The physiologic roles of ERs in the lung are largely unknown; hence, we examined lung function and airway hyperresponsiveness in ER-deficient mice. The results described herein implicate a major role for ERα in modulating lung function and airway hyperresponsiveness, and describe a potential mechanism by which ERα mediates airway responsiveness.
There is considerable evidence supporting a role for sex hormones in the neural control of breathing (24
). Breathing disorders such as obstructive sleep apnea have been linked to sex hormone levels (24
). There is an increase in sleep-disordered breathing after menopause, which can be alleviated by hormone replacement therapy (25
). Respiratory rhythm is generated by medullary neurons in the brainstem (27
), a site in which ERα has been shown to be abundantly expressed (27
). Interestingly, we found a marked reduction in breathing frequency in male and female αERKO mice relative to wild-type control animals. Male wild-type mice have a significantly higher tidal volume than female wild-type mice; however, this pattern is reversed in αERKO mice. Similarly, minute ventilation, peak inspiratory flow, and peak expiratory flow are higher in male versus female wild-type but not αERKO mice. Together, these data indicate that functional disruption of ERα leads to changes in a variety of respiratory parameters and suggest that this nuclear receptor may be a critical regulator of breathing and respiratory rhythmogenesis in mice.
ERβ disruption has no influence on sex differences in tidal volume, minute ventilation, peak inspiratory flow, and peak expiratory flow. However, breathing frequency is significantly lower and peak inspiratory flow is significantly higher in female βERKO relative to female wild-type mice. Tidal volume is higher in both male and female βERKO mice relative to their respective wild-type controls. Consistent with this observation, Massaro and Massaro recently reported that βERKO mice have a higher body mass–specific lung volume relative to wild-type mice (30
). These data suggest that ERβ does play a role in the regulation of breathing, albeit a much less dominant role than ERα.
Airway hyperresponsiveness to cholinergic stimuli is a cardinal feature of asthma and a major risk factor for accelerated decline of lung function and development of COPD in humans (31
). The exact mechanism(s) underlying the development of airway hyperresponsiveness in chronic lung diseases such as asthma remains unknown. Several studies of risk factors associated with airway hyperresponsiveness have reported higher responsiveness in females compared with males (32
), suggesting the involvement of sex hormones in the pathogenesis. Herein we demonstrate that in the absence of immunologic stimulation, αERKO female mice exhibit substantially enhanced airway responsiveness to inhaled methacholine compared with wild-type females, suggesting that ERα is a critical regulator of this process.
Traditionally, airway hyperresponsiveness has been presumed to mainly involve the central airways and not the periphery. However, physiologic and pathologic evidence has emerged in recent years to support the role of the lung parenchyma and distal airways in the pathogenesis of airway hyperresponsiveness (35
). Airway hyperresponsiveness is influenced by properties of the central airways and the surrounding pulmonary parenchyma, which is tethered to the airways, and by interactions between these two compartments (39
). The exact location and precise mechanism for changes in tissue resistance are controversial, but many hypotheses have been proposed, including contraction of parenchymal interstitial cells, contraction of smooth muscle cells within alveolar ducts, and changes in the architecture of the alveoli and alveolar ducts (40
). It has also been suggested that parenchymal changes could be secondary to airway narrowing, either by direct interaction between the airways and parenchyma or indirectly by altering lung volume (39
). Invasive measurement of lung function in the αERKO mice at baseline revealed hyperresponsiveness primarily in the periphery. After allergen challenge, there was marked hyperresponsiveness in both the central and peripheral airways. Interestingly, Massaro and Massaro recently reported that ERs are required for the formation of a full complement of alveoli in female mice (30
). Thus, it is possible that structural abnormalities related to formation and size of alveoli may play a role in the abnormal hyperresponsive phenotype observed in these mice. Alternatively, the parenchymal defect in the αERKO mice could be secondary to airway narrowing.
The reduced airway responsiveness to methacholine after ovariectomy of the αERKO mice suggests that ovarian products may play a role in the hyperresponsive phenotype. However, the lack of an effect of estrogen supplementation or ovariectomy on airway responsiveness in wild-type mice suggests that estrogen alone is not the only culprit. One possible explanation for these findings may be that the absence of ERα allows ERβ to predominate in this model. ERα and ERβ have distinct expression patterns, with some organs having a predominance of one receptor over the other and other tissues having comparable expression of both receptors (16
). Studies with αERKO and βERKO mice have revealed that these receptors have both overlapping and unique (and sometimes opposite) roles in mediating estrogen-dependent action in vivo
. Both receptors are expressed in the lung, with ERβ levels being higher than ERα levels (16
). Studies have revealed that there is a complex interplay between ERα and ERβ in the regulation and autoregulation of their respective promoters (44
) and that some of the biological functions of one receptor may be dependent on the presence of the other receptor (16
). In the present study, estrogen may produce different effects via ERβ, and the observed hyperresponsive phenotype in αERKO mice may be due to an altered estrogen response rather than an absent one, as has been postulated with other phenotypes displayed by these mice (46
). An alternative possibility could be that the hyperresponsive phenotype is driven by an ovarian product other than estrogen, such as an androgen. In this regard, the ovaries of the αERKO mice produce 17β-hydroxysteroid dehydrogenase type III, an enzyme normally found only in testes, which converts androstenedione to testosterone. Indeed, αERKO females have elevated plasma levels of androgens, similar to those seen in wild-type males (23
Estrogen has been shown to modulate the density of muscarinic receptors in vivo
in extrapulmonary tissues (49
), but there are no reports of estrogen modulation of muscarinic receptor expression or function in the lung. Importantly, expression of the M2 muscarinic receptor is markedly reduced in αERKO female mice relative to wild-type control mice. Consistent with this finding, tracheas from αERKO female mice release more ACh in response to electrical field stimulation than tracheas from wild-type control mice. Furthermore, the lack of effect of gallamine, a selective M2 muscarinic receptor antagonist, on the contractile response of αERKO tracheas to electrical field stimulation conclusively demonstrates M2 muscarinic receptor dysfunction in these mice. Together, these data indicate that one potential mechanism for airway hyperresponsiveness in αERKO female mice could be down-regulation of M2 muscarinic receptor expression and function leading to increased ACh in the neuromuscular junction and resulting in enhanced bronchoconstriction after cholinergic agonist stimulation.
In light of our finding that αERKO mice are also hyperresponsive to inhaled serotonin, it is possible that a reflex mechanism may be contributing to the generalized airway hyperresponsiveness in these mice. It is generally assumed that the response to methacholine reflects only a direct effect of the agonist on airway smooth muscle. However, studies suggest that a substantial component of the airway smooth muscle response to cholinergic agonists depends on a vagally mediated reflex (51
). For example, administering cholinergic agonists increases the firing of sensory nerves in the lung, and vagotomy decreases the bronchoconstrictive response to methacholine (52
). Studies suggest that, in the mouse, the respiratory response to serotonin also involves a vagal reflex (54
). Blockade of muscarinic M2 receptors with the subtype selective antagonist gallamine potentiates vagally induced bronchoconstriction by increasing ACh release (55
). In addition, dysfunctional muscarinic M2 receptors have been shown to enhance reflex bronchoconstriction (56
). Thus, if reflex bronchoconstriction is contributing to the enhanced airway hyperresponsiveness to methacholine and serotonin, then dysfunctional M2 receptors could be playing a major role by leading to enhanced ACh release after vagal stimulation. The in vitro
response to carbachol may reflect an additional smooth muscle defect and this may or may not contribute to the airway hyperresponsiveness observed in the intact animal. The combined defects at the level of smooth muscle and nerve may explain why the hyperresponsiveness is so prominent in the αERKO female mice.
Although deletion of ERα has minimal effects on airway inflammation after allergen challenge, it has a profound effect on the development of allergen-induced airway hyperresponsiveness. Indeed, the differences in airway responsiveness between wild-type and αERKO mice are even more pronounced after exposure to ovalbumin. Although airway inflammation is frequently correlated with airway hyperresponsiveness to methacholine, and treatment of inflammation often improves hyperresponsiveness, dissociation between airway inflammation and hyperresponsiveness has been observed in other murine models of allergic airway disease (57
). Of note, allergen challenge induced minimal changes in airway responsiveness in wild-type mice in our study. This is not surprising because other investigators have shown that the C57BL/6 strain is one of the least responsive strains in terms of the development of airway responsiveness after ovalbumin sensitization and exposure (59
). Indeed, the fact that allergen challenge induced such a profound degree of airway hyperresponsiveness in αERKO mice on a C57BL/6 background indicates that ERα is a potent regulator of this process.
Although targeted disruption of specific genes in mice offers a powerful tool to investigators, caution must be exercised in extrapolating from these studies to physiologic effects in humans. Knockout of a gene in mice may produce different effects than inactivation of the same gene in humans. In addition, targeted disruption in mice may lead to compensatory changes in expression of other genes, which may hamper interpretation of the results. Nevertheless, it is of interest to speculate on the potential clinical significance of our findings. There are several polymorphic sites in the human ERα locus and some of these polymorphisms have been associated with diseases such as cancer and osteoporosis (61
). Interestingly, in some instances, the phenotype associated with the polymorphism is also dependent on high levels of estrogen (62
). The role of hormone replacement therapy in the treatment and prevention of cardiovascular disease has been highly controversial due to conflicting findings. It has been suggested that variation in estrogen effects on the cardiovascular system may be related, at least in part, to common variants in the ERα gene, which can significantly alter the way a person responds to endogenous and exogenous estrogen (63
). It is possible that the conflicting human data on estrogen effects in the lung could also be due to genetic variability in ERα. Indeed, genetic variability in ERα with resultant effects on estrogen sensitivity could underlie the greater incidence and severity of asthma and airway hyperresponsiveness in females between the ages of puberty and menopause when circulating estrogen levels are high, and could also underlie the phenomenon of premenstrual asthma as estrogen levels fluctuate during the menstrual cycle. Of interest, the relevance of our findings to humans is supported by a recent publication by Dijkstra and coworkers who found that variations in ERα were associated with airway hyperresponsiveness and more rapid lung function decline in subjects with asthma (64
In conclusion, our data suggest that, in the mouse, lack of ERα leads to airway hyperresponsiveness via defects at the level of airway smooth muscle and nerves, and may involve regulation of M2 muscarinic receptor expression and function. Our data also suggest that the αERKO mouse may prove to be a useful model to study the mechanisms of sex hormone modulation of airway responsiveness in humans.