Role of NK Cells in Early Host Response to Chlamydial Genital Infection

Experimental animals.

Female BALB/c mice at 4 to 5 weeks of age were purchased from Harlan Sprague-Dawley, Indianapolis, Ind. Animals were housed in an environmentally controlled room with a cycle of 12 h of light and 12 h of darkness. Animals were routinely used in experiments when they were 6 weeks old.

Chlamydia culture and infection of mice.

The C. trachomatis biovar MoPn, which was originally obtained from the American Type Culture Collection and has been continually passaged in our laboratory for approximately 20 years, was used throughout this study. MoPn organisms for infection purposes were grown on McCoy cell monolayers, and elementary bodies were purified, titrated, and stored in sucrose-phosphate buffer (2-SP) (23) by established procedures. MoPn for use as an antigen was cultured on HeLa cells in order to avoid cross-reactivity to the host cells.

Seven days prior to infection, mice were subcutaneously injected with a single dose of 2.5 mg of progesterone in the form of Depo-Provera (Upjohn, Kalamazoo, Mich.) in order to induce a state of anestrus in the mice and thus eliminate any potential effect of the stage of the estrous cycle. This was particularly important in this study because we were studying various immunological parameters over a short 5- to 7-day period at the very onset of infection. Lack of estrous cycle synchrony in this study might have introduced major variables which would have made interpretation of the data difficult. For genital infection, mice were anesthetized by the intraperitoneal injection of pentobarbital sodium (50 mg/kg of body weight) and infected intravaginally with 10⁷ inclusion-forming units (IFU) in 30 μl of 2-SP buffer.

Preparation of MNCs.

Mononuclear cells (MNCs) were extracted from the genital tracts and various secondary lymphoid tissue samples from mice at various times after infection. To prepare genital tract-associated MNCs, the genital tracts from infected mice were pooled, minced thoroughly, treated with 10 ml of sterile 0.5% type I collagenase (Sigma Chemicals, St. Louis, Mo.) at 37°C for 1 h with brief agitation every 15 min, and filtered through 70-μm-pore-size nylon cell strainers (Falcon; Becton Dickinson, Paramus, N.J.) with an additional 40 ml of RPMI 1640 medium to remove cell debris. The MNCs were then enriched over a Ficoll-Hypaque gradient (Lympholyte M; Cedarlane Laboratories, Hornby, Ontario, Canada). The contaminating erythrocytes were lysed with a 0.85% NH₄Cl solution (pH 7.2). MNCs of secondary lymphoid tissues were prepared by gentle teasing of the pooled tissues in RPMI 1640 medium and enriched over a Ficoll-Hypaque gradient. MNCs derived from lymph nodes are usually free of erythrocyte contamination and did not require NH₄Cl treatment. The viability of the resulting MNC preparation was routinely greater than 95%, as assessed by trypan blue exclusion.

Phenotyping of genital tract-associated MNCs.

Phenotyping of MNCs was performed by standard flow cytometric methodology. Cells were stained with predetermined dilutions of fluorescein isothiocyanate (FITC)-conjugated anti-CD3 (immunoglobulin G2b [IgG2b]) monoclonal antibodies (MAbs), FITC-conjugated polyclonal anti-mouse Ig (IgA, IgM, and IgG) (Sigma), FITC-conjugated rat anti-mouse Mac-3 MAb (IgG1) (Pharmingen, San Diego, Calif.), and appropriate isotype-matched control antibodies for 30 min at 4°C. The stained cells were washed twice with phosphate-buffered saline (PBS) containing 0.2% bovine serum albumin and 0.02% NaN₃. Flow cytometric analysis was performed on a FACScan, and mean channel fluorescence was calculated by using the Lysis II and WinMDI software packages (Becton Dickinson, Mountain View, Calif.).

Analysis of NK cell activity.

The NK cell activity of cells derived from various tissues was determined by measuring the specific cytolytic activity against ⁵¹Cr-labeled YAC-1 target cells in a standard 4-h chromium release assay at several effector/target (E/T) ratios, as described previously (7) with slight modifications. Briefly, 2 × 10⁶ YAC-1 cells in 10 ml of RPMI 1640 medium (Cell-Gro) containing 10% fetal calf serum, 100 U of penicillin per ml, and 100 μg of streptomycin per ml were labeled with 100 μCi of sodium chromate (ICN) for 12 h at 37°C in a CO₂ incubator. A total of 10⁴ thoroughly washed YAC-1 target cells in 100 μl of complete medium were then distributed into wells of round-bottomed microtiter plates containing effector cells at various concentrations. Each test sample was plated in triplicate. The microtiter plates were incubated at 37°C for 4 h. The percentage of specific ⁵¹Cr release was calculated as follows: 100 × (cpm of test sample − cpm of medium)/(cpm of maximum control − cpm of medium control), where cpm is counts per minute. The spontaneous release of target cells was usually 6 to 7% and never exceeded 10%.

Titration of anti-asialo-GM1 rabbit antibody.

Although asialo-GM1 (Waco Bioproducts, Richmond, Va.) has been widely used as a surface marker to identify NK cells, other cell types also express this surface antigen (12). Therefore, it was necessary to determine the most appropriate concentration of anti-asialo-GM1 antibody to attain maximal depletion of NK-like activity with a minimal effect on other cell types. In this titration experiment, the criteria for selecting the best titer of this antibody was a maximal depletion of in vitro cytotoxicity against YAC-1 cells, concomitant with a minimal effect on the CD3⁺ T-cell population, as determined by a fluorescence-activated cell sorter. A serial dilution of anti-asialo-GM1 antibody and a fixed concentration of freshly prepared rabbit complement (1:10 dilution) were used to selectively deplete asialo-GM1-positive cells in MNCs derived from the spleens of BALB/c mice. Medium alone and complement alone were also included in this titration study as controls. For in vivo use of anti-asialo-GM1 antibody, various amounts of the antibody were injected intraperitoneally into mice at indicated times, and the efficacy of such treatment was determined by analyzing NK-like activity and CD3⁺ T-cell population as above. Control mice received equivalent amounts of an irrelevant rabbit antibody (inactivated rabbit anti-herpes simplex virus [HSV] serum). The experiments indicated that a 1:200 dilution and 50 μl of anti-asialo-GM1 rabbit antibody per injection were the optimal concentrations for in vitro and in vivo use, respectively.

Selective depletion of T cells or NK cells.

Depletion of selective T cells or NK cells was performed both in vitro and in vivo with the appropriate antibodies and dilutions discussed above. In vitro, complement-mediated cytotoxicity was used to selectively deplete T cells or NK cells. Briefly, a minimum of 10⁷ MNCs was incubated with 100 μg of anti-CD3 MAb (Gibco Laboratories, Grand Island, N.Y.) or a 1:200 dilution of rabbit anti-asialo-GM1 antibody for 1 h at 4°C with constant agitation. The treated MNCs were washed and then incubated with appropriately diluted low Tox-M rabbit complement (Accurate Chemical & Scientific Corp., Westbury, N.Y.) for 1 h at 37°C. After the cells were thoroughly washed, the resulting cell suspension was used as a CD3⁺ T-cell-depleted or NK cell-depleted suspension, as described above.

In vivo, polyclonal anti-asialo-GM1 antibody was used to deplete NK cells. Mice were injected intraperitoneally with 50 μl of anti-asialo-GM1 or an equivalent amount of an irrelevant rabbit antibody (inactivated rabbit anti-HSV serum) on days −3, −1, 1, and 3 after MoPn infection. The efficacy of such treatment was evaluated by the standard YAC-1 cell cytotoxicity assay.

Measurement of MoPn-specific antibodies.

IgG1 and IgG2a antibodies to MoPn were measured by a standard enzyme-linked immunosorbent assay as previously described (13). Peroxidase-labeled antibodies to murine IgG1 and IgG2a were obtained from Southern Biotechnology Associates, Inc., Birmingham, Ala.

Analysis of IL-4 and IFN-γ production.

A modification of the ELISPOT assay, originally described by Taguchi et al. (27), was used to determine the profile of cytokine production of genital tract-associated MNCs. Briefly, nitrocellulose-based 96-well plates (Millititer HA; Millipore Corporation, Marlborough, Mass.) were coated with the primary antibody (2 μg/ml) directed against either murine IL-4 or murine IFN-γ (Pharmingen). After the plates were coated overnight at 4°C, they were washed with PBS containing 0.05% Tween 20 (PBS-Tween; pH 7.4), and blocked with PBS containing 5% fetal calf serum for 1 h at 37°C in the CO₂ incubator. To determine the cytokine production, MNCs were extracted from iliac lymph nodes (ILN) at indicated times after infection and were stimulated in vitro with a Renografin-purified preparation of elementary bodies (3) (5 μg/ml), derived from infected HeLa cells, at 37°C overnight. The stimulated MNCs were distributed into each well in triplicate at various concentrations and incubated for an additional 20 h in the incubator. The plates were then washed extensively with PBS-Tween to remove unbound cells, followed by an overnight incubation at 4°C with a secondary biotinylated antibody (4 μg/ml), directed against either IL-4 or IFN-γ. After a thorough washing, the plates were incubated with 2.5 μg of avidin-peroxidase (Vector, Burlingame, Calif.) per ml for 1 h, followed by color development with 3-amino-9-ethylcarbazole. Spots or cytokine-producing cells were enumerated with the aid of a dissecting microscope. The mean number of spots derived from the triplicate samples was used to calculate the spot-forming cells per million cells.

RT-PCR.

Reverse transcription-PCR (RT-PCR) was used to verify the expression of transcripts of specific cytokines in the genital tracts of mice early in the infection. This technique basically involves three major steps, including isolation of total RNA from infected genital tracts, reverse transcription of the first-strand cDNA from total RNA, and amplification of sequences of interest from cDNA by PCR.

(i) RNA extraction.

Total RNA was extracted from the genital tracts of mice at the indicated times after MoPn infection, using commercially available TRI REAGENT (Molecular Research Center, Inc., Cincinnati, Ohio) according to the manufacturer’s recommendation. Briefly, the genital tracts of five mice were pooled, minced, and digested with collagenase, as described earlier for preparation of MNCs. The resulting cell suspensions after digestion were directly lysed with 1 ml of TRI REAGENT. The homogenates were left for 5 min at room temperature to permit complete dissociation of nucleoprotein complex before the addition of 0.1 ml of 1-bromo-3-chloropropane. The resulting mixtures were vigorously shaken, incubated at 25°C for 10 min, and centrifuged at 12,000 × g for 14 min for phase separation. The aqueous phase containing RNA was transferred to a fresh tube. The RNA was precipitated by adding 0.5 ml of isopropanol, and the solution was centrifuged. The resulting precipitates were washed once with 1 ml of 75% ethanol and centrifuged at 7,500 × g for 5 min. After a brief drying period, RNA pellets were dissolved with diethylpyrocarbonate-treated water and quantified at 260 nm in a spectrophotometer (Hitachi Instrument, Inc., Houston, Tex.). The RNA samples were stored at −80°C or immediately used for the first-strand cDNA synthesis.

(ii) First-strand cDNA synthesis.

The cDNA was synthesized directly from total RNA by using a commercially available kit purchased from CLONTECH Laboratories, Inc. (Palo Alto, Calif.). Briefly, a 20-μl mixture contained 1 μg of total RNA, 20 pmol of oligo(dT)₁₈ primer, 0.5 mM each of the four deoxynucleoside triphosphates, 20 U of RNase inhibitor, Tris buffer (50 mM Tris-HCl [pH 8.3], 75 mM KCl, 3 mM MgCl₂), and 200 U of Moloney murine leukemia virus reverse transcriptase were incubated at 42°C for 1 h, followed by 5 min at 95°C to stop the reaction. The final reaction mixture was diluted to a final volume of 100 μl by the addition of 80 μl of diethylpyrocarbonate-treated H₂O and stored either at 4°C for use in 2 weeks or at −20°C for a much later usage.

(iii) Amplification of target signals in cDNA by PCR.

The standard PCR was used to amplify the active transcripts of targeted cytokines indirectly from cDNA. The cytokine-specific oligonucleotides, designed to amplify only cDNA, but not genomic DNA, were either custom-made or purchased from Clontech Laboratories, Inc. The sequences of these amplimers and the expected sizes after amplification are summarized in Table 1. For amplification by PCR, 5 μl of the diluted cDNA sample was typically used as the template in a total volume of a 50-μl PCR mixture, containing 0.2 mM each of the four deoxynucleoside triphosphates, 0.4 μM each of 5′- and 3′-end primers, Tris buffer (10 mM Tris-HCl, [pH 8.4], 50 mM KCl, 1.5 mM MgCl₂) and 2.0 U of Ampli Taq polymerase (CLONTECH Laboratories, Inc.). A DNA thermocycler (model 480; Perkin-Elmer, Norwalk, Conn.) was set to the following conditions for all the amplimers, (except IL-12 p40): (i) an initial denaturation step for 3 min at 94°C, (ii) 30 cycles of PCR, with 1 cycle consisting of denaturation at 94°C for 45 s, primer annealing at 60°C for 45 s, and elongation at 72°C for 2 min, and (iii) a single final extension step of 7 min at 72°C. To amplify IL-12 p40 sequences, the following PCR conditions were used: (i) 3 cycles with each cycle consisting of denaturation at 94°C for 45 s, annealing at 58°C for 1 min and 15 s, extension at 72°C for 1 min 45 s and (ii) 27 cycles, with each cycle consisting of 94°C for 35 s, 58°C for 45 s, and 72°C for 1 min and 15 s. To visualize the amplified signals, an aliquot of 10 μl of the resulting PCR product was electrophoresed onto 2% agarose gels in Tris-acetate-EDTA buffer at pH 8.0. The HaeIII fragments of φX174 replicative-form DNA (Gibco-BRL, Life Technologies) were included as molecular size markers. After electrophoresis, the gels were stained with ethidium bromide and documented with ImageStore 7500 transilluminator (Ultra Violet Products Inc., Upland, Calif.). The assessment of active glyceraldehyde-3-phosphate dehydrogenase (G3PDH) transcripts was used as a positive control and to calibrate the amount of RNA.

Statistics.

Unless otherwise indicated, all data were analyzed by using a one-tailed t test with P of <0.05 as the maximal value for statistical significance. All experiments were repeated at least once.

Characterization of the early cytokine profile in the genital tracts of BALB/c mice intravaginally infected with MoPn.

Previously, we demonstrated that a prominent IFN-γ response could be detected in the genital tract as early as 7 days after intravaginal infection with MoPn (2) and that this response did not appear to be related to a CD4 response (2a). However, since these data represented a single time point and since it has become increasingly clear that the early events in the host response to an infection are significant with regard to the nature of the acquired immune response, we wanted to determine when specific cytokines were expressed following intravaginal infection with chlamydiae. A total of 30 mice were intravaginally infected with MoPn, and groups of five mice each were killed at 6, 12, 18, 24, 36, and 48 h after infection. Five age-matched, uninfected mice were also included as the control, which was designated 0 h. Spleen cells derived from healthy uninfected mice were stimulated with concanavalin A and were used as a source of RNA for a positive control for the expression of tumor necrosis factor alpha (TNF-α), macrophage inflammatory protein (MIP), IFN-γ, IL-10, and IL-4. The genital tracts were pooled and digested with collagenase, and the resulting cell pellets were directly used for the total RNA extraction without further enrichment for MNCs. The RNA was used to assess cytokine expression by RT-PCR.

When the kinetics of expression of each individual cytokine were analyzed, it became clear that the expression of various cytokines occurred quite early in the infection process. TNF-α, MIP, and IL-10 were constitutively expressed in uninfected animals (Fig. 1); however, TNF-α and MIP genes appeared to be up-regulated as early as 6 h after infection. IFN-γ expression was first detected 12 h after infection, and the expression of the inducible p40 subunit of IL-12 was detected at 18 h. Since IL-12 has been shown to be an upstream cytokine, which is required for promoting IFN-γ production by NK cells, the 6-h delay in IL-12 expression, compared to IFN-γ expression, was somewhat unexpected, although it is possible that the RT-PCR conditions used in our study to detect IL-12 were not optimal, thereby reducing the sensitivity in detecting specific transcripts. Similar results were obtained when the experiment was repeated. IL-4 was not detected at any time during the 48-h period after infection. Of interest in this experiment was the observation that IFN-γ was expressed as early as 12 h after infection, suggesting the rapid recruitment of cells capable of producing IFN-γ to the local site. The fact that IFN-γ expression occurs so quickly would also suggest that the responsible cell is normally present either in the local site or in the circulation and does not require induction.

Determination of NK cell activity in tissues of MoPn-infected mice.

Since NK cells are primary effector cells in the innate immune response and have been implicated as a source of IFN-γ in other infectious disease models, we undertook to determine whether they could be detected in the genital tract and ILN early in chlamydial genital infection as well. Thus, in order to determine the kinetics of the NK response, several groups of mice (10 mice/group; 7 weeks old) were intravaginally infected with 10⁷ IFU of MoPn. MNCs were collected from pooled genital tracts, ILN, mesenteric lymph nodes (MLN), and spleens at 2, 4, 6, 10, 15, and 20 days after infection. When the total number of MNCs derived from genital tracts was determined, it was observed that there was a dramatic increase in MNC number from a baseline count of 10⁵ cells to 2.8 × 10⁶ as early as 2 days after infection, with the peak number 6 days after infection (2.8 × 10⁷).

NK cytolytic activity in each of the MNC preparations was determined by the standard NK cytotoxicity assay by using YAC-1 cells as targets. As shown in Fig. 2, NK activity in the genital tract, as judged by YAC-1 cell cytotoxicity, increased over the baseline as early as 12 h after infection. Because of the low cell recovery, we were able to evaluate the NK activity in the genital tracts and ILNs only at 12 h after infection at an E/T ratio of 12:1. NK activity continued to increase rapidly in the genital tract with the peak response occurring 2 days after infection. However, by 10 days after the infection, NK activity had returned to lower, albeit still positive levels. The NK responses in both the ILN and the spleen followed similar kinetics but reached peak levels about 1 day later than cells from the genital track. While the peak response in each tissue was significantly higher than the baseline levels (P < 0.0001 by two-way analysis of variance), the maximum response in the ILN and the spleen was lower than that of the genital tract. Although NK activity was readily detectable in the genital tract, ILN, and spleen, only a minimal NK response was observed in the MLN, indicating that events necessary to activate the NK response did not occur in the MLN. In particular, chlamydiae or chlamydial antigen may not have reached the MLN at this early stage of infection.

Confirmation of NK cell identity.

While NK cells are routinely characterized with regard to their ability to lyse YAC-1 cells, it was important to confirm that this activity in the genital tract cell suspension was indeed caused by NK cells. Rabbit anti-asialo-GM1 antibody has been successfully used to deplete NK cells, both in vitro and in vivo (6, 12). Thus, to verify the identity of NK cells in BALB/c mice, asialo-GM1-positive cells were selectively depleted from MNC derived from either genital tracts or ILN in vitro by complement-mediated cytotoxicity. Since the expression of asialo-GM1 antigen is not restricted solely to NK cells and may be expressed by other cells, such as T cells, monocytes, and liver cells (12), we therefore determined an appropriate dilution of anti-asialo-GM1 antiserum to deplete NK cells and minimally affect other cells. In addition, because T cells are also present at this time, the extracted cell population was also treated with anti-CD3.

When the MNCs obtained from the genital tract and ILN 4 days after infection were treated with anti-asialo-GM1 and complement, the level of NK cytotoxic activity was significantly reduced compared with that in control cells treated with complement alone or in untreated controls (P < 0.01 according to a one-tailed t test) (Fig. 3). The percentage of MNCs from infected ILN which were CD3 positive remained unaltered (P > 0.05 when assessed by a chi-square test) by in vitro treatment with anti-asialo-GM1 (39.1 ± 0.8) compared to controls treated with complement alone (42.7 ± 1.2). There were insufficient cells from the genital tract to determine the percentage of CD3-positive cells. In contrast, genital tract and ILN MNCs depleted of CD3-positive cells (percentage of CD3⁺ MNCs, 17.5 ± 1.6; P < 0.001) retained toxicity for YAC-1 cells. These data suggested that the lysis of YAC-1 cells seen with cells derived early after MoPn infection was indeed mediated by NK cells and not by T cells.

To confirm that NK activity could be removed from mice in vivo by anti-asialo-GM1 treatment, we intraperitoneally injected mice with either a predetermined amount of anti-asialo-GM1 antibody (50 μl) or an equivalent amount of irrelevant rabbit serum at −3, −1, +1, and +3 days after MoPn infection. MNCs were extracted from genital tracts and ILN on day 4 postinfection and assayed for in vitro toxicity against YAC-1 cells. Similar to the in vitro experiments, the percentage of CD3⁺ cells from infected ILN and genital tracts remained unaffected by in vivo treatment with anti-asialo-GM1 compared to that from untreated controls and in contrast to Mac-1-expressing cells which were significantly reduced (Table 2). Mac-1 has been demonstrated to be present on NK cells. However, as shown in Fig. 4, this treatment resulted in a nearly complete depletion of cytotoxic activity against YAC-1 cells from MNCs derived from the genital tracts and ILNs compared to those of control mice (P < 0.01 by one-tailed t-test) in two independent experiments. Therefore, it is clear that multiple injections of mice with anti-asialo-GM1 antibody can efficiently remove NK cells. These data also indicated that the cytolytic activity against YAC-1 cells was NK cell dependent.

NK cells are responsible for the early local IFN-γ response in MoPn-infected mice.

We have shown that a strong IFN-γ response was induced in genital tracts and ILNs of mice within a few days of chlamydial genital infection. Because both T cells and NK cells are potent IFN-γ producers, the identity of the effector cells mediating the early IFN-γ response in MoPn-infected mice was not certain. However, since in a preliminary study, selective depletion of CD4⁺ T cells did not alter the frequency of IFN-γ-producing cells in genital tracts and ILNs (2a), the role of NK cells in the production of the early local IFN-γ response was evaluated.

MNCs were extracted from the ILNs of 20 mice at day 4 after MoPn infection, and T cells or NK cells were selectively depleted in vitro by complement-mediated cytotoxicity with specific anti-CD3 antibody or anti-asialo-GM1 rabbit antibody, respectively. The ELISPOT assay for IFN-γ-producing cells was performed on the resulting cell populations. The ILN were chosen rather than the genital tracts because of difficulties in performing the ELISPOT assay on genital tract lymphocytes. In addition, previous data from our laboratory have shown that cellular and molecular events in the genital tract are accurately reflected by the ILN as well. Controls included cells treated with complement alone and cells that were not treated. As shown in Fig. 5, only the population of cells treated with anti-asialo-GM1 had a significantly lower number of IFN-γ-producing cells. Treatment with anti-CD3, while depleting T cells, did not alter the number of cells secreting IFN-γ. Thus, these data strongly suggest that IFN-γ in the ILN early after chlamydial infection is produced by NK cells.

To further confirm that NK cells are indeed the primary effector cells responsible for the early IFN-γ response in ILN, we also depleted NK cells in vivo before and during the course of chlamydial infection by multiple injections with anti-asialo-GM1 antibody. Age-matched mice receiving an equivalent amount of irrelevant rabbit serum were included as controls. At day 4 after infection, MNCs were extracted from ILN and the total number of IFN-γ-producing cells was determined by the ELISPOT assay. Similar to the in vitro results, animals treated with anti-asialo-GM1 had a marked reduction in the number of IFN-γ-producing cells detected in the ILN (Fig. 5). These results confirmed the results derived from the in vitro depletion study and indicated that NK cells appear to be the major source of IFN-γ early in the course of MoPn genital infection.

Although the cellular response in the ILN reflects the events in the genital tract, it was still important to demonstrate that depletion of NK cells from the genital tract also reduced the level of IFN-γ in this site as well. It has proven difficult to obtain positive results with the ELISPOT assay early in the infection course; therefore, we examined genital tract tissue for the presence of IFN-γ RNA transcripts in animals infected with MoPn and in infected animals treated with anti-asialo-GM1. Because good expression of IFN-γ-specific transcripts was detected in the genital tracts of mice 36 h after infection, we selected this time point to determine the transcription signals of IFN-γ gene in genital tracts. Thus, two groups of five mice each were treated either with anti-asialo-GM1 antibody to deplete NK cells or with irrelevant rabbit serum as a control. At 36 h after infection, genital tracts from five mice of each group were harvested and pooled for preparation of MNCs. Total RNA extracted from the resulting MNCs was processed by RT-PCR. In vivo depletion of NK cells resulted in a lack of IFN-γ expression in genital tracts compared to that in control mice (Fig. 6), thereby supporting the observations in the ILN that NK cells are responsible for the production of IFN-γ early in the course of chlamydial genital infection.

Effect of NK depletion on the course of MoPn genital infection.

IFN-γ produced by Th1 cells has been documented to be important in the resolution of murine chlamydial genital infections; however, as indicated by the above data, IFN-γ is also elaborated by NK cells early in the course of the infection. Since NK cells are present in the tissue so early in the course of infection, it is of interest to determine whether NK cells have an impact on the development of the immune response to chlamydiae and whether they play a role in the control of the early portion of the infection. In order to evaluate the possible contribution to the early host response by NK cells, two groups of animals were infected with 10⁷ IFU of MoPn in the genital tract and were treated either with anti-asialo-GM1 or an irrelevant antibody. The course of the infection was monitored by assessing IFU on cervical swabs at various times after infection. At the end of the experiment, serum was obtained from all mice and the levels of IgG1 and IgG2a were measured to determine the relative Th1 and Th2 responses.

In two separate experiments, there was no significant difference in the number of chlamydial IFU early in the infection (days 3 to 6) (data not shown); however, the infection in anti-asialo-GM1-treated mice was significantly prolonged than in control animals (P < 0.005) (Fig. 7). Interestingly, when IgG1 and IgG2a levels were determined 30 days after infection, there was a significant increase in IgG1 antibody to MoPn in mice treated with anti-asialo-GM1 (P < 0.017) (Fig. 8). These data suggested that depletion of NK cells resulted in a shift from a Th1 dominant response to more of a Th2 response. Thus, it is likely that depletion of NK cells deprived the animal of the early IFN-γ response which is necessary to inhibit the Th2 response. Consequently, the increase in the Th2 response at the expense of the Th1 response resulted in less-efficient clearance of the infection.