Abstract
IL-27 exerts pleiotropic suppressive effects on naive and effector T cell populations during infection and inflammation. Surprisingly, however, the role of IL-27 in restricting or shaping effector CD4+ T cell chemotactic responses, as a mechanism to reduce T cell–dependent tissue inflammation, is unknown. In this study, using Plasmodium berghei NK65 as a model of a systemic, proinflammatory infection, we demonstrate that IL-27R signaling represses chemotaxis of infection-derived splenic CD4+ T cells in response to the CCR5 ligands, CCL4 and CCL5. Consistent with these observations, CCR5 was expressed on significantly higher frequencies of splenic CD4+ T cells from malaria-infected, IL-27R–deficient (WSX-1−/−) mice than from infected wild-type mice. We find that IL-27 signaling suppresses splenic CD4+ T cell CCR5-dependent chemotactic responses during infection by restricting CCR5 expression on CD4+ T cell subtypes, including Th1 cells, and also by controlling the overall composition of the CD4+ T cell compartment. Diminution of the Th1 response in infected WSX-1−/− mice in vivo by neutralization of IL-12p40 attenuated CCR5 expression by infection-derived CD4+ T cells and also reduced splenic CD4+ T cell chemotaxis toward CCL4 and CCL5. These data reveal a previously unappreciated role for IL-27 in modulating CD4+ T cell chemotactic pathways during infection, which is related to its capacity to repress Th1 effector cell development. Thus, IL-27 appears to be a key cytokine that limits the CCR5-CCL4/CCL5 axis during inflammatory settings.
Introduction
Interleukin-27 is a critically important and nonredundant regulator of pathogenic T cell responses during a variety of inflammatory conditions (1, 2). IL-27R (TCCR/WSX-1)–deficient mice develop excessive proinflammatory T cell responses and resultant T cell–dependent immunopathology during a number of infections, including malaria, Toxoplasma gondii, Leishmania major, Mycobacterium tuberculosis, and Trypanasoma cruzi infection (3–7). Although the molecular basis of IL-27–mediated suppression in vivo is still incompletely understood, IL-27 has been shown to attenuate Rorc expression, inhibiting Th17 cell responses, and to limit Th1 and Th2 responses (3, 4, 6–9). Moreover, IL-27 inhibits IL-2 production by effector CD4+ T cells and induces IL-10 production by naive, Tr1, Th1, Th2, and Th17-like cells (10–14).
Despite the number of studies examining the immunoregulatory effects of IL-27 on CD4+ T cells during infection, to date there has been no detailed investigation of whether IL-27 regulates CD4+ T cell trafficking and migration. This is surprising as excessive accumulation of CD4+ T cell populations in peripheral tissues, such as the liver, lung, and brain, is a common pathological feature in infected IL-27R–deficient mice (3, 6, 15, 16), indicating that CD4+ T cell migratory pathways may be dysregulated.
CCR-dependent pathways determine the migration patterns of effector T cells within tissues under both homeostatic and inflammatory conditions (17, 18). Chemokine receptors are heterogeneously displayed by naive and effector/memory T cell populations (17–21). For example, CCR7 is expressed on naive and memory T cell populations, but is downregulated on highly differentiated and migratory effector T cells (20). In contrast, many chemokine receptors, including CXCR3, CCR5, CCR6, and CXCR6, are predominantly expressed by effector T cells (19, 21). Although it has been reported that different CD4+ T cell subsets (i.e., Th1, Th2, Th17, TFH, and T regulatory [Treg]) may express unique repertoires of chemokine receptors (22), it is becoming clear that, in vivo, T cell subset-restricted expression of individual chemokine receptors does not occur. Thus, CCR5, which has been described as a Th1 marker (23), has been shown to be expressed by IL-13–producing cells in the OVA-sensitized lung (24) and is now thought to be expressed on recently activated cells of any subset (22).
A number of studies have examined the pathways that regulate chemokine receptor expression in CD4+ T cell populations. CD3 stimulation of T cell clones in vitro leads to upregulation of CCR7, CCR8, and CXCR5 and downregulation of CCR1, CCR2, CCR3, and CCR5 (21). IFN-γ and TNF upregulate CCR5 and CXCR3 on PBMCs (25, 26). In contrast, there is evidence that IL-10 downregulates CCR5 expression (25) and IL-12 promotes or inhibits CCR5 expression depending on the experimental systems (27–29).
As IL-27 has a profound effect on T cell activation and on their production of IL-2, IFN-γ, IL-17, and IL-10 (3–16), we hypothesized that IL-27R signaling may also modulate the repertoire of chemokine receptor expression on effector CD4+ T cells during infection, and consequently regulate T cell chemotactic behavior. Using Plasmodium berghei NK65 infection as a model systemic inflammatory condition, we show that abrogation of WSX-1 signaling elevates surface expression of CCR5 on CD4+ T cells during infection. Correspondingly, infection-derived WSX-1−/− effector CD4+ T cells displayed significantly enhanced migration to CCL4 and CCL5. Importantly, we show that upregulated expression of CCR5 on CD4+ T cells in WSX-1−/− mice during infection is not simply due to differences in the composition of the effector T cell pool in WSX-1−/− mice compared with wild-type (WT) mice, but is also due to specific alterations in CCR5 expression by individual T cell subsets. These data reveal an important role for IL-27R/WSX-1 in regulating CD4+ T cell chemotactic responses during inflammation.
Materials and Methods
Mice and parasites
C57BL/6N mice were purchased from Harlan. IL-27R–deficient (WSX-1−/−) mice (30) were originally provided by Amgen and were bred at the London School of Hygiene and Tropical Medicine and the University of Manchester. IL-10Rα–deficient mice were provided by W. Muller (University of Manchester). All transgenic strains were fully backcrossed (n > 10) to C57BL/6 mice. Animals were maintained under barrier conditions in individually ventilated cages. Cryopreserved P. berghei NK65 parasites were passaged once through C57BL/6 mice before being used to infect experimental animals. Six- to 10-wk-old mice were infected by i.v. injection of 104 parasitized RBCs (pRBC). In some experiments, 250 μg anti–IL-12 (C17.8) or anti–IL-2 (JES6-5H4), both from BioXCell, was injected i.p. every other day starting on day 7 postinfection (p.i.).
Flow cytometry
Spleens were removed from naive and malaria-infected (day 7 or 14 p.i.) WT and WSX-1−/− mice. Single-cell suspensions were prepared by homogenizing through a 70-μm cell strainer (BD Biosciences). RBCs were lysed using RBC lysis buffer (BD Biosciences). Absolute cell numbers were calculated by microscopy using a hemocytometer, and live/dead cell identification was performed by trypan blue exclusion.
Phenotypic characterization of CD4+ T cell populations was performed by surface staining with the following: anti-mouse CD4 (clone GK1.5), anti-mouse CD44 (IM7), anti-mouse CD62L (MEL-14), anti-mouse KLRG1 (2F1), anti-mouse CCR1 (PA1-41062), anti-mouse CCR5 (7A4), anti-mouse CXCR3 (CXCR3-173), anti-mouse CXCR4 (2B11), and anti-mouse CXCR6 (221002), according to previously published protocols (3, 31
All flow cytometry acquisition was performed using an LSR II (BD Systems). All analysis was performed using Flowjo software (Tree Star). Fluorescence minus one controls were used to validate flow cytometric data.
Transwell migration assays
Migration of purified CD4+
2 for 1 h. CD4+ T cells were purified by magnetic bead separation (Miltenyi Biotec), and 1 × 106 CD4+ T cells were then added to the inner well in a final volume of 100 μl (RPMI 1640 + 5% FCS). Purity of sorted cells was routinely >95%. The plate was incubated at 37°C and 5% CO2 for 3 h, and CD4+ T cell movement toward the recombinant chemokines was quantified by counting, by light microscopy, the number of cells that had migrated into the outer well. Directed movement toward chemokines was calculated by subtracting the number of cells that migrated in response to medium alone from the number that migrated in response to the individual chemokines.Statistics
All data were tested for normal distributions using the D’Agostino and Pearson omnibus normality test. In two group comparisons, statistical significance was determined using the t test or the Mann–Whitney U test, depending on the distribution of the data. For three or more group comparisons, statistical significance was determined using a one-way ANOVA, with the Tukey post hoc analysis for normally distributed data, or a Kruskal–Wallis test, with Dunn post hoc analysis for nonparametric data. All statistical analyses were performed using GraphPad Prism. Differences were considered to be significantly different when p < 0.05.
Results
CD4+ T cells from WSX-1−/− malaria-infected mice demonstrate significantly enhanced chemotaxis toward CCL4 and CCL5
We and others have demonstrated an important role for IL-27R in limiting effector CD4+ T cell accumulation in nonlymphoid tissues during infection (3, 6, 15, 16). As the spleen is an important site for T cell priming, preceding cell migration during a variety of infections, including malaria (32), we hypothesized that WSX-1 signaling may directly suppress splenic CD4+ T cell chemotactic responses during inflammation. To test this, we isolated splenic CD4+ T cells from naive and malaria-infected (day 14 p.i.) WT and WSX-1−/− mice and contrasted their chemotactic responses to a panel of chemokines in an in vitro Transwell migration assay.
CD4+ T cells from naive WT and WSX-1−/− mice failed to migrate toward the majority of the tested chemokines, with the notable exceptions of CCL21 and CXCL12, toward which CD4+ T cells from naive WT and WSX-1−/− mice migrated with equal efficiency (Fig. 1A). These data demonstrate that there are no intrinsic differences in the chemotactic capacities of splenic CD4+ T cells in WT and WSX-1−/− mice under homeostatic conditions (Fig. 1A). In contrast, splenic CD4+ T cells from infected (day 14 p.i.) WSX-1−/− mice displayed significantly increased chemotaxis to CCL4, CCL5, and, to a lesser extent, CXCL16 compared with CD4+ T cells from infected WT mice (Fig. 1B). Interestingly, splenic CD4+ T cells from infected WT and WSX-1−/− mice displayed equivalent chemotactic responses to CCL20, CXCL9, CXCL10, and CXCL12 (Fig. 1B). These data show that lack of IL-27 signaling significantly affects chemotaxis of splenic CD4+ T cells toward specific chemokines during infection.
WSX-1 signaling restricts CD4+ T cell chemotactic responses during malaria infection. WT and WSX-1−/− mice were infected with 104 P. berghei NK65 pRBCs. Splenic CD4+ T cells were purified from naive and malaria-infected mice (D14 p.i.). (A and B) A total of 1 × 106 purified (A) naive, or (B) infection-derived, CD4+ T cells was placed in duplicate in the top chamber of a 5-μm Transwell, and migration of cells toward recombinant chemokines (all at 100 ng/ml) in the bottom chamber was assessed by microscopy. Data are the mean ± SEM of the group and are representative of five independent experiments. *p < 0.05, WT versus WSX-1−/− mice.
WSX-1 restricts CCR5 expression by CD4+ T cells during malaria infection
We hypothesized that the enhanced chemotaxis of splenic CD4+ T cells from infected WSX-1−/− mice toward CCL4, CCL5, and CXCL16 was due to altered chemokine receptor expression by cells from infected WSX-1−/− mice. Consistent with our data showing that CD4+ T cells from naive WT and WSX-1−/− display equivalent in vitro chemotactic responses, and also our previous observations that differences in CD4+ T cell function in WSX-1−/− mice manifest only after day 7 of malaria infection (3), there were no significant differences in chemokine receptor expression by splenic CD4+ T cells from naive WT and WSX-1−/− mice (Fig. 2A, 2B) or by cells obtained from WT and WSX-1−/− mice on day 7 p.i. (data not shown). However, CD4+ T cells obtained from WSX-1−/− mice on day 14 p.i. expressed significantly higher surface levels of CCR5 (the receptor for CCL3, CCL4, and CCL5) than did splenic CD4+ T cells from equivalent WT mice (Fig. 2A, 2C). Splenic CD4+ T cells from infected WSX-1−/− mice also displayed higher surface expression of CCR1 and CXCR6 compared with cells from infected WT mice, but the fold upregulation in expression was significantly less than that observed for CCR5. In contrast, there were no significant differences in the surface expression of CXCR3 or CXCR4 by CD4+ T cells derived from infected WT and WSX-1−/− mice (Fig. 2C). Importantly, similar patterns of dysregulated chemokine receptor expression were observed when intracellular detection was performed (higher expression of CCR5 by CD4+ T cells derived from infected WSX-1−/− mice), showing that the differences in surface CCR5 expression were not simply the result of altered chemokine receptor recycling or internalization (Fig. 2D). Furthermore, CCR5 mRNA levels were higher in splenic CD4+ T cells from infected WSX-1−/− mice than in cells from infected WT mice (results not shown).
WSX-1 signaling restricts CCR5 expression on CD4+ T cells during malaria infection. WT and WSX-1−/− mice were infected with 104 P. berghei NK65 pRBCs. (A) Representative plots (gated on CD4+ T cells) showing the surface expression of chemokine receptors on splenic CD4+ T cells from naive and infected (day 14 p.i.) WT and WSX-1−/− mice. (B and C) Frequencies of splenic CD4+ T cells from (B) naive and (C) malaria-infected (day 14 p.i.) WT and WSX-1−/− mice expressing chemokine receptors on the cell surface. (D) The frequencies of splenic CD4+ T cells from malaria-infected (day 14 p.i.) WT and WSX-1−/− mice expressing intracellular chemokine receptors. (E) The frequencies of splenic CD4+ T cells from malaria-infected (day 14 p.i.) WT and IL-10R−/− mice expressing chemokine receptors on the cell surface. Data are the mean ± SEM of the group with three to five mice per group. Data are representative of three independent experiments. (B–D) *p < 0.05, WT versus WSX-1−/− mice. (E) *p < 0.05, WT versus IL-10R−/− mice.
WSX-1 and IL-10R signaling differentially control chemokine receptor expression by CD4+ T cells during malaria infection
It has been proposed that many of the immunoregulatory properties of IL-27 are directly related to its ability to promote IL-10 production by T cell subsets (11–14). Thus, we tested whether the altered repertoire of chemokine receptor expression by splenic CD4+ T cells from malaria-infected WSX-1−/− mice was phenocopied by CD4+ T cells from malaria-infected IL-10R−/− mice. Significantly higher frequencies of CD4+ T cells from malaria-infected IL-10R−/− mice expressed CCR5 compared with cells from infected WT mice (Fig. 2E), but the fold increase in CCR5 expression was significantly lower than observed in WSX-1−/− mice (2.4 versus 4.3, respectively). Moreover, CCR1, CXCR3, CXCR4, and CXCR6 were expressed at equivalent levels by CD4+ T cells derived from infected WT and IL-10R−/− mice. These data demonstrate that IL-27R exerts stronger suppression of CCR5 expression than IL-10R in vivo during malaria infection. IL-27 may therefore synergize with/amplify the IL-10 signal to limit CCR5 expression.
WSX-1 signaling modulates both the magnitude and the cellular composition of the CD4+CCR5+ T cell population during malaria infection
Our results demonstrate that abrogation of IL-27R signaling primarily alters CCR5 expression by splenic CD4+ T cells during infection. To determine the mechanism by which WSX-1 signaling restricts CCR5 expression by CD4+ T cells during infection, we first characterized the splenic CD4+CCR5+ T cell populations in WT and WSX-1−/−-infected mice, to identify whether the populations are comprised predominantly of a specific cellular subset in each strain of mice and/or whether the absence of WSX-1 also led to a consequential remodeling of the CD4+CCR5+ T cell compartment. Similar frequencies of CD4+CCR5+ T cells from naive WT and WSX-1−/− mice expressed CD62L (naive/central memory), CD44 (activated), GATA-3 (Th2), Foxp3 (Treg), T-bet (Th1), and KLRG-1 (terminal differentiation marker) (Fig. 3A, 3B). Although similar frequencies of CD4+CCR5+ T cells from malaria-infected (day 14 p.i.) WT and WSX-1−/− mice also expressed CD44 and GATA-3, significantly fewer CD4+CCR5+ T cells from WSX-1−/− mice expressed CD62L and Foxp3 (Fig. 3A, 3C). Moreover, compared with infected WT mice, a significantly higher proportion of CD4+CCR5+ T cells from infected WSX-1−/− mice expressed T-bet and KLRG-1 (Fig. 3A, 3C). Indeed, ∼60% of CD4+CCR5+ T cells from WSX-1−/− mice expressed T-bet (Fig. 3A, 3C). As we have previously shown that the majority of CD4+ KLRG-1+ T cells from infected WSX-1−/− mice coexpress T-bet (33), our results indicate that a large proportion of CD4+CCR5+ T cells from infected WSX-1−/− mice are terminally differentiated Th1 cells.
Loss of WSX-1 signaling leads to remodeling of the CD4+CCR5+ T cell compartment and increased numbers of Th1-CCR5+ T cells during malaria infection. WT and WSX-1−/− mice were infected with 104 P. berghei NK65 pRBCs. (A) Representative plots showing the surface expression of T cell markers on splenic CD4+ CCR5+ T cells from naive and infected (day 14 p.i.) WT and WSX-1−/− mice. (B and C) Frequencies and (D, E) numbers of splenic CD4+CCR5+ T cells from (B, D) naive and (C, E) malaria-infected (day 14 p.i.) WT and WSX-1−/− mice expressing the individual markers. (D, E) Numbers of cells calculated out of the 1 × 106 purified CD4+ T cells used in chemotaxis assays. Data are the mean ± SEM of the group with three to five mice per group and are representative of four independent experiments. *p < 0.05, WT versus WSX-1−/− mice.
Using this phenotypic analysis, we calculated the numbers of CD4+CCR5+ T cells expressing the various T cell subset markers among 1 × 106 splenic CD4+ T cells, the number of CD4+ T cells used in the chemotaxis assays. Very few CD4+CCR5+ T cells were observed within 1 × 106 splenic CD4+ T cells from naive WT or WSX-1−/− mice, and there were no significant differences in the number of CD4+CCR5+ T cells expressing CD62L, CD44, Foxp3, T-bet, GATA-3, or KLRG-1 (Fig. 3D). As expected, there were significantly increased numbers of CD4+CCR5+ T cells within the splenic CD4+ T cell population purified from infected WSX-1−/− mice compared with cells from infected WT mice (Fig. 3E). Although there were no differences in the numbers of CD4+CCR5+ cells expressing CD62L, Gata-3, or Foxp3 in the CD4+ T cell populations from infected WT and WSX-1−/− mice, significantly increased numbers of CD4+CCR5+ cells coexpressing CD44, T-bet, and KLRG-1 were present within the CD4+ T cell population from infected WSX-1−/− mice (Fig. 3E). Similar results were obtained when calculating the total numbers of CD4+CCR5+ T cells expressing the subset markers within the spleens of naive and infected WT and WSX-1−/− mice (results not shown).
Taken together, these data indicate that splenic CD4+CCR5+ T cell populations are broadly similar in naive WT and WSX-1−/− mice, but that the composition of the population changes during the course of malaria infection in WSX-1−/− mice, leading to a significant increase in the frequencies and numbers of CD4+CCR5+ T cells displaying markers of terminally differentiated Th1 cells during malaria infection.
WSX-1 signaling represses CCR5 expression by all CD4+ T cell subpopulations during malaria infection
As the composition of the CD4+CCR5+ T cell population was significantly different in malaria-infected WT and WSX-1−/− mice, we examined whether this simply reflected global changes in the CD4+ T cell population within the spleen of malaria-infected WSX-1−/− mice and/or whether WSX-1 specifically regulated chemokine receptor expression on individual CD4+ T cell subtypes, in particular Th1 cells. When compared with WT mice, significantly higher frequencies of splenic CD4+ T cells from WSX-1−/− mice expressed KLRG-1 and T-bet on day 14 of infection, whereas the frequencies of other CD4+ T cell populations (naive/central memory [CD62L+], Th2 [Gata3+], and Treg [Foxp3+]) did not differ between WT and WSX-1−/− mice (Fig. 4A). Consistent with this, significantly higher numbers of splenic CD4+ T cells derived from WSX-1−/− mice expressed KLRG-1 and T-bet compared with cells from WT mice (when calculated out of 1 × 106 purified CD4+ T cells used in the chemotaxis assays) (Fig. 4B). Moreover, as the total number of splenic CD4+ T cells was comparable in malaria-infected WT and WSX-1−/− mice (results not shown), significantly higher numbers of CD4+T-bet+, CD4+KLRG-1+ T cells were observed in the spleens of malaria-infected WSX-1−/− mice compared with infected WT mice (Fig. 4C). There were no significant differences in the frequencies or numbers of any of the CD4+ T cell subsets within naive WT and WSX-1−/− mice (Fig. 4A–C).
Loss of WSX-1 signaling modifies the structure of the CD4+ T cell population, but also leads to unconstrained CCR5 expression on CD4+ T cell subsets during malaria infection. WT and WSX-1−/− mice were infected with 104 P. berghei NK65 pRBCs. (A) Frequencies of splenic CD4+ T cells from naive and infected (day 14 p.i.) WT and WSX-1−/− mice expressing the various T cell markers. (B and C) The numbers of CD4+ T cells derived from naive and infection-derived (D14 p.i.) WT and WSX-1−/− mice expressing the various T cell subset markers out of (B) 1 × 106 CD4+ T cells used in the chemotaxis assay and (C) total splenocytes. (D) Representative histograms (gated on CD4+ marker+ T cells) showing the surface expression of CCR5 on splenic CD4+ T cells from naive and infected (day 14 p.i.) WT and WSX-1−/− mice. (E and F) Frequencies and (G) mean fluorescence intensity of CCR5 expression on (E) naive and (F, G) malaria infection-derived (day 14 p.i.) splenic CD4+ marker+ T cells. Data are the mean ± SEM of the group with three to five mice per group and are representative of four independent experiments. *p < 0.05, WT versus WSX-1−/− mice.
Equivalent proportions of the CD4+ T cell subsets, as defined above, expressed CCR5 in naive WT and WSX-1−/− mice (Fig. 4D, 4E). In contrast, there was a global increase in CCR5 expression by all examined CD4+ T cell subsets in infected WSX-1−/− mice compared with infected WT mice, as measured by both the frequency of subset+ cells expressing CCR5 and the mean fluorescence intensity of CCR5 expressed on subset+ cells on a cell-per-cell basis (Fig. 4D, 4F, 4G). Thus, the increased expression of CCR5 by CD4+ T cells in malaria-infected WSX-1−/− mice is not simply a consequence of skewing of T cell differentiation toward the Th1 phenotype; rather, WSX-1 signaling seems to repress CCR5 expression in all CD4+ T cell subpopulations.
Blocking IL-12p40, but not IL-2, signaling attenuates Th1 polarization and CCR5 expression in WSX-1−/− mice during malaria infection
Because the CD4+CCR5+ T cells accumulating in spleens of malaria-infected WSX-1−/− mice were mainly Th1 cells (Fig. 3C), we hypothesized that attenuating the differentiation of Th1 cells would limit CCR5 expression, thereby reducing CD4+ T cell chemotaxis to CCR5 ligands. Moreover, we expected that CCR5 expression on CD4+ T cells derived from infected WSX-1−/− mice would correlate with IL-12Rβ1 and IL-2Rα (CD25) expression, as IL-12 and IL-2 signaling are required for the development and maintenance of Th1 cells (34). In partial agreement with this expectation, significantly higher frequencies of CD4+CCR5+ T cells from malaria-infected WSX-1−/− mice coexpressed IL-12Rβ1 compared with CD4+CCR5+ T cells from WT mice (Fig. 5A, 5B), but CD25 expression did not differ on CD4+CCR5+ T cells from infected WT and WSX-1−/− mice (Fig. 5A, 5C). These data suggested that IL-12R, but not IL-2, may promote CCR5 expression on CD4+ T cells (and, specifically, on Th1 cells) in WSX-1−/− mice during malaria infection.
Coordinated expression of CCR5 with IL-12R on CD4+ T cells derived from malaria-infected WSX-1−/− mice. WT and WSX-1−/− mice were infected with 104 P. berghei NK65 pRBCs. (A) Representative plots (gated on CD4+ T cells) showing the surface expression of CCR5 versus IL-12Rβ1 and CD25 on splenic CD4+ T cells from naive and infected (day 14 p.i.) WT and WSX-1−/− mice. (B and C) The frequencies of splenic CD4+CCR5+ T cells from naive and malaria-infected (day 14 p.i.) WT and WSX-1−/− mice expressing (B) IL-12Rβ1 and (C) CD25. Data are the mean ± SEM of the group with three to five mice per group and are representative of two independent experiments. *p < 0.05, WT versus WSX-1−/− mice.
Administration of anti–IL-12p40 mAb to WSX-1−/− mice (from day 7 of infection) significantly attenuated T-bet expression by splenic CD4+ T cells (Fig. 6A, 6C) and concomitantly reduced the frequencies of splenic CD4+ T cells that expressed CCR5 (Fig. 6B, 6D). Anti–IL-12p40 also slightly reduced CCR1 expression by splenic CD4+ T cells in WSX-1−/− mice (results not shown). In contrast, anti–IL-2 mAb treatment had no effect on the frequencies of splenic CD4+T-bet+ or CD4+CCR5+ cells in infected WSX-1−/− mice (Fig. 6A–D). As expected, anti–IL-12p40 mAb administration to WSX-1−/− mice led to a reduction in the numbers of total CD4+CCR5+ T cells and CD4+CCR5+ cells expressing CD44, T-bet, and KLRG-1, as calculated within 1 × 106 purified CD4+ T cells used in chemotaxis assays (Fig. 6E), and total splenocytes (results not shown). Anti–IL-12p40 treatment to WSX-1−/− mice did not, however, alter the numbers of CD4+CCR5+ cells coexpressing CD62L, Foxp3, or GATA-3 (Fig. 6E and results not shown). Together, these data demonstrate that IL-12, but not IL-2, is required for the differentiation of splenic CD4+CCR5+ Th1 cells in malaria-infected WSX-1−/− mice. Consistent with this, CD4+ T cells from anti–IL-12p40–treated infected WSX-1−/− mice displayed significantly reduced chemotaxis to CCL4 and CCL5, whereas chemotaxis of CD4+ T cells from anti–IL-2–treated mice was similar to that of cells from control (untreated)-infected WSX-1−/− mice (Fig. 6F and results not shown).
Anti–IL-12p40 mAb administration to malaria-infected WSX-1−/− mice inhibits T-bet expression, reverses CCR5 expression by CD4+ T cells, and attenuates CD4+ T cell chemotaxis. WT and WSX-1−/− mice were infected with 104 P. berghei NK65 pRBCs. Some WSX-1−/− mice were treated with 250 μg anti–IL-12p40 or anti–IL-2 mAbs every second day starting on day 7 p.i. (A and B) Representative plots (gated on CD4+ T cells) showing the surface expression of (A) T-bet and (B) CCR5 on splenic CD4+ T cells from naive and infected (day 14 p.i.) WT and WSX-1−/− mice and WSX-1−/− mice treated with anti–IL-12p40 or anti–IL-2. (C and D) The frequencies of splenic CD4+ T cells from naive, infected control, and infected Ab-treated WT and WSX-1−/− mice expressing (C) T-bet or (D) CCR5. (E) The numbers of CD4+ T cells from WT, anti–IL-12p40 mAb-treated, and control-infected WSX-1−/− mice expressing the various T cell subset markers out of 1 × 106 CD4+ T cells used in the chemotaxis assay. (F) A total of 1 × 106 purified CD4+ T cells from infected WT, WSX-1−/−, and WSX-1−/− mice treated with anti–IL-12p40 or anti–IL-2 was placed in duplicate in the top chamber of a 5-μm Transwell, and migration toward recombinant CCL4 (100 ng/ml) in the bottom chamber was assessed by microscopy. Data are the mean ± SEM of the group with three to five mice per group and are representative of two independent experiments. *p < 0.05, WT versus WSX-1−/− control mice, ∼p < 0.05, WT versus anti–IL-12p40–treated WSX-1−/− mice, +p < 0.05, WT versus anti–IL-2–treated WSX-1−/− mice, †p < 0.05, WSX-1−/− control mice versus anti–IL-12p40–treated WSX-1−/− mice, ‡p < 0.05, anti–IL-12p40–treated WSX-1−/− mice versus anti–IL-2–treated WSX-1−/− mice.
Discussion
In this study, we have demonstrated that IL-27 signaling via WSX-1 modulates the chemokine receptor repertoire expressed by activated CD4+ T cells during malaria infection. In particular, we have shown that IL-27 is an important regulator of the CCR5-CCL4/CCL5 pathway, revealing an entirely novel pathway by which IL-27 can limit CD4+ T cell accumulation in tissues. This observation is likely to have major relevance for a number of inflammatory conditions in which WSX-1 is known to suppress T cell–mediated inflammation [reviewed in (1, 2)].
Remodeling of the CD4+ T cell compartment, in particular enhanced Th1 polarization, was not the sole reason for increased CCR5 expression by CD4+ T cells in infected WSX-1−/− mice. Indeed, our data show that signaling via WSX-1 also suppresses CCR5 expression on CD62L+, T-bet+, GATA-3+, and Foxp3+ CD4+ T cell populations. As CD62L, GATA-3, and Foxp3 expression broadly segregates from T-bet on day 14 of infection (results not shown), we conclude that, in an inflammatory environment, WSX-1 signaling represses expression of CCR5 on most, if not all, CD4+ T cell populations, independent of their polarization or activation status. Thus, IL-27/WSX-1 signaling may be a critical regulator of T cell CCR5 expression during infection, nonspecifically limiting accumulation of effector cells in affected tissues and dampening inflammation. Interestingly, it has very recently been shown that IL-27 directs the development of Th1-adapted CXCR3+Foxp3+ Treg cells during T. gondii infection (35), further supporting the important role for IL-27 in controlling chemokine receptor expression by T cells during infection.
We found that, in the malaria model, the enhanced chemotactic response of splenic CD4+ T cells is dominantly driven by Th1 cells. Specific inhibition of Th1 CCR5+ cell development, and potentially function, in infected WSX-1−/− mice by treatment with anti–IL-12p40 mAbs led to a major reduction in infection-derived CD4+ T cell chemotaxis toward CCL4 and CCL5 in vitro. Interestingly, anti–IL-12p40 mAb treatment of infected WSX-1−/− mice reduced CD4+ T cell chemotaxis to WT levels, but failed to induce the complete downregulation of CCR5 expression. CCR5-mediated chemotaxis requires integrated signaling through a number of intracellular pathways, including PI3K, JAK/STAT, and MAPK pathways (36). Thus, our results suggest that anti–IL-12p40 mAb treatment may also negatively affect the efficiency of CCR5 signal transduction in CD4+ T cells, attenuating the functionality of the remaining CCR5 molecules expressed on the cell surface. Of relevance, IL-12 has been shown to activate PI3K, thus potentially linking IL-12R and CCR5 signaling events (37). Consequently, our data suggest that IL-27R signaling normally limits splenic CD4+ T cell chemotaxis during malaria infection by coordinately inhibiting Th1 cell differentiation and suppressing CCR5 expression and intracellular signaling pathways. Although we predict that Th1 cells from infected WSX-1−/− mice would show enhanced migratory capacity when compared with Th1 cells from WT mice, due to their higher CCR5 expression, we cannot directly test this prediction due to the lack of reliable surface markers for purification of splenic Th1 cells from infected mice (IL-18R and CD226 are good markers of Th1 cells in WSX-1−/− mice, but not WT mice [A. Villegas-Mendez, unpublished observations]). Of note, enhanced accumulation of CCR5-bearing CD4+ T cells in the spleen of WSX-1−/− mice during malaria infection was not simply due to elevated retention of these cells in the spleen of WSX-1−/− mice versus WT mice, and migration of CD4+CCR5+ T cells into nonlymphoid organs is increased in WSX-1−/− mice during infection (3) (E. Gwyer Findlay, A. Villegas-Mendez, J. B. de Souza, C. J. Saris, T. E. Lane, E. M. Riley, and K. N. Couper, manuscript in preparation).
The mechanism(s) by which IL-27R signaling modulates CCR5 expression on CD4+ T cells during infection is not clear. We do not believe that this simply reflects changes in the extent or rate of CCR5 internalization, or (re)cycling between the cell membrane and the intracellular environment, as intracellular staining for CCR5 gave very similar results to those obtained by surface staining. Moreover, the increased mean fluorescence intensity of CCR5 expression on the analyzed CD4+ T cell subsets shows that more CCR5 is present on CD4+ T cells derived from infected WSX-1−/− mice on a cell-per-cell basis. Although CCR5 mRNA transcript levels are higher in CD4+ T cells from infected WSX-1−/− mice than infected WT mice, indicating that IL-27 may regulate CCR5 through transcriptional repression and/or by reducing mRNA stability, the heterogeneity of the analyzed material, with an increased proportion of Th1 cells in the CD4+ T cell population from WSX-1−/− mice, complicates the interpretation of this experiment. Of interest, however, miRNA mimics can suppress CCR5 expression on leukocytes in vitro (38), indicating that CCR5 expression is closely linked to mRNA transcription and/or translation. Whether IL-27 modulates expression of microRNAs that target the 3′ untranslated region of CCR5 transcripts has yet to be examined.
Our data clearly show that IL-12p40 plays an important role in promoting CCR5 expression by CD4+ T cells in WSX-1−/− mice during malaria infection. IL-12 has previously been shown to promote CCR5 expression on CD4+ T cells (28, 29). IL-12p70 production is significantly elevated in WSX-1−/− mice during malaria infection (3), and treatment with anti–IL-12p40 significantly reduced CCR5 expression, at least in part due to reduction in Th1 cell polarization. Recent work in our laboratory suggests that IL-27 limits the magnitude of the Th1 cell response, and Th1 cell terminal differentiation, during malaria infection by inhibiting Th1 cell responsiveness to IL-12p70, which appears to cause instability in the Th1 molecular program, rather than affecting T cell proliferation or apoptosis (33). Of note, however, whereas anti–IL-12p40 treatment of WSX-1−/− mice reduced T-bet expression to WT levels, CCR5 expression, although significantly diminished, was still significantly higher than in WT mice. Thus, we believe our results indicate that other additional signals may act independently from IL-12p40 to promote CCR5 expression on CD4+ T cell subsets in infected WSX-1−/− mice. Alternatively, IL-27R ligation may directly limit CCR5 expression through STAT signaling pathways (39).
Interestingly, the redox status of leukocytes strongly determines CCR5 stability and expression; reactive oxygen species (ROS) promote CCR5 expression, whereas antioxidants inhibit CCR5 expression (40). WSX-1 is known to suppress ROS production by human neutrophils (41). Thus, it is possible that increased IL-12 and IFN-γ production, coupled with loss of direct WSX-1–mediated inhibition of ROS production, in WSX-1−/− mice during malaria leads to enhanced ROS generation, which then nonspecifically promotes CCR5 expression on various CD4+ T cell populations. Nevertheless, we do not believe that WSX-1 regulates CCR5 expression on CD4+ T cells simply and specifically through IL-10–dependent mechanisms, as CCR5 expression is less affected in IL-10R−/− mice during malaria infection than in WSX-1−/− mice.
In summary, we have identified an important and previously unappreciated role for IL-27R in regulating chemokine receptor pathways that control CD4+ T cell chemotactic responses. In particular, we have defined a critical role for IL-27R in suppressing CCR5 expression by CD4+ T cell populations during infection. Our results may have significant implications for understanding how IL-27/WSX-1 regulates CD4+ T cell migration and tissue accumulation in a variety of inflammatory conditions.
Disclosures
The authors have no financial conflicts of interest.
Acknowledgments
We thank Lisa Grady (University of Manchester) for technical support.
Footnotes
This work was supported by Biotechnology and Biological Sciences Research Council Grants 004161 and 020950 and by Medical Research Council Career Development Award G0900487 (to K.N.C.).
Abbreviations used in this article:
- p.i.
- postinfection
- pRBC
- parasitized RBC
- ROS
- reactive oxygen species
- Treg
- T regulatory
- WT
- wild-type.
- Received October 19, 2012.
- Accepted February 25, 2013.
- Copyright © 2013 by The American Association of Immunologists, Inc.