© 2003 American Thoracic Society DOI: 10.1165/rcmb.4853 Resolution of Airway Inflammation following Ovalbumin InhalationComparison of ISS DNA and CorticosteroidsDepartment of Medicine, University of California-San Diego, La Jolla, California Address correspondence to: Dr. David H. Broide, M.B.Ch.B., University of California-San Diego, Basic Science Building, Room 5090, 9500 Gilman Drive, La Jolla, CA 92093-0635. E-mail: dbroide{at}ucsd.edu
In this study we have compared the therapeutic effect of the administration of immunostimulatory DNA sequences (ISS) with that of corticosteroids on the resolution of airway inflammation and airway hyperreactivity (AHR) in a mouse model. Mice which had already developed significant levels of eosinophilic airway inflammation 24 h after allergen challenge were then treated with either ISS or corticosteroids, and the effect on AHR and airway inflammation assessed 6 d later. ISS inhibited AHR as effectively as corticosteroids. Combination therapy with ISS and corticosteroids was more effective than monotherapy with either ISS or corticosteroids in inhibiting AHR. In ovalbumin-challenged mice, levels of bronchoalveolar lavage (BAL) eosinophils were significantly reduced with either ISS or corticosteroids. ISS induced significant levels of BAL interferon- , whereas corticosteroids did not induce expression of BAL interferon- . Both ISS and corticosteroids significantly reduced levels of interleukin-5 in BAL, as well as the number of Periodic Acid Schiffpositive airway epithelial cells. Corticosteroids, but not ISS, increased the number of eosinophils in regional mediastinal lymph nodes. Very few apoptotic peribronchial cells were noted following ovalbumin challenge as assessed by TUNEL assay. Corticosteroids, but not ISS, induced an increase in the small number of apoptotic peribronchial cells. The mechanism by which either ISS or corticosteroids inhibit AHR is likely to be mediated by distinct and shared cellular pathways. The combination of the shared and distinct anti-inflammatory pathways may account for the additive effect of ISS and corticosteroids on inhibiting AHR.
Abbreviations: airway hyperresponsiveness, AHR bronchoalveolar lavage, BAL BAL fluid, BALF dynamic compliance, Cdyn diaminobenzidine, DAB interferon, IFN interleukin, IL immunostimulatory DNA sequence, ISS methacholine, MCh ovalbumin, OVA Periodic Acid Schiff, PAS Toll-like receptor 9, TLR-9
Allergic asthma is characterized by Th2 lymphocyte activation and infiltration of the airways with eosinophils following the inhalation of antigen. Th2 cytokines such as interleukin (IL)-5 promote airway eosinophilia through several different mechanisms, including increased bone marrow production of eosinophils and enhanced eosinophil tissue survival (1, 2). In addition to IL-5, several other Th2 cytokines, including IL-4, IL-9, and IL-13, play an important role in allergic inflammation through their effects on IgE synthesis, mast cell proliferation, endothelial adhesion molecule expression, mucus secretion, and airway responsiveness (3). Thus, it has become increasingly clear that the inflammatory cascade orchestrated by Th2 lymphocytes plays a pivotal role in the pathogenesis and propagation of allergic inflammation in asthma.
As targeting a single Th2 cytokine may not be as effective a therapeutic strategy as targeting multiple Th2 cytokines, we have investigated the ability of immunostimulatory DNA sequences (ISS) to globally inhibit Th2 responses in asthma. Our (46) and other (7, 8) laboratories have demonstrated that ISS containing CpG-rich motifs administered after allergen sensitization but before inhalation allergen challenge, inhibit Th2 responses and airway hyperreactivity (AHR) in mouse models of asthma. ISS inhibits IL-3 (4), IL-4 (7), IL-5 (4), and granulocyte macrophagecolony-stimulating factor (4) production by Th2 cells. In addition to inhibiting Th2 responses, ISS activates the innate immune response to release cytokines including interferon (IFN)- Whereas previous studies have demonstrated an important role for ISS as preventive therapy for asthma, in this study we were interested to compare how effective ISS therapy might be compared with systemic corticosteroids in the treatment of an acute exacerbation of asthma in a mouse model. We designed our current study to compare the therapeutic efficacy of ISS with that of systemic corticosteroids, the current standard therapy for acute exacerbations of asthma. As the mechanism by which inflammation resolves following an acute episode of asthma is incompletely understood, we also investigated the effects of ISS and corticosteroids on eosinophil trafficking into the lung, eosinophil clearance from the lung to regional lymph nodes, and apoptosis of peribronchial airway cells.
Oligonucleotides Endotoxin-free (< 1 ng/mg DNA) phosphorothioate ISS-ODN (5'-TGACTGTGAACGTTCGAGATGA-3') (Trilink, San Diego, CA) synthesized as previously described (4), were used in the in vivo experiments described below.
Animals
Ovalbumin Sensitization and Allergen Challenge
Therapeutic Intervention with ISS or Corticosteroids In comparing the effect of ISS with that of corticosteroids on allergic airway inflammation, we administered a single systemic dose of the long-acting corticosteroid dexamethasone (0.5 mg/kg, the equivalent of 35 mg dexamethasone for a 70-kg patient) or a single 100-µg systemic dose of ISS 24 h after intranasal allergen exposures in sensitized mice (protocol summarized in Figure 1). One day after the final allergen challenge, different groups of mice received systemic therapy with either ISS (100 µg intraperitoneally), corticosteroids (dexamethasone, 0.5 mg/kg intrapeitoneally), the combination (ISS and corticosteroids), or placebo therapy (diluent). We chose the dose of ISS and dexamethasone used in this study based on pilot dose response studies with ISS and dexamethasone. These studies demonstrated that administration of ISS (100 µg) or dexamethasone (0.5 mg/kg) induced maximal inhibition of airway eosinophil responses, with no additional effect noted with higher doses (data not shown). We therefore chose these doses of ISS and dexamethasone for our studies, as we wished to compare the effect of maximal inhibitory responses induced by each therapeutic agent.
We chose to analyze the effects of the different treatment modalities 6 d after therapy was instituted, a time point when untreated mice that are sensitized and challenged still have significant levels of bronchoalveolar lavage fluid (BALF) eosinophilia (
Determination of Airway Responsiveness to MCh In Vivo Noninvasive plethysmography. The enhanced pause (Penh), a dimensionless value which correlates well with pulmonary resistance measured by conventional two-chamber plethysmography in ventilated mice (17), was used to monitor airway responsiveness. In the plethysmograph, mice were exposed for 3 min to nebulized PBS to establish baseline Penh values, and were subsequently exposed to increasing concentrations of nebulized MCh (Sigma) in PBS using an Aerosonic ultrasonic nebulizer (DeVilbiss, Somerset, PA). Following each nebulization, recordings were taken for 3 min. The Penh values measured during each 3-min sequence were averaged and are expressed for each MCh concentration as the percentage of baseline Penh values following PBS exposure (17). Measurement of dynamic compliance. Mouse intubation and ventilation. The mice were anesthetized with pentobarbital sodium (6070 mg/kg intraperitoneally). An anterior, midcervical skin incision was made to expose the trachea, which was cannulated with an 18-gauge blunt needle. Mice were ventilated (Mouse Ventilator, model 683; Harvard Apparatus, Holliston, MA) initially with a tidal volume (VT) of 56 ml/kg at 120 breaths/min. The tidal volume was then adjusted and set to just prevent spontaneous respiration. Peak inspiratory pressure (PI) and peak expiratory pressure (PE) were recorded using Validyne pressure transducers.
MCh challenge.
Dynamic compliance measurement.
BAL, Lung, Bone Marrow, and Blood Eosinophil Counts BALF. The tracheas of the sacrificed mice were surgically exposed and cannulated with 27-gauge silicon tubing attached to a 23-gauge needle on a 1-ml tuberculin syringe. A quantity of 600 µl of sterile PBS was instilled through the trachea into the lung and withdrawn. The BALF was cytospun (3 min at 500 rpm) onto microscope slides and stained with Wright-Giemsa. The percentage of BALF eosinophils was obtained by counting 400 leukocytes on randomly selected portions of the slide by light microscopy (x40 magnification). BALF total white blood cell counts were performed using a hemacytometer. Lung. Lung tissues embedded in TissueTek (O.C.T. Compound; Sakura, Torrance, CA) in 10 x 50 x 50 mm tissue wells were cryosectioned at 5 µ and acetone-fixed onto poly(L-lysine)-coated slides. Diaminobenzidine (DAB) staining was used to detect eosinophil peroxidase, and total eosinophil numbers were enumerated by light microscopy, as described in this laboratory (4). Slides were rinsed in PBS and incubated for 8 min with the peroxidase substrate DAB (Vector Laboratories, Burlingame, CA) and counterstained with hematoxylin, air dried, and examined by light microscopy (x40 magnification). Five random fields were selected and eosinophils were counted to quantitate the number of eosinophils per mm2. Peripheral blood. Blood was collected from the carotid and subclavian arteries. Red blood cells were lysed using a 1:10 solution of 100 mM potassium carbonate, 1.5 M ammonium chloride. The remaining cells were cytospun (3 min at 500 rpm) onto microscope slides and air dried before staining with Wright-Giemsa. Eosinophil counts were performed as described above. Total white blood cell counts were performed using an automated hematology analyzer (Cell-Dyn 4000; Abbot Diagnostics, Abbot Park, IL). Bone marrow. Bone marrow cells were flushed from femurs with 1 ml PBS and cytospun onto microscope slides before staining with Wright-Giemsa. Eosinophil counts were performed as described above. Total bone marrow leukocyte counts were obtained by flushing all of the marrow from a standardized 1-cm length of femur in each mouse with 1 ml of PBS. Manual total white blood cell counts were subsequently performed using a hemacytometer. Lymph nodes. Lymph nodes were dissected from the mediastinum in each mouse. The lymph nodes were embedded in OCT in 10 x 50 x 50 mm tissue wells and were cryosectioned and fixed onto poly(L-lysine)-coated slides. DAB staining was used to detect eosinophils in the lymph node cryosections. Light microscopy examination was performed at x40 magnification. Five random fields were selected and eosinophils were counted to determine the total eosinophil number per mm2.
Periodic Acid Schiff Staining of Airway Mucus Cells
Apoptosis Assays
Cytokine Assays
Statistical Analysis
Resolution of Airway Eosinophilic Inflammation following Allergen Challenge To determine the natural time course of the resolution of airway eosinophilic inflammation following OVA sensitization and allergen challenge, mice were sacrificed at different time points after the final allergen challenge, and the level of BAL eosinophilia was determined. The level of BAL eosinophilia peaked 2 d after the final allergen challenge (83.6 ± 0.7%) and was significantly greater than levels noted in unchallenged mice (< 1%). Six days after the final allergen challenge, the level of BAL eosinophilia was 49.6 ± 3.1%, 10 d after the final allergen challenge the level of BAL eosinophilia was 22.0 ± 3.5%, and 15 d after the final allergen challenge the level of BAL eosinophilia was 4.9 ± 1.2%. To determine the effect of therapy on the resolution of BAL eosinophila we thus chose a time point (6 d after institution of treatment with ISS or corticosteroids) during which there were still significant levels of BAL eosinophilia ( 4050%) to assess the influence of therapy.
ISS Administered as Therapy following Allergen-Induced Airway Inflammation Inhibited Subsequent Development of Airway Hyperresponsiveness
Dexamethasone also significantly inhibited AHR to Mch to a similar degree, as noted with ISS therapy at Mch 24 mg/ml (dexamethasone versus untreated: Mch 24 mg/ml, P = 0.004) (n = 2326 mice/group) (Figure 2). However, at a concentration of 48 mg/ml of Mch, dexamethasone reduced AHR, but this was not statistically significant (dexamethasone versus untreated: Mch 48 mg/ml, P = 0.11). Combination therapy with dexamethasone and ISS had an additive effect on the inhibition of AHR compared with monotherapy with dexamethasone alone (ISS and dexamethasone versus dexamethasone: Mch 48 mg/ml, P = <0.0001) (n = 2126 mice/group) (Figure 2). In selected experiments, we performed invasive measures of changes in dynamic compliance (Cdyn) in response to nebulized MCh in intubated ventilated mice. Mean baseline levels of Cdyn before MCh challenge were not significantly different in the OVA-challenged mice (0.0298 ± 0.0059 ml/cm H2O) compared with the OVA + ISS (0.0275 ± 0.0056 ml/cm H2O), or the OVA + dexamethasonetreated mice (0.0276 ± 0.0017 ml/cm H2O). Measurement of Cdyn in OVA-challenged mice after MCh challenge (0.0109 ± 0.0023 ml/cm H2O) (66% reduction in Cdyn) demonstrated that both ISS (0.0160 ± 0.0026 ml/cm H2O, OVA versus OVA + ISS; P = 0.05) and corticosteroid therapy (0.0165 ± 0.0017 ml/cm H2O, OVA versus OVA + dexamethasone; P = 0.05) improved Cdyn measurements.
ISS Administered as Treatment for Airway Inflammation Inhibited BAL Eosinophilia as Effectively as Dexamethasone
Dexamethasone, like ISS, significantly reduced the absolute number of airway eosinophils (39.0 ± 3.1 x 103 BAL eosinophils, n = 18 mice; P = 0.0002 versus no therapy 402.4 ± 158.0 x 103 BAL eosinophils). Combination therapy with ISS and dexamethasone led to a further reduction in the absolute number of airway eosinophils (24.1 ± 3.1 x 103 BAL eosinophils, n = 17 mice; P < 0.0001), which represented a statistically significant decrease compared with monotherapy with dexamethasone (P = 0.005) but not compared with ISS monotherapy (P = ns). However, as monotherapy with either ISS or dexamethasone inhibited airway eosinophilia by 8090%, the effect of combination therapy on airway eosinophilia was not likely to be significantly better than monotherapy. Therefore, to determine whether the combination of ISS and corticosteroids had an additive effect on inhibiting eosinophilic airway inflammation, we administered a suboptimal dose of ISS (10 µg) and a suboptimal dose of dexamethasone (0.005 mg/kg) alone or in combination to different groups of mice. These studies demonstrated that a suboptimal dose of ISS (10 µg) inhibited BAL eosinophilia by 42% (P = 0.05 versus no therapy, n = 8), and the suboptimal dose of dexamethasone (0.005 mg/kg) inhibited BAL eosinophila by 69% (P = 0.05 versus no therapy, n = 8), whereas the combination inhibited BAL eosinophilia by 73% (P = 0.05 versus no therapy, n = 8). The combination of the suboptimal doses of ISS and dexamethasone was therefore not more effective in inhibiting eosinophilic inflammation compared with a suboptimal dose of dexamethasone alone. The combination of the suboptimal doses of ISS and dexamethasone also inhibited eosinophilic inflammation less (73%) compared with an optimal dose of either dexamethasone or ISS alone (> 90% inhibition).
ISS Administered as Therapy for Airway Inflammation Inhibited Lung Eosinophilia as Effectively as Dexamethasone Dexamethasone, like ISS, significantly reduced the absolute number of lung eosinophils (330.6 ± 53.5 eosinophils/mm2, n = 16 mice; P = 0.0006 versus no therapy, 605.1 ± 41.9 eosinophils/mm2) (Figure 3B). Combination therapy with ISS and dexamethasone led to a further reduction in the absolute number of lung eosinophils (213.8 ± 26.2 eosinophils/mm2, n = 18 mice), which represented a statistically significant decrease compared with monotherapy with dexamethasone (P < 0.05).
ISS Administered as Treatment for Allergic Inflammation Inhibited Bone Marrow and Peripheral Blood Eosinophilia More Effectively than Dexamethasone The reduction in bone marrow eosinophil counts induced by therapy with either ISS or dexamethasone was reflected in reduced peripheral blood eosinophil counts. Compared with the no-treatment group (518 ± 63 eosinophils/µl, n = 7), ISS reduced peripheral blood eosinophilia by 61% (201 ± 16 eosinophils/µl, n = 10; P = 0.001), whereas dexamethasone reduced peripheral blood eosinophilia by a statistically insignificant 20% (412 ± 43 eosinophils/µl, n = 11; P = ns), and combination therapy with ISS and dexamethasone reduced peripheral blood eosinophilia by 60% (206 ± 26 eosinophils/µl, n = 11; P = 0.0006), showing the same effect as ISS monotherapy.
ISS Administered as Therapy for Airway Inflammation Inhibits Generation of IL-5 and Induces Production of IFN-
IFN- was not detected in the BAL of OVA-challenged mice (BAL IFN- , < 9 pg/ml, n = 11). ISS induced significant levels of IFN- (BAL IFN- , 63.1 ± 23.5 pg/ml, n = 12; P = 0.03 versus no therapy) (Figure 4B). In contrast, dexamethasone did not induce IFN- production. Combination therapy with ISS and dexamethasone induced less IFN- (BAL IFN- , 40.2 ± 17.3 pg/ml, n = 12) than ISS therapy alone, but this was not statistically significant.
Dexamethasone Therapy Increases Regional Lymph Node Eosinophilia
Dexamethasone Therapy Increases Apoptosis in the Peribronchial Region and Interstitial Space To investigate whether ISS or dexamethasone may have reduced lung and airway inflammation through increased apoptosis, the level of peribronchial cellular apoptosis was assessed using a TUNEL assay (Figure 6A). In the absence of therapy, the number of apoptotic cells was increased in OVA-challenged compared with unchallenged mice in the peribronchial region (4.0 ± 0.8, versus 1.5 ± 0.3 apoptotic cells/mm2, n = 11; P = 0.001) (Figure 6B) and interstitial compartments (3.4 ± 0.7 versus 0.9 ± 0.2 apoptotic cells/mm2, n = 11; P = 0.004) (Figure 6C).
In the bronchus and the peribronchial space of OVA-challenged mice, the number of apoptotic cells was significantly higher in the dexamethasone-treated mice (8.8 ± 1.5 apoptotic cells/mm2, n = 11) compared with untreated mice (4.0 ± 0.8 apoptotic cells/mm2, n = 11; P = 0.006). The number of apoptotic cells was also noted to be higher in the interstitial compartment of OVA-challenged mice in the dexamethasone-treated mice (6.7 ± 1.2 apoptotic cells/mm2, n = 11) compared with untreated mice (3.4 ± 0.7 apoptotic cells/mm2, n = 11; P = 0.01). Although dexamethasone induced a statistically significant increase in the number of peribronchial and interstitial apoptotic cells, the absolute number of apoptotic cells detected was only a very small percentage of the total number of lung cells ( 1% of the cells examined). As a positive control for our TUNEL assay we used the positive control slides provided by the manufacturer of the apoptosis kit, which contained tissue (female rodent mammary gland obtained 35 d after the weaning of rat pups) known to have 12% apoptotic cells. We detected 12% apoptotic cells in the positive control tissue sections provided by the manufacturer of the apoptosis kit. ISS did not change the number of apoptotic cells in the peribronchial (3.9 ± 1.0 apoptotic cells/mm2, n = 14; P = ns versus control) or interstitial compartments (2.4 ± 0.5 apoptotic cells/mm2, n = 14; P = ns versus control) (Figures 6B and 6C).
ISS and Dexamethasone Reduce the Percentage of PAS-Positive Airway Epithelial Cells
Effect of ISS and Corticosteroids on the Total Number of BAL Lymphocytes The total number of BAL T lymphocytes following OVA challenge (3.1 ± 2.6 x 104 BAL lymphocytes) was significantly reduced in ISS-treated mice (1.3 ± 0.7 x 104 BAL lymphocytes, n = 8; P < 0.001, ISS + OVA, versus untreated OVA challenge), and in dexamethasone-treated mice (2.1 ± 1.4 x 104 BAL lymphocytes, n = 8; P = 0.008, dexamethasone + OVA, versus untreated OVA challenge). The combination of ISS and dexamethasone treatment (1.2 ± 0.7 x 104 BAL lymphocytes) did not reduce the total number of BAL lymphocytes more than ISS alone (1.3 ± 0.7 x 104 BAL lymphocytes, n = 8; P = ns).
In this study, we have demonstrated that ISS administered as therapy for allergic airway inflammation and AHR, such as might occur during an acute episode of asthma, decreases AHR as effectively as dexamethasone, and that the combination of ISS and dexamethasone is more effective than monotherapy with either agent alone in reducing AHR. The mechanism by which either ISS or corticosteroids inhibit AHR is likely to be mediated by distinct and shared cellular pathways. Distinct anti-inflammatory pathways are suggested by the fact that ISS and corticosteroids differ in their mechanism of eosinophil clearance to regional lymph nodes, induction of cellular apoptosis in peribronchial regions, and induction of the Th1 cytokine IFN- , which has antieosinophilic effects (21). Shared anti-inflammatory pathways are suggested by the fact that ISS and corticosteroids both inhibit Th2 cytokine responses (IL-5), the total number of BAL T lymphocytes, and mucus production to a similar degree. The combination of the shared and distinct anti-inflammatory pathways (some of which we have identified), may account for the additive effect of ISS and corticosteroids on inhibiting AHR. As eosinophilic inflammation is one cellular pathway leading to AHR, we have investigated the effect of ISS and corticosteroids on eosinophil production, trafficking into the lung, and clearance from the lung. Both ISS and corticosteroids significantly inhibited airway eosinophilia. ISS was more effective than corticosteroids in inhibiting bone marrow and blood eosinophilia, suggesting that ISS exerted a greater inhibitory effect on the bone marrow production and release of eosinophils into the circulation. However, corticosteroids differed from ISS in that corticosteroids, but not ISS, increased clearance of eosinophils to regional lymph nodes. Previous studies have demonstrated that fluorescently-labeled eosinophils, recovered from both antigen-challenged airways and from the peritoneal cavity of IL-5 transgenic mice, when instilled into the tracheal lumen of normal mice, homed to the peritracheal lymph nodes (10% within 24 h) and were able to stimulate antigen-specific CD4+, but not CD8+, T cell proliferation (22). Our study suggests that corticosteroids might augment the migration of eosinophils from the airway mucosa to regional lymph nodes and thus reduce eosinophilic inflammation in the lung. Apoptosis of recruited airway inflammatory cells such as eosinophils is one mechanism by which eosinophilic inflammation may resolve in asthma. Corticosteroids are known to induce eosinophil apoptosis in vitro (23). Corticosteroid-treated patients with asthma have decreased airway eosinophilia, increased eosinophil apoptosis, and increased expression of Bcl-2, Fas, and Fas ligand (24). Sputum samples obtained during an asthma exacerbation in corticosteroid-treated subjects demonstrate decreased airway eosinophilia and increased eosinophil apoptosis (25). Furthermore, the number of apoptotic eosinophils and macrophages in bronchial biopsy specimens from subjects with asthma is inversely correlated with the severity of asthma symptoms (26). Delayed eosinophil apoptosis (27) may contribute to tissue eosinophilia at sites of allergic inflammation (28). In mouse models of asthma lung eosinophils lavaged from aerosol allergen-challenged mice express the Fas receptor, and when activated in vitro with an anti-Fas mAb induces eosinophil death by apoptosis (29). In vivo administration of an anti-Fas antibody to mice that were sensitized and allergen-challenged with OVA-induced eosinophil apoptosis, and also suppressed AHR (30).
We observed an increased number of apoptotic cells in the peribronchial compartment of allergen-challenged compared with naive mice. As the number of lung eosinophils following allergen challenge ( Several groups (7, 8), including ours (4), have demonstrated that ISS is an effective preventive therapy in mouse models of asthma. The ability of ISS to inhibit airway inflammation and AHR was suggested in a previous study (31), but that study design in which ISS was also administered between two inhalation allergen challenges (preventive therapy for the second inhalation) incorporated components of both preventive therapy (ISS administered after first inhalation of OVA is preventive therapy for the second OVA inhalation 6 d later) and acute therapy following allergen exposure, and precluded definitively assessing whether ISS administered only as therapy after allergen challenge is effective. In this study, we have only administered ISS after both inhalation challenges had been completed to assure that no preventive therapy was incorporated into the study design. In addition, we have compared for the first time the efficacy of ISS compared with systemic corticosteroids, the standard therapy for acute exacerbations of asthma. In summary, we have demonstrated that ISS administered as therapy can significantly reverse allergic airway inflammation and AHR. Combination therapy with ISS and dexamethasone had an additive effect on the inhibition of AHR. The mechanisms by which ISS and dexamethasone inhibit AHR likely use shared and distinct pathways, suggesting that combination therapy might reduce AHR more effectively than monotherapy with either therapy alone.
The authors thank Peter Wagner, M.D. (UCSD) for performing measurements of dynamic compliance. This study was supported by an NIH grant T32 HL07022 (UCSD Pulmonary Division Training Grant) and an American Lung Association of California Research Training Fellowship Award (to R.I.), NIH grants AI 33977 and AI 38425 (to D.H.B.), and NIH grant AI 40682 and Dynavax (to E.R.). Received in original form February 28, 2002
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||