T cell responses to Chlamydia

ABSTRACT

Chlamydia trachomatis is the most commonly reported sexually transmitted infection in the United States. The high prevalence of infection and lack of a vaccine indicate a critical knowledge gap surrounding the host's response to infection and how to effectively generate protective immunity. The immune response to C. trachomatis is complex, with cells of the adaptive immune system playing a crucial role in bacterial clearance. Here, we discuss the CD4⁺ and CD8⁺ T cell response to Chlamydia, the importance of antigen specificity and the role of memory T cells during the recall response. Ultimately, a deeper understanding of protective immune responses is necessary to develop a vaccine that prevents the inflammatory diseases associated with Chlamydia infection.

Here, the authors review the functions by which T cells limit Chlamydia trachomatis infections as well as the ways this pathogen inhibits T cell immunity.

INTRODUCTION

Chlamydia trachomatis is an obligate intracellular Gram-negative bacterium that primarily infects mucosal epithelial cells. Chlamydia species grow inside a membrane-bound vacuole called an inclusion and have a biphasic developmental cycle, where the bacteria alternate between the elementary body and the reticulate body (Ward 1983; Bastidas et al. 2013; Elwell, Mirrashidi and Engel 2016). Genital Chlamydia infection is the most commonly reported bacterial sexually transmitted infection in the United States with an estimated 3 million new cases each year (Wiesenfeld 2017; Centers for Disease Control and Prevention 2018). While the numbers are striking—539.9 cases per 100 000 people—the overall reported rate of Chlamydia infection in the United States drastically underscores the higher rates of infection in women, which are about twice as high compared with men (Centers for Disease Control and Prevention 2018). Young women in particular have extremely high rates of infection—in 2018, 4% of women aged 20–24 tested positive for Chlamydia infection (Centers for Disease Control and Prevention 2018).

Despite these high rates of infection, there is no vaccine against Chlamydia. While readily treated with antibiotics, around 75% of genital infections are asymptomatic and are therefore at a high risk for being left untreated (Farley, Cohen and Elkins 2003; Mylonas 2012; CDC 2017). Additionally, antibiotic treatment does not preclude women from subsequent infections, and repeated or untreated genital infection can lead to excessive inflammation in the upper genital tract resulting in multiple disease sequelae, including pelvic inflammatory disease, which can have long-lasting consequences on reproductive health (Risser and Risser 2007; Hosenfeld et al. 2009; Brunham, Gottlieb and Paavonen 2015). Ultimately, an ideal prophylactic vaccine would prevent not only infection but also the downstream inflammatory diseases associated with Chlamydia infection (Starnbach 2018). In order to generate protection against future infections, most vaccines aim to target the adaptive immune system, as it is the arm of the immune system that is critical for forming immune memory. Indeed, the adaptive immune response to C. trachomatis is absolutely essential for generating future protection. Much of the work on immunity to C. trachomatis has been conducted in mice, and these studies will be the focus of this review. Here, we will focus on the importance of T cell-driven immunity and provide perspective on the different hurdles that need to be overcome in order to generate protective immunity against C. trachomatis.

OVERVIEW OF MURINE ADAPTIVE IMMUNITY AGAINST CHLAMYDIA

Components of the innate immune system are often critical to jump-start adaptive immunity. During Chlamydia infection, natural killer cells (NK cells) traffic to the infected tissue early during infection and are the primary early producers of interferon-γ (IFNγ), which helps to shape the adaptive immune response (Tseng and Rank 1998). Other innate immune cells, including CD103⁻ dendritic cells, have been shown to acquire C. trachomatis antigen in the genital tract and then traffic to the draining lymph node to activate T cells (Stary et al. 2015).

While the innate immune system can rapidly detect and respond to Chlamydia via various pattern recognition receptors and resulting cytokine production (Darville et al. 2003; O'Connell et al. 2006; Derbigny et al. 2012; Barker et al. 2013; Massari et al. 2013), the adaptive immune system is required to mediate Chlamydia clearance and protection against reinfection. Despite the fact that Chlamydia are obligate intracellular bacteria, they can activate multiple arms of the adaptive immune system. Humoral immunity, mediated by B cells, can lead to the production of antibodies. In the context of C. trachomatis infection, B cells appear to be dispensable for clearance during primary infection (Morrison, Feilzer and Tumas 1995). There is some evidence to suggest that B cells may play an important role in clearing secondary C. trachomatis infection, although the mechanism for how this occurs has not yet been elucidated (Su et al. 1997) and is therefore not a focus of this review.

The cellular part of the adaptive immune response is mediated by T cells, which recognize antigen via the membrane-bound T cell receptor (TCR) (reviewed in Davis and Bjorkman 1988; Alcover, Alarcón and Di Bartolo 2018). There are two main subsets of T cells, characterized by the type of TCR present—αβ T cells and γδ T cells—the latter of which are thought to be less critical during Chlamydia infection (Davis and Bjorkman 1988; Perry, Feilzer and Caldwell 1997; Yang, Hayglasst and Brunham 1998). In order for the TCR to recognize foreign Chlamydia antigen, the proteins need to be processed into peptides and presented on major histocompatibility complex (MHC) class I or II. Below, we detail the differences in antigen recognition between MHC I and MHC II, and the importance of antigen-specific responses in combating Chlamydia infection.

CD8⁺ T CELL-MEDIATED IMMUNITY AGAINST CHLAMYDIA

MHC I is expressed on all nucleated cells and classically presents antigen originating from the cytosol to T cells. These cytosolic proteins, which can originate from intracellular pathogens, are degraded by the proteasome and peptides are funneled into the endoplasmic reticulum by the protein TAP. Peptides are then loaded into the binding groove of MHC I by the chaperone protein tapasin and the fully loaded peptide:MHC I complex is exported to the surface of the cell (reviewed in Neefjes et al. 2011). Typically, the TCR on CD8⁺ T cells recognizes antigen presented on MHC I. This recognition results in the upregulation of various effector cytokines, including IFNγ, as well as the release of perforin and granzyme cytotoxic molecules that can result in target cell death (Halle, Halle and Förster 2017). CD8⁺ T cells are therefore thought to play a major role in the immune response to intracellular pathogens, as they can directly deplete the intracellular niche required for pathogen replication.

Indeed, it has been shown that CD8⁺ T cells recognizing C. trachomatis proteins class I accessible protein-1 (Cap1) and cysteine-rich protein A (CrpA) can kill target cells in an antigen-dependent manner (Fling et al. 2000; Starnbach et al. 2003). Despite the ability of CD8⁺ T cells to target infected cells in vitro, CD8⁺ T cells are dispensable for clearance of Chlamydia in mice, as CD8^−/− and perforin-deficient animals clear infection at the same rate as wild-type (WT) mice (Johansson et al. 1997; Perry et al. 1999). Despite this, IFNγ production by CD8⁺ T cells is critical for resolution of Chlamydia infection, as CD8⁺ T cell lines derived from Chlamydia-infected mice are only protective against infection if they can produce IFNγ (Starnbach, Bevan and Lampe 1994; Lampe et al. 1998).

In addition, antigen-specific CD8⁺ T cells can be programmed to be protective against Chlamydia infection. Transgenic NR23.4 T cells that recognize CrpA can protect mice against C. trachomatis when adoptively transferred into mice prior to bacterial challenge (Roan and Starnbach 2006). Additionally, mice that are vaccinated with recombinant vaccinia virus expressing CrpA (VacCrpA) develop a CD8⁺ T cell population specific for CrpA that is protective upon infection with C. trachomatis (Loomis and Starnbach 2006; Zhang and Starnbach 2015). Using MHC I tetramers, it is also possible to track endogenous CrpA-specific CD8⁺ T cells to monitor the kinetics of the C. trachomatis-specific CD8⁺ T cell response over time. For many pathogens, the resolution of primary infection results in the formation of a stable memory T cell population. Following secondary infection, these memory T cells expand more rapidly and to a higher extent compared with primary infection. However, the CD8⁺ T cell response to C. trachomatis infection does not follow this same pattern, as mice that are infected and then re-challenged with C. trachomatis have fewer numbers of CrpA CD8⁺ T cells during the secondary response, suggesting some impairment with natural memory formation (Loomis and Starnbach 2006; Fankhauser and Starnbach 2014).

Recent work has demonstrated that the immunoinhibitory molecule PD-L1 (programmed death-ligand 1) is upregulated on uterine epithelial cells and dendritic cells in the draining lymph node after primary and secondary C. trachomatis infection (Fig. 1). This upregulation skews CrpA-specific CD8⁺ T cells toward a central memory (Tcm) phenotype (Fankhauser and Starnbach 2014). Tcm cells typically circulate through the secondary lymphoid organs (SLOs) and lymphatic vessels but are not well equipped to home to or exert effector functions in peripheral tissues. Effector memory T cells (Tem) are able to circulate from SLOs to peripheral tissues and are thought to aid in clearance of pathogens that infect these tissues (Mueller et al. 2013). When PD-L1 is blocked or deleted in mice during C. trachomatis infection, the T cell population shifts toward a Tem phenotype and C. trachomatis is cleared more rapidly. Additionally, PD-L1 deficiency increases the number of IFNγ producing CD8⁺ T cells during secondary infection and the IFNγ produced by CD8⁺ T cells is sufficient to protect against C. trachomatis infection (Fig. 1) (Fankhauser and Starnbach 2014). However, it is important to consider that immunoinhibitory molecules often act to regulate inflammation by preventing T cell activation and consequent cytokine production. In Chlamydia muridarum-infected mice, blockade of the immunoinhibitory molecules PD-L1 and TIM3 enhances uterine and oviduct pathology (Peng et al. 2011).

Figure 1.

Impact of PD-L1 upregulation on the CD8⁺ T cell response to C. trachomatis. Left: During infection, uterine epithelial cells and lymph node dendritic cells upregulate the immunoinhibitory molecule PD-L1. MHC I presentation of the C. trachomatis antigen CrpA in conjunction with PD-L1 expression results in antigen-specific CD8⁺ T cells being programmed to the Tcm phenotype. Right: When PD-L1 is blocked, antigen-specific CD8⁺ T cells are skewed to the Tem phenotype, resulting in enhanced IFNγ production and Chlamydia clearance.

Although immunoinhibitory molecules have been extensively studied during chronic viral infections and cancer, the upregulation of these molecules during chronic bacterial infections suggests that potentially targeting these pathways to promote T cell reactivation may aid in bacterial clearance but could also contribute to enhanced downstream inflammation. Several studies have asserted that the C. muridarum CD8⁺ T cell response contributes to pathology (Manam, Nicholson and Murthy 2013; Vlcek et al. 2016). However, the basis of these findings is that WT mice have enhanced pathology following infection compared with transgenic mice with fixed CD8⁺ T cell reactivity to ovalbumin (OT-I mice). A comparison of the responses in WT mice versus OT-I mice reflects only the immunopathology caused by OT-I cells and may not reflect the diversity of clonal responses (whether antigen-specific or bystander) stimulated by infection of WT mice. When designing a vaccine against C. trachomatis, it is important to take both of these factors into account in order to adequately generate a long-lasting, effective memory population without excess inflammation that could cause downstream disease.

CD4⁺ T CELL-MEDIATED IMMUNITY AGAINST CHLAMYDIA

While CD8⁺ T cells classically recognize endogenous antigen presented on MHC I, CD4⁺ T cells recognize exogenous antigen presented on MHC II. In contrast to MHC I, MHC II expression is restricted to professional antigen presenting cells (APCs), including dendritic cells, macrophages and B cells (Neefjes et al. 2011). Typically, MHC II presents peptides derived from extracellular proteins that have been endocytosed by APCs and degraded by early endosome proteases. During Chlamydia infection, this can occur through engulfment of extracellular elementary bodies or capture of dying infected epithelial cells. Once the peptide:MHC II complex has assembled, it is then transported to the cell surface to be recognized by the TCR on CD4⁺ T cells (reviewed in Neefjes et al. 2011). Recognition of antigen leads to CD4⁺ T cell differentiation into distinct T cell subtypes (reviewed in O'Shea and Paul 2010), characterized by the upregulation of a master transcription factor that culminates in the production of specific cytokines. CD4⁺ T cell subtype cytokine production ultimately helps to resolve specific types of infection or disease. For example, Th1 cells, which are characterized by their large production of proinflammatory cytokines like IFNγ, are particularly important for clearing viral infections and intracellular bacterial pathogens (Zhu 2018). Indeed, Th1 cells are thought to be the predominant T cell subset in the context of Chlamydia infection. These T cells differentiate following IFNγ and IL-12 production by innate immune cells early during infection (Johansson et al. 1997; Tseng and Rank 1998; Gondek, Roan and Starnbach 2009; Jiao et al. 2011). While there are several other CD4⁺ T cell subtypes, including Th2, Th17, Th22 and Th9 cells (Zhu 2018), it is unclear what role they play during C. trachomatis infection.

CD4⁺ T cells play a critical role during Chlamydia infection. Previous work has demonstrated that CD4⁺ T cells are necessary and sufficient for Chlamydia clearance and protection against reinfection (Su and Caldwell 1995; Gondek et al. 2012). The identification of the CD4⁺ T cell antigen Cta1 (Chlamydia-specific T cell antigen 1) has allowed for further discovery of how these CD4⁺ T cells are capable of clearing infection (Roan et al. 2006). CD4⁺ T cells from TCR transgenic mice that recognize a specific 20-mer peptide from Cta1 (Cta1_133–152), termed NR1 T cells, have been critical for determining the effects of antigen-specific CD4⁺ T cells during C. trachomatis infection. Following genital tract infection with C. trachomatis, NR1 T cells are activated in the genital tract draining iliac lymph nodes (Fig. 2). Upon antigen stimulation, these T cells upregulate the activation marker CD44 and downregulate the lymph node retention marker, CD62L (Roan et al. 2006). In order to traffic to the genital tract, the T cells upregulate the chemokine receptors CXCR3 and CCR5 (Olive, Gondek and Starnbach 2011) and the integrin α4β1 (Fig. 2) (Davila, Olive and Starnbach 2014). These molecules are critical for driving T cell-mediated protection in the genital tract—blockade or deletion of either the chemokine receptors or integrins prevents antigen-specific CD4⁺ T cells from homing to the genital tract, resulting in higher bacterial burden.

Figure 2.

Overview of antigen-specific CD4⁺ T cell immunity to C. trachomatis. Upon adoptive transfer of transgenic NR1 T cells into mice and subsequent transcervical infection with C. trachomatis, NR1 T cells are activated in the draining lymph node. This results in downregulation of the lymph node retention marker CD62L and upregulation of the activation marker CD44. To travel to the genital tract, NR1 T cells also upregulate the integrin α4β1 and chemokine receptors CXCR3 and CCR5. Upon entry into the genital tract, NR1 T cells cluster in specific sections of the genital tract that correlate with where C. trachomatis is located.

T cell activation and trafficking to the site of infection also occur in an antigen-dependent manner. NR1 T cells do not proliferate or become activated in mice infected with Salmonella enterica or Listeria monocytogenes. Similarly, OT-II T cells, which are specific for a peptide fragment in the protein ovalbumin (OVA_323–339) that is not found in C. trachomatis, are unable to become activated and traffic to the genital tract following C. trachomatis infection (Roan et al. 2006). Two-photon microscopy has become a valuable tool for visualizing these CD4⁺ T cell populations in the genital tract in response to C. trachomatis. Using fluorescent NR1 T cells, our lab has shown that the antigen-specific CD4⁺ T cell response to C. trachomatis in the genital tract is not evenly distributed (Helble et al. 2020). Rather, there are distinct sections within the genital tract that contain greater numbers of NR1 T cells (Fig. 2). Sections enriched with NR1 T cells correspond to higher C. trachomatis burdens, suggesting that NR1 T cells are able to home specifically to sections in the genital tract that contain bacteria (Helble et al. 2020). However, it is unknown what drives these gradients and how the NR1 T cells are able to home to specific sections within the genital tract. It is possible that these gradients of antigen-specific CD4⁺ T cells are also reflective of the adaptive immune response to Chlamydia infection in the human female genital tract. There have been a limited number of human studies examining infection and immune responses in the genital tract using stained tissue sections. However, these studies have indicated that lymphoid follicles (similar to germinal centers) do exist in the genital tract tissue (Kiviat et al. 1990b), as well as CD4⁺ T cell clustering in the endometrial tissue (Vicetti Miguel et al. 2013). It is unclear if these patches of immune cells correlate with where the bacteria are localized, but it does provide evidence that the observed NR1 T cell clustering in mice is reminiscent of the immune response in humans. While there has been no direct correlation of adaptive immune cells and C. trachomatis burden in human samples, Kiviat et al. did provide evidence for neutrophil aggregation surrounding C. trachomatis-infected cells (Kiviat et al. 1990a), suggesting that certain immune responses to C. trachomatis may be targeted to specific areas in the genital tract in humans.

Once in the genital tract, NR1 T cells have been shown to be sufficient to protect mice against reinfection (Gondek et al. 2012). It is thought that CD4⁺ T cell-mediated clearance of Chlamydia is driven by IFNγ production from Th1 CD4⁺ T cells. Indeed, NR1 T cells that are pre-skewed as Th2 or Th17 prior to adoptive transfer do not confer protection to mice infected with C. trachomatis (Gondek, Roan and Starnbach 2009). While Th1 T cells can produce other cytokines in addition to IFNγ, the importance of IFNγ production was reenforced when it was discovered that CD4⁺ T cell-mediated clearance of C. trachomatis requires the host to sense IFNγ. Mice that are deficient in the IFNγ receptor (IFNγR^−/− mice) are unable to clear C. trachomatis infection, even in the presence of IFNγ producing CD4⁺ T cells (Gondek, Roan and Starnbach 2009). Additionally, NR1 T cell-mediated IFNγ production but not IFNγ sensing is required for C. trachomatis clearance and preventing NR1 T cell accumulation in the genital tract (Helble et al. 2020).

While the use of NR1 transgenic T cells has been essential to studying antigen-specific CD4⁺ T cells in mice in response to C. trachomatis, the T cells are not cross-reactive to C. muridarum. Although C. muridarum contains an analogous Cta1 protein, there exist several amino acid differences in the 20-mer peptide that NR1 T cells recognize. The differences in these residues may be critical for TCR contact, which could explain why they are not effective against C. muridarum. It is also possible Cta1 is not an effective T cell antigen in C. muridarum. Despite the inability of NR1 T cells to recognize Cta1 in C. muridarum, antigen-specific CD4⁺ T cells are still important for clearance of C. muridarum in mice. Peptides derived from multiple C. muridarum proteins have been shown to bind to MHC II and elicit a CD4⁺ T cell response (Karunakaran et al. 2008, 2015; Yu et al. 2011). Adoptive transfer of dendritic cells pulsed with these peptides also have the potential to protect mice against genital infection, and represent an intriguing approach toward vaccine design (Karunakaran et al. 2008, 2015). In addition, MHC II tetramers against C. muridarum peptides have been useful for studying the kinetics of the CD4⁺ T cell response to infection and the potential for B cell help in shaping protective T cell immunity (Li and McSorley 2013).

In 2017, the development of a new CD4⁺ TCR transgenic mouse specific for C. muridarum has allowed for mechanistic studies on antigen-specific CD4⁺ T cells in this mouse model of infection (Poston et al. 2017). These CD4⁺ T cells become activated in the draining iliac lymph nodes where they upregulate CD44 and downregulate CD62L. Following activation, the T cells traffic to the genital tract using currently unknown chemokine receptors and integrins. Once in the genital tract, these T cells produce IFNγ and can subsequently reduce bacterial burden in mice infected intravaginally with C. muridarum (Poston et al. 2017). Although it is unknown which C. muridarum antigen these T cells recognize, these transgenic T cells are cross-reactive to C. trachomatis (Poston et al. 2017).

The specific requirements for what makes an effective CD4⁺ T cell antigen in the context of Chlamydia infection are unknown. Currently, Cta1 is the most well-characterized CD4⁺ T cell antigen in C. trachomatis (Roan et al. 2006), and very little is known about this protein's function within the infected cell or how it gets processed for antigen presentation on MHC II. Genetic manipulation of Chlamydia is paramount in order to address these questions (Wang et al. 2011; Agaisse and Derré 2013; Johnson and Fisher 2013; Mueller, Wolf and Fields 2016). To assess the role of Cta1 during infection and eliciting a CD4⁺ T cell response, an attractive option would be to simply create a Cta1 deficient C. trachomatis strain. This is theoretically possible using either TargeTron (Johnson and Fisher 2013), a technique involving the insertion of mobile group II introns into the coding sequence of Cta1 and thereby gene inactivation, or FRAEM (fluorescence-reported allelic exchange mutagenesis) (Mueller, Wolf and Fields 2016), using allelic exchange to replace Cta1 with a fluorescent marker, thereby eliminating it from the genome. Alternative approaches involve introducing heterologous antigens that are localized to different sections in the infected cell (host cell cytoplasm, inclusion membrane, etc.) or under different promoters to induce expression at specific points during the developmental cycle in order to determine what role these factors have in proper antigen presentation and stimulation of a protective CD4⁺ T cell response. We have started to address these questions, with the introduction of the MHC II restricted ovalbumin peptide, OVA_323–339, to the cytoplasm of C. trachomatis (Helble and Starnbach 2019). While OT-II CD4⁺ T cells were stimulated in response to C. trachomatis-OVA, these cells could not be programmed to be protective and underscore the complexities surrounding CD4⁺ antigen presentation of a vacuolar pathogen and effective protection against a mucosal pathogen. Future studies should aim to determine when Cta1 is expressed across the developmental cycle, where within the infected Cta1 is expressed, and how this changes between C. trachomatis and C. muridarum in order to identify specific characteristics necessary for appropriate CD4⁺ T cell activation.

IFNγ-MEDIATED CONTROL OF CHLAMYDIA INFECTION

IFNγ produced by T cells can lead to a downstream signaling cascade in infected epithelial cells, culminating in the upregulation of several interferon-stimulated genes (ISGs) that can help control intracellular bacterial replication. Secreted IFNγ binds to the IFNγR complex, comprising two distinct chains (IFNγR1 and IFNγR2) (Pestka, Krause and Walter 2004). This binding event induces the phosphorylation of the two cytoplasmic associated Janus-kinases (JAK), JAK1 and JAK2. JAK2 phosphorylation leads to STAT1 (signal transducer and activator of transcription 1) recruitment and eventual phosphorylation. Dimerized phosphorylated STAT1 can then translocate into the nucleus to act as a transcription factor to upregulate ISGs and other interferon responsive transcription factors (Pestka, Krause and Walter 2004; Bhat et al. 2018).

During C. trachomatis infection in humans, this IFNγ signaling cascade leads to the upregulation of the protein indoleamine 2,3-dioxygenase (IDO) in genital tract epithelial cells (Fig. 3) (Thomas et al. 1993; Chon, Hassanain and Gupta 1996). Once expressed, IDO can degrade intracellular tryptophan stores by catalyzing the breakdown of tryptophan into N-formylkynurenine and kynurenine. Through this, IDO can act to starve C. trachomatis of this essential nutrient, inducing C. trachomatis to enter a state of persistence (Fig. 3) (Beatty et al. 1994). While C. trachomatis replication is limited in the absence of tryptophan, removal of IFNγ and subsequent decreased expression of IDO can cause persistent inclusions to become viable once more (Beatty, Byrne and Morrison 1993). Certain genital tract strains are able to overcome tryptophan depletion, as they express tryptophan synthase and can use exogenous indole to produce their own tryptophan (Aiyar et al. 2014).

Figure 3.

IFNγ signaling induces upregulation of IDO and can impair the CD4⁺ T cell response. In humans, IFNγ production by CD4⁺ T cells can upregulate IDO. IDO breaks down tryptophan into N-formylkynurenine and kynurenine. Because C. trachomatis requires tryptophan to survive, IDO-mediated depletion of this amino acid results in a dormant, persistent inclusion. Kynurenines can go on to inhibit the CD4⁺ T cell response through a variety of mechanisms. Together, upregulation of IDO and inhibition of the CD4⁺ T cell response contribute to the persistence of C. trachomatis in humans.

The by-product of tryptophan catabolism, kynurenine, can have a detrimental effect on the host CD4⁺ T cell response (Fig. 3). Kynurenines can directly and indirectly inhibit the CD4⁺ T cell response through a variety of mechanisms, including limiting their proliferation, promoting the differentiation of naïve T cells into regulatory T cells and enhancing the immunoregulatory function of antigen presenting cells (Terness et al. 2002; Belladonna et al. 2007; Baban et al. 2009; Mezrich et al. 2010). Limited T cell proliferation and regulatory T cell formation would lead to limited IFNγ production. Similarly, enhancing immunoregulatory capabilities of antigen presenting cells could limit T cell activation and proper Th1 T cell formation. Given that CD4⁺ T cells are potent producers of IFNγ, inhibiting this pathway through IDO catabolism of tryptophan can result in the decreased production of IFNγ and ultimately the reactivation of C. trachomatis.

In mice, IDO is neither upregulated by IFNγ nor is it upregulated to sufficient amounts to control pathogen replication during C. trachomatis infection (Roshick et al. 2006). Instead, IFNγ induces the upregulation of a family of interferon-inducible p47GTPases called the immunity-related GTPases (IRGs) (Bernstein-Hanley et al. 2006). The IRG family consists of ∼20 different proteins that all exert cell autonomous control over intracellular pathogen growth. On a molecular level during C. trachomatis infection, murine IRGs work cooperatively to identify the location of the inclusion within the host cell and can then promote the ubiquitination of unknown substrates on the inclusion membrane. As a result, the inclusion membrane breaks, releasing C. trachomatis into the host cell cytosol and ultimately into autolysosomes where the bacteria are degraded (Haldar et al. 2015). In contrast, C. muridarum inclusions in murine cells are not ubiquitinated or targeted for destruction by IRGs (Haldar et al. 2015). This host cell tropism is indicative of the evolution of Chlamydia species that have built up resistance mechanisms against species-specific cell autonomous immunity in host cells.

Species-specific cell autonomous immunity can also account for differences in mouse infection models of Chlamydia. Intravaginal infection with C. muridarum yields a robust infection that is able to ascend to the upper genital tract of mice, in stark contrast to the rapid clearance of C. trachomatis in mice infected intravaginally (Gondek et al. 2012). It is likely that the murine IRGs, to which C. trachomatis is susceptible but C. muridarum is not, play a role in clearing C. trachomatis intravaginally before it can ascend to the upper genital tract of mice. Through removal of these mouse-specific innate immune responses, we can start to build a more accurate mouse model of chronic human C. trachomatis infection. Indeed, when mice lack two important IRG regulatory proteins, Irgm1 and Irgm3, transcervical infection with C. trachomatis results in exacerbated bacterial burden compared with WT mice (Coers et al. 2011). It is unclear if C. trachomatis deposited intravaginally in Irgm1/3^−/− mice would be able to ascend to the upper genital tract, but the removal of these mouse-specific innate immune responses does allow for a better representation of modeled human infection. Species-specific immunity can also be seen with C. muridarum infection in human cells. Although there have been no reports of humans becoming infected with C. muridarum, when human cells are infected with C. muridarum in vitro, C. muridarum can overcome tryptophan starvation caused by IDO expression (Roshick et al. 2006). In summary, IFNγ production by CD4⁺ T cells can bind to the IFNγR on epithelial cells, leading to a downstream signaling cascade culminating in ISG upregulation. In turn, this leads to initial cell autonomous control of Chlamydia infection.

TISSUE RESIDENT MEMORY T CELLS AND MUCOSAL PRIMING

Both naïve and memory antigen-specific CD4⁺ T cells have been studied independently; however, direct comparisons of the two populations in the same animal can determine the specific contributions of each cell type during secondary C. trachomatis infection. Using RNA-sequencing, work from our lab has shown that naïve antigen-specific CD4⁺ T cells in the genital tract draining lymph nodes were more abundant and more proliferative than memory antigen-specific CD4⁺ T cells (Helble, Mann and Starnbach 2020). While these data were surprising, as memory T cells have been shown to proliferate faster in response to antigen (Rogers, Dubey and Swain 2000; Veiga-Fernandes et al. 2000; Hu et al. 2001; Whitmire, Eam and Whitton 2008), it is possible that these results were simply due to the time point and anatomical site investigated. A similar experimental setup can be used to identify transcriptional differences between other T cell populations, and future experiments should aim to compare differences between bulk endogenous CD4⁺ T cells and antigen-specific CD4⁺ T cells in order to identify other genes important for generating protection.

While described above in the context of CD8⁺ T cells, memory CD4⁺ T cells have been traditionally grouped into two separate categories: central memory (Tcm) and effector memory (Tem). Like their CD8⁺ T cell counterparts, CD4⁺ Tcm cells circulate primarily within the lymphatics, while CD4⁺ Tem cells are able to home to peripheral, nonlymphoid tissues (Mueller et al. 2013). As such, it is thought that CD4⁺ Tem cells play a predominant role in clearing genital Chlamydia infections. Recently, a third subset of memory T cells, termed tissue resident memory T cells (Trms), has been discovered. In contrast to Tem cells, which can recirculate into the lymphatics and blood following pathogen clearance, Trm cells remain in nonlymphoid tissue after the pathogen has been cleared and can be maintained at peripheral tissues for long periods of time even in the absence of persisting antigen (Mueller et al. 2013; Shin and Iwasaki 2013; Schenkel and Masopust 2014). Upon re-exposure to pathogens in peripheral tissues, Trm cells act as a first line of defense. These cells are able to respond more rapidly to the pathogenic insult than other memory T cell subsets that need to traffic to the tissue in order to respond.

Similar to Tem cells, Trm cells downregulate the lymph node homing marker CD62L in order to traffic to nonlymphoid organs. In contrast to Tem cells, Trm cells upregulate the C-type lectin CD69 (Schenkel and Masopust 2014). CD69 is briefly expressed on all T cells following initial activation but is then immediately downregulated for Tem and Tcm cells. The role of CD69 in Trm cells is still unclear, but it is known that CD69 expression is also associated with downregulation of sphingosine-1-phosophate receptor 1 (S1P₁) (Shiow et al. 2006; Shin and Iwasaki 2013). S1P₁ expression on T cells is important for T cell egress from the lymphatics, and it is thought that S1P₁ downregulation may be an important step for driving Trm cell formation through preventing recirculation through the lymphatics (Schenkel and Masopust 2014). While less well characterized for CD4⁺ Trm cells, CD8⁺ Trm cells have also been shown to upregulate the αEβ7 integrin (CD103) (Mueller et al. 2013) that plays an important role in Trm cell retention in peripheral tissues. The ligand for CD103, E-cadherin, is expressed on epithelial cells in nonlymphoid tissues (Cepek et al. 1994). As such, it is thought that CD103 expressed on Trm cells can interact with E-cadherin on epithelial cells, leading to Trm cell maintenance in the tissue.

While CD69 (and in some T cell subsets CD103) can be a useful marker to identify Trm cells in the tissue, one of the most definitive methods of proving tissue residence and the importance of Trm cells in clearing infection is through parabiosis studies. Parabiosis is the surgical joining of two organisms resulting in shared circulatory systems in the two conjoined partners (Kamran et al. 2013). Migratory memory T cells (Tem and Tcm cells) are able to circulate between both organisms equally. However, memory T cells that do not travel to the conjoined partner but rather are maintained in the tissue of the host are be considered true Trm cells. Indeed, the discovery of CD4⁺ Trm cells during C. trachomatis infection was established through parabiosis (Stary et al. 2015). Optimal clearance of secondary C. trachomatis infection in mice is dependent on both the establishment of CD4⁺ Trm cells and the recruitment of circulating memory T cells to the upper genital tract mucosa. Despite the fact that circulating memory T cells can contribute to secondary C. trachomatis clearance, mice that only receive circulating memory T cells but not Trm cells are unable to clear secondary C. trachomatis infection as effectively, highlighting the importance of these resident T cells (Stary et al. 2015). The generation of genital tract Trm cells following C. trachomatis infection is dependent on a mucosal priming event. Mice that are infected intranasally or transcervically but not subcutaneously with C. trachomatis are able to establish Trm cells in the genital tract. The importance of Trm cells is not limited to C. trachomatis infection. Protective Trm cells have been well documented during infection with pathogens that infect mucosal peripheral tissues, including influenza virus (lung), herpes simplex virus (skin, genital tract) and Mycobacterium tuberculosis (lung) (Gebhardt et al. 2009; Mackay et al. 2012; Shin and Iwasaki 2012; Iijima and Iwasaki 2014; Perdomo et al. 2016; Pizzolla et al. 2017).

CROSS-MUCOSAL PROTECTION

Mucosal sites are often where pathogens first encounter the host and where pathogens leave the host to infect other individuals. Despite being located in completely different sections of the body, infection or vaccination at one mucosal surface can elicit protection at another, suggesting shared elements of immunity between these two distal mucosal sites. Aside from inducing a Trm population, intranasal infection with C. trachomatis induces both a CD4⁺ and a CD8⁺ T cell response that together can reduce burden in the genital tract (Nogueira et al. 2015). IFNγ also plays a role in eliciting this cross-mucosal protection: mice deficient in IFNγ or treated with an IFNγ depleting antibody are unable to clear transcervical C. trachomatis infection after intranasal priming. The ability of the host to respond to IFNγ is also necessary in order for cross-mucosal protection to occur (Nogueira et al. 2015). This effect has also been demonstrated with other Chlamydia antigens. Mice that are primed with live C. muridarum intranasally are protected against intravaginal C. muridarum infection (Yu et al. 2011). Similarly, mice vaccinated intranasally with recombinant C. muridarum protein CPAF (Chlamydia protease-like activity factor) and then challenged intravaginally had less bacterial shedding and reduced pathology (Murthy et al. 2007). Ultimately, it is clear that Trm and mucosal priming are necessary to generate protection against Chlamydia.

CONCLUDING REMARKS

The prevalence of Chlamydia infections has been steadily on the rise and poses an important public health problem both in the United States and worldwide. The inability of humans to develop protective immunity against Chlamydia has significant ramifications—repeat infections are common and can lead to excessive inflammation, resulting in multiple disease sequelae (Risser and Risser 2007; Hosenfeld et al. 2009). In order to develop an effective vaccine against Chlamydia, it is necessary to understand what constitutes protective immunity versus inflammatory pathology. By enhancing protective immune populations, we can hopefully build an ideal vaccine against Chlamydia, ultimately protecting against both Chlamydia pathogenesis and the associated inflammatory diseases. While this review has largely focused on the protective aspects of antigen-specific T cells following Chlamydia infection, it is important to note that an influx of antigen non-specific bystander CD4⁺ and CD8⁺ T cells has been shown to induce excessive immunopathology in the genital tract of mice (Lijek et al. 2018). Preventing these bulk T cell populations from entering the genital tract has no impact on antigen-specific T cell populations, and can significantly reduce genital tract pathology (Lijek et al. 2018).

Novel imaging techniques and high-throughput sequencing can provide valuable insight into host responses to genital Chlamydia infection. Two-photon microscopy of genital tract tissues allows for visualization of adaptive immune cells in whole tissues. There are multiple applications to this technology, including identifying interactions between antigen-specific T cells and other cells of the genital tract, such as epithelial cells or innate immune cells. Using this in combination with RNA-sequencing, researchers can identify specific components of the adaptive immune response that are critical for promoting protection against Chlamydia infection while simultaneously preventing pathology. Armed with this knowledge, we can build a better model of Chlamydia infection with the ultimate goal of designing a vaccine against this important mucosal pathogen.

ACKNOWLEDGMENT

We wish to apologize to those colleagues whose work we could not cite or discuss due to space limitations.

FUNDING

This work was supported by the National Institutes of Health (AI39558 to MNS).

Conflict of interest

None declared.