Inducible Nitric Oxide Synthase Does Not Affect Resolution of Murine Chlamydial Genital Tract Infections or Eradication of Chlamydiae in Primary Murine Cell Culture

Abstract

Mice lacking inducible nitric oxide synthase (iNOS) or treated with iNOS inhibitors resolved chlamydial genital tract infections. Additionally, treatment of primary murine cell cultures with gamma interferon restricted chlamydial growth in the absence of nitric oxide. From these results, iNOS activity is unnecessary for the resolution of chlamydial genital tract infections in mice and inhibition of chlamydial growth in culture.

Several studies indicate that resolution of primary chlamydial genital tract infections and protection from a secondary infection in female mice require T-helper type 1 (Th1) lymphocytes. Athymic mice or mice lacking major histocompatibility complex class II antigens fail to resolve genital tract infections with the mouse pneumonitis biovar (MoPn) of Chlamydia trachomatis (20, 24), but mice depleted of B cells (22) or deficient in the production of antibody (26) resolve the infection. Also, chronic infection is resolved in MoPn-specific Th1 lymphocytes following adoptive transfer into athymic mice (15, 21), and infection is prolonged in CD4-deficient mice (20). Lastly, gamma interferon (IFN-γ) contributes to the control of murine chlamydial genital tract infections (9, 17, 23) and to inhibition of chlamydial growth in culture (3, 25).

One important mechanism of IFN-γ-dependent inhibition of chlamydial growth in the mouse may be the activity of inducible nitric oxide synthase (iNOS) (10). For example, specific inhibition of iNOS was reported to reverse IFN-γ-mediated inhibition of chlamydial growth in murine fibroblasts (19) and macrophages (7). Igietseme and colleagues (13, 16) found that protective IFN-γ-secreting T-cell clones restrict chlamydial growth by inducing nitric oxide production in MoPn-infected murine epithelial cells. This inhibition was reversed in the presence of the l-arginine analog N^G-monomethyl-l-arginine (L-NMMA), an inhibitor of iNOS activity. However, Chen et al. (7) reported that neither iNOS nor the induction of indoleamine 2,3-dioxygenase (IDO), a tryptophan-decyclizing enzyme responsible for inhibition of chlamydial growth in IFN-γ-treated human cells in culture (4, 27), could fully account for the IFN-γ-mediated effects. Therefore, Th1-mediated protective immunity in the immunocompetent mouse may involve IFN-γ-mediated mechanisms other than iNOS activity.

Thus, the purpose of this study was to further examine the role of iNOS in chlamydial genital tract infections in mice and in primary murine cell culture. Mice with a targeted disruption in the iNOS gene (iNOS KO) or mice treated with specific chemical inhibitors of iNOS were used for this purpose. Breeding pairs of iNOS KO mice were obtained from Edward Balish (University of Wisconsin-Madison Gnotobiotic Laboratory) under a material transfer agreement with John Mudgett (Merck & Co., Rahway, N.J.) to initiate a colony at Midwestern University. The specific disruption in the iNOS gene and promoter are described elsewhere (18). The iNOS KO was confirmed at the molecular level by PCR with genomic DNA by using primer pairs specific for the targeting vector containing the neomycin resistance gene as previously described (9). We also verified the iNOS KO by assessing nitrite accumulation in primary cell cultures derived from various tissues of iNOS KO mice in response to IFN-γ and lipopolysaccharide (LPS) in comparison to that in wild-type controls (11). For wild-type controls, 6- to 8-week-old C57BL/6hsd mice were obtained from Harlan Sprague Dawley, Inc., Indianapolis, Ind. In some experiments, control animals were derived by breeding 129/SvEv female mice with C57BL/6J males to obtain the F₁ parental hybrid strain from which the knockout was derived (18).

Eight- to 12-week-old mice were infected intravaginally with 100 50% infective doses of C. trachomatis MoPn (Weiss strain) in 5 μl of SPG buffer (10 mM phosphate, 0.25 M sucrose, 5 mM l-glutamic acid [pH 7.2]) containing between 5 × 10³ and 1 × 10⁴ inclusion-forming units (IFU) (5, 9) after treatment with 2.5 mg of progesterone, as previously described (8). The infection was assessed by the sequential collection of cervical-vaginal swabs (Calgiswab, type 1; Spectrum Diagnostics, Houston, Tex.) at 4, 7, 10, and 14 days postinfection and every 7 days thereafter until cessation of chlamydial shedding. MoPn was isolated in HeLa cell cultures and enumerated by indirect fluorescence microscopy as previously described (9).

In order to monitor iNOS activity, urine nitrite and nitrate levels, which correlate with nitric oxide production in vivo, were determined. Urine was collected daily from mice housed in metabolism cages and fed a nitrate- and nitrite-free diet. Urine nitrate and nitrite content was determined by the Greiss reaction (11). On the day of infection, the drinking water of the experimental group was spiked with 50 mM L-NMMA (CYCLOPPS Corp., Salt Lake City, Utah), an iNOS inhibitor. Drinking water was replaced daily. iNOS KO mice were not assessed for urine nitrate output and received their normal diet of autoclaved chow and sterile water.

The results presented in Fig. 1 show the course of chlamydial infection in wild-type animals receiving either Millipore drinking water spiked with 50 mM L-arginine (Fig. 1A) or L-NMMA-spiked drinking water (Fig. 1B) and in iNOS KO mice (Fig. 1C). Mice receiving L-NMMA did not have significantly prolonged infection compared with those receiving no treatment despite marked inhibition of iNOS activity as indicated by urinary nitrite excretion. Additionally, in 14 of 15 iNOS KO mice the infection was resolved over the 42-day monitoring period, with the one remaining infected mouse at day 42 shedding less than 100 IFU of viable MoPn. Statistical analysis of the percentage of animals remaining infected at each of the times indicated no significant differences in the duration of infection between any of the three groups (two-factor analysis of the variance with repeated measures of one factor). Nonetheless, in the remaining culture-positive iNOS KO mice at day 14 postinfection and in the L-NMMA-treated wild-type animals at day 10, significantly higher numbers of IFU were isolated compared to those in the wild-type controls (P < 0.02 by a two-tailed t test). Similar results were observed in another experiment comparing animals given aminoguanidine with those receiving Millipore water only (data not shown). These results indicate that while iNOS may be involved in controlling chlamydial replication during the midpoint of the infection, its activity is not essential to resolution of the infection.

FIG. 1.

Role of iNOS in chlamydial genital tract infection in vivo. The course of infection (solid line) and urine nitrate levels (dashed line) in wild-type C57BL/6 mice receiving either 50 mM L-arginine (A) or 50 mM L-NMMA (B) in their drinking water and in iNOS KO mice (C) is shown. Each data point represents the mean IFU from cervical-vaginal swabs of culture-positive mice collected 4, 7, 10, 14, 21, 28, 35, and 42 days postinfection. Above each data point, the ratio of the number of culture-positive animals to the total number of animals in each experimental group is given. The asterisks in panels B and C designate significant differences in the quantitative recovery of viable MoPn at the indicated times postinfection compared to that in the control mice (panel A). Urine was collected daily for nitrate determination. The urinary nitrate response was not assessed in iNOS KO mice.

We then sought to determine the reason for the apparent disparity in the observations in our experiments and in those of previously published works (7, 14, 19). We hypothesized that IFN-γ controlled chlamydial replication independently of iNOS activity. Therefore, we assessed the effect of IFN-γ on chlamydial replication in various cells derived from iNOS KO and control mice.

Following sacrifice of uninfected iNOS KO or control animals, organs were aseptically removed and processed to obtain single-cell suspensions by a standard collagenase D stainless-steel mesh filtration procedure. Cells were plated in Eagle’s minimal essential medium containing 10% heat-inactivated (56°C, 30 min) fetal bovine serum and 2 mM l-glutamine at 1 ml per well in 24-well plates. Primary cell cultures were monitored daily with an inverted microscope and fed by replacing medium at 2- or 3-day intervals. Upon reaching confluency or near confluency (usually 9 to 12 days in culture), the cells were used in the nitric oxide induction and chlamydial growth assays described below. To obtain peritoneal macrophages, cells were harvested by peritoneal lavage 72 h after intraperitoneal injection with 1 ml of 3% Proteose Peptone (DIFCO, Detroit, Mich.). Following enumeration, the cell suspension was diluted to 2 × 10⁵ cells/ml, plated at 1 ml per well in 24-well plates in Eagle’s minimal essential medium with 10% fetal bovine serum, and allowed to adhere for 24 h prior to use.

Initially, single-cell preparations from several tissues were screened for their ability to form monolayers in culture and to support chlamydial growth. From these results, we determined that cell cultures from urinary bladder and lung tissues were best suited for our purposes. These preparations from iNOS KO and control mice were found to be predominantly fibroblasts at confluency by indirect fluorescent-antibody staining with fibroblast-specific goat antivimentin (1:20; ICN Biomedical, Aurora, Ohio) followed by fluorescein-labeled rabbit anti-goat immunoglobulin (1:200; ICN Biomedical).

Treatment of murine cells with IFN-γ and LPS induces iNOS to produce nitric oxide from l-arginine, resulting in the accumulation of nitrite in culture supernatants. This response can be blocked by the presence of the arginine analog L-NMMA (12). Therefore, we assessed nitrite accumulation in the primary cell cultures described above in response to IFN-γ and LPS in the presence and absence of L-NMMA. Culture medium was aspirated from primary murine cell lines and macrophages and replaced with medium containing either IFN-γ (50 ng/ml; Pharmingen, San Diego, Calif.) plus LPS (100 ng/ml; Sigma), IFN-γ plus LPS plus 1 mM L-NMMA (Calbiochem, La Jolla, Calif.), fresh medium plus L-NMMA, or fresh medium only for 48 h. Nitrite accumulation was assessed by the Greiss reaction (11). Treatment of cells from control mice with IFN-γ and LPS resulted in marked nitrite production in the supernatants of each of the primary cultures (P < 0.005, by a two-tailed t test, in each case compared with that in controls with medium only) while the presence of L-NMMA completely reversed this response (data not shown). As expected, no nitrite was detected at any time in supernatants of primary cell cultures derived from iNOS KO mice. We also assessed IDO activity by previously published methods (6) but were unable to detect any activity, as indicated by a lack of accumulated tryptophan catabolites in the culture supernatants (data not shown).

Parallel cultures were also infected with 4 × 10⁵ IFU of HeLa-grown MoPn per well, incubated for 40 h at 37°C in a humidified atmosphere of 5% CO₂, washed once with phosphate-buffered saline, and fixed with methanol. Chlamydial inclusions were stained by an indirect fluorescent-antibody method and enumerated as described elsewhere (9). The results presented in Fig. 2 show that treatment with IFN-γ plus LPS restricts chlamydial growth in primary cultures of each of the tissues derived from both iNOS KO and control animals (P < 0.0001 in each case by comparing the numbers of IFU in IFN-γ-plus-LPS-treated cultures to those in untreated cultures by a two-tailed t test). This inhibition of chlamydial growth was not significantly reversed in the presence of L-NMMA, indicating that mechanisms other than iNOS activity are responsible for IFN-γ restriction of chlamydial growth in cultured murine cells.

FIG. 2.

Chlamydial replication in primary murine cell cultures derived from lung, urinary bladder, and peritoneal macrophages of iNOS KO and F₁ 129/SvEv × C57BL/6 mice treated with medium plus 1 mM L-NMMA (closed bars), γ-IFN plus LPS (open bars), or γ-IFN plus LPS plus 1 mM L-NMMA (hatched bars). All data are standardized to percentage of the mean IFU enumerated in control cultures receiving medium only.

Our findings appear to contradict those of others who have examined the role of nitric oxide in chlamydial infections. For example, Mayer et al. described IFN-γ-mediated inhibition of chlamydial growth in murine fibroblasts that was attributable to nitric oxide production (19). In several studies, Igietseme et al. have described the use of a MoPn-specific Th1 clone to determine the role of nitric oxide-mediated restriction of chlamydial growth in vitro (13, 16) and in vivo (14). The activity of the clone against chlamydiae was inhibited by the presence of L-NMMA. The differences between our current findings and the previous findings described above are likely due to the fact that in the in vivo system used in our study other factors are present and may play a role in the absence of iNOS. These factors may be absent in the athymic mouse model and in the clonal in vitro coculture systems employed by Igietseme et al. Hence, compensatory mechanisms effectively control chlamydial replication in the absence of iNOS activity. Indeed, a multitude of both IFN-γ-dependent and -independent mechanisms likely work together in the immunocompetent mouse to effect clearance of the infection.

In similar studies, Chen et al. (7) found that although inhibition of iNOS activity resulted in a partial reversal of chlamydial growth inhibition, the activity of neither iNOS nor IDO could fully account for the observed effects of treatment with IFN-γ plus LPS in the RAW264.7 murine macrophage cell line or in freshly isolated murine peritoneal macrophages. These findings lend credence to the conclusion that a third, unidentified IFN-γ-mediated mechanism may be in effect.

Interestingly, we did observe a tendency of iNOS KO and wild-type mice treated with chemical inhibitors of iNOS to shed higher numbers of organisms between 10 and 14 days postinfection than did controls. Others have reported similar differences in quantitative recovery of the organism early in the course of infection in mice treated with L-NMMA (14).

Although previous studies indicate that IFN-γ is present in endocervical secretions in humans during chlamydial infection, the precise role of IFN-γ in humans in vivo is not known (2). In vitro, in human cells IFN-γ induces IDO, which is responsible for inhibition of chlamydial growth via tryptophan starvation (4, 6, 27). Others have reported IDO activity in murine macrophages in response to IFN-γ, and this activity appears to be negatively regulated by the activity of iNOS (1). However, we did not detect IDO activity in primary murine cell culture in response to IFN-γ and LPS in iNOS KO or control mice. This finding suggests that tryptophan catabolism is not likely to be the compensatory mechanism used by iNOS KO mice to control the infection. A third, yet-unidentified, IFN-γ-mediated mechanism may be in effect and is currently under investigation.