PDE3 Inhibition Reduces Epithelial Mast Cell Numbers in Allergic Airway Inflammation and Attenuates Degranulation of Basophils and Mast Cells

Epithelial mast cells are generally present in the airways of patients with allergic asthma that are inadequately controlled. Airway mast cells (MCs) are critically involved in allergic airway inflammation and contribute directly to the main symptoms of allergic patients. Phosphodiesterase 3 (PDE3) tailors signaling of cyclic adenosine monophosphate (cAMP) and cyclic guanosine monophosphate (cGMP), which are critical intracellular second messenger molecules in various signaling pathways. This paper investigates the pathophysiological role and disease-modifying effects of PDE3 in mouse bone marrow-derived MCs (bmMCs), human LAD2- and HMC1 mast cell lines, human blood basophils, and peripheral blood-derived primary human MCs (HuMCs). In a chronic house dust mite (HDM)-driven allergic airway inflammation mouse model, we observed that PDE3 deficiency or PDE3 inhibition (PDE3i) therapy reduced the numbers of epithelial MCs, when compared to control mice. Mouse bone marrow-derived MCs (bmMCs) and the human HMC1 and LAD2 cell lines predominantly expressed PDE3B and PDE4A. BmMCs from Pde3−/− mice showed reduced loss of the degranulation marker CD107b compared with wild-type BmMCs, when stimulated in an immunoglobulin E (IgE)-dependent manner. Following both IgE-mediated and substance P-mediated activation, PDE3i-pretreated basophils, LAD2 cells, and HuMCs, showed less degranulation than diluent controls, as measured by surface CD63 expression. MCs lacking PDE3 or treated with the PDE3i enoximone exhibited a lower calcium flux upon stimulation with ionomycine. In conclusion PDE3 plays a critical role in basophil and mast cell degranulation and therefore its inhibition may be a treatment option in allergic disease.


INTRODUCTION
Mast cells (MCs) play a key role in human allergic airway inflammation. The observation of bronchial epithelial MCs in asthmatic patients (Braunstahl et al., 2003) is an indication that the presence and location of MCs are important. Mast cell and basophil dynamics were also observed in the nasal mucosa of allergic rhinitis patients (KleinJan et al., 2000). The findings of elevated numbers of MCs and increased serum tryptase levels, suggesting MC activation, implicate MCs in various lung diseases, including idiopathic pulmonary fibrosis and sarcoidosis (Bargagli et al., 2009;Wygrecka et al., 2013). It has been reported that the presence of pulmonary neuroendocrine cells amplify allergic asthma responses by the release of the neuropeptides calcitonin gene-related peptide (CGRP) and substance P (SP), both of which can act as MC stimulators (Sui et al., 2018).
Localization of MCs is an important pathophysiological feature; in healthy controls they are absent in the epithelium but present in submucosal airways. The presence of MCs in the epithelium or in close proximity to smooth muscle airway cells (Brightling et al., 2002;Bradding et al., 2006;Bradding and Brightling, 2007) contributes to smooth muscle hyperplasia via TGFb and b-tryptase (Woodman et al., 2008). In uncontrolled allergic asthma patients the total number of MCs and MC TC (MC containing tryptase and chymase) in the alveolar parenchyma was found to correlate negatively with FEV1% predicted (Andersson et al., 2011;Andersson et al., 2018). In these patients the numbers of mast cells expressing FceR1 and TGFb are increased. These findings indicate the connection between disease and parenchymal MCs in uncontrolled asthmatics. In addition, the amount of collagen deposition correlates with the number of MCs in the parenchyma (Andersson et al., 2011). In vivo mast cell studies are hampered by the fact that staining for serine proteases is not always easy to interpret because MCs degranulate during allergen challenge; the number of serine protease-positive cells drops, because degranulated cells are not positive anymore (Balzar et al., 2011). Basophil and MC accumulation occurs in the airways after allergen inhalation and/or challenges of allergic patients (Gauvreau et al., 2000;KleinJan et al., 2000;Braunstahl et al., 2003), and in fatal asthma (Perskvist and Edston, 2007;Woodman et al., 2008;Yu et al., 2011). In allergy, mast cell and basophil degranulation is initiated during the early-phase reaction and continues to the late-phase reaction (Togias et al., 1988;Fokkens et al., 1992;de Graaf-in't Veld et al., 1997;KleinJan et al., 2000).
MC activation by immunoglobulin E (IgE)-dependent (i.e., "allergic") or other mechanisms release a diverse spectrum of mediators that induce local effects on blood vessels, nerves, mucous glands, epithelial cells, airway smooth-muscle cells, and immune cells (Bradding et al., 2006). Analyses in chronic asthma mouse models indicated that MCs can contribute to the establishment of chronic eosinophilic airway inflammation (Yu et al., 2011). They also contribute to features of tissue remodeling that resemble those observed in asthma patients, including increased numbers of mucus-secreting goblet cells in the airway epithelium and increased deposition of interstitial collagen (Yu et al., 2011;Li et al., 2019).
In the context of allergic airway inflammation and asthma, phosphodiesterase 3 (PDE3) and PDE4 are widely expressed in airway cells, including epithelial and endothelial cells, smooth muscle cells, and inflammatory cells (Beavo, 1995;Blease et al., 1998;Wright et al., 1998;Myou et al., 1999;Souness et al., 2000;Boswell-Smith et al., 2006). Whereas both PDE3 and PDE4 are cyclic adenosine monophosphate (cAMP)-degrading enzymes, only PDE3 degrades cyclic guanosine monophosphate (cGMP) as well. PDE4 is an asthma-susceptibility gene (Himes et al., 2009). PDE inhibition (PDEi) appears to act effectively on immune cells, as well as on endothelial cells, epithelial cells, and airway smooth muscle cells (KleinJan, 2016). PDE3 deficiency or pharmacological inhibition of PDE3 activity was shown to reduce allergic airway inflammation and to improve airway mucosal barrier function in allergic airway models (KleinJan, 2016). However, the role of PDE3 and the effects of PDE3i on MC and basophil degranulation remained unknown.
This paper investigates the pathophysiological role of PDE3 in MCs and basophils via in vitro studies. Based on the observations that i) PDE3 deficiency or PDE3i reduces allergic airway inflammation (Beute et al., 2018) and that ii) epithelial MCs are induced in a chronic house dust mite (HDM)-driven asthma model (Beute et al., 2018;Li et al., 2019), III) and that elevated cAMP levels have an inhibitory effect on exocytosis (Alm, 1984) we hypothesize that PDE3 activity is important for mast cell function. To determine how PDE3 deficiency or PDE3i affects mast cell function, we analyzed mast cells in a chronic HDM asthma model and investigated in vitro degranulation. We observed that in cell cultures PDE3 inhibitors prevented and/or attenuated stimulation-induced degranulation of bone marrowderived mouse mast cells, human peripheral blood mononuclear cell (PBMC) basophils, LAD2 mast cell lines, and peripheral blood-derived human mast cells.

The Presence of Epithelial Mast Cells Is a Sign of Chronic Allergic Airway Inflammation
Evaluation of PDE mRNA transcripts indicated the expression of PDE3B, PDE4A, PDE4B, and PDE4D in mouse bone marrow-derived MCs. Hardly any PDE3A expression was detected and no PDE4C (Supplementary Figure 1). In earlier studies the presence of airway epithelial MCs was demonstrated in allergic rhinitis and Abbreviations: AAI, allergic airway inflammation; AHR, airway hyperresponsiveness; BAL, broncho-alveolar lavage; cAMP, cyclic adenosine monophosphate; cGMP, cyclic guanosine monophosphate; DC, dendritic cell; HDM, house dust mite; HRP, horseradish peroxidase; MLN, mediastinal lymph node; MC, mast cell; PBS, phosphate buffered saline; PDE, phosphodiesterase; PDE3i, phosphodiesterase 3 inhibitor(s)/inhibition; SP, substance P. allergic asthma patients (KleinJan et al., 2000;Braunstahl et al., 2003) as well as in chronic HDM-driven asthma models (Li et al., 2019). MCs contribute to the development of multiple features of asthma pathophysiology, both in an IgEdependent way and in an IgE-independent neurogenic-driven way (Devos et al., 2016;Sibilano et al., 2016). By toluidine blue staining we investigated the presence of MCs in airway epithelium samples in a chronic HDM-driven asthma model. No epithelial MCs were present in control phosphate buffered saline (PBS) mice, but they did appear following HDM treatment. Degranulated-and therefore activated-MCs were seen in HDM-exposed placebotreated wild-type (WT) mice ( Figure 1A and insert). Mice lacking PDE3 (Pde3 −/− ) or mice that received topical (Supplementary Figure 2) PDE3i therapy showed lower (p < 0.01 and p < 0.005 resp.) epithelial MC numbers and those MCs present were in a more granulated stage ( Figure 1B). Other airway structures, including parenchyma and smooth muscle areas, were negative for mast cells. In conclusion: epithelial mast cell numbers were lower in a HDM-driven allergic airway inflammation mouse model when PDE3 is lacking or when mice were treated with PDE3 inhibitors.

PDE3 Plays a Pathophysiological Role in Mast Cell Degranulation
In vitro-cultured mouse bone marrow-derived MCs (bmMCs) were examined to investigate the role of PDE3 in MC degranulation, using changes in CD107b and CD200r expression as markers. Mast cell expression of surface CD200r was shown to be increased after both IgE-and non-IgE-mediated activation (Larson and Mitre, 2012). Overnight 2,4dinitrophenyl (DNP)-specific IgE-sensitized MCs and PBS control MCs were stimulated with 10 ng/ml DNP-7-OVA for   (Figures 2A-D). Unstimulated WT and PDE3 −/− MCs showed no significant differences in median fluorescence intensity (MFI) for CD107 or CD200r. In conclusion: mouse MCs lacking PDE3 were less activated upon FceR cross-linking.

Clinically Available PDE3 Inhibitors Inhibit Basophil Degranulation
To study the effect of PDE3i on the degranulation of human basophils in the PBMCs fraction, a modified basophildegranulation test was performed. Basophils were defined as FcϵRI + HLA-DR − CD123 + cells, and further analyzed for CD63 and CD203c + expression as degranulation markers. Because CD63 expression is a marker for MC activation (Cop et al., 2018;Taracanova et al., 2018), we determined that, when basophils were activated by FceR cross-linking, the population of CD63 + CD203c + cells increased from~7 to~31% ( Figure  3A). As quantified by the proportions of CD63 + CD203c + cells, anti-IgE-activated basophils showed a diminished degranulation status, when pre-treated with enoximone in a dose-dependent manner ( Figures 3B-D). Attenuation of these degranulation markers was achieved even at a low dose of 4 µM enoximone. Other PDE3i (milrinone 100 µM and cilostazol 100 µM) were able to impair basophil degranulation as well ( Figure 3E).

Mast Cell Line Degranulation Is Impaired When PDE3 Is Targeted
HMC1 and LAD2 mast cell lines express PDE3 and PDE4 transcripts ( Figure 4A). Attenuation of surface degranulation markers in LAD2 cells was observed only if we treated the cells with higher doses of enoximone (Supplementary Figure 3A).
For these experiments, sera obtained from HDM mono-allergic (68-121 KU/l) or grass pollen (GP) mono-allergic (123-211 KU/ l) donors were used. Overnight passively sensitized LAD2 cells with sera (controls: 20% serum or PBS) were stimulated with PBS or HDM. When the cells were pretreated with enoximone in a dose-dependent manner, followed by HDM stimulation (1 ng/ ml), the CD63-and CD203c-expression levels on the surface of LAD2 ( Figure 4B) cells were lower. Control LAD2 cells sensitized with serum IgE obtained from isolated GP-allergic subjects showed no degranulation upon HDM stimulation ( Figure 4C). The marker CD203c presented less sensitive than CD63 in MCs ( Figure 4C).  Figure  3B). Treatment using cGMP (1 mM) analogue and stimulation with SP showed no effect on CD63 expression (Supplementary Figure 3C). One of the downstream targets of the cAMP pathway is protein kinase A (PKA). To investigate the molecular mechanism of PDE3 inhibition in LAD2 cells, we stimulated LAD2 cells that were pretreated with PKA inhibitor (Rp)-8-Br-cAMP (100 µM) with SP, and observed a significant reduction of CD63 expression. This suggests that the LAD2-cell modulation of CD63 expression is regulated, at least in part, via the cAMP-PKA pathway. Next, the Ca 2+ flux in MCs lacking PDE3 or treated with PDE3i was investigated. Antigenic stimulation and neurogenic stimulation were bypassed through the use of ionomycin, a calcium ionophore that opens all Ca 2+ channels in a nonspecific way. Ionomycin induces Ca 2+ flux in BmMCs; it turned out that PDE3-deficient BmMCs showed a lower Ca 2+ flux than control WT cells (Supplementary Figure 3E). Likewise, when WT BmMCs were pretreated with enoximone (20 µM), lower Ca 2+ flux levels were reached (Supplementary Figure 3F). These data suggest that mast cell degranulation is, at least partly, regulated in a cAMP-Ca 2+ dependent manner.

PDE3 Inhibition Inhibits Primary Human Mast Cell Degranulation
To translate our findings to a more physiologically relevant model, we investigated the effects of PDE3 inhibition in primary human MCs. Peripheral blood-derived primary human MCs (HuMCs) were subjected to flow cytometric analysis ( Figure 5A), notwithstanding the relatively high autofluorescence ( Figure 5A left panel), CD117 ( Figure 5A middle panel), and CD63 ( Figure 5A right panel). Overnight passive sensitized (20% serum) HuMCs, with sera obtained from HDM mono-allergic or grass pollen (GP) mono-allergic donors were subjected to flow cytometric analysis ( Figure 5B) after having treated with enoximone or diluent and stimulated with PBS ( Figure 5B left panel), HDM ( Figure 5B middle panel), or Ionomycine ( Figure 5B right panel). The CD63 expression levels of HuMCs changed when the cells were passively sensitized with serum obtained from HDM-allergic donors ( Figure 5C left panel) but not GP mono-allergic donors ( Figure 5C right panel), followed by HDM stimulation (60 ng/ml), illustrating that only specific IgE cross-linking induces MC degranulation. Quantification indicated that HuMCs can, after passive sensitization, be used to measure specific FceR cross-linkingdependent MC degranulation and ionomycine-dependent MC degranulation. CD63 expression was reduced when HuMCs were treated with enoximone 20 µM and subsequently PBS stimulated ( Figure 5D left panel). MC degranulation induced by HDM ( Figure  5D

DISCUSSION
This paper describe that MCs lacking PDE3, as well as LAD2 cells, HuMCs, and human basophils being treated with PDE3i, showed restrained degranulation. High intracellular levels of cAMP are associated with inhibition of MCs and basophil degranulation (Alm, 1984). Transgenic mouse MCs lacking PDE3 demonstrated that MCs without PDE3 react the same as those treated with PDE3i. We used clinically relevant PDE3i to investigate human basophils and human MCs illustrating the role of PDE3 in the degranulation of these cells in allergic airway inflammation.
MC degranulation is a process involving PDE3 and cAMP-PKA. Having investigated three different types of cells in addition to mouse MCs that lack PDE3 activity, all tests confirm the pathophysiological role of PDE3 in MCs/basophils. Our data show that inhibitors of cAMP-degrading enzymes (PDE3) or increased intracellular levels of dibutyryl cAMP (db-cAMP) prevent MC degranulation; this is consistent with the observation by Serra-Pages et al., showing that, when prostaglandin E 2 (EP2)-dependent activation induces cAMP, cAMP/PKA suppresses IgE-/antigen-dependent signals and calcium influx, which is essential for MC degranulation (Serra-Pages et al., 2012). Shichijo et al. observed that cAMP-elevating agents, including PDE3i, inhibit cytokine production of MCs. Unfortunately, the concentrations used (up to 10 µM) were relatively low and were unable to reduce chemical mast cell mediators (Shichijo et al., 1999). PDE inhibition elevated cAMP and prevented MC activation (Shichijo et al., 1999). Measuring CD63 expression or the production of IL1b are accepted methods to determine SP-induced MC activation (Cop et al., 2018;Taracanova et al., 2018). Mast cell activation was measured in relation to CD63 expression. Cell surface CD63 is a marker associated with membranes of intracellular vesicles which is frequently used in basophil activation tests. In the current study cells were treated for 30-60 min with PDEi and subsequently stimulated as well for 30-60 min, which is longer than in previous studies that failed to see any PDE3-inhibitor effect. In addition, dose optimization appeared essential for every specific PDEi in order to find the best effective inhibitory dose in the bioassay. In general 10 µM is taken as a standard concentration for PDEi; however, this study illustrates that 10 µM is not always sufficient. We noticed that high concentrations (1 mM) of enoximone did not affect the cell viability. Sole enoximone treatment without stimulation had no effect on the expression of CD63 and CD203c in LAD2 cells. The initial effect of PDE3 inhibition is smooth muscle relaxation in humans. This effect is already noticeable within 30 s after IV treatment (Beute, 2014;Sobhy et al., 2019). In smooth muscle cells, increased cAMP inhibits myosin light chain kinase (MLCK), preventing Ca 2+ binding, thereby reducing MLC phosphorylation, and hence relaxing the vascular smooth muscle (Prakash, 2016). This mechanism is not unique for smooth muscle cells (Prakash, 2016) or endothelial cells (Rigor et al., 2013) and could also offer an explanation on how, in a cAMP-dependent manner, MCs can be prevented to degranulate, with a role for calmodulin. PDE3 seems to be involved in truncation of FcϵRIb and mediates Ca 2+ -dependent microtubule formation (Cruse et al., 2013), which prevents degranulation and cytokine release. The data in this paper indicate that the release of Ca 2+ via low concentrations of ionomycin can be in part abrogated by PDE3 inhibition. This suggests that cAMP directly prevents binding of Ca 2+ , which in turn prevents microtubule formation and MC degranulation.
The reorganization of the actin cytoskeleton is under the spatial control of Ca 2+ cAMP-PKA signaling in smooth muscle cells (Hocking et al., 2013). PDE3 plays a pathophysiological role in allergic airway inflammation in a translational model (Beute et al., 2018). The present study shows that targeting PDE3 has an inhibitory effect on exocytosis of MCs and can be seen as a valid bonus treatment in allergic asthma-related disease. PDE3 inhibitors are known to induce, within a few seconds, filamentous (F)-actin changes of the cytoskeleton in epithelial and endothelial cells (Liu et al., 2012), leading to enhanced barrier function. Calcium is important for MC degranulation and crucial for the granule migration in the cells, which is in counterbalance with cAMP (Draber et al., 2011). Ionomycin opens Ca 2+ channels, probably via the microtubules and or actin cytoskeleton, leading to MC degranulation and activation. That this process can be inhibited in MCs by PDE3 inhibition is novel; how exactly this inhibition occurs is yet unknown.
Microdomains containing PDE3, PDE4, and adenylate cyclase (AC) balance each other out; using either PDE3-or PDE4-inhibitors, the net outcome might be in favor of immune attenuation. Others suggest that PDE activity is sufficient to restrict cAMP diffusion to specifically defined subcellular compartments in certain cell types (Terrin et al., 2006;Oliveira et al., 2010). These microdomains are formed by cytoskeleton and might constitute an efficient barrier to cAMP diffusion, and promote compartmentalized cAMP and PKA signals by allowing cAMP to accumulate locally to a level sufficient to activate PKA. This would prevent activation of PKA in the other locations of the cytosol (Rich et al., 2000). The actin cytoskeleton would act as a diffusional barrier for cAMP, and PDEs would be in control of local cAMP levels in thus created microdomains (Monterisi et al., 2012). Increased intracellular cAMP levels might account for the decreased reactivity of basophils to venom allergen (Bauer et al., 2007).
An important technique to illuminate this black box might be measuring phosphorylation of cell signaling molecules such as Syk Lyn, PLCg, and PIP 2 , since PIP 2 via activated PLCg leads to IP 3 and DAG. IP3 stimulates the IP 3 R on the ER, leading to the release of Ca 2+ . IP3R are phosphorylated by cAMP-dependent protein kinase (PKA) (Taylor, 2017).
The anti-inflammatory effects of PDE3 inhibitors have been established in other studies, and are illustrated by reduced serum TNFa levels during PDE3 inhibition, when compared to dobutamine-treated septic patients (Kern et al., 2001) and reduction of macrophages as well as CD11b expression on eosinophils (Beute et al., 2018). In alveolar epithelial cells LPSinduced biosynthesis of pro-inflammatory cytokines is regulated by cAMP and tightly controlled by PDEs, and can be reduced by PDE inhibitors (Haddad et al., 2002).
Allergen-driven degranulation of mast cells and basophils is one of the first steps in the allergy cascade resulting in local inflammation. In the context of allergic airway inflammation therapy the first choice to relieve symptoms is targeting basophil and/or MC activation, or blocking the effect of mast cell mediators. PDE4 inhibitors like roflumilast have shown to be beneficial in the context of allergic inflammation; however, because of the serious side effects they are not available for allergy related diseases (Gupta, 2012). PDE3 inhibitors are accompanied with less side effects (Metra et al., 2009) and are in use as therapeutics for intermittent claudication (Barnett et al., 2004). Clinical observations indicate immediate bronchodilation in asthma patients and animal models when treated with PDE3 inhibitors enoximone and milrinone (Howell et al., 1995;Fujimura et al., 1997;Myou et al., 1999).
Enoximone and its metabolites have the advantage over other PDE3/PDE4 inhibitors in that they have been clinically used for over 20 years, peri-operatively and in the ICU, to prevent organ failure, as well as in having minimal side effects. Using low doses of a clinically available PDE3 inhibitor like enoximone makes it possible to avoid the serious side effects that were often seen when asthma patients were treated with the very potent PDE4 inhibitors (Gupta, 2012). Investigational use of the PDE3 inhibitor enoximone shows improvement as well as attenuated cell-driven symptomatology in asthmatic and allergic patients (Beute unpublished). Clinical trials are underway to confirm these observation in asthma patients. PDE3 inhibition is not only a smooth muscle relaxant being beneficial in obstructive lung disease (Leeman et al., 1987), but also improve mucosal barrier function (Kern et al., 2001) and is in use as a rescue medication in severe asthma (Leeman et al., 1987;Kern et al., 2001;Beute, 2014;Sobhy et al., 2019). In conclusion, the present work has established that the small molecule PDE3 inhibitor enoximone is important in preventing basophil and mast cell degranulation. PDE3 is a clinically relevant therapeutic target in the context of allergic inflammation.

Animals
Pde3a −/− and pde3b −/− C57bl/6 mice were previously described (Masciarelli et al., 2004;Choi et al., 2006) and showed no abnormalities regarding T-cell development (data not shown). WT C57bl/6 mice were bought from Charles River. Animals were kept under specific pathogen-free conditions, provided with water and food ad libitum and were used at the age of 6-12 weeks.

Cell Lines and Serum
We took LAD2 cells, kindly provided by Kirshenbaum and Metcalfe (National Institute of Allergy and Infectious Diseases, Bethesda, Maryland), and HMC1 cells, a gift from Dr. J.H. Butterfield (Mayo Clinic, Rochester, MN, USA) as a surrogate for MCs. The human MC line HMC1 was kindly provided by Dr. J.H. Butterfield (Mayo Clinic, Rochester, MN, USA). LAD2 is a stem-cell factor (SCF) dependent human MC line (kindly provided by Kirshenbaum and Metcalfe, National Institute of Allergy and Infectious Diseases, Bethesda, Maryland). Serum obtained from mono-allergic patients containing specific IgE for either house dust mite (62 -> 100 kU/l) or GrassPollen (68 kU/l) or (Birch >100 kU/l) was used for passive sensitization.

Study Approval
Peripheral blood cells were isolated from buffy coats of blood donors (Sanquin Bloedvoorziening, NVT nr. 0014.03, Rotterdam, the Netherlands). Serum samples obtained from HDM and GP mono-allergic patients were initially collected during routine diagnostic analysis in the Laboratory Medical Immunology (Department of Immunology) of the Erasmus MC, Rotterdam, Netherlands, and remaining material was made available for this study. The Erasmus MC Medical Ethics Committee approved this availability without the need for informed consent, based on the use of fully anonymized remaining diagnostic material without the patient being subjected to additional risk or medical procedures. All animal experiments were approved by the Erasmus MC's Animal Ethics Committee.

House Dust Mite Allergic Airway Inflammation Model
To induce chronic airway inflammation, mice were anaesthetized with isoflurane and treated intratracheally (i.t.) with 25 mg house dust mite (HDM; Greer Laboratories) allergen extract in 50 µl phosphate buffered saline (PBS) on day 0 or using PBS only (KleinJan et al., 2014;Vroman et al., 2017)

Therapeutic Intervention by Intratracheal Topical Application of Enoximone
Optimal dose of 25 µg enoximone versus diluent (Beute et al., 2018) was administered with 25 µg HDM solution and used for intratracheal treatment, or diluent (10% ethanol + 40% propylene glycol + 50% Milli-Q pH 12) was administered with 25 µg HDM solution. Controls were treated with enoximone 25 µg administered with PBS or diluent administered with PBS. Mice underwent enoximone treatment during the last 2 weeks of the challenge period; see Supplementary Figure 1.

Histochemical Detection of Mast Cells
Toluidine blue, an aniline dye, stains mas t cells metachromatically. Snap frozen lung tissue cryosections (6 µm) were histochemically stained with toluidine blue (0.5% in HCL 0.5 M) at pH 0.5 for at least 5 min and observed and quantified immediately (KleinJan et al., 1996). Quantification was performed by counting the number of toluidine blue positive cells per mm length basement membrane in the epithelium of bronchioles of the airways, using an Axioskop 20 microscope (Zeiss, Jena Germany) with an eyepiece graticule at a magnification of 200 x (Beute et al., 2018)

Mouse Bone Marrow-Derived Mast Cells
Bone marrow (BM) cells were isolated from tibias and femurs of WT and KO mice. The bones were crushed, cells were filtered (100 µm) and washed. Red blood cells were lysed with lysis buffer. The cells were cultured in RPMI 1640 medium (Life Technologies) supplemented with 10 µg/ml interleukin 3 (IL-3), 10% heat inactivated fetal calf serum, 50 µg/ml gentamycin, 5*10-5 M b-mercaptoethanol at 37°C in 5% CO2 for 4-6 weeks. By week 4, 95-98% of the cells must be MCs (Cop et al., 2018). To confirm, the morphology of the bone marrow-derived MCs (BMMCs) was assessed by toluidine blue staining to observe mast cell granules.

Analysis of Phosphodiesterase Expression
RNA was isolated from mouse bone marrow-derived MCs and LAD2 cells using a Total RNA Purification Kit (Sigma Aldrich). Purity of the extracted RNA was verified using a Nanodrop 2000 Spectrophotometer (Thermo Fisher Scientific). From the RNA, complementary DNA (cDNA) was reverse transcribed using a Revert Aid First Strand cDNA Synthesis Kit (Thermo Fisher Scientific). Quantitative PCR (qPCR) primers (for detailed description of primer sequences see (Liu et al., 2012) specific for all PDE3 and -4 isoforms were designed using the universal probe library (Roche Diagnostics), and checked for efficiency. Using SYBR Select Master Mix (Applied Biosystems), the cDNA was analyzed via qPCR, using a 7300 Real-Time PCR System (Applied Biosystems). Resulting data were normalized to relative expression of household genes [b-actin in human and hypoxanthine-guanine phosphoribosyl transferase (HPRT) in mouse]. Resulting data were further visualized using the comparative Ct (2 −DCt ) method (Wong and Medrano, 2005).

Statistics
Reported values are shown as mean ± SEM. Statistical analyses were performed with SPSS (SPSS Inc., Chicago, IL) using Kruskal-Wallis one-way ANOVA, followed by Mann-Whitney U-test. A Mann-Whitney U-test was only performed when the one-way ANOVA test pointed to significance. A Wilcoxon signrank test was used for paired samples. Resulting p values <0.05 were considered significant. Tests that did not reach significance (p > 0.05) are not indicated.

DATA AVAILABILITY STATEMENT
All datasets generated for this study are included in the article/ Supplementary Material.

ETHICS STATEMENT
The Erasmus MC Medical Ethics Committee approved this availability without the need for informed consent, based on the use of fully anonymized remaining diagnostic material without the patient being subjected to additional risk or medical procedures. All animal experiments were approved by the Erasmus MCs Animal Ethics Committee.

AUTHOR CONTRIBUTIONS
AK and JB contributed to study design and wrote and edited the manuscript. KG, HN, RH, VB, AK and JF contributed to performing the studies and obtaining the samples analyzed in this study. MS, SH, and RWH were involved in reviewing and shaping the manuscript.

SUPPLEMENTARY MATERIAL
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fphar.2020. 00470/full#supplementary-material SUPPLEMENTARY FIGURE 1 | PDE3 and PDE4 are expressed in mice mast cells. Expression of PDE3 and PDE4 isoforms in bone marrow-derived mice mast cells.
SUPPLEMENTARY FIGURE 2 | Experimental design.Experimental chronic HDM-driven asthma design, showing intratracheal exposure to 25 mg of HDM or PBS that was performed three times per week, as indicated by arrows. From 5 weeks onwards HDM/saline was admixed with placebo (diluent) or 25 mg of enoximone. Analyses were performed at day 41, one day after the last challenge.
SUPPLEMENTARY FIGURE 3 | Dose-response of cAMP and cGMP analogs on CD63 expression by SP-activated LAD2 cells. (A-D) CD63 expression of LAD2 cells pre-treated for 30 min with increasing doses of enoximone (A), a cAMP analog (B), a cGMP analog (C) and the PKA inhibitor (Rp)-8-Br-cAMP (D), followed by stimulation with 1 mM substance P (SP) for 30 min. Data are shown as mean values ± SEM. A Mann-Whitney U test was used; * P < 0.05. (E,F) Ca2+ flux experiment with bmMCs obtained from Pde3 -/mice and WT (E) and from WT treated with PDE3i enoximone (20µM) or diluent (F). Data are shown of one representative experiment from three independent experiments.