A Chlamydia trachomatis Strain Expressing Ovalbumin Stimulates an Antigen-Specific CD4⁺ T Cell Response in Mice

Antigen-specific CD4⁺ T cells against Chlamydia are crucial for driving bacterial clearance and mediating protection against reinfection. Although the Chlamydia trachomatis protein Cta1 has been identified to be a dominant murine CD4⁺ T cell antigen, its level of expression during the bacterial developmental cycle and precise localization within the host cell are unknown.

ABSTRACT

Antigen-specific CD4⁺ T cells against Chlamydia are crucial for driving bacterial clearance and mediating protection against reinfection. Although the Chlamydia trachomatis protein Cta1 has been identified to be a dominant murine CD4⁺ T cell antigen, its level of expression during the bacterial developmental cycle and precise localization within the host cell are unknown. Newly developed tools for Chlamydia genetic manipulation have allowed us to generate a C. trachomatis strain expressing a heterologous CD4⁺ T cell epitope from ovalbumin (OVA) consisting of OVA residues 323 to 339 (OVA_323–339). By tagging proteins expressed in C. trachomatis with OVA_323–339, we can begin to understand how protein expression, developmental regulation, and subcellular compartmentalization affect the potential of those proteins to serve as antigens. When OVA_323–339 was expressed as a fusion with green fluorescent protein, we found that we were able to elicit an OT-II T cell response in an antigen-dependent manner, but surprisingly, these T cells were unable to reduce bacterial burden in mice. These data suggest that the subcellular localization of antigen, the level of antigen expression, or the timing of expression within the developmental cycle of Chlamydia may play a crucial role in eliciting a protective CD4⁺ T cell response.

INTRODUCTION

Chlamydia trachomatis is an obligate intracellular bacterial pathogen that is the most common cause of bacterial sexually transmitted infection in the United States (1). Although no vaccines exist to combat C. trachomatis, infection can be treated with antibiotics. However, the infection in humans is often asymptomatic and if left untreated can lead to diseases including pelvic inflammatory disease, ectopic pregnancy, and infertility. Chlamydia has a tightly regulated biphasic developmental cycle, during which bacterial gene expression is carefully regulated at each stage. At the start of the developmental cycle, infectious elementary bodies (EBs) induce their own uptake into epithelial cells that line the surface of the genital tract (2, 3). Once inside, Chlamydia EBs differentiate into the noninfectious reticulate bodies (RBs) and establish their replicative niche inside a membrane-bound vacuole called an inclusion. The inclusion grows in size as RBs divide through binary fission. After 24 to 72 h, RBs redifferentiate into EBs and can ultimately escape the host cell either through cell lysis or through extrusion (4).

Experiments conducted in animal models have demonstrated the importance of CD4⁺ T cells in clearing Chlamydia infection, showing that CD4⁺ T cells are necessary and sufficient for clearing Chlamydia infection (5–7). Following infection, these CD4⁺ T cells differentiate into the Th1 subtype, which is characterized by the generous production of the cytokine interferon gamma (IFN-γ) (7, 8). Th1 T cells that produce IFN-γ are critical for the resolution of infection, as mice that lack the receptor necessary to sense IFN-γ exhibit delayed bacterial clearance (9–11). The use of T cell receptor (TCR) transgenic mice has been key to characterizing the antigen-specific CD4⁺ T cell response to this organism. CD4⁺ TCR transgenic cells that are specific to the C. trachomatis protein Chlamydia T cell antigen 1 (Cta1), denoted NR1 T cells, proliferate in response to C. trachomatis and can traffic to the site of infection (12). These NR1 T cells also have been shown to be protective against C. trachomatis genital infection if preskewed to a Th1 phenotype in vitro (11). Very little is actually known about the characteristics of C. trachomatis proteins, like Cta1, that allows them to be recognized by T cells. Cta1 is a predicted periplasmic protein of unknown function (12, 13), and it is known that the NR1 T cells recognize a specific 20-mer peptide (12). However, little else is known about the Cta1 protein, where the 20-mer peptide domain resides within the inclusion, and when it is expressed during the developmental cycle.

Given the importance of the CD4⁺ T cell response in clearing Chlamydia infection, we sought to utilize recently developed tools for genetic manipulation of C. trachomatis (14, 15) to tag a protein expressed in C. trachomatis with a T cell epitope from ovalbumin (OVA) to determine if T cells would recognize and respond to that tagged protein. When green fluorescent protein (GFP) tagged with OVA residues 323 to 339 (OVA_323–339) was expressed in C. trachomatis and used to infect mice, we found that the tagged strain was able to stimulate a robust OVA-specific response but that the responding T cells could not be programmed to be protective. Our data suggest that stimulating a CD4⁺ T cell response to C. trachomatis is more nuanced than simply inducing expression of an epitope within any selected protein and that there are likely multiple antigen characteristics that are important in order for a protective CD4⁺ T cell response to be elicited.

RESULTS

C. trachomatis GFP-OVA expresses the OVA_323–339 epitope and exhibits no growth defects.

Heterologous T cell epitopes can be fused to proteins expressed by microbial pathogens and thereby serve as a proxy to determine the antigenicity of that protein. Recent advances in Chlamydia genetics allow us now to use this approach to track antigen-specific responses. To start, we modified the existing C. trachomatis pSW2-GFP plasmid vector (15), fusing a CD4⁺ T cell epitope from OVA (OVA_323–339) to the 3′ end of GFP, creating a GFP-OVA_323–339 (GFP-OVA) fusion protein (see Fig. S1 in the supplemental material). We transformed C. trachomatis serovar L2 with this pSW2-GFP-OVA_323–339 plasmid and plaque purified the strain, which we have denoted C. trachomatis GFP-OVA. To confirm that the bacteria were able to successfully produce OVA, we infected McCoy cells at a multiplicity of infection (MOI) of 1 and at 24 h postinfection isolated RNA and assessed OVA expression relative to that of GAPDH (glyceraldehyde-3-phosphate dehydrogenase) through reverse transcription (RT)-quantitative PCR (qPCR) (Fig. 1A). We found that the C. trachomatis GFP-OVA strain was able to express OVA, while C. trachomatis GFP was not.

FIG 1.

C. trachomatis GFP-OVA expresses the OVA_323–339 epitope and exhibits no growth defects. (A) McCoy cells were infected at an MOI of 1:1 and harvested at 24 h postinfection for RT-qPCR of OVA expression relative to GAPDH expression. (B) McCoy cells were infected at an MOI of 5:1 and harvested at 24 h postinfection for bacterial burden, determined through qPCR of Chlamydia 16S relative to host GAPDH. (C) Mice infected transcervically with 5 × 10⁶ IFUs of C. trachomatis GFP or GFP-OVA were sacrificed at the indicated time points, and the bacterial burden in the upper genital tract was assessed. In vitro data are representative of those from two independent experiments with three technical replicates. In vivo data are representative of those from at least two independent experiments with 5 mice per group. All data were analyzed using an unpaired t test. *, P < 0.05; N.S., not significant.

We also confirmed that the addition of the OVA epitope did not attenuate C. trachomatis. We infected McCoy cells at an MOI of 5 and at 24 h postinfection isolated DNA and assessed the bacterial burden in cells infected with C. trachomatis GFP or C. trachomatis GFP-OVA using qPCR. We observed no detectable growth differences in vitro between the two strains, suggesting that the addition of the OVA fragment does not impair the ability of C. trachomatis to establish cellular infection (Fig. 1B). To confirm that C. trachomatis GFP-OVA did not exhibit any growth defects in vivo, 6- to 8-week-old female C57BL/6 mice were transcervically infected with 5 × 10⁶ inclusion-forming units (IFUs) of C. trachomatis GFP or C. trachomatis GFP-OVA and the bacterial burden was assessed at days 3 and 7 postinfection using qPCR. As we observed in vitro, there were no significant differences in the bacterial burden in mice infected with C. trachomatis GFP and those infected with C. trachomatis GFP-OVA (Fig. 1C). These data suggest that this strain is neither attenuated nor hypervirulent.

Mice are not protected against C. trachomatis GFP-OVA following immunization with OVA.

We hypothesized that immunizing mice with OVA protein in conjunction with the adjuvant poly(I·C) would elicit protection against subsequent challenge with C. trachomatis GFP-OVA (16, 17). However, after challenge we observed no difference in the bacterial burden in OVA-vaccinated versus unvaccinated mice (Fig. S2), suggesting either that vaccination with OVA protein provides insufficient immunity to protect against the GFP-OVA-tagged strain or that cells infected with the C. trachomatis GFP-OVA strain are not recognized well enough to mediate protection. Consistent with this, we also found that mice vaccinated against Listeria monocytogenes expressing OVA also were not protected against subsequent C. trachomatis GFP-OVA infection (data not shown).

OVA-specific T cells proliferate in response to C. trachomatis GFP-OVA infection but do not elicit protection during systemic infection.

We next sought to determine if C. trachomatis GFP-OVA could elicit an OVA-specific CD4⁺ T cell response. Isolated OVA_323–339-specific TCR transgenic T cells (OT-II T cells) were adoptively transferred into mice and subsequently infected with C. trachomatis GFP-OVA or C. trachomatis GFP intravenously (i.v.). At 4 days postinfection, spleens were harvested, OT-II T cell proliferation was assessed by flow cytometry (Fig. S3), and the bacterial burden was assessed by qPCR. We observed that there were substantially more OT-II T cells in mice that were infected with C. trachomatis GFP-OVA than in those that were infected with C. trachomatis GFP (Fig. 2A), despite transferring the same number of cells 1 day prior to infection. Using carboxyfluorescein succinimidyl ester (CFSE) dilution to measure T cell proliferation, we found that OT-II T cells preferentially expanded only in response to C. trachomatis GFP-OVA (Fig. 2B). These data show that the engineered C. trachomatis GFP-OVA strain is capable of stimulating an OVA-specific CD4⁺ T cell response and does so without affecting the total CD4⁺ T cell compartment (Fig. 2C).

FIG 2.

OVA-specific T cells proliferate in response to C. trachomatis (Ct) GFP-OVA infection but do not elicit protection during systemic infection. Mice received 10⁶ CFSE-labeled CD45.1⁺ OT-II T cells 1 day prior to infection i.v. with 10⁷ IFUs of C. trachomatis GFP-OVA or GFP. At 4 days postinfection, spleens were analyzed by flow cytometry for OT-II T cells (A), proliferating OT-II T cells (B), and total CD4⁺ T cells (C). (D) The bacterial burden was assessed through qPCR. (E) Mice received 10⁶ Th1-skewed OT-II or NR1 cells prior to infection, and the bacterial burden was assessed through qPCR at day 4 postinfection. The results in panels A to D are representative of those from one experiment with at least 4 mice per group. The results in panel E are representative of those from two independent experiments with at least 5 mice used per group. The results in panels A to C were analyzed using an unpaired t test. The results in panels D and E were analyzed using one-way ANOVA. *, P < 0.05; **, P < 0.01; N.S., not significant.

Because C. trachomatis-specific CD4⁺ T cells can mediate the clearance of C. trachomatis infection, we investigated whether OT-II T cells were able to protect mice against infection with the OVA-tagged C. trachomatis strain. We found that prior transfer of naive OT-II T cells did not mediate a reduction in the bacterial burden following i.v. infection of mice with C. trachomatis GFP-OVA (Fig. 2D). However, it has previously been shown that CD4⁺ T cells against the dominant C. trachomatis antigen Cta1 are able to protect mice against infection only when they are preskewed in vitro toward a Th1 subtype (11). Therefore, we tested whether preskewed OT-II T cells could protect mice against C. trachomatis GFP-OVA i.v. infection and found that at day 4 postinfection, there was no significant difference in the bacterial burden in the spleens of mice that received Th1 OT-II T cells versus those that did not, while mice that received Th1 NR1 T cells were protected against infection (Fig. 2E). Taken together, these data suggest that while C. trachomatis GFP-OVA can elicit OT-II T cells to proliferate in response to antigen, OT-II T cells are unable to protect mice against infection with the GFP-OVA-tagged strain.

OVA-specific T cells proliferate in response to C. trachomatis GFP-OVA genital tract infection but do not provide protection.

While i.v. infection can be a useful model, C. trachomatis infection in the genital tract is more reminiscent of infection in humans. We next sought to determine if infection with C. trachomatis GFP-OVA could stimulate an OT-II-specific T cell response in the genital tract of mice. OT-II T cells were adoptively transferred into naive wild-type female mice, and the mice were challenged transcervically 1 day later with C. trachomatis GFP-OVA or C. trachomatis GFP. At 4 days postinfection, the upper genital tracts and draining lymph nodes were harvested and assessed for OT-II T cell homing and proliferation by flow cytometry (Fig. S4). We found that there were more OT-II T cells in the draining lymph node of mice infected with C. trachomatis GFP-OVA (Fig. 3A) and that this corresponded to higher numbers of proliferating OT-II T cells (Fig. 3B) and activated OT-II T cells (Fig. 3C).

FIG 3.

OVA-specific T cells proliferate in response to C. trachomatis GFP-OVA genital tract infection but do not provide protection. Mice received 10⁶ CFSE-labeled CD45.1⁺ OT-II T cells 1 day prior to transcervical infection with 5 × 10⁶ IFUs of C. trachomatis GFP-OVA or GFP. At 4 days postinfection, draining lymph nodes were assessed by flow cytometry for OT-II T cells (A), proliferating OT-II T cells (B), and activated OT-II T cells (C). Upper genital tracts were also assessed by flow cytometry for OT-II T cells (D) and qPCR for the bacterial burden (E). (F) IFN-γ^−/− mice received 10⁶ Th1-skewed OT-II or NR1 cells 1 day prior to transcervical infection with 5 × 10⁶ IFUs of C. trachomatis GFP-OVA or GFP, and the bacterial burden in the upper genital tract was assessed at 3 days postinfection. The results in panels A to E are representative of those from two independent experiments with at least 6 mice per group. The results in panel F are representative of those from one experiment with at least 5 mice per group. The results in panels A to D were analyzed using an unpaired t test. The results in panels E and F were analyzed using one-way ANOVA. *, P < 0.05; **, P < 0.01; N.S., not significant.

However, we found that there were no differences in OT-II T cell recruitment to the genital tract in mice infected with either C. trachomatis strain (Fig. 3D). It is possible that during the inflammatory conditions of infection, OT-II T cells home to the site of infection regardless of antigen specificity. We also found that naive OT-II T cells did not protect mice against C. trachomatis transcervical infection (Fig. 3E). Given that naive T cells against Cta1 do not protect mice (11), the inability of naive OT-II T cells to protect the mice was expected. However, we found that preskewed Th1 OT-II T cells were also not protective compared to preskewed Th1 NR1 T cells (Fig. 3F).

NR1 T cells are activated faster and traffic to the draining lymph nodes and genital tract better than OT-II T cells.

Given the inability of OT-II T cells to protect mice against both C. trachomatis GFP-OVA transcervical and i.v. infection, we tested whether the protective CD4⁺ T cells specific to the dominant CD4⁺ antigen in C. trachomatis, Cta1, home to the genital tract more efficiently than OT-II T cells. We adoptively transferred equal numbers of OT-II (CD45.1⁺) and NR1 (CD90.1⁺) T cells into naive (CD90.2⁺) mice that were subsequently infected transcervically with C. trachomatis GFP-OVA or C. trachomatis GFP. At 4 days postinfection, the draining lymph nodes and upper genital tracts were harvested and processed for flow cytometry (Fig. S5). We found that there were significantly more NR1 T cells than OT-II T cells (Fig. 4A) and that these NR1 T cells were more activated than the OT-II T cells (Fig. 4B) in the draining lymph node. However, there were no significant differences in the numbers of activated OT-II T cells in the lymph node between mice infected with C. trachomatis GFP and mice infected with C. trachomatis GFP-OVA. This is counter to what we observed previously (Fig. 3C). We hypothesize that a lack of antigen-specific OT-II T cell expansion in the lymph node is due to the preferential expansion of NR1 T cells, which appear to be more effective at responding to the GFP-OVA C. trachomatis strain. There were also significantly more NR1 T cells than OT-II T cells in the genital tract (Fig. 4C). We also determined that the bacterial burdens in these two groups of mice were identical (Fig. 4D). Taken together, these data suggest that the NR1 T cells are better at responding to C. trachomatis infection than OT-II T cells.

FIG 4.

NR1 T cells are activated faster and traffic to the draining lymph nodes and genital tract better than OT-II T cells. Mice received 10⁶ CD45.1⁺ OT-II T cells and 10⁶ CD90.1⁺ NR1 T cells 1 day prior to transcervical infection with 5 × 10⁶ IFUs of C. trachomatis GFP-OVA or GFP. At 4 days postinfection, draining lymph nodes were assessed by flow cytometry for NR1 and OT-II T cells (A) and activated NR1 and OT-II T cells (B). Genital tracts were also assessed for NR1 and OT-II T cells (C) and bacterial burden (D). Data are representative of those from two independent experiments with at least 4 mice per group. The results in panels A through C were analyzed using a paired t test, and those in panel D were analyzed using an unpaired t test. *, P < 0.05; **, P < 0.01; N.S., not significant.

Overall, our findings could be partially explained if there was a significant difference in the expression of the Cta1 versus OVA epitope. To test this, we infected McCoy cells and assessed Cta1 and OVA expression by RT-qPCR at 24 h postinfection. Indeed, while Cta1 levels were not different between the C. trachomatis strains (Fig. 5A), we did find that Cta1 transcripts were expressed at a significantly larger amount than OVA mRNA in the C. trachomatis GFP-OVA strain (Fig. 5B). This suggests that the quantity of antigen expression may play a role in determining the levels of responding T cells.

FIG 5.

Expression of Cta1 is higher than that of OVA in C. trachomatis GFP-OVA. RNA was extracted from McCoy cells infected for 24 h at an MOI of 5 and analyzed for expression of Cta1 in C. trachomatis GFP-OVA and GFP (A) and Cta1 and OVA expression in C. trachomatis GFP-OVA (B). Data are representative of those from one experiment with three technical replicates and were analyzed using an unpaired t test. *, P < 0.05; N.S., not significant.

DISCUSSION

Antigen-specific CD4⁺ T cells against Chlamydia have been shown to reduce the bacterial load in vivo (7, 11, 18), highlighting a crucial role for these cells in clearing infection. Fusing heterologous antigens to different Chlamydia proteins would allow us to determine when during the developmental cycle a particular protein needs to be expressed in order to elicit a protective T cell response. Here we conducted proof-of-principle experiments to determine if expression of a known heterologous T cell epitope in C. trachomatis could stimulate an antigen-specific CD4⁺ T cell response.

We took advantage of newly developed techniques for genetic manipulation of C. trachomatis (14, 15) and created a C. trachomatis strain expressing the antigenic CD4⁺ OVA epitope (OVA_323–339) at the C-terminal end of GFP, which is predicted to reside in the cytoplasm of the bacterium (see Fig. S1 in the supplemental material) (15). We found that while OT-II T cells were able to proliferate in response to C. trachomatis antigen (Fig. 2B and 3B), these T cells were unable to protect mice against infection even when preskewed to a Th1 subtype (Fig. 2E and 3F) or when primed against the OVA antigen (Fig. S2). We also found that Chlamydia-specific CD4⁺ T cells against Cta1 were far superior in homing to the draining lymph node and genital tract than OT-II T cells (Fig. 4) and that Cta1 was expressed more robustly than OVA in vitro (Fig. 5B). We cannot definitively conclude that OVA_323–339 does not generate a protective CD4⁺ T cell response because of its reduced expression level compared to Cta1. Other factors, including when during the developmental cycle these proteins are expressed, may also play key roles in determining antigenicity.

Indeed, it is highly likely that the dominance of Cta1 is due to several factors, including levels of expression. We have very little information about when during the developmental cycle Cta1 is produced, and it is possible that the timing of expression is a factor in antigen presentation. In our system, GFP-OVA_323–339 expression is regulated by the promoter for the incDEFG operon, which controls the expression of a set of inclusion membrane proteins that are thought to be expressed early and throughout the developmental cycle (15, 19). Given that Cta1 and OVA_323–339 are under the control of different promoters, they are potentially expressed at different times during the developmental cycle of C. trachomatis. As a result, understanding precisely when Cta1 is expressed in comparison to the timing of OVA_323–339 expression could help elucidate when antigens need to be expressed in order to elicit a robust CD4⁺ T cell response to C. trachomatis.

The predicted cytoplasmic localization of OVA_323–339, compared with the predicted periplasmic localization of Cta1, could also play a role in antigenic dominance. In order to activate CD4⁺ T cells, antigen-presenting cells must acquire antigen through the endocytic or autophagic route and present peptides on major histocompatibility complex class II (MHC-II). It is currently unknown what subcellular location is the most effective for gaining access to the MHC-II presentation pathway during Chlamydia infection. It would be interesting to place OVA_323–339 in different subcellular compartments through fusion with other Chlamydia proteins of known localization or through the introduction of signal sequences. By generating other strains in which OVA_323–339 is localized to the inclusion membrane, host cytosol, bacterial cytosol, and bacterial membrane, we can start to elucidate which compartment (if any) allows for the most efficient MHC-II antigen presentation.

In future work, we also plan to take advantage of heterologous antigens to query the importance of subcellular localization in eliciting CD8⁺ T cell responses. In order for antigen to be presented on MHC-I and activate CD8⁺ T cells, antigenic proteins/peptides need to gain access to the cytoplasm of the host cell (20). The inclusion membrane has access to the cytoplasm of the host cell, and proteins expressed on the surface of the inclusion membrane can be degraded by the proteasome and processed for MHC-I presentation. Indeed, the CD8⁺ T cell antigen cysteine-rich protein A (CrpA) and class I accessible protein 1 (Cap1) have been shown to localize to the inclusion membrane and stimulate a robust CD8⁺ T cell response (21–24). Using heterologous antigens, such as the CD8⁺ T cell epitope from OVA, we can also begin to address what subcellular compartments within the Chlamydia-infected cell are capable of presenting antigen to CD8⁺ T cells.

Although there remain many unanswered questions about the requirements for antigen presentation and CD4⁺ T cell stimulation, we were able to successfully engineer a strain of C. trachomatis that expresses a heterologous antigen. While we were not able to elicit protection against C. trachomatis using OT-II T cells, these OVA-specific T cells were still capable of responding to C. trachomatis in an antigen-dependent manner. This suggests that the use of heterologous antigens in a mucosal pathogen, such as C. trachomatis, could be useful in assessing the requirements, at a molecular level, needed to elicit a protective CD4⁺ T cell response.

MATERIALS AND METHODS

Growth and isolation of bacteria.

Chlamydia trachomatis serovar L2 (434/Bu; ATCC), C. trachomatis GFP (15) (which was generously provided by Isabelle Derré and Raphael Valdivia), and C. trachomatis GFP-OVA were propagated in McCoy cells as previously described (25, 26). Aliquots of purified elementary bodies were stored at −80°C in medium containing 250 mM sucrose, 10 mM sodium phosphate, and 5 mM l-glutamic acid (SPG buffer) and thawed immediately prior to use.

Generation of Chlamydia trachomatis GFP-OVA.

Overlap PCR was used to generate the GFP-OVA fragment prior to ligation into the existing p2TK2-SW2 IncDProm-RSGFP-IncDTerm plasmid, which was generously provided by Isabelle Derré (15). PCR primers were purchased from Integrated DNA Technologies (IDT), and their sequences are listed in Table S1 in the supplemental material. Primers 1 and 4 were used to amplify a section of GFP from p2TK2-SW2 IncDProm-RSGFP-IncDTerm and to add an overlap to the 5′ end of OVA_323–339. Primers 2 and 5 were used to amplify the IncD terminator from p2TK2-SW2 IncDProm-RSGFP-IncDTerm and to add an overlap to the 3′ end of OVA_323–339. Primers 3 and 7 were used to connect the GFP fragment to the OVA_323–339 fragment. Primers 3 and 6 were then used to connect the GFP-OVA_323–339 fragment to the IncD terminator. The DraIII-HF and NdeI restriction enzymes (New England Biolabs, Ipswich, MA) were used to cut the p2TK2-SW2 IncDProm-RSGFP-IncDTerm backbone and the GFP-OVA_323–339-IncDTerm insert. The two pieces were ligated together and used to transform Escherichia coli (New England Biolabs), and the final plasmid (pSW2-GFP-OVA_323–339) was purified using a HiSpeed maxikit (Qiagen). Plasmid identity was verified through GFP screening in E. coli, and the sequence was verified using primer 3. McCoy cells were used in the transformation of C. trachomatis L2 with pSW2-GFP-OVA_323–339, and L929 cells were used for plaque purification as described previously (14, 15, 27).

Mice.

Six- to 8-week-old female C57BL/6 mice and B6.129S7-Ifngtm1Ts/J mice (IFN-γ^−/−) were purchased from The Jackson Laboratory (Bar Harbor, ME). TCR transgenic OT-II mice that recognize OVA_323–339 were purchased from The Jackson Laboratory and backcrossed onto the CD45.1 background (by U. von Andrian). NR1 mice (Thy1.1 background) that recognize the C. trachomatis Cta1 antigen from residues 133 to 152 (Cta1_133–152) have been described previously (12). All mice were housed in the Harvard Medical School Center for Animal Resources and Comparative Medicine, and all experiments were approved by Harvard’s Institutional Animal Care and Use Committee.

T cell adoptive transfers.

OT-II or NR1 CD4⁺ T cells were isolated from secondary lymphoid organs of transgenic mice. For CFSE labeling, isolated cells were labeled with 10 μM CFSE (Invitrogen, Carlsbad, CA) prior to injection as previously described (11). For Th1 skewing experiments, cells were isolated from all secondary lymphoid organs of transgenic mice and purified using a Dynabeads Untouched mouse CD4 cells kit (Invitrogen). T cells were cultured in RPMI 1640 (Invitrogen) supplemented with 10% fetal calf serum, l-glutamine, HEPES, 50 mM 2-mercaptoethanol, 50 U/ml penicillin, and 50 mg/ml streptomycin. T cells were stimulated using irradiated splenocytes from a C57BL/6 mouse that had been pulsed with 5 μM OVA_323–339 peptide or 5 μM Cta1_133–152 peptide and cocultured with the CD4-enriched populations at a stimulator/T cell ratio of 4:1. Cells were incubated with 10 ng/ml interleukin-12 (IL-12; Peprotech, Rocky Hill, NJ) and 10 μg/ml anti-IL-4 (Bio X Cell, West Lebanon, NH) for 5 days to polarize the T cells to a Th1 phenotype. Recipient mice were injected with ∼10⁶ naive or polarized T cells i.v. 1 day prior to infection. For Th1 skewing experiments, wild-type mice were used for i.v. infection, while IFN-γ^−/− mice were used for transcervical infection.

Infection of mice and preparation of tissue.

For i.v. infection, mice were injected via the tail vein with 10⁷ inclusion-forming units (IFUs) of C. trachomatis in 200 μl SPG buffer. To infect the genital tract, mice were treated with 2.5 mg medroxyprogesterone subcutaneously (s.c.) and then infected transcervically with 5 × 10⁶ IFUs of C. trachomatis in 10 μl SPG buffer using an NSET pipet tip (ParaTechs, Lexington, KY) as described previously (7). At specific times postinfection, the spleens, upper genital tract (uterine horns and ovaries), and iliac lymph nodes were harvested. Single-cell suspensions of secondary lymphoid organs were prepared by grinding the tissue between frosted microscope slides. Uteri were minced with scalpels and enzymatically dissociated in Hanks balanced salt solution-Ca²⁺-Mg²⁺ containing 1 mg/ml type XI collagenase and 50 kunitz units/ml DNase for 30 min at 37°C, washed in Ca²⁺- and Mg²⁺-free phosphate-buffered saline containing 5 mM EDTA, and then ground between frosted microscope slides prior to filtration through a 70-μm mesh. For experiments assessing only the bacterial burden, upper genital tracts were harvested from mice and mechanically homogenized using a tissue homogenizer.

Flow cytometry.

Cells were stained with surface and activation markers along with anti-FcRγ (Bio X Cell) following isolation. Cells were incubated with fluorochrome-conjugated antibodies against mouse CD4 (clone GK1.5; BioLegend, San Diego, CA) or CD4 (clone RM4-5; Invitrogen), CD45.1 (clone A20; BD Biosciences), CD45.2 (clone 104; BioLegend), CD90.1 (clone OX-7; BD Biosciences), CD8 (clone 53-6.7; BioLegend), CD44 (clone 1M7; BioLegend), or CD62L (clone Mel-14; BioLegend) and a LIVE/DEAD Fixable Aqua Dead cell stain kit to exclude dead cells (Invitrogen). AccuCheck counting beads (Invitrogen) were used to determine absolute cell counts. Data were collected on an LSR II flow cytometer (BD Biosciences) and analyzed using FlowJo software (Tree Star, Ashland, OR).

Quantitative PCR.

The bacterial burden was determined through quantitative PCR as previously described (28). Briefly, DNA was isolated from tissue samples using a QIAamp DNA minikit (Qiagen), and the Chlamydia 16S rRNA gene and mouse GAPDH were quantified using primer pairs and dually labeled probes (IDT [San Jose, CA] or Applied Biosciences).

Real-time PCR.

Infected McCoy cells were harvested and resuspended in TRIzol reagent (Invitrogen). RNA was isolated using an RNeasy minikit (Qiagen) and then diluted to 4 ng/ml. Real-time PCR using a QuantiTect SYBR green RT-PCR kit (Qiagen) was used to quantify the levels of Chlamydia Cta1 (forward primer, 5′-ATGAACTCCGGAATGTTCCCA-3′; reverse primer, 5′-CTCGTCTTTTACTGTTCCTGA-3′), GFP-OVA (forward primer, 5′-ATGAGTAAAGGAGAAGCACTT-3′; reverse primer, 5′-ACGTCCAGCTTCATTAATTTC-3′), and host GAPDH (forward primer, 5′-GGTGCTGAGTATGTCGTGGA-3′; reverse primer, 5′-CGGAGATGATGACCCTTTTG-3′).

Vaccinations.

Mice were injected with 10 μg poly(I·C) (InvivoGen, San Diego, CA) and 5 μg EndoFit ovalbumin (InvivoGen) s.c. in the scruff of the neck. At 3 weeks postvaccination, mice were boosted with the same amount of poly(I·C) and OVA protein and were challenged 1 week after the final boost with 10⁷ IFUs of C. trachomatis i.v.

Statistical analysis.

Statistical analysis was performed using Prism software (GraphPad). When comparing NR1 and OT-II levels, data were analyzed using a paired t test. All other data were analyzed using an unpaired t test or one-way analysis of variance (ANOVA). Differences were considered statistically significant if the P value was less than 0.05. Data are represented as the mean ± standard error of the mean (SEM).