The Roles of T cells in Bladder Pathologies

Abstract

T lymphocytes play important roles in the skin and mucosal surfaces such as the gut and lung. Until recently, the contributions of T cells to mammalian bladder immunity were largely unknown. With newer techniques, including single-cell RNA sequencing and reporter mice, an understanding is emerging of T cell roles in bladder diseases (e.g. bacterial infections, bladder cancer, and chronic inflammation). In these pathologies, many bladder T cell responses can be harmful to the host, including the suboptimal clearance of bacteria or cancer cells, or by modulating autoinflammation. Recent findings suggest that T cell behavior might be influenced by resident T cell interactions with the bladder microbiota and other immunostimulants. Thus, regulating bladder T cell functions might emerge as a putative form of immunotherapy to treat certain bladder diseases.

T cells in human bladder diseases

T cells modulate adaptive immune responses in the skin and mucosae, including the gut and lung (1–6). The roles of T cells in bladder immunity have remained largely unknown. This is surprising considering that the bladder is one of the most commonly infected mucosal surface in the human body (7) and also one of the most frequent site of cancer in humans (8). Bladder infections are bacterial infections mostly caused by uropathogenic Escherichia coli (UPEC) and its incidence rate is much higher in females than in males (11, 12). Bladder cancers include urothelial carcinoma, squamous cell carcinoma, adenocarcinoma, and others. Urothelial carcinoma, also known as transitional cell carcinoma (TCC), is the most common type in humans (8). The bladder is also afflicted with multiple inflammatory disorders of unknown etiology, the best known of which is interstitial cystitis (IC) (See Glossary) (9).

Recent studies in humans and mice have revealed distinct activities of different T cell subtypes (Box 1) in several major bladder disorders, which for the most part appear to be detrimental to the host, including the suboptimal clearance of bacteria or cancer cells, and/or boosting autoinflammation. Here, we review these recent findings in different bladder disease contexts and highlight distinct T cell functions. We discuss how in some cases the latter may be influenced by the host’s microbiota. These studies raise the intriguing notion that promoting or inhibiting the recruitment of certain T cell subtypes into the bladder might contribute to the development of putative therapeutic strategies to reduce harmful inflammatory reactions in the bladder.

Box 1. T cell subtypes.

T cells are typically activated by the interaction between antigen presenting major histocompatibility complex (MHC) I or II on the surface of antigen presenting cells (APCs) and the T cell receptor (TCR), and co-receptors including CD8 or CD4 on the surface of T cells (86). Depending on the composition of their TCR, mammalian T cells can be divided into αβ T cells or γδ T cells, although more than 90% of T cells in circulation and in lymphoid organs are αβ T cells (87,88). Mucosal surfaces tend to have more γδ T cells compared to the circulation, but the relative abundance of γδ cells is still much lower than αβ T cells. T cells can also be divided into CD8⁺ T cells or CD4⁺ T cells depending on the expression of either CD8 or CD4 markers, γδ T cells are typically CD8⁻CD4⁻. When activated by antigen-MHC I complex, CD8⁺ T cells can extrude cytotoxic molecules such as perforin and granzymes which can directly kill infected or aging tissue cells (86). When activated by the antigen-MHC II complex, CD4⁺ T cells can differentiate into different subtypes such as Th1, Th2, Th17, Treg, etc., with each subtype secreting distinct cytokines such as IFNγ, IL-4, IL-17A, IL-10, etc. These cytokines can promote neighboring immune cells or tissue cells to perform particular functions, such as killing pathogens, repairing damaged tissues, boosting or suppressing inflammation, killing or helping cancer cells (86). Hence, CD4⁺ T cells are known as T helper (Th) cells. A subtype of CD4⁺ T cells can also express granzymes, termed cytotoxic CD4⁺ T cells. Of note, although γδ T cells appear to display as many diverse functions as αβ T cells based on studies in genetically knock-out (KO) mice, their cytokine expression profile is less well characterized, probably due to their low abundance. One of the most well characterized cytokine expressed by γδ T cells in both humans and mice is IL-17A. T cells which haven’t been activated before are known as naïve T cells, and when stimulated by antigens they exhibit a slow response. Once activated, especially via TCR signaling, they differentiate into different subtypes (as mentioned above) of effector cells or memory cells. Memory T cells have a much faster response than naïve T cells, characterized by faster proliferation and cytokine expression, when re-stimulated by the same antigen encountered before.

T cells in bacterial bladder infections

Bacterial bladder infection, also known as cystitis, is associated with rapid bacterial growth in the urine followed by invasion of the superficial epithelium of the bladder (10). The majority (more than 70%) of these infections in human patients are caused by UPEC (11,12). Urinary frequency, urgency, and dysuria are the most common clinical symptoms of cystitis (11). The annual incidence rate among the human population is estimated to be around 3%-12.6% in females and 0.5%-3% in males (11). Besides the relatively high incidence rate, the recurrence rate is also high: In a study in Finland, 44% of women had a recurrence within 1 year after the first cystitis bout (13); In another study in Michigan, the risk of a secondary infection in patients was 24% within 6 months of the first infection (11). Comparatively, the recurrence rate of bacterial infections in the human respiratory tract is only around 10% (14,15,16) and the rate in the gastrointestinal tract is about 1.5% to 12% (16,17,18).

Bladder infections are typically initiated when bacteria reach the bladder lumen through the urethra, multiply in the urine and successfully attach onto the bladder epithelial surface to resist clearance during voiding (10). Uropathogenic bacteria, especially UPEC, express adhesion organelles called fimbriae. Adhesion molecules such as FimH are expressed at the tips of fimbriae to help UPEC attachment and invasion into bladder epithelial cells (BECs) (10). Although the host evokes a vigorous innate immune response to clear both extracellular and intracellular bacteria in the bladder, the high incidence and recurrence rates suggest that the adaptive immune system in the bladder is not capable of fully coping with infecting bacteria. Findings from microscopic examination of the bladders of UPEC-infected mice during infection and after natural resolution has revealed that a small proportion of intracellular UPEC can still persist within BECs and establish long-term quiescent intracellular reservoirs; this in turn, suggests significant defects in bladder adaptive immune responses, and in recognizing and eliminating persistent bacteria (10,12).

Despite the apparent defect in adaptive immune responses in the bladder, the numbers of both CD4⁺ and CD8⁺ T cells in the mouse bladder following UPEC infection – assessed by immunohistochemistry and immunofluorescent staining, have been found to be markedly increased relative to naive mice (19), suggestive of a vigorous T cell response to infection. This observation is further validated by other reports showing that FimH expressed by uropathogens can stimulate murine BECs to recruit a large number of T cells via the secretion of chemokines (Figure 1) (20). Following specific stimulation by FimH, BECs secrete the chemokine stromal cell-derived factor 1 (SDF-1), whose receptor is CXCR4 –the latter being widely expressed by different subtypes of T cells, primarily CD8⁺ and CD4⁺ T cells (20). In this study, large numbers of T cells were recruited into the infected bladder because of SDF-1/CXCR4 signaling (20). Moreover, subcutaneous injection of the CXCR4 antagonist AMD3100 significantly reduced the number of T cells in the infected bladder relative to the infected bladder without AMD3100 treatment (20).

Figure 1. Multiple bladder T cell subtypes are activated by uropathogens with Th2 cells dominating in mice.

When the bladder epithelial cells are infected by bacteria, they are shed massively as a host defense activity to reduce bacterial load. Thereafter, CD11c⁺CD301b⁺OX40L⁺ DCs are activated and migrate to the uroepithelial region following yet-to-be identified signals to sample bacterial antigens. Then, they traffic to the draining lymph node where antigen presentation to T cells occurs. These DCs preferentially activate Th2 cells, which travel to the lamina propria of the bladder to secrete type 2 cytokines such as IL-4, IL-13. These type 2 cytokines will stimulate macrophages and tissue cells to secrete growth factors (EGF, IGF-1, etc.) to rapidly repair and regenerate the uroepithelium. This enhanced Th2 activity inhibits the emergence of Th1 activities, which are important for bacterial clearance. Consequently, bacteria tend to persist in the bladder for extended periods of time following infection. CD8⁺ T cells are also activated and can migrate into bladder to contribute to the clearance of uropathogens. The chemokine SDF-1 is important for the recruitment of CD8⁺ T and CD4⁺ T cells, γδ T cells residing in the lamina propria are activated during infection. They can secrete IL-17A, which can recruit and activate neutrophils to clear bacteria. Abbreviations: DC, dendritic cell; SDF-1, stromal cell-derived factor 1. This figure was created using BioRender (https://biorender.com).

The reason as to why seemingly strong T cell responses to bladder infection induce suboptimal bacterial clearance has remained a puzzle for the past 15 years. Recently, the subtypes of T cells that are activated and recruited into the bladder following infection and that are markedly deficient in bacterial clearance properties were revealed (16,21,22,23). Using IFNγ and IL-4 reporter mice and employing flow cytometry analysis, investigators showed that the recruited CD4⁺ T cells in the infected mouse bladders were highly biased towards an IL-4 secreting Th2 type, rather than a IFNγ secreting Th1 type (Figure 1) (16). Of note, this Th2 biased response – relevant for tissue repair activities rather than bacterial clearance – was further reinforced with subsequent infections by the same pathogen in a UTI recurrence mouse model (16,24). Accordingly, the Th2 bias also seemed prevalent in UTI-prone patients (21). In general, in contrast to IL-4, IFNγ is the major cytokine that promotes clearance of intracellular pathogens through its actions on CD8⁺ T cells, NK cells, and macrophages (25). In addition, Th1 and Th2 cells can mutually inhibit each other’s activation: for example, IL-4 can activate STAT6-GATA3 signaling which inhibits Th1 associated genes (26). Therefore, in a bacterial infection environment where Th2 responses predominate, the activity of Th1 cells can be significantly repressed. In a recent study, II4^−/− mice exhibited significant increases in IFNγ⁺ Th1 cell numbers in the bladder, concomitant with a corresponding decrease in UPEC bacterial load compared to wild type (WT) control mice (16). Taken together, the limited bacterial clearance observed in the mouse bladder following multiple infections might be attributed – at least in part – to a Th2 biased response in this organ (16). Further analysis employing CD301b⁺ dendritic cell (DC) knock-out mice and antibody neutralization of OX40L (16,24) indicated that this Th2 biased response was initiated by a bladder specific population of CD11c⁺CD301b⁺OX40L⁺ DCs (Figure 1). Of note, DCs are a major type of antigen presenting cells (APCs) activating T cells, and the ligands expressed on the surface of DC are crucial for inducing T cell differentiation into particular T cell subtypes (27). Typically, DCs do not express these ligands until they are stimulated by pathogens or other stimuli (27). However, in the bladder, about 50% of CD11c⁺CD301b⁺ DC constitutively express OX40L – a ligand that preferentially polarizes CD4⁺ T cells into a Th2 type in mice and humans (16, 94). When mice are infected by UPEC, these CD11c⁺CD301b⁺OX40L⁺ DCs in the bladder lamina propria quickly migrate towards the epithelium to sample bacterial antigens, then into lymph nodes that drain the bladder to preferentially activate Th2 cells within 24 hours following infection (16,28). Activated Th2 cells are thereafter recruited to the infected bladder, and since DCs in the bladder do not upregulate ligands for Th1 activation, (e.g. DLL1 and DLL4) within 24 hours after infection, the Th1 response has been deemed to be minimal (16). Therefore, the bladder represents a unique microenvironment whereby Th2 cells appear to be preferentially activated and recruited in response to UPEC infection.

Why has a Th2 biased microenvironment evolved in the bladder? One might argue that this might represent a distinct disadvantage regarding its limited bacterial clearance capability. A possible reason may be that when the bladder epithelium is heavily infected by bacteria, the superficial epithelium sheds massively as a defense mechanism to reduce bacterial load in the bladder (29). And, because the bladder needs to regenerate its superficial epithelium and protect underlying tissue from the urine (replete with multiple harmful and noxious agents such as urea and uric acid), the bladder may shift to inducing a Th2 immune response to promote tissue repair (16,24). While this response helps to protect the underlying bladder tissue from salts, urea, and other deleterious agents in the urine, a side effect of this response might be an ensuing limited bacterial clearance. However, this possibility remains hypothetical and certainly warrants direct testing. Nevertheless, it is reasonable to speculate that the bladder might favor actions to protect itself from the harmful action of urine over those of bacterial clearance following bacterial infection, and thus, it may do so by engaging a tissue repair-oriented Th2 adaptive immune response.

In contrast to extensive work investigating αβ T cells in the bladder as mentioned above, γδ T cell responses in the bladder are less well characterized. An early study suggested that γδ T cells were essential for bacterial clearance in the bladder, as TCR δ chain knock-out mice showed an increased UPEC bacterial load after a bladder infection relative to controls (30). Recent data revealed that γδ T cells were the major producers of IL-17A in the bladder (31, 91). Specifically, relative to WT mice, knocking out IL-17A (Il17a^−/− mice) significantly reduced the transcript expression of inflammatory cytokines such as IL-6, TNFα, CXCL10, etc. as measured by qPCR, which significantly increased the UPEC bacterial load in the bladder after a single infection (31). However, it seems that IL-17A secreted by γδ T cells does not significantly affect adaptive immune responses to infection, as Il17a^−/− mice exhibit similar UPEC bacterial loads compared with WT mice after a secondary infection (31).

In view of the significant bladder dysfunction (frequency, urgency, and dysuria) that has been observed in patients afflicted with recurrent cystitis, it is conceivable that this dysfunction is also linked to the notable Th2 responses described above in the murine bladder (16), but this awaits further investigation. Considering the importance of T cell immunity in intracellular bacterial clearance and the high recurrence rate of bladder infection, we posit that specifically modulating T cell responses (and subsets) in the bladder of patients prone to recurrent UTIs might yield therapeutic benefit, which merits further attention.

T cells in bladder cancer

Bladder cancer is the 9^th most common type of human neoplasia with about 430,000 new cases occurring worldwide every year (8,32,33). Most bladder cancers are of the urothelial carcinoma type, also known as TCC (8,32). Based on how invasive the cancer is, bladder cancer can also be classified as non-muscle-invasive bladder cancer– with limited malignancy in the mucosa and submucosa regions, accounting for 75% of patients– and muscle-invasive bladder cancer, in which cancer cells invade the muscle, accounting for 25% of patients (8,32).

Several studies have been undertaken to characterize T cell responses in bladder cancer in patients and animal models, with the goal of developing efficacious treatment strategies. Indeed, recent work detailed single-cell RNA sequencing analysis of T cell populations in bladder tumors from patients (34,35). As with other types of cancer, a diverse range of T cell subtypes were identified, including pro-inflammatory cytotoxic CD8⁺ T cells and anti-inflammatory regulatory T cells (Treg) (Figure 2), revealing notable T cell heterogeneity in bladder tumors (35).

Figure 2. Treg and immunosuppressive signaling in bladder cancer appear to be effective in inhibiting anti-tumor activities even after immunotherapy in both mice and human patients.

Appreciable numbers of CD8⁺, Th1, and cytotoxic CD4⁺ T cells infiltrate bladder tumors to inhibit tumor growth. However, their anti-tumor activities appear to be inhibited by PD-L1/PD-1 signaling and large numbers of Tregs are recruited into the bladder tumor site by CCL17 and S1P. Immunotherapies such as BCG treatment can activate and recruit more anti-tumor T cells. However, they have limited efficacy in reducing Treg numbers or other immune suppressive signaling such as PD-L1/PD-1. Also, ILC2 can induce MDSC via IL-13, which can suppress anti-tumor T cells. Abbreviations: BCG, Bacillus Calmette–Guérin; S1P, sphingosine 1 phosphate; ILC2, group 2 innate lymphoid cells; MDSC, myeloid derived suppressor cells. This figure was created using BioRender (https://biorender.com).

T cells infiltrating solid tumors can either promote (e.g. Treg) or inhibit tumor growth (e.g. cytotoxic T cells) (35). CD8⁺ T cells recognizing tumor associated antigen (TAA) can directly target and kill tumor cells by secreting cytotoxic molecules such as granzymes (36). IFNγ secreting Th1 cells can stimulate multiple anti-tumor activities, such as enhancing tumor antigen presentation, promoting CD8⁺ T cell function, and polarizing macrophages to a pro-inflammatory phenotype, among other examples (37). However, these anti-tumor T cells might be readily susceptible to suppression by the tumor microenvironment. For example, activated T cells typically express programmed cell death protein 1 (PD-1) and most tumor cells express its ligand PD-L1, and the resulting interaction between PD-L1 and PD-1 tends to inhibit the anti-tumor activities of T cells by limiting their effector functions (38). Indeed, blocking inhibitory signaling (e.g PD-L1/PD-1) on anti-tumor immune cells has gathered great research interest in the bladder cancer field. However, in a recently completed, randomized, open-label phase I clinical treatment trial of locally advanced or metastatic solid tumors (661 participants, NCT01375842)^I, in which primary outcome measures included the number of participants with dose limiting toxicities and the percentage of participants with adverse events, etc. and the secondary outcome measures included the percentage of participants with an objective response, etc., anti-PD-L1 anitbody treatment was efficacious in only approximately 20% of anti-PD-L1 treated TCC patients (35, 89). Therefore, more studies in bladder cancer examining PD-1 signaling (44, 90) and other inhibitory signaling molecules such as CTLA4 (43) may be needed to robustly improve the efficacy of immunotherapy for bladder tumors. In this regard, a recent study suggested that the specific presence of cytotoxic CD4⁺ T cells expressing granzyme B and granzyme K might help determine which patients might be responders versus nonresponders to anti-PD-L1 antibody treatment, although the mechanism for this remains elusive (35). Recently, blockade of other receptors such as tumor necrosis factor receptor 2 (TNFR2) (45) with a specific antibody significantly in a MBT-2 murine bladder cancer model inhibited tumor growth by stimulating the proliferation of recruited CD4⁺ T cells and CD8⁺ T cells and by increasing their secretion of IFNγ and Granzyme B relative to a PBS treated group (45). Currently, numerous translational studies are underway using various approaches to block these inhibitory immune signals or to directly enhance anti-tumor immune signals to modulate effector T cell functions, all of which are relevant for cancer immunotherapy endeavors.

As mentioned above, solid tumors are typically infiltrated by a large numbers of Treg cells, which may inhibit the anti-tumor functions of cytotoxic T cells and other immune cells by secreting inhibitory molecules such as IL-10 (39). A recent study on muscle-invasive TCC patients reported that a large number of TAA-specific and TAA-non-specific Tregs existed in blood samples (40). Among several factors, Treg recruitment into bladder cancers might be promoted by overexpression of various chemoattractants such as sphingosine 1 phosphate (S1P) and CCL17 by tumor cells (41,42). Consequently, there is interest in potentially reducing Treg numbers within bladder tumors by blocking their recruitment. A recent study administered anti-CCR4 antibody (the receptor for CCL17) known as mogamulizumab to treat dogs with spontaneous muscle-invasive bladder cancers and mice with canine bladder cancer xenografts (42). Treatment successfully inhibited Treg accumulation in bladder tumors, resulting in reduced tumor growth and significantly improved survival rates relative to dogs or mice without mogamulizumab treatment, suggesting that inhibition of Treg recruitment into bladder tumors might potentially be an effective therapeutic strategy, awaiting further investigation.

For nearly half a century, immunotherapy involving live Bacillus Calmette–Guérin (BCG) has remained the most common treatment for early stage bladder cancer (46). BCG is a live attenuated strain of Mycobacterium bovis, developed a century ago as a vaccine against tuberculosis. In 1976, it was introduced as a treatment for bladder cancer and since then has remained a standard immunotherapy (46). Despite its moderate efficacy, the presumed mechanism of its anti-tumor activities had remained unclear until recently. A study employing a murine bladder cancer model revealed that after BCG treatment, the numbers of both CD4⁺ T cells (especially Th1 cells) and CD8⁺ T cells were markedly increased in tumors relative to those without BCG treatment (Figure 2) (46,47). Consistent with this finding in mice, a nonrandomized phase I open-label study (NCT01498172)^II of 23 patients with non-muscle invasive bladder cancer in which the primary outcome measure was that of adverse events and the secondary outcome measure included flow cytometric analysis of the number of immune cell types in urine, cystoscopic assessment of disease recurrence, etc., showed that the expression of T cell-secreted cytokines such as IL-2 and IL17 were enhanced in the bladder after BCG treatment (48). Although up to 35% of patients were cured with BCG treatment, the remaining patients continued to experience tumor growth and recurrence (49). Several studies have explored why BCG treatment fails in many patients. One study reported that even before BCG treatment, BCG nonresponders presented a higher percentage of Treg cells and PD-L1⁺ cells in their bladder tumors than responders, and that BCG treatment did not significantly change the percentage of these inhibitory cells in nonresponders (49). Another study reported that myeloid derived suppressor cells (MDSC) might negatively affect the outcome of BCG treatment in non-muscle-invasive bladder cancer patients (50). Further analysis indicated that the recruitment of group 2 innate lymphoid cells (ILC2), which exhibit similar cytokine secretion profiles as Th2 cells, correlated with a low T cell/MDSC ratio in nonresponders. Indeed, after BCG treatment, ILC2 secreted IL-13, recruiting MDSC to the bladder, which inhibited both CD4⁺ and CD8⁺ T cell activation (50). Taken together, data from clinical studies suggest that the presence of multiple immunosuppressive cells including certain PD-L1⁺ cells, Treg, ILC2, and MDSC within bladder tumors might be responsible, at least in part, for the suboptimal activation of T cells upon BCG treatment in patients that are not responders. However, what mediates the recruitment of these immunosuppressive cells into the bladder of BCG nonresponders in the first place remains to be determined.

Enthusiasm over new immunotherapeutic approaches have led to the examination of several novel treatments for bladder cancer. One strategy is to explore a potential substitute for BCG. An attenuated Salmonella enterica serovar Typhi Ty21a strain was recently tested and revealed to recruit both CD4⁺ and CD8⁺ T cells into the bladders of non-muscle-invasive bladder cancer patients, similar to that seen with BCG treatment (51). In this study, although several other immune cells were also recruited into the tumor following Ty21a treatment, only CD8⁺ and CD4⁺ T cells were found necessary for the inhibition of tumor growth (51). This conclusion is based on the finding that antibody-depletion of CD8⁺ or CD4⁺ T cells, but not NK cells or neutrophils, decreased the survival rates of Ty21a-treated mice harboring MB49 bladder tumors (51). Since APCs such as tumor infiltrating B cells have been shown to be important for T cell activation in patients with muscle-invasive bladder cancer (52), another novel putative therapeutic strategy may be to enhance APC function by employing antibodies that target co-stimulatory molecules on APC. Specifically, an agonistic CD40 antibody was recently developed which, upon administration in a MB49 murine bladder cancer model, enhanced local APC secretion of IL-12, an important Th1 cell activator. This antibody engaged co-stimulatory molecule CD40 in APCs, resulting in suppressed tumor growth relative to isotype control antibody-treated mice (53). T cell involvement in bladder tumor suppression was verified when antibody mediated depletion of T cells in these tumor-bearing and CD40-antibody-treated mice exhibited lowered survival rates compared to control mice, in which no T cell depletion was performed (53).

Despite various immunotherapy strategies examined to date, it is apparent that a significant proportion of bladder cancer patients consistently fail to respond to any treatment, pointing to the recalcitrance of bladder cancer. Perhaps, combining different therapies may be an incremental approach to address the problem. Additionally, more single-cell analysis of patient samples may help the development of better candidate biomarkers for outcome prediction.

T cells in interstitial cystitis

The bladder is frequently afflicted by various forms of chronic inflammatory syndromes. Most of these conditions, such as chronic urethral syndrome, overactive bladder, vulvodynia, endometriosis, etc., have poor and overlapping clinical definitions, probably because their underlying causes are diverse and unclear but might potentially involve common inflammatory and non-inflammatory processes^IV (54). Perhaps, the best studied is interstitial cystitis (IC) or bladder pain syndrome (9,55,67). IC is a chronic bladder inflammatory disorder that mainly afflicts subjects aged 50 years and above. It is estimated that 10.6 cases per 100,000 subjects are afflicted by this condition (9,55,56). Typical symptoms include urinary frequency, urgency, pelvic pain, nocturia, among others (9,55,56). Initially, a diagnostic criterion for IC is the presence of a distinct inflammatory lesion, Hunner’s ulcer, detected by cystoscopy. Later findings suggested that these “classical” cases might only account for less than 10% of total IC patients (9,55,56). Moreover, clinical studies have revealed that in almost all IC cases, significant infiltration of T cells into the bladder is evident (57,58). However, how these T cells contribute to the progress of IC has remained elusive.

Studies of IC have been severely impaired by the lack of adequate animal models. In recent years, various models of IC have been described (59–63), and they are generally classified into three types based on how they are created. Since IC patients are typically associated with disruption of the bladder epithelial barrier involving thinning of the epithelium or development of lesions such as Hunner’s ulcer, etc., the first model type involves employing small molecules (e.g. protamine sulfate, potassium chloride, etc.) or peptides that directly damage the bladder epithelium in mice (61). For example, a recent study used a small sialoglycopeptide antiproliferative factor (APF), to successfully damage the bladder epithelium in mice by disrupting expression of uroplakin III and ZO-1, which are important structural proteins forming the uroepithelium structure (61). The second type involves the utilization of WT or genetically deficient animals which will naturally develop IC symptoms (60,64). For example, a proportion of WT domestic cats naturally develop IC, in which the bladder epithelium becomes damaged and chronic inflammation arises (64). Another group recently reported that Fcgr2b^−/−Pdcd1^−/− BALB/c mice, which are deficient in type IIb Fc receptor for IgG and PD-1, spontaneously develop IC-like symptoms before being 10-weeks-old, in contrast to WT BALB/c mice (60). Moreover, these mutant mice exhibited bladder associated pathologies consistent with IC, including disruption of bladder epithelium, generation of anti-urothelial autoantibodies, and massive infiltration of immune cells, especially CD4⁺ T cells (60). The third type of IC model involves administration of autoimmune peptides such as peptides derived from uroplakin (65) and the immunogenic peptide T2 (62). Alternatively, it could involve genetic expression of ovalbumin (OVA) in BECs followed by adoptive transfer of ovalbumin (OVA)- specific OT-II CD4⁺ T cells in mice (59, 66). All of these interventions have resulted in strong T cell infiltration into the bladder epithelium accompanied by chronic bladder inflammation as evidenced by histology. These models have successfully recapitulated several aspects of autoinflammatory responses as seen in IC patients, namely, T cell infiltration and bladder epithelial barrier disruption.

The development of these various models has revealed that T cells are essential for the development of bladder tissue damage and IC-like symptoms (Figure 3). Specifically, in nearly all of these models, several T cell subsets, including CD4⁺ , CD8⁺ , and γδ T cells, appear to infiltrate the bladder and contribute to tissue damage, as shown by histology or flow cytometry analysis, similarly to what has been observed in IC patients (58,67) (Key Figure, Figure 4). Studies using these models have provided clues as to how T cells may be recruited into the bladder and how they may contribute to IC progress. For example, damage to BECs appears to be a key triggering event, resulting in the release of chemokines such as CXCL10 by the damaged cells – a finding also seen in urine samples from IC patients (68,69). T cells typically expressing the cognate chemokine receptor CXCR3 are then recruited into bladder (70). T cells which recognize self-antigens such as Uroplakin II, Uroplakin IIIA generated from damaged BECs can selectively aggregate in the epithelium as shown in histologic and gene expression analysis of Uroplakin II- or Uroplakin IIIA-treated vs control protein-treated mice (65, 85). Recognition of self-antigens caused CD8⁺ T cells and Th1 cells to secrete cytotoxic molecules, such as IFNγ and TNFα, which can further damage the bladder tissue by amplifying the chronic inflammation (65,66,71). A phase III interventional clinical study of 14 IC patients treated with suplatast tosilate (IPD-1151T, a new immunoregulator that selectively suppresses CD4⁺ T cells) (JapicCTI-060257)^III, in which the primary outcome measure was improvement of the IC symptom score and incidence of adverse drug reactions, also suggested that Th2 derived IL-4 might significantly contribute to IC, although the mechanisms remain unclear (72). T cell secreted IFNγ can stimulate epithelial cells to secrete CXCL10, which in turn recruits more Th1 cells (70); T cell secreted TNFα may stimulate mast cells via the receptor for TNFR1 but not TNFR2, which can in turn induce mast cells to produce massive amounts of TNFα, further enhancing T cell activity in murine IC models, such as the model utilizing OVA expression in BEC (71,73). With these positive feedback loops, more T cells and other immune cells including mast cells can accumulate in the bladder and cause chronic inflammatory symptoms resulting in IC in murine models (65,66,71). Many of these observations such as robust T cell infiltration, mast cell activation, and TNFα secretion, are also supported by a wealth of data from IC patients (58,67,68,69,72,74). Recognizing the pivotal roles of T cells, researchers are actively exploring new therapeutic strategies for IC by applying T cell inhibitory molecules. One group tested RDP58, a novel D-amino acid decapeptide effective in treating human ulcerative colitis and chemotherapy-induced diarrhea, in the BEC expressing OVA murine IC model. Researchers found that RDP58 treatment could significantly reduce T cell numbers in the bladder, decrease the production of TNFα, and limit bladder tissue damage in vivo as evidenced from histology analysis (75). In vitro RDP58 treatment of cultured OVA antigen-stimulated T cells has also shown that this agent can inhibit IFNγ and TNFα secretion from T cells (75). Another group treated IC patients with IPD-1151T (JapicCTI-060257)^III, found that IC symptoms such as urinary urgency, frequency, and lower abdominal pain were significantly mitigated after treatment using 2 sample Wilcoxon and student t statistical tests (72).

Figure 3. Interaction between various T cell subtypes and mast cells promote progression of interstitial cystitis in both mice and human patients.

Although the etiology of IC remains unclear, it is possible that this syndrome is triggered by damaged superficial epithelial cells which release selfantigens and powerful chemokines such as CXCL10. These chemokines can recruit a collection of distinct subtypes of T cells including CD8⁺, Th1, Th2, and γδ T cells, etc. T cells such as self-antigen recognizing CD8⁺ T cells and Th1 can secrete pro-inflammatory cytokines such as IFNγ and TNFα. Some of these cytokines (e.g. TNFα) in turn can activate mast cells, causing them to release more TNFα and other pro-inflammatory cytokines. In addition, IFNγ can stimulate the epithelium to secrete more CXCL10. These positive feedback loops can amplify and sustain tissue inflammation, causing more local damage resulting in bladder pathologies such as bladder pain syndrome. Although Th2 cells and γδ T cells are also involved in this inflammation process, their contribution is less well understood. This figure was created using BioRender (https://biorender.com).

Key Figure, Figure 4. Various T cell subtypes can be recruited into the bladder to promote disease in mice and humans.

Th2, CD8⁺ T cells, γδ T cells and a limited number of Th1 cells can be recruited into the bladder following bacterial infections, resulting in accelerated repair of the bladder epithelium but suboptimal bacterial clearance. Although CD8⁺ T cells, cytotoxic CD4⁺ T cells, and Th1 cells infiltrate bladder tumors to combat the tumor, a large number of Tregs are also recruited which, for the most part, can suppress the anti-tumor activities of recruited T cells. Various T cell subtypes including CD8⁺ T cells, Th1, Th2, and γδ T cells are activated in bladder when the epithelium is damaged. These recruited T cells appear to amplify local dysregulated inflammatory reactions, resulting in interstitial cystitis. This figure was created using BioRender (https://biorender.com).

The above research suggests that T cells play various roles in IC, but many questions remain unanswered. First, if disease-causing T cells only infiltrate the bladder when self-antigens and chemokines are released by damaged BECs, what is the initial cause of BEC damage? Several possibilities have been proposed (9,55), including routine renewal of BEC, or the presence of environmental pollutants, as well as mediators released from nerves such as substance P. These relatively minor stimuli might result in the release of small amounts of self-antigen and damage signals, which can be captured and greatly amplified by T cells (9,55). We speculate that another possibility to help explain the seemingly pathologic roles of T cells in IC may be related to the interactions of T cells with different immune cells. But if so, what are these immune cells? As mentioned above, T cells can communicate with mast cells through TNFα signaling (71,73). But whether this results in pathogenic outcomes remains to be more robustly assessed. Finally, because IC describes a heterogeneous collection of symptoms, more in-depth studies in patients and animal models are required to clarify whether distinct subtypes of IC exist, and the roles that T cells play in these conditions.

Concluding Remarks

The prevalence of bladder diseases such as bacterial infection, bladder cancer and chronic inflammation such as IC continues to grow as the population ages. Recent studies clearly show the importance of T cells in the progression of each of these diseases. During bacterial bladder infections, CD8⁺ T cells, γδ T cells, many Th2 cells and a smaller number of Th1 cells are recruited into the bladder to mediate immune responses (Key Figure, Figure 4). Overall, this response appears to be overwhelmingly biased towards tissue repair, and consequently, bacterial clearance (e.g. of UPEC) is suboptimal, which can promote recurrence of infection. Modulation of these T cell responses towards a more balanced one in UTI prone patients by vaccination or immunotherapy might potentially be an approach to prevent infection recurrence. In bladder cancer, both tumor-fighting CD8⁺ T cells, cytotoxic CD4⁺ T cells and Th1 cells, as well as immunosuppressive Tregs infiltrate tumors simultaneously in significant numbers (Key Figure, Figure 4) (35). For the most part, the immunosuppressive activities of Treg and immunosuppressive signaling emanating from the PD-L1/PD-1 pathway appear to dominate within bladder tumors so that even popular immunotherapies such as anti-PD-L1 antibody treatment may have limited efficacy in most patients (35,43,44,49). Bladder cancer patients display a wide range of responsiveness to immunotherapy, therefore, understanding the underlying basis for this heterogeneity in responsiveness could uncover new avenues for study. In IC, as a result of damage incurred to the bladder epithelial barrier, large numbers of self-antigen recognizing T cells including CD8⁺, Th1, Th2, and γδ T cells (Key Figure, Figure 4) can be recruited into the bladder epithelium by, as yet, unknown mechanisms (60–74). These T cells can further amplify inflammation and increase tissue damage (60–74). Several preliminary studies suggest that modulating T cell activities might be effective in treating IC but remain to be rigorously tested. A major limitation with studies relating to IC is that its symptoms overlap with several other chronic bladder inflammatory syndromes, which makes elucidation of its etiology a challenge (see Outstanding Questions). Recently, the existence of microbiota in the bladder of healthy humans was reported (Box 2) (76–80). Considering that T cell responses are highly regulated by microbiota in both humans and in mouse models, it would be interesting to explore the interactions between the bladder microbiota and local bladder T cells with the goal of exploring their therapeutic potential (Box 2). In summary, various T cell subsets can contribute significantly, albeit in different ways, to bladder diseases. However, further in-depth studies on the nature of these activities are needed and are eagerly awaited to envision the development of candidate efficacious immunotherapies that might target specific bladder pathologies

Outstanding Questions.

Is it possible to increase bacterial clearance and reduce infection recurrence by modulating the bladder microenvironment to switch from a predominantly Th2 type to a Th1 type?

What are the molecular mechanisms leading to the development of the Th2 biased microenvironment in the bladder, other than constitutive expression of OX40L on DCs?

Why is there so much heterogeneity in the numbers of immunosuppressive Tregs recruited into the bladder of bladder cancer patients?

Are there any promising biomarkers to might help predict the outcome of bladder cancer immunotherapies (e.g. anti-PD-L1 antibody treatment)?

How do T cells interact with other bladder immune cells to promote interstitial cystitis?

Would it be efficacious to purposefully recruit Treg populations into the bladder to suppress interstitial cystitis?

Can bladder infections, bladder cancer, and interstitial cystitis be modulated through manipulation of the bladder microbiota?

Box.2. The effect of bladder microbiota on T cell responses.

In view of the recent discovery of the presence of a distinct microbiota in the bladder, the contribution of this endogenous microflora to bladder physiology and immunity is of growing interest. Previously, it was believed that urine in healthy bladders was invariably sterile, and subjects who occasionally exhibited positive bacterial urine cultures without inflammatory symptoms were clinically diagnosed as experiencing asymptomatic bacteriuria (11). However, with improvements in bladder urine collection, it is clear that a distinct bladder microbiota exists in healthy individuals (76–80). The microbiota at various regions of the body are known to play a crucial role in educating memory T cells in the blood as well as in lymphoid and non-lymphoid organs (81–83). This interaction between microbiota and T cells is thought to be key in determining the nature and magnitude of subsequent immune responses of the host (83), because microbiota educated T cells can exhibit broad cross-reactivity against different bacteria, parasites, and viruses (81–83, 92). Recent studies in germ-free mice and WT mice suggest that the bladder microbiota might also contribute to the education of local T cells (16). These studies indicate that Th2 type memory cells, possibly primed by bladder microbiota, migrate into bladder draining lymph nodes to establish long term residence (16). Upon UPEC infection of the bladder, these memory T cells are activated by APC resulting in their rapid migration into the bladder to repair damaged bladder epithelium (16). Although it is unclear if they were initially induced by normal bladder microbiota, the presence of small numbers of Th1 and CD8⁺ memory T cells have been reported in the lymph nodes draining the bladder of mice (16,47). Instillation of BCG into the bladder has been shown to moderately increase the number of these preexisting IFNγ secreting Th1 and CD8⁺ cells in mouse models of bladder cancer (47), revealing that bladder microbial flora might dictate prevalence of T cells in draining lymph nodes. However, how T cells are primed by microbiota in the bladder and how bladder microbiota-educated memory T cells engage in immune responses in the bladder are less well understood. Currently, it is hard to assess the specific effects of bladder microbiota on local T cell functions as the interactions between T cells and microbiota elsewhere in the body might also influence what happens in the bladder (81–83). One approach to addressing this question might be to perform total bladder transplantation between regular mice and germ-free mice (84). This might allow to study the putative functional contributions of the bladder microbiota to various bladder activities. Studying these interactions is still at an early stage, and the development of new models and innovative techniques may greatly benefit such work.

Highlights.

During bacterial bladder infections in mammals, the local Th2 response (important for epithelial repair), can be enhanced. This response can negatively impact Th1 responses which are important for bacterial clearance. Consequently, bladder infections are not completely cleared and recur.

In bladder cancer, a diverse T cell population is recruited into the bladder, including pro-inflammatory cytotoxic T cells (presumably beneficial) and anti-inflammatory regulatory T cells (Treg, presumably detrimental). Immunotherapies directed at either promoting anti-tumor T cells or depleting immune inhibitory cells or signals, are under extensive study.

T cells are central to promoting bladder inflammation in interstitial cystitis in patients and animal models that replicate this bladder disorder. Strategies that aim to Inhibit T cell activity and recruitment to the bladder appear to be a promising treatment for IC.

Glossary

αβ T cells

One TCR is composed by two peptide chains chosen from four chains, α, β, γ, δ. If the TCR on a T cell is composed by α chain and β chain, this T cell is a αβ T cell

CD11c⁺CD301b⁺ DC

Dendritic cells are a type of professional antigen presenting cell mostly expressing CD11c. CD301b is a marker for the DC subtype specifically activating CD4⁺ T cells to a Th2 phenotype

Chemokine

proteins secreted by tissue cells or immune cells, which can attract immune cells; can bind one or multiple subtypes of chemokine receptors. Immune cells which express the cognate chemokine receptor can sense the chemokine and migrate towards the high concentration of this chemokine

Chronic urethral syndrome

featured by inflamed or irritated urethra without bacterial infections

Dysuria

Discomfort or pain when urinating

Fimbriae

short, hair-like structure on the surface of bacteria

Group 2 innate lymphoid cells

innate lymphocytes harboring a similar cytokine secretion profile as Th2 cells

γδ T cells

If the TCR on a T cell is composed by γ chain and δ chain, this T cell is a γδ T cell

Interstitial cystitis

This term could be interchangeably used with bladder pain syndrome: a chronic bladder inflammatory disorder with clinical symptoms including pelvic pain, urinary frequency, urgency, among other

Mucosa and submucosa

The bladder is composed by several layers of tissue. From bladder lumen to exterior: mucosa, submucosa, muscularis, and serosa or adventitia. Mucosa refers to the uroepithelium and underlying connective tissue harboring immune cells (lamina propria). Muscularis refers to the thick muscle layers of bladder. Any connective tissue between the mucosa and muscularis is referred as submucosa. However, some researchers do not recognize the submucosa as a separate layer, and any connective tissue between the uroepithelium and muscularis may be included in the mucosa layer

Myeloid derived suppressor cells

heterogeneous population of myeloid cells expanding in cancer, including macrophages, dendritic cells, granulocytes, etc. They can suppress T cell responses through production of several molecules such as nitric oxide, reactive oxygen species, peroxynitrite, etc. but their roles remain rather controversial

Nocturia

Frequent urination at night

Overactive bladder

Describes the symptom of sudden urge to urinate, which may or may not involve inflammation

Th1 type

CD4⁺ T cells secreting IFNγ under the regulation of the master transcriptional factor T-bet

Th2 type

CD4⁺ T cells secreting IL-4, IL-5, IL-13 under the regulation of the master transcriptional factor GATA3

Treg cells

CD4⁺ T cells secreting IL-10 under the regulation of the master transcriptional factor Foxp3

Tumor associated antigen

preferentially expressed in tumor cells in high amounts, although normal cells can express low amounts of these antigens too

Vulvodynia

Chronic pain in the vulva