CCR7 Deficiency Allows Accelerated Clearance of Chlamydia from the Female Reproductive Tract

An effective immune response to mucosal infection depends on prompt activation of the immune system within local draining lymphoid tissue and the subsequent directed migration of effector cells to the site of infection (1, 2). Following infection of the female reproductive tract (FRT), pathogen-specific lymphocyte activation and clonal expansion are initiated within local lymph nodes (LNs) (3–5). Naive T cells encounter cognate Ag carried to the LNs by APCs that have migrated from the inflamed local tissue (4). These activated T cells subsequently undergo functional maturation and acquire the ability to leave the LN and enter infected tissues to combat pathogen replication (1, 2). After the primary infection is resolved, the magnitude of the effector T cell population shrinks markedly, leaving memory T cells in lymphoid and nonlymphoid tissues to provide defense against re-encounter with the same pathogen (2, 6). Thus, the immune response to initial FRT infection involves a coordinated series of events in which naive lymphocytes within local LNs generate effector populations that mediate pathogen clearance and provide long-lived memory.

Chlamydia trachomatis is the most common bacterial sexually transmitted infection in the United States, causing pelvic inflammatory disease and infertility in otherwise healthy women (7, 8). The development of an effective vaccine against Chlamydia is a public health priority, and greater understanding of protective immunity in the FRT will be an essential component of this process (9, 10). The mouse model of Chlamydia infection involves intravaginal infection of inbred mouse strains with the mouse pneumonitis biovar of C. trachomatis, now called C. muridarum (11). C. muridarum rapidly infects vaginal and cervical epithelial cells and ascends the reproductive tract, where it causes upper reproductive tract pathology and postinfection infertility that resemble Chlamydia-associated disease sequelae in women (11, 12). Protective immunity to C. muridarum requires CD4 T cells, although Ab and CD8 T cells can contribute to bacterial clearance during secondary infections (5, 13–16). The development of FRT pathology in the mouse model correlates with bacterial burden, the infiltration of neutrophils, and the production of inflammatory mediators downstream of TLR activation (17–19). Thus, an effective Chlamydia vaccine that maximizes CD4-mediated protection and reduces pathology will require greater understanding of Chlamydia-specific T cell biology within the FRT. In particular, the role of ectopic lymphoid tissues in the FRT is poorly understood. Although the FRT lacks organized lymphoid structures, immune-inductive sites are formed in the tissue after the resolution of infection (20, 21). These ectopic structures can contain naive lymphocytes and are thought to play an important role in secondary immunity (21). Given the absence of organized lymphoid tissues in the FRT, it seems possible that these organized structures are important in accelerating immune responses to heterologous challenge infections. However, the role of ectopic lymphoid tissues in protection against Chlamydia infection has not been carefully examined.

The chemokine receptor CCR7 allows lymphocytes and dendritic cells (DCs) to recognize CCL19 and CCL21 and, thus, sense LN–derived chemokine gradients (22, 23). CCR7 expression is induced on DCs following innate activation and plays an essential role in DC homing to the draining LN to initiate T cell responses (24). CCR7 is also expressed on lymphocytes and is required for LN entry and appropriate anatomical positioning within the LN (22, 23). Therefore, CCR7-deficient mice display defective LN architecture and have a reduced number of lymphocytes in LNs (25). In addition, CCR7-deficient mice display ectopic lymphoid structure within mucosal tissues, such as lung, stomach, and colon (22, 26). Thus, these mice provide a useful model to examine the importance of lymphoid tissue organization in defense against pathogen challenge. The outcome of infection in CCR7-deficient mice varies considerably, depending on the nature of the pathogen studied and the route of challenge infection (27–31). Given recent data suggesting that a protective memory response to Chlamydia infection relies largely upon tissue-resident CD4 T cell populations within the FRT (32), it is of interest to examine how ectopic lymphoid tissues in the FRT of CCR7-deficient mice influence genital Chlamydia infection.

In this article, we report that, under steady-state conditions, CCR7-deficient mice display a marked increase in lymphocytes within the FRT. Following intravaginal Chlamydia infection, CCR7-deficient mice develop dysregulated CD4 T cell and Ab responses that involve a reduction in draining LN responses, combined with enhanced FRT Chlamydia-specific T cell activation and cytokine production. This robust CD4 response in the FRT correlated with rapid bacterial clearance and lower reproductive tract pathology.

C57BL/6 and CCR7-deficient mice [B6.129P2(C)-Ccr7^tm1Rfor/J] mice were purchased from The Jackson Laboratory (Bar Harbor, ME). All mice used for experiments were 8–16 wk old, unless specifically noted. Mice were maintained under specific pathogen–free conditions, and all mouse experiments were performed in accordance with University of California, Davis Research Animal Resource guidelines.

C. muridarum strain Weiss was cultured in HeLa 229 cells in Eagle’s MEM (Invitrogen) supplemented with 10% FCS. Elementary bodies (EBs) were purified by discontinuous density gradient centrifugation, as previously described, and stored at −80°C (33). Purified EBs were titrated by infection of HeLa 229 cells and enumeration of inclusions that were stained with anti-Chlamydia MOMP Ab (Mo33b) (34). A fresh aliquot was thawed and used for every infection experiment. Heat-killed EBs (HKEBs) were prepared by heating purified EBs at 56°C for 30 min.

Mice were synchronized in a diestrus state by s.c. injection of 2.5 mg Depo-Provera (GREENSTONE), 7 d prior to intravaginal infection. For infection, 1 × 10⁵ C. muridarum in 5 μl of SPG buffer were deposited directly into the vaginal vault using a pipette tip. To enumerate bacterial shedding, vaginal swabs were collected and disrupted with glass beads suspended in 1 ml of SPG buffer, and serial dilutions were plated on HeLa 229 cells. To enumerate the bacteria burden within tissues, the upper FRT (ovaries, oviducts, upper one third of uterine horn), the lower FRT (vagina, cervix, lower one third of uterine horn), spleen, and draining LNs were homogenized in SPG buffer, and the tissue homogenate was placed in 2-ml tubes with glass beads. After shaking for 5 min, samples were centrifuged at 500 × gravity for 10 min, supernatants were collected, and serial dilutions were plated on HeLa 229 cells.

Tetramer staining for Chlamydia-specific CD4 T cells was carried out as previously described (5). Spleen and LNs were harvested from naive or infected mice, and single-cell suspensions were prepared in FACS buffer (PBS with 2% FCS) containing Chlamydia MHC class II tetramer in Fc block (culture supernatant from the 24G2 hybridoma, 2% mouse serum, 2% rat serum, and 0.01% sodium azide) for 1 h at room temperature in the dark. Cells were washed, and tetramer⁺ cells were enriched via magnetic selecting LS MACS columns using anti-fluorochrome magnetic beads (Miltenyi Biotec, Auburn, CA). The resulting bound and unbound fractions were stained using a panel of Abs (listed below) and analyzed on a FACSCanto or an LSRFortessa flow cytometer (BD Biosciences, San Jose, CA). To stain for intracellular transcription factors and cytokines, surface stained cells were fixed, permeabilized, and stained using a Foxp3 staining kit (eBioscience, San Diego, CA). Abs used included FITC-CD11b (M1/70), CD11c (N418), F4/80 (BM8), B220 (RA3-6B2), TNF-α (MP6-XT22), PerCP–eFluor 710–CD4 (RM4-5); allophycocyanin-CCR7 (4B12); eFluor 660–T-bet (4B10); Alexa Fluor 700–CD44 (IM7); eFluor 450–CD3 (145-2C11), Foxp3 (FJK-16S), IFN-γ (XMG1.2) (eBioscience); and FITC–IL-17A (TC11-18H1) and allophycocyanin–Cy7–CD8 (53-6.7) (BD Biosciences). Data were analyzed using FlowJo software (TreeStar, Ashland, OR). Endogenous tetramer-specific CD4 T cells were identified using a previously described gating strategy (35).

Mice were infected with Chlamydia, as described above, and mononuclear cells from the FRT were isolated, as previously described (36). Briefly, the mouse vagina, cervix, uterine horns, and oviducts were recovered, minced into small pieces, and digested with collagenase IV (500 mg/l; Sigma-Aldrich), with stirring for 1 h at 37°C for two rounds. Tissue debris was filtered, and mononuclear cells were separated using a Percoll gradient (GE Healthcare). Five million purified cells were incubated in the presence of HKEBs in 96-well round-bottom plates. After 72 h of incubation at 37°C, supernatants were collected and added to 96-well ELISA plates (Costar, Corning, NY) that had been precoated with purified anti–IFN-γ (R4-6A2), anti–IL-2, anti–IL-4, anti–IL-6, anti–IL-17A, anti–IL-23 (eBioscience), or anti–IL-10 (R&D Systems). Cytokine production was detected using biotinylated Abs specific for each cytokine (eBioscience), followed by the addition of ExtrAvidin-Peroxidase and TMB substrate (Sigma-Aldrich, BD Biosciences). Developed ELISA plates were analyzed using a spectrophotometer (SpectraMax M5; Molecular Devices), and cytokine concentrations were calculated according to standard curves. In other experiments, stimulated mononuclear cells recovered from the FRT were stimulated with HKEBs for 48 h and stained using Abs specific for CD4, CD44, IFN-γ, and IL-17A, and examined by flow cytometry, as described above.

Mice were bled retro-orbitally, and vaginal washes were collected at 14 d postinfection. Serum and wash fluid were analyzed by ELISA for the presence of Chlamydia-specific Abs. Briefly, serial dilutions of serum samples were added to HKEB-coated ELISA plates (Costar). Chlamydia-specific Abs were detected using biotinylated isotype-specific Abs (eBioscience, BioLegend) and ExtrAvidin-Peroxidase substrate (Sigma-Aldrich).

Mice, naive or infected intravaginally with C. muridarum as described above, were sacrificed at 7 d postinfection by CO₂ asphyxiation and cardiac exsanguination. The FRT (ovary, oviduct, uterus, cervix and vagina) was immersion fixed in 10% neutral buffered formalin. Fixed tissues were processed routinely, embedded in paraffin, sectioned at 5 μm thickness, and stained with H&E. Histopathologic evaluation was performed by a board-certified veterinary anatomic pathologist (D.M.I.). Masking and randomization of samples were done prior to histopathologic scoring. The samples were evaluated for the presence and severity of acute inflammation, chronic inflammation, erosion, dilation, and fibrosis. Acute inflammation was defined by neutrophilic infiltration and edema. Chronic inflammation was defined by lymphohistiocytic infiltration. Erosion was defined by the loss of mucosal epithelial cells, typically with breach of the basement membrane. Dilation was defined by distention of the lumen. Fibrosis was defined by an increase in fibroblasts or an increase in collagenous connective tissue. Parameters were evaluated using an ordinal scoring system based on lesion severity and distribution on a 0 to 4–point scale. Semiquantitative histopathologic scores were analyzed for statistically significant differences (Mann–Whitney U test) using GraphPad Prism 6.0.

Statistical analysis was performed using an unpaired t test for normally distributed continuous variable comparisons and a Mann–Whitney U test for nonparametric comparisons (Prism; GraphPad).

It has been previously demonstrated that CCR7 deficiency results in ectopic formation of lymphoid structures in the mucosal tissues, such as stomach, lung, and colon (26). To examine whether lymphocyte aggregates also form in the FRT under noninflammatory conditions, we analyzed the frequency of lymphocytes in secondary lymphoid organs and FRTs of naive wild-type (WT) and CCR7-deficient mice. Comparable numbers of lymphocytes, including B cells, CD8 T cells, and NK cells, were found in the spleen of CCR7-deficient mice and WT controls, although a slightly higher number of CD4 T cells was detected in CCR7-deficient mice compared with WT controls (Fig. 1A). The draining iliac LNs (DLNs) of FRT in CCR7-deficient mice were smaller (data not shown), and the lymphocyte counts were drastically reduced in these mice compared with WT mice (Fig. 1B). In contrast, higher numbers of CD4 and CD8 T cells were recovered from the FRT of CCR7-deficient mice (Fig. 1C), demonstrating that lymphocyte accumulation occurs in the FRT, as reported for other mucosal tissues (26). Interestingly, although most CD4 T cells in the FRT of WT mice displayed an effector memory T cell phenotype (CD44^hiCD62L^lo), a significant portion of CD4 T cells in the FRT of CCR7-deficient mice were naive CD4 T cells (CD44^loCD62^hi) (Fig. 1D). In contrast, the frequency of naive CD4 T cells was reduced in the DLNs of CCR7-deficient mice (Fig. 1D), consistent with previous reports showing that memory cells use alternate receptors to enter LN high endothelial venules (37, 38). These results demonstrate that CD4 T cells accumulate within the nonlymphoid FRT of uninfected CCR7-deficient mice.

To determine whether the altered lymphocyte distribution in the FRT of CCR7-deficient mice could affect the host immune response, we examined Chlamydia-specific CD4 T cells in lymphoid tissues after C. muridarum infection. To accomplish this, we used a Chlamydia-specific MHC class II tetramer that directly detects endogenous CD4 T cells specific for C. muridarum polymorphic outer membrane protein G-1 (PmpG-1_303–311:I-A^b) (5). At 14 d postintravaginal infection, PmpG-1–specific CD4 T cells had expanded equally in the spleen of WT and CCR7-deficient mice (Fig. 2A). In contrast, clonal expansion of PmpG-1–specific CD4 T cells was markedly reduced in the draining LNs of infected CCR7-deficient mice compared with WT controls (Fig. 2A). The decreased Chlamydia-specific CD4 T cell response in the LNs of CCR7-deficient mice mirrored an overall reduction in leukocytes within these same LNs (Fig. 2B).

We next examined whether the elevated numbers of CD4 T cells in the FRT of CCR7-deficient mice would enhance the overall response to Chlamydia infection. Total lymphocytes were purified from the FRT of WT and CCR7-deficient mice and stimulated ex vivo with HKEBs to generate an Ag-specific response. Lymphocytes from the spleen and LNs of WT and CCR7-deficient mice produced similar patterns of cytokine production, although a small increase in IL-17 production was detected in CCR7-deficient lymphoid tissues (Fig. 3A, 3B). However, lymphocytes from the FRT of CCR7-deficient mice produced a more prominent IFN-γ, IL-17, and IL-10 response than WT mice (Fig. 3A–D). When examined by flow cytometry, CD44⁺ CD4 T cells produced elevated levels of IFN-γ after stimulation with HKEBs (Fig. 3E, 3F), suggesting that CD4 T cells are more active in the FRT of these mice. Interestingly, the level of IL-6 production was reduced after in vitro stimulation of lymphocytes from CCR7-deficient and WT mice (Fig. 3C). No IL-4, IL-12, or IL-23 production was detected in lymphocyte cultures from WT and CCR7-deficient mice (data not shown). Thus, FRT lymphocytes from Chlamydia-infected CCR7-deficient mice produce heightened IFN-γ and IL-17 responses but have low IL-6 and high IL-10.

Next, we directly examined the pathology scores of histological sections of the FRT recovered from WT and CCR7-deficient mice 7 d post–C. muridarum intravaginal infection. This analysis revealed a reduction in acute inflammation and erosion in tissues from CCR7-deficient mice compared with WT mice (Fig. 4). Thus, the altered production of cytokines from FRT lymphocytes correlates with a reduction in tissue pathology during Chlamydia infection of CCR7-deficient mice.

To determine whether the altered CD4 T cell response in the LNs and FRT of CCR7-deficient mice also affected Ab production, we examined Chlamydia-specific Ig isotypes 14 d after intravaginal infection with Chlamydia. WT and CCR7-deficient mice developed elevated titers of Chlamydia-specific IgM and IgG isotypes; however, we observed lower levels of IgM, IgG2c, and IgG3 in Chlamydia-infected CCR7-deficient mice, whereas IgG1 and IgG2b levels were modestly elevated compared with WT mice (Fig. 5A–E). Surprisingly, although Chlamydia-specific IgA was not detected in WT serum, a systemic Chlamydia-specific IgA response was evident in CCR7-deficient mice (Fig. 5F); however, IgA was not elevated in vaginal washes from infected CCR7-deficient mice (Fig. 5G). Overall, these results suggest that a heightened immune response occurs at mucosal surfaces in CCR7-deficient mice (Figs. 1–3), whereas the increased systemic IgA response may indicate colonization of Chlamydia in the intestine (39, 40).

Given the increased number of Chlamydia-specific CD4 T cells and enhanced IFN-γ and IL-17 production observed in CCR7-deficient mice, we hypothesized that this would be beneficial for bacterial clearance from the FRT. Therefore, we examined the shedding of Chlamydia from vaginally infected WT and CCR7-deficient mice over the course of primary infection. As expected, WT mice shed bacteria for a prolonged period and eventually resolved primary Chlamydia infection around 4–5 wk postinfection (Fig. 6A, 6B). In marked contrast, CCR7-deficient mice resolved Chlamydia infection at a faster rate, with all mice clearing bacteria within 20 d of intravaginal infection (Fig. 6A, 6B). Furthermore, when the bacterial burden in the upper and lower FRT tissue was directly examined after tissue homogenization, accelerated bacteria clearance was detected in the upper and lower FRT of CCR7-deficient mice (Fig. 6C, 6D). In addition, we monitored the possibility that bacteria could have disseminated from the FRT of CCR7-deficient mice, but we did not detect greater numbers of bacteria in systemic sites (data not shown). Thus, CCR7-deficient mice are more capable of resolving primary Chlamydia infection within the FRT.

A localized bacterial infection causes activation of tissue innate immune cells and initiates DC homing to the draining LNs (24). Bacterial Ags are presented to low-frequency naive T cells in the paracortex of the LN, initiating rapid T cell activation and clonal expansion (1, 41). These expanded T cells gain relevant effector capabilities and traffic to the infected tissues where bacterial killing can occur (42, 43). In all of these events, the efficiency of the immune response depends on appropriate anatomical positioning of immune cells, which is guided by chemokine gradients (44).

The homeostatic chemokine receptor CCR7 plays multiple roles in the process of immunity to infection by guiding activated DCs to the LN T cells area, positioning naive T cells within this same region, and allowing activated B cells to interact with activated T cells at the follicular border (22, 23). Despite this critical role in host immunity, CCR7 deficiency has variable effects on the immune response to different pathogens. Although CCR7-deficient mice display reduced clonal expansion of CD8 T cells to lymphocytic choriomeningitis virus and Listeria infection, viral and bacterial clearance occurred, and protective memory was induced in both cases (27, 28). Thus, CCR7 deficiency modestly impedes the ability of CD8 T cells to eliminate these pathogens. In a mouse model of Mycobacterium tuberculosis infection, CCR7-deficient mice actually had lower bacterial burdens in the spleen at later time points of infection, although this difference was not observed in the lungs (30). A similar reduction in bacterial burden was observed in the mesenteric LNs of CCR7-deficient mice following Salmonella infection (31), although this was attributed to the role of CCR7 in recruiting infected DCs to the lymphoid tissues. Thus, CCR7-deficient mice have profound deficiencies in the generation of T and B cell responses but display only modest deficiencies in the clearance of some pathogens.

Our findings add to these reports by showing that CCR7-deficient mice have accelerated clearance of Chlamydia from the FRT after intravaginal infection. It seems likely that the enhanced efficiency of Chlamydia clearance largely derives from the elevated number of lymphocytes within the FRT prior to infection, although it should be noted that these mice also have a deficiency in DC migration and regulatory T cell function (25, 45). Given impaired bacterial shedding in CCR7-deficient mice, efficient Chlamydia growth may rely upon the absence of organized lymphoid tissues within the FRT of WT laboratory mice. The formation of FRT tertiary lymphoid clusters is a hallmark of protective memory (20), suggesting that the anatomical proximity of lymphocytes to infected epithelia is critical for rapid clearance. Recent data suggest an explanation for this observation, because these organized structures likely play an important role in retaining protective tissue-resident memory T cells within the FRT (32, 46). However, our data add to this by suggesting that lymphocyte accumulation within the FRT is protective prior to the generation of specific memory. It should be noted that this accumulation of lymphocytes is most likely due to the inability of effector or memory cells to egress to draining LNs (47), thus generating a larger pool of effector and memory T cells in the tissue. Therefore, it will be of interest to determine whether development of tertiary lymphoid tissues by heterologous FRT infection can accelerate Chlamydia clearance.

Our data show that CCR7 deficiency impedes the expansion of Chlamydia-specific CD4 T cells within the local draining LNs. This finding contrasts somewhat with a study of systemic Listeria infection in which Ag-specific CD8 T cells priming was reduced, but CD4 responses were unaffected (28). It is not yet clear whether the reduced Chlamydia-specific CD4 T cell response is due to inefficient DC migration to the LN or to the disrupted lymphoid architecture within secondary lymphoid tissues of CCR7-deficient mice. Either way, this deficiency did not prevent the development of an efficient protective response to Chlamydia in the FRT, a model that is dependent on CD4 T cells for primary clearance (48).

Our data examining effector development of Chlamydia-specific CD4 T cells within the LN and local tissue show an elevation in IFN-γ and IL-17 production, consistent with a study reporting higher IFN-γ and IL-17 in the gastric mucosa of CCR7-deficient mice (38). Another study has suggested that CCR7 signaling favors Th2 differentiation and IL-4 production (49); however, in the Chlamydia model, IL-4 production by FRT lymphocytes was barely detectable. In contrast, a protective role for Chlamydia-specific CD4 Th1 cells has been clearly demonstrated using gene-deficient mice and Ab depletion (14, 15, 50–52), and expanded T-bet⁺ Th1 cells have been detected in the draining LNs of vaginally infected mice (5, 53). Our understanding of IL-17 in immunity to Chlamydia is less well developed, and a recent study reported that IL-17 responses were low in seropositive women (54). However, CD4 Th17 cells have been detected in pulmonary and genital Chlamydia infection models (5, 55, 56). In the pulmonary model, IL-17 neutralization reduced Th1 responses and impeded bacterial clearance but otherwise enhanced immune pathology (55, 57), suggesting that Th17 cells could be protective by promoting Th1-mediated immunity. Other vaccine studies have noted that IFN-γ/IL-17 double-producing cells correlate with protective immunity (58). Thus, the enhancement of Chlamydia-specific IFN-γ and IL-17 detected in CCR7-deficient mice seems likely to explain the rapid bacterial clearance from the FRT of these mice.

An interesting feature of the rapid Chlamydia clearance in CCR7-deficient mice is the fact that this occurred with a corresponding reduction in FRT pathology. This also correlated with lower IL-6 production and greater IL-10 production from lymphocytes recovered from the FRT, two cytokines associated with inflammatory and anti-inflammatory responses, respectively. The generation of CD4 regulatory T cell populations has been reported during Chlamydia infection (59–61), but otherwise the suppression of inflammatory responses is incompletely understood. One possibility is that highly effective bacterial clearance by expanded Th1 and Th17 responses simply reduces the bacterial Ag available to drive innate immune activation.

In summary, our data show that CCR7-deficient mice display a marked increase in naive lymphocytes within the FRT and that they develop dysregulated CD4 T cell and Ab responses to genital Chlamydia infection. Surprisingly, these mice develop enhanced adaptive immune responses within the FRT and rapidly clear Chlamydia, with reduced reproductive tract pathology. These data suggest that efficient T cell trafficking to the FRT encourages bacterial clearance and should be a key goal of vaccine development.

We thank Dr. Richard Morrison and Sandra Morrison for helpful discussions and Dr. Harlan Caldwell for the kind gift of C. muridarum mAb.

The authors have no financial conflicts of interest.