ATP6V0d2 Suppresses Alveoli Macrophage Alternative Polarization and Allergic Asthma via Degradation of PU.1.

PURPOSE: Macrophages are important regulators of environmental allergen-induced airway inflammation and asthma. ATP6V0d2 is a subunit of vacuolar ATPase highly expressed in macrophages. However, the functions of ATP6V0d2 in the regulation of pathogenesis of allergic asthma remain unclear. The aim of this study is to determine the function and related molecular mechanisms of macrophage protein ATP6V0d2 in allergic asthma.
METHODS: We compared the disease severity between female C57BL/6 wild-type and ATP6V0d2-/- mice in an ovalbumin (OVA)-induced asthma model. We also investigated the association of expression of ATP6V0d2, PU.1 and CCL17 with disease severity among asthmatic patients.
RESULTS: The expression of ATP6V0d2 in sputum cells of asthmatic patients and in the lungs of OVA-challenged mice was enhanced compared to healthy subjects and their counterparts, respectively. However, ATP6V0d2-deficient mice exaggerated inflammatory cell infiltration as well as enhanced alternative activated macrophage (AAM) polarization and mucus production in an OVA-induced asthma model. Furthermore, we found that Atp6v0d2 promoted lysosomal degradation of Pu.1, which induced AAM polarization and Ccl17 production. Among asthma patients, ATP6V0d2 expression was inversely associated with disease severity, whereas PU.1 and CCL17 expression was positively associated with disease severity.
CONCLUSIONS: Our results identify macrophage Atp6v0d2, as an induced feedback inhibitor of asthma disease severity by promoting Pu.1 lysosomal degradation, which may in turn leads to reduced AAM polarization and Ccl17 production.

INTRODUCTION

Asthma is a chronic inflammatory disease of the lower airways of the lungs characterized by bronchospasm and inflammatory infiltration.12 Macrophages are the most abundant immune cell type in the lungs and play crucial roles in environmental allergen-induced airway inflammation and asthma.34 Alveolar macrophages (AMs) are the major effector cells of immune responses in the lungs, which process both proinflammatory and anti-inflammatory properties.56 In allergic asthma, after allergen exposure there is a rapid recruitment of monocytes to the airways and lung tissues,7 which differentiate into a subpopulation of AMs called monocyte-derived AMs (Mo-AMs).8 Mo-AMs undergo further polarization and promote acute inflammatory responses.910

Despite its oversimplification and limitation in explaining all macrophage phenotypes and functions, polarized macrophages can be divided into classic activated macrophages (CAM) or alternative activated macrophages (AAMs) cell lineages.1112 AAM polarization is driven by cytokines involved in Th2 responses including IL-4 and IL-13. The ETS family transcriptional factor Pu.1 plays critical roles in macrophage differentiation and promotes AAM polarization.13 Inducible ablation of Pu.1 attenuated eosinophil infiltration and airway hyperinflammation, which can be reversed by adoptive transfer of wild-type (WT) macrophages.1415 AAMs express Th2-related chemokines, CCR4 ligands such as Ccl17 and Ccl22 the latter of which has been reported to be induced by Pu.1.16 In keeping with their association with atopic disease and Th2 associated cytokines, AAMs were significantly increased in bronchoalveolar lavage fluid (BALF) from patients with asthma and promoted the allergic airway responses and the pathogenesis of asthma.17 Understanding the molecular details of regulation of AAM polarization and functions will provide a new insight into the treatment of asthma.

Atp6v0d2 is a subunit of vacuolar ATPase that acts as a proton pump and regulates extracellular and intracellular organelle acidification such as lysosome.1819 However, Atp6v0d2 is dispensable for the acidification for lysosomes in macrophages.20 We recently identified that Atp6v0d2 senses inflammatory stimuli and tumor cell-derived lactate and regulates AAM differentiation in inflammation and tumor microenvironment.21 In addition, Atp6v0d2 regulates leucine-induced mTORC1 activation and suppresses AAM polarization.22

Given the sentinel role of Atp6v0d2 in macrophages, we speculated that Atp6v0d2 might regulate macrophage-associated allergic asthma. Using sputum samples from asthmatic patients and ATP6V0d2 mice in an ovalbumin (OVA)-induced asthma model, we identified a novel ATP6V0d2-PU.1-CCL17 axis that regulates asthma pathogenesis and provided the mechanism that Atp6v0d2 mediates the lysosomal degradation of Pu.1, which directly binds to the Ccl17 promoter and regulates the expression of Ccl17. More importantly, ATP6V0d2 expression was reversely associated with disease severity in human asthmatic patients.

MATERIALS AND METHODS

Mice

ATP6V0d2−/− mice were generated using transcription activator-like effector nuclease (TALEN) technology as previously described.21 C57BL/6 WT mice were purchased from Huafukang (Beijing, China). All the mice were bred and housed in a specific-pathogen-free facility at Tongji Medical College, Huazhong University of Science and Technology (Wuhan, China) and used in accordance with institutional guidelines.

Patients and specimens

Induced sputum samples were obtained from healthy volunteers (n = 16) and asthmatic patients (n = 22) enrolled in Tongji Hospital. All of the subjects provided informed consent. The diagnosis of asthma was based on asthma guidelines provided by the American Thoracic Society.23 None of the patients in the asthma group had undergone inhaled corticosteroid therapy before blood and lung function tests. All studies with human subjects were approved by the Institutional Ethical Review Board of Tongji Hospital (TJ-IRB20180518). The clinical characteristics of all subjects are provided in Supplementary Table S1.

Sputum induction and sputum cell isolation

Sputum induction was performed as described before.24 In brief, the sputum was induced in healthy volunteers and asthmatic patients by inhalation of nebulized hypertonic (4.5%, 25 mL) sterile saline solution. Patients expectorated whenever it is needed or every 5 minutes. If sputum amount was inadequate, induction was continued for another 5 minutes. The whole procedure lasted for 20 minutes. The sputum was isolated from saliva and dispersed with 0.1% dithiothreitol and then filtered through a 40-μm Nitex filter to remove mucus. The resulting suspension was centrifuged at 1,000 rpm for 5 minutes at 4°C and the cell pellets were washed once and re-suspended with phosphate buffered saline (PBS). The cell suspension was centrifuged by a cytospin for immunohistochemistry staining or was used directly for RNA isolation or flow cytometry analysis.

Murine bone marrow-derived macrophage (BMDM) induction and activation

Primary BMDMs were generated by culturing mouse bone marrow cells in RPMI 1640 medium supplemented with 10% fetal calf serum, 100 U/mL penicillin G and 100 µg/mL streptomycin, in addition to recombinant mouse M-CSF (20 ng/mL, PeproTech, Rocky Hill, NJ, USA) for 7 days. On day 7, macrophages were washed and stimulated with IL-4 (20 ng/L, PeproTech) for different hours.

Quantitative reverse transcription polymerase chain reaction (RT-PCR)

Total RNAs were extracted from lung tissues or cells with Trizol reagent (Invitrogen, Carlsbad, CA, USA). cDNA was synthesized with 1 μg RNA isolated from mouse lung tissues or with 0.3 μg RNA from human sputum samples by using the Reverse Transcription Kit (Toyobo, Osaka, Japan) following the manufacturer's protocol. All quantitative RT-PCR was performed with the SYBR green method on a Bio-Rad CFX. The quantification of the results was performed by the comparative Ct (2−∆∆Ct) method. The Ct value for each sample was normalized by the value for Gapdh or β-Actin gene. Primer sequences for mouse genes and human genes were provided in Supplementary Table S2.

OVA-induced asthma model in mice

Asthma was induced in female WT and ATP6V0d2 mice (8–12 weeks old) as previously described.25 Briefly, mice were sensitized by intraperitoneal injection of 100 µL OVA liquid in which 20 mg of OVA (grade V, Sigma, A5503; Sigma-Aldrich, St. Louis, MO, USA) was emulsified in 4 mg of aluminum hydroxide (Sigma, 239186; Sigma-Aldrich) in a total volume of 100 mL on days 0, 7, and 14. Mice were injected with an equal volume of saline served as control animals. After sensitization, 1% OVA (grade II, Sigma, A5253; Sigma-Aldrich) was atomized and inhaled by mice for 7 consecutive days to induce asthma.

Lung tissue histological analysis

For histological analysis, lungs were harvested and fixed in 4% buffered formalin for 24 hours at room temperature. Tissues were then embedded in paraffin; 6-µm sections were cut with a micrometer and stained with hematoxylin and eosin, Periodic acid-Schiff (PAS) and Masson. The images were assessed by 2 independent pathologists using a bright field microscope in a blinded fashion. Inflammatory cell infiltration was scored with a scale of 0 to 4, with 0 representing no inflammation, 1 representing light cell infiltration in only a few areas, 2 and 3 representing moderate cell infiltration encompassing < 20% and 50%, respectively, of the airways and vessels, and 4 representing dense cell infiltration encompassing > 50% of the airways and vessels. All histomorphometric measurements were quantified within 3 randomly selected visual fields per slide from each group (n = 4–6).

Flow cytometry

After perfusion with 5 mL of PBS through the right ventricle, lung tissues were cut into small pieces and digested with RPMI 1640 medium containing 1 mg/ml collagenase D (Roche, 11088866001; Roche, Basel, Switzerland) for 1 hour at 37°C. Single lung cell suspensions were obtained after filtration and red cell lysis. Before staining, cells were incubated with Fc blocker anti-CD16/32 (Biolegend, clone 93; Biolegend, San Diego, CA, USA). Cells were stained with the following surface anti-mouse antibodies: CD11b-FITC (clone M1/70), F4/80-Percp/cy5.5 (clone BM8), CD206-PE/cy7 (clone C068C2), CD3-APC/cy7 (clone 17A2), CD19-BV421 (clone 6D5) (all from Biolegend), SiglecF-PE (clone, 1RNM44N) and CD64-APC (clone 10.1) from eBioscience (San Diego, CA, USA).

For human sputum cell staining, the single cell suspensions were blocked with anti-human Trustain FcXTM (Biolegend), then stained with the following antibodies: CD33-FITC (clone HIM3-4), CD3-APC (clone OKT3), CD3-PE (clone HIT3a), CD19-APC (clone HIB19), CD68-PE (clone Y1/82A) (all from eBioscience), and CD68-APC (clone Y1/82A, Biolegend). Samples were collected on a BD Verse Flow cytometer, and data were analyzed using FlowJo software (Tree Star, San Carlos, CA, USA).

Immunofluorescence and immunoblotting

BMDMs were seeded at 3 × 104 cells per well on glass slides and rested overnight for proper attachment. Lung AMs from control mouse BALF were seeded at 1 × 104 cells per well on glass slides, and rested 3 hours for attachment. Then, the cells were stimulated with IL-4 (20 ng/mL) for the indicated time, followed by incubation with 1:5,000 PBS-diluted LysoTracker Red DND-99 dye (Cat: L7528, Life Technologies, Carlsbad, CA, USA) for 30 minutes. Afterwards, cells were washed twice with sterile PBS and fixed with 4% paraformaldehyde, permeabilized with 0.05% Triton X-100, and blocked in 5% BSA. Anti-Pu.1(2266, CST) was incubated overnight in 4°C. Secondary fluorescent antibodies FITC (from Life Technologies or Jackson Laboratories, Bar Harbor, ME, USA) were added for 1 hour and 4',6-diamidino-2-phenylindole (DAPI) was used for nuclear counterstaining. Samples were imaged with a SP5 confocal (Leica, Wetzlar, Germany) after 24-hour mounting. Immunoblotting was performed according to the standard method. Cell lysates were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and proteins were transferred from gel onto polyvinylidene difluoride membranes. Membranes were incubated with the following primary antibodies: anti-Atp6v0d2 (SAB2103221; Sigma), anti- Atp6v0d1 (ab56441; Abcam, Cambridge, UK), anti- Atp6v1a (ab137574; Abcam), p-NF-κb (3033; Cell Signaling Technology, Danvers, MA, USA), Pu.1 (2258; Cell Signaling Technology), Stat6 (5397; Cell Signaling Technology), phosphor-Stat6 (9362; Cell signaling Technology), p62 (PM045; MBL, Woburn, MA, USA), and β-actin (4970S; Cell Signaling Technology). After washing, membranes were incubated with appropriate secondary horseradish peroxidase-conjugated antibodies, and then developed with enhanced chemiluminescence (GE Healthcare, Chicago, IL, USA).

Statistical analysis

All values are expressed as the mean ± standard error of the mean for each group. Statistical significance was assessed using an unpaired Student's t-test or Mann-Whitney U test. Using Prism software for data comparisons were made between 2 groups. When groups contained more than 2 samples, they were analyzed using one-way analysis of variance with Tukey's post hoc test. GraphPad Prism was used for all statistical calculation. A P value of < 0.05 was considered statistically significant (GraphPad, San Diego, CA, USA).

RESULTS

ATP6V0d2 expression is enhanced in the sputum cells of asthmatic patients as well as in lung tissue and BALF cells of OVA-induced asthmatic mice

To investigate the potential role of ATP6V0d2 in asthma, we first compared the mRNA expression of ATP6V0d2 in the cell pellets of induced sputum from healthy donors and asthmatic patients. Asthmatic patients exhibited significantly higher ATP6V0d2 mRNA expression in the sputum cells compared to healthy controls (Fig. 1A). This was confirmed at the protein level using immunohistochemistry in the sputum cells (Fig. 1B and C). Notably, the percentage of CD68+ macrophage population was increased in the sputum of asthmatic patient (Supplementary Fig. S1A). Immunofluorescence data showed that the expression of ATP6V0d2 was mainly co-localized with CD68 staining in sputum cells from asthmatic patient (Fig. 1D). Similar to the human results, we found that OVA-challenged mice showed significantly higher expression of Atp6v0d2 mRNA within lung tissues (Fig. 1E). Around 89% cells within BALF from control mouse were F4/80+CD11c+ AMs (Supplementary Fig. S1B). Similarly, Atp6v0d2 expression was enhanced in the OVA-treated BALF cells compared to unchallenged controls (Fig. 1F and G). Taken together, these data demonstrate that ATP6V0d2 expression is enhanced in both asthmatic humans and mice.

Fig. 1

ATP6V0d2 expression is increased in the sputum cells of asthmatic patients and in the lung tissues and BALF cells of OVA-challenged mice. (A) qPCR analysis of ATP6V0d2 expression in sputum samples from asthmatic patients (n = 22) and healthy controls (n = 16). (B, C) A representative image of IHC analysis of ATP6V0d2 in the sputum samples from healthy donors and asthmatic patients as well as quantification of the percentages of ATP6V0d2-positive cells in each sputum sample in healthy controls and asthmatic patients. (D) Immunofluorescence staining of isolated sputum cells from asthmatic patients' saliva with DAPI (blue), anti-CD68 (green), and anti-ATP6V0d2 (red). (E) The expression of Atp6v0d2 in the lungs of mice after OVA sensitization (n = 10–11). (F, G) A representative image of IHC analysis of Atp6v0d2 in the BALF between WT mice and OVA-induced mice and the percentages of Atp6v0d2-positive cells in BALF cells sample were analyzed (scale bar: 100 µm). The percentage of ATP6V0d2 in sputum samples from human (n = 9–11) or in BALF from mice (n = 9–10) was analyzed by ImageJ software, and the mean value of 3 image fields per slides was plotted. Data are shown as the ratio compared to Gapdh or β-ACTIN to 2−ΔCt for mRNA levels. Data were assessed by the Mann-Whitney U test and are presented as mean ± standard error of the mean.

BALF, bronchoalveolar lavage fluid; OVA, ovalbumin; qPCR, quantitative polymerase chain reaction; IHC, immunohistochemistry; DAPI, 4',6-diamidino-2-phenylindole; WT, wild-type.

*P < 0.001.

IL-4 induces Atp6v0d2 expression via activation of Stat6 in macrophages

IL-4 is a major cytokine in driving Th2 differentiation and AAM polarization. The BALF supernatants of OVA-sensitized mice had significantly enhanced IL-4 production, compared to control mice (Fig. 2A). Stimulation of WT BALF cells with IL-4 enhanced ATP6V0d2 mRNA expression (Fig. 2B). IL-4 induced ATP6V0d2 mRNA expression and protein expression in BMDMs (Fig. 2C and D). However, IL-4 had no effect on the expression of Atp6v0d1 and Atp6v1a, suggesting that IL-4 specifically regulates Atp6v0d2 expression in macrophages. By sequence comparison, we identified 4 putative Stat6 binding sites in the ATP6V0d2 promoter (Fig. 2E). Stimulation of IL-4 in macrophages led to enhanced Stat6 binding in the regions amplified by primer pairs 1, 2, and 3 (Fig. 2F). This was associated with enhanced trimethylation of histone 3 lysine 4 (H3K4me3), a transcriptional activation marker in the primer 1 region (Fig. 2G). While IL-4 promotes the AAM phenotype, lipopolysaccharide (LPS) induces macrophage polarization into CAM phenotype. Consistent with this, LPS suppressed Atp6v0d2 expression, but not Atp6v0d1 and Atp6v1a, in an NF-kB-dependent manner (Supplementary Fig. S2A and B). The suppression of IL-4-induced Atp6v0d2 by LPS was dose-dependent (Supplementary Fig. S2C and D) and the addition of LPS resulted in significantly reduced Stat6 binding to the promoter (Fig. 2H). Conversely, increasing dose of IL-4 could overcome the suppressive effect of LPS on Atp6v0d2 expression (Supplementary Fig. S2E and F).

Fig. 2

ATP6V0d2 expression is induced by IL-4 through activation of Stat6 in macrophages. (A) IL-4 concentrations were measured in control animals and OVA-treated mice. (B) BALF cells isolated form control mice were stimulated with IL-4 (20 ng/mL) for 3 hours and the expression of Atp6v0d2 was detected by qPCR. (C) Macrophages were stimulated with IL-4 (20 ng/mL) for the indicated time and the expression of Atp6v0d2 was determined by qPCR. (D) Immunoblotting of p-Stat6, Atp6v0d2, Atp6v0d1 and Atp6v1a in macrophages stimulated with IL-4 (20 ng/mL) for the indicated time. (E) Schematic representation showing primers used for the CHIP analysis. (F, G) Chromatin immunoprecipitation analysis of Stat6 and H3K4me3 occupancy to the Atp6v0d2 locus in macrophages that were stimulated with IL-4 for 2 hours. Precipitated DNA was amplified by qPCR for primer sites p1–p4 and results were presented relative to input DNA. (H) Chromatin immunoprecipitation analysis for Stat6 occupancy to the Atp6v0d2 locus in macrophages that were stimulated with IL-4 plus LPS (100 ng/mL) for 2 hours. Data are representative of 3 independent experiments. Data were assessed by unpaired Student's t-test and are presented as mean ± standard error of the mean.

OVA, ovalbumin; BALF, bronchoalveolar lavage fluid; qPCR, quantitative polymerase chain reaction; LPS, lipopolysaccharide.

*P < 0.05; †P < 0.01; ‡P < 0.001.

Taken together, these data demonstrate that Atp6v0d2 expression is counter-regulated by IL-4 and LPS, suggesting it may serve as a focal point for macrophage polarization and function.

ATP6V0d2-deficient mice exhibit exacerbated OVA-induced airway inflammation, remodeling and mucus production

To investigate the function of Atp6v0d2 in asthma, we employed an OVA-induced allergic asthma model in WT and ATP6V0d2-deficient mice. We found significantly enhanced infiltration of inflammatory cells surrounding the airway and blood vessels of the lungs in ATP6V0d2 mice compared with control mice (Fig. 3A). The inflammatory scores in the lungs of ATP6V0d2 mice were significantly higher (Fig. 3B). Using PAS staining, we found aggravated mucus hyperplasia in the bronchi in OVA-challenged ATP6V0d2 mice (Fig. 3C and D). Next, we used Masson trichrome staining of the collagen deposition to evaluate the airway wall thickening. The lungs of Atp6v0d2-deficient mice had enhanced collagen deposition (Fig. 3E and F). Moreover, the expression of Muc5ac and Muc5b was further potentiated in the lungs of OVA-challenged ATP6V0d2 mice despite the difference of Muc5b was not significant (Fig. 3G and H). Together, these data demonstrate that Atp6v0d2 suppresses OVA-induced airway inflammation, remodeling, and mucus production.

Fig. 3

ATP6V0d2-deficient mice were more susceptible to OVA-induced asthma. (A, B) Hematoxylin and eosin staining of lung tissues in WT and ATP6V0d2 mice after saline or OVA treatment, and the mean scores of inflammation were analyzed. (C, D) PAS staining on lung tissues in WT and ATP6V0d2 mice after saline or OVA treatment and the PAS-positive cells were counted. (E, F) Masson trichrome staining of lung tissues in WT and ATP6V0d2 mice after saline or OVA treatment, and the positive areas were quantified. (G, H) qRCR analysis of muc5ac and muc5b in the lung tissues from WT and ATP6V0d2 mice after saline or OVA treatment (scale bar: 100 µm). Data are representative of 2 independent experiments, and shown as the ratio compared to Gapdh to 2−ΔCt for mRNA levels. Quantification was analyzed for 3 randomly selected visual fields per slide (n = 4–6). Data were assessed by one-way analysis of variance with Turkey's test and are presented as mean ± standard error of the mean.

OVA, ovalbumin; WT, wild-type; PAS, Periodic acid-Schiff; qPCR, quantitative polymerase chain reaction; KO, knockout.

*P < 0.05; †P < 0.01; ‡P < 0.001.

ATP6V0d2 mice enhance inflammatory cell infiltration and cytokine productions in OVA-induced asthma

Next, we compared the inflammatory cell infiltration in the BALF supernatants of OVA-challenged WT and ATP6V0d2−/− mice. ATP6V0d2 mice had significantly increased numbers of total infiltrated inflammatory cells compared to WT mice, and the numbers of eosinophils, macrophages, and T lymphocytes were higher in the BALF supernatants of ATP6V0d2 mice (Fig. 4A). Flow cytometry showed enhanced presence of SiglecF+CD64− eosinophils among the populations of CD11b+ cells in the lungs of ATP6V0d2 mice (Fig. 4B). T-cell infiltration in the lungs of ATP6V0d2 mice after OVA challenge was enhanced; however, the CD19+ B-cell infiltration was comparable (Supplementary Fig. S3A). Next, we compared the levels of type 2 cytokines involved in the asthma process in lung tissue homogenates as well as BALF of WT and ATP6V0d2 mice. OVA-challenged WT mice had elevated levels of IL-4, IL-5, IL-13, and Ccl17 in the lungs compared to control animals. The expressions of IL-4, IL-5, IL-13 and Ccl17 were significantly further enhanced in the absence of ATP6V0d2 upon OVA challenge (Fig. 4C). Similar results were obtained for IL-4, IL-5, IL-13, and Ccl17 production in the BALF supernatants (Fig. 4D).

Fig. 4

Deletion of Atp6v0d2 exacerbated lung inflammation and elevated production of inflammatory mediators upon asthma induction. (A) Numbers of total cells or different types of immune cells, including eosinophils, macrophages, and T cells, in BALF of saline or OVA-treated WT and ATP6V0d2 mice. (B) Flow cytometry analysis eosinophils in the lungs from WT and ATP6V0d2 mice after saline or OVA treatment. (C, D) The levels of IL-4, IL-5, IL-13 and Ccl17 in the lungs and BALF, respectively, were determined by ELISA (n = 4–6). (E, F) Serum levels of total IgE and IgG1were determined in WT and ATP6V0d2 mice treated with saline or OVA using ELISA (n = 4-6). Data are representative of 2 independent experiments. Data were assessed by one-way analysis of variance with Turkey's test and are presented as mean ± standard error of the mean.

BALF, bronchoalveolar lavage fluid; OVA, ovalbumin; WT, wild-type; ELISA, enzyme-linked immunosorbent assay; Ig, immunoglobulin; KO, knockout.

*P < 0.05; †P < 0.01; ‡P < 0.001.

Despite the enhanced expression of IL-6 and IL-10 in WT mice after OVA treatment, the concentrations of IL-6 and IL-10 in the lungs were comparable between WT and ATP6V0d2 mice (Supplementary Fig. S3B). Similar results were obtained in the BALF (Supplementary Fig. S3C). Meanwhile, the serum levels of immunoglobulin E (IgE) and IgG1were also enhanced in ATP6V0d2 mice compared to control animals despite the difference of IgG1 was not significant (Fig. 4E and F). Taken together, these data demonstrate that ATP6V0d2-deficiency led to enhanced infiltration of eosinophils, macrophages and T lymphocytes type 2 cytokines as well as IgE production during asthma.

Deletion of ATP6V0d2 results in enhanced AAM polarization in asthma

We next asked whether the enhanced asthma disease severity seen in the absence of Atp6v0d2 was due to enhanced alternative macrophage polarization. Immunofluorescence staining of lung sections after OVA-challenge showed the presence of higher numbers of F4/80+CD206+AAM cells in the group of ATP6V0d2 mice compared to WT mice (Fig. 5A and B). This was further confirmed by flow cytometry analysis in the lung tissues of OVA-treated ATP6V0d2 mice compared to WT mice (Fig. 5C). In the lungs of OVA-treated ATP6V0d2 mice, the expressions of AAM associated genes, including Fizz-1, Mrc-1, Ym-1, and Il-4r, were significantly higher compared to lungs from WT mice (Fig. 5D). However, the expressions of CAM-associated genes, including Il-1b, Il-6, Il12 p40, and iNos, were comparable (Fig. 5E).

Fig. 5

ATP6V0d2-deficiency enhanced AAM polarization in the OVA asthma model. (A, B) Whole-mount immunofluorescence staining of DAPI (blue), F4/80 (red), and CD206 (green) in the lung tissues of WT and ATP6V0d2 mice treated with OVA and the percentages of F4/80+CD206+ among F4/80+ were quantified with 2 randomly selected visual fields per slide (n = 5). (C) Flow cytometry analysis of singlets, SiglecF+F4/80+, and CD206+CD64+ cells in lungs from OVA-challenged WT and ATP6V0d2 mice and histogram plots of percentages of CD206+CD64+ alveolar macrophages among total cells. (D, E) Expression of AAM-related markers (Fizz-1, Mrc-1, Ym-1, and Il-4r) and CAM-related markers (Il-1b, Il-6, Il-12p40 and iNos) in lungs from OVA-treated WT and ATP6V0d2 mice was determined by qPCR (n = 4–6) (scale bar: 100 µm). Data are representative of 2 independent experiments and shown as the ratio compared to Gapdh to 2ΔCt for mRNA levels. Data were assessed by the Mann-Whitney U test and are presented as mean ± standard error of the mean.

AAM, alternative activated macrophage; OVA, ovalbumin; DAPI, 4',6-diamidino-2-phenylindole; WT, wild-type; CAM, classic activated macrophage; qPCR, quantitative polymerase chain reaction; KO, knockout.

*P < 0.05; †P < 0.01; ‡P < 0.001.

Taken together, these data demonstrate that ATP6V0d2 intrinsically suppresses AAM polarization in asthma.

ATP6V0d2 promotes Pu.1 lysosomal degradation and suppressed Pu.1-mediated Ccl17 production

Pu.1 acts as a critical transcriptional factor for macrophage alternative polarization and regulates asthma pathology.1426 Treatment of IL-4 induced Pu.1 in AMs (Fig. 6A). Consistently, IL-4 also induced enhanced Pu.1 protein level in WT macrophage, however, this phenotype was further increased in the absence of ATP6V0d2, although the total and phosphorylated Stat6 levels were comparable (Supplementary Fig. S4A), indicating the differential expression of Pu.1 was not due to impaired IL-4 signaling. Thus we asked whether Atp6v0d2 affects Pu.1 stability. In the presence of cycloheximide that blocked protein synthesis, Pu.1 degradation was reduced in the absence of ATP6V0d2 (Fig. 6B and C). Treatment of Bafilomycin, a V-ATPase inhibitor led to accumulation of Pu.1 in WT BMDMs, which was abolished in the absence of ATP6V0d2 (Fig. 6D). Next, we isolated BALF cells, most of which were AMs, from WT mice and treated the cells with IL-4. IL-4-induced Pu.1 was co-localized with lysosomes stained with LysoTracker in WT AMs (Fig. 6E) and deletion of ATP6V0d2 led to significant reduction of co-localizations of Pu.1 with lysosome (Fig. 6F). We repeated the same experiments with BMDMs and found similar results (Supplementary Fig. S4B and C). Meanwhile, OVA challenge induced Pu.1 expression in the lung tissues of WT mice and this was further potentiated in the ATP6V0d2-deficient mice (Fig. 6G and H). Immunohistochemistry staining showed similar results (Supplementary Fig. S4D and E). We found IL-4 induced Ccl17 expression in AMs (Fig. 6I), which is consistent with previous study showing critical role of Ccl17 for recruitment of Th2 cells into the lungs during asthma.27 Furthermore, IL-4-induced recruitment of Pu.1 into the Ccl17 promoter and deletion of Atp6v0d2 led to enhanced IL-4-induced Ccl17 secretion in BMDMs (Fig. 6J and K). Taken together, these data suggest that ATP6V0d2 promotes Pu.1 lysosomal degradation, thus limiting Ccl17 production and asthma disease severity.

Fig. 6

Deletion of Atp6v0d2 led to Pu.1 accumulation that promotes Ccl17 production. (A) BALF cells isolated from control mice were stimulated with IL-4 (20 ng/mL) for 3 hours and the expression of Pu.1 was determined by qPCR. (B, C) Immunoblotting of Stat6, P62, Pu.1 and Atp6v0d2 in WT and ATP6V0d2 BMDMs that were incubated with cycloheximide (100 µg/mL) for the indicated time and band intensities relative to β-actin were plotted. (D) WT and ATP6V0d2 BMDMs were untreated or treated with bafilomycin A (100 nM) for 2 hours. The amounts of Pu.1 and Atp6v0d2 were determined by immunoblotting. (E, F) BALF cells from control mice were treated with or without IL-4 (20 ng/mL) for 1 hour. Cells were stained with anti-Pu.1 (green) and LysoTracker (red); and percentages of Pu.1 co-localized with LysoTracker among PU.1-positive cells were quantified. Fifteen randomly visual fields were chosen for quantification. (G, H) Immunoblotting of p-Stat6, Stat6, Pu.1 and Atp6v0d2 in lungs from OVA-challenged WT and ATP6V0d2 mice and the band intensities relative to β-actin were quantified. (I) BALF cells isolated from control mice were stimulated with IL-4 (20 ng/mL) for 3 hours and the expression of Ccl17 was determined by qPCR. (J) Schematic presentation of the Ccl17 locus and primers location. The amounts of Pu.1 bound to the Ccl17 in unstimulated or Il-4 (20 ng/mL, 2h) stimulated WT and ATP6V0d2 BMDMs were determined by chromatin immunoprecipitation. (K) WT and Atp6v0d2 BMDMs were stimulated with IL-4 (20 ng/mL) for 48 hours; the amounts of Ccl17 secreted in the supernatants were determined by enzyme-linked immunosorbent assay. Data are representative of 3 independent experiments. Data were assessed by one-way analysis of variance with Turkey's test (F, H, and K) and unpaired Student's t test (A, C and I), and are presented as mean ± standard error of the mean.

BALF, bronchoalveolar lavage fluid; qPCR, quantitative polymerase chain reaction; WT, wild-type; BMDM, bone marrow-derived macrophage; KO, knockout.

*P < 0.05; †P < 0.01; ‡P < 0.001.

ATP6V0d2 expression is inversely correlated with expression of PU.1 and CCL17 and asthma severity in human patients

Next, we investigated whether the ATP6V0d2-PU.1-CCL17 axis is relevant in human asthma. We compared PU.1 and CCL17 mRNA expression in the sputum cells from healthy controls and asthmatic patients and found significantly increased expression of PU.1 and CCL17 in the induced sputum from asthmatic patients, compared to those from healthy controls (Fig. 7A and B). Using a Pearson correlation assay, we found that PU.1 and CCL17 expressions were negatively correlated with ATP6V0d2 expression (Fig. 7C and D), but PU.1 expression was positively correlated with CCL17 expression (Fig. 7E). More importantly, we found that the disease severity measured by the ratio between forced expiratory volume (FEV1) and forced vital capacity (FVC) (FEV1/FVC, less means severer) was inversely correlated with ATP6V0d2 expression (Fig. 7F), suggesting ATP6V0d2 negatively regulates asthma disease severity. In contrast, the disease severity was positively associated with the expression of PU.1 and CCL17 (Fig. 7G and H). Taken together, these data suggest that ATP6V0d2-PU.1-CCL17 axis regulates asthma severity in both humans and mice.

Fig. 7

Data were assessed by one-way ANOVA with Turkey's test (F, H, and K) and unpaired Student's t-test (A, C and I), and are presented as mean ± SEM. Expression in induced sputum from patients was inversely associated with the severity of asthma. (A, B) qRCR analysis of PU.1 and CCL17 expression in sputum samples from healthy subjects (n = 16) and asthma patients (n = 22). (C-E) Pearson correlation assay for expression of ATP6V0d2 versus PU.1. Data were assessed by one-way ANOVA with Turkey's test (F, H, and K) and unpaired Student's t-test (A, C, and I), and are presented as mean ± SEM. CCL17 and PU.1 versus CCL17 in sputum cells from asthma patients. (F-H) Correlation analysis of FEV1/FVC ratio versus expression of ATP6V0d2, PU.1 and CCL17 in sputum cells from asthmatic patients. Data were assessed by Mann-Whitney U test (A, B) and Pearson correlation test (C-G). Data are presented as mean ± SEM.

ANOVA, analysis of variance; SEM, standard error of the mean; FEV1, forced expiratory volume in 1 second; FVC, forced vital capacity.

*P < 0.05; †P < 0.01.

DISCUSSION

In this study, we found that a subunit of the V-type ATPase pump ATP6V0d2 is induced during asthma, but serves as an inhibitor of hyperresponsiveness and inflammation in the airway and lung tissues. ATP6V0d2-deficient mice had enhanced AAM polarization in an allergic asthma model through reduced degradation of Pu.1. More importantly, ATP6V0d2 expression was negatively correlated with PU.1 expression and asthma disease severity in humans.

Cellular compositions in sputum and BALF may vary and be affected by age, atopy, and other factors.2829 Nevertheless, we found that ATP6V0d2 expression was enhanced in the induced sputum of asthmatic patients and in the lung tissues of OVA-sensitized mice. In addition, we could recapitulate the induction of Atp6v0d2 by IL-4 in both BALF-derived AMs and BMDMs.

Pu.1 has been shown to promote Th9 differentiation and AAM differentiation, which exacerbates asthma inflammation.3031 Deletion of Pu.1 alleviated IL-4-mediated AAM polarization and eosinophilic asthmatic inflammation both in vitro and in vivo.14

In addition, Pu.1 also regulates Ccl17 expression, which promotes the recruitment of Th2 and Th9 cells into the upper respiratory tracts and further manifests disease severity.3233 Consistently, we found Il-4-induced Pu.1 directly binds to Ccl17 promoter in murine BMDMs. Recently, it has been shown that LPS can enhance Pu.1 binding to Ccl22 promoter in BMDMs and dendritic cells.16 However, in our study, we could not find enhanced Pu.1 binding to Ccl22 locus upon IL-4 treatment (data not shown), indicating differential chromatin conformations of Ccl22 loci between LPS and IL-4 treatments.

Pu.1 possesses a PEST domain that could be targeted by proteasome degradation; however, truncation of the Pu.1 PEST domain did not lead to protein stabilization suggesting proteasome-independent degradation.3435 In line with this, Pu.1 has been shown to be subject to K63 ubiquitination and degradation through p62-dependent autophagy in Th9 cells.36 Consistent with these, we found that the ATP6V0d2 mediates Pu.1 degradation in a lysosome-dependent manner. We recently showed that ATP6V0d2 mediates degradation of hypoxia-inducible factor (HIF)-2α, but not HIF-1α, in tumor- associated macrophages and regulates leucine-induced mTORC1 activation and M2 polarization in vitro.2122 The molecular mechanisms of how ATP6V0d2 selectively regulates lysosomal degradation of transcriptional factors under different pathological conditions remain unclear, which warrants further biochemical investigations.

Despite the induction of ATP6V0d2 during asthma, the expression of ATP6V0d2 was negatively correlated with the asthma severity. It is likely that ATP6V0d2 acts as a feedback inhibitor for constraining airway hyper-inflammation. Hosts have developed multiple mechanisms for feedback inhibition of inflammation. For example, IL-6 family cytokines induce suppressor of cytokine signaling and IL-1-family cytokines and TLR agonists induce A20, both of which suppress their signaling pathways that initiate their expression.3738

There are at least 3 different types of macrophages including AM, intestinal macrophages and monocyte-derived macrophages in the lung tissues.39 The polarization of macrophages under different disease conditions adds another layer of complexity to the function of macrophages during asthma. Both protective and pathogenic roles of AAM have been reported in the literature.40 Our data indicate a pathogenic role of AAMs in asthma and there is a feedback inhibitory pathway by the Atp6v0d2-Pu.1-Ccl17 axis in regulation of AAMs polarization, which is consistent with previous literature.1441

Atp6v0d2 is not expressed in T cells, B cells, and cancer cells in our study.21 However, we could not completely rule out the possible role of Atp6v0d2 in other myeloid cells in addition to macrophages. Another limitation of the study is that we could not completely differentiate the tissue-resident AMs and Mo-AMs due to the scarce of the cells. Asthma can be also classified as neutrophils-dominant or eosinophils-dominant types.29 It remains unclear how ATP6V0d2 expression regulates the numbers of neutrophils and eosinophils within the respiratory tract and lung tissues. Further studies using sophisticated fate-tracking system of AMs will shed more light on the functions of subpopulations of AMs in the regulation of asthma.

In summary, we uncovered a novel regulatory axis of ATP6V0d2-PU.1-CCL17 in AM polarization, which is associated with human asthma severity.