INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

ENDOTHELIN (ET)-1 is a 21-amino-acid peptide isolated from vascular endothelial cells (37). ET-1 acts on two G protein-coupled receptors (1, 29) and has been reported to be one of the most potent agonists of airway smooth muscle (23, 36). It has been also suggested that ET-1 may have pathophysiological roles in asthma (10, 19, 32), pulmonary inflammation (22), and pulmonary vascular disease (9). Meanwhile, it has been recently shown that there exists the physiological and pathophysiological importance of ET-1. Kurihara et al. (14) have disrupted the mouse Edn1 locus encoding ET-1 by gene targeting. They have demonstrated that Edn1^/ homozygous mice present morphological abnormalities of the pharyngeal arch-derived craniofacial tissues and organs, indicating that ET-1 is essential to normal embryonic development. They have further reported that Edn1^/ knockout mice display cardiovascular malformations including aortic arch malformations and ventricular septal defect (13). Recently, it has been also postulated that ET-1 might have substantial physiological roles in maintaining normality of respiratory system. For example, it has been demonstrated that Edn1^+/ heterozygous knockout mice, which genetically produce lower levels of ET-1, present methacholine (MCh) hyperresponsiveness compared with Edn1^+/+ wild-type mice (25).

Airway hyperresponsiveness is one of the most cardinal features of bronchial asthma. It is postulated that both functional and structural factors potentially contribute to etiology of airway hyperresponsiveness (16, 31). Structural mechanisms include airway wall thickening caused by airway remodeling and airway edema. Possibly, airway smooth muscle dysfunction makes a functional contribution to asthmatic airway hyperresponsiveness (31). Increased mass of airway smooth muscle may be both a functional and structural factor of airway hyperresponsiveness (16, 31). In severe asthma, thickening of all layers, including the airway inner wall and smooth muscle, has been demonstrated, suggesting that structural alterations in the airway wall may be particularly important in terms of contributing factors of asthma (15).

Currently, it remains unclear whether ET-1 per se has either proliferative or antiproliferative roles in the development of airway remodeling. ET-1 has been shown to cause proliferation of cultured smooth muscle cells (35), indicating the possibility that smooth muscle cell content in the airways might be reduced by the disruption of ET-1 gene. Meanwhile, it has been reported that ET-1 per se has antiproliferative effects in certain cells such as hepatic stellate cells, which are involved in hepatic remodeling (8). In addition, ET-1 induces the production of prostacyclin and nitric oxide (4), which are reported to be antiproliferative mediators (3, 38). On the basis of these reports, one could also assume that ET-1 gene disruption might provoke airway remodeling in vivo.

In the present study, we hypothesized whether ET-1 and ET-1 gene might play a role in airway remodeling, leading to the altered airway responsiveness. To test this hypothesis, we investigated airway function in vivo and airway wall structure in Edn1^+/ heterozygous knockout mice and Edn1^+/+ wild-type mice.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

ET-1 gene knockout mice. ET-1 knockout mice were established by gene targeting (14). Mice with the genetic background of the 129Sv/J × ICR hybrid were backcrossed to ICR mice to achieve a uniform genetic background. These mice (more than 10th generation of backcross) heterozygous for Edn1 mutant allele were mated. The animals were maintained on a 12:12-h light-dark cycle with light from 0900 to 2100 at 25°C. Mice were fed with a normal diet and water ad libitum. Offspring were genotyped at 3 wk of age. For genotyping, genomic DNAs were isolated from biopsied tail and subjected to 28 cycles of PCR amplification (1 min at 95°C; 1 min at 60°C; 1 min at 72°C). The primers were sense 5'-TGTCTTGGGAGCCGAACTCA-3' and anti-sense 5'-GCTCGGTTGTGCGTCAACTTCTGG-3'. The PCR product consisted of 537 bp.

No viable Edn1^/ homozygotes were observed, as previously reported (14). Eight-week-old male littermates (Edn1^+/ or Edn1^+/+) were used in the present study.

Animal preparation. Animals were anesthetized with pentobarbital sodium (25 mg/kg ip) and ketamine hydrochloride (25 mg/kg ip) in combination and then paralyzed with pancuronium bromide (0.3 mg/kg ip). Anesthesia and paralysis were maintained by supplemental administration of 10% of the initial dose every hour. After tracheostomy, an endotracheal metal tube (inside diameter of 1 mm, length of 8 mm) was inserted in the trachea. Animals were mechanically ventilated (model 683, Harvard Apparatus, South Natick, MA) with tidal volumes of 8 ml/kg and frequencies of 2.5 Hz. The thorax was widely opened by means of midline sternotomy, and a positive end-expiratory pressure (PEEP) of 3 cmH₂O was applied by placing the expired line underwater. During the experiments, oxygen gas was continuously supplied to the ventilatory system. Under these ventilatory conditions, arterial pH, PO₂, and PCO₂ were 7.35-7.45, 100-180 Torr, and 30-45 Torr, respectively, at the end of experiments. A heating pad was used to maintain the body temperature of animals.

Tracheal pressure was measured with a piezoresistive microtransducer (Endevco 8510B-2, San Juan Capistrano, CA) placed in the lateral port of the tracheal cannula. Tracheal flow was measured by means of a Fleisch pneumotachograph (model no. 00000). All signals were amplified, filtered at a cutoff frequency of 100 Hz, and converted from analog to digital with a converter (DT2801-A, Data Translation, Marlborough, MA). The signals were sampled at a rate of 200 Hz and stored on an IBM-AT-compatible computer.

Lung elastance and resistance were measured as previously described (2, 23, 25). From flow, volume (V), and tracheal pressure (Ptr), lung elastance (EL) and total lung resistance (RL) were calculated by finding the best fit for the equation of motion
where K is a constant, the value of which was also estimated by multiple linear regression and was <0.5 cmH₂O different from the real value of PEEP.

Airway responsiveness to administration of agonists. We chose serotonin (5-HT) and MCh as agonists in this study. Inhalations of saline and agonists were administered at PEEP of 3 cmH₂O. At the start of the protocol, two deep inhalations (three times tidal volume) were delivered to standardize volume history. All animals were then challenged with saline aerosol for 2 min. Aerosols were generated by an ultrasonic nebulizer (Ultra-Neb100, DeVilbiss, Somerset, PA) and delivered through the inspiratory line into the trachea. Measurements of 10-s duration were sampled during tidal ventilation 1 min after administration of saline aerosol. This represented the baseline measurement. Then, each dose of 5-HT or MCh aerosol was administered for 2 min in a dose-response manner (5-HT, 0.15-20 mg/ml; MCh, 0.31-80 mg/ml). Airway responsiveness was assessed by using the concentration of agonists required to double lung resistance (EC₂₀₀RL), which was calculated by interpolation.

Morphometric study. In five animals from each group, airway smooth muscle was quantitated by using morphometric techniques. The lungs were removed intact and fixed with 10% Formalin at 25 cmH₂O inflation pressure for 48 h. After fixation, the tissue blocks obtained from midsagittal slices of the lungs were embedded in paraffin. Blocks were cut 4 µm thick by using a microtome. Slides were stained with hematoxylin-eosin. We assessed tissue shrinkage, and subsequent measurements were corrected for shrinkage.

Photomicrographs of airways were obtained from all slides. Using a digitizer, we subsequently performed morphometric analyses of airway wall. To assess whether airways were cut in cross section, the maximal diameter of the airway (D₁) and the diameter at the widest point perpendicular to this axis (D₂) were measured. Airways with a D₂/D₁ ratio of >0.7 were analyzed in the present study. We measured the length of the epithelial basement membrane (L_bm) and the areas enclosed by basement membrane and by inner and outer borders of smooth muscle (A_bm, A_mi, and A_mo, respectively) (12, 34). Inner wall area (WA_i), the areas of lamina propria and airway smooth muscle were calculated as A_mo  A_bm, A_mi  A_bm, and A_mo  A_mi, respectively. The ideal area of the lumen of the relaxed airway (A_bm^*) was then calculated as A_bm^* = L_bm²/4. We normalized each area of airway components to the ideally relaxed area (area/A_bm^*) to adjust for differences in airway size (12).

Data analysis. Comparisons of physiological and morphometric data between the experimental groups were carried out with Student's t-test or ANOVA (Scheffé's test). Data are expressed as means ± SE. P values <0.05 were taken as significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Responsiveness to agonists. There were no significant differences in baseline EL and RL between Edn1^+/ and Edn1^+/+ mice (EL, 12.2 ± 0.7 and 12.0 ± 0.4 cmH₂O/ml; RL, 0.386 ± 0.018 and 0.383 ± 0.009 cmH₂O · ml¹ · s, respectively). Figure 1 shows 5-HT dose-response curves for EL in the two groups. There was no difference in elastic responses to 5-HT. MCh dose-response curves for EL are demonstrated in Fig. 2. Each dose of MCh induced a significantly greater response in Edn1^+/ mice compared with Edn1^+/+ mice.

View larger version (12K): [in this window] [in a new window]   Fig. 1.   Serotonin (5-HT) dose-response curves for lung elastance (EL) in Edn1+/ and Edn1+/+ mice (n = 4 for each group). SAL, saline.

View larger version (13K): [in this window] [in a new window]   Fig. 2.   Methacholine (MCh) dose-response curves for EL in Edn1+/ and Edn1+/+ mice (n = 6 for each group). # P < 0.05, * P < 0.01 compared with Edn1+/+ mice.

Responses for RL are summarized in Fig. 3. Log scale was applied to EC₂₀₀RL to normalize this parameter. As shown, mice were more sensitive to 5-HT administration compared with MCh in terms of doses, whereas there were no differences in 5-HT responsiveness between the two groups. Meanwhile, MCh airway responsiveness in Edn1^+/ mice was significantly greater than that in Edn1^+/+ mice (logEC₂₀₀RL: 0.223 ± 0.037 vs. 1.259 ± 0.161, P < 0.0001).

View larger version (9K): [in this window] [in a new window]   Fig. 3.   Airway responsiveness expressed as concentration of agonists required to double lung resistance (EC200RL). * P < 0.0001 compared with Edn1+/+ mice.

Morphometric study. Table 1 summarizes the morphometric data of airway size and roundness. There were no significant differences in the number of airways, airway size, and airway roundness between Edn1^+/ and Edn1^+/+ mice, indicating that there were no significant biases between the experimental groups in terms of airway selection.

As shown in Fig. 4, there were no significant differences in thickness of either lamina propria or airway smooth muscle between the two groups. In addition, no significant difference in WA_i was observed between Edn1^+/ and Edn1^+/+ mice (WA_i/A_bm^*, 0.208 ± 0.011 and 0.210 ± 0.016, respectively), suggesting that airway structure of Edn1^+/ mice is not altered compared with that of Edn1^+/+ mice.

View larger version (35K): [in this window] [in a new window]   Fig. 4.   Thickness of lamina propria and airway smooth muscle in Edn1+/ and Edn1+/+ mice. Area/Abm*, area normalized with the ideally relaxed airway size.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The results of the present experiments show that pulmonary responsiveness to MCh, but not to 5-HT, was markedly enhanced in Edn1^+/ mice compared with Edn1^+/+ mice. There was, however, no difference in morphometrically assessed airway structure between Edn1^+/ and Edn1^+/+ mice. These findings suggest that the decrease in ET-1 concentration may induce MCh airway hyperresponsiveness via possible airway dysfunction but not airway remodeling.

Before dicussion of the results, technical issues warrant consideration. In this study, we used Edn1^+/ heterozygous mice, but not Edn1^/ homozygous mice, as ET-1 gene-disrupted mice. Although Edn1^/ mice would be most desirable to examine the present hypothesis that ET-1 and the gene encoding ET-1 might be associated with airway remodeling, Edn1^/ mice died of apnea at birth, and viable Edn1^/ mice were unavailable (14). Meanwhile, Edn1^+/ heterozygous mice, in which ET-1 level in plasma was reduced by 40% compared with the wild-type mice (14), were all viable. Therefore, Edn1^+/ mice were used in the present study. In addition, a uniform genetic background was essential to make meaningful comparisons of the experimental groups. In the present model, we backcrossed the mice to ICR mice to achieve a uniform genetic background (more than 10th generation of backcross). Therefore, the mice used in the present study were of the ICR background. Whereas ICR strain is not strictly inbred but rather closed-circle strain, we used littermates of knockout heterozygotes and wild-type mice to minimize potential effects of minute genetic variabilities within the ICR strain.

It has been described that changes in EL reflect lung parenchymal alterations and stiffening of the lungs induced by various contractile stimuli (7, 27), although the contraction of the conducting airways can also cause changes in EL (20). Meanwhile, increases in RL represent decreases in airway luminal cross-sectional area (16, 27). In this study, enhanced responses in both EL and RL to MCh administration were observed in Edn1^+/ mice, suggesting that heterozygous disruption of ET-1 gene elicits increased responses to MCh at both lung parenchyma and airways.

One of the possible mechanisms to explain this finding is that ET-1 and ET-1 gene expression might affect airway structure, especially, lamina propria and airway smooth muscle thickness. Airway remodeling including thickening of airway smooth muscle is a feature in asthmatic subjects and could be involved in bronchial hyperresponsiveness (16, 31). It has been shown by Lambert and Paré (16) that a marked increase in airway responsiveness is theoretically induced by thickening of airway wall including lamina propria and smooth muscle layers. In the present study, however, no significant difference in either lamina propria or airway smooth muscle thickness was observed between ET-1 knockout and wild-type mice. These results suggest that heterozygous disruption of ET-1 gene has little effect on airway remodeling in mice. Although Edn1^+/ mice represent normal airway structure, this finding might still reflect a balanced state between proliferative and antiproliferative effects of ET-1 in vivo. It seems that the effect of ET-1 on altered airway responsiveness depends on airway dysfunction, but not airway remodeling, in the present model. Possibly, the deficiency in ET-1 may affect airway function by modulating the production of bronchodilating mediators such as nitric oxide, resulting in this observed MCh hyperresponsiveness in Edn1^+/ mice (4, 28).

Airway responsiveness to 5-HT was not affected by disruption of ET-1 gene. These observations suggest that ET-1 and ET-1 gene may be specifically related to bronchial responsiveness to MCh but not to 5-HT. Recently, it has been demonstrated that 5-HT and ACh airway hyperresponsiveness is inherited independently in mice and that murine nonspecific airway hyperresponsiveness is determined by multiple genes (17, 18). Of interest, it has been reported that 90% of the response to 5-HT could be blocked with atropine in 5-HT-hyperresponsive mice, suggesting that the genetic defect responsible for 5-HT airway hyperresponsiveness could be proximal to the muscarinic receptor (18). In the preliminary study, we observed no effects of atropine pretreatment on 5-HT-induced responses in Edn1^+/ mice. Currently, we do not know whether this lack of effect was due to the relative hyporesponsiveness to 5-HT, although it seems unlikely that 5-HT and MCh (ACh) exactly share the same pharmacological pathway, i.e., muscarinic ligand-receptor couplings. In addition, it has been shown that AKR/J strain mouse is hyperreactive to ACh but hyporeactive to 5-HT (18). In comparison with the control mouse, Edn1^+/ mouse is also hyperreactive to MCh but rather hyporeactive to 5-HT. We do speculate that Edn1^+/ mice may have similar genotype of AKR/J strain, although this notion requires genome-analysis study. The present results indicate that the mutation of ET-1 gene affects muscarinic receptor-specific responsiveness but not general bronchial responsiveness.

Hereditary factors potentially contribute to the etiology of bronchial asthma. A number of complex genes are suggested to be involved in the pathogenesis of asthma (30). On the basis of the reports that airway hyperresponsiveness in the mouse resembles human asthma (17, 18), murine models of asthma have been extensively used to approach candidate genes and loci (5). Recently, genetically engineered animals, including knockout and transgenic mice, have been used to study asthma-associated genes encoding bioactive mediators (6, 11, 24, 33). The present knockout mice, which genetically produce lower levels of ET-1 throughout development and life, may be useful to study the genetic contribution of ET-1 to the etiology of asthma. Meanwhile, whether overexpression of the ET-1 gene may affect airway responsiveness remains to be elucidated and warrants further study.

Recently, the physiological and pathophysiological importance of ET-1 in the respiratory system has been reported (14, 21, 25). Previous reports suggest that increased plasma level of ET-1 induced by various stimuli, such as exogenous administration, is related to bronchoconsriction, whereas decreases in ET-1 plasma level also evoke MCh pulmonary hyperresponsiveness. Taken together, ET-1 may have an important role in maintaining homeostasis in the respiratory system, and the metabolism of ET-1 might be strictly regulated to keep ET-1 level constant within physiological ranges.

In summary, Edn1^+/ mice, which generate lower levels of ET-1 than do wild-type mice, represent pulmonary hyperresponsiveness to MCh. However, neither airway remodeling nor airway smooth muscle thickening was detected in Edn1^+/ mice. These findings suggest that ET-1 gene expression is involved in MCh pulmonary responsiveness by acting as a functional mediator. The ET-1 knockout mice may provide appropriate models to study diseases related to ET-1 metabolism.