Stimulator of IFN gene is critical for induction of IFN-beta during Chlamydia muridarum infection

Abstract

Type I IFN signaling has recently been shown to be detrimental to the host during infection with Chlamydia muridarum in both mouse lung and female genital tract. However, the pattern recognition receptor and the signaling pathways involved in chlamydial-induced IFN-beta are unclear. Previous studies have demonstrated no role for TLR4 and a partial role for MyD88 in chlamydial-induced IFN-beta. In this study, we demonstrate that mouse macrophages lacking TLR3, TRIF, TLR7, or TLR9 individually or both TLR4 and MyD88, still induce IFN-beta equivalent to wild type controls, leading to the hypothesis that TLR-independent cytosolic pathogen receptor pathways are crucial for this response. Silencing nucleotide-binding oligomerization domain 1 in HeLa cells partially decreased chlamydial-induced IFN-beta. Independently, small interfering RNA-mediated knockdown of the stimulator of IFN gene (STING) protein in HeLa cells and mouse oviduct epithelial cells significantly decreased IFN-beta mRNA expression, suggesting a critical role for STING in chlamydial-induced IFN-beta induction. Conversely, silencing of mitochondria-associated antiviral signaling proteins and the Rig-I-like receptors, RIG-I, and melanoma differentiation associated protein 5, had no effect. In addition, induction of IFN-beta depended on the downstream transcription IFN regulatory factor 3, and on activation of NF-kappaB and MAPK p38. Finally, STING, an endoplasmic reticulum-resident protein, was found to localize in close proximity to the chlamydial inclusion membrane during infection. These results indicate that C. muridarum induces IFN-beta via stimulation of nucleotide-binding oligomerization domain 1 pathway, and TLR- and Rig-I-like receptor-independent pathways that require STING, culminating in activation of IFN regulatory factor 3, NF-kappaB, and p38 MAPK.

FIGURE 1

IFN-β expression is unimpaired in TLR4MyD88-DKO macrophages during chlamydial infection. IFN-β mRNA (A), IFN-β protein (B), and TNF-α mRNA (C) were quantified in WT and TLR4-MyD88 DKO peritoneal macrophages infected in vitro with C. muridarum for the indicated times or treated where noted for 8 h with either the TLR4 ligand LPS (1 μg/ml) or the TLR9 ligand CpG DNA (100 nM). Supernatants were assayed for IFN-β protein following 24 h of infection. Error bars in A and C signify the mean ± SD of samples assayed in duplicate. A representative of four independent experiments is shown. Error bars in B represent the mean ± SD from three independent experiments.

FIGURE 2

The host TLR3-TRIF pathway is dispensable for chlamydial-induced IFN-β expression. IFN-β mRNA was quantified by quantitative RT-PCR in bone marrow derived macrophages from WT controls and TLR3 KO (A) or TRIF^lps2 (B) mice infected with C. muridarum for the indicated times or treated with the TLR3 ligand poly I:C (25 μg/ml) for 6 h. Supernatants from WT and TRIF^lps2 macrophages were also assayed for IFN-β protein after 24 h of infection (C). Error bars in A and B signify the mean ±SD of samples assayed in duplicate. A representative of three independent experiments is shown. Error bars in C represent the mean ± SD from three independent experiments.

FIGURE 3

TLR7 and TLR9 are not essential for MyD88 dependent IFN-β expression during chlamydial infection. IFN-β mRNA was quantified by quantitative RT-PCR in peritoneal macrophages from WT controls and TLR9 KO (A) or TLR7 KO (B) mice infected with C. muridarum for the indicated times. Error bars signify the mean ± SD of samples assayed in duplicate. A representative of three independent experiments is shown for both panels.

FIGURE 4

Chlamydial-induced IFN-β requires IRF-3, NF-κB, and the p38 MAPK pathway. IFN-β mRNA (A) and protein (B) expression was quantified in C. muridarum-infected peritoneal macrophages isolated from WT, IRF3 KO, and IRF7 KO mice. IFN-β mRNA (C) induced by C. muridarum at 24 h postinfection was quantified in WT and IRF-3 KO macrophages using the same infection conditions as in A or alternatively pretreated for 6 h with 2 ng/ml recombinant IFN-β where indicated. IFN-β mRNA upregulation was quantified in macrophages that were pretreated with either a proteasomal inhibitor (D) or inhibitors targeting MAPK pathways individually or simultaneously (E) and then infected with C. muridarum for 24 h. Error bars for A, C, D, and E represent the mean ± SD of samples assayed in duplicate. A representative of three experiments is shown for A and E. Error bars in) represent the mean ± SD from three independent experiments. *p < 0.05; **p < 0.01.

FIGURE 5

Nod1 is required for maximal expression of IFN-β during chlamydial infection. HeLa cells were treated with siRNA targeting Nod1 (42–49% knockdown) prior to infection with C. muridarum. IFN-β (A) and IL-8 (B) upregulation at 24 h postinfection was quantified by quantitative RT-PCR. Chlamydial rs16 was quantified by quantitative RT-PCR at the same time point (B). Error bars in A and B represent the mean ± SD of samples assayed in duplicate. A representative of three experiments is shown.

FIGURE 6

IFN-β expression during chlamydial infection is mediated by STING, but not MAVS in HeLa cells. HeLa cells were treated with a non-targeting control siRNA duplex or an oligo-targeting STING (68–82% knockdown). IFN-β mRNA was quantified 4 h after transfection of the dsDNA analog poly dA:dT (A) or 24 h postinfection with C. muridarum (B). In addition, chlamydial rs16 expression (C) and IL-8 mRNA levels (D) were quantified from infected cells at 24 h postinfection. HeLa cells were treated with a nontargeting control siRNA duplex or an oligo-targeting MAVS (63% knockdown). To measure the functional implications of MAVS knockdown, IFN-β mRNA level measured 4 h after transfection of dsRNA into cells treated with control or MAVS siRNA (E). Chlamydial-induced IFN-β levels were measured at 24 h postinfection as quantified by quantitative RT-PCR (F). Error bars represent the mean ± SD of samples assayed in duplicate. A representative of at least three experiments is shown.

FIGURE 7

IFN-β expression during chlamydial infection is mediated by STING in mouse oviduct epithelial cells. STING knockdown (69–72%) was performed in mouse oviduct epithelial cells (Bm1.11). IFN-β (D) CXCL10 (E) and chlamydial rs16 (F) expression were quantified by quantified RT-PCR 24 h postinfection. Error bars for all panels represent the mean ± SD of samples assayed in duplicate. A representative of two independent experiments is shown for each.

FIGURE 8

STING colocalizes with the ER and accumulates in the vicinity of the inclusion during infection. HeLa cells were transfected with a plasmid vector expressing human flag tagged STING and infected for 20 h with C. muridarum. A, Single color confocal microscopy staining for STING (red), C. muridarum (green), nuclei (blue), and a merged picture. B, Single color staining for STING (red), ER resident protein PDI (green), nuclei (blue), and a merged picture. Arrows indicate chlamydial inclusions. C and D, Staining for endogenous STING and Sec61α (red) respectively, chlamydial inclusion (green), nuclei (blue), and merged pictures. UI, uninfected cells.

FIGURE 9

Schematic representation of pathways contributing to IFN-β induction during chlamydial infection.