Myeloid-Restricted AMPKα1 Promotes Host Immunity and Protects against IL-12/23p40–Dependent Lung Injury during Hookworm Infection

Wildaliz Nieves, Li-Yin Hung, Taylor K. Oniskey, Louis Boon, Marc Foretz, Benoit Viollet and De’Broski R. Herbert
J Immunol June 1, 2016, 196 (11) 4632-4640; DOI: https://doi.org/10.4049/jimmunol.1502218

Abstract

How the metabolic demand of parasitism affects immune-mediated resistance is poorly understood. Immunity against parasitic helminths requires M2 cells and IL-13, secreted by CD4+ Th2 and group 2 innate lymphoid cells (ILC2), but whether certain metabolic enzymes control disease outcome has not been addressed. This study demonstrates that AMP-activated protein kinase (AMPK), a key driver of cellular energy, regulates type 2 immunity and restricts lung injury following hookworm infection. Mice with a selective deficiency in the AMPK catalytic α1 subunit in alveolar macrophages and conventional dendritic cells produced less IL-13 and CCL17 and had impaired expansion of ILC2 in damaged lung tissue compared with wild-type controls. Defective type 2 responses were marked by increased intestinal worm burdens, exacerbated lung injury, and increased production of IL-12/23p40, which, when neutralized, restored IL-13 production and improved lung recovery. Taken together, these data indicate that defective AMPK activity in myeloid cells negatively impacts type 2 responses through increased IL-12/23p40 production. These data support an emerging concept that myeloid cells and ILC2 can coordinately regulate tissue damage at mucosal sites through mechanisms dependent on metabolic enzyme function.

Introduction

The AMP-activated protein kinase (AMPK) complex (αβγ), comprised of an α catalytic subunit with serine/threonine kinase activity, β glycogen–binding domain, and γ adenylate–binding region, coordinates glucose metabolism with fatty acid oxidation to regulate energy availability (1, 2). AMPK may regulate genes in muscle and adipose tissue by phosphorylation of specific transcription factors involved in glucose metabolism, mitochondrial biogenesis, or fatty acid oxidation (3). However, the roles served by AMPK in the myeloid lineage, particularly during pathogen-specific immune responses, are not well understood.

Mononuclear phagocytes (macrophages and dendritic cells [DCs]) can serve influential roles in the development of type 2 immunity against parasitic helminths (4). Humans infected with hookworms present with symptoms of dysregulated metabolic homeostasis, including anemia, fatigue, malnutrition, and impaired T cell responses (5), any of which could be due to altered functions of myeloid phagocytes. Infection-associated immunosuppression due to worms is closely associated with IL-4/IL-13–driven M2c polarization characterized by increased production of Ym-1 (chitinase-like protein 3) and resistin-like molecule-α (6). AMPK promotes an M1 to M2 conversion, as demonstrated in mouse models of muscle tissue damage, multiple sclerosis, and LPS-induced inflammation (7–9), but whether myeloid-intrinsic AMPK controls the outcome of type 2 responses during infection has not been addressed.

Similar to many helminth species, hookworms cause transient, alveolar lung damage (distal lung space) within the host (10). Nippostrongylus brasiliensis, a rodent hookworm, causes hemorrhagic lung injury within the first 3 days postinfection (dpi), which causes the release of cytokines such as IL-33 to initiate type 2 immunity. IL-33–driven host resistance relies upon IL-13 secreted by CD4+ Th2 and group 2 innate lymphoid cells (ILC2) for resistin-like molecule-β–dependent killing (11–13). Although macrophage depletion exacerbates Th17-associated inflammation and lung damage (14), the question of whether metabolic control of M2 cells promotes lung tissue repair has not been answered. In this study, we demonstrate that AMPKα1 is required for CD11c-expressing myeloid cells to promote ILC2 expansion, type 2 immune resistance, and lung tissue repair during N. brasiliensis infection. AMPK deficiency in macrophages and conventional DCs (cDCs) impaired host protection, marked by increased numbers of adult worms and fecal egg numbers, reduced IL-13 cytokine secretion, and reduced ILC2 expansion in an IL-12/23p40–dependent manner. Thus, our data reveal a previously unrecognized role for myeloid-restricted AMPK as an important driver of type 2 responses involving ILC2 expansion that precedes lung tissue repair and expulsion of gastrointestinal nematodes from the intestinal tract.

Materials and Methods

Ethics statement

Animal experiments were conducted in accordance with University of California at San Francisco Institutional Animal Care and Use Committee protocol AN088478-03 (PI-DRH) and Public Health Service assurance no. A3400-01. These procedures were approved under University of California at San Francisco administrative policy no. 100-17 (Research and Instruction Using Animal Subjects).

Animals and parasites

CD11cCre mice (The Jackson Laboratory, no. 008068) were bred with previously described AMPKflox/flox mice on a C57BL/6 background (7). Either CD11cCre or AMPKα1flox/flox littermates were used for wild-type (WT) controls. Age-matched mice between 6 and 12 wk old were used and housed at the San Francisco General Hospital and Trauma Center vivarium. N. brasiliensis life cycle and techniques for s.c. inoculation with 750 N. brasiliensis third-stage larvae, quantitation of worm burdens, and fecal eggs has been described previously (11).

ELISA

A pAMPKα (T172)–specific ELISA kit (Invitrogen) was used to quantify pAMPK levels in lung tissue homogenates following the manufacturer’s instructions. pAMPK levels were normalized to the total protein content of the sample as determined by a BCA assay (Pierce). Cytokine ELISAs IL-13 (eBioscience) and Ym-1 (R&D Systems) were carried out per the manufacturers’ protocols.

Abs and flow cytometry

To determine the frequency of pAMPKα+ cells, whole lungs were processed as previously described (12). To obtain a single suspension of whole splenocytes, tissue was passed through a 100-μm sieve prior to lysis of RBCs. Once a single-cell suspension was achieved, cells were stained with the cell viability dye eFluor 506 (eBioscience) prior to myeloid cell surface staining using CD11c (N418, BioLegend), Siglec-F (3D6.112, BioLegend), MHC class II (MHC-II; M5-114.15.2, eBioscience), CD103 (2E7, eBioscience), and CD11b (M170, Tonbo Biosciences) or T cell staining CD62L (MEL-14, Tonbo Biosciences), CD44 (IM7, BioLegend), CD4 (GK1.5, eBioscience), and CD8a (5H10, Invitrogen). Cells were stained intracellularly per the manufacturer’s protocol using rabbit anti-mouse pAMPKα (Y365) mAb (Abcam) or isotype control rabbit anti-mouse IgG primary Ab and anti-rabbit allophycocyanin secondary Ab.

To determine the frequency of IL-13–producing CD4+ T cells, mice were administered 0.5 mg/ml brefeldin A (Sigma-Aldrich) i.p. 5–6 h prior to processing of whole splenocytes to obtain a single-cell suspension. Cells were stained with the cell viability dye eFluor 506 (eBioscience) prior to T cell surface staining as described above, including B220 (RA36B2, eBioscience), F4/80 (BM8, eBioscience), CD11b (M1/70, eBioscience), and CD11c (N418, eBioscience) lineage-negative markers. Following fixation and permeabilization using the BD Cytofix/Cytoperm kit (BD Biosciences), intracellular cytokine staining of IL-13 (I3A, eBioscience) was performed overnight at 4°C.

To determine the percentage and total lung ILC2, whole lungs were processed to obtain a single-cell suspension, and live cells were quantified using a Guava cell counter (EMD Millipore). A lineage-negative cell isolation kit was used (Miltenyi Biotec) (anti-CD4, -CD11c, -B220, -NK1.1) to enrich for ILC2 by magnetic bead–based purification (negative selection) followed by enumeration of the live cells in the flowthrough. This was used as the total lung cell number for calculation of ILC2 percentage. Cells were incubated with Fc Block (clone 2.4G2) prior to surface staining with a lineage-specific Ab mixture: anti-B220 (RA3-6B2), -CD4 (GK1.5), -CD8a (53-6.7), -CD11b (M1/70), -CD11c (N418), -Gr1 (RB6-8C5), and -CD49b (DX5). Lineage-negative cells were further analyzed using fluorescently conjugated mAbs specific for anti-Thy1.2 (30-H12), -CD45 (30-F11), –IL-7Rβ (A7R34), and -GATA3 (TWAJ) to identify the ILC2 population. An LSR II (BD Biosciences) flow cytometer was used to acquire cells. Data were analyzed using FlowJo (Tree Star) software.

Neutralization of IL-12/23p40 was achieved by administering 1 mg anti–IL-12/23p40 mAb (C17.8) or isotype control (GL117) i.p. every 48 h for 9 d.

Histology

Perfused formalin-fixed, paraffin-embedded whole-lung tissue cross-sections were stained with H&E. Composite stitched images (×5 magnification) were generated using LAS v4.3 imaging acquisition software (Leica Microsystems) to visualize an entire lobe.

Pulse oximetry

Blood oxygen levels (percentage of blood oxygen saturation [% SpO2]) were measured using MouseOx Plus with a small CollarClip (Starr Life Sciences). Mouse hair on the sensor site was removed 1 d before oximetry per the manufacturer’s recommendations. On the day of measuring, mice were anesthetized and 1 min of oximetry was recorded and analyzed. Results were obtained as % SpO2, which was a readout from MouseOx Plus. The mean ± SE % SpO2 was obtained per time point for each mouse and the change in % SpO2 was calculated by subtracting from baseline values.

Real-time PCR analysis

Total RNA was harvested from tissues using an RNeasy mini kit (Qiagen) following the manufacturer’s protocol. Total RNA (500 ng) was reverse transcribed using SuperScript II (Invitrogen) following the manufacturer’s protocol. Diluted cDNA samples were added to SsoAdvanced SYBR Green supermix (Bio-Rad), and RT-PCR reactions were run on a CFX96 RT-PCR detection system (Bio-Rad). Primer sequences are listed in Supplemental Table I. Gene expression is normalized to Gapdh, and data are presented as means ± SEM from the replicates.

Statistical analysis

The statistical significance of differences between two or more groups was determined by an unpaired Student t test, the Mann–Whitney U test, one-way ANOVA, or two-way ANOVA. A p value <0.05 was considered significant. All analyses were performed using GraphPad Prism 5.0 software.

Results

Myeloid cell-specific AMPKα1 activation exacerbates hookworm-mediated lung injury

Using pulse oximetry, we established a noninvasive method for tracking the change in % SpO2 during the course of hookworm infection. WT mice infected with 750 N. brasiliensis larvae had a 30% reduction in % SpO2 by 3 dpi, but % SpO2 levels rebounded to baseline within 9–12 dpi (Fig. 1A). To investigate whether parasitized tissue underwent any change in AMPK activity, experiments were designed to compare whole-lung AMPK activity between naive and N. brasiliensis–infected WT mice. Results from a phospho-AMPKα–specific ELISA revealed a 2-fold increase in AMPK activation within whole-lung tissue of WT mice at 3 dpi (Fig. 1B). Increased AMPK activation corresponded to the nadir in % SpO2 assessed via pulse oximetry. Next, to determine whether myeloid cells were a functionally important source of AMPK, mice with a conditional defect of AMPKα1 in alveolar macrophages and cDCs (CD11cAMPK) were generated following an intercross between CD11cCre and AMPKα1flox/flox mouse strains and compared with WT controls (CD11cCre).

FIGURE 1.
FIGURE 1.

Hookworm infection increases lung injury and AMPKα activation. (A) Percentage SpO2 of N. brasiliensis–infected WT mice. (B) Data show pAMPKα levels in whole-lung tissue lysates from WT C57BL/6 mice left untreated (naive) or at 3 dpi with 750 N. brasiliensis third-stage larvae. Data represent two independent experiments (n = 3 per group) (±SEM). **p < 0.01.

To confirm CD11cCre specificity, intracellular flow cytometry was used to identify the AMPKα-deleted cell populations. As expected, staining with pAMPKα-specific mAb in naive alveolar macrophages revealed a 3-fold reduction in mean fluorescence intensity (MFI) compared with WT (Fig. 2A, 2B). Similarly, following N. brasiliensis infection, alveolar macrophages, but not eosinophils, showed a 2- to 3-fold reduction in pAMPKα MFI when isolated from CD11cAMPK mice compared with WT mice (Fig. 2C–F). The CD11cAMPK strain also deleted AMPKα protein in MHC-IIhiCD11b+ DCs, but not in CD103+ DCs or interstitial macrophages (Fig. 2G–J). Next, we quantified AMPK activation within whole-lung tissue of CD11cAMPK mice following N. brasiliensis infection. Curiously, the infection induced increase in AMPK activity was unabated in lung tissues of the CD11cAMPK strain (Fig. 2K), which was most likely due to the compensatory increase in pAMPKα MFI levels within CD4+ or CD8+ T lymphocytes, a feature noted both at baseline and during infection (Fig. 2L–R). CD11cAMPK mice displayed no defect in ILC2 expression of AMPKα1 (data not shown). We postulated that although overall tissue AMPK levels were not reduced in CD11cAMPK mice, selective AMPK deficiency in the myeloid compartment could limit the nature of the immune response against worm infection and potentially impact the rate of lung recovery following N. brasiliensis infection.

FIGURE 2.
FIGURE 2.

Characterization of the AMPK activation status in CD11cAMPK mouse-derived pulmonary leukocytes during hookworm infection. (A–J) AMPKα-specific intracellular staining to compare CD11cAMPK (dotted line) and CD11cCre (gray histogram) for MFI expression levels in and alveolar macrophages of naive mice (A and B). (C–E) MFI levels for AMPKα intracellular cytokine staining at 9 dpi in autofluorescent alveolar macrophages and (F) nonautofluorescent eosinophils (n = 4 per group). (G–J) MFI expression levels in MHC-II+, (H) CD103+ DCs, (I) interstitial macrophages (Mϕ), and (J) CD11b DCs (n = 4). (K) Data show pAMPKα levels in whole-lung tissue lysates from WT C57BL/6 mice left untreated (naive) or at 3 dpi with 750 N. brasiliensis third-stage larvae (n = 3 per group). (L) Contour plot shows gating strategy for splenic CD4+ and CD8+ T cell gating. (M–P) Representative MFI histograms from two independent experiments (n = 4 per group) show (M and N) CD4+ T cells and (O and P) CD8+ T cells from naive (dotted line) versus 9 d N. brasiliensis–infected (red line) CD11cCre and CD11cAMPK mice. Means ± SEM of MFI levels of AMPKα staining in gated (Q) CD4+ T cells and (R) CD8+ T cells from four mice per group are shown. All data represent two to three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001.

Consistent with our hypothesis, although there were no baseline defects in % SpO2 levels among naive mice of control or AMPK mutants (Fig. 3A), we found that CD11c-restricted AMPKα1 deletion had a striking impact upon the rebound of % SpO2 levels following hookworm-induced lung injury (Fig. 3B). The % SpO2 nadir caused by N. brasiliensis was significantly greater in CD11cAMPK mice at 3 dpi compared with WT mice, and also the late-phase rebound in lung function at day 8 was significantly impaired in the former compared with the latter (Fig. 3B). Comparative analysis of lung histological tissue sections from naive versus N. brasiliensis–inoculated mice at 9 dpi revealed numerous denuded tissue areas with large distended alveoli, whereas WT lung tissues had normal compact alveolar structure that was consistent with resolution of tissue damage (Fig. 3C–G). Myeloid-restricted AMPK also changed the inflammatory response, as determined by differential cell staining of bronchoalveolar lavage (BAL) fluid recovered from WT and CD11cAMPK mice at 9 dpi. Data show that although both strains had similar numbers of macrophages, there was a reduction in eosinophils and an increase in neutrophils within infected CD11cAMPK mice compared with their WT counterparts (Supplemental Fig. 1A). Thus, loss of AMPKα1 activity in myeloid cells exacerbated lung tissue injury and skewed the balance of inflammatory leukocytes that entered the lung at baseline and following hookworm infection.

FIGURE 3.
FIGURE 3.

Deletion of AMPKα in myeloid cells exacerbates lung injury. (A) Raw % SpO2 of naive WT (CD11cCre and AMPKflox/flox) and CD11cAMPK mice (n = 3–10). Data are representative of three independent experiments. (B) Change in % SpO2 from baseline levels at 3 and 8 dpi (n = 6–10). (C and D) H&E-stained left lobe of naive or (E and F) 9 d infected lung tissue. Arrowheads indicate areas of tissue injury. Original magnification ×5. (G) H&E-stained histological cross-sections of the left lobe from naive or 9 d infected WT and CD11cAMPK mice. Arrowheads indicate areas of tissue injury and distended alveoli. Original magnification ×20. Scale bars, 100 mm. Data represent two to four experiments (±SEM). *p < 0.05, **p < 0.01.

Type 2 immunity requires AMPKα1 expression in myeloid APCs

AMPK deficiency has been shown to increase Ag-specific Th1/Th17 responses (9, 15, 16). Consistent with these reports, we found that helminth Ag-specific recall and T cell mitogen responses were marked by highly elevated IFN-γ production in CD11cAMPK mice compared with WT mice when assessed at 9 dpi (Supplemental Fig. 1B). However, it has not been investigated whether AMPK serves as a driver of infection-induced type 2 responses, and because type 2 cytokine production and M2 macrophage development have been proposed to drive tissue repair, we asked whether the CD11cAMPK strain had any defects in the production of canonical type 2 cytokines and/or M2-associated molecules.

N. brasiliensis infection–induced release of Ym-1 (chitinase 3–like 3) in BAL fluid (Fig. 4A) and lung mRNA transcript levels for Retnla (Fig. 4B) were significantly reduced in CD11cAMPK versus the WT strain. Ag-induced IL-13 production from whole-splenocyte cultures was also significantly reduced in CD11cAMPK compared with WT mice (Fig. 4C). Lung Il4 and Ccl17 mRNA transcripts were significantly reduced in CD11cAMPK mice compared with WT mice at 9 dpi (Fig. 4D, 4E).

FIGURE 4.
FIGURE 4.

Myeloid pAMPKα1 deficiency impairs M2 responses and hookworm-induced IL-13 production. Comparison of naive and 9 d infected CD11cCre and CD11cAMPK mice for (A) Ym-1 protein levels in BAL fluid, (B) lung mRNA transcript levels for Retnla. Fold expression of infected over respective naive controls. (C) IL-13 levels in splenocyte culture supernatants following stimulation with crude N. brasiliensis Ag extract (N. brasiliensis Ag) or left untreated (control). (D) Fold change of lung il4 and (E) Ccl17 mRNA transcript levels. Fold expression of infected over respective naive controls. Data show means ± SEM and represent three independent experiments with three to eight mice per group. *p < 0.05, **p < 0.01, ***p < 0.001. N.b., N. brasiliensis.

Next, we asked whether the CD11cAMPK strain had any defect in the accumulation of ILC2, a subset of cytokine-secreting innate lymphoid cells that promote the development of host-protective type 2 immunity (17–19). Identification of ILC2 (defined as lineage-negative CD45+IL-7R+Thy1.2+IL-17Rβ+GATA3+) in whole-lung tissue digests prepared from naive (Fig. 5A) and infected (Fig. 5B) WT and CD11cAMPK strains revealed that CD11cAMPK mice had a significant defect in lung ILC2 expansion by day 3 (Fig. 5C), whereas there were no differences between groups at baseline. Defective ILC2 expansion was corroborated by fewer Areg lung tissue mRNA transcripts, an epidermal growth factor family cytokine (20) (Fig. 5D). Double-immunofluorescence staining for Ym-1 and CRTH2 (GPR44) on hookworm-damaged lung tissue was used to determine whether the reduced numbers of ILC2 correlated with any alteration in the potential association between M2 and ILC2 in situ. Consistent with this hypothesis, data indicated that whereas many foci of Ym-1/CRTH2 clusters were observed in WT mice, these interactions were less apparent within lung tissues of CD11cAMPK mice and their respective naive controls (Supplemental Fig. 2).

FIGURE 5.
FIGURE 5.

Myeloid pAMPKα1 deficiency impairs hookworm-induced ILC2 expansion. (A) Flow cytometry dot plot shows the percentage of lung ILC2 within lung tissue digests from CD11cCre (top) and CD11cAMPK (bottom) naive and (B) infected mice 3 dpi. Percentage of lung population indicated on plot. (C) Total ILC2 cell number per lung of naive and infected mice. (D) Lung mRNA transcript levels for Areg. Fold expression of infected over respective naive controls. (E) Fecal egg counts per gram of tissue 6, 7, and 8 dpi and (F) number of adult intestinal worms 9 dpi with N. brasiliensis. Data show mean ± SEM and represent three independent experiments with three to five mice per group. *p < 0.05, **p < 0.01.

Next, parasitological analyses were conducted to assess the functional impact of the defective type 2 immune responses on host protection against hookworm infection. Evaluation of CD11cAMPK and WT mice at 9 dpi revealed that CD11cAMPK mice had significantly higher hookworm egg and adult intestinal worm burdens compared with WT controls (Fig. 5E, 5F). Given that CD4+ T cells displayed abnormally elevated AMPK levels in CD11cAMPK mice and that host immunity is CD4 dependent (21), we asked whether this T cell subset could be partially responsible for the impaired host-protective phenotype of the CD11cAMPK strain. Data show that RAG1−/− mice passively transferred with CD4+ T cells purified from WT or CD11cAMPK strains and infected with N. brasiliensis showed an equivalent ability to reduce worm and egg burdens (Supplemental Fig. 3A and 3B, respectively) and to promote a rebound in pulmonary function (Supplemental Fig. 3C) as compared with nontransferred infected RAG1−/− mice. Thus, defective M2 and ILC2-associated responses in CD11cAMPK mice were likely due to myeloid-intrinsic defects in AMPK activation following hookworm infection.

AMPKα1 suppresses IL-12–dependent tissue injury and drives M2 polarization following hookworm infection

Immunity against N. brasiliensis is propelled by IL-4Rα–dependent type 2 immunity but is strongly suppressed by IL-12p40 (22). We reasoned that CD11cAMPK mice would show defaulted IL-12p40–associated inflammation. To investigate, magnetically sorted infected lung tissue CD11c+ cell mRNA transcripts were analyzed for IL-12/23p40 expression levels. Compared to WT animals, infected CD11cAMPK mice expressed 3-fold greater levels of Il12b cytokine transcript (Fig. 6A), indicating an intrinsic defect due to AMPK deficiency. Anti–IL-12/23p40 neutralizing Ab (C17.8) was used to assess whether type 2 immunity could be restored by neutralization of IL-12/23p40–associated inflammatory responses. Compared to isotype-matched control mAb (GL117), anti–IL-12/23p40 Ab administration increased the percentage of IL-13 cytokine-secreting splenic CD4+CD62LloCD44hi cells (Fig. 6B, 6C). Anti–IL-12/23p40 Ab administration also significantly increased lung Areg expression (Fig. 6D) and reversed the histological features of hookworm-induced lung injury (Fig. 6E–H). Lastly, IL-12/23p40 neutralization increased the % SpO2 in infected CD11cAMPK mice compared with those treated with GL117 control (Fig. 6I). These findings demonstrated that myeloid-specific deletion of AMPK led to an IL-12–dependent suppression of IL-13–producing CD4+ T cells and ILC2 expansion.

FIGURE 6.
FIGURE 6.

Neutralization of IL-12/23p40 in CD11cAMPK mice restores type 2 responses and lung repair. (A) Comparison of whole-tissue mRNA transcript levels for Il12b from sorted pulmonary CD11c+ cells between CD11cCre and CD11cAMPK mice at 9 dpi with N. brasiliensis. (B) Representative dot plots showing the percentage of CD4+CD62loCD44hiIL-13+ cells within the spleen of naïve mice (top) or at 9 dpi following IgG2a isotype mAb (middle) or anti–IL-12/23p40 mAb treatment (bottom) mice. (C) Dot plot showing the percentage of CD4+CD62loCD44hiIL-13+ cells within the spleen of naive mice (white) or at 9 dpi following isotype mAb (gray) or anti–IL-12/23p40 mAb treatment (charcoal) mice. (D) Areg mRNA transcript of lung tissue from CD11ccre (solid) and CD11cAMPK (hatched) mice 9 dpi with N. brasiliensis third-stage larvae. Graph indicates relative expression over the Gapdh reference gene. H&E-stained cross-sections of left lobe of lung (original magnification ×200) from (E and F) isotype mAb or (G and H) anti–IL-12/23p40 mAb treatment are shown. Arrowheads indicate areas of lung injury. (I) Change in % SpO2 at 8 dpi following treatment with three doses of 1 mg anti–IL-12/23p40 mAb or IgG2a isotype mAb. Data represent two independent experiments with three to five mice per group. *p < 0.05.

Discussion

Although it is well established that metabolic status regulates tissue homeostasis at the organ level, it is only recently becoming recognized that metabolic enzymes can direct leukocyte function in a cell lineage–specific manner (15, 23–26). Our work demonstrates that AMPK promotes M2 development and restrains myeloid cell-driven IL-12 responses during infectious tissue injury, which has a vitally important role in promoting functional organ recovery following hookworm infection. AMPK deletion in alveolar macrophages and cDCs reduced production of canonical M2 molecules, Th2 cytokine production, and ILC2 expansion. Curiously, this phenotype was marked by reduced Ccl17 production and fewer interactions between M2 and ILC2, which may imply a regulatory role for myeloid AMPK in producing chemokines required for ILC2 recruitment and/or expansion. Moreover, CD11c-specific AMPK deficiency led to increased production of IL-12/23p40, a Th1/Th17-inducing cytokine, which, in turn, limited IL-13 production from CD4+ T cells and reduced ILC2-associated gene expression. Taken together, these data support the emerging concept that myeloid cells can regulate ILC2 and that, mechanistically, myeloid cells use AMPK to promote type 2 responses that are essential for parasite clearance and tissue repair following hookworm infection.

Following skin penetration, the lung is the first major organ system damaged by parasitic worms, which in some cases leads to acute respiratory distress (27). N. brasiliensis rapidly migrates into the distal lung compartment within hours of infection where larvae destroy normal alveolar structure (10). AMPKα activity was upregulated within 3 d, corresponding to peak hemorrhagic injury that precedes type 2 cytokine production (14). Although worm infection did not increase pAMPK MFI levels in alveolar macrophages on a per cell basis, increased myeloid cell accumulation most likely contributed to increased AMPK levels in tissues. However, CD11c-mediated deletion led to an increase in AMPK levels, most likely because pAMPK levels were highly induced within CD4+ and CD8+ T cells by day 9, consistent with increasing energy demands during effector T cell development (28). Nonetheless, loss of AMPKα within alveolar macrophages and cDCs abrogated the spontaneous rebound in lung function observed between 7 and 9 dpi with N. brasiliensis (13). As compared with WT controls, % SpO2 levels were lower in CD11cAMPK mice by day 3. It is unlikely that the differences in % SpO2 levels between strains were due solely to inflammatory leukocyte infiltration, because peak type 2 inflammation and eosinophilia normally occur at day 9, when % SpO2 levels are no different from baseline. Instead, % SpO2 differences were likely due to defects in macrophage-dependent repair of pulmonary gas exchange. Histological comparison of hookworm-injured lungs revealed that CD11cAMPK mice had large distended alveoli, in contrast to WT lung pathology. However, it remains unclear whether macrophages can promote epithelial repair or only suppress cytokine release associated with the M1 pathway. The protective efficacy of anti–IL-12/23p40 treatment suggests the latter, and evidence that IFN-γ, a canonical type 1 cytokine, can block type 2 effector lymphocytes such as ILC2 (29) suggests that a predominate M1 cell response in situ may impair innate type 2 responses.

The M2-associated molecules Ym-1 and Relm-α were produced in an AMPKα1p-dependent manner. However, it remains unclear whether AMPK also controls specific expression of molecules such as PD-L2 and Aldha1/2, which can distinguish between M2 cells derived from different sources (30). AMPK may directly favor M2 development through regulation of PPARγ coactivators 1 α and β, which are induced by IL-4/13 and support metabolic reprogramming associated with M2 polarization (31). Perhaps AMPK may function downstream of IL-4 to promote Arg1 transcription, but this remains unknown. Whether there are discrete mediators released from M2 cells that promote tissue regeneration remains unknown and is the topic of future study (6, 32–35). Thus, although M2 cells may have secreted epithelial regenerative molecules, these factors have not been identified. Future work is needed to discern whether M2 cells derived from different sources or whether other macrophage phenotypes such as IL-10 or TGF-β–secreting cells or even N2 cells can elaborate selective mediators of epithelial cell regeneration.

Even though pharmacological activation of AMPK suppresses LPS-induced acute lung injury (36), the explanations for these effects were previously lacking. Our data are consistent with evidence that myeloid-specific AMPK drives IL-10/STAT-3 and PI3K/Akt/mTORC1-mediated suppression of proinflammatory cytokine production and CD40 expression (9, 15, 37). We demonstrate that neutralization of IL-12/23p40 (essential subunit for IL-12p70 and IL-23) partially reversed susceptibility of CD11cAMPK mice to severe lung injury and impaired respiratory function. This is consistent with evidence that IL-12 administration promotes susceptibility to N. brasiliensis (22). Lack of myeloid-derived AMPKα1 also impaired infection-induced ILC2 expansion and lung Areg expression, which was reversed by IL-12/23p40 neutralization, suggesting that myeloid cells can regulate ILC expansion owing to the cytokines they produce. Indeed, defective Ccl17 expression and defective ILC2 responses are similar to findings in a recent report in the context of allergic lung inflammation, where a coordinately regulated mechanism among Th2, ILC2, and M2 drives the memory type 2 response (38). Our demonstration that M2 and ILC2 are intimately associated in hookworm-damaged lungs is consistent with the view that these cells may modulate each other’s activity. Indeed, increased IL-12 in parasitized CD11cAMPK mice, accompanied by defective ILC2 responses, is consistent with data showing that type 1 inflammatory responses dominated by IFN-γ block ILC2 function (39). It will be important to determine whether myeloid-derived IL-12/23p40 directly antagonizes ILC2 expansion, as Areg expression levels increased following anti–IL-12/23 p40 mAb administration. Similarly, CD4+ T cell–derived IL-13 was restored upon IL-12/23p40 neutralization in CD11cAMPK mice. Thus, we extend findings that cDC depletion impairs Th2 development (4) by revealing AMPK as an intrinsic determinant shaping the ability of myeloid APCs to promote type 2 immunity.

Macrophages in the pulmonary compartment have dynamic roles in homeostasis, defense, and repair, perhaps due to their unique lipid-rich environment (40). Indeed, lipolysis drives M2 polarization, a function that may occur through AMPK-dependent oxidative phosphorylation (25). Inclusive with studies demonstrating that IL-4 cytokine promotes type 2 responses, including proliferation and activation of type 2 macrophages during Litomosoides sigmodontis infection (41), we propose that myeloid-derived AMPK contributes to the IL-4–driven pathway that promotes M2 polarization in the context of hookworm infection and lung injury.

Disclosures

The authors have no financial conflicts of interest.

Acknowledgments

We thank Dr. Jill Suttles for critical reading of this manuscript and members of the Herbert laboratory for technical support.

Footnotes

  • This work was supported by National Institutes of Health Grants R01 A1095289 and R01 GM83204, Burroughs Wellcome Fund Award CA-0062619 (to D.R.H.), as well as by National Institutes of Health Immunology Research Training Grant 2T32AI007334-16 and Diversity Supplement Award (to W.N.).

  • The online version of this article contains supplemental material.

  • Abbreviations used in this article:

    AMPK
    AMP-activated protein kinase
    BAL
    bronchoalveolar lavage
    cDC
    conventional DC
    CD11cAMPK mice
    intercross between CD11cCre and AMPKα1flox/flox mice
    DC
    dendritic cell
    dpi
    day postinfection
    ILC2
    group 2 innate lymphoid cell
    MHC-II
    MHC class II
    MFI
    mean fluorescence intensity
    % SpO2
    percentage of blood oxygen saturation
    WT
    wild-type.

  • Received October 14, 2015.
  • Accepted March 18, 2016.

References

  1. ↵
    1. Xiao B.,
    2. M. J. Sanders,
    3. E. Underwood,
    4. R. Heath,
    5. F. V. Mayer,
    6. D. Carmena,
    7. C. Jing,
    8. P. A. Walker,
    9. J. F. Eccleston,
    10. L. F. Haire,
    11. et al
    . 2011. Structure of mammalian AMPK and its regulation by ADP. Nature 472: 230–233.
    OpenUrlCrossRefPubMed
  2. ↵
    1. Towler M. C.,
    2. D. G. Hardie
    . 2007. AMP-activated protein kinase in metabolic control and insulin signaling. Circ. Res. 100: 328–341.
    OpenUrlAbstract/FREE Full Text
  3. ↵
    1. Jäger S.,
    2. C. Handschin,
    3. J. St-Pierre,
    4. B. M. Spiegelman
    . 2007. AMP-activated protein kinase (AMPK) action in skeletal muscle via direct phosphorylation of PGC-1α. Proc. Natl. Acad. Sci. USA 104: 12017–12022.
    OpenUrlAbstract/FREE Full Text
  4. ↵
    1. Phythian-Adams A. T.,
    2. P. C. Cook,
    3. R. J. Lundie,
    4. L. H. Jones,
    5. K. A. Smith,
    6. T. A. Barr,
    7. K. Hochweller,
    8. S. M. Anderton,
    9. G. J. Hämmerling,
    10. R. M. Maizels,
    11. A. S. MacDonald
    . 2010. CD11c depletion severely disrupts Th2 induction and development in vivo. J. Exp. Med. 207: 2089–2096.
    OpenUrlAbstract/FREE Full Text
  5. ↵
    1. Hotez P. J.,
    2. P. J. Brindley,
    3. J. M. Bethony,
    4. C. H. King,
    5. E. J. Pearce,
    6. J. Jacobson
    . 2008. Helminth infections: the great neglected tropical diseases. J. Clin. Invest. 118: 1311–1321.
    OpenUrlCrossRefPubMed
  6. ↵
    1. Van Dyken S. J.,
    2. R. M. Locksley
    . 2013. Interleukin-4- and interleukin-13-mediated alternatively activated macrophages: roles in homeostasis and disease. Annu. Rev. Immunol. 31: 317–343.
    OpenUrlCrossRefPubMed
  7. ↵
    1. Mounier R.,
    2. M. Théret,
    3. L. Arnold,
    4. S. Cuvellier,
    5. L. Bultot,
    6. O. Göransson,
    7. N. Sanz,
    8. A. Ferry,
    9. K. Sakamoto,
    10. M. Foretz,
    11. et al
    . 2013. AMPKα1 regulates macrophage skewing at the time of resolution of inflammation during skeletal muscle regeneration. Cell Metab. 18: 251–264.
    OpenUrlCrossRefPubMed
    1. Meares G. P.,
    2. H. Qin,
    3. Y. Liu,
    4. A. T. Holdbrooks,
    5. E. N. Benveniste
    . 2013. AMP-activated protein kinase restricts IFN-γ signaling. J. Immunol. 190: 372–380.
    OpenUrlAbstract/FREE Full Text
  9. ↵
    1. Sag D.,
    2. D. Carling,
    3. R. D. Stout,
    4. J. Suttles
    . 2008. Adenosine 5′-monophosphate-activated protein kinase promotes macrophage polarization to an anti-inflammatory functional phenotype. J. Immunol. 181: 8633–8641.
    OpenUrlAbstract/FREE Full Text
  10. ↵
    1. Craig J. M.,
    2. A. L. Scott
    . 2014. Helminths in the lungs. Parasite Immunol. 36: 463–474.
    OpenUrlCrossRefPubMed
  11. ↵
    1. Herbert D. R.,
    2. J. Q. Yang,
    3. S. P. Hogan,
    4. K. Groschwitz,
    5. M. Khodoun,
    6. A. Munitz,
    7. T. Orekov,
    8. C. Perkins,
    9. Q. Wang,
    10. F. Brombacher,
    11. et al
    . 2009. Intestinal epithelial cell secretion of RELM-β protects against gastrointestinal worm infection. J. Exp. Med. 206: 2947–2957.
    OpenUrlAbstract/FREE Full Text
  12. ↵
    1. Hung L. Y.,
    2. I. P. Lewkowich,
    3. L. A. Dawson,
    4. J. Downey,
    5. Y. Yang,
    6. D. E. Smith,
    7. D. R. Herbert
    . 2013. IL-33 drives biphasic IL-13 production for noncanonical type 2 immunity against hookworms. Proc. Natl. Acad. Sci. USA 110: 282–287.
    OpenUrlAbstract/FREE Full Text
  13. ↵
    1. Wills-Karp M.,
    2. R. Rani,
    3. K. Dienger,
    4. I. Lewkowich,
    5. J. G. Fox,
    6. C. Perkins,
    7. L. Lewis,
    8. F. D. Finkelman,
    9. D. E. Smith,
    10. P. J. Bryce,
    11. et al
    . 2012. Trefoil factor 2 rapidly induces interleukin 33 to promote type 2 immunity during allergic asthma and hookworm infection. J. Exp. Med. 209: 607–622.
    OpenUrlAbstract/FREE Full Text
  14. ↵
    1. Chen F.,
    2. Z. Liu,
    3. W. Wu,
    4. C. Rozo,
    5. S. Bowdridge,
    6. A. Millman,
    7. N. Van Rooijen,
    8. J. F. Urban Jr..,
    9. T. A. Wynn,
    10. W. C. Gause
    . 2012. An essential role for TH2-type responses in limiting acute tissue damage during experimental helminth infection. Nat. Med. 18: 260–266.
    OpenUrlCrossRefPubMed
  15. ↵
    1. Carroll K. C.,
    2. B. Viollet,
    3. J. Suttles
    . 2013. AMPKα1 deficiency amplifies proinflammatory myeloid APC activity and CD40 signaling. J. Leukoc. Biol. 94: 1113–1121.
    OpenUrlAbstract/FREE Full Text
  16. ↵
    1. Carroll K. C.,
    2. B. Viollet,
    3. J. Suttles
    . 2013. AMPKa1 deficiency amplifies proinflammatory myeloid APC activity and CD40 signaling. J. Leukoc. Biol. 94: 1113–1121.
    OpenUrlAbstract/FREE Full Text
  17. ↵
    1. Barlow J. L.,
    2. R. J. Flynn,
    3. S. J. Ballantyne,
    4. A. N. McKenzie
    . 2011. Reciprocal expression of IL-25 and IL-17A is important for allergic airways hyperreactivity. Clin. Exp. Allergy 41: 1447–1455.
    OpenUrlCrossRefPubMed
    1. Neill D. R.,
    2. A. N. McKenzie
    . 2011. Nuocytes and beyond: new insights into helminth expulsion. Trends Parasitol. 27: 214–221.
    OpenUrlCrossRefPubMed
  19. ↵
    1. Neill D. R.,
    2. S. H. Wong,
    3. A. Bellosi,
    4. R. J. Flynn,
    5. M. Daly,
    6. T. K. Langford,
    7. C. Bucks,
    8. C. M. Kane,
    9. P. G. Fallon,
    10. R. Pannell,
    11. et al
    . 2010. Nuocytes represent a new innate effector leukocyte that mediates type-2 immunity. Nature 464: 1367–1370.
    OpenUrlCrossRefPubMed
  20. ↵
    1. Monticelli L. A.,
    2. G. F. Sonnenberg,
    3. M. C. Abt,
    4. T. Alenghat,
    5. C. G. Ziegler,
    6. T. A. Doering,
    7. J. M. Angelosanto,
    8. B. J. Laidlaw,
    9. C. Y. Yang,
    10. T. Sathaliyawala,
    11. et al
    . 2011. Innate lymphoid cells promote lung-tissue homeostasis after infection with influenza virus. Nat. Immunol. 12: 1045–1054.
    OpenUrlCrossRefPubMed
  21. ↵
    1. Finkelman F. D.,
    2. T. Shea-Donohue,
    3. J. Goldhill,
    4. C. A. Sullivan,
    5. S. C. Morris,
    6. K. B. Madden,
    7. W. C. Gause,
    8. J. F. Urban Jr..
    1997. Cytokine regulation of host defense against parasitic gastrointestinal nematodes: lessons from studies with rodent models. Annu. Rev. Immunol. 15: 505–533.
    OpenUrlCrossRefPubMed
  22. ↵
    1. Finkelman F. D.,
    2. K. B. Madden,
    3. A. W. Cheever,
    4. I. M. Katona,
    5. S. C. Morris,
    6. M. K. Gately,
    7. B. R. Hubbard,
    8. W. C. Gause,
    9. J. F. Urban Jr..
    1994. Effects of interleukin 12 on immune responses and host protection in mice infected with intestinal nematode parasites. J. Exp. Med. 179: 1563–1572.
    OpenUrlAbstract/FREE Full Text
  23. ↵
    1. Pearce E. L.,
    2. M. C. Poffenberger,
    3. C. H. Chang,
    4. R. G. Jones
    . 2013. Fueling immunity: insights into metabolism and lymphocyte function. Science 342: 1242454.
    OpenUrlAbstract/FREE Full Text
    1. Pearce E. L.,
    2. E. J. Pearce
    . 2013. Metabolic pathways in immune cell activation and quiescence. Immunity 38: 633–643.
    OpenUrlCrossRefPubMed
  25. ↵
    1. Huang S. C.,
    2. B. Everts,
    3. Y. Ivanova,
    4. D. O’Sullivan,
    5. M. Nascimento,
    6. A. M. Smith,
    7. W. Beatty,
    8. L. Love-Gregory,
    9. W. Y. Lam,
    10. C. M. O’Neill,
    11. et al
    . 2014. Cell-intrinsic lysosomal lipolysis is essential for alternative activation of macrophages. Nat. Immunol. 15: 846–855.
    OpenUrlCrossRefPubMed
  26. ↵
    1. O’Sullivan D.,
    2. G. J. van der Windt,
    3. S. C. Huang,
    4. J. D. Curtis,
    5. C. H. Chang,
    6. M. D. Buck,
    7. J. Qiu,
    8. A. M. Smith,
    9. W. Y. Lam,
    10. L. M. DiPlato,
    11. et al
    . 2014. Memory CD8+ T cells use cell-intrinsic lipolysis to support the metabolic programming necessary for development. Immunity 41: 75–88.
    OpenUrlCrossRefPubMed
  27. ↵
    1. Phills J. A.,
    2. A. J. Harrold,
    3. G. V. Whiteman,
    4. L. Perelmutter
    . 1972. Pulmonary infiltrates, asthma and eosinophilia due to Ascaris suum infestation in man. N. Engl. J. Med. 286: 965–970.
    OpenUrlCrossRefPubMed
  28. ↵
    1. Blagih J.,
    2. F. Coulombe,
    3. E. E. Vincent,
    4. F. Dupuy,
    5. G. Galicia-Vázquez,
    6. E. Yurchenko,
    7. T. C. Raissi,
    8. G. J. van der Windt,
    9. B. Viollet,
    10. E. L. Pearce,
    11. et al
    . 2015. The energy sensor AMPK regulates T cell metabolic adaptation and effector responses in vivo. Immunity 42: 41–54.
    OpenUrlCrossRefPubMed
  29. ↵
    1. Molofsky A. B.,
    2. F. Van Gool,
    3. H. E. Liang,
    4. S. J. Van Dyken,
    5. J. C. Nussbaum,
    6. J. Lee,
    7. J. A. Bluestone,
    8. R. M. Locksley
    . 2015. Interleukin-33 and interferon-γ counter-regulate group 2 Innate lymphoid cell activation during immune perturbation. Immunity 43: 161–174.
    OpenUrlCrossRefPubMed
  30. ↵
    1. Gundra U. M.,
    2. N. M. Girgis,
    3. D. Ruckerl,
    4. S. Jenkins,
    5. L. N. Ward,
    6. Z. D. Kurtz,
    7. K. E. Wiens,
    8. M. S. Tang,
    9. U. Basu-Roy,
    10. A. Mansukhani,
    11. et al
    . 2014. Alternatively activated macrophages derived from monocytes and tissue macrophages are phenotypically and functionally distinct. Blood 123: e110–e122.
    OpenUrlAbstract/FREE Full Text
  31. ↵
    1. Vats D.,
    2. L. Mukundan,
    3. J. I. Odegaard,
    4. L. Zhang,
    5. K. L. Smith,
    6. C. R. Morel,
    7. R. A. Wagner,
    8. D. R. Greaves,
    9. P. J. Murray,
    10. A. Chawla
    . 2006. Oxidative metabolism and PGC-1β attenuate macrophage-mediated inflammation. Cell Metab. 4: 13–24.
    OpenUrlCrossRefPubMed
  32. ↵
    1. Jang J. C.,
    2. M. G. Nair
    . 2013. Alternatively activated macrophages revisited: new insights into the regulation of immunity, inflammation and metabolic function following parasite infection. Curr. Immunol. Rev. 9: 147–156.
    OpenUrlPubMed
    1. Osborne L. C.,
    2. L. A. Monticelli,
    3. T. J. Nice,
    4. T. E. Sutherland,
    5. M. C. Siracusa,
    6. M. R. Hepworth,
    7. V. T. Tomov,
    8. D. Kobuley,
    9. S. V. Tran,
    10. K. Bittinger,
    11. et al
    . 2014. Coinfection. Virus-helminth coinfection reveals a microbiota-independent mechanism of immunomodulation. Science 345: 578–582.
    OpenUrlAbstract/FREE Full Text
    1. Ferrante C. J.,
    2. S. J. Leibovich
    . 2012. Regulation of macrophage polarization and wound healing. Adv. Wound Care (New Rochelle) 1: 10–16.
    OpenUrlCrossRefPubMed
  35. ↵
    1. Vannella K. M.,
    2. L. Barron,
    3. L. A. Borthwick,
    4. K. N. Kindrachuk,
    5. P. B. Narasimhan,
    6. K. M. Hart,
    7. R. W. Thompson,
    8. S. White,
    9. A. W. Cheever,
    10. T. R. Ramalingam,
    11. T. A. Wynn
    . 2014. Incomplete deletion of IL-4Rα by LysMCre reveals distinct subsets of M2 macrophages controlling inflammation and fibrosis in chronic schistosomiasis. PLoS Pathog. 10: e1004372.
    OpenUrlCrossRefPubMed
  36. ↵
    1. Zhao X.,
    2. J. W. Zmijewski,
    3. E. Lorne,
    4. G. Liu,
    5. Y. J. Park,
    6. Y. Tsuruta,
    7. E. Abraham
    . 2008. Activation of AMPK attenuates neutrophil proinflammatory activity and decreases the severity of acute lung injury. Am. J. Physiol. Lung Cell. Mol. Physiol. 295: L497–L504.
    OpenUrlAbstract/FREE Full Text
  37. ↵
    1. Zhu Y. P.,
    2. J. R. Brown,
    3. D. Sag,
    4. L. Zhang,
    5. J. Suttles
    . 2015. Adenosine 5′-monophosphate-activated protein kinase regulates IL-10-mediated anti-inflammatory signaling pathways in macrophages. J. Immunol. 194: 584–594.
    OpenUrlAbstract/FREE Full Text
  38. ↵
    1. Halim T. Y.,
    2. Y. Y. Hwang,
    3. S. T. Scanlon,
    4. H. Zaghouani,
    5. N. Garbi,
    6. P. G. Fallon,
    7. A. N. McKenzie
    . 2016. Group 2 innate lymphoid cells license dendritic cells to potentiate memory TH2 cell responses. Nat. Immunol. 17: 57–64.
    OpenUrlCrossRefPubMed
  39. ↵
    1. Kudo F.,
    2. M. Ikutani,
    3. Y. Seki,
    4. T. Otsubo,
    5. Y. I. Kawamura,
    6. T. Dohi,
    7. K. Oshima,
    8. M. Hattori,
    9. S. Nakae,
    10. K. Takatsu,
    11. S. Takaki
    . 2016. Interferon-γ constrains cytokine production of group 2 innate lymphoid cells. Immunology 147: 21–29.
    OpenUrlCrossRefPubMed
  40. ↵
    1. Whitsett J. A.,
    2. T. Alenghat
    . 2015. Respiratory epithelial cells orchestrate pulmonary innate immunity. Nat. Immunol. 16: 27–35.
    OpenUrlPubMed
  41. ↵
    1. Jenkins S. J.,
    2. D. Ruckerl,
    3. P. C. Cook,
    4. L. H. Jones,
    5. F. D. Finkelman,
    6. N. van Rooijen,
    7. A. S. MacDonald,
    8. J. E. Allen
    . 2011. Local macrophage proliferation, rather than recruitment from the blood, is a signature of TH2 inflammation. Science 332: 1284–1288.
    OpenUrlAbstract/FREE Full Text
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section