INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Asthma is generally characterized by airways hyperresponsiveness, chronic inflammation, and reversible airways obstruction (1). Disease severity has long been accepted to correlate with the degree of airway hyperresponsiveness (2) yet the etiology of hyperresponsiveness is still poorly understood. There is compelling evidence that airways inflammation may provoke the secondary development of hyperresponsiveness in allergic, occupational, and ozone-induced asthma (1, 3), but exercise- and cold air-induced asthma can develop without prior inflammation (1, 4). As well, segregation and linkage analyses in humans (5, 6) and mice (7, 8) have indicated that airway responsiveness is a polygenic trait that segregates independently from atopy (9). Thus, airway hyperresponsiveness appears to be a component of asthma that may interact with, but be independent of, other characteristics of the condition.

Enhanced airway smooth muscle contractility has been widely suspected to be an important determinant of airway hyperresponsiveness in asthma, but inconsistencies between the behaviour of asthmatic tissue in vivo and in vitro have put into question this association. Contractile agonists have been observed to evoke hyper- (10), normo-, and even hyporesponsiveness (13) in various preparations of excised asthmatic airway smooth muscle. These discrepancies may partly be attributable to the difficulty in controlling for confounding variables affecting experimental samples because all the asthmatic tissue was collected postmortem or from lungs with another preexisting disease. In addition, most of these investigators measured isometric force development as the index of responsiveness, but in animal models of allergic airway hyperresponsiveness, sensitization has no effect on isometric force generation; airway smooth muscle shortening capacity and velocity are increased instead (9, 14). Consequently, these parameters are proposed to be the relevant markers of in vitro airway smooth muscle responsiveness (15). In agreement with this, shortening velocity has been reported to be faster in asthmatic than in nonasthmatic bronchial preparations (12).

It is plausible that in vivo airway hyperresponsiveness could arise from intrinsically exaggerated contractility of airway smooth muscle. Passively sensitized airway smooth muscle excised from humans and animals displays hyperresponsive contractile behavior (16), confirming that a hyperresponsive phenotype can be observed in vitro. However, while these results provided a causal link between allergic sensitization and pathophysiological changes in airway smooth muscle, it is also possible that natural variations in the properties of airway smooth muscle may underlie airway hyperresponsiveness in the absence of inflammation and may predispose a subject to develop disease in conjunction with other risk factors.

To examine the mechanisms of hyperresponsiveness in normal animals, a model of spontaneous, innate airway hyperresponsiveness was characterized based on findings that responsiveness is a heritable, polygenic trait. A number of normal, highly inbred rat strains were screened for their responsiveness to inhaled methacholine, and Fisher 344 rats were found to be the most responsive whereas Lewis rats were the least responsive of all the strains studied (20). Morphometric studies demonstrated that Fisher rats had more smooth muscle in their airways, but increased smooth muscle mass alone did not appear to entirely account for this strain's enhanced responsiveness (20). Isolated tracheal rings from Fisher rats were subsequently found to also be more sensitive and to generate greater maximal isometric tension in response to carbachol than rings from Lewis rats (21). Furthermore, the difference in responsiveness of excised airways was not limited to the trachea, but was also observed in isolated Fisher intraparenchymal airways which constricted to a greater degree and at a faster rate to methacholine than Lewis airways (22). These observations suggest that differences between Lewis and Fisher in vivo responsiveness may be associated with differences in the contractile properties of their airway smooth muscle.

Airway smooth muscle contracts when external stimuli trigger an increase in cytosolic [Ca²⁺]_i. Ca²⁺ binds to calmodulin to activate myosin light chain kinase, which phosphorylates myosin and thereby initiates actinomyosin crossbridge cycling and hence contraction (23). Because the increase in cytosolic [Ca²⁺]_i is the signaling event where all modalities of contractile stimulation (i.e., depolarization, receptor-mediated, mechanotransduction) first converge, alterations of this step could be a common abnormality in all types of nonspecific airway hyperresponsiveness. In support of this idea, altered smooth muscle Ca²⁺ regulation seems to be present in various models of hypertension. Basal and agonist-stimulated levels of Ca²⁺ have been reported to be higher in hypertensive vascular smooth muscle than in normal control sprecimens (24, 25).

We propose that differences in in vivo airway responsiveness among different rat strains originate in part from differences in the contractility of their airway smooth muscle. We further postulate that differences in intracellular Ca²⁺ mobilization may account for disparities in airway smooth muscle contractility. These hypotheses were tested by relating responsiveness in vivo with dynamic airway narrowing ex vivo and with airway smooth muscle Ca²⁺ mobilization in Fisher and Lewis strains of rat as well as in ACI rats, a rat strain of intermediate responsiveness.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals

Highly inbred, 7- to 9-wk-old male Fisher 344, ACI, and Lewis rats were obtained from Harlan-Sprague-Dawley (Indianapolis, IN) and housed at a conventional animal facility at McGill University (Montreal, PQ, Canada).

Measurement of Airway Responsiveness In Vivo

Rats were anesthesized intraperitoneally with a combination of xylazine (7 mg/kg) and sodium pentobarbital (35 mg/kg). Once sedated, the animals were intubated by inserting 6 cm of polyethylene tubing into their tracheae. A water-filled polyethylene catheter (PE-160) was then placed in the lower esophagus for measuring transpulmonary pressure (PL) and the rats were positioned in the lateral decubitus posture with the open end of the endotracheal tube inserted into one port of a plexiglass chamber. The esophageal catheter was attached to one port of a differential pressure transducer (Transpac II; Abbott, Chicago, IL), the second port of which was referenced to the plexiglass chamber. A 0 size Fleisch pneumotachograph connected to a piezoresistive differential pressure transducer (Micro-Switch 163PC01D36; Honeywell, Scarborough, ON, Canada) was attached to the plexiglass chamber in order to measure flow. Pressure and flow signals were amplified, filtered, and recorded by a 12-bit analogue-to-digital converter and analyzed with a commercially available software package (RHT-Infodat Inc., Montreal, PQ, Canada). Lung resistance (RL) was calculated by applying multiple linear regression to fit the equation of motion of the lung using 10-s epochs of spontaneous tidal breathing (26).

5-Hydroxytryptamine (5HT) was aerosolized at 8 L/min for 30 s into the plexiglass chamber using a Hudson nebulizer (Hudson Respiratory Care Inc., Temecula, CA). After recording baseline and aerosolized saline control responses, the 5HT challenge started with 0.0625 mg/kg followed by progressively doubling concentrations until the baseline pulmonary resistance at least doubled. PL and flow were recorded at baseline and after each dose of 5HT, and the concentration required to double RL (EC₂₀₀RL) was determined by interpolation.

Lung Explant Cultures

Preparations of lung explants were made as previously reported (22, 27). The lungs were surgically excised and were inflated to a standard volume (50% of total lung capacity) with buffered culture medium (BCM) (9.62 g/L Eagle's minimum essential medium powder [MEM] with Earle's salts and L-glutamine supplemented with 2.2 g/L sodium bicarbonate, 20 ml/L of 50× MEM amino acid solution, 10 ml/L sodium pyruvate, 10 ml/L of 100× vitamin solution, 0.1 µg/ml vitamin A, 0.1 µg/ml hydrocortisone, and 50 µg/ml gentamicin) containing 1% agarose type VII. The lungs were then cooled to 4° C for 30 min to allow the agarose to gel inside the airways, and the lobes were separated and embedded upright in a syringe filled with MEM containing 4% agarose type VII. After cooling at 4° C for another 30 min, the lobes were sliced cross-sectionally into 0.5- to 1-mm-thick explants by a hand-held microtome blade (Cambridge Instruments, Buffalo, NY). Agarose was removed from the airways immediately after slicing. Explants were cultured in BCM and incubated overnight in a 5% CO₂ atmosphere.

Measurement of Dynamic Airway Constriction in Lung Explants

Suitable explants were screened for study based on the presence of clearly visualized, agarose-free, cross-sections of airways when viewed under ×10 magnification of an inverted microscope. Clean airways were characterized by intact epithelium with beating cilia. The chosen explants were individually transferred to single wells of six-well culture plates containing 500 µl HEPES buffered culture medium (HCM) (the same constituents as BCM except that 5.96 g/L HEPES replaced sodium bicarbonate) and viewed under ×4 magnification on the stage of an inverted microscope (model IMT-2; Olympus, Tokyo, Japan). The microscopic images of the explanted airways were observed on a video monitor and acquired onto optical disk via a CCD-200-R camera (Videoscope International, Washington, DC) and optical disk recorder (model TQ 2026F; Panasonic, Osaka, Japan). Images were recorded immediately before administration of 5HT, then at every second for the first 15 s, at 20s, 25 s, 30 s, 60 s, and finally at every minute until 5 min postagonist. The agonist was applied directly onto the airway of interest. The images were digitized with an 8-bit frame grabber (model PIP1024B; Matrox, Montreal, PQ, Canada) and the internal airway luminal perimeters were measured with Galileo software (Inspiraplex, Montreal, PQ, Canada). Explants were maintained at 37° C throughout the experiment.

Responses in explanted airways were compared between rat strains in terms of the peak constriction, the average velocity of airway constriction from rest to 50% of the peak constriction, and the time to reach peak constriction. Peak constriction was defined as the greatest reduction in airway luminal area from baseline during the 300-s recording period and was obtained from A = P_i²/4, where P_i is the internal perimeter. The average velocity of constriction was calculated as the change in luminal area over time from rest to 50% of the peak constriction (A_i/t). Both parameters were normalized to baseline A_i to eliminate variability related to baseline airway size.

In experiments designed to define the different contributions of intracellular and extracellular Ca²⁺ to airway constriction, narrowing was observed as previously described in the presence of Ca²⁺-free Hanks' buffer (137.7 mM NaCl, 4.2 mM NaHCO₃, 10 mM glucose, 3 mM Na₂HPO₄, 5.4 mM KCl, 0.4 mM KH₂PO₄, 0.5 MgCl₂, 0.8 mM MgSO₄, 5 mM HEPES, and 1 mM ethyleneglycol-bis-(-aminoethyl ether)- N,N'-tetraacetic acid [EGTA]).

Airway Smooth Muscle Cell Culture

Airway smooth muscle cells were cultured as previously described (28). Fisher, ACI, and Lewis rats were overdosed with sodium pentobarbital and their tracheae were aseptically excised. After removal of visible fascia, the tracheae were washed in Hanks' balanced salt solution (HBSS) composed of 5 mM KCl, 0.3 mM KH₂PO₄, 138 mM NaCl, 4 mM NaHCO₃, and 5.6 mM Na₂HPO₄. The tracheae were cut longitudinally and the trachealis muscle was dispersed by digestion with 0.2% collagenase IV and 0.05% elastase IV in HBSS for 30 min at 37° C. Smooth muscle cells were isolated from the supernatant by centrifugation at 500 g for 6 min and the pellet was resuspended in 1:1 Dulbecco's modified Eagle's medium (DMEM):Ham's F12 fortified with 10% fetal bovine serum (FBS), 100 U/ml penicillin, and 100 µg/ml streptomycin. The cells were plated and incubated in a humidified 5% CO₂ environment at 37° C.

When the cells grew to confluence, they were detached from the plates with 0.25% trypsin and 0.02% ethylenediaminetetraacetic acid (EDTA) and subcultured onto 25-mm-diameter round glass coverslips at a seeding density of 5,000 cells/ml, 4 ml/well, in 6-well plates. Only confluent, first passaged cells were used for experimentation. The cells were characterized as smooth muscle by typical morphology, by positive immunohistochemical staining with murine smooth muscle specific anti--actin and by the presence of contractile responses when exposed to 5HT.

Spectrofluorimetric Imaging of Calcium

Cells were loaded with 5 µM fura-2 acetoxymethyl ester (AM) and 0.02% pluronic acid in Hanks' buffer (137 mM NaCl, 4.2 mM NaHCO₃, 10 mM glucose, 3 mM Na₂HPO₄, 5.4 mM KCl, 0.4 mM KH₂PO₄, 1.3 mM CaCl₂, 0.5 mM MgCl₂, 0.8 mM MgSO₄, and 5 mM HEPES) for 30 min at 37° C. The cells were washed with Hanks' buffer and rested for 10 min to allow for full deesterification of fura-2AM (25, 29). Each coverslip was subdivided into sections by affixing three rings with vacuum grease to the central area in order to increase the number of measurements made on each coverslip. The coverslips were mounted in a Leiden chamber (Medical Systems Corp., Greenville, NY) then onto the stage of an epifluorescence-equipped inverted microscope with a ×40 oil immersion objective (Nikon, Tokyo, Japan). Cells were excited at wavelengths of 340 and 380 nm with a PTI Deltascan 1 dual monochromator illuminator (Photon Technology International Inc., Princeton, NJ) and fluorescence emissions were detected at 510 nm by a PTI D104 microphotometer. Ca²⁺ responses in individual cells were detected by imaging a group of single cells within a ring with a CCD200R videoscope camera (Videoscope International, Herndon, VA) and were subsequently analyzed with PTI software. 5HT was added separately to each ring and fluorescence emission was recorded every 2 s during imaging. The cells were maintained at 37° C throughout the recording.

Free intracellular Ca²⁺ concentration was obtained from the following equation (30):

where R is the ratio of Ca²⁺-free to Ca²⁺-bound fura-2 fluorescence in the sample while R_min and R_max are the same ratios obtained by calibrating Ca²⁺-free and Ca²⁺-saturated cells, respectively.  is the ratio of fura-2 fluorescence of Ca²⁺-free to Ca²⁺-saturated cells when they are excited at 380 nm. R_min was 0.64, R_max was 6.74, and  was 16.7. The K_d was 113 nM by in vitro calibration.

Reagents

Almost all the reagents for lung explant and cell culture were purchased from GIBCO (Burlington, ON, Canada) except for sodium pyruvate, hydrocortisone, agarose, collagenase IV, and elastase IV (Sigma Chemical Co., St. Louis, MO), Ham's F12 (Flow Lab, Mississauga, ON, Canada); trypsin (Worthington, NJ), and fura-2AM and pluronic acid (Molecular Probes, Eugene, OR). All other general biochemical reagents were supplied by Sigma Chemical Company.

Statistical Analysis

The Kolmogorov-Smirnoff test was used to test the difference between cumulative frequency distributions of the lumen size of explanted airways and the velocity of airway narrowing as well as of increases in airway smooth muscle [Ca²⁺]_i.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Pulmonary Resistance to 5HT In Vivo

The changes in pulmonary resistance from baseline levels in response to progressively doubling concentrations of aerosolized 5HT were measured in Lewis, ACI, and Fisher rats in vivo (Figure 1). The geometric mean EC₂₀₀RL to 5HT was 16.51 mg/ml in Lewis, 8.16 mg/ml in ACI, and 1.13 mg/ml in Fisher rats. The EC₂₀₀RL of Fisher rats was significantly lower than that of Lewis rats (p < 0.001) and of ACI rats (p < 0.005). There was no significant difference between the EC₂₀₀RL of Lewis and ACI rats.

View larger version (14K): [in this window] [in a new window]   Figure 1.   Effect of 5HT on lung resistance in vivo. Aerosolized 5HT was administered to anesthesized rats in doubling concentrations until RL at least doubled above baseline levels. Base = baseline, Sal = saline. Open circles: Lewis (n = 6); open triangles: ACI (n = 5); closed circles: Fisher (n = 6). Results show mean ± SEM.

Airways Constriction to 5HT Ex Vivo

Lung explants were used to compare the ex vivo responsiveness of intraparenchymal airways from Lewis, ACI, and Fisher rats. Explanted airways from all three rat strains were treated with 0.1 to 10 µM 5HT and their dynamic constriction was recorded for 5 min. Figure 2 shows the median constriction with the 25th and 75th quartile variability about the median in explanted airways from all three strains over time. Lewis airways (top panels, open circles) were insensitive to 0.1 µM 5HT but narrowed significantly in response to 0.3, 1, and 10 µM 5HT (p < 0.001). The peak constriction to 0.3 µM 5HT was 85.2% baseline area with < 15% interquartile variability. Peak constriction was 60.8% baseline area with < 20% interquartile variability in response to 1 µM 5HT and 64.1% with < 12% interquartile variability to 10 µM 5HT, indicating that narrowing reached a plateau with 1 µM 5HT but variability decreased with 10 µM 5HT. Airways from ACI rats (middle panels, open triangles) constricted significantly to all concentrations of 5HT tested (p < 0.001). 5HT 0.1 µM narrowed the airways to 93.6% baseline area with < 20% interquartile variability, whereas 0.3 µM 5HT reduced airways to 89.9% baseline area with < 15% interquartile variability. Increasing 5HT concentrations to 1 and 10 µM caused progressively greater median peak constriction and interquartile variability of ACI airways (49.8% baseline area with < 23% variability and 23.9% baseline area with up to 45% variability, respectively). As with ACI, Fisher airways (bottom panels, closed circles) responded significantly to all concentrations of 5HT tested (p < 0.005). However, as with Lewis airways, median peak constriction in Fisher airways reached a plateau with 1 µM 5HT. Peak constriction to 0.1 µM and to 0.3 µM 5HT were 92.5% baseline area with < 25% interquartile variability and 91.5% baseline area with 20% interquartile variability, respectively. Peak constriction to 1 µM 5HT was 46.9% of baseline area with 15 to 25% variability; increasing to 10 µM 5HT reduced the variability to  15% but peak constriction remained at 42.0% baseline area.

View larger version (25K): [in this window] [in a new window]   Figure 2.   Change in luminal area of intraparenchymal airways in response to 5HT. Constriction of explanted airways to increasing [5HT] was visualized with video microscopy. Changes in the airway luminal area were normalized to baseline and plotted over time. Top panels: Lewis airways. Middle panels: ACI airways. Bottom panels: Fisher airways. The symbols indicate the median constriction of all the airways tested at that specific [5HT] (n = 16 to 40); the dotted lines represent the 25th and 75th quartiles of constrictions. Time is shown on a logarithmic scale.

There were no differences in peak constriction between Fisher and ACI explanted airways to any of the 5HT concentrations tested (p > 0.05). However, Fisher and ACI peak constrictions were both significantly greater than in Lewis airways in response to 0.1 and 10 µM 5HT (p < 0.05). Fisher airways had consistently greater responses than Lewis to 1 µM 5HT, but the difference did not quite reach statistical significance (p = 0.07). There were no differences in airway constriction to 1 µM 5HT between Lewis and ACI, nor to 0.3 µM 5HT among any of the rat strains (p > 0.05).

The narrowing velocity from baseline until 50% peak constriction was unaffected by increasing 5HT from 1 to 10 µM in all rat strains (Table 1). Constriction rates were similar between Lewis and ACI airways and between ACI and Fisher airways to 1 and 10 µM 5HT stimulation. The rates were significantly faster in Fisher than Lewis airways in response to both 1 and 10 µM 5HT (p < 0.05).

The latency to peak constriction with 0.3 µM 5HT stimulation differed significantly between Lewis and ACI airways (p < 0.05) but was similar between ACI and Fisher airways as well as between Lewis and Fisher airways (Table 2). There was no difference between the three strains of rat with 1 µM 5HT stimulation either. However, all three strains displayed significantly different latencies from each other in response to 10 µM 5HT (p < 0.05).

Relationship Between In Vitro Airway Narrowing and In Vivo Airways Responsiveness

The correlations between intraparenchymal airways narrowing capacity and velocity ex vivo with airways responsiveness in vivo were explored among the three rat strains in order to assess the relevance of these parameters to overall responsiveness. A strong correlation was found between narrowing velocity and the EC₂₀₀RL with 1 and 10 µM 5HT stimulation (Figure 3, upper left and right panels, r ² = 0.85 and 0.92, respectively). A positive correlation was also observed between narrowing capacity and the EC₂₀₀RL with 1 µM 5HT stimulation (Figure 3, lower left panel, r ² = 0.93), but not with 10 µM 5HT (Figure 3, lower right panel, r ² = 0.35).

View larger version (14K): [in this window] [in a new window]   View larger version (11K): [in this window] [in a new window]   Figure 3.   Relationship between airways responsiveness in vivo and in vitro. Upper panels: Correlation between EC200RL and airway narrowing velocity to 1 and 10 µM 5HT. Lower panels: Correlation between EC200RL and airway narrowing capacity to 1 and 10 µM 5HT measured as the minimal median area. Open circles: Lewis; open triangles: ACI; closed circles: Fisher.

5HT Mobilization of Ca²⁺ in Airway Smooth Muscle

5HT-mediated Ca²⁺ mobilization in cultured airway smooth muscle cells was visualized with digitized imaging microscopy (Figure 4). Cumulative frequency distributions comparing the increase in [Ca²⁺]_i above baseline levels stimulated by 0.1 to 10 µM 5HT in Lewis, ACI, and Fisher rat airway smooth muscle revealed both substantial intrastrain heterogeneity of responses among cells and interstrain heterogeneity of cell populations. The increase in [Ca²⁺]_i was concentration-dependent in the population of Lewis airway smooth muscle cells, as indicated by the rightward shift of the distribution curve between 0.1 and 10 µM 5HT. In contrast, only a portion of the Fisher and ACI airway smooth muscle cell populations shifted rightward to increasing concentrations of 5HT; the rest had reached peak increases in [Ca²⁺]_i with 0.1 µM 5HT. Overall, [Ca²⁺]_i increases were significantly higher in Fisher and ACI airway smooth muscle cells than in Lewis cells at all 0.1 and 1.0 µM 5HT (p < 0.01, p < 0.05, respectively). [Ca²⁺]_i increases were significantly higher in ACI than Fisher airway smooth muscle cells with 0.1 µM 5HT stimulation (p < 0.05) but similar with 1 and 10 µM 5HT. The mean increases in [Ca²⁺]_i in Lewis, ACI, and Fisher cells were 32.4 ± 2.0, 108.3 ± 6.0, and 93.6 ± 7.7 nM, respectively, to 0.1 µM 5HT, 100.0 ± 5.5, 134.6 ± 5.0, and 118.9 ± 5.3 nM, respectively, to 1.0 µM 5HT, and 100.1 ± 5.2, 106.1 ± 4.9, and 113.0 ± 5.3 nM, respectively, to 10 µM 5HT.

View larger version (17K): [in this window] [in a new window]   Figure 4.   5HT stimulated [Ca2+]i increase (peak  baseline) in cultured airway smooth muscle cells. Cell monolayers were loaded with fura-2AM and visualized with fluorescence microscopy. Images of 5HT- mediated changes in [Ca2+]i were converted to spectrofluorimetrical tracings for quantification of [Ca2+]i. Results are plotted as cumulative frequency distributions; symbols represent the fraction of the total cell population with  [Ca2+]i increase indicated. Left panel: 0.1 µM 5HT. Middle panel: 1.0 µM 5HT. Right panel: 10 µM 5HT. Open circles: Lewis (n = 139 to 196 cells); open triangles: ACI (n = 192 to 197 cells); closed circles: Fisher (n = 88 to 156 cells). Cell populations were composed of cells originating from 2 to 5 animals. The inset shows baseline [Ca2+]i. Lewis, open bar; ACI, hatched bar; Fisher, solid bar. *p < 0.05.

The inset (Figure 4) shows mean baseline [Ca²⁺]_i in resting airway smooth muscle cells. Mean baseline [Ca²⁺]_i was significantly lower in Fisher than in ACI or Lewis cells (p < 0.001). ACI baseline [Ca²⁺]_i was also lower than Lewis (p < 0.001).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The objectives of this study were to examine the possibility that enhanced airway smooth muscle contractility may determine airway hyperresponsiveness in vivo and to establish the relationship between airway smooth muscle Ca²⁺ mobilization and contractility. These hypotheses were addressed by relating in vivo and ex vivo airway responsiveness with airway smooth muscle Ca²⁺ mobilization in three highly inbred strains of rat that had demonstrated different degrees of airway responsiveness to methacholine inhalation challenge (20). We found that the relative degree of Lewis and Fisher maximal contractile responses of the airways ex vivo correlated with their relative degree of pulmonary responsiveness in vivo but that ACI airways were more responsive ex vivo than expected from their in vivo responses. Agonist sensitivity in vivo correlated with intraparenchymal airways narrowing velocity ex vivo in all three strains of rat. However, the relationship between agonist sensitivity and intraparenchymal airways narrowing capacity ex vivo was less clear because although a positive correlation was observed with 1 µM 5HT stimulation, these parameters were not correlated with 10 µM. The degree of 5HT-mediated release of intracellular Ca²⁺ in airway smooth muscle cells from each strain of rat mirrored the degree of their airways responsiveness ex vivo, suggesting a link between the extent of airway smooth muscle Ca²⁺ mobilization and the degree of airways narrowing.

Fisher rats have been previously shown to be hyperresponsive to methacholine compared with Lewis and ACI in vivo (20). Because the hyperresponsiveness underlying human asthma is not agonist-specific, we wished to verify that the same nonspecificity was also true of the rat. Serotonin was the chosen agonist because our objective was to relate three levels of responsiveness from whole animals to excised tissue to single cells and 5HT is effective under all three experimental conditions. In the current study, Fisher rats were also found to be more responsive in vivo to 5HT than either Lewis or ACI, confirming the nonspecificity of their hyperresponsiveness.

We next wished to determine whether ex vivo airway responsiveness reflected in vivo responsiveness in these rat strains. Earlier findings had demonstrated that tracheal rings from Fisher rats contracted with greater isometric force in response to carbachol than rings from Lewis or ACI rats (21), consistent with the rank order of their in vivo responsiveness. However, airway smooth muscle shortening capacity and velocity are believed to be more relevant markers of bronchial reactivity (15), so we wanted to explore these parameters in our strains of rat. We were also interested in examining the responsiveness of intraparenchymal airways because they are the site of airflow limitation during bronchoconstriction (31). Therefore, the maximal airway narrowing capacity and velocity of intraparenchymal airways from Lewis, ACI, and Fisher rats were compared by measuring the dynamic narrowing of airways in lung explants, a preparation that maintains airway-parenchymal coupling and thus the mechanical tension that normally constrains airway smooth muscle shortening in vivo (22, 32). Despite the high degree of intrastrain variability in airway narrowing capacity to 5HT among the sampled intraparenchymal airways, peak narrowing capacities as well as velocities to 50% peak constriction were greater in Fisher than in Lewis airways in response to different concentrations of 5HT. These results are consistent with the observed differences in in vivo responsiveness to 5HT and are similar to responses to methacholine in explanted Fisher and Lewis airways (22).

Airways from ACI rats, conversely, were as responsive as Fisher airways in terms of maximal narrowing capacity and narrowing velocity ex vivo, in contrast to their lower in vivo responses. However, the time to reach the same degree of peak constriction with 10 µM 5HT stimulation took twice as long in ACI than in Fisher airways. This discrepancy appears to reflect interstrain differences during the latter half of the constriction because their narrowing velocities during the first half were similar. Because explanted airways constrict against an auxotonic load, these results suggest that the progressive resistance to narrowing may be lower in Fisher than in ACI lungs. This is consistent with earlier suggestions that the mechanical coupling of airways to parenchyma in Fisher lungs may be reduced compared with Lewis lungs (32) and/or that Fisher airway smooth muscle can generate higher tension (29). Consequently, although ACI airways have the potential to be as responsive as Fisher ex vivo, the longer latency to peak constriction may allow enough time for the activation of other mechanisms to limit bronchoconstriction in vivo. Tidal breathing oscillations might be one such mechanism (33). The slower narrowing rate of ACI airways might be expected to prevent them from reaching the same degree of constriction as Fisher airways before inspiration increased the impedance so as to prevent further constriction. In addition, neural and other modulatory influences may be absent from the explant preparation, which could also create disparities between in vivo and ex vivo responses in the ACI strain.

To confirm that the ex vivo narrowing capacities and velocities were indeed related to in vivo airway responsiveness in our strains of rat, we correlated the agonist sensitivity of lung resistance with peak constriction and early narrowing velocity at 1 and 10 µM 5HT. Agonist sensitivity strongly correlated with narrowing velocity of all three strains of rat at both concentrations of 5HT, indicating that ex vivo narrowing velocity, and by extension smooth muscle shortening velocity, are reliable predictors of in vivo responsiveness. Maximal narrowing capacity, in contrast, did not appear to be as reliable because a correlation could not be drawn between it and agonist sensitivity at 10 µM 5HT, even though the relationship was present at 1 µM 5HT.

To assess whether airway smooth muscle contractility might be related to intracellular Ca²⁺ responses, the mobilization of intracellular Ca²⁺ was compared in airway smooth muscle cells cultured from Lewis, ACI, and Fisher rats. We found that the relative degree of [Ca²⁺]_i increases to 5HT paralleled the relative degrees of explanted airways constriction among all three rat strains insofar as Fisher and ACI displayed higher cellular Ca²⁺ responses as well as greater airways narrowing and narrowing velocity than Lewis. The consistency of these findings suggests that airway smooth muscle contractility is associated with the extent of intracellular Ca²⁺ mobilization. It is worth noting that enhanced Ca²⁺ uptake into airway smooth muscle may be involved in allergen-induced airway hyperresponsiveness as well (17), so alterations in airway smooth muscle Ca²⁺ signaling may be a common feature of airway hyperresponsiveness arising from different origins.

Interestingly, populations of Lewis cells appeared to contain more cells with low agonist sensitivity compared with Fisher or ACI populations. This is evident from the positions of the frequency distributions of Ca²⁺ responses and from the 5HT concentration-dependent shifts in these distributions. The entire Lewis distribution shifted rightward as [5HT] increased whereas only a fraction of Fisher and ACI distributions shifted rightward; the majority of Lewis cells thus were only submaximally activated whereas the majority of Fisher and ACI cells already reached their maximal responses at a low concentration of 5HT. If these cells correctly represent a random sample of whole tissue, then the proportional distribution of heterogeneously responsive cells in a population may determine the responsiveness of the whole tissue.

It is important to note, however, that altered intracellular Ca²⁺ signaling is likely only one of a number of contributors to enhanced airway smooth muscle contractility, particularly in light of the Ca²⁺ sensitization of the contractile apparatus in which the force:Ca²⁺ ratio in smooth muscle is higher in response to contractile agonists than to K⁺ depolarization (34). This occurs because contractile agonists activate two additional pathways which inhibit myosin light chain phosphatase (SMPP-1M) activity, resulting in increased myosin light chain phosphorylation (34, 35). One route of SMPP-1M inhibition is protein kinase C (PKC)-dependent (36) while the other is Rho-dependent (37). In addition, recent evidence indicates that Rho-associated kinase (Rho-kinase) can directly phosphorylate ser19 on myosin light chain (40, 41), further amplifying the capacity of Rho to activate smooth muscle contraction independently of Ca²⁺. Given that selective inhibition of Rho-kinase reduces high blood pressure in spontaneously and experimentally induced hypertensive rats (42), possible variations in the Ca²⁺ sensitization pathways might contribute to the enhanced contractility of Fisher airway smooth muscle as well. Nevertheless, the relationship between Ca²⁺ sensitization of force and shortening velocity is still unclear whereas initial isotonic shortening velocity appears to follow the extent of the initial Ca²⁺ peak in canine tracheal smooth muscle strips stimulated with full and partial muscarinic agonists (43). In the same study, isotonic shortening velocity did become sensitized to Ca²⁺ after several minutes of agonist activation, although this time course may be irrelevant to airway caliber modified by tidal breathing oscillations (44).

In conclusion, we have described co-incidence of in vivo and ex vivo airways responsiveness with airway smooth muscle intracellular Ca²⁺ mobilization in Fisher and Lewis rats. We have also reported co-incidence of ex vivo airways responsiveness with airway smooth muscle intracellular Ca²⁺ mobilization but not with in vivo responsiveness in ACI rats. Our results suggest that enhanced Ca²⁺ signaling in airway smooth muscle is associated with enhanced contractility ex vivo and that this in turn may result in airway hyperresponsiveness.