Advances in our understanding of murine airway physiology have been hindered by the lack of suitable, ex vivo, small airway bioassay systems. In this study, we introduce a novel small murine airway bioassay system that permits the physiological and pharmacological study of intrapulmonary bronchial smooth muscle via a bronchial ring (BR) preparation utilizing BR segments as small as 200 μm in diameter. Using this ex vivo BR bioassay, we characterized small airway smooth muscle contraction and relaxation in the presence and absence of bronchial epithelium. In control BRs, the application of mechanical stretch is followed by spontaneous bronchial smooth muscle relaxation. BRs pretreated with methacholine (MCh) partially attenuate this stretch-induced relaxation by as much as 42% compared with control. MCh elicited a dose-dependent bronchial constriction with a maximal tension (Emax) of 8.7 ± 0.2 mN at an EC50 of 0.33 ± 0.02 μM. In the presence of nifedipine, ryanodine, 2-aminoethoxydiphenyl borate, and SKF-96365, Emax to MCh was significantly reduced. In epithelium-denuded BRs, MCh-induced contraction was significantly enhanced to 11.4 ± 1.0 mN with an EC50 of 0.16 ± 0.04 μM (P < 0.01). Substance P relaxed MCh-precontracted BR by 62.1%; however, this bronchial relaxation effect was completely lost in epithelium-denuded BRs. Papaverine virtually abolished MCh-induced constriction in both epithelium-intact and epithelium-denuded bronchial smooth muscle. In conclusion, this study introduces a novel murine small airway BR bioassay that allows for the physiological study of smooth muscle airway contractile responses that may aid in our understanding of the pathophysiology of asthma.
- bronchial ring
- airway smooth muscle
most in vitro studies of murine airway physiology have used isolated tracheal segments because of the relative technical ease by which these tissues can be retrieved and prepared for measurements of large airway smooth muscle function (24). Typically, these studies have involved the preparation of isolated tracheal strips, usually suspended in tissue baths, where tension is generated and continuously monitored through the use of a force transducer (11, 24). On the other hand, research with murine bronchial small airways is much less common, no doubt because of the increased difficulty of handling these microscopic tissues ex vivo. Nonetheless, studies that utilize small airway bronchial smooth muscle are more likely to accurately represent the physiological and/or pathophysiological events in small airways diseases, such as asthma, simply because of their anatomical proximity to relevant involved tissues. The mechanical responses of large airways, such as the trachea, are hampered by cartilage. Importantly, intrapulmonary bronchioles may differ from larger airways, for example, in the smooth muscle content or in the density of receptors and/or innervation (34). Previous studies have already shown that the lesser amount of cartilage in bronchial smooth muscle results in increased airway wall compliance, which, in turn, enhances the degree of airway narrowing after agonist-mediated airway constriction (25). However, studies with bronchial tissues, especially from small animal models such as mice, have been hampered by the lack of a suitable ex vivo bioassay system.
The myograph system is commonly used in vascular studies for the measurement of contractile responses in vessels (13, 28). To our knowledge, there are only a few studies reported that have applied the myograph system for the study of isolated tracheal and bronchial segments from rats and guinea pigs (31, 33). However, none of these studies have been performed on small murine intrapulmonary airways. In this study, we introduce an ex vivo murine bronchial ring (BR) bioassay that should allow for a better assessment of small airway responsiveness under physiological and pathophysiological conditions.
MATERIALS AND METHODS
Methacholine (MCh), Substance P (SP), nifedipine, ryanodine, papaverine, and xestospongin C were obtained from Sigma Chemical (St. Louis, MO). 2-Aminoethoxydiphenyl borate (2-APB) was obtained from Cayman Chemical (Ann Arbor, Michigan). SKF-96365 was obtained from Calbiochem. Stock solutions of drugs were prepared fresh each day and stored at 4°C during the experiment. All drug concentrations are expressed as final molar concentrations (mol/l) in the chamber superfusate (1).
Wild-type (C57BL/6XC3) F1 mice, 5–10 wk old and weighing 20–25 g, were used in this study. Animals were purchased from Taconic Laboratories (Germantown, NY) and stored in an approved animal facility under the care of a licensed veterinarian at Duke University Medical Center. All protocols and procedures were approved by the Duke University Institutional Animal Care and Use Committee in accordance with National Institutes of Health guidelines.
BR isometric contraction bioassay.
Mice were killed by an intraperitoneal injection of pentobarbital sodium (Nembutal, 80 mg/kg). The chests were opened, and the entire respiratory tree was rapidly removed and immersed in Krebs-Ringer bicarbonate solution (in mM: 118.3 NaCl, 4.7 KCl, 1.2 MgSO4, 1.2 KH2PO4, 25 NaHCO3, 2.5 CaCl2, and 10 glucose). BR, 200–400 μm in diameter, were isolated from mouse intrapulmonary bronchi using a dissection microscope. Isolated bronchial segments, 2 mm in length, were mounted as ring preparations in a small-vessel (18, 19) wire myograph chamber (Danish Myo Technology, Aarhus, Denmark) by threading them on two steel wires (40 μm in diameter) secured to two supports. One support is attached to a micrometer allowing control of ring circumference while the other support is attached to a force transducer for measurements of isometric contraction tension (Danish Myo Technology). Isometric tension is initially calibrated to 0 mN, and the BR is allowed to equilibrate for 10 min. The BR is next stretched by applying a total of 7.5 mN tension (in 3 discrete steps of 2.5 mN tension with 5 min between intervals). After equilibration, baseline tension was recorded. Compounds were then added directly to the chamber, and BR isometric tension was monitored for possible contractile or relaxant effects. The preparation is kept within the chamber, immersed in 6 ml Krebs-Ringer bicarbonate solution (pH 7.35–7.45), bubbled with 21% O2-5% CO2-balance N2, and maintained at 37°C while a Plexiglas cover was placed over the chamber to control oxygen tension in the superfusate. Temperature and BR tension were recorded using a data acquisition and analysis program (Myonic Technology, Aarhus, Denmark).
Removal of epithelium from BR.
BR epithelial cells were removed by gently rubbing the intraluminal surface with a steel wire (40 μm in diameter), followed by perfusion with 2 ml air bubbles and then 2 ml Krebs-Ringer bicarbonate solution (perfusion pressure <5 mmHg), before mounting the tissue in the myograph chamber. This method is similar to that described for removal of endothelium from isolated murine coronary artery or isolated pulmonary artery (17, 19). BRs were stained with hematoxylin and eosin for histological examination.
Dose-challenge response curves were generated by plotting data points in a standard four-parameter logistic curve (Sigma Plot 8.0). Two-way repeated-measures ANOVA and ANOVA Student-Newman-Keul's test (Sigma Stat 2.03) were used for statistical analysis of data as appropriate. P values <0.05 indicate significant differences between two groups. Values given in text are means ± SE, and n equals the number of animals.
Isolating and optimizing BR contractile response to different agonists.
The entire respiratory tree was excised from a mouse and immediately immersed in Krebs-Ringer bicarbonate solution (Fig. 1, A and B). Ring segments (200–400 μm in diameter and 2.0 mm long) of intrapulmonary bronchi were isolated and mounted in the myograph chamber (Fig. 1C). Isometric tension was calibrated to 0 mN, and the BR was allowed to equilibrate for 10 min. The BR was mechanically stretched (preloading tension, Fig. 2A) to optimize agonist-induced muscle contraction. Preload tension appeared most effective when mechanical tension was increased in small increments and allowed to equilibrate to a new baseline tension before it is distended again. These studies show that preloading the BR with a total of 7.5 mN (in 3 discrete steps of 2.5 mN with 5 min between intervals) optimized the smooth muscle contractility as assessed by their response to KCl (Figs. 1D and 3). Sequential stretch of the BR, up to 20 mN total, continued to show optimal contraction to KCl. Stretch >20 mN began to demonstrate reductions in contraction to KCl (Fig. 1D).
Mechanical stretching of the BR in this manner was followed immediately by bronchial relaxation (Fig. 2, A and C). Stretch-induced relaxation, measured as percent dilation from total tension applied, increased from 61.9 ± 1.2 to 64.8 ± 0.9% after the second 2.5-mN stretch and to 66.7 ± 1.0% after the third stretch (Fig. 2C). This bronchial smooth muscle dilation after mechanical stretch was partially abolished if the BR was pretreated with MCh (10 μM) before stretching (Fig. 2, B and C). In BR pretreated with MCh, stretch-induced relaxation was reduced when compared with control, only rising from 25.8 ± 2.8 to 32.5 ± 3.6% after two stretches, and to 54.7 ± 6.9% by the last stretch (P < 0.05; Fig. 2C). In addition, removal of BR epithelium significantly increased stretch-induced relaxation on stretch 1 (78.6 ± 6.5%, P < 0.05) compared with the BR epithelium intact group, but was not different on stretch 2 (64.0 ± 4.8%) or stretch 3 (72.7 ± 7.1%, Fig. 2C).
KCl-induced constriction in isolated mouse BR.
After a total tension of 7.5 mN was loaded, baseline tension was recorded and allowed to equilibrate. KCl (60 mM) was then added in two 30-mM aliquots to assess and standardize depolarization-dependent (i.e., agonist-independent) smooth muscle contractile responses (Fig. 3). The addition of 60 mM KCl increased BR tension to 5.7 ± 0.7 mN above baseline (Fig. 3).
MCh-induced constriction in isolated mouse BR.
To assess viability and to normalize BR contractility, BRs were exposed to 30 and 60 mM KCl, followed by three washes with fresh Krebs-Ringer. After washing (15 min), MCh (0.01–10 μM) was added to the chamber in a cumulative fashion to establish BR contractility to the agonist. The isometric contractile response of the BR to cumulative MCh doses (0.1–10 μM) was assessed (Fig. 4A). MCh induced a concentration-dependent increase in tension, reaching a maximal tension (Emax) of 8.7 ± 0.2 mN at an EC50 of 0.33 ± 0.02 μM (Fig. 4B). Maximal bronchial contraction was induced by 10 μM MCh. Larger doses of MCh (30 or 100 μM) did not elicit significantly higher responses from the isolated BR (data not shown).
We used nifedipine, ryanodine, and 2-APB to characterize the effects of calcium on MCh-induced BR contraction in our ex vivo bioassay system. In BR pretreated with nifedipine (10 μM, a voltage-dependent L-type calcium channel inhibitor), the maximal contractile response to MCh was lowered by 51.5% (Emax of 4.5 ± 0.6 mN) with an EC50 of 0.85 ± 0.24 μM (Fig. 4B). Ryanodine (10 μM, an inhibitor of intracellular ryanodine-sensitive calcium release) diminished MCh-induced Emax by 67.1% to 2.9 ± 0.6 mN with an EC50 of 0.75 ± 0.24 μM (Fig. 4B). The combinatorial effect of nifedipine (10 μM) plus ryanodine (10 μM) was even more pronounced, reducing the isolated BR contractile response to MCh by 86.8% (1.2 ± 0.3 mN) with an EC50 of 0.98 ± 0.19 μM compared with control (P < 0.05; Fig. 4B). Because most of the previous publications that study bronchial and tracheal rings have been done under 95% O2 conditions, we have conducted additional studies comparing BR responsiveness at 21 vs. 95% O2 (Fig. 4C). As shown in Fig. 4C, BR treated with 95% O2 did not significantly alter the MCh-induced BR contractile responses with or without pretreatment with nifedipine, ryanodine, or the combination of nifedipine plus ryanodine.
The addition of 100 μM 2-APB [an inositol 1,4,5-trisphosphate (IP3) receptor and/or store-operated calcium channel inhibitor (36)] markedly decreased MCh-induced Emax by 71.3% to 3.0 ± 0.9 mN with an EC50 of 0.78 ± 0.22 μM (Fig. 5A). Xestospongin C, a cell-permeant IP3 receptor antagonist (27), and SKF-96365, an inhibitor of store-operated calcium channels (37), were used to determine the underlying mechanism of the MCh-induced contractile response. Xestospongin C did not affect the MCh-induced constriction (Fig. 5B). SKF-96365, however, markedly reduced MCh-induced constriction (Fig. 5C), suggesting that MCh-induced mouse bronchial smooth muscle constriction is mediated by store-operated calcium channels rather than via IP3 receptors.
We assessed the reproducibility of the ex vivo BR responses to agonist-dependent and -independent contraction. BR contractility to KCl (60 mM) and MCh (0.1–10 μM) was tested repeatedly (Fig. 6). These results demonstrate that both KCl- and MCh-induced constriction in the bronchial smooth muscle were virtually identical between sessions and were, therefore, highly reproducible (Fig. 6).
Role of epithelium in BR reactivity.
To test whether airway epithelium can modulate small airway bronchial smooth muscle responsiveness to MCh, we used our newly adapted small vessel myograph to assess the BR contractile response after removal of bronchial epithelium (Fig. 7A). Our results demonstrate that bronchial contraction in epithelium-denuded BR (Emax = 11.4 ± 1.0 mN) was significantly increased compared with epithelium-intact BR (Emax = 8.7 ± 0.2 mN; Fig. 7B).
To study the role of the epithelium on BR relaxation, BRs with and without epithelium were precontracted with MCh (10 μM) and subsequently treated with SP (1 μM) followed by papaverine (30 μM). Under normal conditions, the administration of SP (10 μM) reduced MCh-induced BR constriction by ∼62.1%, whereas papaverine (30 μM) virtually abolished MCh-induced contraction by almost 100% (P < 0.05; Fig. 7C). In epithelium-denuded BRs, SP was unable to elicit any relaxation, highlighting the critical role bronchial epithelium plays in transducing SP effects. Papaverine, on the other hand, continued to demonstrate complete relaxation, even in epithelium-denuded BR (Fig. 7C).
It is well established that airway smooth muscle from different locations in the conducting airways of the respiratory tree can elicit diverse responses to pharmacological stimuli (5). Marthan and colleagues (22) have indicated that there are spatial differences in the types of ion channels and pathways of calcium signaling involved in the mechanical activity of airways throughout the bronchial tree. Indeed, as indicated by others, the distribution and innervations of different receptors can differ between larger airways, such as the trachea and bronchi, and smaller distal airways, such as the bronchiole (34). Most data on airway smooth muscle function are based on ex vivo tension development and contractility of isolated airway tissues under different pathophysiological conditions (14, 26). These studies, especially those performed on small animal models, generally use tracheal and occasionally main bronchi segments as their primary focus of study (10). To our knowledge, there are no data on the use of intrapulmonary bronchi from mice, whereas data from other small animal models, such as rats, is scarce (34).
Previous studies on isometric tension development from isolated smooth muscle segments from different animal models have determined that preloading the tissue with an optimal loading tension enhances their contractility (22). In one study, Van de Voorde and Joos (34) reported that preloading rat trachea and main stem bronchial segments of 2–3 mm in length with 1.5 mN/mm optimized their subsequent isometric contraction studies. The optimal stretch required for rat intrapulmonary bronchiole segments of 1–2 mm in length was shown to be 0.8 mN/mm. In adapting our small vessel myograph system for studies on BR isometric tension, we have determined that preloading the mouse BR with a mechanical tension of 3.75 mN/mm, applied in 2.5-mN increments, optimizes the contractile responses. It is interesting to note that our maximal contractions were five- to sixfold higher in our mouse BR preparations compared with the rat BR preparations reported by Van de Voorde.
Under normal conditions, loading tension on BR by stretching the murine bronchial smooth muscle in vitro (similar to deep inspiration in vivo) did not induce myogenic constriction, but it was followed by bronchial relaxation. This is consistent with reports from Noble et al. (25) that indicate deep inspiration transiently dilates the airways. This stretch-induced relaxation of airway smooth muscle has been previously reported by investigators studying isolated sheep tracheal strips and canine tracheal smooth muscle strips (6, 31). In isolated sheep tracheal strips, Thulesius and Mustafa (31) observed that pretreatment with the muscarinic bronchoconstrictor carbachol (0.01 μM) and histamine (100 μM) prevented this stretch-induced bronchial relaxation and induced a protracted myogenic contraction. To some extent, in our mouse BRs, pretreatment with MCh (10 μM) partially inhibited this stretch-induced relaxation. These results suggest that the protracted myogenic contraction, evident in tracheal smooth muscle pretreated with cholinergic agonists, is also present in our smaller bronchial conducting airways, a feature that might be relevant to asthma.
Calcium activity (extracellular calcium influx and/or intracellular calcium release) during MCh- or Ach-induced airway smooth muscle constriction has been studied for many years. Most studies have suggested that muscarinic receptor agonists induce tonic contraction in airway smooth muscle by causing transient increases of intracellular calcium through calcium release from the sarcoplasmic reticulum (SR) after IP3 formation (12). Marthan et al. (21) reported that ACh was able to elicit contractions in isolated human bronchial smooth muscle after 20 min of calcium-free perfusion and suggested that an intracellular calcium store may be the major component in human bronchoconstriction. Furthermore, in mouse tracheal studies, tumor necrosis factor-α and interleukin-13 have been shown to enhance airway smooth muscle contraction by modulating G protein-coupled signal transduction, ultimately increasing calcium signaling during agonist-mediated contraction (2, 32).
Our results, however, demonstrate that extracellular calcium influx and intracellular calcium release (which was mediated by store-operated calcium release) provide approximately equivalent contractile impetus during MCh-induced BR constriction. MCh-induced bronchial constriction was significantly reduced when the rings were pretreated with nifedipine, ryanodine, or 2-APB. MCh-induced BR constriction was virtually completely inhibited when our isolated mouse BR were treated with both nifedipine and ryanodine (Fig. 4B). Xestospongin C did not affect MCh-induced constriction (Fig. 5B), whereas SKF-96365 markedly reduced the MCh-induced constriction (Fig. 5C). Taken together, these results suggest that MCh-induced constriction in isolated mouse BR is mediated by both an extracellular calcium influx (primarily via voltage-dependent L-type calcium channels) and an intracellular calcium release (via ryanodine-sensitive calcium storage and/or store-operated calcium releases). Extracellular calcium influx (via voltage-dependent calcium channels) may enhance intracellular calcium release through intracellular calcium stores. Further studies are required to elucidate the mechanism by which extracellular calcium influx induces intracellular calcium release during MCh-induced BR constriction.
It is well known that bronchial epithelium can regulate airway smooth muscle function. One of the several ways in which the epithelium can modulate airway smooth muscle reactivity to contractile agonists is by releasing epithelium-derived relaxing factors that reduce smooth muscle reactivity by hyperpolarizing the membrane (7). Using our ex vivo murine model, we studied isometric contraction in epithelium-free BR to elucidate mechanisms by which bronchial epithelium might modulate smooth muscle sensitivity to MCh-induced contraction. Our results demonstrate that epithelium-denuded BR have significantly higher contraction and increased sensitivity to MCh compared with epithelium-intact BR. These results are mostly consistent with studies by Vanhoutte (35) who have shown that removing the epithelium in canine bronchi potentiates the smooth muscle reactivity to histamine, 5-hydroxytryptamine, and ACh, without altering the maximal responsiveness to these bronchoconstrictors. Thus damage to the epithelial layer in distal small airways is predicted to exacerbate bronchoconstriction and may contribute to the phenotypic development of airway diseases such as asthma.
Tachykinin neuropeptides, such as SP, have potent but variable effects on bronchial smooth muscle tone and epithelial cell function (23). Prior studies have shown that administration of SP to isolated human airways induces smooth muscle relaxation, presumably through NK1-receptor activation (4). However, when given to asthmatic subjects, aerosolized SP results in bronchoconstriction and airway hyperresponsiveness (3, 9). The mechanism by which SP exerts this differential effect on airway smooth muscle tone has not been fully elucidated (30). In our current study, we show that SP relaxes MCh-contracted BR and that this bronchodilation effect is completely dependent on intact epithelium. This finding is consistent with previous studies demonstrating that SP and ATP can evoke epithelium-dependent relaxation of rat trachea and that the removal of the epithelium in the tracheal smooth muscle abolishes this relaxant response (8, 9). In murine main stem bronchi, SP-induced bronchodilation effects can be blocked by selective NK1-receptor antagonists and by the addition of indomethacin (20). In rat bronchi, SP-induced relaxation was found to be mediated by epithelial cell-derived PGE2 release (16, 23, 30).
Papaverine, on the other hand, is a potent nonspecific smooth muscle relaxant whose mechanism of action remains to be fully elucidated. It is commonly thought that papaverine acts as a vasodilator by increasing smooth muscle intracellular cAMP levels through its action as a nonselective cAMP phosphodiesterase inhibitor (29). Studies by Iguchi et al. (15) have also indicated that papaverine might actually induce smooth muscle relaxation in guinea pig by inhibiting voltage L-type calcium channels and calcium-activated oscillatory K+ currents. In our experiments, papaverine (30 μM) abolished MCh-induced constriction in epithelium-intact and epithelium-denuded BR, indicating that its mechanism of smooth muscle relaxation is independent of epithelium-derived factors.
In summary, our mouse bioassay system provides a suitable ex vivo model for the study of murine small airway physiology. We have successfully adapted a small vessel myograph system for the study of small bronchial smooth responsiveness to agonist-mediated contraction and relaxation. We anticipate that the use of this ex vivo BR bioassay system should further our understanding of factors regulating airway smooth muscle tone under both normal and pathological conditions such as those found in asthma.
This work was supported, in part, by National Institutes of Health Grants ES-08698 and HL-64894 and the American Heart Association Grant-in-Aid.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2006 the American Physiological Society