The results of the present study indicate that the PDE 4 inhibitor rolipram inhibits chlorine-induced lung injury and may represent a potential rescue treatment for chlorine inhalation. Rolipram administered 1 h after exposure inhibited chlorine-induced pulmonary edema and airway hyperreactivity. I.p. and i.n. routes of administration were effective in inhibiting pulmonary edema, and i.p., i.n., and i.m. routes produced inhibition of airway hyperreactivity. No inhibitory effect of rolipram on plasma protein leakage or inflammatory parameters (neutrophils and KC in lavage fluid) was observed.
Rolipram inhibited chlorine-induced pulmonary edema, and this is effect is consistent with the known mechanism of action of PDE inhibitors. Chlorine inhalation injures lung epithelial and endothelial cells leading to fluid leakage and pulmonary edema. The effects of PDE inhibitors on pulmonary edema are thought to be related to the known stimulation of alveolar fluid clearance by cAMP (Matthay et al., 2005
). Consistent with this, Song et al. demonstrated that i.n. administration of the long-acting β agonist R-formoterol to mice reversed chlorine-induced impairment of alveolar fluid transport, although the effects on pulmonary edema were not reported (Song et al., 2011
). In the present study, rolipram did not reduce the concentration of IgM in lavage fluid, which was an indication that it did not inhibit plasma protein leakage. This result would be expected if rolipram stimulated alveolar fluid clearance but did not promote repair of the disrupted epithelial/endothelial barrier.
Phosphodiesterase inhibitors are known to produce a spectrum of anti-inflammatory effects in lung injury models (Souness et al., 2000
), although the underlying molecular mechanisms are not well characterized. In some models, PDE inhibitors selectively decrease the production of a subset of inflammatory mediators (Herbert et al., 2008
). In the present study, anti-inflammatory effects of rolipram following chlorine exposure were assessed by measuring neutrophils and KC in lavage fluid. The highest dose of rolipram resulted in increased lavage fluid KC for three of the four delivery methods tested. However, rolipram treatment had either no effect or a minor inhibitory effect on lavage fluid neutrophils. It therefore appears that the observed increases in chemokine expression are not sufficient to affect neutrophilic inflammation. Previous studies have shown that PDE 4 inhibitors can increase cytokine expression, including KC, in some cell types (McCluskie et al., 2006
; Hertz et al., 2009
). Therefore the effects of PDE 4 inhibitors on inflammation may depend on the context of the specific injury or proinflammatory stimulus and on the specific cell types involved. Because an inhibitory effect of rolipram on chlorine-induced neutrophilic inflammation was not observed, it is possible that increased efficacy may be obtained by combined treatment with rolipram and an anti-inflammatory agent.
Our results indicated that rolipram treatment did not affect baseline lung mechanics in chlorine-exposed mice, but did inhibit methacholine-induced increases in respiratory system resistance. Chlorine inhalation alters baseline lung function and causes airway hyperreactivity to methacholine in mice (Martin et al., 2003
; Hoyle et al., 2010a
; Song et al., 2011
). Agents such as PDE inhibitors that raise intracellular cAMP levels can produce bronchodilation through a direct relaxant effect on airway smooth muscle. cAMP signaling pathways appear to have a minor effect on basal airway tone, but can be targeted with β-agonists or PDE inhibitors to counteract bronchoconstriction or airway hyperreactivity that occur in pathological states (Deshpande and Penn, 2006
). This was similar to published findings with R-formoterol, which inhibited airway hyperreactivity to inhaled methacholine in chlorine-exposed mice (Song et al., 2011
). Chlorine inhalation causes extensive injury to the epithelium of the central airways (Tian et al., 2008
; Song et al., 2011
). One potential mechanism of airway hyperreactivity to inhaled methacholine in this model would be the enhanced availability of aerosolized methacholine to the airway smooth muscle. Other injury models involving increases in airway epithelial permeability exhibited this phenomenon with an associated increase in Rn
(Bates et al., 2006
; Allen et al., 2009
). In contrast, our analysis of lung mechanics revealed that Rn
was not increased by the doses of methacholine we used and that rolipram treatment, which had a profound effect on Rrs
, did not inhibit Rn
. The results suggest that rolipram was not inhibiting resistance by relaxing the large airways; rather, the data (e.g. the significant inhibition of tissue damping) are consistent with effects on the lung periphery, which could potentially occur through the inhibitory effects of rolipram on small airways or on pulmonary edema.
In the present study mice were used to model chlorine injury to the lung and to investigate the ability of rolipram to ameliorate aspects of the injury. Because of the functional and anatomical differences between the mouse and human respiratory tracts, caution is necessarily required when extrapolating to effects in humans. In response to irritants such as chlorine, mice exhibit pronounced concentration-dependent changes in breathing patterns which are mediated by irritant-responsive sensory nerves. This effect, in concert with the differences in anatomy, makes it difficult to determine a dose in humans that would be equivalent to that used in this study in mice. In a disaster scenario involving large-scale chlorine release, humans will experience a variety of exposures with respect to concentration and time. Because of this, it is more important to model the spectrum of injuries that is typical following high-level exposure rather than targeting an exact human exposure dose. The general aspects of lung injury that we assessed in the mouse model, including pulmonary edema, inflammation, and impaired lung function, are typical of those documented in humans after large-scale chlorine release (Van Sickle et al., 2009
Both β-agonists and PDE inhibitors have been used therapeutically for treatment of lung diseases (Hoyle, 2010
). In theory, both types of compounds can raise cAMP levels and provide beneficial effects for treating lung injury. In practice, differential effects of the two classes of drugs have been observed. β-agonists have a long history of use as bronchodilators in asthma patients, and have also shown to be effective as inhibitors of acute lung injury in animal models (McAuley et al., 2004
; Wang et al., 2004
; Litvan et al., 2006
; Song et al., 2011
). In clinical trials, β-agonist treatment showed efficacy in an initial trial for treatment of acute lung injury/adult respiratory distress syndrome (Perkins et al., 2006
), but had no effect in a subsequent larger trial (Matthay et al., 2009
). As the majority of patients in such clinical trials develop lung injury associated with sepsis, the results do not rule out the possibility that β-agonists may be effective in treating acute lung injury of other etiologies such as acid aspiration or inhalation of irritant chemicals such as chlorine. A disadvantage of treatment with β-agonists is that these agents lose effectiveness with continued treatment as a result of receptor desensitization or compensatory increases in PDE activity (Johnson, 2006
). Therefore patients already receiving β-agonist therapy, e.g. for asthma, may be refractory to treatment of acute lung injury with these same agents. In contrast, PDE inhibitors raise cAMP levels, and no compensatory mechanisms that limit cAMP accumulation with chronic PDE treatment have been identified. Oral administration of PDE 4 inhibitors has been clinically tested for the treatment of COPD. Such inhibitors appear to have clinical efficacy, but their use is limited by side effects including gastrointestinal irritation and nausea (Rennard et al., 2008
; Calverley et al., 2009
; Giembycz and Field, 2010
). When considering the use of PDE 4 inhibitors as countermeasures against chlorine-induced lung injury, such adverse effects may be better tolerated or minimized in light of the limited time the drug will be taken, the potentially life-threatening nature of the illness, and administration by a route other than oral.
Ideal characteristics of an agent to be used as a countermeasure against chlorine-induced lung injury include efficacy against multiple aspects of injury, the ability to be administered quickly to large numbers of individuals, and efficacy when given as a rescue treatment subsequent to the exposure. Our experiments compared systemic and local delivery of rolipram and showed that both were effective in inhibiting pulmonary edema and airway hyperreactivity. Direct delivery to the respiratory tract has the advantage that locally high levels of drug can be targeted to the organ of interest. A disadvantage is that fluid leakage into the lungs caused by chlorine injury may interfere with the drug reaching the alveolar epithelium. In addition, delivery to unconscious victims may be difficult by this route. For countermeasure use in a mass casualty situation, a method of delivery that can be accomplished quickly by personnel with limited medical training is preferred. For treatment of human casualties, systemic delivery by intramuscular administration is likely a preferred route that combines speed and simplicity. The present study shows proof-of-principle for systemically delivered rolipram as an inhibitor of chlorine-induced lung injury. Future studies can be targeted toward developing formulations optimal for this route of delivery. An effective countermeasure must also provide therapeutic benefit when administered after exposure once lung injury has already begun to develop. In the present study, we selected a 1 h interval between the end of the chlorine exposure (2 h after the start of the exposure), and rolipram treatment was able to inhibit multiple aspects of lung injury when given at this time. The choice of the timing of treatment was based on the knowledge that lung injury, e.g. as evidenced by hypoxia, is present by 1 h after chlorine exposure (Gunnarsson et al., 1998
; Batchinsky et al., 2006
; Leustik et al., 2008
) and on the fact that this would be a reasonable time frame during which many patients could be reached by first responders in a disaster scenario. Overall, the results of the study indicated that rolipram is an effective rescue treatment for chlorine-induced lung injury and that both systemic and targeted administration to the respiratory tract were effective routes of delivery.