Cell Host & Microbe Article

A2B Adenosine Receptor Induces Protective Antihelminth Type 2 Immune Responses

Nirav Patel,1,2 Wenhui Wu,1,2,5 Pankaj K. Mishra,1,2,5 Fei Chen,1,2 Ariel Millman,1,2 Bala´ zs Cso´ ka,2,3 Bala´ zs Koscso´ ,2,3 Holger K. Eltzschig,4 Gyo¨ rgy Hasko´ ,2,3 and William C. Gause1,2,* 1Department of Medicine 2Center for Immunity and Inflammation 3Department of Surgery New Jersey Medical School, Rutgers, the State University of New Jersey, Newark, NJ 07101, USA 4Department of Anesthesiology, Mucosal Inflammation Program, University of Colorado School of Medicine, Aurora, CO 80045, USA 5These authors contributed equally to this work *Correspondence: [email protected] http://dx.doi.org/10.1016/j.chom.2014.02.001

SUMMARY Hp inoculation results in chronic infection, but after clearance with antihelminthic drugs, secondary inoculation triggers a The type 2 immune response evoked by intestinal strong type 2 memory response, with worm expulsion by 2 weeks nematode parasites contributes to worm expulsion (Anthony et al., 2007; Reynolds et al., 2012). and tolerance to associated tissue damage. We Adenosine is an endogenous purine nucleoside, a catabolite investigated whether this host response is affected of ATP, and can regulate multiple physiological processes by by blocking signaling by the putative endogenous binding to and activating one or more of four transmembrane danger signal adenosine, which can be released adenosine G protein-coupled cell surface receptors, denoted A ,A ,A , and A adenosine receptors (ARs) (Jacobson and during inflammation and host cell damage. Specific 1 2A 2B 3 Gao, 2006). Inflammation, hypoxia, and cell damage can all blockade of the A2B adenosine receptor (A2BAR) result in accumulation of extracellular adenosine, which may inhibited worm elimination and the development of act as an endogenous regulator of innate immunity (Hasko´ and innate and adaptive components of the type 2 pri- Cronstein, 2004; Hasko´ et al., 2004; Linden, 2001). Several mary and memory response. Infected mice lacking recent studies suggest that adenosine may promote specific

A2BAR exhibited decreased M2 macrophage and components of type 2 immunity. A2BARs can activate mast cells eosinophil recruitment and reduced IL-4 and IL-13 in vitro to secrete IL-4 and dendritic cells to promote CD4 production. Additionally, shortly after infec- IL-4 production (Ryzhov et al., 2004; Yang et al., 2010). A2BAR tion, upregulation of the alarmin IL-33, which drives signaling has further been shown to promote IL-4-induced M2 type 2 immunity, and activation of innate lymphoid macrophage activation in vitro (Cso´ ka et al., 2012). A2BAR type 2 (ILC2) cells was inhibited, while exogenous signaling in macrophages can also induce IL-10 gene expression in vitro (Ne´ meth et al., 2005) and enhance allergen-induced IL-33 restored ILC2 cell activation and type 2 cyto- chronic pulmonary inflammation in an ovalbumin model system kine expression. Thus, adenosine acts as a danger- in vivo (Zaynagetdinov et al., 2010). associated molecular pattern (DAMP) that initiates The factors that initially trigger the type 2 immune response helminth-induced type 2 immune responses through remain uncertain. IL-33, produced by cells in barrier tissues, A2BAR. including epithelial cells, can initially drive both innate and adaptive components of the type 2 mucosal immune response (Mirchandani et al., 2012). But the actual signals that trigger INTRODUCTION the production and/or release of IL-33 remain unclear. Helminth excretory/secretory (ES) products may trigger type 2 immunity Gastrointestinal parasites affect global health, inducing chronic (Everts et al., 2009; Grainger et al., 2010; Hewitson et al., 2009; malnutrition, morbidity, and susceptibility to infection with other Tawill et al., 2004). Tissue damage resulting from parasite traf- infectious agents. Increasing evidence suggests that these path- ficking through tissues may induce release of danger-associated ogens trigger polarized type 2 immune responses (Gaze et al., molecular patterns (DAMPs) (Gause et al., 2013). As the type 2 2012), similar to many experimental mouse models. Heligmoso- response progresses during helminth infection, its potency moides polygyrus (Hp) is a strictly enteric natural murine parasite increases, presumably due to amplification of the immune that is a widely used model for intestinal nematode parasite response and associated sustained signals triggered by hel- infection. Hp induces a polarized and potent type 2 immune minth parasites, which may include DAMPs. response characterized by Th2 cell differentiation and the activa- As yet specific DAMPs have not been identified as playing a tion of multiple immune protective components that contribute significant role in the development of the in vivo type 2 immune to worm expulsion at both the tissue-dwelling and luminal stages response to helminths. In these studies, we examine whether of the life cycle (Herbert et al., 2009; Patel et al., 2009). Primary the complex type 2 immune response evoked by Hp infection

Cell Host & Microbe 15, 339–350, March 12, 2014 ª2014 Elsevier Inc. 339 Cell Host & Microbe Adenosine Promotes Antihelminth-Induced Immunity

Figure 1. Host Protective Immune Response Leading to Worm Expulsion Requires A2BAR Signaling / (A and B) A2BARÀ À and wild-type (C57BL/6) mice were orally inoculated with 200 Hp (HP) L3, 14 days later treated with the antihelminthic drug pyrantel pamoate to expulse parasites, and 4 weeks later given a secondary HP inoculation (Hp20). Controls included mice given only a primary inoculation (Hp10). Mice were sacrificed at 14 days after HP inoculation, and parasite and egg numbers determined (A). Serum immunoglobulins (IgE and IgG1) were measured by ELISA (B). / (C–J) A2BARÀ À and wild-type mice were given an Hp10 or Hp20 inoculation as just described. On day 11 after Hp treatment, prior to worm expulsion, total parasite and egg numbers were enumerated (C). Individual parasites were isolated, and male and female parasites were recovered from corresponding treatment groups at day 11 after HP infection. Protein (D) and ATP (relative luminescence units [RLU]) (E) levels were determined for five to ten male or female worms per mouse. Small intestine Th2 cytokine gene expression was determined by qRT-PCR and expressed as treated/untreated (trt/untrt) (F). In a separate experiment after WT / and A2BARÀ À Hp20, small intestine tissue was collected and stained for periodic acid Schiff (PAS) (G). Small intestines were also stained for wheat germ agglutinin (WGA-PE, red) (H). The mucous index was measured by counting mucous-secreting cells/100 cells/villi (I). Goblet cell RELMb and Muc2 mRNA were measured by qRT-PCR (J). All experiments were repeated two times with similar results. Mean and SE of four to five mice per treatment group are shown. **p < 0.01.

was affected by specific blockade of adenosine signaling A2BAR signaling is essential in the development of the host- through the A2BAR. Surprisingly, our studies demonstrated pro- protective type 2 memory immune response, which is required / found impairment in the absence of A2BAR signaling of the type 2 for Hp expulsion, A2BARÀ À and wild-type (WT) (BL/6) mice protective immune response to Hp. They provide evidence of were orally inoculated with 200 L3. Two weeks later, worms a DAMP essential for the development of the type 2 immune were expelled by administration of the antihelminthic pyrantel response to helminths. pamoate, and four weeks later, mice were challenged with Hp. Worm expulsion was inhibited and egg production was / RESULTS increased in A2BARÀ À mice compared to WT mice at day 14 after Hp secondary inoculation (Figure 1A).

A2BAR Receptors Contribute to Enhanced Worm To examine humoral immunity, we measured total serum Expulsion and Associated Development of the Type 2 immunoglobulins by ELISA at day 14 after Hp secondary inocu- Immune Response lation. Total serum IgE and IgG1 were significantly reduced / Hp triggers a strong, highly polarized type 2 immune response in A2BARÀ À mice compared to WT mice after Hp secondary (Anthony et al., 2007; Svetic et al., 1993). To examine whether inoculation (Figure 1B). Collectively, these findings indicate that

340 Cell Host & Microbe 15, 339–350, March 12, 2014 ª2014 Elsevier Inc. Cell Host & Microbe Adenosine Promotes Antihelminth-Induced Immunity

–/– Figure 2. Alternatively Activated Macrophages (M2) and Eosinophils Are Reduced in Type 2 Granuloma in Hp-Inoculated A2BAR Mice / WT and A2BARÀ À mice were given a primary (Hp10) or secondary Hp inoculation (Hp20), as described in Figure 1. At day 11 after Hp20, small intestine samples were as follows: cryosectioned and stained for macrophages (F4/80, green) and IL-4Ra (red), a marker for M2 cells (A); formalin-fixed and stained with H&E (B); and cryosectioned and stained for eosinophils (anti-MBP, green) (C). Samples are representative of results from four to five mice per treatment group. Arg1, RELMa, and Ym1 iNOS mRNA were analyzed in small intestine by real-time qRT-PCR (D), and the mean and SE from four to five mice/treatment group is shown. Experiments were performed two times with similar results. **p < 0.01.

A2BARs contribute to enhanced worm expulsion and associated IL-4, IL-13, IL-5, and IL-9 mRNA were observed (Figure 1F). serum Ig elevations. Histological analysis showed reduced mucus producing goblet / cells in the small intestine of A2BARÀ À mice compared to WT –/– A2BAR Mice Show Impaired Worm Damage and mice (Figure 1G–1I). mRNA levels of RELMb, secreted by goblet Compromised Peripheral Intestinal Type 2 Memory cells in response to IL-13 (Artis et al., 2004), and Mucin 2 Responses (Muc2), a major mucin secreted by intestinal goblet cells, were

To assess whether the type 2 immune response was affected at also reduced significantly in the absence of A2BAR signaling earlier stages of luminal infection and to directly assess effects of (Figure 1J). / A2BAR deficiency on luminal parasites, A2BARÀ À mice were M2 macrophages contribute to the formation of a type 2 gran- examined at day 11 after primary and secondary Hp inocula- uloma surrounding the tissue-dwelling parasite (Anthony et al., tion—a time point when cytokine elevations are readily detect- 2006; Patel et al., 2009). Swiss roll cryosections were stained able during luminal infection and parasites are still present. Total for F4/80 and IL-4Ra to detect M2 macrophages, as previously / worms and egg production were increased in A2BARÀ À mice described (Anthony et al., 2006). As shown in Figure 2A, immu- at this early time point during luminal infection (Figure 1C). Total nofluorescent imaging of the granuloma showed dual-stained worm ATP and protein levels were assessed to determine M2 cells in WT mice after Hp secondary inoculation; in contrast, / + metabolic activity and growth (a direct measurement of parasitic in A2BARÀ À mice, F4/80 cells, although abundant, were IL- impairment), as previously described (Herbert et al., 2009). Both 4RaÀ. Hematoxylin and eosin (H&E) staining further showed female parasite total protein and ATP levels were significantly multinucleated giant cells in the region vacated by the parasite / increased (p < 0.01) in A2BARÀ À mice compared to WT controls after Hp secondary inoculation of WT mice. In contrast, giant / (Figures 1D and 1E). In intestinal tissue significant reductions in cells were largely absent in granulomas of A2BARÀ À mice

Cell Host & Microbe 15, 339–350, March 12, 2014 ª2014 Elsevier Inc. 341 Cell Host & Microbe Adenosine Promotes Antihelminth-Induced Immunity

Figure 3. A2BAR Deficiency Inhibits Upregulated Th2 Cytokine Gene and Protein Expression A2BAR deficiency inhibits upregulated Th2 cytokine gene and protein expression in draining lymph nodes at day 11 after Hp secondary inoculation. / (A) A2BARÀ À and wild-type mice were given a primary (Hp10) or secondary (Hp20) Hp inoculation as described in Figure 1. At day 11 after inoculation, MLN IL-4, IL-13, IL-5, and IL-9 mRNA were analyzed by qRT-PCR and expressed as treated/untreated (trt/untrt). (B–H) MLN CD4+ T cells were magnetically cell sorted and analyzed for IFN-g, IL-4, and IL-13 mRNA by qRT-PCR (B–D). Total and magnetically sorted CD4+ T cells were isolated and assessed for IL-4 secretion (E and F) and IFN-g secretion (G and H) by ELISPOT. Experiments were repeated two times with similar results, and the mean and SE of four to five mice/treatment group are shown. **p < 0.01.

(Figure 2B). Both H&E (data not shown) and immunofluorescent ure 3C) and IL-13 (Figure 3D) mRNA were reduced in CD4+ / staining for major basic protein (MBP) showed markedly reduced T cells in A2BARÀ À compared to WT mice. The number of IL- / + eosinophils in A2BARÀ À mice compared to WT mice after Hp 4- and IFN-g-secreting MLN cells and sorted CD4 and CD4À secondary inoculation (Figure 2C). Gene expression analysis MLN cells were assessed by ELISPOT. Marked reductions in for M2 markers, Arg1, Ym1, and RELMa, in the small intestine CD4+ IL-4-secreting cells while significant increases in IFN-g- / (Figure 2D), further confirmed greatly reduced M2 cell activation secreting cells were observed in A2BARÀ À mice, compared to / in A2BARÀ À mice, compared to WT mice. In contrast, iNOS, an WT mice, after Hp secondary inoculation (Figures 3E–3H). These M1 macrophage marker, remained low. findings indicate that CD4+ Th2 cell development is impaired in

A2BAR-deficient mice during the memory response to Hp. A2BAR Signaling Is Required for the Polarized Type 2 Response in the Mesenteric Lymph Nodes during Early A2BAR Signaling Is Essential for the Host Protective Luminal Stage Response against Tissue-Dwelling Parasites The compromised immunity to parasites in the intestinal lumen Shortly after entering the small intestine, Hp L3 penetrate the in- after A2BAR blockade suggested that either the localized periph- testinal epithelium and migrate to the submucosal region, where eral effector response at the host-parasite interface was directly they reside as tissue-dwelling larvae for 7 days, when they return impaired or, alternatively, that initial Th2 cell differentiation in the to the lumen as mature adults. Previous studies have suggested draining lymph node, required for optimal peripheral responses, that immune mechanisms mediating host protection at the tis- was inhibited. To address this, mesenteric lymph nodes (MLNs) sue-dwelling phase are distinct from those promoting expulsion / were collected from WT and A2BARÀ À mice at day 11 after Hp at the luminal phase (Anthony et al., 2006; Herbert et al., 2009). secondary inoculation and prepared for quantitative RT-PCR Using previously described techniques (Liu et al., 2010), we

(qRT-PCR). As shown in Figure 3A, IL-4, IL-13, IL-5, and IL-9 next examined whether A2BAR signaling contributes to host pro- / mRNA were significantly reduced in A2BARÀ À mice compared tection at the tissue-dwelling phase. Tissue-dwelling parasites to WT mice. To assess CD4+ T cell , MLN CD4+ and isolated at day 7 after secondary inoculation from individual CD4À cells were isolated using magnetic bead cell sorting of granulomas were decreased in number and length compared pooled suspensions from five mice per treatment group. IFN-g to mice given a primary inoculation only (Figures 4A and 4B), indi- mRNA levels were increased (Figure 3B), while both IL-4 (Fig- cating effective host protection in the tissue-dwelling phase, as

342 Cell Host & Microbe 15, 339–350, March 12, 2014 ª2014 Elsevier Inc. Cell Host & Microbe Adenosine Promotes Antihelminth-Induced Immunity

Figure 4. Protective Type 2 Immune Response Induced by Tissue Dwelling Parasites Is Dependent on A2BAR Signaling In Vivo / (A and B) A2BARÀ À and wild-type mice were given a primary (Hp10) and then secondary (Hp20) inoculation as previously described in Figure 1. At day 7 after inoculation, tissue-dwelling Hp larvae were isolated, sexed, counted (A), and their length determined (B). (C and D) Small intestine IL-4, IL-13, IL-5, Arg1, RELMa, and Ym1 were detected by qRT-PCR and expressed as treated/untreated (trt/untrt) after secondary inoculation. / (E) In a separate experiment, small intestines were collected at day 7 after Hp20 from WT and A2BARÀ À mice and frozen sections stained with F4/80-Alexa488 (green) and anti-IL-4Ra-PE (red) (upper panel) or major basic protein (MBP) (green) (lower panel). (F and G) In a separate experiment at day 8 after primary inoculation, increases in gene expression of type 2 cytokines (F) and M2 markers (G) were also inhibited in / A2BARÀ À mice, which were largely restored by exogenous IL-33 administration. Similar results were obtained in two independent experiments. Mean and SE ([A]–[D], [F] and [G])) are shown for four to five mice/treatment group. *p < 0.05; **p < 0.01. previously described (Liu et al., 2010). However, larvae isolated ary inoculation, analysis of MLN gene expression also showed / from A2BARÀ À mice were significantly higher in number and pronounced decreases in type 2 cytokines and M2 markers at / length compared to WT controls after secondary inoculation, day 7 after Hp inoculation of A2BARÀ À mice compared to WT indicating that A2BAR signaling contributes to host protection mice (data not shown). These results indicate that A2BAR defi- leading to impaired larval development during the tissue dwelling ciency markedly impairs the helminth-induced localized second- phase (Figures 4A and 4B). We also observed reduced IL-4, IL- ary type 2 immune response at the host:parasite interface and 13, and IL-5 mRNA (Figure 4C) and decreased M2 macrophage also significantly reduces effective protective immunity during / markers (Arg1, RELMa, and Ym1)(Figure 4D) in A2BARÀ À the tissue dwelling stage. mice after secondary inoculation in the small intestine. Immuno- fluorescence imaging of granulomas revealed that F4/80+ cells Type 2 Responses after Primary Inoculation Are Blocked double stained with IL-4Ra were absent or reduced dramatically at Tissue Dwelling Stage in A2BAR-Deficient Mice / + in the A2BARÀ À mice (Figure 4E, upper panel). Eosinophil (MBP ) To examine whether type 2 responses are inhibited at early / / accumulation was also blunted in A2BARÀ À mice (Figure 4E, stages after primary inoculation, WT and A2BARÀ À mice were lower panel). Similar to results obtained at day 11 after second- inoculated with Hp and the peripheral intestinal type 2 cytokine

Cell Host & Microbe 15, 339–350, March 12, 2014 ª2014 Elsevier Inc. 343 Cell Host & Microbe Adenosine Promotes Antihelminth-Induced Immunity

Figure 5. Exogenous Administration of IL-33 Rescues Adaptive Components of the Type 2 Immune Response in A2BAR-Deficient Mice after Hp Inoculation / / (A and B) WT and A2BARÀ À mice were orally inoculated with Hp (Hp10). Recombinant IL-33/vehicle was administered to A2BARÀ À mice and 8 days later mesenteric lymph node (MLN) cell suspensions were collected from individual mice and analyzed for IL-4 (A) and IFN-g (B) secretion by ELISPOT. + (C) IL-4 secretion was also assessed for sorted CD4 and CD4À T cells. (D–F) MLN MHCII surface expression, as shown by a representative histogram (D) and mean MFI (E) and the percentage of MLN CD4+ T cells expressing phosphorylated STAT6 (F). In all assays, samples were individually assessed for four to five animals/treatment group. Mean and SE are shown. **p < 0.01. Figure 5, related to Figure S1A. responses assessed at day 8 after inoculation, a time point at pooled from five mice per treatment group showed inhibition of which the MLN T cell cytokine response is consistently high after elevations in CD4+ T cell IL-4 secretion (Figure 5C). Previous primary inoculation (Lu et al., 1996; Svetic et al., 1993). Increases studies have shown that increased B cell MHC-II expression af- in type 2 cytokines (Figure 4F) and M2 markers (Figure 4G) were ter Hp inoculation is IL-4 dependent (Liu et al., 2002; Lu et al., / largely inhibited after primary Hp inoculation of A2BARÀ À mice 1996). B cell surface MHC-II expression was markedly higher / compared to inoculated WT mice. Previous studies have indi- in WT mice compared to A2BARÀ À mice after Hp inoculation cated that IL-33 may drive the development of both innate (Figures 5D and 5E), indicating that A2BAR deficiency impairs and adaptive type 2 immune responses (Hung et al., 2013). IL-4 bioactivity in vivo. We next performed flow cytometric anal- To examine whether exogenous IL-33 administration could ysis to assess CD4+ T cell STAT6 phosphorylation (pSTAT6), as rescue the impaired type 2 immune response following A2BAR previously described (Perona-Wright et al., 2010). A significant / + blockade, A2BARÀ À mice were exogenously administered IL- reduction in the percentage of CD4 T cells expressing pSTAT6 / 33 and inoculated with Hp. At day 8 after primary inoculation, was observed after Hp inoculation in A2BARÀ À compared to IL-33 administration restored elevations in type 2 cytokines (Fig- WT mice, suggesting reduced CD4+ T cell IL-4R signaling in ure 4F) and M2 markers (Figure 4G). As IL-33 administration is the absence of A2BARs (Figure 5F). IL-33 administration in the sufficient to rescue the type 2 response, our findings suggest absence of Hp inoculation was not sufficient to sustain elevated / that adenosine interactions with other more downstream targets IL-4 secretion in A2BARÀ À mice (Figure S1A available online). In of A2BAR signaling, such as M2 macrophages, may not be all these studies, exogenous administration of IL-33 substantially essential if sufficient levels of IL-33 are available to support the restored the type 2 immune response to Hp, raising the possibil- development of the type 2 immune response. ity that one important mechanism of A2BAR-induced type 2 im- MLN cell suspensions were also assessed for cytokine protein munity may be through early upregulation and release of IL-33. expression at day 8 after primary inoculation. As shown in Fig- ure 5A, elevations in IL-4 in unsorted cell suspensions were A2BAR Signaling Is Essential for Initiation of the Type 2 / markedly inhibited in Hp-inoculated A2BARÀ À mice, while IFN- Immune Response Shortly after Hp Inoculation g remained at low levels (Figure 5B). ELISPOT analysis of mag- The immune response to Hp includes both innate and adaptive + netic-bead-sorted CD4 and CD4À T cells from cell suspensions components that are induced shortly after infection and likely

344 Cell Host & Microbe 15, 339–350, March 12, 2014 ª2014 Elsevier Inc. Cell Host & Microbe Adenosine Promotes Antihelminth-Induced Immunity

Figure 6. Early Intestinal Cytokine Response to Invading Parasites Is Dependent on A2BAR Signaling / (A–C) WT and A2BARÀ À mice were given a primary Hp inoculation (Hp10) and 2 days later small intestines were collected and RNA was isolated. IL-25, IL-33, TSLP, IL-4, IL-13, IL-5, RORa, Id2, RORgc, and GATA3 mRNA levels were determined by qRT-PCR and expressed as treated/untreated (trt/untrt) (A–C).

(D–F) In parallel experiments, C57BL/6 mice were given a primary Hp inoculation and administered A2BAR antagonist (MRS1754, 1mg/kg/mouse) or DMSO vehicle i.p. Gene expression for type 2 cytokines (D and E) and transcription factors (F) were determined as above. / (G–I) WT and A2BARÀ À mice were given a primary Hp inoculation (Hp10) and 2 days later small intestine lamina propria cell suspensions were pooled and LinÀ IL- 7Ra+ Sca-1+ c-Kit+ ILC2 cells were isolated by electronic cell sorting. qRT-PCR was used to assess mRNA levels for IL-4, IL-5, and IL-13(G) and Id2 and RORa (H). / Frequency of ILC2 cells was assessed in the small intestine lamina propria in Hp-inoculated WT and A2BARÀ À mice (I). Experiments were repeated two times with similar results, and the mean and SE are shown for four to five mice/treatment group. **p < 0.01.

/ provide an essential microenvironment supporting the develop- A2BARÀ À, mice. Furthermore, elevations in IL-5 and IL-13, but ment of effector and memory Th2 cells. Th2 initiating cytokines not IL-4, were also markedly increased in Hp-inoculated WT may include IL-25, TSLP, and IL-33 and can be produced by controls, consistent with activation of ILC2 cells. In marked epithelial cells through as yet undetermined mechanisms of contrast, elevations in IL-5 and IL-13 were blocked at 48 hr after + / stimulation. In addition to triggering CD4 Th2 cell differentiation, inoculation of A2BARÀ À mice (Figure 6B). To further confirm ILC2 these cytokines also stimulate ILC2 cells (Saenz et al., 2010; activation in the small intestine by 48 hr after inoculation, induc- Wills-Karp and Finkelman, 2011). We next examined whether tion of transcription factors, associated with ILC2 cell activation initial upregulation of Th2-inducing cytokines, or possibly ILC2 (Wong et al., 2012), were assessed. As shown in Figure 6C, intes- / cell activation, was affected by A2BAR deficiency shortly after tinal elevations in RORa and Id2 were decreased in A2BARÀ À Hp infection. Hp larvae typically penetrate the small intestine mice compared to WT mice at 48 hr after Hp inoculation. To epithelial cell barrier within 48 hr after inoculation. As shown in corroborate these findings, the specific A2BAR antagonist, Figure 6A, marked increases in IL-33, but not TSLP or IL-25, MRS1754, was administered to WT mice every 24 hr starting were detected at 48 hr after inoculation in WT, but not in 3 days before Hp inoculation until the termination of the

Cell Host & Microbe 15, 339–350, March 12, 2014 ª2014 Elsevier Inc. 345 Cell Host & Microbe Adenosine Promotes Antihelminth-Induced Immunity

Figure 7. Exogenous Treatment of IL-33 Rescues the Type 2 Innate Immune

Response in A2BAR-Deficient Mice Shortly after Hp Inoculation / / (A and B) WT, A2BARÀ À mice, and A2BARÀ À mice administered exogenous IL-33 were orally

inoculated with Hp (Hp10). Two days later, small intestines were collected, and RNA was isolated. IL-4, IL-13, IL-5, RORa, Id2, RORgc, and GATA3 mRNA levels were determined by qRT-PCR and expressed as treated/untreated (trt/untrt). (C and D) Small intestine lamina propria cell sus- pensions were collected from individual mice, + + + pooled, and ILC2 cells (LinÀ IL-7Ra Sca-1 c-Kit ) were isolated by electronic cell sorting. qRT-PCR was used to assess mRNA levels for IL-4, IL-5, and IL-13(C) and Id2 and RORa (D) and expressed / as treated A2BARÀ À Hp10. Experiments were repeated two times with similar results, and the mean and SE are shown for four to five mice/ treatment group. **p < 0.01.

parable to type 2 cytokines elevations in Hp-inoculated WT control mice. Further- more, the ILC2-activating transcription factors, RORa and Id2, were also upregu- lated in small intestines of IL-33-treated / A2BARÀ À mice (Figure 7B). To specifically analyze the effects of IL-33 on ILC2 cells, + + experiment (day 2). As shown in Figures 6D–6F, at day 2 after ILC2 cells were electronically sorted (LinÀ, IL-7Ra , Sca-1 , and + / inoculation, upregulation of IL-33, IL-5, IL-13, RORa, and Id2 c-kit ) from Hp-inoculated A2BARÀ À mice after exogenous IL-33 were significantly reduced in Hp-inoculated WT mice adminis- treatment. As shown in Figures 7C and 7D, ILC2 cells isolated / tered MRS1754. To specifically assess ILC2 activation and from A2BARÀ À mice expressed high levels of IL-5, IL-13, + + + cytokine expression, ILC2 cells (LinÀ, IL-7Ra , Sca-1 , c-kit ) RORa, and Id2 at 48 hr after Hp inoculation when they were were electronically sorted. As shown in Figure 6G, ILC2 cells treated with IL-33, and these increases were similar to Hp-inoc- expressed IL-5 and IL-13, but not IL-4, mRNA at 48 hr after Hp ulated WT mice. Collectively, these data indicate that exogenous inoculation of WT mice, and these increases were large inhibited administration of the alarmin, IL-33, rescues initiation of the type / in Hp inoculated A2BARÀ À mice. Similarly, increases in Id2 and 2 immune response in the absence of A2BAR signaling, including RORa in ILC2 cells observed in WT mice were largely blocked ILC2 activation and cytokine production, after Hp inoculation. / in A2BARÀ À mice (Figure 6H). Flow cytometric analysis further ATP is a predominantly intracellular molecule; stressful and showed that increases in ILC2 cell frequency in the small injurious events result in extracellular ATP release, and the / intestine after Hp inoculation were reduced in A2BARÀ À mice extracellular ATP is degraded to adenosine via cell surface (Figure 6I). Taken together, these studies indicate that A2BAR ectonucleotidases, including CD39 (nucleoside triphosphate signaling triggers IL-33 upregulation and the initial development diphosphorylase [NTPDase]) and CD73 (50-ectonucleotidase of the innate type 2 response after Hp inoculation. [Ecto50NTase]) (Yegutkin, 2008). As shown in Figure S1B, cell surface CD39 and CD73 levels were increased in the intestinal Exogenous Administration of IL-33 Restores the epithelial cells (IELs), but not lamina propria (LP), at 24 hr after Development of Innate Type 2 Immune Response in the inoculation. These findings indicate that within 24 hr after Hp

Absence of A2BAR Signaling inoculation, surface expression of ectonucleotidases is elevated, A2BAR signaling thus promotes intestinal elevations in IL-33 providing indirect evidence for increased adenosine production expression and subsequent increases in Th2 cytokines shortly at this early time point. after Hp inoculation. This finding raised the possibility that one mechanism through which A2BAR signaling promotes the type The Immune Response to Nippostrongylus brasiliensis –/– 2 response may be through enhanced expression of this alarmin. Is also Impaired in A2BAR Mice To directly test this possibility, we next examined whether exog- It was possible that the immune response to Hp preferentially enous administration of IL-33 could rescue the innate type 2 required A2BAR interactions, compared to type 2 immune re- / immune response in Hp-inoculated A2BARÀ À mice. As shown sponses elicited by other parasites. The protective immune in Figure 7A, elevations in IL-5 and IL-13 mRNA were restored response to the intestinal nematode parasite Nippostrongylus / in A2BARÀ À mice after Hp inoculation such that they were com- brasiliensis (Nb) results in parasite expulsion between 9 and

346 Cell Host & Microbe 15, 339–350, March 12, 2014 ª2014 Elsevier Inc. Cell Host & Microbe Adenosine Promotes Antihelminth-Induced Immunity

/ 11 days after primary inoculation. To examine whether Nb egg is reduced in Hp-inoculated A2BARÀ À mice. Furthermore, IL- production and parasite expulsion were influenced by A2BAR 4R signaling was also inhibited, as evidenced by a reduction / + deficiency, A2BARÀ À mice and WT controls were inoculated in the frequency of MLN CD4 T cells expressing pSTAT6 subcutaneously (s.c.) with Nb L3. At day 9 after primary Nb inoc- (Perona-Wright et al., 2010) and also decreased B cell MHCII ulation, the number of intestinal parasites and the egg count expression (Liu et al., 2002; Lu et al., 1996). Intriguingly, at day / + were markedly elevated in A2BARÀ À mice compared to WT con- 11 after Hp inoculation, CD4 T cell IFN-g mRNA and protein trols (Figure S2A). Furthermore, type 2 cytokines (Figure S2B) secretion were elevated, suggesting immune deviation toward were significantly reduced in the small intestine of Nb-inoculated a type 1 response—a surprising result, as the robust type 2 / A2BARÀ À mice compared to inoculated WT controls. Taken immune response is generally refractory to immune deviation, together, these studies indicate that the protective type 2 pri- as observed following specific blockade of Th2 cytokines and mary immune response to the quite different intestinal nematode costimulatory molecules (Gause et al., 2003). Previous in vitro / parasite, Nb, is impaired in A2BARÀ À mice. studies have shown an inhibitory effect of A2BAR signaling on T cell IFN-g production (Yang et al., 2010). DISCUSSION The general impairment in adaptive type 2 immunity raised the

possibility that A2BAR signaling may be required for initiation Our findings demonstrate an essential role for adenosine inter- of the type 2 immune response during Hp infection. Previous acting with A2BAR in the development of effector immune cell studies have shown that the type 2 immune response to aller- populations that mediate resistance against the intestinal nema- genic extracts can trigger IL-33 release in the lung through tode parasite Hp. They indicate that A2BAR signaling is required ATP-binding P2 purinergic receptors (Kouzaki et al., 2011). Our for the initiation of the type 2 immune response, and they further studies now show that adenosine (a catabolite of ATP) signaling suggest that an important mechanism involves A2BAR-induced through the A2BAR is essential for IL-33 upregulation and asso- expression of the alarmin IL-33, which subsequently triggers ciated activation of ILC2 cells following intestinal helminth infec- ILC-2 and Th2 cell activation and cytokine expression. These tion. Hp is a large multicellular pathogen thought to potentially studies identify a single DAMP required for the development produce multiple factors that together trigger type 2 immunity of the potent and complex helminth-induced type 2 immune (Gause et al., 2013). It is thus perhaps surprising that other response, thereby indicating the importance of danger signals potentially redundant parasite-associated factors cannot substi- in inducing type 2 immunity. tute for its absence. In future studies, A2BAR antagonists could The ability of Hp to trigger a potent type 2 response during be administered at specific time intervals to examine whether the larval tissue-dwelling phase and the adult luminal phase A2BAR specifically contributes to development af- provides a spectrum of elicited immune components, many of ter primary inoculation or memory effector cell activation after which contribute to worm damage and eventual expulsion. In secondary inoculation. Adenosine accumulates in the extracel- the luminal phase, Th2 cytokine-induced increases in goblet lular space following metabolic stress and cell damage (Hasko´ cell secretion of mucus and Relmb impair parasite feeding and and Cronstein, 2004; Hasko´ et al., 2004), primarily as a result contribute to worm expulsion (Herbert et al., 2009). Our findings of CD39 and CD73 enzyme-mediated conversion of ATP to showed expression of these effector molecules was reduced in adenosine (Hasko´ et al., 2008). Adenosine is notoriously difficult / Hp-inoculated A2BARÀ À mice at day 11 after secondary inocu- to measure in vivo due to its rapid degradation by adenosine lation, consistent with our findings that A2BAR signaling was deaminase and uptake into cells through nucleoside trans- required for reduced worm metabolism. The tissue-dwelling porters. In addition, differentiating between extracellular and phase is characterized by a type 2 granuloma, surrounding the intracellular adenosine in solid tissues in vivo requires employ- invading larvae in the submucosa, and is composed primarily ment of techniques such as microdialysis, which at this point of M2 cells, neutrophils, and eosinophils (Patel et al., 2009). has not been tested in a motile organ, such as the gut. Thus,

These studies indicate that A2BAR blockade results in an overall CD39 and CD73 expression, which are regulated primarily by impairment of effector immune cell populations and associated increased cell surface expression, are widely used as sentinels effector molecules required for the development of a protective of adenosine production in the extracellular space (Antonioli immune response culminating in worm expulsion. Although it et al., 2013). Our findings that CD39 and CD73 were elevated remains possible that A2BAR signaling directly inhibits specific in worm-infected mice thus indirectly indicate increased adeno- type 2 immune effector functions, the generally compromised sine production. / nature of the protective response in Hp-inoculated A2BARÀ À Recent studies have shown that particulate structures are suf- mice raised the possibility that Th2 cytokines, required for ficient to act as adjuvants promoting the development of type 2 many of these effector functions, were directly affected. A2BAR immunity (Kuroda et al., 2011; Mishra et al., 2011), suggesting signaling in intestinal tissues has previously been shown to influ- that interactions with inert structures may stress tissues in ence control of inflammation (Frick et al., 2009), raising the pos- such a manner that characteristic DAMPs are released that sibility that effector T cells in peripheral intestinal tissues may then initiate innate and adaptive type 2 responses. Similar dam- preferentially require adenosine and A2BAR signaling for a sus- age may result from migrating multicellular parasites, triggering tained response. It was also possible that the development of release of characteristic DAMPs, including adenosine, which cytokine-producing Th2 cells in the draining MLN was compro- then promote type 2 immunity. Although such interactions may mised. Few studies have yet examined whether CD4+ Th2 cell be necessary for the development of type 2 immunity, they differentiation in vivo requires A2BAR signaling; our findings may not be sufficient, and other factors released by the migrating now show that MLN CD4+ Th2 cell cytokine expression in vivo parasite, including putative pathogen-associated molecular

Cell Host & Microbe 15, 339–350, March 12, 2014 ª2014 Elsevier Inc. 347 Cell Host & Microbe Adenosine Promotes Antihelminth-Induced Immunity

patterns (Reese et al., 2007) and cytokine mimics (Grainger et al., response to the intestinal nematode parasite Hp. They raise 2010; Hewitson et al., 2009), may further influence the course the possibility that adenosine functions as a DAMP that plays a of the type 2 immune response. It was also possible that the significant role in the development of type 2 immunity during immune response to Hp was uniquely dependent on A2BAR helminth infection. signaling and that other parasites may trigger additional initiating signals for type 2 immunity that may substitute for this specific EXPERIMENTAL PROCEDURES AR. However, our findings that the immune response and asso- / Mice ciated resistance to Nb was similarly impaired in A2BARÀ À mice C57BL/6 mice were purchased from the Jackson Laboratory (Bar Harbor, indicate a general significance of adenosine interacting with this / Maine). A2BARÀ À BL/6 mice were bred and housed in a specific pathogen- specific receptor for the development of the helminth-induced free animal facility in the Comparative Medicine Resources center at New Jer- type 2 immune response. Previous studies have also suggested sey Medical School (NJMS). Four to five mice were used per treatment group, that trefoil factor 2 (TFF2) contributes to IL-33 upregulation and if not otherwise indicated. These studies conform to the principles for labora- resistance to Nb (Wills-Karp et al., 2012). In future studies, it tory animal research outlined by the Animal Welfare Act and the Department of will be interesting to examine possible interactions between Health, Education, and Welfare (National Institutes of Health) guidelines for the experimental use of animals and were approved by the IACUC at New Jersey TFF2 and adenosine-A2BAR signaling at early stages of the Medical School. type 2 immune response, as both ligand interactions appear to play a significant role in the generation of this helminth-induced Parasite Inoculation, Length, Burden, IL-33 Administration, and response. A2BAR Antagonist Treatment While adenosine has been previously shown to be involved Mice were inoculated perorally with 200 infective Hp L3 using a rounded in recruitment of immune cells to the site of injury (Hasko´ et al., gavage tube, and adult worm numbers and egg production were quantitated 2008; Schnurr et al., 2004), other studies have suggested that as described previously (Anthony et al., 2006). An antihelminthic drug, pyrantel it generally plays a protective role by dampening inflammatory pamoate (1–2 mg), was administered orally to expulse Hp adults from the gut after primary infection and prior to secondary challenge infection. For adult responses following tissue injury, in this way promoting wound parasite burdens, worms were collected from the small intestine and egg healing and homeostasis (Linden, 2001; Macedo et al., 2007). burdens determined from fecal contents, as previously described (Anthony Similarly, several studies now suggest that a number of compo- et al., 2006). Developing larvae were also sexed and measured in the small in- nents of the type 2 immune response also promote tissue repair, testine at 7 days after Hp primary and secondary inoculations as previously / thereby enhancing the capability of the host to tolerate invading described (Liu et al., 2010). A2BARÀ À mice were injected intraperitoneally multicellular parasites (Chen et al., 2012; Gause et al., 2013). It (i.p.) with 1 mg IL-33 (R&D Systems) daily starting from 1 day before Hp inocu- lation until the day of sacrifice. The specific A AR antagonist N-(4-cyan- may be that adenosine functions as an essential DAMP, trig- 2B ophenyl)-2-[4-(2, 3, 6, 7-tetrahydro-2, 6-dioxo-1, 3-dipropyl-1H-purin-8-yl) gered through tissue damage, that initiates the development of phen-oxy] acetamide (MRS1754) (1 mg/kg body weight) was injected i.p. daily a type 2 immune response during helminth infection that then (Ne´ meth et al., 2007). In a separate experiment, mice were inoculated s.c. with contributes to tissue repair and associated homeostasis. Recent infective Nb L3. Parasite egg numbers and adult worm numbers were evalu- findings that Trichinella-secreted products can block nucleotide- ated as above. mediated mast cell activation suggest that helminths may target this signal as an evasion mechanism (Afferson et al., 2012). Uric Total ATP Assay and Protein Measurement Five to ten male and female worms were isolated after Hp inoculation of acid may also function as a DAMP to trigger type 2 immunity dur- / WT and A2BARÀ À mice, incubated in 100 ml RPMI1640 with 100 ml Cell-Glo ing allergic lung inflammation, and its release is associated with reagents (Promega, Madison), and ground with a motorized pestle. After tissue stress and hypoxia (Kool et al., 2011). Although the mech- centrifugation at 5,300 rpm for 5 min, 100 ml supernatant was transferred to anism through which uric acid triggers type 2 immunity is uncer- wells of 96-well plate and luminescence measured in 1420 multilabel counter tain, recent studies suggest it occurs independently of purinergic luminometer (Perkin Elmer, Waltham). Controls included the following: PBS receptors (Kool et al., 2011). Our studies indicate an important alone, heat killed worms; worms incubated without Cell-Glo reagent. For pro- tein measurement, worms were isolated in 100 l RPMI1640, and the mixture functional link between adenosine signaling through the A AR m 2B ground with a motorized pestle. Parasite extract was centrifuged at 5,300 rpm and upregulation of IL-33, an important inducer of type 2 for 5 min, and total proteins from parasite extracts measured by NanoDrop immunity (Hung et al., 2013), thereby suggesting a potential 2000 Spectrophotometer (Thermo scientific, Wilmington). mechanism for A2BAR signaling in initiating the type 2 immune response. Indeed IL-33 administration effectively restored Real-Time PCR, Serum Immunoglobulins, and ELISPOT / Total RNA was prepared from tissues or sorted cells and then reverse tran- the type 2 primary response in Hp-inoculated A2BARÀ À mice. scribed. Real-Time PCR kits (PE Applied Biosystems, Foster City), specific A2BAR signaling on individual effector cells, such as M2 macro- for individual cytokines or ribosomal RNA, were used to quantify differences phages, may thus not be essential in the context of helminth in gene expression. All data were normalized to constitutive ribosomal RNA infection. It should also be noted that in future studies it will be values and expressed as mRNA fold changes relative to untreated controls, important to examine whether immune regulatory components as previously described (Mishra et al., 2013). Total serum IgE and IgG1 levels of the helminth response, including immune cell expression of were measured by ELISA, and the frequency of IL-4 and IFN-g-producing cells

IL-10 and TGF-b, are also influenced by A2BAR signaling, partic- was determined by ELISPOT assay as previously described (Mishra et al., 2013). ularly given recent studies suggesting that these regulatory cytokines may be induced independently of the canonical type Flourescent Immunohistochemistry and Histology Cryosections were stained for macrophages (F4/80-Alexa 488, green; Caltag 2 immune response (Hu¨ bner et al., 2012; Mishra et al., 2013). Laboratories, Carlsbad), IL-4Ra (mIL-4R-M1-PE, red; BD Pharmingen, San In summary, our studies suggest that specific interactions of Jose), and eosinophils (anti-MBP-Alexa Fluor 488, green; Mayo Clinic, Scotts- adenosine with the A2BAR receptor are essential for the optimal dale). Small intestine sections were formalin fixed and embedded in paraffin, development of the potent and highly polarized type 2 immune and 5–6 mm sections were stained with H&E and PAS. Using previously

348 Cell Host & Microbe 15, 339–350, March 12, 2014 ª2014 Elsevier Inc. Cell Host & Microbe Adenosine Promotes Antihelminth-Induced Immunity

described methods (Anthony et al., 2006), images were taken using either a Anthony, R.M., Rutitzky, L.I., Urban, J.F., Jr., Stadecker, M.J., and Gause, Leica DM6000B or Nikon A1R-A1 confocal microscope and were tiled together W.C. (2007). Protective immune mechanisms in helminth infection. Nat. Rev. using Image-Pro stitching software (Media Cybernetics, Rockville). Immunol. 7, 975–987. Antonioli, L., Pacher, P., Vizi, E.S., and Hasko´ , G. (2013). CD39 and CD73 in Flow Cytometry immunity and inflammation. Trends Mol. Med. 19, 355–367. IEL and LP cells were isolated as previously described (Lefrancois and Lycke, Artis, D., Wang, M.L., Keilbaugh, S.A., He, W., Brenes, M., Swain, G.P., Knight, 2001). Cells were lysed with ACK lysing buffer to remove erythrocytes, P.A., Donaldson, D.D., Lazar, M.A., Miller, H.R., et al. (2004). RELMbeta/FIZZ2 and 0.1 3 106 cells were blocked with Fc Block and directly stained with is a goblet cell-specific immune-effector molecule in the gastrointestinal tract. anti-CD3- FITC (145-2c11, BD PharMingen, San Jose), anti-B220- FITC Proc. Natl. Acad. Sci. USA 101, 13596–13600. (AR3-6B2, BD PharMingen), anti-Gr1- FITC (RB6-8C5, BD PharMingen), anti-CD11b- FITC (M1/70, Biolegend), anti-Sca-1-PE (E13-161.7, BD Chen, F., Liu, Z., Wu, W., Rozo, C., Bowdridge, S., Millman, A., Van Rooijen, N., PharMingen), anti-c-Kit- allophycocyanin(APC) (2B8, BD PharMingen), anti- Urban, J.F., Jr., Wynn, T.A., and Gause, W.C. (2012). An essential role for TH2- CD73-PE (TY/23, BD PharMingen), anti-CD39-Alexa Fluor 647 (24DMS1, type responses in limiting acute tissue damage during experimental helminth eBioscience), and anti-CD127 (IL-7Ra)- PerCP-cy5.5 (A7R34, BD PharMingen) infection. Nat. Med. 18, 260–266. and analyzed by flow cytometry. MLN cell suspensions were collected and Cso´ ka, B., Selmeczy, Z., Koscso´ , B., Ne´ meth, Z.H., Pacher, P., Murray, P.J., / prepared from Hp1- and Hp2-inoculated WT and A2BARÀ À mice; washed; Kepka-Lenhart, D., Morris, S.M., Jr., Gause, W.C., Leibovich, S.J., and blocked with Fc Block (BD PharMingen); and stained with anti-CD4-FITC Hasko´ , G. (2012). Adenosine promotes alternative macrophage activation (RM4-5, BD PharMingen), anti-pSTAT6-PE (pY641,BD PharMingen), anti- via A2A and A2B receptors. FASEB J. 26, 376–386. b B220-FITC(AR3-6B2, BD PharMingen), and anti-MHCII-PE (IA , BD Everts, B., Perona-Wright, G., Smits, H.H., Hokke, C.H., van der Ham, A.J., PharMingen). Phosphorylation of STAT6 at tyrosine 641 was detected by intra- Fitzsimmons, C.M., Doenhoff, M.J., van der Bosch, J., Mohrs, K., Haas, H., cellular staining with PE-conjugated anti-phospho-STAT6 using PhosFlow Fix et al. (2009). Omega-1, a glycoprotein secreted by Schistosoma mansoni Buffer I and Perm Buffer III reagents (BD PharMingen). Cells were acquired on eggs, drives Th2 responses. J. Exp. Med. 206, 1673–1680. a FACSCalibur or LSRII Flow cytometer (BD Biosciences, San Jose), and Frick, J.S., MacManus, C.F., Scully, M., Glover, L.E., Eltzschig, H.K., and analyzed using FlowJo software (Treestar, Ashland). Colgan, S.P. (2009). Contribution of adenosine A2B receptors to inflammatory parameters of experimental colitis. J. Immunol. 182, 4957–4964. Sorting of CD4+ and Innate Lymphoid Cells + Gause, W.C., Urban, J.F., Jr., and Stadecker, M.J. (2003). The immune MLN CD4 and CD4À cells were isolated from MLN using anti-CD4 micro beads, as previously described (Mishra et al., 2011). For sorting ILCs, LP cells response to parasitic helminths: insights from murine models. Trends were isolated from small intestines, as previously described (Lefrancois and Immunol. 24, 269–277. Lycke, 2001), and stained with anti-LinÀ FITC, anti-Sca-1-PE, anti-c-Kit-allo- Gause, W.C., Wynn, T.A., and Allen, J.E. (2013). Type 2 immunity and wound phycocyanin (APC), and anti-CD127 (IL-7Ra)-PerCP-cy5.5. Cells negative healing: evolutionary refinement of adaptive immunity by helminths. Nat. Rev. for Lin and positive for IL-7Ra and Sca-1 were sorted with FACS ARIA 2 cell Immunol. 13, 607–614. sorter (BD Biosciences). Gaze, S., McSorley, H.J., Daveson, J., Jones, D., Bethony, J.M., Oliveira, L.M., Speare, R., McCarthy, J.S., Engwerda, C.R., Croese, J., and Loukas, A. (2012). Statistical Analyses Characterising the mucosal and systemic immune responses to experimental Data were analyzed using SigmaPlot 12 (SYSTAT Software) and are reported human hookworm infection. PLoS Pathog. 8, e1002520. as means ±SEM. Differences between multiple groups were assessed by one- Grainger, J.R., Smith, K.A., Hewitson, J.P., McSorley, H.J., Harcus, Y., Filbey, way analysis of variance, and individual comparisons were analyzed using pro- K.J., Finney, C.A., Greenwood, E.J., Knox, D.P., Wilson, M.S., et al. (2010). tected Fisher-LSD tests. Differences of p < 0.05 were considered statistically Helminth secretions induce de novo T cell Foxp3 expression and regulatory significant. function through the TGF-b pathway. J. Exp. Med. 207, 2331–2341. Hasko´ , G., and Cronstein, B.N. (2004). Adenosine: an endogenous regulator of SUPPLEMENTAL INFORMATION innate immunity. Trends Immunol. 25, 33–39.

Supplemental Information includes two figures and can be found with this Hasko´ , G., Sitkovsky, M.V., and Szabo´ , C. (2004). Immunomodulatory and article online at http://dx.doi.org/10.1016/j.chom.2014.02.001. neuroprotective effects of inosine. Trends Pharmacol. Sci. 25, 152–157. Hasko´ , G., Linden, J., Cronstein, B., and Pacher, P. (2008). Adenosine recep- ACKNOWLEDGMENTS tors: therapeutic aspects for inflammatory and immune diseases. Nat. Rev. Drug Discov. 7, 759–770. This work was supported in part by the National Institutes of Health grants Herbert, D.R., Yang, J.Q., Hogan, S.P., Groschwitz, K., Khodoun, M., Munitz, RO1AI031678 and R01GM066189. The authors have no competing financial A., Orekov, T., Perkins, C., Wang, Q., Brombacher, F., et al. (2009). Intestinal interests. epithelial cell secretion of RELM-beta protects against gastrointestinal worm infection. J. Exp. Med. 206, 2947–2957. Received: August 30, 2012 Hewitson, J.P., Grainger, J.R., and Maizels, R.M. (2009). Helminth immunoreg- Revised: October 22, 2013 ulation: the role of parasite secreted proteins in modulating host immunity. Accepted: February 4, 2014 Mol. Biochem. Parasitol. 167, 1–11. Published: March 12, 2014 Hu¨ bner, M.P., Shi, Y., Torrero, M.N., Mueller, E., Larson, D., Soloviova, K., Gondorf, F., Hoerauf, A., Killoran, K.E., Stocker, J.T., et al. (2012). Helminth REFERENCES protection against autoimmune diabetes in nonobese diabetic mice is inde- Afferson, H.C., Eleftheriou, E., Selkirk, M.E., and Gounaris, K. (2012). pendent of a type 2 immune shift and requires TGF-b. J. Immunol. 188, Trichinella spiralis secreted enzymes regulate nucleotide-induced mast cell 559–568. activation and release of mouse mast cell protease 1. Infect. Immun. 80, Hung, L.Y., Lewkowich, I.P., Dawson, L.A., Downey, J., Yang, Y., Smith, D.E., 3761–3767. and Herbert, D.R. (2013). IL-33 drives biphasic IL-13 production for noncanon- Anthony, R.M., Urban, J.F., Jr., Alem, F., Hamed, H.A., Rozo, C.T., Boucher, ical Type 2 immunity against hookworms. Proc. Natl. Acad. Sci. USA 110, J.L., Van Rooijen, N., and Gause, W.C. (2006). Memory T(H)2 cells induce 282–287. alternatively activated macrophages to mediate protection against nematode Jacobson, K.A., and Gao, Z.G. (2006). Adenosine receptors as therapeutic parasites. Nat. Med. 12, 955–960. targets. Nat. Rev. Drug Discov. 5, 247–264.

Cell Host & Microbe 15, 339–350, March 12, 2014 ª2014 Elsevier Inc. 349 Cell Host & Microbe Adenosine Promotes Antihelminth-Induced Immunity

Kool, M., Willart, M.A., van Nimwegen, M., Bergen, I., Pouliot, P., Virchow, Patel, N., Kreider, T., Urban, J.F., Jr., and Gause, W.C. (2009). J.C., Rogers, N., Osorio, F., Reis e Sousa, C., Hammad, H., and Lambrecht, Characterisation of effector mechanisms at the host:parasite interface during B.N. (2011). An unexpected role for uric acid as an inducer of T helper 2 cell the immune response to tissue-dwelling intestinal nematode parasites. Int. J. immunity to inhaled antigens and inflammatory mediator of allergic asthma. Parasitol. 39, 13–21. Immunity 34, 527–540. Perona-Wright, G., Mohrs, K., and Mohrs, M. (2010). Sustained signaling by Kouzaki, H., Iijima, K., Kobayashi, T., O’Grady, S.M., and Kita, H. (2011). canonical helper T cell cytokines throughout the reactive lymph node. Nat. The danger signal, extracellular ATP, is a sensor for an airborne allergen Immunol. 11, 520–526. and triggers IL-33 release and innate Th2-type responses. J. Immunol. 186, Reese, T.A., Liang, H.E., Tager, A.M., Luster, A.D., Van Rooijen, N., 4375–4387. Voehringer, D., and Locksley, R.M. (2007). Chitin induces accumulation in Kuroda, E., Ishii, K.J., Uematsu, S., Ohata, K., Coban, C., Akira, S., Aritake, K., tissue of innate immune cells associated with allergy. Nature 447, 92–96. Urade, Y., and Morimoto, Y. (2011). Silica crystals and aluminum salts regulate Reynolds, L.A., Filbey, K.J., and Maizels, R.M. (2012). Immunity to the production of prostaglandin in macrophages via NALP3 inflammasome- the model intestinal helminth parasite Heligmosomoides polygyrus. Semin. independent mechanisms. Immunity 34, 514–526. Immunopathol. 34, 829–846. Lefrancois, L., and Lycke, N. (2001). Isolation of mouse small intestinal intrae- pithelial lymphocytes, Peyer’s patch, and lamina propria cells. Curr. Protoc. Ryzhov, S., Goldstein, A.E., Matafonov, A., Zeng, D., Biaggioni, I., and Immunol. 17, 3.19.1–3.19.16. Feoktistov, I. (2004). Adenosine-activated mast cells induce IgE synthesis by B lymphocytes: an A2B-mediated process involving Th2 cytokines IL-4 and Linden, J. (2001). Molecular approach to adenosine receptors: receptor- IL-13 with implications for asthma. J. Immunol. 172, 7726–7733. mediated mechanisms of tissue protection. Annu. Rev. Pharmacol. Toxicol. 41, 775–787. Saenz, S.A., Noti, M., and Artis, D. (2010). Innate immune cell populations Liu, Z., Liu, Q., Pesce, J., Whitmire, J., Ekkens, M.J., Foster, A., VanNoy, J., function as initiators and effectors in Th2 cytokine responses. Trends Sharpe, A.H., Urban, J.F., Jr., and Gause, W.C. (2002). Nippostrongylus Immunol. 31, 407–413. brasiliensis can induce B7-independent antigen-specific development of Schnurr, M., Toy, T., Shin, A., Hartmann, G., Rothenfusser, S., Soellner, J., IL-4-producing T cells from naive CD4 T cells in vivo. J. Immunol. 169, Davis, I.D., Cebon, J., and Maraskovsky, E. (2004). Role of adenosine recep- 6959–6968. tors in regulating chemotaxis and cytokine production of plasmacytoid den- Liu, Q., Kreider, T., Bowdridge, S., Liu, Z., Song, Y., Gaydo, A.G., Urban, J.F., dritic cells. 103, 1391–1397. Jr., and Gause, W.C. (2010). B cells have distinct roles in host protection Svetic, A., Madden, K.B., Zhou, X.D., Lu, P., Katona, I.M., Finkelman, F.D., against different nematode parasites. J. Immunol. 184, 5213–5223. Urban, J.F., Jr., and Gause, W.C. (1993). A primary intestinal helminthic infec- Lu, P., Urban, J.F., Zhou, X.D., Chen, S.J., Madden, K., Moorman, M., Nguyen, tion rapidly induces a gut-associated elevation of Th2-associated cytokines H., Morris, S.C., Finkelman, F.D., and Gause, W.C. (1996). CD40-mediated and IL-3. J. Immunol. 150, 3434–3441. stimulation contributes to proliferation, antibody production, Tawill, S., Le Goff, L., Ali, F., Blaxter, M., and Allen, J.E. (2004). Both free-living eosinophilia, and mastocytosis during an in vivo type 2 response, but is not and parasitic nematodes induce a characteristic Th2 response that is depen- required for T cell IL-4 production. J. Immunol. 156, 3327–3333. dent on the presence of intact glycans. Infect. Immun. 72, 398–407. Macedo, L., Pinhal-Enfield, G., Alshits, V., Elson, G., Cronstein, B.N., and Wills-Karp, M., and Finkelman, F.D. (2011). Innate lymphoid cells wield a dou- Leibovich, S.J. (2007). Wound healing is impaired in MyD88-deficient mice: ble-edged sword. Nat. Immunol. 12, 1025–1027. a role for MyD88 in the regulation of wound healing by adenosine A2A recep- tors. Am. J. Pathol. 171, 1774–1788. Wills-Karp, M., Rani, R., Dienger, K., Lewkowich, I., Fox, J.G., Perkins, C., Lewis, L., Finkelman, F.D., Smith, D.E., Bryce, P.J., et al. (2012). Trefoil factor Mirchandani, A.S., Salmond, R.J., and Liew, F.Y. (2012). Interleukin-33 and the 2 rapidly induces interleukin 33 to promote type 2 immunity during allergic function of innate lymphoid cells. Trends Immunol. 33, 389–396. asthma and hookworm infection. J. Exp. Med. 209, 607–622. Mishra, P.K., Wu, W., Rozo, C., Hallab, N.J., Benevenia, J., and Gause, W.C. Wong, S.H., Walker, J.A., Jolin, H.E., Drynan, L.F., Hams, E., Camelo, A., (2011). Micrometer-sized titanium particles can induce potent Th2-type re- Barlow, J.L., Neill, D.R., Panova, V., Koch, U., et al. (2012). Transcription factor sponses through TLR4-independent pathways. J. Immunol. 187, 6491–6498. RORa is critical for nuocyte development. Nat. Immunol. 13, 229–236. Mishra, P.K., Patel, N., Wu, W., Bleich, D., and Gause, W.C. (2013). Prevention of type 1 diabetes through infection with an intestinal nematode parasite Yang, M., Ma, C., Liu, S., Shao, Q., Gao, W., Song, B., Sun, J., Xie, Q., Zhang, requires IL-10 in the absence of a Th2-type response. Mucosal Immunol. 6, Y., Feng, A., et al. (2010). HIF-dependent induction of adenosine receptor A2b 297–308. skews human dendritic cells to a Th2-stimulating phenotype under hypoxia. Immunol. Cell Biol. 88, 165–171. Ne´ meth, Z.H., Lutz, C.S., Cso´ ka, B., Deitch, E.A., Leibovich, S.J., Gause, W.C., Tone, M., Pacher, P., Vizi, E.S., and Hasko´ , G. (2005). Adenosine aug- Yegutkin, G.G. (2008). Nucleotide- and nucleoside-converting ectoenzymes: ments IL-10 production by macrophages through an A2B receptor-mediated important modulators of purinergic signalling cascade. Biochim. Biophys. posttranscriptional mechanism. J. Immunol. 175, 8260–8270. Acta. 1783, 673–694. Ne´ meth, Z.H., Bleich, D., Cso´ ka, B., Pacher, P., Mabley, J.G., Himer, L., Vizi, Zaynagetdinov, R., Ryzhov, S., Goldstein, A.E., Yin, H., Novitskiy, S.V., E.S., Deitch, E.A., Szabo´ , C., Cronstein, B.N., and Hasko´ , G. (2007). Goleniewska, K., Polosukhin, V.V., Newcomb, D.C., Mitchell, D., Morschl, Adenosine receptor activation ameliorates type 1 diabetes. FASEB J. 21, E., et al. (2010). Attenuation of chronic pulmonary inflammation in A2B aden- 2379–2388. osine receptor knockout mice. Am. J. Respir. Cell Mol. Biol. 42, 564–571.

350 Cell Host & Microbe 15, 339–350, March 12, 2014 ª2014 Elsevier Inc.