Neisseria gonorrhoeae epithelial cell interaction leads to the activation of the transcription factors nuclear factor kappaB and activator protein 1 and the induction of inflammatory cytokines

Abstract

We have studied the effect of human bacterial pathogen Neisseria gonorrhoeae (Ngo) on the activation of nuclear factor (NF)-kappaB and the transcriptional activation of inflammatory cytokine genes upon infection of epithelial cells. During the course of infection, Ngo, the etiologic agent of gonorrhea, adheres to and penetrates mucosal epithelial cells. In vivo, localized gonococcal infections are often associated with a massive inflammatory response. We observed upregulation of several inflammatory cytokine messenger RNAs (mRNAs) and the release of the proteins in Ngo-infected epithelial cells. Moreover, infection with Ngo induced the formation of a NF-kappaB DNA-protein complex and, with a delay in time, the activation of activator protein 1, whereas basic leucine zipper transcription factors binding to the cAMP-responsive element or CAAT/enhancer-binding protein DNA-binding sites were not activated. In supershift assays using NF-kappaB-specific antibodies, we identified a NF-kappaB p50/p65 heterodimer. The NF-kappaB complex was formed within 10 min after infection and decreased 90 min after infection. Synthesis of tumor necrosis factor alpha and interluekin (IL)-1beta occurred at later times and therefore did not account for NF-kappaB activation. An analysis of transiently transfected IL-6 promoter deletion constructs suggests that NF-kappaB plays a crucial role for the transcriptional activation of the IL-6 promoter upon Ngo infection. Inactivation of NF-kappaB conferred by the protease inhibitor N-tosyl--phenylalanine chloromethyl ketone inhibited mRNA upregulation of most, but not all, studied cyctokine genes. Activation of NF-kappaB and cytokine mRNA upregulation also occur in Ngo-infected epithelial cells that were treated with cytochalasin D, indicating an extracellular signaling induced before invasion.

Figure 1

Ngo infection induces proinflammatory cytokines in epithelial cells. (A) Shown is an analysis of cytokine mRNA levels in ME180 cells in response to Ngo infection (the invasive Opa⁺ strain, the piliated, noninvasive P⁺ strain, and the P⁻ Opa⁻ control strain) by duplex RT-PCR. Total RNA was isolated at the indicated time points after infection and reverse transcribed into cDNA, and cytokine mRNAs as well as β-actin mRNA were semiquantitated by several cycles of PCR using cytokine-specific primers so that products were below the saturation stage of amplification. Equal RNA was amplified for each sample within an infection kinetic as indicated by the internal β-actin amount. DNA products were separated by electrophoresis on a agarose gel and visualized with ethidium bromide. Shown is an experiment representative of at least three. *, β-actin. Not shown: IL-2, IL-3, IL-4, IL-5, IL-10, IL-13, I-309, and IFN-γ RT-PCR reactions that recognized no transcripts. (B) The secretion of the cytokines TNF-α, GM-CSF, and IL-8 was assayed in ELISA systems 3 h and 6 h after infection (p.i.). Values are means, the standard errors of the means are representative for three experiments.

Figure 1

Ngo infection induces proinflammatory cytokines in epithelial cells. (A) Shown is an analysis of cytokine mRNA levels in ME180 cells in response to Ngo infection (the invasive Opa⁺ strain, the piliated, noninvasive P⁺ strain, and the P⁻ Opa⁻ control strain) by duplex RT-PCR. Total RNA was isolated at the indicated time points after infection and reverse transcribed into cDNA, and cytokine mRNAs as well as β-actin mRNA were semiquantitated by several cycles of PCR using cytokine-specific primers so that products were below the saturation stage of amplification. Equal RNA was amplified for each sample within an infection kinetic as indicated by the internal β-actin amount. DNA products were separated by electrophoresis on a agarose gel and visualized with ethidium bromide. Shown is an experiment representative of at least three. *, β-actin. Not shown: IL-2, IL-3, IL-4, IL-5, IL-10, IL-13, I-309, and IFN-γ RT-PCR reactions that recognized no transcripts. (B) The secretion of the cytokines TNF-α, GM-CSF, and IL-8 was assayed in ELISA systems 3 h and 6 h after infection (p.i.). Values are means, the standard errors of the means are representative for three experiments.

Figure 2

The effect of Ngo infection on the activation of transcription factors. (A) HeLa cells were analyzed in a gel retardation assay using a ³²P-labeled H-2K gene NF-κB–binding site oligonucleotide. Nuclear extracts were prepared at different time points after infection with the Ngo P⁺ strain and the DNA-binding activity was analyzed (lanes 1–5). The identity of the NF-κB transcription factor was investigated by competition with increasing amounts of the unlabeled oligonucleotide (lanes 6–9). The composition of the Ngo-induced NF-κB complex was investigated by antibody supershifting and inhibition using different amounts (0.5 μl and 2 μl) of anti-p50 (lanes 10 and 11), anti-p65 (lanes 12 and 13), or anti–c-Rel antisera (lanes 14 and 15). (B) The NF-κB DNA-binding activity was assayed in response to Ngo strains, different MOIs, and different time points (lanes 1–5). As controls, HeLa cells were stimulated with TNF-α (10 ng/ml) or treated with LPS (10 μg/ml). (C) AP-1 DNA-binding activity was investigated in response to Ngo infection at different time points (lanes 1–5) using a ³²P-labeled AP-1 DNA-binding site oligonucleotide as a probe. As controls cells were stimulated with PMA (40 nM) or treated with LPS (10 μg/ml). (D) DNA-binding activity at the CRE, C/EBP, and octamer binding sites were studied in extracts from Ngo P⁺ strain–infected HeLa cells at different time points (lanes 1–5). Only sections of the autoradiograms containing the protein– DNA complexes are shown. The position of protein–DNA complexes are indicated with arrows.

Figure 2

The effect of Ngo infection on the activation of transcription factors. (A) HeLa cells were analyzed in a gel retardation assay using a ³²P-labeled H-2K gene NF-κB–binding site oligonucleotide. Nuclear extracts were prepared at different time points after infection with the Ngo P⁺ strain and the DNA-binding activity was analyzed (lanes 1–5). The identity of the NF-κB transcription factor was investigated by competition with increasing amounts of the unlabeled oligonucleotide (lanes 6–9). The composition of the Ngo-induced NF-κB complex was investigated by antibody supershifting and inhibition using different amounts (0.5 μl and 2 μl) of anti-p50 (lanes 10 and 11), anti-p65 (lanes 12 and 13), or anti–c-Rel antisera (lanes 14 and 15). (B) The NF-κB DNA-binding activity was assayed in response to Ngo strains, different MOIs, and different time points (lanes 1–5). As controls, HeLa cells were stimulated with TNF-α (10 ng/ml) or treated with LPS (10 μg/ml). (C) AP-1 DNA-binding activity was investigated in response to Ngo infection at different time points (lanes 1–5) using a ³²P-labeled AP-1 DNA-binding site oligonucleotide as a probe. As controls cells were stimulated with PMA (40 nM) or treated with LPS (10 μg/ml). (D) DNA-binding activity at the CRE, C/EBP, and octamer binding sites were studied in extracts from Ngo P⁺ strain–infected HeLa cells at different time points (lanes 1–5). Only sections of the autoradiograms containing the protein– DNA complexes are shown. The position of protein–DNA complexes are indicated with arrows.

Figure 2

The effect of Ngo infection on the activation of transcription factors. (A) HeLa cells were analyzed in a gel retardation assay using a ³²P-labeled H-2K gene NF-κB–binding site oligonucleotide. Nuclear extracts were prepared at different time points after infection with the Ngo P⁺ strain and the DNA-binding activity was analyzed (lanes 1–5). The identity of the NF-κB transcription factor was investigated by competition with increasing amounts of the unlabeled oligonucleotide (lanes 6–9). The composition of the Ngo-induced NF-κB complex was investigated by antibody supershifting and inhibition using different amounts (0.5 μl and 2 μl) of anti-p50 (lanes 10 and 11), anti-p65 (lanes 12 and 13), or anti–c-Rel antisera (lanes 14 and 15). (B) The NF-κB DNA-binding activity was assayed in response to Ngo strains, different MOIs, and different time points (lanes 1–5). As controls, HeLa cells were stimulated with TNF-α (10 ng/ml) or treated with LPS (10 μg/ml). (C) AP-1 DNA-binding activity was investigated in response to Ngo infection at different time points (lanes 1–5) using a ³²P-labeled AP-1 DNA-binding site oligonucleotide as a probe. As controls cells were stimulated with PMA (40 nM) or treated with LPS (10 μg/ml). (D) DNA-binding activity at the CRE, C/EBP, and octamer binding sites were studied in extracts from Ngo P⁺ strain–infected HeLa cells at different time points (lanes 1–5). Only sections of the autoradiograms containing the protein– DNA complexes are shown. The position of protein–DNA complexes are indicated with arrows.

Figure 2

The effect of Ngo infection on the activation of transcription factors. (A) HeLa cells were analyzed in a gel retardation assay using a ³²P-labeled H-2K gene NF-κB–binding site oligonucleotide. Nuclear extracts were prepared at different time points after infection with the Ngo P⁺ strain and the DNA-binding activity was analyzed (lanes 1–5). The identity of the NF-κB transcription factor was investigated by competition with increasing amounts of the unlabeled oligonucleotide (lanes 6–9). The composition of the Ngo-induced NF-κB complex was investigated by antibody supershifting and inhibition using different amounts (0.5 μl and 2 μl) of anti-p50 (lanes 10 and 11), anti-p65 (lanes 12 and 13), or anti–c-Rel antisera (lanes 14 and 15). (B) The NF-κB DNA-binding activity was assayed in response to Ngo strains, different MOIs, and different time points (lanes 1–5). As controls, HeLa cells were stimulated with TNF-α (10 ng/ml) or treated with LPS (10 μg/ml). (C) AP-1 DNA-binding activity was investigated in response to Ngo infection at different time points (lanes 1–5) using a ³²P-labeled AP-1 DNA-binding site oligonucleotide as a probe. As controls cells were stimulated with PMA (40 nM) or treated with LPS (10 μg/ml). (D) DNA-binding activity at the CRE, C/EBP, and octamer binding sites were studied in extracts from Ngo P⁺ strain–infected HeLa cells at different time points (lanes 1–5). Only sections of the autoradiograms containing the protein– DNA complexes are shown. The position of protein–DNA complexes are indicated with arrows.

Figure 3

Inhibition of invasion of Ngo Opa⁺ bacteria using cytochalasin D does not affect NF-κB activation. Time-dependent activation of NF-κB in response to Ngo Opa⁺ strain infection in HeLa cells (lanes 1–5) and in cells treated for 30 min before the infection with cytochalasin D (lanes 6–10) is shown. Only a section of the autoradiogram containing the protein–DNA complex is shown. The position of the NF-κB–DNA complex is indicated with an arrow.

Figure 4

Kinetics of NF-κB transactivation activity in Ngo-infected epithelial cells. Epithelial cells at 50–70% confluence were transfected with 1 μg of a luciferase expression plasmid containing four repeats of the NF-κB H-2K–binding site as enhancer element as described in Materials and Methods. Luciferase assays from whole cellular extracts were performed at the indicated time points after infection. The results are expressed as fold induction against nontreated cells. (A) HeLa cells were either infected with different Ngo strains, stimulated with TNF-α (10 ng/ml), treated with LPS (10 μg/ml), or left untreated. (B) HeLa cells were infected with Ngo P⁺ strain at a MOI of 5 and 50, or left uninfected. (C) Different epithelial cells (HeLa, HaCaT, ME180) were infected with Ngo at a MOI of 50 or left untreated. The results are expressed as fold induction against nontreated cells from each cell line. The data presented are representative of more than three separate experiments.

Figure 5

Ngo infection induces expression of IL-6 promoter–hGH constructs transfected in HeLa cells. (A, left) Schematic representation of the deleted IL-6 promoter constructs. The transcription start site is designated with a black arrow. (A, right, and B) HeLa cells were transfected by cationic liposomes with 3 μg of plasmids. Cells were maintained in medium containing 10% FCS for 24 h. 1 h before the infection, the medium was exchanged and the cells either untreated or infected with the Ngo P⁺ strain at a MOI of 100. hGH activity was assessed in culture supernatants by ELISA, and the results of three independent experiments expressed as fold induction against noninfected cells.

Figure 5

Ngo infection induces expression of IL-6 promoter–hGH constructs transfected in HeLa cells. (A, left) Schematic representation of the deleted IL-6 promoter constructs. The transcription start site is designated with a black arrow. (A, right, and B) HeLa cells were transfected by cationic liposomes with 3 μg of plasmids. Cells were maintained in medium containing 10% FCS for 24 h. 1 h before the infection, the medium was exchanged and the cells either untreated or infected with the Ngo P⁺ strain at a MOI of 100. hGH activity was assessed in culture supernatants by ELISA, and the results of three independent experiments expressed as fold induction against noninfected cells.

Figure 6

Inhibition of NF-κB activation by TPCK blocks the induction of cytokine genes in Ngo-infected epithelial cells. (A) Nuclear extracts from HeLa cells were prepared at different time points after infection (Ngo P⁺ strain), incubated with a ³²P-labeled oligonucleotide containing the NF-κB H-2K DNA-binding site, and analyzed for NF-κB activation in an EMSA (lanes 1–5). Additionally, cells were treated with the serine protease inhibitor TPCK 30 min before the infection (lanes 6–10). Only a section of the autoradiogram containing the protein– DNA complexes is shown. The position of the NF-κB–DNA complexes is indicated with an arrow. (B) Shown is an analysis of cytokine mRNA levels in HeLa cells in response to Ngo P⁺ strain infection by duplex RT-PCR either in the absence or presence of the protease inhibitor TPCK. Total RNA was isolated at the indicated time points after infection and reverse transcribed into cDNA, and cytokine mRNAs as well as β-actin mRNA were semiquantitated by several cycles of PCR using cytokine-specific primers so that products were below the saturation stage of amplification. Equal RNA was amplified for each sample within an infection kinetic as indicated by the internal β-actin amount. DNA products were separated by electrophoresis on a agarose gel and visualized with ethidium bromide. Shown is an experiment representative of at least three. *, β-actin.

Figure 6

Inhibition of NF-κB activation by TPCK blocks the induction of cytokine genes in Ngo-infected epithelial cells. (A) Nuclear extracts from HeLa cells were prepared at different time points after infection (Ngo P⁺ strain), incubated with a ³²P-labeled oligonucleotide containing the NF-κB H-2K DNA-binding site, and analyzed for NF-κB activation in an EMSA (lanes 1–5). Additionally, cells were treated with the serine protease inhibitor TPCK 30 min before the infection (lanes 6–10). Only a section of the autoradiogram containing the protein– DNA complexes is shown. The position of the NF-κB–DNA complexes is indicated with an arrow. (B) Shown is an analysis of cytokine mRNA levels in HeLa cells in response to Ngo P⁺ strain infection by duplex RT-PCR either in the absence or presence of the protease inhibitor TPCK. Total RNA was isolated at the indicated time points after infection and reverse transcribed into cDNA, and cytokine mRNAs as well as β-actin mRNA were semiquantitated by several cycles of PCR using cytokine-specific primers so that products were below the saturation stage of amplification. Equal RNA was amplified for each sample within an infection kinetic as indicated by the internal β-actin amount. DNA products were separated by electrophoresis on a agarose gel and visualized with ethidium bromide. Shown is an experiment representative of at least three. *, β-actin.