Research ArticleInflammation

An anti-inflammatory eicosanoid switch mediates the suppression of type-2 inflammation by helminth larval products

See allHide authors and affiliations

Science Translational Medicine  22 Apr 2020:
Vol. 12, Issue 540, eaay0605
DOI: 10.1126/scitranslmed.aay0605

Tamping down type-2 inflammation

Type-2 immunity in allergy and asthma is driven by nonredundant pathways, sometimes making therapeutic targeting difficult. De los Reyes Jiménez et al. investigated whether factors from a known mediator of type-2 immunity, the helminth Heligmosomoides polygyrus bakeri, could have added beneficial anti-inflammatory effects. They found that helminth larvae modulated eicosanoids in human macrophages and granulocytes. Both the larvae and the helminth protein glutamate dehydrogenase alleviated inflammation in a house dust mite model of asthma. These results suggest that helminth products have the potential to broadly modulate pathways in type-2 immune diseases.

Abstract

Eicosanoids are key mediators of type-2 inflammation, e.g., in allergy and asthma. Helminth products have been suggested as remedies against inflammatory diseases, but their effects on eicosanoids are unknown. Here, we show that larval products of the helminth Heligmosomoides polygyrus bakeri (HpbE), known to modulate type-2 responses, trigger a broad anti-inflammatory eicosanoid shift by suppressing the 5-lipoxygenase pathway, but inducing the cyclooxygenase (COX) pathway. In human macrophages and granulocytes, the HpbE-driven induction of the COX pathway resulted in the production of anti-inflammatory mediators [e.g., prostaglandin E2 (PGE2) and IL-10] and suppressed chemotaxis. HpbE also abrogated the chemotaxis of granulocytes from patients suffering from aspirin-exacerbated respiratory disease (AERD), a severe type-2 inflammatory condition. Intranasal treatment with HpbE extract attenuated allergic airway inflammation in mice, and intranasal transfer of HpbE-conditioned macrophages led to reduced airway eosinophilia in a COX/PGE2-dependent fashion. The induction of regulatory mediators in macrophages depended on p38 mitogen-activated protein kinase (MAPK), hypoxia-inducible factor-1α (HIF-1α), and Hpb glutamate dehydrogenase (GDH), which we identify as a major immunoregulatory protein in HpbE. Hpb GDH activity was required for anti-inflammatory effects of HpbE in macrophages, and local administration of recombinant Hpb GDH to the airways abrogated allergic airway inflammation in mice. Thus, a metabolic enzyme present in helminth larvae can suppress type-2 inflammation by inducing an anti-inflammatory eicosanoid switch, which has important implications for the therapy of allergy and asthma.

INTRODUCTION

Severe type-2 inflammation, such as in asthma or nasal polyposis, represents a major clinical need, which is insufficiently targeted by current treatments (13). Eicosanoids are bioactive metabolites of the polyunsaturated fatty acid (PUFA) arachidonic acid (AA), which play major roles in severe and therapy-resistant forms of type-2 inflammation (4, 5). However, current therapeutics fail to sufficiently modulate eicosanoid pathways (6), resulting in poor clinical efficacy against asthma and nasal polyps (3, 7).

Among the eicosanoids, leukotrienes (LTs), formed via the 5-lipoxygenase (5-LOX) pathway, as well as the cyclooxygenase (COX) metabolite prostaglandin D2 (PGD2) are the key drivers of type-2 inflammation (810). In contrast, prostaglandin E2 (PGE2) and prostacyclin (PGI2) can suppress allergic inflammation and asthma symptoms (1114). Recently, eicosanoids have also been suggested to participate in the type-2 immune response to helminth parasites (1517).

Infection with helminths, tissue damage, or exposure to allergens can trigger type-2 immune responses, which, if not properly controlled, can result in chronic type-2 inflammation (18). However, helminths can also counter-regulate type-2 immune responses, e.g., by inducing regulatory T cells or by targeting innate effector mechanisms, such as interleukin-33 (IL-33), type-2 innate lymphoid cells (ILC2s), or M2 macrophages (1922). Heligmosomoides polygyrus bakeri (Hpb) is a natural parasite of mice with particularly potent regulatory effects on type-2 immune responses. Products of adult Hpb worms can suppress allergic airway inflammation in mice (21, 22), and Hpb larvae can interfere with the innate type-2 immune response that is initiated early after infection (23). On the basis of their potent immunoregulatory potential, helminth products have been suggested as remedies for inflammatory diseases, including allergy and asthma (20, 24). However, whether helminth products can modulate eicosanoid pathways and thus interfere with type-2 inflammation had not been studied.

RESULTS

Treatment with Hpb larval extract suppresses allergic airway inflammation in mice

Hpb larvae modulate innate type-2 immunity (23), and eicosanoids are a critical innate component of the type-2 immune response to house dust mite (HDM) (8, 13, 25). Thus, we hypothesized that Hpb larvae may produce factors that modulate eicosanoid-driven type-2 inflammation. To mimic a desirable therapeutic application, we administered an extract of infective Hpb L3 larvae (HpbE) topically (intranasally) during HDM-induced allergic airway inflammation in mice (Fig. 1A). Intranasal treatment with HpbE together with HDM reduced hallmarks of type-2 inflammation, including airway eosinophilia, goblet cell hyperplasia, and mucin maturation (Fig. 1, B and C). Consistent with increased eosinophil numbers, 15-HETE, a major AA metabolite of eosinophils, was increased in bronchoalveolar lavage fluid (BALF) of HDM-sensitized mice, and treatment with HpbE decreased 15-HETE as well as pro-inflammatory cytokines and chemokines (IL-5, IL-6, eotaxin, and RANTES) (Fig. 1D). Hence, topical treatment with HpbE could suppress the inflammatory response to HDM in the airways.

Fig. 1 Topical treatment with Hpb larval extract (HpbE) modulates type-2 airway inflammation in mice.

(A) Experimental model of house dust mite (HDM)–induced allergic airway inflammation and intranasal (i.n.) treatment with HpbE. (B) BALF cell counts in mice sensitized (1 μg) and challenged (10 μg) with HDM ± intranasal treatment with HpbE (5 μg), 48 hours after the last challenge and treatment. (C) Representative hematoxylin and eosin (H&E)– or periodic acid–Schiff (PAS)–stained lung tissue from mice sensitized to HDM ± treatment with HpbE. Scale bars, 100 μm. (D) Concentrations of 15-HETE (LC-MS/MS) or IL-5, IL-6, eotaxin, and RANTES (Bioplex) in BALF from mice sensitized to HDM ± treatment with HpbE. Results are pooled from two independent experiments (B and D) or representative of stainings performed for two independent experiments (C). Results are presented as means ± SEM; n = 3 to 9 (naïve) or n = 5 to 17 (treated) per group. Statistical significance was determined by Kruskal-Wallis test followed by Dunn’s multiple comparison test. *P < 0.05; **P < 0.01; ***P < 0.001.

HpbE-treated macrophages produce reduced LTs and modulate allergic airway inflammation via PGE2

Genetic ablation or pharmacological inhibition of eicosanoid pathways affects the development of allergic airway inflammation itself (8, 13, 26), thus impeding the assessment of HpbE effects in such models. However, macrophage transfer experiments can provide valuable insights into the role of eicosanoids in allergic airway inflammation (12) because macrophages represent particularly plastic eicosanoid-producing cells that determine the inflammatory response to HDM in the airways (25, 27).

First, we characterized the eicosanoid profile of HpbE-treated bone marrow–derived macrophages (BMDMs) by liquid chromatography–tandem mass spectrometry (LC-MS/MS). To elicit AA release and analyze the full capacity of eicosanoid formation, cells were stimulated with ionophore (A23187). Consistent with the anti-inflammatory potential of HpbE, we observed a shift from type-2–inducing metabolites (PGD2 and LTs) to regulatory metabolites (PGE2) after treatment with HpbE (Fig. 2A). This shift was most likely a result of transcriptional changes in AA-metabolizing enzymes as HpbE induced COX-2 (gene: Ptgs2) and microsomal PGE synthase (mPGES-1, gene: Ptges) while suppressing 5-LOX (Alox5) and leukotriene C4 synthase (Ltc4s) gene expression (Fig. 2B).

Fig. 2 HpbE-treated macrophages produce less LTs and modulate allergic airway inflammation via PGE2.

(A) Eicosanoids (LC-MS/MS) produced by mouse bone marrow–derived macrophages (BMDMs) after treatment with Hpb larval extract (HpbE) (24 hours) and stimulation with A23187 (10 min). (B) Relative gene expression of AA-metabolizing enzymes [quantitative polymerase chain reaction (qPCR)] in mouse BMDM treated with HpbE. (C) Heat map showing major PUFA metabolites (LC-MS/MS) produced by human monocyte-derived macrophages (MDMs) ± treatment with HpbE followed by A23187 (10 min). (D) Amounts of major bioactive AA metabolites (LC-MS/MS) produced by human MDM ± treatment with HpbE +A23187 (10 min). (E) Relative gene expression of eicosanoid pathway proteins (qPCR) in human MDM ± treatment with HpbE. (F) Frequencies of total leukocytes and total numbers of eosinophils in BALF of mice sensitized to HDM + intranasal transfer of HpbE-conditioned or untreated BMDM (MΦ) [wild type (wt) or Ptges−/−], 18 hours after the last challenge and transfer. Dashed lines indicate eosinophil prevalence in HDM-sensitized mice. (G) Representative H&E-stained lung tissue from mice sensitized to HDM ± intranasal transfer of untreated or HpbE-conditioned BMDM (wt or Ptges−/−). Scale bars, 100 μm. Data are presented as means ± SEM; n = 8 BMDM from C57BL/6 mice (A and B); n = 10 to 15 MDM from healthy human blood donors (C to E); n = 7 to 9 mice per group (F and G). Statistical significance was determined by Wilcoxon test (A to E) or Mann-Whitney test (F). *P < 0.05; **P < 0.01; ***P < 0.001.

To investigate whether the eicosanoid-modulatory effects of HpbE could be translated to human macrophages, we treated human monocyte-derived macrophages (MDMs) with HpbE and assessed their lipid mediator profile after stimulation with A23187. Using an LC-MS/MS eicosanoid screen [including 200 different eicosanoids and PUFAs (table S1)], we confirmed that HpbE treatment resulted in fundamental changes in PUFA metabolites, with the largest changes observed for COX metabolites (Fig. 2C and fig. S1, A and B). Consistent with mediator profiles of murine BMDM, HpbE triggered a shift from pro-inflammatory 5-LOX metabolites (LTB4, LTC4, and 5-HETE) to PGE2, TXB2, and 12-hydroxyheptadecatrenoic acid (12-HHT) (Fig. 2D). It also enhanced the synthesis of the pro-resolving mediator lipoxin A5 (LXA5) (Fig. 2D). Thus, HpbE induced a broad and potentially anti-inflammatory eicosanoid switch.

In line with HpbE-induced transcriptional changes in BMDM, human macrophages responded to HpbE by inducing the expression of enzymes involved in the biosynthesis of PGE2: COX-2 (PTGS2) and mPGES-1 (PTGES) (Fig. 2E). In contrast, HpbE reduced the expression of PGD2 synthase (PTGDS) as well as of LT biosynthetic enzymes ALOX5, leukotriene A4 hydrolase (LTA4H), and LTC4S and the high-affinity receptor for cysLTs [cysteinyl leukotriene receptor-1 (CYSLTR1)] (Fig. 2E). Together, HpbE triggered a switch from a pro-inflammatory 5-LOX–dominated to a type-2–suppressive COX-dominated eicosanoid profile.

To define the in vivo relevance of the HpbE-driven induction of COX metabolites in macrophages, we intranasally transferred BMDM from wild-type (wt), COX-2–deficient (Ptgs2−/−), or mPGES-1–deficient (Ptges−/−) mice during HDM-induced airway inflammation.

Mice that received HpbE-treated wt BMDM during challenges with HDM showed reduced eosinophil numbers and airway inflammation as compared to mice that received untreated wt BMDM (Fig. 2, F and G). In contrast, transfer of HpbE-treated Ptges−/− BMDM failed to reduce eosinophil numbers and eosinophilic airway inflammation compared to transfer of untreated Ptges−/− BMDM (Fig. 2, F and G). Last, HpbE-treated Ptgs2−/− BMDM transfer led to exaggerated HDM-induced airway inflammation (fig. S1C). This suggested that macrophage-derived COX metabolites, particularly mPGES-1–derived PGE2, can contribute to the anti-inflammatory effects of HpbE in vivo.

HpbE induces type-2–suppressive cytokines and prevents M2 polarization

To investigate whether treatment with HpbE also modified cytokine profiles and the polarization of macrophages, we quantified cytokines implicated in macrophage polarization and the regulation of type-2 inflammation. Treatment of human MDM with HpbE resulted in the induction of IL-10, IL-1β, IL-12, IL-18, IL-27, and tumor necrosis factor–α (TNF-α) (Fig. 3, A and B), all known to modulate M2 polarization and type-2 immune responses (23, 2832). However, HpbE hardly affected the production of mediators of type-2 inflammation (IL-33 or CCL17) by macrophages (Fig. 3B). The HpbE-triggered induction of IL-10 and IL-1β also occurred in murine BMDM, albeit at 10- to 100-fold lower amplitude as compared to human MDM (Fig. 3C).

Fig. 3 HpbE induces type-2–suppressive cytokines and prevents M2 polarization.

(A) Amounts of IL-10 and IL-1β (ELISA) produced by human MDM (n = 14 to 15) ± treatment with HpbE. (B) Amounts of TNF-α, IL-6, IL-12p70, IL-18, IL-27, IL-33, and CCL17/TARC (Bioplex) produced by human MDM (n = 3) after treatment with HpbE. (C) Amounts of IL-10 and IL-1β (Bioplex) produced by mouse BMDM (n = 8) ± treatment with HpbE. (D) Gene expression of M2 markers (qPCR) in human MDM (n = 9 to 10) ± treatment with HpbE. (E) Gene expression of M2 markers (qPCR) in mouse BMDM (n = 5) ± treatment with HpbE. Data are presented as means ± SEM. Statistical significance was determined by Wilcoxon test. **P < 0.01; ***P < 0.001. n.d., not detectable.

In addition, HpbE down-regulated the expression of M2 markers {ALOX15 [15-lipoxygenase (15-LOX)] and MRC1 (mannose receptor C-type 1, MR/CD206)}, but not transglutaminase-2 (TGM2) in human MDM, suggesting that it could partially counteract M2 polarization (Fig. 3D). Because human and mouse M2 macrophages are defined by distinct sets of markers (33), we also investigated the effect of HpbE on murine M2 polarization. However, HpbE did not significantly affect M2 marker genes (Tgm2, Arg1, Mrc1, Tmed1/St2l, or Retnla/Fizz1) in mouse BMDM (Fig. 3E). Together, these data suggest that HpbE can broadly modulate the polarization and mediator output of human macrophages to induce a regulatory, type-2–suppressive phenotype.

HpbE has a unique potential to induce type-2–suppressive mediators

For the treatment of complex type-2 inflammatory diseases such as allergy, asthma, and nasal polyps, regulation of multiple pathways is often superior to targeting single mechanisms. Glucocorticoids (GCs), which regulate a broad array of inflammatory pathways, are widely used in the treatment of these diseases and still represent the first-line therapy for most patients. Thus, we compared the immunoregulatory effects of HpbE to those of GCs [dexamethasone (Dex) and fluticasone propionate (FP)] with a focus on eicosanoid pathways and the anti-inflammatory cytokine IL-10. While HpbE triggered a shift from pro-inflammatory genes (ALOX5, LTC4S, and PTGDS) and 5-LOX metabolites to anti-inflammatory pathways (PTGS2, PTGES, PGE2, and IL-10), GCs failed to significantly regulate these pathways (fig. S2, A to C).

In human granulocytes [polymorphonuclear leukocytes (PMNs)], which also represent an important source of eicosanoids, HpbE and FP, but not Dex, induced a similar shift from pro- to- anti-inflammatory metabolites of AA or linoleic acid (fig. S2D) (34). Together, this suggested that compared to GCs, HpbE is more efficient in inducing a regulatory mediator profile that can counteract type-2 inflammation.

During their life cycles, most helminths develop from free-living infective larval stages (L1 to L3) via tissue-migratory stages (L4) to an adult or juvenile stage (L5). Because larval stages of Schistosoma mansoni as well as excretory secretory products of adult (L5) Hpb (HES) can induce type-2–suppressive mediators (35, 36), we compared S. mansoni– or HES-elicited effects on eicosanoids and IL-10 to those of HpbE. An extract of S. mansoni larvae (SmE) failed to induce a shift from 5-LOX to COX metabolism and was less potent in triggering IL-10 production as compared to HpbE (fig. S3, A and B). Similarly, adult-stage HES failed to induce the COX pathway or IL-10 (fig. S3, C and D). In addition, extracts of L4 and L5 stages of Hpb did not show any induction of PGE2 and exhibited only minor suppressive effects on cysLTs as compared to L3 stage extract (HpbE) (fig. S3E). In contrast to L3 extract, L4 and L5 extracts did not induce type-2–suppressive cytokines (IL-1β and IL-10; fig. S3E).

As changes in the microbiota contribute to the suppression of type-2 inflammation by Hpb infection (37), we further identified HpbE-associated bacteria by aerobic or anaerobic culture of plated HpbE followed by MS and assessed whether a mix of HpbE-associated bacteria would exert similar effects as HpbE. However, COX metabolites, IL-10, and COX pathway genes remained unaffected by treatment with HpbE-associated bacteria (fig. S3, F and G). To further exclude that the HpbE-triggered induction of regulatory mediators was due to lipopolysaccharide (LPS) contamination, we additionally quantified mediator profiles of macrophages treated with LPS at the concentration present in HpbE (60 ng/ml). However, LPS alone failed to significantly induce COX metabolites (fig. S3H). Together, this suggested that HpbE has a unique immunoregulatory profile, which is distinct from GCs and somatic extracts from S. mansoni larvae or more mature stages of Hpb as well as from Hpb-associated bacteria.

HpbE modulates eicosanoid production and chemotaxis of human granulocytes

Next, we determined whether HpbE would also affect the PUFA metabolism of human granulocytes by LC-MS/MS analysis. To limit apoptosis and mimic an inflammatory setting, granulocytes were cultured in the presence of granulocyte-macrophage colony-stimulating factor (GM-CSF), which resulted in 40 to 50% viability after 24 hours. Treatment with HpbE reduced neutrophil viability but had no effect on the viability of eosinophils (fig. S4A). We confirmed that neutrophils and eosinophils show abundant expression of transcripts for LT biosynthetic enzymes (ALOX5, LTA4H, and LTC4S) at baseline (38), allowing them to synthesize high amounts of LTs upon ionophore stimulation (Fig. 4, A to E, and fig. S4B). In line with the profiles observed for macrophages, granulocytes showed an induction of COX metabolites (particularly 12-HHT and TXB2) after treatment with HpbE (Fig. 4, A and B, and fig. S4C). Furthermore, the abundance of 5-LOX metabolites was reduced by HpbE treatment in ionophore-stimulated mixed human granulocytes as well as in purified eosinophils (Fig. 4, A and C). Similar to HpbE-driven changes in AA metabolism genes in macrophages, transcripts for enzymes involved in the synthesis of pro-inflammatory mediators (ALOX5, LTA4H, and PTGDS) were down-regulated, whereas PTGS2 and PTGES were induced by HpbE in human granulocytes (Fig. 4, D and E).

Fig. 4 HpbE modulates eicosanoid profiles and chemotaxis of granulocytes in a human setting of type-2 inflammation.

(A) Heat map showing major PUFA metabolites (LC-MS/MS) produced by mixed human granulocytes ± treatment with HpbE (24 hours) followed by A23187 (10 min). (B) Amounts of major eicosanoids (LC-MS/MS) produced by mixed human granulocytes ± treatment with HpbE (24 hours) + A23187 (10 min). (C) Amounts of cysLTs [enzyme immunoassay (EIA), validated by LC-MS/MS] produced by purified human eosinophils ± treatment with HpbE (24 hours) + A23187 (10 min). (D) Relative gene expression of AA-metabolizing enzymes (qPCR) in mixed human granulocytes ± treatment with HpbE. (E) Expression of LT synthetic enzymes (LTC4S and LTA4H) (flow cytometry) in human eosinophils ± treatment with HpbE. (F) Chemotaxis of granulocytes from patients with AERD toward nasal polyp secretions ± treatment with HpbE/fluticasone propionate (FP)/montelukast (MK). Dashed line depicts basal migration. (G) Surface expression of chemotactic receptors (CCR3 and CRTH2) (flow cytometry) in human eosinophils ± treatment with HpbE. (H) Chemotaxis of mixed human granulocytes ± pretreatment with conditioned medium from MDM (±HpbE, ±COX inhibitor indomethacin). Dashed line depicts basal migration. Data are pooled from at least three independent experiments and presented as means ± SEM; n = 6 to 9 mixed granulocytes or purified eosinophils from human blood donors. Statistical significance was determined by Wilcoxon test (two groups) or Friedman test (four groups). *P < 0.05; **P < 0.01.

Eicosanoid-driven granulocyte recruitment represents a key event in type-2 inflammation (5, 8). Thus, we studied how HpbE would affect granulocyte recruitment in a clinically relevant setting of type-2 inflammation, in which AA metabolites play a major role. We collected granulocytes and nasal polyp secretions from patients suffering from aspirin-exacerbated respiratory disease (AERD) and assessed the effects of HpbE on the migration of patient granulocytes toward nasal polyp secretions ex vivo. Pretreatment of AERD granulocytes with HpbE resulted in a marked reduction in cell recruitment, an effect not achieved by FP or the cysLT1R antagonist montelukast (MK), which is commonly used to treat AERD symptoms (Fig. 4F). In keeping with the suppression of granulocyte chemotaxis, HpbE reduced surface expression of chemotactic receptors [C-C chemokine receptor type 3 (CCR3) and PGD2 receptor 2 (CRTH2)] on human eosinophils (Fig. 4G). To investigate whether COX metabolites released by HpbE-treated human macrophages could affect granulocyte recruitment, we performed chemotaxis assays in the presence of conditioned medium from MDM treated with HpbE and the nonselective COX inhibitor indomethacin. Conditioned medium from HpbE-treated human macrophages reduced granulocyte chemotaxis, and indomethacin partially restored cell recruitment (Fig. 4H). Thus, either directly or by acting on macrophages, HpbE can suppress the chemotaxis of granulocytes, including those from patients experiencing severe type-2 inflammation.

HpbE induces IFN-γ, IL-10, and an anti-inflammatory eicosanoid switch in human PBMCs

To test whether the regulatory potential of HpbE extended to type-2 cytokines, we analyzed IL-4, IL-5, and IL-13 expression in human peripheral blood mononuclear cells (PBMCs) after treatment with HpbE. Type-2 cytokines were hardly affected by HpbE, which instead triggered a marked induction of interferon-γ (IFN-γ) and IL-10 (fig. S5, A and B). In line with eicosanoid modulation in macrophages and granulocytes, HpbE treatment of PBMCs also triggered the synthesis of prostanoids (PGE2 and TXB2) while decreasing 5-LOX metabolites (5-HETE and LTB4) (fig. S5C). However, in contrast to macrophages and granulocytes, HpbE-treated PBMCs produced high amounts of 12-/15-LOX metabolites (fig. S5C), which can be metabolized into pro-resolving mediators (39). The modulation of IL-10 and eicosanoids in PBMCs was entirely dependent on CD14+ monocytes, as CD14 PBMCs produced low amounts of these mediators with no apparent regulation by HpbE (fig. S5D). Thus, HpbE acted predominantly on monocytes/macrophages to induce a regulatory and potentially pro-resolving eicosanoid profile.

The HpbE-induced regulatory eicosanoid switch depends on HIF-1α and the COX pathway

To identify mechanisms by which HpbE could trigger the production of type-2–suppressive mediators, we genetically or pharmacologically targeted regulatory pathways and studied eicosanoid profiles and macrophage polarization. Because hypoxia-inducible factor-1α (HIF-1α) is a positive regulator of the COX pathway (40), we first assessed the effect of HpbE on HIF-1α activation and COX-2 expression in BMDM. After treatment with HpbE, nuclear translocation of HIF-1α and expression of COX-2 were increased (Fig. 5A). In contrast to wt BMDM, HIF-1α–deficient BMDM (HIF-1αfl/flxLysMCre) failed to up-regulate TXB2 and PGE2 in response to HpbE, whereas the suppression of (PGD2 and LTB4) remained intact (Fig. 5B). In addition, HIF-1α–deficient BMDM showed a reduced HpbE-driven induction of IL-6, TNF-α, and IL-10 as well as of the M2 markers Tgm2 and Arg1 (Fig. 5, C and D). Expression of Mrc1 and Retnla was generally higher in BMDM lacking HIF-1α, but HpbE down-regulated Mrc1 expression regardless of HIF-1α (Fig. 5D). Thus, the induction of type-2–suppressive mediators in BMDM was largely dependent on HIF-1α.

Fig. 5 Induction of type-2–suppressive macrophages by HpbE is mediated via HIF-1α, p38 MAPK, and COX.

(A) Representative immunofluorescence staining of HIF-1α, COX-2, 4′,6-diamidino-2-phenylindole (DAPI) (cell nuclei), and F4/80 in BMDM ± treatment with HpbE. (B) Eicosanoid production (LC-MS/MS) in BMDM (wt or HIF-1αfloxed/floxedxLysMCre) ± treatment with HpbE (24 hours) + A23187 (10 min). (C) Amounts of IL-6, TNF-α, IL-1β, or IL-10 (Bioplex) in BMDM (wt or HIF-1αfloxed/floxedxLysMCre) ± treatment with HpbE. (D) Gene expression of M2 markers (qPCR) in BMDM (wt or HIF-1αfloxed/floxedxLysMCre) ± treatment with HpbE. (E) Protein amounts of phospho-p38, total p38, COX-2, or β-actin (Western blot) in MDM ± treatment with HpbE. Left, representative blots for n = 3 blood donors; right, quantification for n = 5 to 9 donors. (F and G) Amounts of IL-10 or IL-1β (ELISA) in MDM ± treatment with HpbE ± inhibitors of p38 (VX-702), COX (indomethacin), or HIF-1α (acriflavine). (H) Fold change of eicosanoid enzymes in MDM treated with HpbE ± inhibitors of p38 (VX-702), COX (indomethacin), or HIF-1α (acriflavine). Dotted lines indicate expression in untreated cells. Data are pooled from at least two independent experiments and presented as means ± SEM; n = 5 to 9. Statistical significance was determined by two-way ANOVA (A to D), Wilcoxon test (E), or Friedman test (F to H). *P < 0.05; **P < 0.01; ***P < 0.001.

Because HIF-1α is positively regulated by the mitogen-activated protein kinase (MAPK) p38, we studied the involvement of p38 signaling in the induction of type-2–suppressive mediators by HpbE. In human MDM, p38 was phosphorylated upon treatment with HpbE, correlating with the induction of COX-2 (Fig. 5E). A p38 inhibitor (VX-702) abrogated the induction of IL-10, IL-1β, and PGE2 synthetic enzyme transcripts (PTGS2 and PTGES) (Fig. 5, F to H). In line with HIF-1α–dependent regulation in murine BMDM, a pharmacological inhibitor of HIF-1α (acriflavine) attenuated the HpbE-induced expression of IL-10, IL-1β, and COX pathway enzymes in human MDM (Fig. 5, F to H). However, p38 and HIF-1α were not responsible for the modulation of the 5-LOX pathway, which was still active in the presence of inhibitors (Fig. 5H).

To investigate whether the HpbE-triggered production of IL-10 and IL-1β occurred downstream of the COX pathway, we studied whether COX inhibitors could modify the induction of these cytokines in MDM. A nonselective COX inhibitor (indomethacin), but not a selective COX-2 inhibitor (CAY10404), reduced the induction of IL-10, IL-1β, and PTGES (Fig. 5, F to H, and fig. S6, A and B). In contrast, the HpbE-triggered expression of COX-2 was reduced both by indomethacin and by selective inhibition of COX-2, suggesting that COX-2 metabolites could drive an autocrine feedback loop to promote COX-2 expression (Fig. 5H and fig. S6B).

Because the transcription factor nuclear factor κB (NF-κB) and the kinases phosphatidylinositol 3-kinase (PI3K), protein kinase A, and PTEN can regulate eicosanoid pathways, we additionally assessed the contribution of these mechanisms to the induction of type-2–suppressive mediators by HpbE. Inhibition of NF-κB (by BAY 11-7085) reduced PGE2, IL-10, and IL-1β production as well as gene expression of PGE2 synthetic enzymes and IL-10 in HpbE-treated human MDM (fig. S6, C and D). In contrast, inhibitors of PI3K, protein kinase A, or PTEN did not interfere with the induction of PGE2, IL-10, or IL-1β (fig. S6E). Thus, the HpbE-driven induction of type-2–suppressive mediators largely depended on the activation of p38 MAPK, HIF-1α, NF-κB, and the COX pathway.

TLR2, dectin-1, and dectin-2 contribute to the induction of the COX pathway by HpbE

To further elucidate the upstream mechanisms underlying prostanoid and cytokine modulation by HpbE, we blocked IL-1β or pattern recognition receptors [PRRs; Toll-like receptor 2 (TLR2), dectin-1, and dectin-2], which had all been previously linked to helminth-driven immunoregulation (15, 23, 41, 42). Blockade of IL-1β in MDM affected the HpbE-driven modulation neither of IL-10 nor of eicosanoid pathways (fig. S7A). However, neutralizing antibodies against TLR2, dectin-1, or dectin-2 reduced the induction of PGE2 synthetic enzymes by HpbE, whereas the modulation of IL-10 or 5-LOX was not affected (fig. S7, A and B). This suggested that TLR2, dectin-1, and dectin-2 contributed to the induction of the COX pathway but not to other immunoregulatory effects by HpbE.

Glutamate dehydrogenase is a major immunoregulatory protein in HpbE

As the above-defined mechanisms did not provide a molecular explanation for the immunoregulatory effects of HpbE, we further characterized its active components. Heat inactivation of HpbE attenuated the induction of prostanoids, IL-10, and IL-1β in MDM as well as the HpbE-driven suppression of granulocyte recruitment (Fig. 6, A and B). In addition, the induction of IL-10 by HpbE was abrogated if the extract was pretreated with proteinase K (Fig. 6C). This suggested that the induction of type-2–suppressive mediators by HpbE was largely dependent on heat-labile and proteinase K–digestible molecules, most likely proteins.

Fig. 6 GDH is a major immunoregulatory factor in HpbE.

(A and B) Amounts of prostanoids (EIA) or IL-10 and IL-1β (ELISA) in human MDM (A) or chemotaxis of human PMNs (B) ± treatment with HpbE or heat-inactivated HpbE (HpbE 90°C). (C) Amounts of IL-10 (ELISA) in human MDM ± treatment with HpbE ± pretreatment with proteinase K (prot K). (D) Size exclusion chromatogram for fractionation of HpbE. (E) Amounts of TXB2 (EIA) or IL-10 (ELISA) in MDM ± treatment with HpbE fractions. (F) Summary of results from MS identification of proteins in active fractions of HpbE. (G) Amounts of PGE2 (EIA) or IL-10 (ELISA) in MDM ± treatment with HpbE ± inhibitor of GDH (GDHi; bithionol, 20 μM). (H) Amounts of PGE2 or total COX metabolites (LC-MS/MS) in MDM ± treatment with HpbE ± inhibitor of GDH (bithionol, 20 or 100 μM). (I) Amounts of PGE2 (EIA) or IL-10 (ELISA) in MDM ± treatment with HpbE ± different dilutions of a monoclonal antibody (clone 4F8) against Hpb GDH. Data are pooled from at least two independent experiments and presented as means ± SEM for MDM from n = 2 to 10 healthy human blood donors. Statistical significance was determined by Friedman test. *P < 0.05; **P < 0.01; ***P < 0.001.

To identify immunoregulatory proteins present in HpbE, we fractionated the extract by size exclusion chromatography and identified active fractions based on the capacity to induce the COX metabolite TXB2 as well as IL-10 (Fig. 6, D and E). We then identified proteins present in active and non-active fractions by MS. Hpb glutamate dehydrogenase (GDH) was uniquely present in active fractions of HpbE, qualifying it as an immunoregulatory candidate (Fig. 6F and table S2). An inhibitor of GDH (bithionol), which is also used as an anti-helminthic, reduced the HpbE-triggered induction of prostanoids and IL-10 in a dose-dependent manner without affecting cell viability (Fig. 6, G and H, and fig. S8A). In line with the unique anti-inflammatory properties of L3 (HpbE) versus L4 and L5 extracts (fig. S3E), HpbE showed higher GDH activity as compared to L4 and L5 preparations (fig. S8B).

To further validate Hpb GDH as a major immunoregulatory component of HpbE, we generated monoclonal antibodies specific for Hpb GDH [i.e., not cross-reactive with mammalian (human/mouse) GDH] (fig. S9A). Administration of an anti-Hpb GDH antibody (clone 4F8), but not an isotype control antibody, resulted in a dose-dependent reduction of the HpbE-induced production of IL-10 and PGE2 in human MDM (Fig. 6I and fig. S9B).

Recombinant Hpb GDH reduces allergic airway inflammation in mice

Last, we developed a strategy for the recombinant production of Hpb GDH, allowing us to directly assess immunoregulatory effects of the protein in vitro and in vivo. Recombinant Hpb GDH was active, and its activity was partially inhibited by bithionol (fig. S9C). In human MDM, treatment with Hpb GDH induced PGE2 and IL-10, but reduced cysLT production, thus recapitulating key anti-inflammatory effects of total HpbE (Fig. 7A).

Fig. 7 Recombinant Hpb GDH induces anti-inflammatory mediators in human macrophages and suppresses allergic airway inflammation in mice.

(A) Amounts of eicosanoids (LC-MS/MS/EIA) or IL-10 (ELISA) in human MDM ± treatment with recombinant Hpb GDH (5 μg/ml) (24 hours) ± A23187 (10 min). (B) BALF cell counts in mice sensitized (1 μg) and challenged with HDM (10 μg) ± intranasal treatment with Hpb GDH (10 μg), 48 hours after the last challenge and treatment. (C) Representative H&E- or PAS-stained lung tissue from mice sensitized to HDM ± treatment with Hpb GDH. Scale bars, 100 μm. Results are pooled from at least two independent experiments (A and B) or representative of stainings performed for two independent experiments (C). Results are presented as means ± SEM for MDM from n = 7 to 9 healthy human blood donors or n = 9 to 10 mice per group. Statistical significance was determined by Friedman test (A) or Kruskal-Wallis test followed by Dunn’s multiple comparison test (B). *P = 0.05; **P = 0.01; ***P < 0.001.

To validate the use as a potential therapeutic enzyme in vivo, mice were treated intranasally with Hpb GDH during HDM-induced allergic airway inflammation. Administration of Hpb GDH attenuated the HDM-triggered eosinophil infiltration and airway inflammation as well as goblet cell hyperplasia and mucin maturation (Fig. 7, B and C).

DISCUSSION

Searching for new approaches to target eicosanoids in type-2 inflammation, we identified an extract of a parasitic nematode (HpbE) and its active component Hpb GDH. HpbE showed broad immunoregulatory effects in various myeloid cell types and when administered topically to the airways of allergen-sensitized mice. In a human ex vivo setting of type-2 inflammation, HpbE effectively reduced the chemotaxis of granulocytes toward nasal polyp secretions. The effects of HpbE on the chemotaxis of granulocytes from patients with AERD were particularly notable because this disease is characterized by an aberrant AA metabolism and AA signaling as well as GC-resistant type-2 inflammation (5).

Of note, although HpbE is derived from a murine parasite, it showed largely overlapping anti-inflammatory effects on eicosanoids and granulocyte activation in human cells ex vivo and in mice in vivo. HpbE likely affected granulocytes directly (e.g., by modulating eicosanoids and chemotactic receptors) as well as indirectly by inducing granulocyte-regulatory factors such as PGE2 in macrophages. Although other cell types can also contribute to eicosanoid production during allergic airway inflammation, we focused on macrophages as a particularly plastic, long-lived, and thus therapeutically accessible source of these mediators. The macrophage pool in inflamed airways consists of resident and infiltrating monocyte-derived cells, which develop into alveolar macrophages under the control of GM-CSF and transforming growth factor–β1 (TGF-β1) in the lung microenvironment (43, 44). Thus, although the BMDM and MDM used in this study may not entirely reflect airway macrophages, they likely represent a relevant model for infiltrating, inflammatory monocytes, which crucially contribute to allergic airway inflammation (27).

When treated with HpbE or Hpb GDH, monocytes and macrophages produced high amounts of regulatory prostanoids, which can modulate type-2 immune responses by activating a regulatory macrophage phenotype and limiting type-2 cytokine production (12, 45, 46). In contrast, eicosanoid pathways that typically drive type-2 inflammation (5-LOX/LTs and PTGDS/PGD2) were suppressed by treatment with HpbE or Hpb GDH. HpbE also triggered the production of TXB2, a mediator involved in platelet and vascular function as well as tissue repair. Thus, in addition to modulating type-2 inflammation, helminth-induced eicosanoids may limit bleeding and tissue damage. Given the broad eicosanoid-modulatory potential of HpbE and Hpb GDH, it appears unlikely that only one metabolite or receptor is responsible for their anti-inflammatory effects. Transfer of HpbE-conditioned wt, but not Ptges−/−, BMDM reduced eosinophilic airway inflammation as compared to transfer of untreated BMDM of the same genotype, suggesting that macrophage-derived PGE2 may contribute to HpbE-driven immunoregulation. In addition, the exaggerated allergic airway inflammation observed after transfer of HpbE-conditioned Ptgs2−/− BMDM supports a role for macrophage-derived COX metabolites, including PGE2, in the anti-inflammatory effects of HpbE. However, because we only obtained limited numbers of Ptgs2−/− BMDM and could thus not transfer untreated cells of this genotype, the strong inflammation observed after transfer of HpbE-conditioned Ptgs2−/− BMDM may also be due to COX-2 deficiency per se. Transfer of Ptges−/− BMDM led to reduced HDM-induced airway inflammation regardless of HpbE conditioning. Thus, when transferred into an inflammatory environment, mPGES-1/COX-2–deficient BMDM may respond differently as compared to wt BMDM. Of note, when transferred into the airways during HDM challenge, untreated wt BMDM tended to increase eosinophilic inflammation compared to HDM-sensitized mice without BMDM transfer, possibly due to HDM-induced chemokine production by the transferred macrophages (47). In combination with the COX-dependent reduction of human granulocyte chemotaxis, the BMDM transfer data support an important role for macrophage-derived COX metabolites in the modulation of granulocyte responses during type-2 inflammation.

To achieve therapeutic efficacy in allergy, asthma, or nasal polyposis, the modulation of multiple eicosanoid pathways (e.g., cysLTs, PGD2, and PGE2) may be favorable, as targeting single pathways has failed to provide a substantial clinical benefit (7, 48). Although the 5-LOX pathway represents a key drug target in multiple chronic diseases, there are currently no known drugs that work by simultaneously preventing LT generation and inducing anti-inflammatory eicosanoids. The regulatory eicosanoid switch triggered by HpbE and its active component Hpb GDH thus represents a key asset compared to current anti-inflammatory drugs.

Given that nematode larvae (Caenorhabditis elegans and Nippostrongylus brasiliensis) had been reported to induce LT production and eosinophil recruitment (16, 47), the LT-suppressive effects of HpbE were unexpected. However, our findings suggest that distinct helminth species and stages may differentially affect eicosanoid pathways and thus either promote or suppress type-2 inflammation. The regulated LT response induced by Hpb products may explain why Hpb is not rapidly expelled by a LT-dependent type-2 immune response as observed for N. brasiliensis (17).

To potentially harness the eicosanoid-modulatory effects of HpbE and Hpb GDH, we chose a topical administration in a model of HDM-induced allergic airway inflammation, where the central role of prostanoids and LTs is well documented (6, 8, 12). Intranasal treatment with HpbE or Hpb GDH attenuated hallmarks of HDM-induced type-2 inflammation. This was in line with previous studies, showing that Hpb infection or treatment with excretory secretory products from the Hpb adult stage (“HES”) can suppress allergic inflammation by modulating type-2 cytokine responses (21, 35, 49). However, in comparison to HES, HpbE and Hpb GDH showed a distinct potency to induce PGE2 and IL-10, which can act in concert to induce regulatory macrophages and suppress type-2 inflammation (12). HpbE appeared to preferentially target early, innate immune mechanisms (e.g., granulocyte recruitment), whereas HES can modulate adaptive type-2 immune responses, e.g., by inducing regulatory T cells (19). HpbE or Hpb GDH may thus be particularly suited to target eicosanoids and myeloid cells as crucial innate components of type-2 inflammation.

Recently, protein components of HES (HpARI and Hp-TGM) have been identified as major immunoregulatory factors (21, 50). Particularly, HpARI and the hookworm protein AIP-2 were shown to have potent allergy-suppressive effects (20, 21), but their effects on eicosanoids have not been studied. Using fractionation and MS, the enzyme Hpb GDH was identified as a major immunoregulatory candidate in HpbE. Recombinant Hpb GDH could recapitulate all major anti-inflammatory effects of total HpbE in vitro and in vivo, suggesting a potential use as a therapeutic enzyme. However, because Hpb GDH only shows 60% identity to human GDH (fig. S10), immunogenicity may represent a limitation to its clinical development as a new biotherapeutic. Thus, future studies should assess whether treatment with Hpb GDH results in the generation of neutralizing antibodies and design strategies to reduce the immunogenicity of the protein.

In addition, it will be important to define the mechanism of action by which Hpb GDH exerts its effects on eicosanoid pathways. Several PRRs (TLR2 and dectin-1/2) participated in the induction of PGE2 synthetic enzymes by HpbE. As TLR2 and dectins bind to carbohydrate structures, the full immunoregulatory potential of HpbE might rely on co-factors (e.g., carbohydrates or glycoproteins), which may act in concert with Hpb GDH or promote its activity. On the basis of its sequence, the natural, worm-derived Hpb GDH may also be O-glycosylated, which may result in recognition by C-type lectins and contribute to the immunoregulatory activities of HpbE. This would be in line with a recent study, showing dectin-dependent induction of PGE2 in dendritic cells by Schistosoma egg antigens (15). However, recombinant Hpb GDH from Escherichia coli, which is most likely not glycosylated, was also immunologically active, suggesting that glycosylation is not the major factor determining its immunoregulatory activity.

In keeping with anti-inflammatory roles for myeloid NF-κB and HIF-1α in airway inflammation (51, 52), both transcription factors contributed to the HpbE-triggered induction of type-2–suppressive mediators (PGE2, IL-10, and IL-1β). In addition, p38 MAPK, which mediated the induction of COX metabolites, had previously been implicated in the modulation of macrophage activation by a protein from a filarial nematode (AvCystatin) (53). However, if and how Hpb GDH may be involved in the activation of p38 MAPK, NF-κB, or HIF-1α pathways remains to be studied.

Last, because eicosanoids represent important mediators in severe and therapy-resistant type-2 inflammatory diseases (5, 54, 55), their broad anti-inflammatory modulation by HpbE may represent a promising future therapeutic approach. The identification of Hpb GDH as a major eicosanoid-modulatory component of HpbE provides a basis for the development of a new helminth-based therapeutic enzyme with a unique immunoregulatory profile.

MATERIALS AND METHODS

Study design

The aim of this study was to investigate if and how helminth products could modulate eicosanoid pathways to regulate type-2 inflammation. We characterized eicosanoid profiles of multiple myeloid cell types after treatment with helminth preparations by LC-MS/MS and selected the most promising candidates (HpbE and its active component Hpb GDH) for in vivo and ex vivo testing. To define in vivo effects of HpbE, Hpb GDH, or HpbE-conditioned macrophages, we treated mice intranasally with HpbE or recombinant Hpb GDH or transferred HpbE-conditioned macrophages during experimental HDM allergy and assessed type-2 inflammation by differential cell counts, histology, and multiplex cytokine analysis. All readouts were performed by a blinded experimenter. For the human part of our study, healthy volunteers (total n = 15) and AERD patients (n = 6) (Caucasian men and women) were recruited. Sample sizes, replicates, and statistical methods are specified in the figure legends. Primary data are reported in data file S1.

Mice

C57BL/6J mice were bred and maintained under specific pathogen–free conditions at the École Polytechnique Fédérale de Lausanne (EPFL) or at the Centre Hospitalier Universitaire Vaudois (CHUV). Alternatively, BALB/c and C57BL/6J mice were obtained from Charles River Laboratories (Sulzfeld, Germany). Unless stated otherwise, 6- to 12-week-old mice of both sexes were used. All animal experiments were approved by the local authorities (Swiss Veterinary Office or Regierung von Oberbayern).

Hpb infection and preparation of larval extract

Infective stage-three larvae (L3) of Hpb were obtained from the eggs of Hpb-infected mice as previously published (56). For preparation of Hpb larval extract (HpbE), L3 larvae were homogenized in two cycles at 6000 rpm for 60 s in a Precellys homogenizer using Precellys tough micro-organism lysing kits VK05 (Bertin Pharma). Remaining debris were removed by centrifugation (20 min, 14,000 rpm, 4°C). When indicated, heat-inactivated HpbE (HpbE 90°C) was prepared by heating at 90°C overnight.

HDM-induced allergic airway inflammation and intranasal treatment with HpbE or Hpb GDH

Eight-week-old female C57BL/6J mice were sensitized on day 0 by bilateral intranasal instillations of HDM extract from Dermatophagoides farinae [1 μg of extract in 20 μl of phosphate-buffered saline (PBS); Stallergenes SA] and challenged on days 8 to 11 with 10 μg of the same extract dissolved in 20 μl of PBS. Control animals received the same amount of PBS. HpbE treatment (5 μg of Hpb extract in 20 μl of PBS) was performed intranasally before sensitization and challenges, and Hpb GDH treatment (10 μg of protein in 20 μl of PBS) was performed intranasally before challenges. In the absence of HpbE or Hpb GDH treatment, the mice received 20 μl of PBS. Three days after the last challenge, the airways of the mice were lavaged five times with 0.8 ml of PBS. Aliquots of cell-free BALF were frozen immediately with or without equal volumes of methanol. Viability, yield, and differential cell count of BAL cells were performed as described before (57).

Human blood and tissue samples

PBMCs or PMNs were isolated from the blood of healthy human donors or patients with AERD. Nasal polyp tissues were obtained during polypectomy of patients suffering from chronic rhinosinusitis with nasal polyps. Nasal polyp secretions were obtained from cultured nasal polyp tissues as described previously (6). All blood and tissue donors participated in the study after informed written consent. Blood and tissue sampling and experiments including human blood cells were approved by the local ethics committee at the University clinic of the Technical University of Munich (internal reference: 422/16).

Macrophage cultures

MDMs or BMDMs were generated by culture in the presence of human or murine recombinant GM-CSF (10 ng/ml) (Miltenyi Biotec) and human recombinant TGF-β1 (2 ng/ml) (PeproTech) as previously described (6, 58). On day 6, cells were harvested and used for further experiments. More detailed procedures can be found in the Supplementary Materials.

Eicosanoid and cytokine analysis

Eicosanoids were quantified by LC-MS/MS similar to a previously published method (47). Cytokines were quantified using commercially available multiplex assays or enzyme-linked immunosorbent assay (ELISA) kits according to the manufacturer’s instructions. More detailed procedures can be found in the Supplementary Materials.

Chemotaxis assays

PMNs were resuspended to a concentration of 1 × 106 cells/ml in the presence of human GM-CSF (100 ng/ml; Miltenyi Biotec) and stimulated overnight with HpbE (10 μg/ml). When mentioned, PMNs were pretreated with 1 μM FP (Sigma-Aldrich), 10 μM MK (Cayman Chemical), or conditioned medium from MDM stimulated overnight with Hpb extract (10 μg/ml) ± 100 μM indomethacin for 1 hour. Nasal polyp secretions or a chemokine cocktail [RANTES (2 ng/ml), IL-8 (20 ng/ml; Miltenyi Biotec), and LTB4 (2 ng/ml; Cayman Chemical)] was placed in the lower wells of a chemotaxis plate (3 μm pore size; Corning). After mounting the transwell unit, 2 × 105 PMNs were added to the top of each well and migration was allowed for 3 hours at 37°C, 5% CO2. The number of cells migrating to the lower well was counted microscopically. In some experiments, manual counting was validated by flow cytometry.

Statistical analysis

All statistical analyses for biological data were performed using GraphPad Prism (GraphPad Software Inc.). Where two groups were compared, statistical significance was determined by Wilcoxon-Mann-Whitney test. Where more than two groups were compared, Kruskal-Wallis followed by Dunn’s multiple comparisons test, Friedman test (paired samples), or two-way analysis of variance (ANOVA) (unpaired samples) was used. Heat maps were generated using MetaboAnalyst version 3.0 (McGill University), a free online tool for metabolomics data analysis.

SUPPLEMENTARY MATERIALS

stm.sciencemag.org/cgi/content/full/12/540/eaay0605/DC1

Supplementary Materials and Methods

Fig. S1. PUFA metabolites produced by human MDM in response to treatment with HpbE and impact of HpbE-treated Ptgs2−/− BMDM on eosinophilic airway inflammation.

Fig. S2. HpbE has stronger eicosanoid-modulatory effects than glucocorticosteroids.

Fig. S3. HpbE has a distinct potential to induce type-2–suppressive mediators compared to other helminth products or helminth-associated bacteria.

Fig. S4. Viability and LTA4H expression in human eosinophils and neutrophils and PUFA metabolites produced by human PMNs in response to treatment with HpbE.

Fig. S5. HpbE induces a regulatory eicosanoid and cytokine profile in mixed and isolated CD14+ human PBMCs.

Fig. S6. Effect of COX-2, NF-κB, PI3K, PTEN, or PKA inhibition on HpbE-driven modulation of cytokines and eicosanoid pathways.

Fig. S7. Effect of neutralizing antibodies against PRRs (TLR2 and dectin-1/2) or IL-1β on HpbE-driven modulation of eicosanoids and IL-10 in human MDM.

Fig. S8. Bithionol does not affect cell viability and L3 stage HpbE shows a higher GDH activity as compared to L4 or L5 extracts.

Fig. S9. Newly generated monoclonal antibodies recognize Hpb GDH, clone 4F8 reduces HpbE-induced PGE2 and IL-10 production, and bithionol partially inhibits activity of recombinant Hpb GDH.

Fig. S10. Sequence of Hpb GDH is distinct from human GDH and contains potential predicted glycosylation sites.

Table S1. LC-MS/MS panel of PUFAs and PUFA metabolites.

Table S2. Proteins present in active fractions of HpbE identified by MS.

Table S3. Primer sequences for qPCR.

Table S4. Reagents and resources.

Data file S1. Primary data.

References (5963)

REFERENCES AND NOTES

Acknowledgments: We thank M. Haid and J. Adamski (Helmholtz Center Munich) for support with LC-MS/MS analysis. We also thank the animal caretakers of the University of Lausanne and Helmholtz Center Munich for animal husbandry. We further acknowledge technical assistance by M. Kulagin, J. Steinmetz, M. Korotkova, N. Dehne, F. Lauffer, M. Schiener, and J. Grösch. We also thank the staff of the monoclonal antibody facility at the Helmholtz Center Munich (R. Federle, A. Schepers, and A. Flatley) for generating and providing Hpb GDH–specific monoclonal antibodies. Funding: This study was supported by the German Research Foundation (DFG) with grants ES 471/2-1 and FOR 2599_A07 (ES 471/3-1), the Else Kröner-Fresenius-Stiftung (grant 2015_A195), the Fritz Thyssen Foundation (Az. 10.17.2.017MN), and a Helmholtz Young Investigator grant (VH-NG-1331) to J.E.-v.B. Work by D.T. was supported by the DFG (SFB1039). Author contributions: Conceptualization: J.E.-v.B. and C.B.S.-W.; methodology: M.d.l.R.J., A.L., F.A., S.B., D.T., B.B., R.N., M.H., P.-J.J., D.V., M.J.F., A.G., A.M., N.L.H., B.J.M., and J.E.-v.B.; investigation; M.d.l.R.J., A.L., F.A., S.B., S.S., P.H., F.H., W.A.N., A.T., C.P.d.C., A.M.C., M.H., C.O., and J.E.-v.B.; writing: M.d.l.R.J., A.L., and J.E.-v.B.; funding acquisition: J.E.-v.B. Competing interests: J.E.-v.B. and C.B.S.-W. have submitted a patent application (PCT/EP2019/058610) related to the immunoregulatory effects of HpbE and its components. C.B.S.-W. received grant support from Allergopharma, PLS Design, as well as Zeller AG and received speaker honoraria from Allergopharma. A.M.C. reports to have given lectures sponsored by Allergopharma, ALK, and GSK and received grant support or consultant arrangements via Technische Universität München with Allergopharma, ALK, HAL-Allergy, Mundipharma, Lofarma, Zeller AG, and Novartis. Data and materials availability: All data associated with this study are present in the paper or the Supplementary Materials.

Stay Connected to Science Translational Medicine

Navigate This Article