Main

Infection of mice with the mouse pathogen Citrobacter rodentium mimics infection with enteropathogenic Eshcerichia coli in humans1. These bacterial pathogens attach intimately to intestinal epithelium and cause subcellular attaching-and-effacing lesions. T helper type 1 cytokines such as interferon-γ and interleukin 12 (IL-12) are important for host defense against attaching-and-effacing pathogens2. The TH17 subset of helper T cells has also been found to be critical for the clearance of C. rodentium3,4. TH17 cells probably function through their secretion of cytokines such as IL-17A, IL-17F and IL-22 (refs. 5,6,7,8,9,10,11,12). IL-22, a member of the IL-10 family of cytokines, signals through the IL-22 receptor (IL-22R) paired with IL-10Rβ13. As IL-22R is expressed exclusively on tissue-resident cells, IL-22 has a considerable effect on epithelial cells and mediates epithelial innate immunity13. IL-22 is critical for mucosal host defense against the Gram-negative bacteria C. rodentium and Klebsiella pneumoniae14,15. IL-17A and IL-17F are also important for intestinal immunity to infection with C. rodentium16. Furthermore, IL-17A or IL-17F acts together with IL-22 to synergistically or additively induce antibacterial peptides in keratinocytes17.

IL-17A and IL-17F belong to the IL-17 family, which also includes IL-17B, IL-17C, IL-17D and IL-17E18,19,20,21. IL-17A (also called IL-17) is the best-characterized member. It functions as an important proinflammatory cytokine by contributing to the pathogenesis of a variety of inflammatory diseases, whereas it is also important for host defenses against bacterial and fungal infections18,19,20,21. IL-17F is the member most similar to IL-17A in sequence. However, genetic studies of mice deficient in IL-17A or IL-17F or both have shown these cytokines have some similar functions but also different functions in the regulation of inflammatory responses, as well as in host-defense systems16,22,23,24. IL-17E (IL-25) has a very different function, in that it is a cytokine that promotes type 2 helper T cells and is involved in the pathogenesis of type 2 helper T cell–related inflammatory disorders such as asthma19. Although IL-17A, IL-17E and IL-17F have been well investigated, the other members of the IL-17 family are poorly characterized18,19,20,21. IL-17B and IL-17C are able to stimulate the release of tumor necrosis factor (TNF) and IL-1β from the THP-1 human monocytic cell line25, whereas adoptive transfer of CD4+ T cells with ectopic overexpression of IL-17B and IL-17C exacerbates collagen-induced arthritis26, which indicates these cytokines may have proinflammatory roles. Therefore, because of the limited studies so far, the physiological functions of IL-17B, IL-17C and IL-17D remain unclear18,19,20,21.

The IL-17R family consists of five members (IL-17RA–IL-17RE)18. IL-17RA (also called IL-17R) is the best characterized and has important roles in the pathogenesis of inflammatory disorders as well as in host defense against bacterial and fungal infections, similar to the functions of IL-17A14,18,27,28. IL-17RA binds with much higher affinity to IL-17A than to IL-17F, whereas IL-17RC 'preferentially' associates with IL-17F rather than IL-17A18,29. However, IL-17RA and IL-17RC can form a heterodimeric complex required for signaling mediated by IL-17A or IL-17F15,29,30,31,32. IL-17A-mediated signaling is unique and tightly controlled18,33,34,35,36,37,38. IL-17RB can bind to IL-17E18,19. Both IL-17RB and IL-17RA are essential for IL-17E-induced activity in airway hypersensitivity39. IL-17RD (SEF) has been found to inhibit signaling mediated by fibroblast growth factor through association with its receptor18,21. IL-17RE is the least-well-characterized receptor of the family. Ectopic overexpression of IL-17RE in a myeloid cell line might promote proliferation in a ligand-independent manner40. The physiological function of IL-17RE remains to be identified. Neither the receptors for IL-17C and IL-17D nor the ligand for IL-17RE have been identified.

Here we show that the orphan receptor IL-17RE is the specific functional receptor for IL-17C. IL-17C activated downstream signaling through the IL-17RE–IL-17RA heterodimeric complex. IL-17RE was specifically required for IL-17C-mediated induction of genes encoding antibacterial peptides, proinflammatory cytokines and chemokines both in vivo and ex vivo. IL-17C was induced substantially in colonic epithelial cells (CECs) but not in colonic stromal cells (CSCs) after challenge with C. rodentium. IL-17C acted in synergy with IL-22 to induce antibacterial peptides in CECs for host defense. Because of blockade of IL-17C-mediated signaling, IL-17RE-deficient mice had much lower expression of genes encoding antibacterial molecules, much greater bacterial burdens and total mortality after infection with C. rodentium. Therefore, our results identify the IL-17C–IL-17RE pathway as a crucial regulator in innate immunity to intestinal bacterial pathogens in mice.

Results

IL-17RE is essential for mucosal immunity to C. rodentium

To investigate the physiological function of IL-17RE, we studied IL-17RE-deficient (Il17re−/−) mice. Genotyping of the Il17re−/− mice showed successful deletion of IL-17RE mRNA (Supplementary Fig. 1). Il17re−/− mice were born normally at the expected Mendelian frequency (data not shown). To investigate the potential function of IL-17RE, we assessed its expression pattern in mouse tissues and found high expression of IL-17RE in the colon (Fig. 1a). IL-17RA is known to have important roles in host defense against bacterial infection14,18,27,28. To determine whether IL-17RE might have a role in intestinal immunity to infection, we challenged wild-type and Il17re−/− mice with C. rodentium. The Il17re−/− mice began to lose weight and died by day 13, whereas wild-type mice did not lose weight and all survived throughout the observation period of 22 d (Fig. 1b). Further phenotypic analysis showed that the spleen and intestine weights of Il17re−/− mice were significantly greater than those of wild-type mice (Fig. 1c,d), and bacterial numbers were considerably higher, as much as 100-fold greater, in the spleen, colon and feces of Il17re−/− mice than in those of wild-type mice (Fig. 1e–g). Consistent with their higher bacterial counts, Il17re−/− mice had enhanced structural disruption of the colon epithelium (Fig. 1h). Together these data suggested that IL-17RE has a crucial role in host mucosal defense against the intestinal pathogen C. rodentium.

Figure 1: IL-17RE is essential for intestinal immunity to infection with C. rodentium.
figure 1

(a) Real-time quantitative PCR analysis of IL-17RE mRNA in various wild-type mouse tissues; mRNA expression is presented relative to expression of the housekeeping gene Rpl13a (encoding ribosomal protein L13a) throughout. (b) Survival (left) and body weight change (right) in 7- to 8-week-old Il17re−/− mice (n = 4) or wild-type mice (WT; n = 5) infected orally with 2 × 109 colony-forming units (CFU) of C. rodentium. *P < 0.05, **P < 0.01 and ***P < 0.001 (Student's t-test). (c,d) Spleen weight (c) and weight of cecum and colon (d) from Il17re−/− mice (n = 6) or wild-type mice (n = 5) at day 0 (uninfected) or day 10 after infection as in b. *P < 0.01 (Student's t-test). (e,f) Bacterial titers in homogenates of spleen (e) or colon (f) from Il17re−/− mice (n = 6) or wild-type mice (n = 5) at day 10 after infection as in b. *P < 0.05 and **P < 0.001 (Student's t-test). (g) Bacterial titers in fecal homogenates at 4–10 d (horizontal axis) after oral infection of wild-type mice (n = 6) or Il17re−/− mice (n = 6) with C. rodentium as in b. *P < 0.05 and **P < 0.01 (Student's t-test). (h) Histopathology of colons obtained from Il17re−/− or wild-type mice at day 0 (uninfected (UN)) or day 10 after infection as in b and stained with hematoxylin and eosin. Original magnification, ×10 (left and middle) or ×20 (right). Data are representative of two independent experiments (mean and s.e.m. in a–g).

Full size image

The much higher bacterial count in the Il17re−/− mice indicated that IL-17RE may regulate the expression of genes encoding antibacterial molecules for host defense. To assess this, we obtained colon RNA from Il17re−/− or wild-type mice at day 10 after C. rodentium infection and analyzed the RNA by real-time PCR for expression of genes encoding antibacterial peptides and proinflammatory cytokines and chemokines, which are important for antibacterial defense. Indeed, expression of most genes assessed, including those encoding antibacterial peptides (S100A8, S100A9, LCN2, RegIIIβ and RegIIIγ), proinflammatory cytokines (TNF and IL-1β) and chemokines (KC, CXCL2 and CCL20), was significantly lower in the infected colons of Il17re−/− mice than in those of wild-type mice (Fig. 2a and Supplementary Fig. 2a). CECs or intestinal epithelial cells are the main components of intestinal barrier that encounters and protects the organism from pathogens1. To determine if IL-17RE regulates genes encoding antibacterial molecules in CECs, we infected Il17re−/− or wild-type mice with C. rodentium, isolated cells from their colons and analyzed gene expression. Expression of genes encoding antibacterial peptides, cytokines and chemokines was much lower in Il17re−/− CECs than in wild-type CECs (Fig. 2b and Supplementary Fig. 2b). These results indicate that IL-17RE regulates genes encoding antibacterial molecules in the intestinal epithelium for mucosal immunity to bacterial pathogen.

Figure 2: IL-17RE is critical for the induction of genes encoding antibacterial molecules during C. rodentium infection.
figure 2

(a) Real-time quantitative PCR analysis of the expression of mRNA encoding antimicrobial peptides, proinflammatory chemokines and cytokines in the colons of Il17re−/− or wild-type mice (n = 6 per group) at day 0 (uninfected) or day 10 after oral infection with 2 × 109 CFU of C. rodentium. *P < 0.05, **P < 0.01 and ***P < 0.001 (Student's t-test). (b) Real-time quantitative PCR analysis of the expression of mRNA (as in a) in CECs of Il17re−/− or wild-type mice (n = 6 per group) at day 0 or day 10 after infection as in a. *P < 0.05 and **P < 0.01 (Student's t-test). Data are representative of two independent experiments (mean and s.e.m.).

Full size image

IL-17C-induced gene expression requires IL-17RE

As IL-17RE was critical for mucosal defense against C. rodentium, we assessed the regulation of all IL-17 ligands (IL-17A–IL-17F) in the colon after infection with this bacteria (Fig. 3a). There was considerable induction of IL-17C and IL-17F at an early time point (day 4), whereas there was considerable induction of IL-17A at a late time point (day 12) after infection. There was only slight upregulation of IL-17B but slight downregulation of IL-17E at the later time point (day 12). IL-17D expression was not regulated much. We observed similar regulation in the cecum of genes encoding those IL-17 ligands after infection, although the gene induction was much lower (Supplementary Fig. 3). As IL-17A, IL-17C and IL-17F were upregulated by infection, we sought to determine if IL-17RE was required for responses to those three cytokines. We isolated CECs of Il17re−/− mice and investigated the possibility of a role for IL-17RE in IL-17A- or IL-17F-mediated gene induction. IL-17A- or IL-17F-induced expression of genes encoding KC, CXCL2 and CCL20 was not obviously altered in Il17re−/− CECs (Supplementary Fig. 4a). Similarly, the induction of genes encoding KC, CXCL2 and CCL20 by IL-17A, IL-17F or TNF alone or by IL-17A together with TNF was not obviously altered in Il17re−/− mouse embryonic fibroblasts (Supplementary Fig. 4b,c), which indicated that IL-17RE was not essential for effects mediated by IL-17A or IL-17F. We then investigated the potential role of IL-17RE in effects mediated by IL-17C. It is not clear what cells IL-17C targets or what kinds of genes are induced by IL-17C. Because the colon had high expression of IL-17RE, where IL-17C was upregulated by bacterial infection, we cultured mouse colon tissues ex vivo and tested their responsiveness to IL-17C and IL-17A. Notably, IL-17C induced the expression of genes encoding antibacterial peptides, proinflammatory cytokines and chemokines in wild-type colon tissues but not in Il17re−/− colon tissues (Fig. 3b), which indicated that IL-17RE is the functional receptor for IL-17C. As expected, IL-17A-induced gene expression was not altered in Il17re−/− colon tissues (Fig. 3c). Together these results suggested that IL-17RE is a functional receptor for IL-17C that can induce genes encoding antibacterial peptides and proinflammatory molecules for host defense.

Figure 3: IL-17RE is required for IL-17C-induced gene expression in colons cultured ex vivo.
figure 3

(a) Real-time quantitative PCR analysis of mRNA encoding members of the IL-17 family in C57BL/6 wild-type mouse colons at day 0, 4, 8 or 12 after oral infection with 2 × 109 CFU of C. rodentium. *P < 0.05 and **P < 0.01 (Student's t-test). (b) Real-time PCR quantitative analysis of mRNA encoding antibacterial peptides, cytokines and chemokines in Il17re−/− or wild-type mouse colons cultured ex vivo and treated for 24 h with IL-17C (100 ng/ml). *P < 0.05, **P < 0.01 and ***P < 0.001 (Student's t-test). (c) Real-time quantitative PCR analysis of the S100A8 mRNA and CCL20 mRNA in Il17re−/− or wild-type mouse colons cultured ex vivo and treated for 24 h with IL-17A (50 ng/ml). Data are representative of three independent experiments (mean and s.e.m).

Full size image

To investigate whether IL-17RE is required for IL-17C-mediated functions in vivo, we generated adenovirus encoding mouse IL-17C or empty virus and injected these intravenously into Il17re−/− mice or wild-type control mice. IL-17C mRNA was efficiently expressed in colon (Fig. 4a) and kidney (Supplementary Fig. 5) with similar expression in both wild-type and Il17re−/− mice. Similar to the results of the ex vivo colon cultures, IL-17C-induced production of mRNA encoding antibacterial peptides, proinflammatory cytokines and chemokines was absent from the colons of Il17re−/− mice, in contrast to the mRNA induction in the colons of wild-type mice (Fig. 4a). Similarly, IL-17C-mediated gene induction was blocked in the kidneys of Il17re−/− mice (Supplementary Fig. 5). Because CECs were responsible for IL-17RE-mediated induction of genes encoding antibacterial molecules in the colon during infection with C. rodentium (Fig. 2b), we assessed the function of IL-17RE in CECs in IL-17C-mediated gene induction. We isolated CECs from Il17re−/− or wild-type mice after adenovirus-mediated IL-17C expression. We consistently found that IL-17C-induced expression of mRNA encoding antibacterial peptides, proinflammatory cytokines and chemokines was absent from CECs of Il17re−/− mice, whereas IL-17C mRNA had similar expression in the CECs of both Il17re−/− and wild-type mice (Fig. 4b). Therefore, our data demonstrated that IL-17RE was required for IL-17C-mediated induction of genes encoding antibacterial molecules for intestinal mucosal immunity in vivo.

Figure 4: IL-17RE is essential for IL-17C-induced gene expression in vivo.
figure 4

(a,b) Real-time quantitative PCR analysis of the expression of mRNA encoding antibacterial peptides, cytokines and chemokines in the colons (a) or CECs isolated from the colons (b) of Il17re−/− or wild-type mice (n = 4 per group) at day 4 after intravenous injection with adenovirus encoding empty vector (Ad-EV) or mouse IL-17C (Ad–IL-17C). *P < 0.05 and **P < 0.01 (Student's t-test). Data are representative of two independent experiments (mean and s.e.m.).

Full size image

IL-17C signals through the IL-17RE–IL-17RA complex

To prove that IL-17RE is the receptor for IL-17C, we checked for direct interaction by glutathione S-transferase (GST) precipitation assay of HEK293 human embryonic kidney cells. IL-17C associated with both IL-17RE and IL-17RA, but not with IL-17RB, IL-17RC or IL-17RD (Fig. 5a). Further analysis showed that IL-17C bound directly to the extracellular domains of IL-17RE or IL-17RA (Fig. 5b,c). As a control, we found that IL-17A and IL-17F associated with IL-17RA and IL-17RC but not with IL-17RE (Fig. 5d and Supplementary Fig. 6a,b). Similarly, IL-17E bound to IL-17RB but not to IL-17RE (Supplementary Fig. 6c). These data indicated that IL-17RE is the specific receptor for IL-17C. IL-17RA interacts with IL-17RB, IL-17RC and IL-17RD18,21, which suggests that IL-17RA may be a shared receptor of the IL-17R family that probably functions through a dimeric receptor complex27. As IL-17C associated with both IL-17RE and IL-17RA, we assessed the potential interaction of IL-17RE with IL-17RA. Indeed, IL-17RE associated with IL-17RA (Fig. 5e), which indicated that IL-17C probably bound to the IL-17RE–IL-17RA heterodimeric complex to activate downstream signaling.

Figure 5: IL-17C binds to IL-17RE and IL-17RA.
figure 5

(a) GST-precipitation analysis (GST ppt) of the interaction between GST-tagged empty vector (GST-EV) or IL-17C (GST–IL-17C) and ectopically expressed hemagglutinin-tagged (HA-) receptors of the IL-17R family (above lanes) in HEK293 cells. IB, immunoblot. (b) GST-precipitation analysis of the interaction between GST-tagged empty vector or IL-17C and ectopically expressed hemagglutinin-tagged IL-17RE (HA–IL-17RE) or its extracellular domain (HA–IL-17RE-dCyt) in HEK293 cells. (c) GST-precipitation analysis of the interaction between GST-tagged empty vector or IL-17C and ectopically expressed hemagglutinin-tagged IL-17RA (HA–IL-17RA) or its extracellular domain (HA–IL-17RA-dCyt) in HEK293 cells. (d) GST-precipitation analysis of the interaction between GST-tagged empty vector or IL-17A and ectopically expressed hemagglutinin-tagged IL-17RE or IL-17RA in HEK293 cells. (e) Coimmunoprecipitation analysis of Flag-tagged IL-17RA (M2–IL-17RA) and hemagglutinin-tagged IL-17RE (HA–IL-17RE) in HEK293 cells. IP, immunoprecipitation; IgG, immunoglobulin G (control); M2, anti-Flag. Data are representative of three independent experiments.

Full size image

It is still unknown what signaling pathways IL-17C can activate18,20,21. IL-17A can activate NF-kB pathways and mitogen-activated protein kinase pathways (p38, Erk and Jnk)36. We sought to determine whether IL-17C activated similar signaling pathways. Because the cell types responsive to IL-17C are unknown, we assessed IL-17RE expression in various mouse colon cell populations, including intraepithelial lymphocytes (IELs), lamina propria mononuclear cells (LPMCs), CSCs and CECs. Although IELs, CSCs and CECs had similar expression of IL-17RA, the expression of IL-17RE was much higher in CECs than in IELs, LPMCs and CSCs (Supplementary Fig. 7), which suggested that CECs were the main IL-17C-responsive cells. IL-17C activated NF-kB pathways (phosphorylation of the NF-kB subunit p65 and the NF-kB inhibitor IkBα) and mitogen-activated protein kinase pathways (phosphorylation of p38, Erk and Jnk) in wild-type CECs, whereas IL-17C-mediated activation of these pathways was absent from Il17re−/− CECs (Fig. 6a), consistent with our data showing that IL-17C-mediated downstream gene induction was absent from Il17re−/− colon tissues. As a control, IL-17A-mediated activation of NF-kB was not altered in Il17re−/− CECs (Fig. 6b). We knocked down IL-17RE with small interfering RNA (siRNA) to determine whether IL-17RE mediated IL-17C signaling in human cells. We tested several gut epithelial cell lines and found that HT-29 human colon cancer cells were responsive to IL-17C (data not shown). Indeed, IL-17C-induced signaling was almost completely blocked in HT-29 cells when IL-17RE was efficiently knocked down (Fig. 6c,d), whereas IL-17A-, IL-17F- and IL-17E-mediated activation of NF-kB was not altered (Supplementary Fig. 8a–c); this suggested IL-17RE was also specifically required for IL-17C signaling in human cells. IL-17C-mediated signaling was also considerably suppressed when IL-17RA was efficiently knocked down (Fig. 6c,e). Together our data suggested that IL-17C activates downstream signaling through the IL-17RE–IL-17RA heterodimeric complex.

Figure 6: IL-17RE is indispensable for IL-17C-induced signaling pathways.
figure 6

(a,b) Immunoblot analysis of lysates of primary CECs from Il17re−/− and wild-type colons, treated for 0–30 min with IL-17C (100 ng/ml; a) or IL-17A (50 ng/ml; b), probed with antibody to phosphorylated (p-) p65, IkBa, Erk, p38 and Jnk (a) or to phosphorylated IkBa (b). GAPDH, glyceraldehyde phosphate dehydrogenase (loading control). (c) Real-time PCR analysis of IL-17RE and IL-17RA mRNA in HT-29 cells treated with control siRNA with scrambled sequence (NC) or siRNA designed to knock down IL-17RE (Si-IL-17RE) or IL-17RA (Si-IL-17RA). *P < 0.001 (Student's t-test). (d,e) Immunoblot analysis of HT-29 cells treated with control siRNA or siRNA specific for IL-17RE (d) or IL-17RA (e) as in c and stimulated for 0–30 min with IL-17C (100 ng/ml), probed with antibodies as in a. Data are representative of three independent experiments (error bars (c), s.e.m.).

Full size image

Specific induction of IL-17C in CECs after bacterial infection

We next sought to determine how IL-17C expression is regulated in the colon. We found that IL-17C mRNA was induced in the colon but not in the spleen or liver during infection with C. rodentium (Fig. 3a and Supplementary Fig. 9a,b). We examined IL-17C mRNA expression in T lymphocytes and B lymphocytes, because IL-17A is produced mainly by T cells. We found that IL-17C induction was not altered, whereas IL-17A induction was lower, after C. rodentium infection of mice deficient in recombination-activating gene 1, which lack lymphocytes (Supplementary Fig. 9c); this indicated that T cells and B cells were not the main producers of IL-17C. We assessed induction of IL-17C mRNA in various cell populations in the colon after infection with C. rodentium. IL-17C mRNA was considerably induced in CECs, as was mRNA encoding the proinflammatory cytokines TNF and IL-1β (Fig. 7a). However, IL-17C was not induced in IELs or LPMCs after infection, although TNF was induced in these cell populations (Supplementary Fig. 9d,e). The expression of IL-17C mRNA was also induced in colon tissue cultured ex vivo after infection with C. rodentium (Fig. 7b). To determine if CECs were directly responsible for the IL-17C induction, we isolated and cultured CECs and CSCs and analyzed IL-17C expression after infection with C. rodentium. The expression of IL-17C mRNA was induced specifically in CECs but not in CSCs, although the expression of TNF mRNA and IL-1β mRNA was induced in both cell populations (Fig. 7c,d). We similarly observed infection-induced IL-17C mRNA in the human gut epithelial cell lines HT-29 and SW480 (Fig. 7e,f). Together our data suggested that CECs are the main IL-17C-producing cells in the gut after infection with C. rodentium.

Figure 7: IL-17C is upregulated in CECs after they encounter bacteria.
figure 7

(a) Real-time quantitative PCR analysis of the expression of IL-17C mRNA, TNF mRNA and IL-1b mRNA in CECs isolated from C57BL/6 wild-type mice infected orally for 0, 4 or 8 d (horizontal axes) with 2 × 109 CFU of C. rodentium. *P < 0.05 and **P < 0.001 (Student's t-test). (b) Real-time quantitative PCR analysis of the expression of mouse IL-17C mRNA in colons cultured ex vivo for 0 or 4 h with C. rodentium. *P < 0.05 (Student's t-test). (c,d) Real-time quantitative PCR analysis of the expression of IL-17C mRNA, TNF mRNA and IL-1b mRNA in mouse primary CECs (c) or CSCs (d) incubated for 0 or 4 h with C. rodentium (multiplicity of infection, ∼100). *P < 0.05, **P < 0.01 and ***P < 0.001 (Student's t-test). (e,f) Real-time quantitative PCR analysis of IL-17C mRNA expression in the human epithelial cell lines HT-29 (e) and SW480 (f) incubated as in c,d. *P < 0.05 and **P < 0.001 (Student's t-test). (g) Real-time quantitative PCR analysis of IL-17C mRNA expression in mouse primary CECs treated for 4 h with LPS (1 mg/ml), flagellin (200 ng/ml), TNF (10 ng/ml), IL-1b (10 ng/ml), IL-17A (50 ng/ml) or IL-17F (50 ng/ml). *P < 0.05, **P < 0.01 and ***P < 0.001 (Student's t-test). Data are representative of three independent experiments (mean and s.e.m.).

Full size image

Pathogen-associated molecular patterns (PAMPs) are recognized by the gut immune system (including CECs) for the activation of host defense41. Both lipopolysaccharide (LPS) and flagellin, which are PAMPs associated with infection with C. rodentium, induced IL-17C production in CECs (Fig. 7g). Similarly, IL-17C was upregulated by LPS or flagellin in human SW480 cells (Supplementary Fig. 10). TNF is reported to induce IL-17C expression in human keratinocytes42. We tested the effect of TNF on gut epithelial cells and found that it induced the expression of IL-17C mRNA in both mouse primary CECs and human SW480 gut epithelial cells (Fig. 7g and Supplementary Fig. 10). In addition to TNF mRNA, mRNA encoding other antibacterial inflammatory cytokines, such as IL-1β, IL-17A and IL-17F, was also upregulated by infection with C. rodentium (Figs. 3a and 7a), and these cytokines all induced the expression of IL-17C mRNA in CECs and SW480 cells, although with different induction ability (Fig. 7g and Supplementary Fig. 10). Together these data indicated that both bacterial PAMPs and inflammatory cytokines induced during infection contributed to the potent induction of IL-17C in gut epithelial cells for host defense.

IL-17C acts in synergy with IL-22

Cytokines secreted by TH17 cells, such as IL-17A, IL-17F and IL-22, are important for host defense against infection7,14,15,16. IL-17A and IL-17F have also been shown to act together with IL-22 to synergistically or additively induce antibacterial peptides in keratinocytes17. The phenotype of early mortality of Il17re−/− mice after infection with C. rodentium resembled that of IL-22-deficient mice. We assessed whether deficiency in IL-17RE affected IL-22 induction during infection. As reported before15, IL-22 mRNA was upregulated in the colon and cecum of mice early after infection with C. rodentium (Fig. 8a,b). The induction of IL-22 mRNA was normal in Il17re−/− mice after infection (Fig. 8c). The kinetics of IL-22 upregulation were similar to those of IL-17C (Figs. 3a and 8a,b and Supplementary Fig. 3). Because both IL-22 and IL-17C were induced in the colon at the early stage of infection, we sought to determine whether they worked together to induce antibacterial peptides in the colon. To address this, we cultured colon tissues ex vivo and then treated them with IL-22 or IL-17C or both. IL-17C and IL-22 synergistically induced the expression of mRNA encoding antibacterial peptides, including S100A8, S100A9, RegIIIβ and RegIIIγ, in the cultured colons (Fig. 8d). We also observed the synergistic effect of IL-17C and IL-22 in cultured CECs (Fig. 8e), which suggested that CECs are the in vivo targets of IL-17C and IL-22. To prove that the IL-17C-mediated synergy with IL-22 was mediated through IL-17RE, we assessed the synergistic effect in cultured colon tissues of Il17re−/− mice. Indeed, whereas IL-17C acted in synergy with IL-22 to induce antibacterial peptides in wild-type colons, such synergy was absent from Il17re−/− colon cultures (Fig. 8f), which further confirmed that IL-17RE is the functional receptor for IL-17C.

Figure 8: IL-17C and IL-22 synergistically induce genes encoding antibacterial peptides.
figure 8

(a,b) Real-time quantitative PCR analysis of IL-22 mRNA in the colon (a) or cecum (b) of C57BL/6 wild-type mice at day 0, 4, 8 or 12 after infection with C. rodentium. *P < 0.05 (Student's t-test). (c) Real-time quantitative PCR analysis of IL-22 mRNA in the colons of Il17re−/− or wild-type mice at day 0, 4 or 10 after infection with C. rodentium. (d,e) Real-time quantitative PCR analysis of the expression of mRNA encoding S100A8, S100A9, RegIIIb and RegIIIg in wild-type mouse colon tissues (d) or mouse primary CECs (e) cultured ex vivo and treated for 24 h with IL-17C (100 ng/ml) or IL-22 (20 ng/ml) alone or in combination. *P < 0.05, **P < 0.01 and ***P < 0.001 (Student's t-test). (f) Real-time quantitative PCR analysis of the expression of mRNA (as in d,e) in Il17re−/− or wild-type mouse colon cultures treated for 24 h as in d,e. NS, not significant. *P < 0.05, **P < 0.01 and ***P < 0.001 (Student's t-test). Data are representative of three independent experiments (mean and s.e.m.).

Full size image

Discussion

Here we have identified the orphan receptor IL-17RE as a functional receptor for IL-17C. We found that IL-17C activated downstream signaling through an IL-17RE–IL-17RA heterodimeric complex. IL-17RE was required for IL-17C-induced production of antibacterial peptides, proinflammatory cytokines and chemokines both ex vivo and in vivo. IL-17C was induced specifically in CECs during infection with C. rodentium and acted in synergy with IL-22 in the induction of antibacterial peptides. Because of the loss of IL-17C-mediated functions, IL-17RE-deficient mice had much lower expression of genes encoding antibacterial molecules, greater bacterial burdens, enhanced colon pathology and eventually mortality after C. rodentium infection.

Although one study has reported that IL-17C is able to stimulate the release of TNF and IL-1β from the THP-1 monocytic cell line25, it is still unclear how IL-17C functions. Our data suggest that IL-17RE is the specific receptor for IL-17C and that IL-17C activates downstream signaling and gene induction through the IL-17RE–IL-17RA heterodimeric complex. As IL-17RA is probably a shared receptor in the IL-17R family, the phenotype of IL-17RA-deficient mice, which have been used extensively to elucidate the roles of IL-17A and IL-17F in various inflammatory conditions, could be attributable in part to the combined effects of knocking out signaling in response to IL-17 cytokines, including IL-17A, IL-17F, IL-17E and IL-17C, which require this receptor for signaling18,19,20,21. Therefore, knocking out individual cytokines or their corresponding 'affinity-conversion' receptors, such as IL-17RC, IL-17RE and IL-17RB, would be more informative about the roles of the individual cytokines in host defense and the pathogenesis of inflammatory diseases.

Here we have demonstrated that IL-17C induced genes encoding antibacterial peptides and inflammatory molecules, which indicates that the IL-17C–IL-17RE axis may be important for host defense against bacterial infection. Because colon tissues had high expression of IL-17RE, we chose infection with the intestinal pathogen C. rodentium as a model with which to investigate the physiological function of IL-17RE. So far, IL-17RC is the only receptor of IL-17R family that has been evaluated in this model, and it is reported to be dispensable for early host defense against infection with C. rodentium15. Thus, IL-17RE represents the first member, to our knowledge, of the IL-17R family shown to be critical for host defense against C. rodentium.

IL-17C, the ligand for IL-17RE, was substantially upregulated in the colon early after infection, consistent with the critical role of IL-17RE in mucosal immunity to C. rodentium. IL-17C was specifically induced in mouse CECs but not in mouse IELs, LPMCs or CSCs after challenge with C. rodentium, which suggested that CECs were the main producers of IL-17C. IL-17C was also induced in human gut epithelial cell lines after infection. The induction of IL-17C in the gut by C. rodentium probably involves both direct and indirect effects. The proposal of a direct effect was supported by the evidence that IL-17C was induced substantially in mouse CECs and human gut epithelial cell lines by direct infection with C. rodentium or stimulation with PAMPs (LPS and flagellin) in vitro. The proposal of an indirect effect was supported by the following evidence. First, in addition to IL-17C, proinflammatory cytokines such as TNF were induced in CECs by direct infection with C. rodentium in vitro. Those proinflammatory cytokines induced IL-17C expression in CECs. Second, the proinflammatory cytokines induced in other gut cell populations (IELs, LPMCs and CSCs) after infection stimulated CECs to induce IL-17C production. Thus, it is likely that bacterial PAMPs directly, together with proinflammatory cytokines produced indirectly during infection, contributed to the potent induction of IL-17C in CECs in vivo. However, it remains to be determined why IL-17C is induced in a cell type–specific manner in the epithelial cells studied here or in keratinocytes, as reported before42.

The early mortality of Il17re−/− mice after infection with C. rodentium resembled that of IL-22-deficient mice15. IL-22 induces the expression of antibacterial peptides, including S100A8, S100A9, RegIIIβ and RegIIIγ15. IL-17C also induced production of those antibacterial peptides in colon cultures ex vivo, although IL-17C-mediated induction of RegIIIβ and RegIIIγ was much weaker than that of S100A8 and S100A9. IL-22 acts together with IL-17A or IL-17F for synergistic induction of S100A9 and for additive induction of S100A7 and S100A8 in keratinocytes17. Here we have shown that IL-17C also acted in synergy with IL-22 to induce S100A8, S100A9, RegIIIβ and RegIIIγ in ex vivo colon cultures as well as in CECs. Therefore, it is likely that both IL-17C and IL-22 contribute to the observed phenotype of Il17re−/− mice after infection. It still remains to be determined how IL-17C acts with IL-22 in this synergy.

In conclusion, we have identified IL-17RE as a functional receptor specifically for IL-17C. IL-17C activated downstream signaling through the IL-17RE–IL-17RA complex for the induction of genes encoding antibacterial molecules. IL-17C, induced in CECs by infection, acted together with IL-22 to synergistically induce antibacterial peptides in CECs. Loss of IL-17C-mediated function in Il17re−/− mice resulted in early mortality after infection. Our results have demonstrated an essential role for the IL-17C–IL-17RE pathway in host mucosal immunity to intestinal pathogens.

Methods

Mice.

Il17re−/− mice on the C57BL/6 background were from ZymoGenetics. Unconditional knockout of Il17re was created by deletion of the coding sequence from exon 1 to exon 6. A β-galactosidase-encoding sequence was inserted into exon 1, in-frame with the original ATG start codon. A phosphoglycerate kinase–neomycin-resistance selection cassette was inserted downstream of the lacZ reporter and the construct was transfected by electroporation into C57BL/6 Bruce-4 embryonic stem cells. Il17re−/− mice and littermate controls 6–8 weeks of age were used for experiments. All mice were maintained in specific pathogen–free conditions. All animal studies were approved by the Institutional Animal Care and Use Committee of the Shanghai Institutes for Biological Sciences (Chinese Academy of Sciences).

Reagents and constructs.

Recombinant IL-17C was from eBioscience; recombinant IL-17A, IL-17F, IL-17E, IL-1β and TNF were from R&D Systems; LPS was from Sigma; and flagellin was from Invivogen. Anti-Flag (M2; F7425) and anti-GAPDH (G9545) were from Sigma. Anti-hemagglutinin (MMS-101R) was from Covance. Antibodies to phosphorylated IkBα (2859L), p65 (3033S), p38 (9211S) and Jnk (92516) were from Cell Signaling. Antibody to phosphorylated Erk (sc-7383) was from Santa Cruz Biotechnology.

Hemagglutinin- or Flag-tagged IL-17RE or IL-17RA and their deletion mutants were generated by PCR and then cloned into the pcDNA3.1 plasmid. Hemagglutinin-tagged IL-17RB, IL-17RC and IL-17RD were also generated by PCR and cloned into the pcDNA3.1 plasmid.

Bacterial infection of mice.

C. rodentium strain DBS100 (ATCC51459; American Type Culture Collection) were prepared by shaking of bacteria overnight at 37 °C in Luria-Bertani broth. Bacterial cultures were serially diluted and plated on MacConkey agar plates so the CFU dose administered could be confirmed. For infection, mice were made to fast for 8 h before oral inoculation with 2 × 109 CFU C. rodentium in a total volume of 100 ml per mouse. Mortality was monitored daily throughout the infection. Body weights were assessed at the beginning of infection and every 2 d after infection.

Tissue collection, histology and CFU counts.

Colons were dissected from mice and fixed with 4% (vol/vol) paraformaldehyde. Paraffin-embedded tissue sections were stained with hematoxylin and eosin for evaluation of tissue pathology. Colons and spleens were removed aseptically, weighed and homogenized in PBS. Homogenates were serially diluted and plated on MacConkey agar plates for determination of CFU counts. C. rodentium colonies (pink with white rings) were counted after overnight incubation at 37 °C. Fecal specimens were collected, weighed and homogenized in PBS and then the CFU counts were determined.

RNA isolation and real-time quantitative PCR.

Total RNA was extracted from cells or mouse tissues with TRIzol reagent according to the manufacturer's instructions (Invitrogen). For cDNA synthesis, RNA was reverse-transcribed with a PrimeScript RT Reagent kit (TaKaRa), then cDNA was amplified by real-time PCR (primers, Supplementary Table 1) with a SYBR Premix ExTaq kit (TaKaRa) on an AbiPrism 7900 HT cycler (Applied Biosystems). The expression of target genes was normalized to expression of housekeeping gene Rpl13a.

Cell culture.

Human HT-29 and SW480 cells were grown in RPIM-1640 medium supplemented with 10% (vol/vol) FBS, penicillin (100 mg/ml) and streptomycin (100 mg/ml). HEK293 cells and wild-type and Il17re−/− mouse embryonic fibroblasts were maintained in DMEM supplemented with supplemented with 10% (vol/vol) FBS, penicillin (100 mg/ml) and streptomycin (100 mg/ml). CSCs were purified as described43. IELs and LPMCs were isolated as described15.

Isolation of mouse primary CECs.

Cut colon pieces were incubated for 3 h in 37 °C in DMEM digestion buffer containing 1% (vol/vol) FBS, collagenase type XI (75 U/ml; Sigma) and dispase (20 mg/ml; Sigma). Digestion mixtures was allowed to settle for 1 min. Suspensions containing epithelial crypts were collected for future sedimentation with S-DMEM (2% (vol/vol) D-sorbitol in DMEM containing 5% (vol/vol) FBS). After centrifugation 200g for 4 min, suspensions were discarded and pellets were resuspended in S-DMEM, then the centrifugation step was repeated five times until the supernatants were clear. Isolated crypts were collected for future experiments. For culture of the isolated cells, crypts were seeded into plates coated with Matrigel (Invitrogen) with DMEM containing 5% (vol/vol) FBS, penicillin (100 mg/ml) and streptomycin (100 mg/ml).

Ex vivo colon culture.

Longitudinally cut colons were placed in PBS containing 1% (vol/vol) FBS, penicillin (100 mg/ml) and streptomycin (100 mg/ml). After being cut into pieces 1–2 mm in length, colons were cultured in 24-well plates with RPMI-1640 medium containing 10% (vol/vol) FBS, penicillin (100 mg/ml) and streptomycin (100 mg/ml).

Knockdown with lentivirus-delivered siRNA.

The siRNA sequence for knockdown of the gene encoding human IL-17RE was 5′-GGCATACCCTTTGCAAAGA-3′; the siRNA sequence for knockdown of the gene encoding human IL-17RA was 5′-CAGGAAGGTCTGGATCATCTA-3′ (ref. 44); the scrambled sequence of control siRNA was 5′-GGATCCTTGACAATACCAA-3′. The siRNA sequences were cloned into the pLSLG lentivirus vector. Lentivirus vectors and helper vectors (delta 8.9 and VSVG) were transfected into 293FT cells (a clonal isolate of HEK293 cells transformed with the SV40 large T antigen; Invitrogen) for viral packaging. At 60 h after transfection, virus was collected for infection of target cells in the presence of polybrene (10 mg/ml; Sigma). At 4 d after infection, cells were used for experiments.

Adenovirus-mediated IL-17C expression in mice.

Sequence encoding mouse IL-17C was cloned into the plasmid pAdTrack-CMV, then recombined with the plasmid pAdEasy-1. Recombinant plasmid Ad-mIL-17C or empty vector was transfected into QBI-HEK 293A cells (an immortalized line of HEK293 cells transformed by human adenovirus DNA; Qbiogene). Virus was packaged and amplified as described36. After titration, 2 × 109 adenovirus particles were injected intravenously into wild-type or Il17re−/− mice. Mouse IL-17C expression was evaluated 4 d after injection.

GST-precipitation assay.

Mouse IL-17A, IL-17C, IL-17E and IL-17F were cloned into the pGEX-4T-1 plasmid. Recombinant GST–IL-17A, GST–IL-17C, GST–IL-17E and GST–IL-17F were expressed in E shcerichia coli BL21 cells and purified with Glutathione Sepharose (GE Healthcare). IL-17RA, IL-17RE and their extracellular domains were ectopically expressed in HEK293 cells and incubated overnight at 4 °C with purified recombinant GST proteins. The Glutathione Sepharose beads were then washed, followed by immunoblot analysis.

Statistics.

A two-tailed Student's t-test was used for analysis of differences between the groups. A one-way analysis of variance was initially done to determine whether an overall statistically significant change existed before analysis with a two-tailed paired or unpaired Student's t-test. P values of less than 0.05 were considered statistically significant.