Decorin (Dcn), a small leucine-rich proteoglycan, is present in the extracellular matrix of the airways and lung tissues, contributes to lung mechanical properties, and its deposition is altered in asthma. The effect of Dcn deficiency on airway parenchymal interdependence was examined during induced bronchoconstriction. Studies were performed in C57Bl/6 mice in which the Dcn gene was disrupted by targeted deletion (Dcn−/−) and in wild-type controls (Dcn+/+). Mice were mechanically ventilated, and respiratory system impedance was measured during in vivo ventilation at positive end-expiratory pressure (PEEP) = 2 and 10 cmH20, before and after aerosol delivery of methacholine (MCh). Length vs. tension curves in isolated tracheal rings were measured in vitro. Dcn distribution in +/+ mice airways was characterized by immunofluorescence; differences in collagen structure in Dcn+/+ and Dcn−/− mouse lungs was examined by electron microscopy. MCh caused similar increases in airway resistance (Raw) and tissue elastance (H) in Dcn+/+ and Dcn−/− mice. During MCh-induced constriction, increasing PEEP caused a decrease in Raw that was greater in Dcn−/− mice and a decrease in H in Dcn−/− mice only. Tracheal ring compliance was greater in Dcn −/− mice. Imaging studies showed that Dcn was deposited primarily in the airway adventitial layer in Dcn+/+ mice; in Dcn−/− mice, collagen had an irregular appearance, especially in the lung periphery. These results show that lack of Dcn alters the normal interaction between airways and lung parenchyma; in asthma, changes in Dcn could potentially contribute to abnormal airway physiology.
- airway responsiveness
excessive airway narrowing following a smooth muscle agonist challenge is a characteristic feature of asthma. The degree to which airways narrow is determined by the balance between the shortening capacity of airway smooth muscle (ASM) and the mechanical load opposing it. The mechanical load includes stress related to the deformation of the airway wall, and, because of the mechanical interdependence between airways and parenchyma, the recoil of the surrounding parenchymal attachments (5, 15, 18). The structural composition of the airway wall and of the lung parenchyma will therefore affect the load the ASM must overcome to shorten and the degree of narrowing achieved for a given stimulus.
The extracellular matrix (ECM) is a major component of the lung tissues and comprises the fiber and the interfiber compartment. Decorin is an important constituent of the interfiber compartment of the ECM and contributes to the mechanical properties of the airway wall and parenchyma (8). In addition, decorin plays a significant role in tissue assembly by regulating collagen fibrillogenesis (12, 13). In the decorin-deficient (Dcn−/−) mouse, collagen abnormalities have been described in the skin and tendons, resulting in altered compliance and reduced tensile strength (3, 6). In recent work from this laboratory, Fust et al. (8) showed that lung mechanics in Dcn−/− mice were altered. Specifically, Dcn−/− mice had decreased airway resistance (Raw) and increased lung compliance. Immunohistochemistry in decorin-replete mice showed that decorin was localized mainly around the airway and vessel wall (8).
The mechanical consequences of changes in decorin has particular relevance to asthma, since there is now considerable data in the literature showing that deposition of decorin is altered in the asthmatic airway wall (4, 11, 16, 24). We reasoned that, since decorin was an important determinant of airway mechanics and because it was distributed peribronchially, decorin would influence the interaction between airways and lung parenchyma. Hence its abnormal distribution in the asthmatic airway wall might contribute to the enhanced bronchial responsiveness characteristic of this disease.
Specifically, we questioned whether altered decorin would affect the mechanical behavior of the airway wall and/or the link between airways and parenchyma. The availability of a Dcn−/− mouse allowed us to directly examine this question. We hypothesized that methacholine (MCh) responsiveness and the effect of changing lung volume would be altered in these animals. To investigate this question, lung mechanics were assessed in vivo under baseline conditions and during MCh-induced constriction in wild-type (Dcn+/+) and Dcn−/− mice. Positive end-expiratory pressure (PEEP) and, thereby, lung volume was changed, both under baseline conditions and after delivery of MCh aerosol. To further characterize the mechanical behavior of the airway itself, we tested the length-tension characteristics of isolated tracheal rings from Dcn+/+ and Dcn−/− mice.
Determination of genotype.
Breeding was initiated with a male C57Bl/6 mouse heterozygous for decorin (Dcn+/−) and Dcn+/+ C57Bl/6 females. Mice were housed in a regular animal facility at McGill University (Montreal, PQ). The mutation of the decorin gene was created by targeted disruption of exon 2 of murine decorin by the insertion of the Pgk-neo cassette (3). The genotype of the offspring was determined by extracting DNA from the tail tissue using a Genomic DNA Purification Kit (Promega, Madison, WI) according to the manufacturer's instructions. Real-time PCR (Roche, Basel, Switzerland) was used to analyze DNA. Sense and antisense primers corresponding to exon 2 of murine decorin and an additional primer corresponding to the PGK promoter of the PGK-neo cassette were generated (3). The reaction mixture consisted of 1 ul DNA, 1 ul sense primer, 1 ul antisense primer or 1 ul PGK primer, 7 ul H2O, and 10 ul SBr-green mix (Qiagen, Hilden, Germany and Qiagen, Mississagua, Ontario, Canada). Reaction conditions were as follows: melting at 95°C for 900 s; PCR (45 cycles) at 94°C for 15 s for segment 1, 57°C for 20 s for segment 2, and 72°C for 25 s for segment 3; melting curve: 57°C for 45 s; and cooling at 30°C for 30 s. To verify that the products of the RT-PCR reaction were of the appropriate size, PCR products were run on 1.8% agarose gel for 1.5 h at 60 volts.
Dcn+/+ (mean age = 22.3 ± 8.7 wk) and Dcn−/− (mean age = 23.0 ± 7.9 wk) C57Bl/6 mice of either sex were studied. Mice were anesthetized with an injection of xylazine (12 mg/kg ip) followed 5 min later by an injection of pentobarbital sodium (40 mg/kg ip). After tracheostomy, an 18-gauge metal cannula was inserted in the trachea and tightly bound. The mouse was connected via the tracheal cannula to a computer-controlled small-animal ventilator (FlexiVent; Scireq, Montreal, Quebec; see Ref. 26). Mice were mechanically ventilated at 150 breaths/min with a tidal volume of 6 ml/kg at a PEEP of 2 cmH2O. The animals were paralyzed with an injection of pancuronium bromide (1.2 mg/kg ip). All animals received humane care in compliance with the Guide to the Care and Use of Experimental Animals formulated by the Canadian Council of Animal Care, and an institutional animal ethics committee approved the protocol.
Measurement of complex impedance and experimental protocol.
Regular mechanical ventilation was interrupted for measurement of complex impedance. A computer-generated volume signal comprised of 19 mutually primed sinusoids ranging from 0.25 to 19.625 Hz was applied to the airway opening. The amplitudes of the sinusoids decreased hyperbolically with frequency such that the flow had the same power at each frequency. The phase of each component was chosen so as to minimize the peak-to-peak amplitude excursions of the complex signal. The signal had a peak-to-peak volume of 0.17 ml and lasted 16 s. Piston displacement (ml) and cylinder pressure (cmH2O) were measured during the application of the signal. Three total lung capacity breaths (TLC ∼25 cmH2O transpulmonary pressure) were delivered before measurement of complex impedance to standardize volume history. Later (1 min), baseline impedance of the respiratory system (ZRS) was measured at PEEP = 2 cmH2O. Measurements at each PEEP were repeated three times, and the parameter estimates from each of the signal applications were averaged. Normal ventilation was resumed, and PEEP was then increased to 10 cmH2O. ZRS was then measured at the new PEEP level. Normal ventilation was resumed again, and PEEP was reset to 2 cmH2O. Two TLC breaths were given, and aerosols of saline and MCh in increasing concentrations (0.1, 0.33, 1.0, 3.3 and 10 mg/ml) delivered for 15 s duration (Hudson RCI, Teleflex Medical). Two TLC breaths were given before delivery of each concentration of MCh aerosol. After the aerosol was finished (1 min), ZRS was measured at 2 and then at 10 cmH2O as described above. Measurements were repeated after each concentration of aerosol was delivered. The same order of PEEP was maintained so as to not ablate the bronchoconstrictor response to MCh. Because we were interested in the difference in the volume effects between the two groups of mice, we felt this approach was appropriate.
Calculation of parameters.
ZRS was determined using the equation: (1) where P is cylinder pressure (cmH2O) and V is piston displacement volume (ml). Both are functions of angular frequency (ω). The respiratory system impedance [ZRS(ω)] was fit with the constant-phase model (10): (2) where R is airway (flow dependent) resistance, I is airway inertance, G is tissue damping, H is tissue elastance, j is the imaginary unit, and α = (2/π)tan−1(H/G).
Measurement of length vs. tension curves in isolated tracheal rings.
Tracheal rings were isolated from Dcn+/+ (mean age = 22.8 ± 12.7 wk) and Dcn−/− (mean age = 20.0 ± 12.7 wk) mice. With the use of a dissecting microscope, loose connective tissue was carefully removed, and the height and diameter of the rings were measured (height ∼1.7 mm and diameter ∼1.3 mm). The rings were mounted on hooks, one end of which was fixed, and the other was connected to a force transducer that measured both changes in tension and length (300B-LR Dual Mode Force transducer; Aurora Scientific, Aurora, ON). Rings were placed in an organ bath, filled with a modified Krebs solution at 37°C, pH ∼7.35, and bubbled with 95% O2-5% CO2. The initial tension was set at zero, and the tissue was allowed to equilibrate for 30 min, after which three stretches to 5 g tension were performed. To generate a length vs. tension curve, rings were stretched to a tension of 7 g and relaxed in a stepwise fashion at 1-min intervals until zero tension. For each curve, the Salazar Knowles equation (L = A − Be−kt, where L is length, t is tension, and k is an exponential constant. see Ref. 25) was fit to the raw data to determine the length corresponding to a given tension.
Tissue fixation and immunofluorescent localization of decorin.
In Dcn+/+ mice, the lungs were examined for immunofluorescent localization of decorin. [Previous studies in Dcn−/− mice showed no immunolocalization of decorin antibody (8).] Lungs were excised at the end of the in vivo physiological measurements, filled with histocon, submerged in optimum cutting temperature compound, and frozen in isopentane cooled in liquid nitrogen. Lung tissue was cut into 5-μm sections with a cryostat. Tissue sections were fixed with ice-cold acetone-methanol and then dried. The sections were blocked by incubating with universal blocking solution for 1 h, followed by incubation with 10% normal goat serum. Slides were then rinsed with Tris-buffered saline (TBS) and incubated overnight with primary antibody for decorin (rabbit anti-mouse, polyclonal from L. Fisher, NIH; see Ref. 7) at a dilution of 1:500 at 4°C. After being washed with TBS, sections were incubated with secondary antibody (Alexa Fluor 546 goat anti-rabbit; Molecular Probes, Burlington, ON, Canada) at a dilution of 1:1,000 for 1 h at room temperature. After being washed with TBS, sections were sealed with crystal mount and viewed with an Olympus fluorescence microscope (Model BX51; Carson Group, Markham, Ontario).
Tissue fixation and electron microscopy.
In Dcn+/+ and Dcn−/− mice, lungs were excised and fixed in 2% glutaradehyde in 0.2 M phosphate buffer followed by postfixation in 1% osmium tetroxide in 0.2 M phosphate buffer.
The tissue fragments were dehydrated in graded alcohols and infiltrated with mixtures of embedding medium and alcohol. The tissue fragments were embedded in Spurr embedding medium and polymerized over night at 80°C. One-micrometer-thick sections were cut on a Reichert Ultracut S ultramicrotome using glass knives. The sections were transferred to glass microscope slides, stained with Toluidine blue stain, and reviewed using a light microscope. One hundred-nanometer-thick sections were cut on a Reichert Ultracut S ultramicrotome using a diamond knife. Sections were collected on copper grids and stained with uranyl acetate and lead citrate. The 100-nm-thick sections were evaluated and photographed using a JEOL 100CX II electron microscope. The negatives were scanned using a computer system and Adobe Photoshop.
All in vivo data manipulations were performed with FlexiVent software (Scireq, Montreal, Quebec). Student's t-tests were used to assess differences in the baseline mechanical parameters between the two groups and to assess the effect of PEEP on baseline mechanical parameters within and between groups. To assess the effect of MCh on the mechanical parameters, ANOVA for repeated measures (concentration of MCh) blocked by mice was used. The same approach was used to test the effect of PEEP during induced constriction. Finally, we applied two-way ANOVA to look for an effect of mouse strain and concentration of MCh on the PEEP effect. For in vitro data, differences between the two curves, at a given tension, were examined using the Student's t-test with Bonferroni correction for multiple comparisons. Results were considered statistically significant at a probability level of 5%. Values are reported as means ± SE.
Measurements of in vivo mechanics under baseline conditions.
Baseline values of Raw, H, and G at PEEP 2 and 10 cmH2O are shown in Table 1. Raw tended to be lower in Dcn−/− mice at both PEEP, although these differences did not achieve statistical significance. The decrease in Raw (standardized to its baseline value for each individual animal) with the change in PEEP from 2 to 10 cmH2O was significant in both groups (P < 0.005). The decrease in Raw tended to be greater in the Dcn−/− mice (P = 0.065). Under baseline conditions, H and G were not significantly affected by PEEP.
Effect of MCh on in vivo lung mechanics.
The effect of increasing doses of MCh on Raw, H, and G as a percentage of baseline is presented in Fig. 1, A–C. MCh caused a dose-dependent increase in Raw, H, and G. The effect of MCh on the mechanical parameters was not different between groups at either PEEP.
Effect of PEEP on in vivo lung mechanics during MCh-induced bronchoconstriction.
Increasing PEEP from 2 to 10 cmH2O during MCh-induced constriction resulted in a significant decrease in Raw in both Dcn+/+ and Dcn−/− mice (P < 0.001; Fig. 2A). Increasing PEEP caused a decrease in H in the Dcn−/− group (P < 0.001) but had no effect in the Dcn+/+ mice (Fig. 2B). Two-way ANOVA showed that the decrements in Raw and H were significantly greater in the Dcn−/− mice compared with Dcn+/+ mice (P < 0.01 and 0.05, respectively).
Differences in airway mechanics in vitro.
Relaxation curves for Dcn−/− and Dcn+/+ tracheal rings between 5.0 and 0 g tension are shown in Fig. 3. Tracheal rings from Dcn−/− mice demonstrated increased changes in length for a given change in tension compared with tracheal rings from Dcn+/+ mice (P < 0.01–0.001; Fig. 3). These data are consistent with increased airway compliance in Dcn−/− mice.
Decorin immunolocalization by fluorescence microscopy.
The deposition of Dcn in a Dcn+/+ mouse airway is shown in Fig. 4. Dcn was deposited primarily in the adventitial layer of the airway wall, to a lesser extent in the reticular basement membrane and, notably, spared the smooth muscle layer. Decorin staining was also detected in the alveolar wall.
Ultrastructural analysis of Dcn+/+ and Dcn−/− lungs.
Electron micrographs (Fig. 5, A–F) showed appreciable differences in the collagen fibrils in the Dcn−/− mouse lung compared with Dcn+/+ mouse lung. In the Dcn+/+ lungs, the subpleural connective tissue was composed primarily of thin and regular collagen fibers (Fig. 5, A and B). In contrast, the Dcn−/− lungs showed prominent deposits of collagenous material. The deposited collagen was irregular and often showed interwoven contours (Fig. 5, C–F).
The main finding of this study is that, in mice deficient in the proteoglycan decorin, increases in transpulmonary pressure had a greater effect in reversing MCh-induced changes in Raw and H. Changes in physiology were accompanied by changes in collagen structure and deposition.
Mechanical load is an important factor modulating airway narrowing during smooth muscle agonist challenge (5, 15, 18). The mechanical load opposing ASM shortening includes the impedance of the airway wall to shape change and the impedance of the surrounding lung parenchyma. Under normal conditions, the lung parenchyma exerts a dilating effect on the intraparenchymal airways through the airway-parenchymal attachments. The number and integrity of parenchymal attachments, as well as the structural composition of the lung parenchyma itself, may all impact on the impedance offered by the lung parenchyma to ASM shortening. Other sources of impedance within the airway wall relate to the distortion and compression of the structure internal to the ASM layer (basement membrane, subepithelial matrix) and the mechanical impedance within the ASM layer itself (17, 19).
The composition of the ECM affects the viscoelastic properties of the tissue (13). Decorin is involved in collagen fibril formation and collagen assembly, which influences the tensile strength of the connective tissue (13). In addition, decorin is an important component of the extracellular proteoglycan matrix, a main source of the viscoelastic behavior of tissues (2, 6). In previous animal studies, decorin deficiency has been shown to be associated with abnormal collagen morphology in tissues such as skin and tendon (3, 6). In in vitro experiments, the skin of Dcn−/− mice had reduced tensile strength and increased fragility (3); tendons showed altered viscoelastic properties (6).
In the current study, we examined the role of decorin on airway and lung parenchymal mechanics, and in their mechanical interactions. Modifications in the elastic and resistive behavior of the airways and the lung parenchyma may alter the impact of parenchymal tethering on the airways under baseline conditions, and even more so during induced bronchoconstriction. In addition, when transpulmonary pressure is changed, the dilating effect of the parenchyma on the airway may be different. The degree to which airway calibre is affected by lung volume or transpulmonary pressure change will depend on the properties of the airways, the properties of the lung parenchyma, and on their mechanical interdependence. An “optimal coupling stiffness” (1) can be postulated, which describes the condition when linking of the airways and parenchyma favors the maximal transmission of strain to modulate bronchoconstriction.
In a previous study from this laboratory (8), baseline Raw was found to be lower in Dcn−/− than in Dcn+/+ mice. We speculated that the lack of decorin in the airway wall made the airway more compliant. The data of the current experiment in isolated tracheal rings showing increased length changes for a given change in tension (Fig. 3) support this hypothesis. We also reasoned that, in so far as bronchoconstriction is limited by mechanical load, if airway parenchymal interdependence were altered in Dcn−/− mice, then airway responsiveness to MCh would be affected. Furthermore, increasing transpulmonary pressure would have a greater effect on measurements of airway calibre, both under baseline conditions and, especially, during induced bronchoconstriction.
Although there was no difference in MCh-induced bronchoconstriction in Dcn−/− and decorin-replete mice, increasing PEEP had a greater effect on Raw and tissue stiffness in the Dcn−/− mouse after induced bronchoconstriction (Fig. 2). There are several possible explanations for this observation. One is that increased lung volume resulted in enhanced recruitment of closed airways in the Dcn−/− mice. Because Dcn−/− mice have increased lung compliance, an equivalent change in PEEP would result in a greater change in actual lung volume. Airways that were previously closed at low lung volume would be subject to a greater change in actual volume and would reopen. This mechanism could be especially pertinent in the Dcn−/− mice, since their airways were more compliant and therefore more prone to closure, especially during induced constriction. The effect of increasing PEEP on H also supports this interpretation. Usually, increasing lung volume results in “stiffer” lungs; the decrease in H at higher PEEP in the Dcn−/− mice is consistent with increased recruitment of closed airways causing a decrease in H.
Another explanation for our findings is that the link between airways and parenchyma may be different in the Dcn−/− mouse. Immunohistochemical assessment of decorin in the airway wall (Fig. 4) showed that decorin was localized primarily in the adventitial layer, the level at which the parenchyma attaches to the airway wall. A lack of decorin in this layer may affect the way in which the parenchyma is linked to the ASM layer, resulting in an enhanced transmission of force from the parenchyma to the ASM. Alternately, a more compliant airway, as was demonstrated in the current study, may undergo more airway dilation, for the same amount of “pull” by the parenchyma.
We have previously shown that the elastic properties of the lung parenchyma are altered in the Dcn−/− mouse (8). This increased compliance may impact on airway parenchymal interactions because of enhanced volume change for a given change in PEEP or because of changes in the recoil properties of the attachment. The EM images are particularly interesting in this regard (Fig. 5). They show increased collagenization in the peripheral lung tissues as well as altered collagen morphology; fibrils were irregular and interwoven. This change in collagen structure may explain the changes in parenchymal mechanics observed in our previous experiment (8). One might hypothesize that increased collagenization would result in stiffer lungs; however, our previous data (8) demonstrated that quasistatic compliance was increased in peripheral lung strips. This apparent contradiction underscores the idea that it is the nature of the collagen, rather than the quantity, per se, that is important in determining the mechanical characteristics of tissues. For example, alterations in the way the collagen fibrils are organized or cross-linked could have an impact on the mechanical properties of the lung tissues. A recent study (9) has shown that decorin protects collagen fibrils from cleavage by collagenase. Increased turnover of collagen in the Dcn−/− lung could lead to abnormal induction of collagen synthesis, which might result in the interwoven collagen bundles in the lungs we describe here.
Defining the impact of altered matrix on airway and parenchymal dynamics may be pertinent to understanding the pathophysiology of asthma. Asthma is a disease characterized by chronic airway remodeling (14). Several studies have now shown that alterations in proteoglycan deposition contribute to this process (11, 20, 21). The potential role of decorin is somewhat complicated by the observation that, in human studies, decorin deposition seems to be decreased or unaltered in the airway wall of asthmatics compared with normal subjects (4, 16, 20, 22), whereas in animal models, decorin deposition is increased (21, 23). Some of the confusion may arise from the difficulty of sampling the decorin-rich adventitial airway layer in human subjects. Nonetheless, examining the role this molecule and other members of its proteoglycan subclass play in determining the mechanical properties of the airway wall is important. Although the Dcn−/− mouse is, by no means, an animal model of asthma, the data of Matsushita et al. (16) and de Kluijver et al. (4) showing decreased decorin deposition in the asthmatic airway wall underscores the importance of the potential role decorin may play in asthma pathophysiology. Our results showing an enhanced bronchodilating effect of PEEP in the Dcn−/− mouse are of particular interest.
These studies were supported by the J. T. Costello Memorial Research Fund, the Canadian Institutes of Health Research, and National Cancer Institute Grant RO1 CA-39481.
We thank Sarah Heath for excellent technical assistance and Heberto Ghezzo for help with statistical analysis.
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 © 2007 the American Physiological Society