Up-Regulation of the JAK/STAT1 Signal Pathway during Chlamydia trachomatis Infection¹

Chlamydia trachomatis is the most common cause of sexually transmitted infections (1, 2). Urogenital infection of C. trachomatis can lead to female pelvic inflammatory disease, tubal blockage, and infertility, whereas ocular infection causes the most preventable blindness worldwide. Chlamydia spp. are Gram-negative, obligate intracellular bacterial parasites that infect and multiply within a broad range of eukaryotic cells, including macrophages and smooth muscle, epithelial, and endothelial cells (3). Chlamydia sp. exists as either an infectious elementary body (EB)³ or a metabolically active, but noninfectious, reticulate body (RB). A typical infection cycle is initiated by EB attachment to host cells. After internalization, the EBs differentiate into RBs for replication through binary fission. Mature RBs then redifferentiate into the infectious EBs and are released 48–72 h after infection for the initiation of a new infection cycle (3, 4, 5).

Chlamydial infection of nonimmune cells produces proinflammatory factors, including IL-8 (6, 7, 8, 9) and IFNs (10, 11, 12, 13, 14, 15), which play a critical role in the disease process of Chlamydia infection (6, 11, 16). IL-8, also known as neutrophil-activating peptide-1, induces inflammatory responses and promotes angiogenesis, resulting in the tissue damage and scarring that are characteristic of chlamydial disease (6, 16). In contrast, IFN treatment inhibits Chlamydia growth by blockage of Chlamydia transformation (17, 18, 19), and mice deficient in IFN-γR develop severe disseminated infections and fail to clear genital tract infection (54). Moreover, IFN production by infected epithelial cells (10), fibroblasts (12, 13), and macrophages (15) inhibits Chlamydia infection (15), indicating the importance of IFN-mediated immunity in host clearance of Chlamydia infection.

To better understand the immune responses of mucosal epithelial cells to chlamydial infection, we examined the gene transcriptional profiles of cervical epithelial cells to Chlamydia infection. We used the newly released Affymetrix HG-U133 Plus 2.0 Gene Chip, which completely covers the human genome and allows for comprehensive studies of host response to Chlamydia infection. In addition to induction of genes for proinflammatory cytokines and chemokines, infection of HeLa 229 cells by C. trachomatis LGV2 up-regulated IFN-β and genes of the JAK/STAT signaling pathway (20, 21, 22). Both type I (αβ) and type II (γ) IFNs are known to induce and activate JAK/STAT to directly modulate the innate immune response through transcriptional gene regulation (20, 21, 22, 23). Mutations of STAT1 in humans are associated with severe mycobacterial and viral diseases (24, 25). STAT1-null mice display a complete lack of responsiveness to IFN-α and IFN-γ (26). These animals exhibit markedly increased mortality to viral and bacterial infections due to impaired immunity. We report in this study the characterization of STAT1 induction and JAK/STAT signal pathway activation by Chlamydia infection. Our data demonstrate that the host activates an innate immune response against chlamydial infection.

The HeLa 229 cell line (CCL-2.1; American Type Culture Collection) was maintained in DMEM (Invitrogen Life Technologies) supplemented with 10% heat-inactivated FBS (defined as 10 endotoxin units (EU)/ml endotoxin or less; HyClone), 2 mM nonessential amino acids (Invitrogen Life Technologies), and 2 mM l-glutamate. The U3A and 2fTGH cells were provided by G. Stark (Cleveland Clinic Foundation, Cleveland, OH), C. trachomatis serovar L2 (434/Bu) was provided by G. Zhong (University of Texas, San Antonio, TX), serovar D was supplied by H. Caldwell (National Institutes of Health, Bethesda, MD), and anti-major outbreak membrane protein (MOMP) Ab was provided by G. McClarty (University of Manitoba, Winnepeg, Canada). The STAT/JAK Ab sampler kit (STAT1–6, IFN-stimulated transcription factor 3γ (ISGF3γ), JAK1, and Tyk2) was purchased from BD Pharmingen. The MAPK Ab kit (ERK, JNK, and p38) and the phosphorylation-specific Abs against STAT1 and MAPK were obtained from Cell Signaling. The anti-chlamydial LPS mAb was purchased from Chemicon International. Recombinant proteins of human IL-6, IL-8, TNF-α, IFN-β, and IFN-γ were purchased from PeproTech, and the corresponding Abs were obtained from eBioscience or from the National Institute of Arthritis and Infectious Diseases, National Institutes of Health, Reagent Reference Repository Program. The ELISA kits for cytokine and chemokine detection were obtained from eBioscience or R&D Systems. The sensitivity of the ELISA kits was 10 pg/ml or better, except for IFN-β which was 250 pg/ml.

C. trachomatis serovars L2/434/Bu and D/UW3/Cx were propagated in HeLa 229 cells (American Type Culture Collection; CCL2). Briefly, cultures were grown in DMEM growth medium supplemented with 10% heat-inactivated FBS, nonessential amino acids, l-glutamic acid, 60 μg/ml vancomycin, and 20 μg/ml gentamicin at 37°C in 95% air and 5% CO₂. Cultures infected with C. trachomatis L2 were grown for 48 h, and those infected with Chlamydia trachomatis serovar D were grown for 72 h before harvesting. Infected cells were detached by scraping, then pelleted and sonicated to lyse the host cells. Cellular debris was removed by differential centrifugation as described previously. Chlamydial EBs were pelleted, resuspended in isotonic sucrose-phosphate-glutamate buffer, and frozen at −80°C. Infectious titers, expressed as inclusion-forming units (IFUs), were determined by titration on HeLa 229 cell monolayers by immunostaining with an Ab against chlamydial LPS (Chemicon International), followed by Alexa Fluor-labeled secondary Ab (Invitrogen Life Technologies).

Monolayers of HeLa 229 cells were infected with chlamydial organisms at a multiplicity of infection (MOI) of 1 IFU/cell or as specified in individual experiments. The chlamydial inocula were removed 2 h after infection, and cells were fed with complete DMEM supplemented with 25 μg/ml gentamicin. For immunoblotting analysis, the soluble proteins were extracted with a lysis buffer containing 50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 1% Triton X-100, 0.1% SDS, and a mixture of protease inhibitors (Roche) on ice for 30 min. The cell lysates were cleaned by centrifugation at 13,000 × g, and the proteins were separated by SDS-PAGE. After transfer to a polyvinylidene difluoride membrane (Immobilon-P; Millipore), the proteins were detected by incubating with a primary Ab, followed by HRP-conjugated secondary Ab and the ECL reagent (Pierce).

To test the role of overexpressed STAT1 protein against chlamydial infection, HeLa 229 cells were first treated with IFN-β for 24 h to up-regulate STAT1 expression. We chose IFN-β treatment instead of transient transfection for STAT1 overexpression because this approach resulted in uniform expression of STAT1 protein among the cell population. To test the inhibitory effect of STAT1 induction on chlamydial infection, IFN-β-treated cells (10 pg/ml for 24 h) were infected with C. trachomatis at various MOIs. The infected cells were incubated at 37°C for 36 h before collection for quantitation of infectious EB production.

For determination of C. trachomatis production efficiency, we performed a secondary infection assay (27). Aliquots of the combined HeLa cell lysates and culture medium were serially diluted in DMEM and used to infect fresh HeLa cell monolayers in 16-well chamber slides. Recovered IFU were enumerated after being immunostained for chlamydial LPS or stained for bacterial DNA with a DNA-binding dye, SYTO-16 (Invitrogen Life Technologies). The yield of recovered C. trachomatis was expressed as IFU per milliliter of duplicate samples.

Culture supernatants from Chlamydia-infected or uninfected HeLa cells were collected 24 h postinfection (p.i.). The supernatants were precleaned by centrifugation at 10,000 × g at 4°C for 30 min, followed by filtration with a 0.2-μm low protein-binding filter. The absence of infectious Chlamydia in the supernatants was examined on HeLa 229 cells with an IFU-forming assay. To assay for STAT1 induction activity, HeLa 229 cells in 12-well plates were incubated with the precleaned culture supernatants for 24 h. STAT1 protein expression and induction were detected by immunoblotting analysis.

For the neutralization assay, HeLa 229 cells were infected with LGV2 in the presence or the absence of a functional blocking Ab against IFNs at 20 μg/ml. STAT1 expression was determined by immunoblotting analysis 24 h p.i.

We chose the SMARTpool silent RNA (siRNA) reagents (Dharmacon) for gene knockdown studies. SMARTpool uses a sophisticated algorithm to combine four siRNA duplexes in a single pool. These reagents are selected with the propriatry SMARTselection technology, and at least three of the four individual duplexes will silence target gene expression at the mRNA level by at least 75%. Briefly, monolayers of HeLa 229 cells were transfected with sequence-specific siRNA or control oligos using Oligofectamine (Invitrogen Life Technologies) as the manufacturer instructed. Protein expression was determined 40–48 h after transfection by immunoblotting analysis.

Total RNA from LGV2-infected or uninfected HeLa 229 cells was isolated with TRIzol reagent (Invitrogen Life Technologies) and cleaned with Qiagen RNeasy columns. The samples were processed and analyzed at the DNA Microarray Core Facility at The Scripps Research Institute. The integrity of the RNA was verified on the Agilent Bioanalyzer system. We used the U133 Plus 2.0 Affymetrix Gene Chip arrays containing >47,000 human genes and transcripts. The call rates for the uninfected control and infected samples collected at 16 and 34 h p.i. were 32.8, 31.2, and 30.0%, respectively (17,971, 17,047, and 16,427 of 54,675 total probe sets). The accuracy of the microarray data was validated with RT-PCR and quantitative RT-PCR. The transcripts for IL-6 and IL-8 were positive in both infected and uninfected cells and were up-regulated in the infected samples. The JNK1 and JNK2 transcripts remained unchanged in infected and uninfected cells, whereas no inducible NO synthase was detected (data not shown). The baseline expression was standardized and analyzed for identification of genes with altered expression.

To evaluate the cervical epithelial cell response to Chlamydia infection, we performed microarray analysis using the newly released Affymetrix Gene Chip Human Genome U133 Plus 2.0. This gene chip completely covers the human genome and 6500 additional genes and allows for systemic evaluation of the host response to Chlamydia infection. HeLa 229 cells were infected with Chlamydia trachomatis LGV2 at 1 IFU/cell. Total RNA was extracted from the infected and uninfected control cells. We chose 16 and 34 h as representative for the middle and late stages of infection, respectively.

Using two different filtering methods, we identified 77 genes and gene transcripts with significantly increased levels of expression (≥2-fold) 16 h p.i. Chlamydia infection induced genes of transcription, cell growth, cell differentiation, biosynthesis and metabolism, signal transduction, and inflammatory and host defense responses. In addition to strong induction of genes of inflammation, including chemokine IL-8, CXCL1, CXCL2, CCL2, CCL5 (RANTES), and cytokine IL-6, we found that C. trachomatis infection up-regulated IFN regulatory factor 7 (IRF7), and components of the JAK/STAT signal pathway (28, 29). The levels of STAT1 and ISGF3γ (IRF9) were induced by 2.0- and 2.6-fold, respectively, at 16 h p.i. We also detected up-regulation of IFN-induced protein with tetratricopeptides (IFIT1, IFIT2, and IFIT4), myxovirus resistance 1 and 2 (MX1 and MX2), and 2′,5′-oligoadenylate synthetases (OAS1, OAS2, and OASL), genes whose expression is dependent on JAK/STAT signal pathway activation. The up-regulated genes at 16 h are listed in Table I.

By contrast, Chlamydia infection repressed eight genes by ≥2-fold at 16 h p.i., including nuclear protein 220, Ras-related nuclear protein-binding protein-like protein (RANBP2L1), and the guanine nucleotide-binding protein (GNAO1).

The gene expression profile was more profoundly influenced by Chlamydia infection at 34 h p.i. Approximately 80% of the 470 induced unigenes are involved in one or more biological processes, including transcriptional regulation, metabolism, inflammation, and host defense. The up-regulated genes were subjected to further analysis for biological annotations using the National Institutes of Health online Database for Annotation, Visualization and Integrated Discovery (DAVID, accessed via 〈http://apps1.niaid.nih.gov/david〉). This program uses LocusLink accession numbers to link gene accessioning systems such as GenBank, Unigene, and Affymetrix identifiers and allows for assignment of gene functions as well as delineation of biochemical and signal transduction pathways. Of particular interest, we were able to identify a signal transduction pathway of host defense. We found that Chlamydia infection up-regulated IFN-β (512×), JAK2 (15.7×), STAT1 (104×), STAT2 (12×), and ISGF3γ (8.2×), important components of the JAK/STAT signal transduction pathway for transcriptional regulation of host defense genes. We also observed strong induction of IFN-inducible genes, including OAS1 (25×), OAS2 (87×), OAS3 (30×), OASL (91×), tripartite motif-containing protein (TRIM) 5 (12×), TRIM19 (140×), TRIM22 (111×), MX1 (66×), MX2 (101×), viperin (149×), and INDO (727×), whose expression is transcriptionally regulated by IFN-JAK/STAT-mediated signal transduction, indicating that the host up-regulated innate immunity in response to Chlamydia infection. Although their functions in antimicrobial responses have not been well-defined, ISG20 (538×), IFN-inducible protein (IFI27; 50×), IFI44 (25×), IFN-inducible guanylate-binding protein 1 (GBP1; 60×), and IFIT1, -2, and -4 (101, 313, and 93×, respectively) were also strongly induced by Chlamydia infection at 34 h p.i.

JAK/STAT signal transduction plays a critical role in host defense against viral and bacterial infections. Patients with subtle mutations of STAT1 protein are susceptible to mycobacterial infection (24, 25), and animals that lack STAT1 or STAT2 are prone to microbial infections because of impairment of IFN-induced immunity (26, 30). Human pathogens, such as hepatitis C virus, target the STAT1 and STAT2 signal pathways for immune evasion (31, 32, 33). We hypothesized that the host up-regulated JAK/STAT for the antimicrobial response. We then applied biochemical approaches to verify the results from DNA microarray studies. HeLa 229 cells were infected with LGV2 at 1 IFU/cell for varying times. The expression and induction of STAT proteins were evaluated by immunoblotting analysis. As shown in Fig. 1,A, the level of STAT1 expression in the uninfected HeLa 229 cells was very low. C. trachomatis infection markedly induced STAT1 expression at 1 IFU/cell. STAT1 induction was detected at ∼8 h p.i., and became significant ∼24 h p.i. STAT1 induction was not restricted to LGV2 serovar, because both serovar D and Chlamydia pneumoniae also induced STAT1 expression in HeLa 229 cells (Fig. 1,B). Moreover, Chlamydia infection induced STAT1 expression in primary cells. As shown in Fig. 1 C, human PBMC up-regulated STAT1 expression in response to Chlamydia infection.

To extend these findings, we next investigated whether STAT1 induction was dependent on Chlamydia replication. Chlamydia-infected HeLa 229 cells remained untreated or were treated with chloramphenicol to block chlamydial translation or with penicillin, which does not inhibit Chlamydia replication but does inhibit the differentiation of RB to EB (34). As shown in Fig. 1 D, treatment with penicillin did not markedly affect STAT1 induction, although chloramphenicol treatment significantly blocked Chlamydia-induced STAT1 expression, indicating that STAT1 up-regulation was dependent on bacterial growth. This was consistent with our observation that heat-inactivated C. trachomatis LGV2 did not induce STAT1 expression (data not shown). Together, these results demonstrated that Chlamydia infection up-regulated STAT1 protein expression of the JAK/STAT signal pathway. The induction of STAT1 was dependent on Chlamydia growth.

The JAK/STAT signal pathway is composed of the nonreceptor tyrosine kinases, such as JAK1, JAK2, and Tyk2, and STAT proteins (20). We next investigated whether Chlamydia infection up-regulated components of the JAK/STAT signal pathway. HeLa 229 cells were infected with C. trachomatis LGV2 for 24 h at varying MOIs, and protein expression and induction were determined by immunoblotting analysis. As shown in Fig. 2,A, both STAT1 and ISGF3γ were strongly induced at MOIs of 0.1, 0.3, and 1.0 IFU/cell, whereas the level of STAT2 was moderately up-regulated by C. trachomatis. In contrast, HeLa 229 cells constitutively expressed STAT3, and the levels of STAT3 expression were not affected by Chlamydia infection. Similar to STAT3 expression, HeLa 229 cells also expressed JAK1, Tyk2, and STAT6 proteins, and their expression was not altered by Chlamydia infection (Fig. 2 B). No expression or induction of tissue-specific STAT4 (35) and STAT5 (28) was detected in the infected cells (data not shown). Together, these data demonstrated the induction/presence of the JAK/STAT signal transduction pathway for type I and type II IFN signaling in Chlamydia-infected cells.

The JAK/STAT pathway transmits extracellular signals directly to the nucleus to target gene promoters. The activity of STAT proteins is regulated by phosphorylation, followed by dimerization and nuclear translocation (29, 36). We next determined the activation status of STAT1 protein in C. trachomatis-infected cells using phosphorylation-specific Abs (Fig. 2 C). C. trachomatis infection induced phosphorylation of STAT1 at both Y701 and S727 residues, indicating that Chlamydia infection induced STAT1 expression and promoted STAT1 activation.

STAT1 protein was originally identified as an IFN-induced transcriptional factor (21, 37). Epithelial cells (10), fibroblasts (12, 13), and macrophages (15) all secrete IFNs in response to Chlamydia infection. From our DNA microarray studies we observed strong induction of IFN-β in Chlamydia-infected HeLa 229 cells. To determine whether a secreted factor such as IFN was responsible for STAT1 up-regulation, HeLa 229 cells were treated with conditioned medium from Chlamydia-infected cells. As shown in Fig. 3,A, culture medium from C. trachomatis-infected, but not uninfected control, cells contained STAT1-inducing activity. Unlike STAT1 induction during Chlamydia infection, which was dependent on bacterial growth, chloramphenicol treatment did not inhibit STAT1-inducing activity, indicating that a secreted factor(s), not contamination by viable Chlamydia in the medium, promoted STAT1 expression. Furthermore, the activity was heat-liable, because pretreatment by heat inactivation abolished the STAT1-inducing activity (Fig. 3 A).

Chlamydia infection was reported to produce IFNs, and we observed significant up-regulation of IFN-β from the microarray study. We therefore addressed the question of whether Chlamydia induced STAT1 expression through secretion of IFNs. HeLa 229 cells were infected with LGV2 in the presence or the absence of function-blocking Abs against IFN-α, IFN-β, or IFN-γ. The expression and induction of STAT1 were determined by Western blotting analysis. As shown in Fig. 3,B, Chlamydia infection induced a 12-fold increase in STAT1 expression, whereas coincubation with the anti-IFN-β Ab blocked Chlamydia-induced STAT1 expression by ∼70%. The Abs against IFN-α or IFN-γ had a marginal effect on STAT1 expression in Chlamydia-infected cells. To extend this finding, we next treated HeLa 229 cells with rIFN-β and tested for STAT1 protein expression. As shown in Fig. 3 C, treatment of IFN-β significantly induced STAT1 expression, indicating the potential role of IFN-β production in Chlamydia-induced STAT1 expression.

STAT proteins are critical in host innate defense against viral and bacterial infections. We hypothesized that the host up-regulated and activated the JAK/STAT signal pathway against Chlamydia infection. We therefore determined Chlamydia growth in cells that were pretreated with IFN-β to induce STAT1 expression. Chlamydia growth was determined in a secondary infection assay by measuring the production of infectious organisms (27). As shown in Fig. 4, up-regulation of STAT1 expression before Chlamydia infection significantly blocked Chlamydia replication. Although IFNs are known to induce INDO, an IFN-inducible enzyme that catalyzes tryptophan oxidization and depletion, the inhibitory effect was not caused by tryptophan depletion, because supplementation of tryptophan to culture medium did not significantly reverse the inhibitory effect of IFN-β on Chlamydia infection. Thus, up-regulation of STAT1 protein expression inhibited Chlamydia growth.

To extend this finding, we next investigated whether inhibition of STAT1 expression promoted Chlamydia infection. HeLa 229 cells were treated with sequence-specific siRNA reagents (SMARTpool) to inhibit STAT1 expression. These cells were infected with Chlamydia or treated with IFN-β. As shown in Fig. 5,A, Chlamydia infection induced STAT1 expression in mock-transfected cells. Transfection with siRNA against STAT1 at both 60 and 100 nM blocked STAT1 induction by Chlamydia infection and IFN-β stimulation. The inhibitory effect of STAT1 induction was selective, because treatment with siRNA against STAT3 did not block STAT1 induction. When examined for Chlamydia growth by immunoblotting for MOMP protein expression, we found that Chlamydia grew more effectively in STAT1 knockout cells (Fig. 5 B), indicating that induction of STAT1 expression inhibits Chlamydia infection.

This conclusion was supported by additional evidence from STAT1-null cells. The human U3A cell line lacks the expression of endogenous STAT1 and is nonresponsive to IFN stimulation (55). Both the parental 2fTGH and U3A cell lines expressed STAT3 protein at similar levels and responded to Chlamydia-induced signal transduction, as determined by ERK/MAPK phosphorylation (Fig. 5,C). Infection of C. trachomatis markedly induced STAT1 protein expression in 2fTGH cells; no STAT1 expression or induction was detected in U3A cells (Fig. 5,C). Consistent with the observations after siRNA treatment, we found that ablation of STAT1 expression in U3A cells permitted more rapid C. trachomatis propagation, as determined by immunoblotting for MOMP production (Fig. 5,C) as well as infectious EB production assays (Fig. 5 D).

We investigated host cellular responses to C. trachomatis infection. We chose the Affymetrix HG-U133 Plus 2.0 Gene Chip for this study. This gene chip contains >47,000 genes and gene transcripts and allows for more comprehensive evaluation of the host gene response to Chlamydia infection compared with the gene chips used in previous studies (38, 39). Consistent with observations from these studies, we found that Chlamydia infection induced genes of cell proliferation, transcription, biosynthesis and metabolism, signal transduction, inflammation, and host defense. In addition, we identified distinct signal pathways that lead to lipid metabolism for proinflammatory IL-8 production (40) and innate antimicrobial response, as described in this study. We showed in this study that Chlamydia infection strongly induced IFN-β, STAT1, and ISGF3γ (IRF-9 or p48). The biochemical characterization of STAT1 activation combined with the detection of IFN-inducible genes and transcripts from Chlamydia-infected cells indicate the activation of a functional JAK/STAT signal pathway. Indeed, we found that the host up-regulates STAT1 protein for the antichlamydial response, because a deficiency of STAT1 expression promotes Chlamydia growth, whereas induction of STAT1 protein before C. trachomatis infection inhibits bacterial infection.

C. trachomatis infects a variety of cells, including macrophages, fibroblasts, and endothelial and epithelial cells. The cellular response of cervical epithelial cells, the primary target of chlamydial infection, plays a critical role in the disease process of Chlamydia infection (6, 11, 16, 41, 42, 43). We used HeLa 229 cells of human cervical carcinoma and C. trachomatis LGV2 serovar for DNA microarray studies. We found that the induction of STAT1 was not restricted to HeLa 229 cells or to the LGV2 serovar (Fig. 1,B), because LGV2 infection also up-regulated STAT1 in HeLa, A549 lung carcinoma, A431 epidermoid carcinoma cells (data not shown), and primary PBMC cells (Fig. 1 C).

IFNs exert antiviral and antimicrobial activities through JAK/STAT-mediated gene regulation. Humans with the defective STAT1 gene are more susceptible to mycobacterial infection, and STAT1-null animals die from viral and bacterial inoculations (24, 25, 26, 30). IFN-γ activates STAT1 to regulate IFN-γ-activating sequence-containing gene transcription. Type II IFNs induce STAT2 phosphorylation and ISGF3 complex formation, which contains STAT1, STAT2, and ISGF3γ, and gene activation (23, 44). We observed that the host up-regulates STAT1, STAT2, and ISGF3γ protein expression in response to Chlamydia infection. Thus, the host could potentially mobilize both type I and type II IFN responses for antimicrobial activity against Chlamydia infection. This is consistent with observations from a recent report that C. pneumoniae induces IFN-αβ-dependent IFN-γ secretion in murine bone marrow-derived macrophages that leads to control of intracellular bacterial growth (15).

We provided evidence to show that Chlamydia promotes STAT1 expression through the secretion of IFN-β. We found that a secreted factor was responsible for STAT1 induction, and treatment with an Ab against IFN-β inhibited STAT1 up-regulation. STAT proteins were originally identified as IFN-induced proteins. Jenkins and Lu (10) first reported IFN secretion by HeLa 229 cells that were infected with the Bour strain of trachoma. Macrophages, fibroblasts, and epithelial cells all produce IFNs in response to Chlamydia infection. We detected strong induction of IFN-β in Chlamydia-infected HeLa 229 cells with DNA microarray and RT-PCR; however, we failed to detect IFN-β production from these cells with ELISA. It is possible that the accumulation of secreted IFN-β by these cells was too low for detection by the ELISA kit used in this study. Indeed, we found that the minimal concentration of IFN-β required to induce significant STAT1 expression was ∼3–10 pg/ml (Fig. 3 C), a concentration ∼25 times below the detection limit of the commercial ELISA kit used for this study. Thus, the induction of STAT proteins probably involves a two-step process, similar to viral infection (45). Upon infection, a small amount of IFN-β is initially produced through the activation of an unknown factor(s). The IFN-β then stimulates the signal transduction pathways to activate ISGF for antiviral and antimicrobial activities (22, 45).

IFN-γ has been shown to inhibit Chlamydia infection in fibroblast cells (reviewed in Ref. 11) through tryptophan depletion and/or the production of microbicidal NO. IFN-γ treatment induces the expression of IDNO, an enzyme that catalyzes tryptophan degradation (46); however, the antichlamydial activity was not dependent on tryptophan concentration (47, 48). In addition to activation of the JAK/STAT pathway and up-regulation of IDNO in Chlamydia-infected cells, we observed strong induction of OASs, TRIM5, MX1 and MX2 GTPases, and viperin/cig5 (52), genes that have been linked to an antiviral response. OASs catalyze the conversion of ATP for the production of 2-5 A (49, 50, 51, 52 , 56), resulting in the depletion of ATP pools and a general degradation of viral and cellular RNA through 2-5 A-mediated RNase L activation (53). It is plausible to anticipate that the host up-regulates OASs to restrict Chlamydia growth via a similar mechanism. Hence, the activation of JAK/STAT in response to Chlamydia infection potentially leads to depletion of tryptophan and ATP, materials vital for bacterial replication, and production of antimicrobial compounds, such as the 2-5 A polynucleotides.

We thank Dr. Q. Pan for comments and helpful discussion, and Drs. Q. Pan and J. de Silva for reagents.

The authors have no financial conflict of interest.