Neisseria gonorrhoeae Infection Protects Human Endocervical Epithelial Cells from Apoptosis via Expression of Host Antiapoptotic Proteins

Abstract

Several microbial pathogens can modulate the host apoptotic response to infection, which may contribute to immune evasion. Various studies have reported that infection with the sexually transmitted disease pathogen Neisseria gonorrhoeae can either inhibit or induce apoptosis. N. gonorrhoeae infection initiates at the mucosal epithelium, and in women, cells from the ectocervix and endocervix are among the first host cells encountered by this pathogen. In this study, we defined the antiapoptotic effect of N. gonorrhoeae infection in human endocervical epithelial cells (End/E6E7 cells). We first established that N. gonorrhoeae strain FA1090B failed to induce cell death in End/E6E7 cells. Subsequently, we demonstrated that stimulation with N. gonorrhoeae protected these cells from staurosporine (STS)-induced apoptosis. Importantly, only End/E6E7 cells incubated with live bacteria and in direct association with N. gonorrhoeae were protected from STS-induced apoptosis, while heat-killed and antibiotic-killed bacteria failed to induce protection. Stimulation of End/E6E7 cells with live N. gonorrhoeae induced NF-κB activation and resulted in increased gene expression of the NF-κB-regulated antiapoptotic genes bfl-1, cIAP-2, and c-FLIP. Furthermore, cIAP-2 protein levels also increased in End/E6E7 cells incubated with gonococci. Collectively, our results indicate that the antiapoptotic effect of N. gonorrhoeae in human endocervical epithelial cells results from live infection via expression of host antiapoptotic proteins. Securing an intracellular niche through the inhibition of apoptosis may be an important mechanism utilized by N. gonorrhoeae for microbial survival and immune evasion in cervical epithelial cells.

A number of microbial pathogens have evolved complex immune evasion strategies to establish infection in human hosts (30). These include antigenic variation (68), degradation of secretory immunoglobulin (47), inhibition of complement activation (62), and modulation of apoptosis (24, 48). Apoptosis, a host defense strategy that eliminates the infected host cell via clearance by phagocytosis, is often triggered by bacterial infection (57). However, pathogen-directed induction of apoptosis is also a successful strategy for bacteria to escape from immune cells or to prevent the tissue destruction and inflammation associated with cell necrosis, thereby avoiding recognition by the immune system (69). Conversely, pathogen-directed inhibition of apoptosis has been well characterized for the obligate intracellular bacteria of the genus Chlamydia, which likely use this strategy to secure an intracellular niche for replication (1, 18, 22, 28, 48, 59).

Cells that undergo apoptosis are characterized by distinct morphological and chemical changes that result from signaling cascades initiated through two distinct pathways, the extrinsic and intrinsic pathways (3, 12). Engagement of death receptors on the cell surface leads to activation of the extrinsic pathway of apoptosis (63). In contrast, intracellular events, such as formation of reactive oxygen species or acidification of the cytosolic pH, cause mitochondrial release of cytochrome c into the cytoplasm (9, 37), which activates the intrinsic pathway (11, 61). In both pathways, apoptosis proceeds via the activation of the caspase cascade (10, 12) and results in DNA degradation, cell shrinkage, and membrane blebbing (34). Common effects of bacterial infection include the production of reactive oxygen species and induction of intracellular stress, which may lead to intrinsic pathway activation (38, 65). Families of pro- and antiapoptotic proteins also exist to regulate the process of apoptosis (67). Antiapoptotic proteins, which in some cases are induced by bacterial infection (28), act by directly inhibiting a variety of apoptotic events, including caspase activity (17) and mitochondrial depolarization (2, 27).

The host cell response to infection with the sexually transmitted disease pathogen Neisseria gonorrhoeae includes activation of transcription factors such as NF-κB and AP-1 (54), upregulation of cell adhesion molecules (35, 50), production of inflammatory cytokines (20, 41, 54), and modulation of apoptosis (7, 49, 52, 53). While N. gonorrhoeae infection of HeLa endocervical adenocarcinoma cells has been reported to induce apoptosis (52, 53), a number of other studies have reported that N. gonorrhoeae does not induce apoptosis (7, 31, 32, 49, 51, 64). Furthermore, incubation of N. gonorrhoeae with a variety of cell types other than HeLa cells has been reported to protect host cells from apoptosis (7, 31, 32, 49). These discrepancies may be due to the specific responses of different cell types, culture conditions, and bacterial strains.

Gonococcal infection begins at mucosal surfaces (66), typically the urethra in males and the cervix in females (16). The cervix is comprised of the ectocervix, which consists of stratified squamous epithelium, and the endocervix, which consists of a single layer of columnar epithelial cells. It is thought that N. gonorrhoeae establishes infections in the sterile endocervix (58). The response to bacterial challenge within the reproductive tract is likely due to the specific microenvironment encountered. Our laboratory and others have previously established cell lines derived from the epithelia of the ectocervix and endocervix as a model system to study N. gonorrhoeae proinflammatory responses (19-21). These human epithelial cells exhibit the common morphological and cytochemical characteristics of primary cervical epithelial cells and also differ significantly from HeLa cells (21). Importantly, these cells represent the first susceptible host cell types encountered during natural infection with N. gonorrhoeae in females. We demonstrate here that N. gonorrhoeae stimulation of endocervical epithelial cells (End/E6E7 cells) protects these cells from apoptosis. This response was specific to cells incubated with live bacteria and in direct association with N. gonorrhoeae and correlated with induction of the host antiapoptotic protein cIAP-2.

MATERIALS AND METHODS

Cell culture and bacterial strains.

An endocervical epithelial (End/E6E7) cell line (21) was maintained at 37°C in a 5% CO₂ incubator in keratinocyte serum-free medium (KSFM) (Gibco-Invitrogen, Grand Island, NY) supplemented with 50 μg of bovine pituitary extract per ml, 0.1 ng/ml of epidermal growth factor, and 0.4 mM CaCl₂. Penicillin and streptomycin (Cellgro-Mediatech, Inc., Herndon, VA) were used at concentrations of 100 U/ml and 100 mg/ml, respectively. N. gonorrhoeae strain FA1090B (obtained from J. G. Cannon, University of North Carolina School of Medicine, Chapel Hill) expresses the OpaB adhesin protein in the absence of all other Opa proteins (23), thereby eliminating any variations of the invasion process due to Opa phase variation. For confocal microscopy studies, green fluorescent protein (GFP)-expressing N. gonorrhoeae strain F62 (F62-GFP) was utilized (20). Prior to infection, bacteria were grown overnight on gonococcal medium base agar plates at 37°C in a 5% CO₂ incubator. Both FA1090B and F62-GFP were piliated, as determined by colony morphology.

Infection of cells and induction of apoptosis.

End/E6E7 cells were seeded in six-well tissue culture plates (10⁵ cells/well) and allowed to grow to confluence (10⁶ cells/well) in antibiotic-free KSFM. An overnight culture of N. gonorrhoeae was used to make a bacterial cell suspension of approximately 5 × 10⁷ CFU/ml in KSFM by optimizing the optical density at 660 nm to 0.030 to 0.035. A 2-ml volume of this suspension was added to the cells to obtain a multiplicity of infection (MOI) of 100. Appropriate dilutions were made to obtain MOIs of 10 and 1. The cocultures were maintained at 37°C in a 5% CO₂ incubator. At 2 h postinfection, the cells were washed three times with phosphate-buffered saline (PBS; Cellgro-Mediatech, Inc., Herndon, VA). Medium with gentamicin (50 μg/ml) was added to each well, and the cultures were further incubated for 1 h. The cells were washed and incubated with medium without antibiotics for the remainder of 48 h. For induction of apoptosis, 1 μM staurosporine (STS) (Sigma, St. Louis, MO) was added at 24 h postinfection and incubated overnight for approximately 20 to 24 h. At various times postinfection, intracellular and extracellular viable bacteria were measured by counting CFU in cell lysates. Briefly, supernatants from selected wells were collected, and infected cells were washed with PBS. A 1.0-ml volume of PBS supplemented with 0.5 mM EDTA was added to the cell monolayer, scraped off with a sterile tissue culture scraper to facilitate lysis, and collected. Dilutions of cell lysates and extracellular supernatants were plated onto gonococcal medium base agar plates and incubated at 37°C in a 5% CO₂ incubator for ∼48 h, and bacterial CFU were counted to determine the number of viable bacteria present.

Confocal microscopy.

A sterile coverslip was placed into each six-well tissue culture plate prior to seeding of End/E6E7 cells. The cells were allowed to reach 90 to 100% confluence before infection with N. gonorrhoeae strain F62-GFP at an MOI of 100. The plasma membrane was labeled using DiI (CellTracker CM DiI; Molecular Probes, Eugene, OR) according to the manufacturer's protocol. The stained monolayers were then fixed with a 3.7% formaldehyde solution in PBS for 10 min at room temperature. Finally, the coverslips were washed three times with PBS, air dried, mounted in Vectashield mounting medium for fluorescence (Vector Laboratories, Burlingame, CA), and stored in the dark at 4°C. Reading for GFP and DiI was performed at 485 and 570 nm, respectively, using a Zeiss LSM 510 confocal laser scanning microscope.

Caspase-3 activity assay.

Cell culture supernatants were collected, followed by addition of End/E6E7 cells collected by trypsin treatment. Caspase-3 activity was determined with an EnzChek caspase-3 assay kit no. 1 (Molecular Probes, Eugene, OR), in which a fluorogenic substrate, Z-DEVD-AMC, becomes fluorescent upon cleavage by active caspase-3. Briefly, cells were lysed for 5 min in a dry ice-ethanol bath, thawed, and centrifuged at 5,000 rpm for 5 min to pellet cell debris. Supernatants were transferred to a 96-well plate and incubated with Z-DEVD-AMC for 45 min. The plate was read in a fluorescence microplate reader, using an excitation wavelength of 365 nm and emission detection at 465 nm, and data were collected using Magellan software. As a positive control for the assay, recombinant caspase-3 was used.

Determination of DNA degradation.

Cellular DNA content was determined by staining with the fluorescent DNA-intercalating dye propidium iodide (PI; Sigma, St. Louis, MO). Cell culture supernatants were collected, followed by the addition of cells collected by trypsin treatment. Samples were centrifuged at 1,600 rpm for 5 min and washed once in PBS plus 2% fetal bovine serum (FBS). Cells were resuspended in 300 μl of PBS-2% FBS, permeabilized with 700 μl of ethanol for 15 min at 4°C, centrifuged and washed again, and then stained with 300 μl of PI staining solution (PBS-2% FBS, 50 μg/ml PI, 20 mg/ml RNase A [Sigma, St. Louis, MO]) at 4°C for 15 min. Cells were washed once, resuspended in 300 μl of PBS-2% FBS, and analyzed by flow cytometry using a FACScan flow cytometer and Cell Quest acquisition software (Becton Dickinson, Mountain View, CA), with gating to exclude cell debris associated with necrosis and nonapoptotic cells. Data were analyzed using WinMDI or Cell Quest software.

Mitochondrial membrane potential assay.

Rhodamine 123 (Molecular Probes, Eugene, OR), a fluorescent mitochondrial membrane potential dye, was incubated with cell culture samples at a concentration of 1 μM for 30 min at 37°C. The cells were then collected as described above, resuspended in 300 μl PBS plus 2% FBS, and immediately analyzed by flow cytometry.

Treatment of End/E6E7 cells with porin.

Porin type PIB from N. gonorrhoeae strain FA1090B was purified according to the methods of Massari et al. (44). The porin from Neisseria meningitidis, PorB, was used as a positive control. Purified porin (PorB and PIB) was incubated with either HeLa or End/E6E7 cells at a concentration of 10 μg/ml for 24 h at 37°C prior to the addition of STS.

NF-κB activation assay.

End/E6E7 cells were incubated with N. gonorrhoeae for 1 h, and nuclear extracts were obtained using a nuclear extraction kit (Active Motif, Carlsbad, CA) according to the manufacturer's specifications. NF-κB translocation was measured in the nuclear extracts by use of a Trans-AM NF-κB family kit (Active Motif), an enzyme-linked immunosorbent assay (ELISA)-based assay. Briefly, a 96-well plate was coated with DNA oligonucleotides containing an NF-κB consensus sequence (5′-GGGACTTTCC-3′). A 2-μg sample of total nuclear proteins was added to the plate to allow for NF-κB subunits to bind the DNA. As a control, excess oligonucleotides containing the wild-type NF-κB consensus sequence were incubated with nuclear extracts to demonstrate competitive binding. In addition, an oligonucleotide with a mutated sequence was also used to demonstrate the specificity of binding.

Gene expression profiling.

Quantitative reverse transcription-PCR (RT-PCR) was performed on uninfected End/E6E7 cells, cells infected for 8 h with live bacteria, or cells incubated for 8 h with heat-killed bacteria. Briefly, RNAs were isolated and DNase treated using a Qiagen RNeasy kit (Qiagen, Valencia, CA) and then used to synthesize cDNAs by use of an RT² PCR array first-strand kit (Super Array Bioscience, Frederick, MD). The cDNAs were added to a 96-well plate from the RT² Profiler PCR array system (human apoptosis PCR array; Super Array Bioscience), in which 84 wells contain individual primer sets specific to different genes involved in apoptosis and the remaining wells serve as genomic DNA contamination controls, RT controls, and positive PCR controls. Quantitative RT-PCR was performed using an ABI Prism 7000 sequence detection system (Applied Biosystems, Foster City, CA). Data analysis was performed using the ΔΔC_T method, with the housekeeping gene β2-microglobulin (β2 M) as the normalization factor. Changes in gene expression were determined by comparing gene transcription levels in experimental samples to those in uninfected control samples.

Western blot analysis.

Whole-cell lysates were generated by suspension of cells in lysis buffer containing 50 mM Tris-HCl, pH 7.4, 1% NP-40, 0.25% sodium deoxycholate, 150 mM NaCl, and 1 mM EGTA supplemented with a protease inhibitor cocktail (Sigma, St. Louis, MO). Total protein concentration was determined using a bicinchoninic acid kit for protein determination (Sigma-Aldrich, St. Louis, MO), and sodium dodecyl sulfate-polyacrylamide gel electrophoresis was carried out using 20 μg of total protein. Western blots were performed using primary antibodies purchased from Abcam (Cambridge, MA) and diluted in 5% bovine serum albumin in Tris-buffered saline plus Tween 20 according to the manufacturer's instructions, followed by incubation with the secondary antibody, goat anti-rabbit immunoglobulin G conjugated to horseradish peroxidase (Cell Signaling Technology, Danvers, MA). Immunoreactive bands were detected using chemiluminescent reagents (ECL Plus Western blotting detection system; GE Healthcare, Piscataway, NJ) and then exposed to film (Thermo Scientific, Rockford, IL).

RESULTS

N. gonorrhoeae infection prevents STS-induced apoptosis in End/E6E7 cells.

To determine the susceptibility of End/E6E7 cells to apoptosis following infection with N. gonorrhoeae, a variety of markers of apoptosis were examined. First, End/E6E7 cells were incubated with live N. gonorrhoeae FA1090B (23) at various MOIs (100, 10, and 1) for 2 h, followed by a 1-h gentamicin treatment to kill extracellular bacteria and to allow for replication of intracellular bacteria. N. gonorrhoeae FA1090B expresses the outer membrane adhesins pili and Opa, which typically undergo phase-variable expression, and was chosen due to its constitutive expression of OpaB, which binds to CEACAM receptors expressed on cervical epithelial cells and facilitates invasion (5, 15, 23). Viable bacteria were recovered from cell cultures infected for up to 48 h, and phase microscopic analysis determined that infection with N. gonorrhoeae FA1090B for up to 48 h did not induce cell morphology changes associated with apoptotic cell death (data not shown).

STS, a protein kinase C inhibitor that induces apoptosis via the intrinsic pathway, was used to induce apoptosis in both infected and uninfected cell cultures. Apoptosis in response to STS treatment was first examined by measuring mitochondrial depolarization, an upstream marker of apoptosis characteristic of the intrinsic pathway (27), and caspase-3 activation, a marker common to both the intrinsic and extrinsic apoptotic pathways. End/E6E7 cells were infected for 24 h as described above or left uninfected prior to the addition of STS for an additional 24 h to induce apoptosis, followed by staining with rhodamine 123, a fluorescent mitochondrial membrane potential dye. Infection with N. gonorrhoeae FA1090B at an MOI as high as 100 for 48 h did not induce significant apoptotic mitochondrial depolarization compared to that in uninfected cells (Fig. 1A). In addition, infected cells exhibited resistance to STS-induced mitochondrial depolarization (Fig. 1A).

FIG. 1.

N. gonorrhoeae infection prevents STS-induced mitochondrial depolarization and caspase-3 activation in End/E6E7 cells. (A) Mitochondrial membrane depolarization was determined by staining with rhodamine 123 and fluorescence-activated cell sorter analysis. The shaded gray histogram represents uninfected cells, the gray line represents cells infected with N. gonorrhoeae at an MOI of 100 for 48 h, the black line represents uninfected cells with the addition of 1 μM STS at 24 h, and the dashed line represents cells infected with N. gonorrhoeae at an MOI of 100 for a total of 48 h, with the addition of STS at 24 h. Results are representative of three experiments performed in duplicate. (B) Caspase-3 activation was determined in End/E6E7 cells following infection with N. gonorrhoeae FA1090B at various MOIs (100, 10, and 1), as indicated, for 48 h. STS (1 μM) was added to the cultures at 24 h postinfection, where indicated. Cell lysates were tested for caspase-3 activity by incubation with a fluorogenic substrate, and fluorescence units were determined by excitation at 365 nm and emission detection at 465 nm (indicated on the y axis). The data represent combined means from three individual experiments performed in duplicate; error bars indicate the standard errors of the means.

Next, caspase-3 activity was measured in cell lysates by incubation with a fluorogenic substrate. As shown in Fig. 1B, STS-induced caspase-3 activation was reduced in End/E6E7 cells infected with N. gonorrhoeae, in a dose-dependent manner (Fig. 1B). Infection at higher MOIs (100 and 10) prevented the induction of caspase-3 activity following STS treatment, while cells infected at an MOI of 1 had an increased level of caspase-3 activation which was comparable to that in uninfected cells treated with STS. As expected, we detected viable N. gonorrhoeae organisms in association with End/E6E7 cells throughout the incubation period (data not shown). These results suggest that a distinct number of bacteria may be required to protect cells from induction of apoptosis.

We next examined a marker of apoptosis downstream of caspase activation, i.e., DNA degradation (34). Cells were incubated with N. gonorrhoeae FA1090B, and apoptosis was induced with STS as described above. DNA degradation was determined by fluorescence staining with PI followed by flow cytometry to analyze fluorescence intensity, which diminishes upon DNA breakdown. Infection of End/E6E7 cells with N. gonorrhoeae did not induce DNA degradation per se, as the percentage of cells containing hypodiploid DNA after infection with N. gonorrhoeae at an MOI of 100 was comparable to that for uninfected control cells (Fig. 2A). Furthermore, following induction of apoptosis, we observed a decrease in the percentage of cells undergoing apoptosis in N. gonorrhoeae-infected cells compared to cells treated with STS alone (Fig. 2B).

FIG. 2.

N. gonorrhoeae infection protects End/E6E7 cells from STS-induced DNA degradation. End/E6E7 cells were left uninfected or infected with N. gonorrhoeae FA1090B at an MOI of 100 for 48 h (A), and some infections were followed by the addition of 1 μM STS at 24 h (B). Hypodiploid DNA content, indicative of DNA degradation, was examined by PI staining and fluorescence-activated cell sorter analysis. The percentage of cells with hypodiploid DNA is indicated in each histogram. Data are representative of three duplicate experiments.

Apoptosis is prevented in End/E6E7 cells directly associated with N. gonorrhoeae.

To address whether the antiapoptotic effect of N. gonorrhoeae infection was due to absolute numbers of infected cells within the cell cultures, we first examined the extent of intracellular bacteria by confocal microscopy. To identify individual infected cells within a population, we utilized a GFP-expressing strain of N. gonorrhoeae (strain F62-GFP) (20) and infected cultures at an MOI of 100 as described above. As shown in Fig. 3A, confocal microscopy revealed that only a portion of the End/E6E7 cells in an infected culture were associated with bacteria following 24 h of infection. To determine if only those cells in association with N. gonorrhoeae were resistant to apoptosis or if there was a broader effect extending to all cells in the culture, we analyzed the DNA content of GFP⁺ or GFP⁻ infected End/E6E7 cells by flow cytometry. We found that while the GFP⁺ population was protected from apoptosis, the GFP⁻ population (uninfected cells within the infected cell sample) was not protected from apoptosis (Fig. 3B). These results suggest that the antiapoptotic effect induced by N. gonorrhoeae requires a direct interaction of the bacterium with target cells.

FIG. 3.

The antiapoptotic response of N. gonorrhoeae-stimulated End/E6E7 cells is specific to cells in association with bacteria. (A) Confocal microscopy was performed on End/E6E7 cell cultures infected with N. gonorrhoeae F62-GFP at an MOI of 100 for 24 h and stained with a red fluorescent plasma membrane marker, DiI. The image represents a 1-μm z-stack section (6 of 12). Bar = 10 μm. (B) EndE6/E7 cells were infected with N. gonorrhoeae F62-GFP at an MOI of 100 for 24 h, followed by the addition of 1 μM STS for an additional 22 h, as indicated. Uninfected cells were used as a control, and PI staining was used to determine DNA content. Flow cytometry was performed by gating on GFP⁺ cells from the infected cell cultures. The GFP⁻ population was also assessed for PI fluorescence by analysis of a “dead gate” following flow cytometry. Apoptosis is expressed on the y axis as the percentage of total cells containing hypodiploid DNA. Samples not treated with STS had hypodiploid DNA levels of <3% (data not shown). Results represent the means for triplicate samples from two individual experiments. Error bars represent standard deviations.

Antiapoptotic effect is mediated by live N. gonorrhoeae.

To determine if protection from apoptosis via a direct cell interaction with N. gonorrhoeae requires live bacteria, the effect of heat-killed or antibiotic-killed N. gonorrhoeae on End/E6E7 cells was examined. Cells were infected with live bacteria (MOI, 100) or incubated with heat-killed bacteria prior to the addition of STS, and the mitochondrial membrane potential was examined as a marker of apoptosis. As expected, End/E6E7 cells infected with live N. gonorrhoeae at an MOI of 100 were protected from STS-induced mitochondrial depolarization (Fig. 4A). However heat-killed bacteria were unable to prevent mitochondrial depolarization (Fig. 4B).

FIG. 4.

The antiapoptotic response of N. gonorrhoeae-stimulated End/E6E7 cells is dependent on live bacteria. N. gonorrhoeae FA1090B was incubated with End/E6E7 cells at an MOI of 100 for 24 h prior to the addition of 1 μM STS for an additional 22 h. Mitochondrial membrane potential was measured by staining with rhodamine (R123) followed by analysis of fluorescence intensity by flow cytometry. (A) Infection with live N. gonorrhoeae (GC). (B) Infection with heat-killed N. gonorrhoeae. (C) Prolonged gentamicin treatment during infection. Shaded gray histograms represent uninfected cells (plus gentamicin in panel C), thin black lines represent uninfected cells treated with STS (plus gentamicin in panel C), thick black lines represent cells incubated with N. gonorrhoeae (live, heat killed, or gentamicin treated), and dashed lines represent cells incubated with N. gonorrhoeae and treated with STS. Results are representative of three experiments performed in triplicate.

Since heat killing N. gonorrhoeae could damage bacterial surface components involved in the modulation of apoptosis (51, 52), an alternative method of antibiotic killing was examined. End/E6E7 cells were infected with live N. gonorrhoeae organisms for 2 h, incubated with a standard concentration of gentamicin (50 μg/ml) for 1 h to kill extracellular bacteria, and then cultured in the presence of low levels of gentamicin for the remainder of the 48-h experiment. This antibiotic treatment resulted in a time-dependent decrease of viable bacteria, as determined by declining numbers of CFU over time (data not shown). Although extracellular and intracellular bacteria were killed after 12 h of incubation, this treatment had no consequences on End/E6E7 cell survival (Fig. 4C). However, in agreement with our results demonstrating that only actively infected cells are protected from cellular apoptosis, a prolonged antibiotic treatment of 22 h prior to the addition of STS abrogated the antiapoptotic effect of N. gonorrhoeae (Fig. 4C). These results indicate that protection from apoptosis can be induced only by live N. gonorrhoeae.

Several reports suggest that the neisserial outer membrane protein porin protects cells from apoptosis (6, 42, 43). We tested the effect of incubation of gonococcal PIB purified from N. gonorrhoeae FA1090B on End/E6E7 cells and found that PIB induced some degree of protection against STS-induced apoptosis (data not shown). The extent of protection induced by purified PIB appeared lower than that observed with whole live bacteria, but the amount of porin used was approximately equivalent to the amount of porin which would be found in a culture of live bacteria (MOI, 100). This suggests that PIB partially contributes to the N. gonorrhoeae-induced antiapoptotic effect in End/E6E7 cells.

Contributions of cellular antiapoptotic genes and proteins in N. gonorrhoeae-infected cells.

Several reports on bacterial infection of human cells have demonstrated that activation of the transcription factor NF-κB leads to upregulation of antiapoptotic genes (8, 26, 40, 55, 56). We thus measured NF-κB nuclear translocation in End/E6E7 cells following stimulation with N. gonorrhoeae. Using an ELISA-based system to measure NF-κB nuclear translocation by subunit binding of oligonucleotides containing the NF-κB consensus sequence, we determined that the canonical pathway of NF-κB (p65 and p50 subunits) was activated in response to incubation with N. gonorrhoeae strain FA1090B for 1 h. Figure 5A and B depict nuclear translocation of the NF-κB p65 and p50 subunits, respectively, as measured spectrophotometrically in nuclear extracts of End/E6E7 cells obtained from untreated, uninfected cultures or cultures infected with N. gonorrhoeae. As expected, stimulation of End/E6E7 cells with tumor necrosis factor alpha (TNF-α) also resulted in nuclear translocation of the NF-κB p65 and p50 subunits (Fig. 5). We also observed that translocation of the NF-κB p65 and p50 subunits was sustained through 24 h (data not shown).

FIG. 5.

The canonical pathway of NF-κB is activated and sustained upon stimulation of End/E6E7 cells with N. gonorrhoeae FA1090B. End/E6E7 cells were left untreated and uninfected, infected with gonococcal (GC) strain FA1090B at an MOI of 100, or treated with 25 ng/ml TNF-α for 1 h. A Trans-AM ELISA kit was used to detect NF-κB nuclear translocation, and the colorimetric reaction was read as the optical density (OD) at 450 nm. Nuclear proteins were incubated alone (gray bars), with an oligonucleotide containing a mutated NF-κB consensus sequence (black bars), or with an oligonucleotide containing a wild-type NF-κB consensus sequence for competitive binding (white bars). The p65 (A) and p50 (B) subunits were detected in nuclear extracts from 1-h samples following treatment with gonococci or TNF. Untreated, uninfected cells showed a basal nuclear level of each subunit. Error bars represent standard deviations for triplicate experimental samples.

To determine if NF-κB activation induced a cellular antiapoptotic response, gene transcription profiles for proteins involved in apoptosis were examined using real-time quantitative RT-PCR. In total, we examined the expression of 84 genes grouped into 12 protein families that included key ligands, receptors, intracellular modulators, and transcription factors involved in apoptosis (see Table S1 in the supplemental material). We compared transcription in uninfected End/E6E7 cells and End/E6E7 cells following an 8-h incubation with either live N. gonorrhoeae or heat-killed bacteria. Live N. gonorrhoeae infection induced a twofold or greater change in expression for approximately 30% of the genes examined (see Table S2 in the supplemental material). Among these genes, the NF-κB-regulated antiapoptotic genes bfl-1 (upregulated 44-fold), cIAP-2 (upregulated 26-fold), and CFLAR/c-FLIP (upregulated 5-fold) were the most highly induced (Fig. 6A; see Table S2 in the supplemental material).

FIG. 6.

Expression of host genes and proteins following infection of End/E6E7 cells with N. gonorrhoeae. Gene expression was measured for representative genes associated with the inhibition of apoptosis (A), the induction of apoptosis (B), and inflammation (C). Transcript levels are represented as changes compared to uninfected control RNA. Genes were normalized to the β-2 microglobulin (b-2-M) housekeeping control. Results represent the means for four experimental samples. *, P < 0.05; **, P < 0.005 (Student's t test). Black bars, infection with live bacteria for 8 h; white bars, treatment with heat-killed bacteria for 8 h. (D) End/E6E7 cells were infected with N. gonorrhoeae FA1090B at an MOI of 100, and at 8 h and 24 h postinfection, whole-cell lysates were prepared. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis was performed using 20 μg of total protein, followed by Western blot analysis using anti-Bfl-1, -c-FLIP, -cIAP-2, and -β-2-microglobulin antibodies. GC, N. gonorrhoeae.

Heat-killed N. gonorrhoeae induced the expression of a similar subset of genes, but to a much lesser extent (Fig. 6A; see Table S2 in the supplemental material). A modest increase in a subset of proapoptotic genes in response to stimulation with live bacteria was also detected, including the casp10 gene (fourfold) and the casp7 gene (threefold) (Fig. 6B; see Table S2 in the supplemental material). Although the expression of caspase genes was increased, this did not correlate with a proapoptotic effect. Interestingly, TNFSF10, which encodes the TNF-related apoptosis-inducing ligand (TRAIL), was also upregulated in End/E6E7 cells incubated with both live and heat-killed N. gonorrhoeae (see Table S2 in the supplemental material); however, a TRAIL decoy receptor gene (TNFRSF11B) which inhibits TRAIL-induced apoptosis was significantly upregulated only by live bacteria (Fig. 6C; see Table S2 in the supplemental material). Finally, the induction of several inflammatory genes was also detected (Fig. 6C; see Table S2 in the supplemental material), in agreement with previous data showing TNF-α production induced by N. gonorrhoeae infection in various cell types (41, 46, 54, 60).

To correlate gene expression with protein expression, Western blot analysis of whole-cell lysates from N. gonorrhoeae-infected cultures was performed to investigate corresponding levels of these antiapoptotic proteins. The bfl-1 gene was highly upregulated following stimulation with N. gonorrhoeae (Fig. 6), but we did not observe an increase in Bfl-1 protein level by Western blot analysis of End/E6E7 cells obtained either 8 h or 24 h after stimulation with N. gonorrhoeae (Fig. 6D), as well as at earlier time points (data not shown). Similar to the results observed with Bfl-1, we did not detect an increase in c-FLIP protein level in End/E6E7 cells obtained following stimulation with N. gonorrhoeae (Fig. 6D), suggesting that the contribution of these proteins to the antiapoptotic effect of N. gonorrhoeae may be limited. In addition, there is the possibility that these genes may be regulated posttranscriptionally or posttranslationally, which may account for the discrepancies between gene upregulation and protein levels. However, increased levels of cIAP-2, which functions to block caspase activity, were detected in End/E6E7 cells obtained following stimulation with N. gonorrhoeae for 8 h, and these levels were sustained 24 h after stimulation with N. gonorrhoeae (Fig. 6D). These results indicate that cIAP-2 is among the cellular factors induced by N. gonorrhoeae infection of human End/E6E7 cells.

DISCUSSION

Entry and survival of N. gonorrhoeae in epithelial cells play an important role in the survival and spread of this organism. During infection in women, cells from the ectocervix and endocervix are among the first host cells encountered by this pathogen. Because the female genital tract is such a unique anatomical site, there is a need for female-specific host cells to study such sexually transmitted disease pathogens. The cell lines used in this study were created in order to provide a better model system for studying the response of cervical epithelial cells to pathogens, and they exhibit the common morphological and cytochemical characteristics of primary cervical epithelial cells (21). Furthermore, these cells were shown to differ significantly from HeLa adenocarcinoma cells (21), making them an excellent model for the studies described here. In this study, we demonstrate that live N. gonorrhoeae infection of cells derived from the lower female genital tract induces an antiapoptotic effect. We determined that infection of End/E6E7 cells with N. gonorrhoeae prevented specific apoptotic events induced via the intrinsic apoptosis pathway. In particular, mitochondrial depolarization, caspase-3 activation, and DNA degradation induced by STS were either reduced or prevented by infection of cells with N. gonorrhoeae. We determined that the antiapoptotic effect induced by N. gonorrhoeae was specific to cells associated with bacteria. Some partial protection from apoptosis was observed at low MOIs, possibly due to smaller numbers of cells in direct association with bacteria. Importantly, we demonstrated that heat-killed and antibiotic-killed bacteria failed to induce protection from STS-induced apoptosis. We also determined that the antiapoptotic response induced by live N. gonorrhoeae may be due in part to gonococcal PIB.

Prevention of apoptosis may be a general mechanism of immune evasion that N. gonorrhoeae employs during infection, as similar findings have been reported for other cell types. Morales et al. (49) reported protection against apoptosis in N. gonorrhoeae-infected fallopian tube epithelial cells and showed that infection with high MOIs of N. gonorrhoeae strain P9-17 protected cells from TNF-α-induced apoptosis. These studies support our observations and suggest that protection from apoptosis is not an artifact of immortalized cell lines. However, fallopian tubes are not the primary site for N. gonorrhoeae infection. Hence, our studies demonstrate that the antiapoptotic effect may play a role in the initial infection of host cells upon bacterial invasion of the lower female genital tract. N. gonorrhoeae has also been reported to inhibit apoptosis in T84 colonic epidermoid cells (31, 32), ME-180 cervical carcinoma cells (51), polymorphonuclear leukocytes (64), and both primary and transformed human male urethral epithelial cells (6, 7).

Antiapoptotic responses of host cells are often attributed to cellular activation of the transcription factor NF-κB, a known regulator of antiapoptotic genes (8, 55). We confirmed in this study that live N. gonorrhoeae infection of End/E6E7 cells induced NF-κB nuclear translocation via the canonical pathway of NF-κB (p65 and p50 subunits). Furthermore, the NF-κB-regulated antiapoptotic genes bfl-1, cIAP-2, and c-FLIP were significantly upregulated in these cells in response to live infection. In general, the response of End/E6E7 cells to heat-killed bacteria was not as pronounced as that observed with live bacteria. This could be due to the lack of internalization of the heat-killed gonococcal preparation. Studies with human T84 cells have demonstrated that live N. gonorrhoeae induces cytoprotective effects via gonococcal pilus retraction (31, 32). For male urethral cells and polymorphonuclear leukocytes, live N. gonorrhoeae infection has also been reported to induce the expression of host cell antiapoptotic genes (7, 64). Thus, the ability of N. gonorrhoeae to induce the expression of genes encompassing multiple antiapoptotic pathways may represent a common cellular response to live N. gonorrhoeae infection of host cells.

It has been hypothesized that the antiapoptotic host cell response to infection is a host defense that allows for complete and sustained cellular innate inflammatory responses (30, 69). Of particular interest was the identification in this study of several genes related to innate immunity that were regulated in response to N. gonorrhoeae. Ripk2, also known as Rip2 or Rick, is an adaptor molecule that was initially identified as a component of the TNF receptor 1 signaling complex that participates in the activation of NF-κB (45). Rip2 also has an important role in innate immune recognition via nucleotide-binding oligomerization domain (Nod) proteins, Nod1 and Nod2, which recognize bacterial pathogen-associated molecular patterns (14). Therefore, it seems logical that Rip2 would be activated in N. gonorrhoeae-infected cervical epithelial cells. In addition, casp1 was upregulated >4-fold in response to live N. gonorrhoeae, albeit without statistical significance. Although the caspase-1 protein shares homology with other apoptotic caspases, caspase-1 actually has a proinflammatory function by participating in the production of the cytokine interleukin-1β. This also correlates with expected results, as interleukin-1β production has been demonstrated in response to N. gonorrhoeae infection both in vivo and in vitro (20, 29, 54, 60).

One final observation from this study, regarding levels of antiapoptotic protein expression in End/E6E7 cells, deserves particular attention. We found that the levels of Bfl-1 and c-FLIP proteins, expressed in End/E6E7 cells under normal growth conditions, were not altered following N. gonorrhoeae infection, although gene transcription increased. It is not known whether this could be due to posttranscriptional or posttranslational modifications, but this observation possibly suggests a nonessential role for these antiapoptotic proteins in the effect of N. gonorrhoeae. However, we observed that cIAP-2 protein levels were increased in End/E6E7 cells incubated with N. gonorrhoeae. The IAP family of proteins plays multiple roles, including protecting cells from apoptotic stimuli and modulating innate immunity (13, 25). In Drosophila, dIAP2 is required for the antimicrobial function of the Imd pathway (25, 33, 36, 39), which is activated by peptidoglycan recognition proteins. Innate immune signaling pathways are well conserved from Drosophila to humans, suggesting that IAP proteins may also play a role in mammalian innate immunity. Indeed, cIAP-2 was demonstrated to exacerbate endotoxic shock in mice (13). Furthermore, the X-linked IAP was recently reported to be required for innate immune control of Listeria monocytogenes infection (4). Another IAP family member, NAIP5, was found to mediate caspase-1 activation in response to cytosolic bacterial flagellin. Thus, the IAPs may function as cytosolic surveillance systems that play roles in both inflammation and protection from cell death. While these recent studies point to a pivotal role of IAPs in the regulation of innate immune responses to bacterial pathogens, a specific role for cIAP-2 in the antiapoptotic (and inflammatory) response to N. gonorrhoeae remains to be determined.

In conclusion, we have demonstrated in this study that endocervical epithelial cells, which are specific targets of N. gonorrhoeae infection in female hosts, are protected from STS-induced apoptosis. This response was specific to End/E6E7 cells incubated with live bacteria and in direct association with N. gonorrhoeae and correlated with increased expression of selected host antiapoptotic genes. Current studies are aimed at defining the role of cIAP-2 in antiapoptotic and inflammatory responses of endocervical epithelial cells to intracellular gonococci. Importantly, the cellular antiapoptotic response of these cells to N. gonorrhoeae infection may also be an important innate immune defense mechanism.