Abstract
Viral respiratory infections are considered one of the triggers of exacerbations of asthma. In a model of virus-induced airway hyper-responsiveness (AHR), mice infected with human respiratory syncytial virus (RSV) were shown to develop AHR accompanied by lung eosinophilia. Inhibitors of cyclic nucleotide phosphodiesterase (PDE) have been shown to affect airway responsiveness and pulmonary allergic inflammation. In this study, we assessed the effects of type 4 PDE (PDE4) inhibitors on AHR following RSV infection and compared them with a PDE3 inhibitor. In mice infected by intranasal inoculation of RSV, treatment with the PDE4 inhibitor rolipram or Ro-20-1724 reduced both AHR and the eosinophil infiltration of the airways. In contrast, the PDE3 inhibitor, milrinone, did not influence airway responsiveness or eosinophilic inflammation. These results demonstrate that PDE4 inhibitors can modulate RSV-induced AHR and lung eosinophilia and indicate that they have a potential role in treating exacerbations of asthma triggered by viral infection.
The pathology of asthma is complex and many factors contribute to its development. Infection with respiratory viruses has been well recognized as a trigger for acute asthma symptoms (Cypar et al., 1992), and viral pathogens have been found in greater than 80% of asthma exacerbations in adults (Nicholson et al., 1993). Respiratory syncytial virus (RSV), rhinoviruses, parainfluenza virus, and coronavirus have all been implicated (McIntosh et al., 1973), but the mechanisms underlying virus-induced wheezing are not well defined. Clinical studies suggest that eosinophils play a role in triggering and sustaining lung inflammation following RSV infection. Lymphocytes and eosinophils predominated in the airways on autopsy of two patients who died as a consequence of severe RSV bronchiolitis (Kim et al., 1969), and children with severe bronchiolitis demonstrated an increase in eosinophil numbers in the blood after RSV infection (Chin et al., 1969). Furthermore, high levels of eosinophil cationic protein have been demonstrated in nasopharyngeal secretions of children with RSV bronchiolitis (Garofalo et al., 1992).
Recently, a murine model of virus-induced altered airway function was described using human RSV (Schwarze et al., 1997). In this model, eosinophils, interleukin-5 (IL-5), and CD8+ T cells were shown to be essential for development of airway hyper-responsiveness (AHR) (Schwarze et al., 1997, 1999). In IL-5-deficient mice or following administration of anti-IL-5, AHR failed to develop. In addition, administration of anti-VLA4 antibody, while preventing eosinophil accumulation in the lung, also attenuated AHR (Schwarze et al., 1999).
Type 4 phosphodiesterase inhibitors (PDE4) have been demonstrated to exhibit anti-asthma effects due in part to bronchodilatory actions and anti-inflammatory activities (Torphy, 1998). PDE4 is the major PDE isotype in human (Dent et al., 1994; Hatzelmann et al., 1995) and guinea pig (Dent et al., 1991; Souness et al., 1991) eosinophils, and inhibition of PDE4 leads to a reduction in the production of proinflammatory mediators and eosinophil chemotaxis (Lagente et al., 1994, 1995; Tenor et al., 1996). There have been several reports demonstrating inhibitory effects of PDE4 inhibitors on AHR and lung eosinophilia in allergen-induced airway obstruction in guinea pigs (Raeburn et al., 1994; Santing et al., 1995; Danahay and Broadley, 1997; Manabe et al., 1997) and monkeys (Turner et al., 1994).
Given the relationship between eosinophil accumulation in the lung and virus-induced alterations of airway function, we tested the potential suppressive effects of PDE inhibitors on RSV-induced AHR. In these studies, we evaluated the effects of PDE3 and PDE4 inhibitors on RSV-induced inflammatory cell infiltration of the lung and changes in airway responsiveness to inhaled methacholine (MCh).
Materials and Methods
Animals.
Female BALB/c mice (20 to 30 g weight and 8 to 12 weeks of age), free of specific pathogens, were obtained from Jackson Laboratories (Bar Harbor, ME). All experimental animals used in this study were under a protocol approved by the Institutional Animal Care and Use Committee of the National Jewish Medical and Research Center.
Virus and Infection.
Human RSV A (long strain) was obtained from the Viral Diagnostics Laboratory, Health Sciences Center, University of Colorado (Denver, CO). The virus was cultured on Hep2 cells from American Type Culture Collection (Rockville, MD) in medium containing fetal calf serum from Life Technologies, Inc. (Gaithersburg, MD). Titers for infectiousness of the stock virus were determined using quantitative plaque-forming assay.
Mice were infected under light anesthesia (Avertin 2.5%, 0.015 ml/g b.wt.) by intranasal inoculation of RSV [105plaque-forming units (PFU) in 50 μl of PBS]. Controls were sham-infected with PBS in the same way.
Determination of Airway Responsiveness.
Airway responsiveness was assessed as described (Hamelmann et al., 1997), using a single chamber whole body plethysmograph obtained from Buxco (Troy, NY). This approach correlated closely with pulmonary resistance measured by conventional two-chamber plethysmography in ventilated animals (Hamelmann et al., 1997). Enhanced pause (Penh) was used as the measure of airway responsiveness in this study. In the plethysmograph, mice were exposed for 3 min to nebulized PBS and subsequently to increasing concentrations of nebulized MCh (Sigma Chemical Co., St. Louis, MO) in PBS using the AeroSonic ultrasonic nebulizer. After each nebulization, recordings were taken for 3 min. The Penh values measured during each 3-min sequence were averaged. Shown are the absolute Penh values in response to inhaled PBS or increasing concentrations of MCh.
Measurement of Cytokine Levels in Bronchoalveolar Lavage Fluids.
After sacrificing the mice, lungs were lavaged with 1-ml aliquots of Hanks' balanced salt solution (room temperature) through a polyethylene syringe attached to the tracheal cannula. Bronchoalveolar lavage fluid (BALF) was centrifuged (500g for 5 min), and the supernatants were collected and frozen at −20°C until analysis. The concentrations of interferon-γ (IFN-γ) and IL-5 in the supernatants were assessed by enzyme-linked immunosorbent assay as described (Schwarze et al., 1997). Cytokine levels were calculated by comparison with known cytokine standards (PharMingen, San Diego, CA). The limit of detection in the assay was 10 pg/ml for each cytokine.
Lung Cell Isolation.
Lung cells were isolated by collagenase digestion as described previously (Oshiba et al., 1996) and counted with a Coulter counter. Slides prepared with Cytospin 3 (Shandon, Pittsburgh, PA) were stained with Leukostat from Fisher Diagnostics (Pittsburgh, PA), and differential cell counts were performed by counting approximately 300 cells under light microscopy.
Histology.
Before removal, the lungs were fixed by inflation with 10% neutral buffered formalin. The fixed lung specimens were stored in 10% neutral buffered formalin, dehydrated in 70% ethanol, and parafin embedded. Sections (5 μm) were cut, deparafinized, stained with H&E, and viewed by light microscopy.
Experimental Protocols.
Mice were infected with RSV on day 0. In a previous study (Schwarze et al., 1997), AHR was found to peak on day 6, and AHR was assessed similarly in this study. Mice were sacrificed on day 7 for collection of BALF and lung cells. Drugs were administered i.p. twice a day for 6 days. As a control, mice were administered PBS.
Drugs.
Ro-20-1724 (4-(3-butoxy-4-methoxybenzyl)-2-imidazolidinone) was purchased from Calbiochem (San Diego, CA). Rolipram and milrinone were purchased from Sigma. PDE inhibitors were dissolved in ethanol and diluted with PBS. The final concentrations of ethanol were less than 1%.
Data and Statistical Analyses.
Values for all measurements were expressed as the mean ± S.E. The inhibitory effects of drugs on the increase in numbers of cells are indicated as a percentage reduction. Pairs of groups were compared by Student's ttest; comparison of more than two groups was performed by the Dunnett test for parametric analyses or Steel test for nonparametric analyses.P values for significance were set at .05.
Results
Effects of Rolipram on RSV-Induced Airway Hyper-responsiveness, Leukocyte Infiltration, and Cytokine Levels in BALF.
The airway response to inhaled MCh in mice infected with RSV and in sham-infected controls was assessed on day 6 after infection. The mice infected with RSV were significantly more reactive than the sham-infected controls (Fig. 1A). The Penh response to 50 mg/ml MCh in RSV-infected, vehicle-treated mice was 2.1-fold higher than in sham infected-mice. In mice treated with rolipram, at 0.03, 0.1, or 0.3 mg/kg for 6 days, a significant decrease in airway responsiveness to MCh following RSV infection was observed.
Inhibitory effects of rolipram on RSV-induced airway hyper-responsiveness and numbers of eosinophils and neutrophils in the lungs. Mice were inoculated with RSV (105 PFU) or PBS (sham). RSV-infected mice were treated with rolipram or vehicle for 6 days (twice per day). Airway responsiveness to MCh was assessed on day 6 (A), and lung digests were prepared on day 7 (B, eosinophils; C, neutrophils). The results are expressed as mean ± S.E. (n = 12). #P < .05, ##P < .01 compared with the sham-infected group; *P < .05, **P < .01 compared with vehicle-treated group.
To investigate the effects of rolipram on pulmonary inflammatory cell infiltration induced by RSV infection, lung cells were isolated by collagenase digestion and differential cell counts were performed. In RSV-infected, vehicle-treated mice, the numbers of eosinophils and neutrophils were increased significantly in lung cell isolates compared with sham-infected controls (Fig. 1, B and C). The number of eosinophils and neutrophils was increased by 2.8- and 1.7-fold, respectively, in RSV-infected mice. In rolipram-treated mice (0.1 and 0.3 mg/kg), the numbers of eosinophils were lower (by up to 73%) than in vehicle-treated mice (Fig. 1B). There were no significant differences in neutrophil numbers between the vehicle-treated group and the rolipram-treated groups (Fig. 1C).
The concentrations of IFN-γ in BALF are shown in Table1. The concentrations of IFN-γ were higher in the BALF of RSV-infected, vehicle-treated mice than in sham-infected mice. IL-5 could not be detected in any of the groups. The concentrations of IFN-γ were lower in the BALF of RSV-infected/rolipram-treated mice than in vehicle-treated mice, but the differences between vehicle- and rolipram-treated groups were not significant.
Influence of treatment with PDE4 inhibitors for 6 days on RSV-induced IFN-γ levels in BALF
Effects of Ro-20-1724 on RSV-Induced AHR, Leukocyte Infiltration, and Cytokine Levels in BALF.
As shown in Fig.2A, when mice were treated with Ro-20-1724 (3 mg/kg) for 6 days, they demonstrated a significant inhibition of AHR to MCh following RSV infection. The responses of these mice were significantly lower than those of vehicle-treated mice. Ro-20-1724 at a concentration of 1 mg/kg did not alter airway responsiveness to RSV infection.
Inhibitory effects of Ro-20-1724 on RSV-induced airway hyper-responsiveness and numbers of eosinophils and neutrophils in the lungs. Mice were inoculated with RSV (105 PFU) or PBS (sham). RSV-infected mice were treated with Ro-20-1724 or vehicle for 6 days (twice per day). Airway responsiveness to MCh was assessed on day 6 (A) and lung digests were prepared on day 7 (B, eosinophils; C, neutrophils). The results are expressed as mean ± S.E. (n = 12). #P < .05, ##P < .01 compared with the sham-infected group; **P < .01, *P < .05 compared with the vehicle-treated group.
The number of eosinophils in the lung were increased in the RSV-infected, vehicle-treated mice compared with sham-infected mice (Fig. 2B). In Ro-20-1724 (3 mg/kg)-treated mice, the number of eosinophils was significantly lower than in the vehicle-treated group. The number of eosinophils following Ro-20-1724 (3 mg/kg) was similar to the sham-infected controls. Ro-20-1724 did not influence the number of neutrophils (Fig. 2C). Ro-20-1724 (1 or 3 mg/kg) treatment had no significant effect on IFN- γ concentrations in the BALF (Table 1).
Effects of Milrinone on Acute RSV-Induced Airway Hyper-reactivity and Leukocyte Infiltration.
Administration of the PDE3 inhibitor, milrinone, at a dose of 3 mg/kg for 6 days had no significant effect on RSV-induced AHR throughout the MCh dose-response curve (Fig.3A). Similarly, the number of eosinophils in the lungs in response to RSV infection was not affected by milrinone (Fig. 3B). There was a trend toward reduced number of lung neutrophils in milrinone-treated mice, but the differences were not statistically different (Fig. 3C).
Failure of milrinone on RSV-induced airway hyper-responsiveness and numbers of eosinophils and neutrophils in lungs. Mice were inoculated with RSV (105 PFU) or PBS (sham). RSV-infected mice were treated with milrinone or vehicle for 6 days (twice per day). Airway responsiveness to MCh was assessed on day 6 (A), and lung digests were prepared on day 7 (B, eosinophils; C, neutrophils). The results are expressed as mean ± S.E. (n = 8). #P < .05 compared with the sham-infected group.
Histopathological Investigations in RSV-Infected Mice.
RSV infection induced a peribronchial infiltration of inflammatory cells as is illustrated in Fig. 4 (B and C) compared with sham-infected mice (shown in Fig. 4A). In the lungs of mice treated with rolipram (0.3 mg/kg, twice a day) or Ro-20-1724 (3 mg/kg, twice a day) for 6 days, these inflammatory changes were not observed following RSV infection (Fig. 4, D and E).
Histological appearance of lung tissue from RSV-infected mice. A, lung tissue from a sham-infected mouse shows normal alveolar structure. B and C, lung tissue from RSV-infected mouse illustrating an inflammatory infiltrate. Lung from RSV-infected and rolipram (0.3 mg/kg) (D) or Ro-20-1724 (3 mg/kg)-treated (E) mouse for 6 days. Tissues were stained with H&E. Magnifications are 100× (A, B, D, E) and 400× (C).
Effect of PDE4 Inhibitors on MCh-Induced Bronchoconstriction.
To ensure that the effects of the PDE4 inhibitors were not simply on MCh-induced bronchoconstriction, mice were treated with rolipram or Ro-20-1724 for 6 days at the doses that reduced AHR (0.3 or 3 mg/kg, respectively). As shown in Fig. 5, neither rolipram nor Ro-20-1724 influenced MCh-induced bronchoconstriction when measured on day 6, excluding a direct effect of the drugs on smooth muscle contraction.
Influence of rolipram or Ro-20-1724 on MCh-induced bronchoconstriction. Mice were treated with rolipram (0.3 mg/kg), Ro-20-1724 (3 mg/kg), or vehicle for 6 days (twice per day). Airway responsiveness to MCh was assessed on day 6. The results are expressed as mean ± S.E. (n = 8). There were no significant differences between the groups at each concentration of MCh.
Discussion
In this study, we evaluated the potential of different PDE inhibitors in preventing airway hyper-responsiveness and inflammation following RSV infection. In this well-characterized murine model, RSV infection triggered a significant inflammatory response in the lung with increased numbers of eosinophils and neutrophils as well as altered airway responsiveness to inhaled MCh. It has been reported that PDE4 inhibitors are more effective when administered on a twice daily basis and, preferably, for several days in guinea pigs (Banner and Page, 1995). For these reasons, we utilized a twice daily regimen for 6 days. In the present study, two PDE4 inhibitors, rolipram and Ro-20-1724, inhibited the induction of AHR and eosinophil influx in the lungs induced by RSV infection. Milrinone, a PDE3 inhibitor, at a dose known to inhibit PDE3 did not affect any of these parameters. Rolipram was effective at doses of 0.03 to 0.3 mg/kg, whereas Ro-20-1724 was effective at 3 mg/kg. The doses of rolipram were lower than those used in other murine models of lung inflammation (Klemm et al., 1995; Miotla et al., 1998) but were similar to their effective use in antigen-induced airway obstruction models in guinea pigs (Santing et al., 1995; Danahay and Broadley, 1997).
The intracellular concentrations of cyclic nucleotides in most cell types is determined by the balance between surface receptor stimulation and intracellular breakdown of cyclic nucleotides by PDEs. Five distinct isoenzyme families have been identified based on the specificity of substrate interactions and the activities of selective inhibitors. The main PDE isotype in eosinophils is type 4 (Souness et al., 1991; Hatzelmann et al., 1995), and because eosinophils have been closely correlated with disease activity in human asthma and murine models, PDE4 was an obvious target. This is supported by the significant inhibitory effects of both of the PDE4 inhibitors and the poor inhibitory activity of the type 3 inhibitor, milrinone, on eosinophil infiltration seen in this study. Moreover, the results of the inhibitory effect of PDE4 inhibitors on eosinophil infiltration support the previous reports in other antigen-induced models (Underwood et al., 1994; Danahay and Broadley, 1997). Milrinone administered i.p. twice daily for 6 days at a dose of 3 mg/kg did not have any effect on eosinophil numbers but marginally reduced the number of neutrophils in the lungs following RSV infection. In an antigen-induced model in guinea pigs, the PDE3 inhibitors similarly reduced the number of neutrophils but did not affect eosinophil numbers (Danahay and Broadley, 1997). At a dose of 3 mg/kg administered orally, milrinone did reduce formation of an occlusive thrombus in mice (Kondo et al., 1999). Together these results suggest that PDE3 inhibitors are not preventing allergic responses in the lung (although direct evidence for milrinone reaching the lung in sufficient quantities is lacking).
We previously demonstrated an essential role for eosinophils in the development of AHR following both allergic sensitization and challenge (Hamelmann et al., 1997) as well as following RSV infection (Schwarze et al., 1997, 1999). In models of RSV-induced AHR, eosinophils recruited into the lung have been suggested to be essential to the development of AHR. In the study of Schwarze et al. (1999), IL-5-deficient mice did not develop AHR nor lung eosinophilia, but this could be overcome following IL-5 reconstitution. Furthermore, anti-VLA-4 antibody inhibited eosinophil infiltration and AHR as well (Schwarze et al., 1999). In parallel to the reduction in eosinophil numbers, RSV-induced AHR to inhaled MCh could be inhibited by PDE4 inhibitors but not by the PDE3 inhibitor. The mechanism of induction of AHR by eosinophils has not been defined. Instillation of human eosinophil-derived major basic protein has been shown to induce AHR in rats (Coyle et al., 1994), and cationic proteins released from activated eosinophils are likely involved in the pathogenesis of AHR. Based on these results and the assumption that eosinophils play a major role in RSV-induced AHR, it is presumed that PDE4 inhibitors are at least partially effective in this model by inhibiting eosinophil influx into the lung.
PDE4 inhibitors may also affect eosinophil activation. By increasing intracellular concentrations of cAMP, activated eosinophils may be inhibited from discharging their contents (i.e., degranulation). There are other possible sites of action of these compounds, including effects on the development of eosinophils from their stem cells in the bone marrow, the permeability of the vascular endothelium to leukocytes, the production and release of inflammatory mediators, and cytokine synthesis. Other potential mechanisms for PDE4 inhibitory activity in this RSV model include effects on superoxide release or leukotriene production (Kimpen et al., 1992) and on eosinophil chemotaxis (Barnette et al., 1995; Banner et al., 1996; Cohan et al., 1996). In in vitro experiments, PDE4 inhibitors may preferentially inhibit Th2 type cytokine production (Essayan et al., 1997). In the present study, little if any IL-5 was detected in the BALF of RSV-infected mice; IFN-γ levels, which were increased following RSV infection, were not significantly affected by the PDE4 inhibitors.
One other mechanism of action of the inhibitors may be through reduced expression of adhesion molecules resulting in impaired homing of eosinophils to the lung, as seen in the reduced numbers of eosinophils in the lungs. These findings are supported by the observations that PDE4 inhibitors can reduce the expression of the binding proteins on endothelial cells and, as a result, reduce the eosinophil and lymphocyte numbers in the lung. In these studies, the PDE4 inhibitor rolipram inhibited expression of E-selectin by human lung microvascular endothelial cells (Blease et al., 1998).
It is well established that PDE4 inhibitors have a bronchodilatory effect on guinea pig and human airway smooth muscle (Howell et al., 1992; Fujii et al., 1997). To eliminate this possibility as a contributing mechanism to the attenuation of AHR, we evaluated the effect of PDE4 inhibitors on MCh-induced bronchoconstriction in naive mice. Under these conditions, the PDE4 inhibitors had no effect on the altered airway responsiveness induced following inhalation of increasing concentrations of MCh, indicating that the inhibitory effects of PDE4 inhibitors on RSV-infected mice were not related to a direct bronchodilatory effect.
In summary, PDE4 inhibitors administered during an ongoing RSV-induced inflammatory response in the lung significantly reduced the virus-induced eosinophilic inflammatory response in the lung and the alterations in airway responsiveness to inhaled MCh. These results support the further evaluation of these potent inhibitors in pulmonary eosinophilic inflammatory disorders that lead to altered airway responsiveness.
Acknowledgments
We are grateful to Makiko Ikemura for assistance and Diana Nabighian for help in preparing the manuscript.
Footnotes
-
Send reprint requests to: Erwin W. Gelfand, M.D., National Jewish Medical and Research Center, 1400 Jackson St., Denver, CO 80206. E-mail: gelfande{at}njc.org
-
↵1 This work was supported in part by National Institutes of Health Grants HL-61005 and HL-36577 (to E.W.G.).
- Abbreviations:
- RSV
- respiratory syncytial virus
- AHR
- airway hyper-responsiveness
- BALF
- bronchoalveolar lavage fluid
- MCh
- methacholine
- PDE
- phosphodiesterase
- PFU
- plaque-forming units
- IFN-γ
- interferon-γ
- IL-5
- interleukin-5
- Penh
- enhanced pause
- Received January 13, 2000.
- Accepted April 10, 2000.
- The American Society for Pharmacology and Experimental Therapeutics