate complement) that were either incubated with 20 μM cytochalasin D for 10 min at 37°C prior to the addition of bacteria and subsequently throughout the assay or not treated, (ii) approximately 5 × 10⁶ CFU of the strain being assessed (resuspended in 100 μl of 1× PBS), (iii) 10 μl of autologous serum (final concentration, 1%), (iv) 10 μl of rabbit anti-CP9 serum (final concentration, 1%), and (v) 1× PBS (380 μl) was incubated for 60 min at 37°C on a Nutator. For assays in which we assessed the survival of HB101, the autologous serum and rabbit anti-CP9 serum were omitted since HB101 is serum sensitive. Next, gentamicin (final concentration, 200 μg/ml) was added; in addition, 20 μM cytochalasin D was added to the untreated neutrophils in order to inhibit further phagocytosis. Then the preparations were incubated for an additional 30 min at 37°C. Pilot experiments established that addition of cytochalasin D at the time that it was added had no effect on survival. Samples were then placed on ice, and neutrophils were washed twice with 5 ml of 1× PBS to remove the gentamicin, which could kill intracellular bacteria when they were subsequently released by lysis of neutrophils. Initially, neutrophils were lysed with saponin (final concentration, 0.5 mg/ml) to release intracellular bacteria. However, a comparison of the bacterial titers obtained for neutrophils that were lysed with saponin and the bacterial titers obtained for neutrophils that were not treated with saponin showed that the titers were similar; therefore, saponin-mediated lysis was not performed. Presumably, neutrophils lyse when they are spread onto LB plates. Bacterial titers were determined by 10-fold serial dilution in 1× PBS. The level of intracellular survival was calculated by determining the difference between the bacterial titer obtained for untreated neutrophils and the bacterial titer obtained for cytochalasin D-pretreated neutrophils.

Standard assay used to measure survival of E. coli within neutrophils over time.In the standard assay used to measure intracellular survival of E. coli within neutrophils, we incubated neutrophils and bacteria for 60 min. To determine the survival of CP9 within neutrophils over time and to determine whether CP9 was capable of intracellular replication, neutrophils and CP9 were incubated together as described above for the standard assay, except that the volume of the mixture was increased to 2 ml. After 60 min gentamicin and cytochalasin D were added to prevent further phagocytosis. At 90, 120, 150, 180, and 210 min, 200-μl aliquots were removed to measure intracellular survival as described above. In addition, neutrophils were concomitantly enumerated since high titers of CP9 have been determined to cause neutrophil lysis (26).

Standard assay used to measure survival of E. coli within neutrophils in which the oxidative burst was inhibited.The standard survival assay was performed utilizing neutrophils that were (i) not treated, (ii) pretreated with the NADPH oxidase inhibitor diphenylene iodonium (DPI) (10 μM DPI suspended in dimethyl sulfoxide [DMSO], preincubated with neutrophils for 10 min at 37°C) (5), (iii) pretreated with cytochalasin D (20 μM for 10 min at 37°C), and (iv) pretreated with cytochalasin D and DPI. A pilot experiment was performed to confirm the effectiveness of DPI using the Phagoburst assay (Phagoburst [Orpegen Pharma], distributed by BioCarta, San Diego, CA) according to the manufacturer's instructions. In brief, 1 × 10⁶ purified human neutrophils that either were not treated or were pretreated with DPI were stimulated with either 8.1 μM phorbol 12-myristate 13-acetate (positive control) or 5 × 10⁷ CFU of live CP9 (wild type) for 10 min at 37°C. Untreated neutrophils or neutrophils exposed to DMSO (DPI solvent) were used as negative controls. The generation of reactive oxygen species was measured by flow cytometry by gating on 10,000 neutrophil events and determining the proportion of these cells for conversion of the fluorogenic substrate dihydrorhodamine 123 to rhodamine 123. In the presence of DPI, nearly complete inhibition of the generation of rhodamine 123 was observed (Fig. 2), confirming the effectiveness of DPI as an inhibitor of the oxidative burst.

• Open in new tab
• Download powerpoint

FIG. 2.

Confirmation of the efficacy of DPI for inhibiting the neutrophil oxidative burst. A total of 1 × 10⁶ purified human neutrophils that were either not treated or pretreated with DPI were stimulated with various reagents. The generation of reactive oxygen species was measured by flow cytometric analysis using a Phagoburst assay kit. (Top left panel) Neutrophils only, autofluorescence. (Top right panel) Neutrophils exposed to only DMSO (DPI diluent). (Middle left panel) Neutrophils exposed to phorbol 12-myristate 13-acetate (PMA). (Middle right panel) Neutrophils pretreated with DPI and exposed to phorbol 12-myristate 13-acetate. Inhibition is expressed as the percent decrease compared with the untreated sample (middle left panel). (Bottom left panel) Neutrophils exposed to CP9. (Bottom right panel) Neutrophils pretreated with DPI and exposed to CP9. Inhibition is expressed as the percent decrease compared with the untreated sample (bottom left panel).

Measurement of ExPEC binding to neutrophils.Differences in bacterial binding to untreated neutrophils may affect phagocytosis, which in turn may affect the survival. Therefore, we examined whether the strains whose survival within neutrophils was assessed in this study bound differently to neutrophils. A total of 5 × 10⁵ neutrophils were mixed with 8.0 × 10⁶ to 1.2 × 10⁷ CFU of CP9, CFT073, IA2, K1/Y, 743, 470, and 104 at 4°C for 30 min. At this temperature bacteria can bind to but are not phagocytosed by neutrophils. Subsequently, differential centrifugation (110 × g for 11 min) was performed to separate neutrophil-bound and nonbound bacteria, and the bacteria bound to neutrophils were enumerated. The percentage of bacteria bound was calculated by determining the ratio of the number of neutrophil-bound bacteria to the initial bacterial input titer. The percentages of bacteria that bound to neutrophils were similar for all strains, as follows: CP9, 3.3%; CFT073, 2.7%; IA2, 4.2%; K1/Y, 3.5%; 743, 3.0%; 470, 2.7%; and 104, 4.2%. Therefore, this factor did not appear to be a confounding variable in our survival assay.

Serum sensitivity assay.To determine the effect of complement-mediated bactericidal activity on CFT073 and IA2, serum sensitivity assays were performed as described previously (32), except that 1% active serum and 1% active anti-CP9 rabbit polyclonal antibody were utilized. Bacterial titers were determined at 0, 30, and 60 min.

Assessment of neutrophil phagocytosis.Laser scanning confocal microscopy was performed in order to measure the phagocytosis of various ExPEC strains. ExPEC strains were stained with 400 μM carboxyfluorescein diacetate (succinimidyl ester) (Invitrogen) for 2 h and then washed three times in 1× PBS. Neutrophils were isolated from human blood, and the intracellular survival assay was performed as described above using bacteria stained with carboxyfluorescein diacetate (succinimidyl ester). After the final wash, the neutrophil pellet was resuspended in 20 μl of 1× PBS and allowed to adhere to poly-l-lysine (Sigma)-pretreated coverslips for 20 min at room temperature. Nonadherent neutrophils were removed by washing the preparations with 1× PBS. In order to identify the outer membrane of neutrophils, cells were incubated for 5 min at room temperature with 2 μM 1,1′-dioctadecyl-6,6′-di(4-sulfophenyl)-3,3,3′,3′-tetramethylindocarbocyanine (Invitrogen). The cells were washed twice in 1× PBS and were then fixed using 3.5% paraformaldehyde (Sigma) for 5 min at 37°C. Fixation was followed by 3 min of permeabilization in 0.3% Triton-X-1× PBS. After washing, cell nuclei were stained with 1 μM TO-PRO-3 (Invitrogen) for 15 min at room temperature. The cells were then washed and mounted on glass slides using Vectashield (Vector) mounting medium. Neutrophils were examined at the Confocal Microscopy and 3-D Imaging Facility (School of Medicine and Biomedical Sciences, University at Buffalo) using a Zeiss LSM 510 Meta confocal microscope. Z stacks were analyzed in order to distinguish intracellular bacteria from extracellular bacteria. One hundred neutrophils were counted per slide, and the percentage of phagocytosis was calculated. For assessment of phagocytosis by light microscopy we utilized cytoslides. One hundred neutrophils were examined for the presence of intracellular bacteria (within the limits of the methodology).

Transmission electron microscopy (TEM) of the neutrophil interaction with CP9.A total of 5 × 10⁵ human neutrophils were exposed to CP9 (1 × 10⁷ CFU/ml) and mixed with a Nutator at 37°C for 1 h. After this incubation, cells were fixed in 4% glutaraldehyde in 1× PBS. Pellets were rinsed in 1× PBS, postfixed in 1% OsO₄, dehydrated with a graded alcohol series, and embedded in Embed-Araldite plastic. Ultrathin sections of samples were prepared using standard techniques (12). The ultrathin sections were examined with a JEOL 100Cx-II electron microscope.

Intracellular survival assay for inhibition of macropinocytosis.We examined the effects of the macropinocytosis inhibitors methylamine (10 to 100 mM), amiloride (3 mM), dimethyl amiloride (500 μM), and rottlerin (2 μM) in our intracellular survival assays to determine their effects on the intracellular survival of CP9. Unfortunately, methylamine was found to be toxic to our bacteria and therefore could not be used in our assay. In a recent paper the researchers noted that cytochalasin D, amiloride, and dimethyl amiloride lacked specificity for blocking phagocytosis and/or macropinocytosis in other phagocytic cells (33), but the protein kinase C inhibitor rottlerin was found to be an effective specific inhibitor of macropinocytosis; thus, we used this inhibitor in our assay as well. The standard survival assay for measuring E. coli survival within neutrophils was used. Neutrophils were (i) not pretreated, (ii) pretreated with 20 μM cytochalasin D for 10 min at 37°C, (iii) pretreated with amiloride for 15 min at 37°C, (iv) pretreated with dimethyl amiloride for 15 min at 37°C, (v) pretreated with rottlerin for 15 min at 37°C, (vi) pretreated with amiloride and cytochalasin D, or (vii) pretreated with dimethyl amiloride and cytochalasin D.

ExPEC model pathogen CP9 is relatively resistant to neutrophil-mediated bactericidal activity.Neutrophil-mediated bactericidal activity against various titers of CP9 was assessed in the presence of complement and CP9-specific antisera as opsonins. While some killing was observed from 60 to 90 min, the magnitude was only approximately 0.25 log (Fig. 3). These data demonstrate that CP9 is relatively resistant to neutrophil-mediated bactericidal activity.

• Open in new tab
• Download powerpoint

FIG. 3.

Neutrophil-mediated bactericidal activity against ExPEC strain CP9. Approximately 1 × 10³, 1 × 10⁴, and 1 × 10⁵ CFU of CP9 in 1 ml of 1× PBS containing 10% heat-inactivated autologous plasma, 1% anti-CP9 rabbit polyclonal antibody, and 1% active autologous plasma were incubated with and without 5 × 10⁵ human neutrophils isolated from peripheral blood. Samples were mixed with a Nutator at 37°C, and at 0, 30, 60, and 90 min aliquots were removed and bacterial titers were determined. CP9 survival over time in the presence and absence of neutrophils is shown for each titer.

ExPEC model pathogen CP9 survives within human neutrophils.Strains of ExPEC are thought to be classical extracellular bacterial pathogens, whose primary defense against neutrophil-mediated bactericidal activity is avoidance of phagocytosis; once phagocytosed, ExPEC strains are believed to be killed. However, if strains of ExPEC were able to avoid intracellular killing and survive within neutrophils, this would give them a significant advantage and increase their pathogenic capability. In neutrophil morphological studies we noted that ExPEC strain CP9 appeared to be able to survive within neutrophils (26). Therefore, we hypothesized that an additional mechanism by which ExPEC may subvert PMN function involves the ability to survive within PMNs after phagocytosis. To test this hypothesis, we assessed the survival of ExPEC wild-type strain CP9 within human neutrophils.

To assess the survival of CP9 within neutrophils, we used a modification of the cell invasion assay (14). The premise of this assay is that bacteria that are able to survive within host cells can be identified by adding gentamicin, which kills all extracellular bacteria but not intracellular bacteria since gentamicin is unable to enter cells. To control for extracellular bacteria that may not have been killed by gentamicin, intracellular survival was calculated by determining the difference between the bacterial titer obtained for untreated neutrophils and the bacterial titer obtained for neutrophils pretreated with cytochalasin D, an inhibitor of phagocytosis. However, this provided a conservative estimate of survival since some (or all) of the bacteria counted when neutrophils were pretreated with cytochalasin D may actually have been intracellular since we were not able to completely inhibit phagocytosis (see Materials and Methods). The survival of CP9 within neutrophils is shown in Fig. 4. The number of CP9 CFU obtained for untreated neutrophils reflected both the number of intracellular CP9 cells and the number of extracellular CP9 cells (if any) that survived treatment with gentamicin (Fig. 4A), and the number of CP9 CFU obtained for neutrophils pretreated with cytochalasin D included the extracellular CP9 cells that were not killed by gentamicin and/or the intracellular bacteria that may have been present if cytochalasin D did not mediate complete inhibition of neutrophil phagocytosis (Fig. 4A). The calculated difference between these titers was a conservative estimate of intracellular survival (Fig. 4B). The level of intracellular survival was dependent on the input titer of CP9, increasing as the input titer increased (Fig. 4B). When 1.5 × 10⁷ CFU of CP9 were added to 5 × 10⁵ neutrophils, we calculated that 1.6 × 10⁵ ± 6.7 × 10⁴ CFU of CP9 survived. However, since the efficiency of phagocytosis of CP9 is unknown, it is not clear what proportion of CP9 cells survived within neutrophils once they were phagocytosed. At the lowest input titer at which we were able to detect intracellular survival of CP9 (3.5 × 10⁵ CFU), 1.8 × 10² ± 2.5 × 10² CFU of CP9 were calculated to survive. Whether survival occurs at lower input titers is not clear, since we were not able to decrease the background titer of gentamicin-resistant CP9 determined when neutrophils were pretreated with cytochalasin D to less than 1 × 10³ CFU. These results clearly demonstrate that the model ExPEC strain CP9 is able to survive within human neutrophils.

• Open in new tab
• Download powerpoint

FIG. 4.

Survival of ExPEC strain CP9 within human neutrophils. The standard assay for measuring intracellular survival was used (see Materials and Methods). The level of intracellular survival was calculated by determining the difference between the bacterial titer obtained for untreated neutrophils and the bacterial titer obtained for cytochalasin D-pretreated neutrophils. (A) Bacterial titers obtained for neutrophils that were pretreated (cytochalasin D-pretreated neutrophils) and were not pretreated (untreated neutrophils) with cytochalasin D. (B) Calculated levels of survival for various titers of CP9.

ExPEC strain CP9 survives but does not replicate within neutrophils.In previous studies (26) microscopic examination of neutrophils that had interacted with CP9 revealed multiple intracellular bacteria. The observation that CP9 was able to survive within neutrophils raised the question of whether CP9 could also replicate within neutrophils. To answer this question, we performed the gentamicin protection assay as described above but assessed survival over time. A total of 5 × 10⁶ CFU of CP9 were incubated with 5 × 10⁵ purified human neutrophils that either were pretreated or were not pretreated with cytochalasin D. After 1 h of incubation with CP9, neutrophils were treated with gentamicin to kill the extracellular bacteria and with cytochalasin D (only neutrophils that were not pretreated) to prevent further phagocytosis. Neutrophils were then harvested to assess the intracellular survival of CP9 at 30, 60, 90, 120, and 150 min after the addition of gentamicin (Fig. 5A). For the times examined, the survival of CP9 over time was constant. These data can be explained by either (i) a lack of intracellular replication of CP9, (ii) an equilibrium between intracellular bacterial replication and death, and/or (iii) intracellular bacterial replication and lysis of neutrophils over time, since it was shown in previous studies (26) that CP9 was capable of lysing neutrophils. Neutrophil lysis would expose intracellular CP9 to gentamicin, which in turn would decrease the titer of CP9. To assess possible neutrophil lysis under the conditions used in the survival assay, neutrophils were counted in parallel with the enumeration of CP9. As Fig. 5B shows, the numbers of neutrophils were constant during this experiment. Therefore, CP9 did not undergo intracellular replication, or intracellular replication and death were in equilibrium. These results also demonstrated that CP9 was able to survive within neutrophils for up to 210 min.

• Open in new tab
• Download powerpoint

FIG. 5.

Survival of ExPEC strain CP9 within human neutrophils over time. The standard assay for measuring survival of E. coli within neutrophils over time was used (see Materials and Methods). (A) Calculated levels of survival of CP9, obtained by determining the difference between the bacterial titer obtained for untreated neutrophils (PMNs plus CP9, not pretreated) and the bacterial titer obtained for cytochalasin D-pretreated neutrophils (PMNs plus CP9, pretreated). (B) Multisizer Coulter Counter readings for neutrophils over time, obtained concomitant with the intracellular survival assay. PMNs-no bacteria represents neutrophils that were incubated at 37°C but not exposed to CP9 or cytochalasin D.

Survival of ExPEC within neutrophils was observed when additional ExPEC strains were used.Although previous studies suggested that various strains of ExPEC have common virulence factors that enable them to cause extraintestinal infections (18), it was important to examine whether CP9 was unique or whether other ExPEC strains could survive in neutrophils. Therefore, six additional ExPEC strains and E. coli laboratory strain HB101 were assessed to determine whether they could survive within neutrophils. Not surprisingly, HB101 was not able to survive within human neutrophils. In contrast, four of the six additional ExPEC strains were capable of survival within neutrophils (Fig. 6). These studies demonstrated that even when our conservative method was used, the majority of ExPEC strains assessed to date (5 of 7 strains, or 71%) are able to survive within human neutrophils and that this phenotype is not limited to CP9. We were not able to demonstrate that CFT073 and IA2, which are well-studied ExPEC isolates, survived. One potential explanation for this is that these strains actually survive within neutrophils but we were unable to establish this fact with our methodology and because of the sensitivity of our assay. Alternative explanations were also considered. Therefore, studies were performed to determine if serum sensitivity, resistance to phagocytosis, rapid neutrophil-mediated killing, or toxic effects of neutrophils on the strains could explain our findings. However, CFT073 and IA2 were resistant to the concentration of serum used in the survival assay, were not killed rapidly by neutrophils, did not lyse neutrophils, and were not resistant to phagocytosis (data not shown). Therefore, the reason(s) for the lack of survival of CFT073 and IA2 within human neutrophils is unclear.

• Open in new tab
• Download powerpoint

FIG. 6.

Intracellular survival of additional E. coli strains. The standard intracellular survival assay was performed with ExPEC strains CP9 (O4:K54), 743 (O6:K2), 470 (O25:K5), K1/Y (O7:K1), 104 (unknown serotype), IA2 (O4:K12), and CFT073 (O6:K2) and laboratory strain HB101, as described in Materials and Methods. The levels of intracellular survival were calculated by determining the difference between the bacterial titer obtained for untreated neutrophils and the bacterial titer obtained for cytochalasin D-pretreated neutrophils.

ExPEC wild-type strain CP9 is partially susceptible to the oxidative burst of neutrophils.One mechanism by which CP9 may be able to survive within neutrophils involves complete or partial resistance to the oxidative burst. To examine this possibility, we measured the intracellular survival of CP9 in neutrophils in which NADPH oxidase, the enzyme responsible for production of reactive oxygen species, was and was not inhibited by DPI. The results of these studies demonstrated that inhibition of NADPH oxidase resulted in an approximate 0.5-log increase in the intracellular survival of CP9 (the mean survival titer increased from 6.77 × 10⁴ ± 2.2 × 10⁴ CFU/ml to 4.56 × 10⁵ ± 5.15 × 10⁴ CFU/ml) (Fig. 7). These findings demonstrate that CP9 is at least partially susceptible to the neutrophil oxidative burst. Therefore, absolute resistance to the oxidative burst is not the mechanism by which CP9 survives within neutrophils.

• Open in new tab
• Download powerpoint

FIG. 7.

Survival of ExPEC strain CP9 within neutrophils in which the oxidative burst was inhibited. The standard assay for measuring survival of E. coli within neutrophils in which the oxidative burst has been inhibited was used (see Materials and Methods). Neutrophils were (i) not treated, (ii) pretreated with the NADPH oxidase inhibitor DPI, (iii) pretreated with cytochalasin D, and (iv) pretreated with cytochalasin D and DPI. The levels of intracellular survival were calculated by determining the difference between the bacterial titer obtained for untreated neutrophils and the bacterial titer obtained for cytochalasin D-pretreated neutrophils and the difference between the bacterial titer obtained for DPI-pretreated neutrophils and the bacterial titer obtained for DPI- and cytochalasin D-pretreated neutrophils. The asterisk indicates that the P value was <0.05, as determined by an unpaired t test.

ExPEC appears to be incorporated into phagosomes within neutrophils.To assess whether avoidance of uptake into a phagosome and escape from the phagosome were mechanisms that contributed to CP9's survival within human neutrophils, we performed TEM with neutrophils that had phagocytosed CP9 under the conditions used in our standard survival assay. The findings of these studies demonstrated that CP9 always appeared to be within a phagosome. Some of the phagosomes appeared to be spacious (Fig. 8A), while others appeared to be tight phagosomes (Fig. 8B). Therefore, avoidance of neutrophil uptake via the phagosome and subsequent escape from the phagosome do not appear to be mechanisms by which CP9 survives within neutrophils.

• Open in new tab
• Download powerpoint

FIG. 8.

Visualization of ExPEC strain CP9 within neutrophils by TEM. TEM was performed with neutrophils from a standard assay to measure survival of E. coli as described in Materials and Methods. Both images were obtained from the same sample. (A) CP9 in a “spacious” phagosome. (B) CP9 in a “tight” phagosome.

ExPEC is not taken up via macropinocytosis.To assess whether CP9 was taken up by macropinocytosis and was subsequently confined in a protective vacuole within the neutrophil (a potential survival mechanism), we inhibited neutrophil macropinocytosis using three known inhibitors, amiloride, dimethyl amiloride, and rottlerin. No change in intracellular survival was observed with dimethyl amiloride or rottlerin, but a slight increase in survival was observed with amiloride (Fig. 9). These data do not support the hypothesis that uptake of via macropinocytosis is a mechanism that contributes to the survival of ExPEC within neutrophils.

• Open in new tab
• Download powerpoint

FIG. 9.

Effects of macropinocytosis inhibitors on the survival of ExPEC strain CP9 within neutrophils. Purified human neutrophils were not treated or were pretreated with either cytochalasin D (20 μM), amiloride (3 mM), dimethyl amiloride (500 μM), rottlerin (2 μM), a combination of cytochalasin D and amiloride, or a combination of cytochalasin D and dimethyl amiloride (DMA) for 15 min before the intracellular survival assay was performed. The intracellular survival assay was performed as described in the text.