Repeated beta-2 adrenergic receptor agonist therapy attenuates the response to rescue bronchodilation in a hyperoxic newborn mouse model

Abstract

Background

Preterm infants with neonatal lung injury are prone to wheezing and are often treated with beta-2 adrenergic receptor (β-AR) agonists although any benefits of β-AR agonists may be lost with chronic use.

Objective

To investigate if repeated β-AR agonist exposures would down-regulate β-ARs in the immature lung resulting in a decreased response to bronchodilator rescue and whether hyperoxic exposure would aggravate this response.

Methods

Newborn mice were raised for 21 days in 60% or 21% oxygen and received daily aerosols of formoterol or saline. Respiratory system resistance (Rrs) and compliance (Crs) were measured in response to methacholine challenge and rescue bronchodilation with levalbuterol. Western blot analysis quantified the relative amount of lung β-ARs.

Results

Hyperoxia increased airway reactivity to methacholine. Animals raised in hyperoxia that received daily formoterol were most sensitive to methacholine and exhibited a blunted response to levalbuterol bronchodilation. Hyperoxia exposed animals receiving daily formoterol vs saline showed a significant decrease in the relative amount of lung β-ARs.

Conclusions

In this hyperoxia exposed neonatal mouse model, repeated β-AR agonist treatments increased airway reactivity and attenuated the response to a rescue bronchodilator. The blunted bronchodilator response could be explained by a reduced quantity of lung β-ARs. Our findings may account for a time-dependent decrease in therapeutic benefit of β-AR agonists in preterm infants with neonatal lung injury, which may have clinical consequences for patients already prone to airway hyperreactivity.

Introduction

Bronchopulmonary dysplasia (BPD) and wheezing disorders are major pulmonary morbidities of prematurity [1, 2]. The causes are likely multifactorial, including prematurity, oxygen toxicity/oxidative injury, volutrauma, infection, and inflammation [2]. Acute lung injury in the vulnerable infant results in airway remodeling, decreased alveolarization, alvealor simplification, and parenchymal changes [1, 2]. Increased reports of respiratory wheezing and asthma are common long-term pulmonary consequences of preterm birth and BPD [1–3]. Indeed, a consistent observation has been that preterm infants, with or without BPD, continue to be at very high risk for airway hyperreactivity in infancy and childhood [4–6].

Although often diagnosed with asthma, infants and children with BPD are frequently less responsive to conventional asthma therapies [7]. The airway hyperreactivity observed in former preterm neonates is strongly associated with a history of prolonged supplemental oxygen and BPD, compared to the airway hyperreactivity observed in term controls which instead is associated with a history of inheritance, allergy, inflammation, and cigarette exposures [5]. Wheezing preterm infants, with and without BPD, display decreased airway reversibility, atopy, inflammation, and exhaled nitric oxide than asthmatics [1, 6–9]. Yet, school-age preterm infants are twice as likely to be prescribed asthma medications than their full term peers [10]. Furthermore, beta-2 adrenergic receptor (β-AR) agonists may paradoxically increase airway resistance in some of these former premature patients [11].

There is not yet a consensus on how to manage preterm patients with developing or established BPD. Therapies include respiratory support with positive pressure and/or supplemental oxygen, diuretics, bronchodilators, inhaled and systemic corticosteroids, caffeine, and inhaled nitric oxide [12]. Clinically, β-AR agonists are used both in the neonatal intensive care unit [13] and prescribed at discharge [14]. Both short and long-acting β-AR agonists have been shown to transiently improve airway resistance and compliance of ventilated neonates [15, 16] but these effects have not facilitated earlier weaning of respiratory support nor decreased the rates of BPD [17]. The potential benefits of β-AR agonists may also diminish with repeated or continued use [18] since prolonged exposures can result in tolerance [19].

At the time of birth, newborn mice have developmentally immature lungs similar to the premature human lung in sequence and relative timing of alveolarization [20]. Hyperoxic exposure in newborn mice creates a lesion very similar to human BPD [21], manifesting increased airway reactivity [22]. As such, we used a neonatal mouse model to test the hypothesis that repeated exposure to a β-AR agonist will down-regulate β-ARs in the immature lung resulting in a decreased response to bronchodilator rescue and that hyperoxic exposure will aggravate this response. These findings may provide future insight into mechanisms related to a perceived desensitized/diminished response of preterm infants to long-term β-AR agonist use.

Materials and Methods

Animal Model

Animal protocols were approved by the Institutional Animal Care and Use Committee at Case Western Reserve University. Timed pregnant C57BL/6 mice (Charles River) were purchased and housed in an AAALAC accredited animal facility at Case Western Reserve University. Animals were maintained on standard 12 hour light-dark cycles with ad libitum standard food and water. Two or more litters of newborn pups were pooled within 24 hours of birth and randomly redistributed into treatment groups. Litters, paired with a nursing dam, were placed in 60% oxygen or room air (21%). Hyperoxia exposed animals were housed in standard cages placed in a 38 L Plexiglas chamber with a continuous flow of oxygen (2 L/min) for 21 days. Oxygen concentrations were monitored twice daily via an oxygen analyzer (MiniOX I; MSA Medical). To control for oxygen toxicity, nursing dams were rotated between paired litters during biweekly cage changes. The long-acting β-AR agonist formoterol was dissolved in DMSO and resuspended to a final concentration of 1 mg/10 mls formoterol in 0.1% DMSO saline (Sigma Aldrich). Subgroups of the hyperoxic and room air exposed animals were treated daily with either aerosolized formoterol or normal saline with 0.1% DMSO vehicle (10 mls). Solutions were delivered to caged animals in 38 L Plexiglas chambers over 60 minutes using a continuous flow nebulizer (Global Medical Holdings LLC). On day 21, animals were all weaned to room air and aerosol treatments discontinued. Animals were weighed and experimental procedures or tissue preparations were performed after 24–48 hours had elapsed from the last aerosol treatment.

Lung Mechanics

Under anesthesia (intraperitoneal ketamine/xylazine), 3-week old mice were placed supine on a heated surgical table, tracheostomized, and ventilated via a 19G blunt tip cannula with a commercial rodent ventilator (flexiVent, SCIREQ) at a tidal volume of 10 ml/kg, rate of 150 breaths/min, PEEP of 3 cm H₂O, and FiO₂ of 50%. Animals were then paralyzed (intraperitoneal pancuronium bromide) and respiratory system resistance (Rrs) and compliance (Crs) were calculated in the flexiVent software using a 1.2 second, 2.5 Hz single-frequency forced oscillation maneuver [23]. After two recruitment breathes of sustained inspiration up to a pressure of 30 cm H₂0 for 3 seconds, saline control and methacholine at 12.5, 25, 50, 100, and 200 mg/ml (Sigma Aldrich) were aerosolized using an ultrasonic nebulizer diverted into the ventilator’s inspiratory flow over 10 seconds (Aeroneb; SCIREQ). Measurements of average Rrs and Crs were made every 40 seconds over 5 minutes after each dose. Following preconstriction with the full methacholine dose-response (200 mg/ml), the short-acting β-AR agonist levalbuterol (2.5 mg/ml; Sigma Aldrich) was similarly aerosolized over 10 seconds to assess bronchodilator rescue as indicated by the average decrease in Rrs and increase in Crs 2 minutes after the aerosol was administered.

Western Blot

Western blot was used to quantify the relative amount of β-AR in lung homogenates of animals from each treatment group that did not undergo lung mechanics as previously described [24]. After terminal anesthesia (intraperitoneal ketamine/xylazine), harvested lungs were rinsed in ice-cold PBS, snap-frozen in liquid nitrogen, and stored at −80° C. Tissue was resuspended in ice-cold commercial lysis buffer containing protease inhibitors (Sigma Aldrich) and then homogenized. Protein levels of lung homogenates were determined by BCA assay (Thermo-Scientific). Samples of 15 μg of protein were loaded and separated by 10% polyacrylamide gels and transferred to nitrocellulose membranes (Bio-Rad). Membranes were blocked with 5% Milk/5% BSA TBS-Tween and incubated in β-AR primary antibody (55 kDa, rabbit polyclonal, 1:1000; Abcam) overnight at 4° C then anti-rabbit HRP-conjugated secondary antibody for 1 hour at room temperature (goat polyclonal, 1:5000; Abcam). As a loading control, β-actin primary antibody (42 kDa, mouse monoclonal, 1:5000; Abcam) and anti-mouse HRP-conjugated secondary antibody (donkey polyclonal, 1:4000; Abcam) were used. β-AR band intensities were quantified and normalized to β-actin using ECL (Thermo Scientific). Relative intensities were measured using densitometry software (Image J).

Statistical Analysis

Data are expressed as means ± standard errors of the mean (SEM). Animal body weights were compared by one way analysis of variance. Alterations in airway reactivity with increasing doses of methacholine were compared via two way analysis of variance repeated measure analysis with post hoc comparisons using a fixed sequence method. Bronchodilatory response was compared by Student’s T-test. β-AR density was compared between groups using Kruskal-Wallis one way analysis of variance rank test with post hoc Student-Newman-Keuls test. Statistical significance was defined as p < 0.05.

Results

Animal Body Weight

In total, 46 animals completed full lung mechanics testing and another 56 animals were harvested for tissue collection from 20 litters. The average body weights did not significantly differ between groups (table 1).

Table 1.

Body weights of animals exposed for 21 days to 60% or 21% oxygen with daily treatments of aerosolized saline or formoterol.

Body Weight (grams) Mean ± SEM	21%+Saline	21%+Formoterol	60%+Saline	60%+Formoterol	p-value
Lung Physiology (n)	9.9 ± 0.4 (12)	9.9 ± 0.5 (8)	10.2 ± 0.4 (18)	10.5 ± 0.3 (8)	0.81
Western Blot (n)	9.5 ± 0.4 (14)	8.7 ± 0.5 (14)	9.2 ± 0.4 (14)	8.6 ± 0.4 (14)	0.44
Total (n)	9.2 ± 0.5 (26)	8.5 + 0.5 (22)	9.5 ± 0.5 (32)	8.6 ± 0.5 (22)	0.46

Airway Responsiveness to Methacholine

Following aerosolized saline, baseline Rrs and Crs did not statistically differ between exposure groups. Within each group there were observed significant increases in Rrs (fig. 1a) and decreases in Crs (fig. 1b) when challenged with the methacholine dose response. The hyperoxic cohorts displayed significantly increased Rrs at the highest methacholine doses (100 and 200 mg/ml) when compared to the animals raised in room air (p < 0.05). The animals exposed to both hyperoxia and daily formoterol were particularly reactive to methacholine, displaying significantly increased reactivity at even the lower methacholine doses (≥ 25 mg/ml; p < 0.05) compared to the room air controls. There was no effect of hyperoxia and/or formoterol on the Crs dose response curves between groups.

Fig. 1.

Lung physiology in mechanically ventilated mice. Significant methacholine dose-dependent increases in respiratory system resistance (Rrs) and decreases in compliance (Crs) were observed within each group. a) Rrs: Formoterol administration alone had no effect on the response of Rrs to methacholine when compared between room air cohorts. Significant increases were seen between the dose response curves of both hyperoxic cohorts compared to room air control (brackets). These differences were seen at the highest methacholine doses (100 and 200 mg/ml) with hyperoxia exposure alone. In contrast, the combined exposure of hyperoxia and formoterol increased the sensitivity to methacholine, displaying significantly higher reactivity at lower doses of methacholine (≥ 25 mg/ml) when compared to room air controls. b) Crs: Significant differences in Crs methacholine dose response curves were not observed between groups. p < 0.05: * 21% + Saline vs 60% + Saline, ** 21% Saline vs 60% + Formoterol.

Bronchodilator Rescue with Levalbuterol

Following preconstriction with methacholine, levalbuterol was administered to assess bronchodilator rescue. All treatment groups showed a significant decrease in Rrs following bronchodilator rescue (fig. 2a). Significant changes in Crs were not observed in the group raised in hyperoxia and treated with daily formoterol (fig. 2b). To account for potential differences in maximum Rrs following preconstriction, the percent change in Rrs was also calculated and compared between the hyperoxic cohort receiving saline and formoterol (fig. 3). In this manner, the hyperoxia exposed mice receiving daily β-AR agonist treatments displayed an attenuated response to the rescue bronchodilator when compared to those raised in hyperoxia receiving saline (p < 0.05).

Fig. 2.

Rescue bronchodilation with levalbuterol following preconstriction with 200 mg/ml methacholine. a) Rrs: Significant decreases in respiratory system resistance were observed in all groups following bronchodilator administration. b) Crs: Significant increases in compliance were observed in the 60% + Saline, 21% + Saline, and 21% + Formoterol groups, but Crs did not increase in the 60% + Formoterol group after administration of a bronchodilator. p < 0.05: * pre vs. post bronchodilator.

Fig. 3.

Percent change in respiratory system resistance with levalbuterol rescue following preconstriction with 200 mg/ml methacholine. Bronchodilator response, as measured by a percent decrease in Rrs, was significantly attenuated in the hyperoxic cohort treated with daily aerosolized formoterol compared to saline. * p < 0.05.

Lung Beta-2-adrenergic Receptor Expression

Since β-AR agonist response depends, in part, on expression of β-AR, we measured β-AR density in the lung by Western Blot analysis (fig. 4). Treatment with daily formoterol compared to saline for 21 days significantly decreased the relative amount of β-AR:β-actin among both the room air and hyperoxic groups (both p < 0.05). Compared to room air saline controls, hyperoxic exposure with and without formoterol significantly reduced the relative amount of β-AR (p < 0.05 for both hyperoxic groups).

Fig. 4.

Whole lung beta-2-adrenergic receptor expression by Western Blot. Hyperoxic exposure significantly reduced the relative quantity of receptor compared to room air controls (* p < 0.05). Treatment with daily formoterol for 21 days showed a decreased relative quantity of receptor among both the room air and the hyperoxic cohorts (** p < 0.05). n = 14 animals in each group.

Discussion

In this newborn mouse model, hyperoxia increased airway reactivity. Hyperoxic exposure, when coupled with repeated β-AR agonist treatments, further increased sensitivity to methacholine, attenuated the bronchodilator response to levalbuterol, and was associated with a reduced quantity of receptors in the lung. Taken together, these data may indicate detrimental effects of prolonged courses of β-AR agonists on preterm infants and children in supplemental oxygen therapy. Furthermore, chronic inhaled β-AR agonists may diminish the efficacy of a rescue bronchodilator in a patient population already at risk for airway hyperreactivity.

Moderate hyperoxia (FiO₂ 60%) as compared to 100% oxygen was used in this model because of the movement in clinical practice to utilize lower levels of oxygen in neonatal intensive care units. Increased reactivity to methacholine challenge was found in our hyperoxic cohorts compared to room air controls, which is consistent with other studies of moderate hyperoxia (FiO₂ 40–70%) in newborn rodents [2, 22, 25]. Neonatal hyperoxia exposure (FiO₂ 65% for just 7 days) has been shown to cause lasting changes to the bronchiolar wall and bronchiolar-alveolar attachments in adult animals recovered in room air [26]. Further investigation of the long term effects of early hyperoxia exposure, β-AR agonist treatment, and airway reactivity are warranted.

Although we did not observe significant differences in baseline Rrs or Crs, animals exposed to the combination of hyperoxia and chronic β-AR agonists seemed to be particularly sensitive to methacholine challenge, with observed increases in resistances at even the lower doses of methacholine. In the experimental adult asthma literature, increased airway hyperreactivity has similarly been observed with the combination of chronic β-AR agonist treatments and lung inflammation [27, 28]. Future studies targeted at blockade of cytokine driven lung inflammation may provide insight into some of the underlying mechanisms of this observed airway hyperreactivity.

Our animal findings revealed a relative decrease in β-AR quantity following either daily treatments with a β-AR agonist or sustained hyperoxia exposure. Agonist stimulation of β-ARs classically results in receptor inactivation and internalization with eventual down-regulation of the receptor following chronic stimulation. Under normoxic conditions, this decrease in β-ARs did not appear to have a physiological effect on baseline Rrs or Crs, nor response to a rescue bronchodilator. Hyperoxia exposure alone significantly reduced the quantity of β-ARs when compared to saline room air controls. The mechanisms of hyperoxic related decreases in β-ARs have not yet been well studied. We speculate that airway remodeling, alveolar simplification, and altered transcription may all contribute to a decreased whole lung receptor pool, and as such future studies are warranted. Despite a decreased β-AR population, a physiologic effect was not observed on baseline mechanics in these hyperoxia exposed animals, and a marked decrease in Rrs and increase in Crs was observed following rescue bronchodilation. We hypothesize that there likely exists an effective threshold of β-ARs required for agonist stimulated bronchodilation. Conversely, daily treatments with formoterol in hyperoxia exposed mice further decreased the receptor population, with these animals displaying a markedly blunted response to an equivalent rescue bronchodilator dose and no observed improvement in Crs. We speculate that the combination of sustained hyperoxic and β-AR agonist exposures may cumulatively impair the expression and/or activity of an already decreased receptor population. This may be of particular clinical concern in infants and children with underlying hyperoxic lung injury who receive regular treatments with a β-AR agonist.

Formoterol was chosen for our study design because of ease of administration and clinical concerns raised that chronic formoterol use in humans may cause β-AR tachyphylaxis [29]. In utilizing a long-acting β-AR agonist, we attempted to create a phenotype of chronic stimulation of the β-AR in the developing lung. Future studies are needed to refine this model to correlate with common clinical practices, including drug selection, timing, and frequency. Although there is much practice variability, preterm infants with BPD are often maintained on scheduled β-AR agonists (such as albuterol) in the NICU and are frequently discharged home on bronchodilators [14]. Indeed, repeated stimulation of β-ARs, even with short-acting agonists, has been associated with desensitization of the receptor and a decline in bronchodilator response in asthma [19, 30]. Similar studies are needed in the neonate with and without BPD.

Conclusions

In our immature mouse model, hyperoxia exposure resulted in an increase in airway reactivity. When hyperoxia exposure was coupled with repeated β-AR agonist treatments, animals demonstrated increased airway reactivity at lower doses of methacholine. Furthermore, animals exposed to both hyperoxia and β-AR agonists displayed an attenuated response to rescue bronchodilation. We speculate that such an effect may be related to the associated down-regulation of β-ARs in the lung. Our findings may account for a time-dependent decrease in therapeutic benefit of β-AR agonists with prolonged or repeated use in preterm infants with and without BPD, which may have clinical consequences for patients already prone to airway hyperreactivity and bronchospasm. Additionally, chronic treatment with β-AR agonists in children with hyperoxic lung injury may further sensitize them to provoked airway bronchoconstriction. In cases where chronic bronchodilator therapies are necessary, methods for restoring receptor quantity and function in the developing lung can serve as novel therapeutic targets. Further studies are needed to investigate the effects and safety of prolonged courses of β-AR agonists on the immature human lung.