Dynamic airway constriction in rats: heterogeneity and response to deep inspiration

Abstract

Airway narrowing due to hyperresponsiveness severely limits gas exchange in patients with asthma. Imaging studies in humans and animals have shown that bronchoconstriction causes patchy patterns of ventilation defects throughout the lungs, and several computational models have predicted that these regions are due to constriction of smaller airways. However, these imaging approaches are often limited in their ability to capture dynamic changes in small airways, and the patterns of constriction are heterogeneous. To directly investigate regional variations in airway narrowing and the response to deep inspirations (DIs), we utilized tantalum dust and microfocal X-ray imaging of rat lungs to obtain dynamic images of airways in an intact animal model. Airway resistance was simultaneously measured using the flexiVent system. Custom-developed software was used to track changes in airway diameters up to generation 19 (~0.3–3 mm). Changes in diameter during bronchoconstriction were then measured in response to methacholine (MCh) challenge. In contrast with the model predictions, we observed significantly greater percent constriction in larger airways in response to MCh challenge. Although there was a dose-dependent increase in total respiratory resistance with MCh, the percent change in airway diameters was similar for increasing doses. A single DI following MCh caused a significant reduction in resistance but did not cause a significant increase in airway diameters. Multiple DIs did, however, cause significant increases in airway diameters. These measurements allowed us to directly quantify dynamic changes in airways during bronchoconstriction and demonstrated greater constriction in larger airways.

INTRODUCTION

Airway narrowing in response to allergens or pharmacological agents causes increased airway resistance, which reduces regional airflow and, thereby, results in patchy ventilation. Airway hyperresponsiveness is an exaggerated response to stimuli that is one of the hallmark characteristics of asthma, but the precise nature of changes in the mechanisms that regulate bronchial contractility is unclear (1, 46). The degree of bronchoconstriction is determined by complicated interactions involving smooth muscle tone, localized inflammation, and mechanical forces, such as oscillatory mechanical stretch and deep inspirations (DIs). For example, DIs promote airway dilation in healthy individuals, but this response is altered in patients with asthma.

Because the factors that regulate bronchial caliber vary spatially throughout the airway tree, there is substantial regional heterogeneity in the response to bronchoconstricting agents (12, 32, 38). In patients with asthma, this regional heterogeneity causes severe ventilation defects that contribute to ventilation-perfusion mismatching and reduced gas exchange. In patients and in animal studies, several different imaging approaches, including computed tomography (CT), functional MRI, and PET scans, have been used to demonstrate these ventilation defects (8, 13, 27, 53, 54, 59, 60, 62). However, while these approaches are useful for identifying regional ventilation defects, they are typically limited in their ability to distinguish bronchoconstriction in individual airways across a wide range of sizes, especially smaller airways. In addition, some of these imaging techniques, such as hyperpolarized helium MRI, require a static breath hold for imaging, which limits imaging of dynamic changes in airways. To overcome these limitations, we used tantalum dust as a contrast agent, along with microfocal X-ray imaging, to visualize changes in airway diameters throughout the rat airway tree during bronchoconstriction while simultaneously measuring overall lung resistance.

The goal of this study was to measure airway narrowing in an intact rat in response to a methacholine (MCh) challenge in both large and small airways while simultaneously measuring lung mechanics. While MCh induced significant increases in lung resistance, only the larger airways experienced significant decreases in diameter. In addition, a single DI caused a significant decrease in airway resistance but did not have a significant effect on airway diameters.

MATERIALS AND METHODS

Animal preparation.

Brown Norway rats were anesthetized (14 mg/kg xylazine ip + 35 mg/kg pentobarbital ip) and paralyzed (0.8 mg/kg pancuronium), and the trachea was cannulated with polyethylene tubing. The rats, in the right lateral decubitus position, were placed on the object stage of a microfocal X-ray imaging system and mechanically ventilated with a tidal volume of 6 ml/kg, positive end-expiratory pressure (PEEP) of 5 cmH₂O, and a respiratory rate that kept arterial Pco₂ at 35–45 mmHg. Airway pressures were continuously measured and recorded using a flexiVent system (SCIREQ, Montreal, QC, Canada). All experiments were performed in accordance with animal protocols approved by the University of Tennessee Health Science Center and the Zablocki Veterans Affairs Medical Center Institutional Animal Care and Use Committees.

Microfocal X-ray imaging.

The microfocal X-ray imaging system was designed and constructed at the Zablocki Veterans Affairs Medical Center/Keck Functional Imaging Center (Milwaukee, WI). Details of the imaging system are available elsewhere (40, 51). The apparatus consists of a microfocal X-ray source in conjunction with an image-intensifier detector coupled to a digital charge-coupled device (CCD) camera. The specific components include an X-ray source (3-µm focal spot; FeinFocus 100.50, Comet North America, Stamford, CT), an image intensifier (model AI-5830-HP, North American Imaging, Camarillo, CA) coupled to a CCD camera (Silicon Mountain Design SMD1M-15, DALSA, Waterloo, ON, Canada), and a specimen micromanipulator stage (1-µm translational accuracy; New England Affiliated Technologies/Danaher Precision Systems, Salem, NH), all mounted on a precision rail (10-µm relative positional accuracy). Planar projection images were averaged from seven frames to minimize noise and maximize contrast. The image data from the CCD camera were transmitted to a frame grabber board. Image acquisition, positional information recording, and stage control were performed using custom-developed software.

Airway imaging.

Tantalum dust, passed through a 5-µm filter, was introduced into the ventilator circuit at the onset of inflation via a timed valve. Labeled airways were imaged during mechanical ventilation. Approximately 300 images at low and high magnification were obtained at 27 frames/s during each ventilation maneuver. Bronchoconstriction was induced using increasing doses (1–50 mg/ml) of MCh, introduced via a nebulizer attached the ventilator circuit. A control nebulization was performed using phosphate-buffered saline before the first MCh challenge. After each bronchoconstriction maneuver, images were also recorded for DI via a total lung capacity (TLC) maneuver followed by a pressure-volume (P-V) loop maneuver. A TLC maneuver involved a ramped increase in airway pressure to 30 cmH₂O that was held for 3 s. A P-V maneuver involved a stepwise increase in pressure from 0 to 35 cmH₂O followed by a return to 0 cmH₂O over a period of 16 s. Additional images for spatial calibration were obtained with a 2,000- and a 500-µm movement of the rat for each magnification, respectively. Measurements made from images were used to calculate percent change in airway diameter for different generations of airways, as previously described (51). The sensitivity of the measurements is limited by the resolution of the imaging system and the ability to identify the airway boundaries. Using the high-magnification images calibrated at ~25 μm/pixel, we could identify changes in airways as small as 150–200 μm in diameter.

Airway generation and changes in diameter.

Images of airways at end exhalation before and after MCh nebulization, as well as a TLC maneuver, were used to measure airway diameters. Airway diameter was measured perpendicular to the airway wall at landmark locations along the airway tree in the selected images. These landmarks included bifurcations in the airway tree and tantalum patterns that were easily distinguishable. In addition, we selected locations that remained within the frame throughout the imaging protocol. Percent change in airway diameter was calculated relative to diameter before the maneuver. The airway diameters were grouped by branching generation, with the trachea designated generation 1. Using low-magnification images, we could identify airways up to generation 13, while high-magnification images showed branches up to generation 19.

Lung function measurements.

Lung mechanics were measured using the flexiVent system while images were recorded simultaneously. Total respiratory system resistance and compliance were measured using a single-frequency, forced-oscillation maneuver and a single-compartment model. Newtonian (central airway) resistance, tissue damping (energy dissipation in the peripheral tissue), and tissue elastance (energy conservation in the peripheral tissue) were measured using a low-frequency, forced-oscillation maneuver and applying the constant phase model. After MCh nebulization, the resistance values would peak and then decline. TLC and P-V maneuvers were initiated after the decline, and subsequent measurements were made. These measurements were delayed at least as long as the length of time required for the single-compartment measurement (~1 s) or the forced oscillation (~3 s).

Data analysis.

Values are means ± SE, unless indicated otherwise. Comparisons between two groups were assessed by a t-test, while comparisons of multiple groups were assessed by one- or two-way ANOVA followed by Tukey’s post hoc testing. A threshold of P < 0.05 was used to determine significant differences. Statistical comparisons were made using Sigmastat 3.5 (Jandel Scientific, San Rafael, CA) or Prism 7.05 (GraphPad, La Jolla, CA).

RESULTS

Imaging MCh-induced bronchoconstriction.

To visualize bronchoconstriction throughout the rat lungs, we first dusted the airways with tantalum to outline the airway walls. We standardized the lung volume history for each rat by performing two sequential TLC maneuvers. We then imaged the airways during nebulization of saline and increasing doses of MCh while simultaneously measuring lung mechanics using the flexiVent system. At 10 mg/ml MCh, an additional high-magnification set of images was recorded during a second MCh challenge. The high-magnification images were taken toward the periphery of the lung to visualize higher-generation/smaller airways. Figure 1 shows representative baseline images at end exhalation for low and high magnification (Fig. 1, A and E) compared with images in the same lung after nebulization of 10 mg/ml MCh (Fig. 1, B and F). Bronchoconstriction is shown throughout the airway tree. A video of the time course of bronchoconstriction (in a different rat) in response to 10 mg/ml MCh at low magnification is included in the online supplement (Supplemental Material Video S1, available at https://doi.org/10.6084/m9.figshare.7643273.v2). Figure 2 shows the lung resistance and the Newtonian resistance during a sequence of nebulizations in a representative single animal. After each nebulization, a series of alternating TLC and P-V maneuvers was performed and imaged until the lung resistance returned to baseline levels (typically 3–7 maneuvers). Figure 1, C and G, shows the airways at TLC during the first maneuver, while Fig. 1, D and H, shows the airways following the first TLC maneuver. A second video showing the changes in airways (in a different rat) during the first TLC maneuver following MCh is also included in the online supplement (Supplemental Material Video S2, available at https://figshare.com/s/e87e7463166186b3780a).

Fig. 1.

Tantalum dust was used to label airways for microfocal X-ray imaging to visualize airway diameter during bronchoconstriction and deep inspiration. A–D: images of airways from the same rat captured at end exhalation at baseline (A), after nebulization with 10 mg/ml methacholine (MCh; B), at total lung capacity (TLC) during a deep inspiration (C), and after the first TLC maneuver (D). E–H: higher-magnification images of the region in the inset in A obtained during a subsequent nebulization (10 mg/ml) and imaging sequence.

Fig. 2.

Measurements of total respiratory system resistance (R, A) and Newtonian resistance (R_N, B) from a representative rat following tantalum dusting and subsequent inhalation of nebulized saline and increasing doses (1–50 mg/ml) of methacholine. ▲, Total lung capacity maneuvers.

Changes in lung mechanics and airway diameters.

To evaluate the effects of each dose on lung mechanics, we averaged each parameter measured during the peak response (between nebulization and the first TLC maneuver, ~1–2 min) and summarized the results for all animals in Fig. 3. Compared with the saline control, there was a significant increase in respiratory system resistance and Newtonian resistance at 10 and 50 mg/ml, but not 1 mg/ml, MCh (Fig. 3, A and B). Tissue damping was also significantly increased at the two higher MCh doses, while tissue elastance was significantly increased at all three doses (Fig. 3, C and D).

Fig. 3.

Lung physiology measurements following exposure to saline and increasing doses (1, 10, and 50 mg/ml) of methacholine (MCh). Total respiratory system resistance, R (A), central airway resistance, R_N (B), damping, G (C), and tissue elastance, H (D), were measured in 6 different rats. R was measured during a single-frequency inspiration and calculated using a single-compartment model of the lungs. R_N, G, and H were measured using a low-frequency, forced-oscillation technique and calculated using the constant-phase model of the lung. Mean value for each rat was calculated from 1–3 measurements during the peak response; error bars indicate SE; n = 6 rats. *P ≤ 0.05 vs. saline, #P ≤ 0.05 vs. 1 mg/ml MCh.

To determine the distribution of responses in airways of different sizes, we measured the change in diameter following saline or MCh nebulization compared with the baseline diameter before nebulization. Figure 4 shows changes in diameter in response to saline and 10 mg/ml MCh for all airways visualized in five rats. As shown in Fig. 4A, saline treatment caused relatively small changes (1.4 ± 0.8%), both positive and negative, in diameter across all airways, with larger variations in smaller (<700-μm) airways. MCh (10 mg/ml) caused significantly larger changes in diameter (−15.6 ± 1.1%), with 79.3% of the airways constricting (Fig. 4B). The largest (>1,800-μm) airways were consistently constricted, but there was more variability in the response in smaller airways.

Fig. 4.

Changes in airway diameter as a function of baseline diameter in response to nebulized saline (A) or methacholine (MCh, 10 mg/ml) (B) in 5 different rats. Percent change in diameter for each airway was measured by comparing diameter at end exhalation after MCh with diameter at end exhalation under baseline conditions. A negative percent change indicates airway constriction; a positive change indicates airway expansion. Each color represents measurements from a specific rat. Measurements of the response to saline were taken from only low-magnification images; measurements of the response to MCh were taken from a combination of low- and high-magnification images captured in subsequent MCh nebulizations.

To further investigate the changes in diameter and the dose dependence, diameter measurements from low-magnification images were binned according to airway generation (generations 2–13). Figure 5 shows the mean change in diameter as a function of generation for each dose of MCh compared with saline. The overall mean change in diameter (dashed line) was significantly greater for all doses of MCh compared with saline; this is in contrast with the changes in lung parameters in Fig. 3 that were not affected by the lowest dose. There were no differences in the mean change in diameter between the different doses of MCh. As a further comparison, we evaluated the differences between generations within a given dose. The tables within Fig. 5 indicate significant differences between changes in lower and upper generations at each dose (there were no differences in the saline group). Consistent with Fig. 4, these results demonstrate significantly greater airway constriction in the larger airways than the smaller airways that were visualized.

Fig. 5.

Dose-dependent effects of methacholine (MCh) on bronchoconstriction as measured by changes in airway diameter compared with saline. Airways were measured in generations 2–13 of branching using the lower-magnification tantalum bronchograms. MCh caused significant constriction in airway generations 2–8 at all doses (1, 10, and 50 mg/ml) compared with saline nebulization. Values are means ± SE; n = 3–5 rats per generation. Dashed lines indicate mean value from all measurements for each specific treatment. *P < 0.05 vs. saline, +P < 0.05 between different generations at indicated dose of MCh. Only low-magnification images were used for these comparisons.

Single vs. multiple DIs.

As described in materials and methods, we performed alternating TLC and P-V maneuvers following each peak response to MCh nebulization, with the initial maneuver a TLC maneuver. (These multiple-DI maneuvers were not performed following saline nebulization.) We then assessed the effect of these DIs on changes in lung mechanics and diameters. As shown in Fig. 6, after saline nebulization there were modest changes in total respiratory system resistance (−10% ± 3%), central airway (Newtonian) resistance (+2% ± 14%), tissue damping (−4% ± 3%), and tissue elastance (−1% ± 2%) following a single DI, but a single DI following 10 mg/ml MCh caused significant changes in total respiratory system resistance (−30% ± 4%), central airway resistance (−30% ± 5%), and tissue damping (−24% ± 3%). Subsequent multiple DIs (typically 3–7 alternating TLC and P-V maneuvers) caused further significant changes in total respiratory system resistance, central airway resistance, and tissue elastance compared with a single DI.

Fig. 6.

Deep inspirations (DIs) caused reductions in lung physiology parameters following methacholine (MCh)-induced bronchoconstriction. After treatment with MCh (10 mg/ml) or saline, a single DI was induced using a total lung capacity (TLC) maneuver, and measurements were made of total respiratory system resistance, R (A), central airway (Newtonian) resistance, R_N (B), tissue damping, G (C), and tissue elastance, H (D). Plots show change in each parameter from peak response to post-DI values. After these measurements, multiple DIs (typically 1–2, in addition to the first TLC maneuver) consisting of alternate TLC and pressure-volume maneuvers were induced, and measurements were repeated. A single DI following MCh caused a significant decrease in R, R_N, and G compared with the effect of a DI following saline. Multiple DIs following MCh caused a significant decrease in R, R_N, and H compared with a single DI. Values are means ± SE; n = 3–6 rats. *P ≤ 0.05 vs. a single DI following saline, #P ≤ 0.05 vs. a single DI following MCh.

To evaluate the distribution of the response to DIs in individual airways, we compared changes in the response to single and multiple DIs in each generation following saline or 10 mg/ml MCh. Surprisingly, there were no significant differences in diameter change in response to a single DI following saline compared with MCh at any generation (Fig. 7), despite the significant changes in lung parameters shown in Fig. 6. Subsequent multiple DIs did, however, cause significant increases in diameters in generations 4–8, 10, and 11 following MCh (Fig. 7).

Fig. 7.

Multiple deep inspirations (DIs) stimulated bronchodilation following methacholine (MCh) treatment. In rats treated with saline or MCh (10 mg/ml), no significant differences were observed in any airway generations following a single DI. Multiple DIs (typically 3–7) caused significant increases in diameter in generations 4–8, 10, and 11 following MCh-induced airway constriction. Values are means ± SE; n = 3–5 rats per generation. Measurements were obtained for all 6 rats, but generations 2 and 13 could not be visualized in all rats. Only low-magnification images were used for these comparisons. *P < 0.05 vs. respective airway generation following a single DI.

DISCUSSION

In this study we simultaneously measured changes in lung mechanics and imaged changes in airway diameters in response to MCh throughout the airway tree in intact anesthetized rats. The main findings of this study are as follows. 1) MCh caused dose-dependent changes in lung mechanical parameters, but changes in airway diameters were similar at all doses. 2) Larger airways constricted more than smaller airways in response to MCh. 3) A single DI caused significant changes in resistance and lung mechanics but not in diameters in any generation. 4) Multiple DIs caused further changes in lung resistance and mechanics, as well as significant changes in diameters.

There are several limitations to our approach, which we discussed in detail in our previous study (51), but we will mention some important considerations here. During dynamic imaging, airway segments may distend out of the plane of the image. However, this does not limit the measurements of diameters, since the tantalum dust outlines the circumference of the airway, and we can continuously visualize its projection, because the depth of the focal plane in our system is greater than the width of the lungs. We previously imaged a tantalum-dusted tube comparable in diameter to large (~1,500 µm) airways and positioned it at different locations between the X-ray source and the detector spanning the approximate width of a lung. Measured diameters varied by <3–4% of the diameter of the tube. We visualized airways perpendicular to the imaging plane and portions of airway segments that move out of the field of view in real time during image acquisition, and we avoided measurements of airways that distended out of the field. It has long been recognized that significant folding of airway walls can occur during bronchoconstriction, and this is observed as narrowing of the lumen (17, 57, 61). Such folding of the wall would be difficult to visualize using our planar system, unless the airways were captured in cross section.

Our approach allowed us to visualize substantial heterogeneity in the response to MCh in airways down to generation 19 (in high-magnification images). While the largest (central) airways underwent consistently high constriction (~40–70% in Fig. 4) in response to MCh, the smaller airways exhibited much greater heterogeneity, with responses ranging from comparatively small (<20%) dilation to very large constriction (Figs. 1, 4, and 5). Some of this variability within generations may be due to the significant asymmetric branching structure of the rat lung (34, 48), where airways of similar diameter branch from the larger airways but are included in many different generations. In addition, we observed variations in constriction within the same bronchial segment, as regions near branch points constricted less than regions in the center of the segment; this was more apparent in Supplemental Video S1. The increases in diameter that we observed in some of the smaller airways in response to MCh could be due to interdependent interactions with nearby larger airways or, potentially, to air trapping within that region.

Despite the importance of determining airway distension and its role in regulating airway responsiveness (1, 10, 17, 21, 23, 39, 44, 50, 56–58, 61), there is only limited information regarding changes in airway diameters in vivo and how airway distension may vary in patients with asthma or in animal models (3, 6, 7, 9, 18, 28, 43). While important insights have been gained, these studies were limited to isolated tissue or to images of mainly large airways acquired under static conditions. Previous studies using computational modeling and/or imaging of ventilation defects have suggested that constriction of small airways is likely important in airway dysfunction in patients with asthma (8, 19, 27, 49, 54, 62, 63), but few studies have directly examined small airway dynamics in vivo. Furthermore, while there has been extensive investigation of the response of constricted airways to DIs, there is little direct information about the response of airways of varying size.

Airway narrowing that occurs during an asthma attack or in response to bronchoconstricting agents results in heterogeneous ventilation patterns that lead to impaired gas exchange. Previous imaging of ventilation patterns and computational modeling have suggested that the decreased ventilation is largely due to severe constriction of smaller bronchioles (4, 5, 32, 33, 38, 60), but these imaging studies did not include direct measurements of the smaller airways. Venegas et al. (60) pointed out that the relatively large ventilation defects observed in patients with asthma would be consistent with narrowing of larger airways in a symmetrical branching model but that heterogeneity of peripheral airway narrowing in a network model could account for the larger-than-expected poorly ventilated regions. However, there is only limited direct information about the heterogeneity of airway narrowing in response to bronchoconstricting agents. Our direct measurements of diameters show that the largest airways constrict the most and that the smaller airways constrict less and exhibit greater heterogeneity (Figs. 4 and 5). Our approach could be applied to disease models of asthma to determine if constriction patterns are altered and to evaluate whether therapeutic interventions are effective in different-sized airways.

A potential limitation of our measurements is that the MCh may not have been distributed uniformly throughout the lung. If, for example, the aerosolized MCh was deposited in greater amounts in larger airways and regions with better ventilation, then our measurements would indicate greater changes in diameter in those regions. However, MCh was delivered using ventilator settings similar to those used to dust the airways with tantalum, and the measurements were made throughout the tree of airways that were labeled with tantalum. Tantalum particles (<5 μm) are similar in size to the aerosolized droplets in which MCh was delivered (~3.9–4.7 μm) (37), suggesting that their distribution would be similar. Also, the tantalum labeling remained consistent throughout the course of the experiments and was unaffected by subsequent MCh exposure. Although not every airway was labeled in every experiment, our approach allowed measurements of a significant number of airways in multiple generations and spanning a large range of diameters within each lung (45 ± 5 airways measured per rat, with an average of 1–6 measurements per rat in generations 2–13). However, it is still possible that the localized dose of MCh delivered to the smaller airways was less than that delivered to the larger airways. In comparison with structures observed from casting and CT imaging (12, 30), our tantalum images provided good representation of airways out to the terminal bronchioles. Another possibility is that MCh caused airway closure in some segments, preventing delivery to distal airways. As shown in Fig. 1, we were able to effectively label the largest airways that branch through many generations to more distal regions of the lung. Although difficult to show definitively, we did not observe complete closure of these central airways (no diameters constricted to 0). The maximum constriction was 74%, 72%, and 77% for 1, 10, and 50 mg/ml MCh, respectively. In addition, all our measurements were made with a PEEP setting of 5 cmH₂O, which would likely prevent substantial airway closure. In a previous study we were able to visualize airway closure when we applied negative pressure to the airways (51). Also, it is important to note that studies of MCh challenge in human subjects typically involve aerosolized delivery, and the distribution of MCh would be limited by the same considerations as in our study.

In previous studies, high-resolution CT imaging of canine airways demonstrated heterogeneous responses to aerosolized histamine, both in the whole lung and when delivered locally, but diameters were not measured in the smaller distal airways (4, 32, 33). These studies suggested that increases in airway resistance during bronchoconstriction may be dominated by changes that occur in the lung periphery, but the authors also raised the possibility that constriction of central airways might affect their surrounding parenchyma due to interdependence. High-resolution CT imaging was also used to demonstrate heterogeneous changes in central airways in human subjects exposed to inhaled MCh (35), rabbits exposed to inhaled histamine (2), and mice exposed to intravenous MCh (12), with some airways undergoing constriction and others dilation. However, these studies were also limited by the necessity to perform imaging during a static breath hold. Our measurements were made during dynamic ventilation and imaging at end exhalation over a wide range of airway sizes. The study by Dubsky et al. (12) included measurements of diameters covering both large central and smaller peripheral airways in mice as well (using a static breath hold), but they observed a paradoxical dilation of airways in response to an initial intravenous injection of MCh, followed by constriction with a second MCh injection. The differences in response may be due to species-specific responses or, possibly, differences in the delivery of MCh.

To further investigate how changes in diameter affected lung mechanics, we measured the MCh dose-dependent responses to both a single-frequency forced oscillation and a multiple-frequency forced oscillation. We measured a dose-dependent increase in both total respiratory system resistance and Newtonian (airway) resistance compared with the response to saline (Fig. 3), with no significant increase in the lowest dose (1 mg/ml). However, when we compared changes in diameter in different generations, we measured significantly greater constriction with all doses of MCh in the largest airways (generations 2–8) than with saline (Fig. 5). There were no significant differences between the doses at any generation, indicating that the level of airway constriction was similar at each dose. At the lowest dose (1 mg/ml), the lack of a significant change in resistance when there were significant changes in diameter suggests that other factors, such as mucus bridging or airway closure, may have contributed to the resistance. When we compared lung tissue damping, a reflection of distal tissue energy loss and, potentially, mechanical heterogeneity, we found a significant increase at 10 and 50 mg/ml, but not 1 mg/ml, MCh. Changes in tissue elastance can reflect tissue stiffness or loss of volume due to airway closure. Elastance was significantly increased at each dose of MCh, indicating a potential effect of airway closure. However, as mentioned above, PEEP may have minimized potential closure, and we did not observe any airway diameters that went to zero. Another potential limitation in comparing changes in diameter with changes in these parameters is that the models are idealized representations of the lungs and may not capture the complexity of the changes that occurred. For example, the changes in lung tissue damping and tissue elastance due to MCh may be linked indirectly to changes in larger airways not captured by our imaging.

The results in Figs. 6 and 7 were surprising. We expected that significant changes in lung resistance following a single DI would correlate with significant changes in diameter, but changes in diameter occurred only after multiple DIs. There were also significant changes in elastance and tissue damping with a single DI that were further increased by subsequent DIs. One potential explanation for these findings is that the initial DI reduced resistance by redistributing mucus plugs and/or opening atelectatic regions adjacent to the tantalum-labeled airways. Subsequent DIs caused relaxation of the smooth muscle and airway dilation. However, we have no direct evidence for this possibility. Nadel and Tierney first demonstrated that a DI caused a rapid, but transient, decrease in airway resistance in human subjects exposed to a bronchoconstrictor, but there was no effect on airway resistance under baseline conditions (42). A later study showed a similar response in control subjects but then showed an important difference in patients with asthma (14): a DI caused a transient increase in resistance in the baseline state but had no effect during MCh-induced bronchoconstriction. Subsequent studies confirmed these findings and further demonstrated that the timing of the recovery from DI also differed between patients with asthma and control subjects (11, 29, 31, 36, 45, 47, 55). Other studies showed that when healthy subjects voluntarily refrained from DIs for 15 min, their responses to MCh were similar to those of patients with asthma, and these effects were persistent (41, 52). The precise mechanisms responsible for the DI-induced bronchodilation in healthy subjects and the lack of this response in patients with asthma are not fully understood. Several potential mechanisms have been proposed for these differences that point toward changes in smooth muscle cell (SMC) function (1, 15). One proposed mechanism is based on the premise that oscillatory tidal stretch of a SMC regulates airway responsiveness because of differences in cross-bridge cycling and reorganization of the cytoskeleton and contractile filaments relative to the oscillation rate of the SMC (16, 20, 22, 24–26). Our study does not directly address these potential mechanisms, but our finding shows that changes in resistance, elastance, and tissue damping could occur without significant changes in airway diameters up to generation 13.

In summary, our studies demonstrate that while airway diameter changes were similar in response to increasing doses of MCh, there was a dose-dependent change in lung mechanics. In contrast with previous predictions based on mathematical modeling of airway responses, we found that larger airways constricted to a greater degree than smaller airways. A single DI caused significant changes in lung mechanics without significant changes in airway diameters, but multiple DIs did cause airway dilation.

GRANTS

These studies were supported by National Heart, Lung, and Blood Institute Grants HL-094366, HL-123540, and HL-131526 (C. M. Waters).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

S.E.S., P.S.M., and C.M.W. conceived and designed research; S.E.S., P.S.M., R.C.M., and C.M.W. performed experiments; T.-K.N.P., S.E.S., and C.M.W. analyzed data; T.-K.N.P., S.E.S., and C.M.W. interpreted results of experiments; T.-K.N.P. and C.M.W. prepared figures; T.-K.N.P. and C.M.W. drafted manuscript; T.-K.N.P., P.S.M., R.C.M., and C.M.W. edited and revised manuscript; and T.-K.N.P., S.E.S., P.S.M., R.C.M., and C.M.W. approved final version of manuscript.