We previously reported the ability of isoprostanes to induce airway hyperresponsiveness (AHR). In this study, we examined the signaling mechanisms underlying that phenomenon with the standard muscle bath technique. Responses to a threshold concentration of carbachol (CCh, 3 × 10−9 M) were significantly augmented by pretreatment for 20 min with 8-isoprostaglandin E2 (15-E2t-IsoP, 10−6 M): this AHR was obliterated in tissues pretreated with the selective Rho kinase (ROCK) inhibitor Y-27632 added 20 min before isoprostane, but not by cyclopiazonic acid (CPA). Increasing the CCh concentration to 3 × 10−8 M (still considerably less than the half-maximally effective concentration of CCh) evoked larger contractions that were also augmented significantly by 15-E2t-IsoP: this AHR was completely abolished in tissues pretreated with CPA as well as those pretreated with Y-27632. We noted, however, that Y-27632 and CPA profoundly effect baseline tone and the cholinergic response per se, which confounds the interpretation of the data summarized above. We therefore modified the protocol by using combinations of CCh and blocker (CPA, Y-27632, or nifedipine) that were equieffective. In this way, we found that AHR could not be demonstrated under conditions in which Rho/ROCK signaling or Ca2+ release was abolished (by Y-27632 and CPA, respectively). Likewise, other autacoids that act through G protein-coupled receptors via Rho/ROCK and Ca2+ release (serotonin, histamine) mimicked this effect of isoprostane, whereas bradykinin did not. We conclude that isoprostane-induced AHR is mediated in part through an action on Rho/ROCK signaling. This novel finding may contribute to a better understanding of the mechanisms underlying AHR and asthma.
- Rho kinase
- airway smooth muscle
- cholinergic responsiveness
- sarcoplasmic reticulum
the hallmark features of asthma include variable direct bronchoconstriction as well as a general or nonspecific increase in the sensitivity of the airway smooth muscle (ASM) to a variety of bronchoconstrictor stimuli (airway hyperresponsiveness, AHR) (39, 40). The former is due in part to a direct action of inflammatory mediators on the ASM. However, the mechanism(s) underlying the latter phenomenon is as yet unclear. There is a growing literature suggesting the involvement of various cytokines present in the airways during asthma, particularly interleukin-13 (36). We recently proposed (19) that isoprostanes may also be involved.
Isoprostanes are a class of molecules that arise in large part via peroxidative attack of membrane lipids and have therefore been used extensively as markers of oxidative stress. Many groups have now documented the rise in levels of isoprostanes in the bronchoalveolar lavage and breath condensates of asthmatic patients (4, 5, 25, 31, 43). Other groups, including our own, have described the powerful biological effects of these agents on essentially every cell type present in the lungs (as reviewed in Ref. 19). In ASM, we have described both direct constrictor and relaxant effects (8, 22) as well as direct electrophysiological effects (16). More recently, we described (7) the capacity of isoprostanes to enhance the cholinergic sensitivity of ASM. In particular, we found that responses to submaximal stimulation were augmented, but not those to maximally effective concentrations of carbachol (CCh) (i.e., there was a leftward shift of the lower portion of the concentration-response relationship but no change in the upper portion). As such, isoprostanes appear to be present in asthma and to trigger some of the hallmark features of asthma, namely direct bronchoconstriction and indirect nonspecific hyperresponsiveness.
In this study, we sought to examine further this phenomenon of isoprostane-induced AHR. In particular, our objective was to elucidate the signaling mechanism(s) underlying this effect. We used a variety of pharmacological tools to block or mimic hyperresponsiveness induced by 8-isoprostaglandin E2 (15-E2t-IsoP), finding Rho kinase (ROCK) and release of internally sequestered Ca2+ to be particularly important in this respect. These findings could have particular importance for diseases characterized by oxidative stress, such as asthma, cystic fibrosis, and pulmonary hypertension.
Preparation of isolated tissues.
Tracheae were obtained from cows (200–500 kg) euthanized at a local abattoir and immediately put in ice-cold physiological solution for transport to the laboratory. Tracheal smooth muscle was isolated by removing connective tissue, vasculature, and epithelium and then cut into strips parallel to the muscle fibers (∼1 mm wide).
Muscle bath technique.
Tracheal strips were tied with silk suture (Ethicon 4-0) to a Grass FT.03 force transducer on one end and to a Plexiglas rod that served as an anchor on the other end. These were bathed in Krebs-Ringer buffer (see below for composition) containing indomethacin (10 μM) and Nω-nitro-l-arginine (l-NNA; 10−4 M), bubbled with 95% O2-5% CO2, and maintained at 37°C; tissues were passively stretched to impose a preload tension of ∼1 g. Isometric changes in tension were amplified, digitized (2 samples/s), and recorded online (DigiMed System Integrator, MicroMed, Louisville, KY) for plotting on the computer. Tissues were equilibrated for 1 h before the experiments commenced, during which time they were challenged with 60 mM KCl three times to assess the functional state of each tissue.
Solutions and chemicals.
Tissues were studied with Krebs-Ringer buffer containing (in mM) 116 NaCl, 4.2 KCl, 2.5 CaCl2, 1.6 NaH2PO4, 1.2 MgSO4, 22 NaHCO3, and 11 d-glucose, bubbled to maintain pH at 7.4. l-NNA (10−4 M) and indomethacin (10 μM) were also added to prevent generation of nitric oxide and of cyclooxygenase metabolites of arachidonic acid, respectively.
All chemicals were obtained from Sigma Chemical and prepared as 10 mM stock solutions, either as aqueous solutions (carbachol) or in absolute EtOH [nifedipine; (R)-(+)-trans-4-(1-aminoethyl)-N-(pyridyl)cyclohexanecarboxamide dihydrochloride (Y-27632)] or DMSO (cyclopiazonic acid, CPA). Aliquots were then added to the muscle baths; the final bath concentration of solvents did not exceed 0.1%, which we have found elsewhere to have little or no effect on mechanical activity.
Cholinergic contractions were expressed as a percentage of the response to 60 mM KCl added during the equilibration period (immediately before onset of the experiment). In general, the peak magnitudes of sustained contractions were used to quantify contractile responses to the agents used here. The one exception was data obtained in the presence of CPA plus CCh, where we were forced to digitally average the recording traces as described in ⇓⇓⇓⇓Fig. 5. Data are reported as means ± SE; n refers to the number of animals. Statistical comparisons were made with ANOVA (with Student-Newman-Keuls post hoc test) or Pearson's correlation analysis, as appropriate; P < 0.05 was considered statistically significant.
Pharmacological dissection of isoprostane-induced hyperresponsiveness.
We previously described (7) the capacity of isoprostanes to enhance ASM sensitivity to cholinergic stimuli, a phenomenon that is exemplified in Fig. 1A. To reliably demonstrate this effect, we challenged tissues two times with a test pulse of CCh (CCh1 and CCh2, respectively) at a threshold concentration (3 × 10−9 M; see Fig. 1B) and then pretreated with the isoprostane 15-E2t-IsoP (10−6 M) for 20 min before a third test pulse challenge with the cholinergic agonist (CCh3). The pretreatment with isoprostane typically evoked a small constrictor response of its own (Tpretreatment in Fig. 1A), which was only a fraction of the maximal cholinergic response (Fig. 1B). The response to CCh, however, was markedly greater in the presence of isoprostane than it was before isoprostane pretreatment (CCh3 vs. CCh2; Fig. 2A); in quantifying CCh3, we omitted the small increase in baseline tension evoked by the isoprostane (Tpretreatment in Fig. 1A). We used this experimental protocol to probe the signaling pathways underlying isoprostane-induced AHR in the present study. We first investigated the effects of depleting the internal Ca2+ store with CPA (10−5) or of inhibiting the Rho/ROCK pathway with the ROCK inhibitor Y-27632 (10−5); both agents were added 20 min before CCh3. Two concentrations of carbachol were used, 3 × 10−9 and 3 × 10−8 M, both of which are well below the half-maximally effective concentration of CCh (Fig. 1B).
When test pulses of 3 × 10−9 M CCh were used (Fig. 2A), the contractile responses were barely discernable in control tissues but were significantly augmented by 15-E2t-IsoP (Fig. 2A, right; see also positive value for CCh3 − CCh2 in Fig. 2A, left).
When the pretreatment included CPA, the isoprostane induced a greater change in baseline tone, as reflected in the tendency for the points in the scatterplot to be displaced more to the right. The isoprostane-induced change in cholinergic responsiveness, however, was markedly augmented in the presence of CPA (Fig. 2A, right, 2nd pair of bars). Y-27632, however, obliterated the isoprostane-induced change in baseline tone as well as the isoprostane-induced AHR (reflected in a clustering of all the points to the origin of the scatterplot and suppression of 3rd pair of bars in Fig. 2A, right).
Figure 2B summarizes the results obtained when test pulses of 3 × 10−8 M CCh were used to evoke larger cholinergic contractions (∼75% of the response to 60 mM KCl, consistent with the results shown in Fig. 1B). Pretreatment of those tissues with 15-E2t-IsoP again elevated baseline tone (points in scatterplot tend to right of y-axis; Fig. 2B) and augmented the cholinergic responses (points shifted upward; 1st pair of bars, Fig. 2B, right). CPA tended to further increase the baseline tone (rightward shift in scatterplot; Fig. 2B) but abolished the isoprostane-induced augmentation of the cholinergic response (net zero value for CCh3 − CCh2 in scatterplot; 2nd pair of bars, Fig. 2B, right). Y-27632, once again, abolished the isoprostane-induced change in baseline tone (clustering of points in scatterplot on y-axis) as well as the isoprostane-induced AHR (net negative value for CCh3 − CCh2) (Fig. 2B).
Pharmacological manipulation of AHR with equieffective cholinergic stimuli.
While the experimental approach and data analysis summarized above highlight isoprostane-induced AHR and its sensitivity to CPA or to Y-27632, they also clearly show how the latter interventions profoundly alter the experimental condition before cholinergic stimulation (Tpretreatment is a mixture of spontaneous activity, tone evoked by the blocker, and tone elicited by the isoprostane). Thus it is difficult to distinguish between the effects of those blockers on cholinergic responsiveness from their effect on AHR.
To overcome this methodological problem, we looked for combinations of CCh plus blockers that evoked comparable degrees of contraction (to make the magnitudes of the test responses uniform in each treatment group). We therefore assessed the cholinergic concentration-response relationship in the presence vs. absence of CPA (10−5 M), Y-27632 (10−5 M), or nifedipine (10−6 M) (Fig. 3). Comparison of those relationships showed that the response to 10−8 M CCh alone is comparable in magnitude to challenge with 10−7 M CCh in the presence of Y-27632, with 10−8 M CCh in the presence of nifedipine, or with 3 × 10−9 M CCh in the presence of CPA (Fig. 3). It goes without saying, however, that the signaling pathways activated under those four experimental conditions are quite different. In this way, in each treatment group the control response involved pretreating the tissue for 10 min with the blocker and then challenging with CCh (CCh2), whereas the test response involved treatment with the blocker (10 min), then with isoprostane (10 min), and finally with CCh (CCh3). An example of this novel experimental approach is given in Fig. 5A, and the data so obtained are summarized in Figs. 4 and 5.
15-E2t-IsoP augmented the response to 10−8 M CCh in the absence or the presence of nifedipine (Fig. 4A, 1st and 3rd pairs of bars, respectively). However, it had no statistically significant effect on the mean magnitude of the response to 10−7 M CCh plus Y-27632 (Fig. 4A, 2nd pair of bars). A plot of the individual data for each tissue in the three treatment groups (Fig. 4B) highlights the fact that with this experimental protocol we normalized the change in baseline tone brought on by the blockers (Tpretreatment is generally much less than 20% of KCl here but ranged up to 250% of KCl with the protocol exemplified in Fig. 2). Correlation analysis showed no statistically significant relationship between the increase in baseline tone and the degree of AHR in the Y-27632-treated tissues (R = 0.332; P = 0.29); the correlation between these two parameters was extremely high but not statistically significant in control tissues (P = 0.06; R = 0.581). When voltage-dependent Ca2+ influx was blocked with nifedipine, on the other hand, there was a highly statistically significant correlation between change in baseline tone and AHR (P < 0.0001; R = 0.949).
The effect of CPA on AHR was more difficult to quantify, however, because CCh plus CPA generally evokes repetitive twitch contractions that gradually fade in amplitude rather than a simple sustained contraction (see, e.g., Fig. 5A; also described in more detail previously in Refs. 1, 21, and 24). Instead, to quantify these data, we took all 10 digitized recordings (pairs of tissues from 5 different animals) and melded them digitally into one in order to smooth out the phasic twitches (see Fig. 5A for details). Figure 5B shows 5-min-long segments of the composite trace under the different experimental conditions, as well as the mean amplitude of those segments. When the internal Ca2+ store was depleted with CPA, the mean cholinergic response over and above the tone evoked by the isoprostane was not significantly different from the cholinergic response in the absence of the isoprostane: in other words, AHR was not seen in the presence of CPA.
Mimicry of AHR by other constrictor agents.
Collectively, the data summarized above suggest that 15-E2t-IsoP augments cholinergic responsiveness by enhancing Ca2+ handling and/or Rho/ROCK signaling. Because both of these signaling pathways are also used by other autacoids that act through G protein-coupled receptors (39), we hypothesized that those other autacoids should be able to mimic the effects of 15-E2t-IsoP on cholinergic responsiveness. To test this hypothesis, we first determined the concentrations of serotonin, histamine, bradykinin, and ATP, which evoke contractions comparable in magnitude to that of 10−6 M 15-E2t-IsoP. Figure 6 summarizes the concentration-response relationships for these agonists; more importantly, these data show that 10−7 M serotonin, 10−6 M histamine, and 10−5 M bradykinin are all approximately equieffective with 10−8 M CCh.
We adapted the experimental protocol summarized in Fig. 1 to assess agonist-induced AHR: tissues were challenged two times with 3 × 10−9 M CCh (CCh1 and CCh2) and then preconditioned with those equieffective stimuli before cholinergic responsiveness was assessed a third time (CCh3). Figure 7 summarizes the results, showing isoprostane-induced AHR to indeed be mimicked by serotonin and histamine. Interestingly, bradykinin had no such effect on cholinergic responsiveness (Fig. 7), even though it nonetheless activated the contractile apparatus on its own (Fig. 6).
Isoprostanes are a very biologically active, and therefore clinically relevant, group of autacoids. Our group (19) and many others have described their direct actions on smooth muscle, evoking either a constrictor or a relaxant response depending on the specific nature of the isoprostane and of the tissue being studied. Our group also described (7) their ability to act indirectly on smooth muscle, modifying the responsiveness to other agonists. In particular, that study documented the ability of isoprostanes to augment responsiveness to cholinergic and histaminergic stimuli, as well as to potassium chloride, apparently via an action on prostaglandin F2α-selective prostanoid receptors on the plasmalemma (7). However, in that study we did not examine the intracellular mechanisms by which this effect was exerted.
Gerthoffer (14) previously described augmentation of cholinergic responsiveness in ASM by histaminergic or serotonergic stimuli. This hyperresponsiveness was modeled mathematically and inferred to involve some postjunctional component of excitation-contraction coupling, including sensitization of the contractile apparatus to Ca2+ (14). However, the specific contribution of Rho/ROCK or of Ca2+ release was not investigated directly, as we have done here. Although fura-2 fluorimetry was used to show that the phenomenon was accompanied by an increase of intracellular Ca2+ concentration ([Ca2+]i), that event could be interpreted to involve voltage-dependent Ca2+ influx as well as release of internal Ca2+.
Although it was relatively straightforward to demonstrate isoprostane-induced AHR per se in the present study and our previous study (7), it was technically difficult to probe the signaling mechanisms underlying this phenomenon because the pharmacological interventions also directly and seriously impact cholinergic responsiveness itself: the large increase in “baseline tone” (Tpretreatment in Fig. 1) induced by CPA and the powerful suppression of all constrictor activity induced by Y-27632 were particularly troublesome. To overcome this obstacle, we revised the experimental protocol in such a way as to strengthen the cholinergic stimulus (by increasing the concentration of CCh) in order to overcome the inhibitory effect of the pharmacological blocker. This allowed us to compare cholinergic responses of similar magnitude but constituting very different signaling pathways. For example, when using Y-27632 it was necessary to increase CCh concentration 10-fold to get responses similar in magnitude to the control responses; although this might have been accompanied by a greater degree of Ca2+ release, at least it could be said that the response did not involve ROCK.
In this way, we were able to show that isoprostane-induced AHR was sensitive to Y-27632 as well as to depletion of the internal Ca2+ store with CPA, suggesting that this AHR involves stimulation of the Rho/ROCK signaling pathway and/or release of internally sequestered [Ca2+]i. We were also able to rule out any involvement of voltage-dependent Ca2+ channels [a hypothesis that could not be ruled out in the previous study (14)]. In other work from our lab (23, 27), we have documented the interaction and synergism of these two apparently distinct signaling events in ASM.
The finding that Rho/ROCK activation and/or Ca2+ release were involved in isoprostane-induced AHR led us to test whether AHR could also be produced by other agonists that act through these same signaling pathways: consistent with this prediction, threshold stimulation with either serotonin or histamine did markedly enhance cholinergic responsiveness. We were intrigued to find, however, that bradykinin did not cause AHR even though it did elevate baseline tone: although maximal bradykinin-induced contractions are only a fraction of the magnitude of those to cholinergic, serotonergic, or histaminergic stimuli (Fig. 6), the same can be said for 15-E2t-IsoP (Fig. 1). Bradykinin elevates [Ca2+]i (13, 41, 45), as do cholinergic agonists (20, 24, 41), serotonin (42, 46), and histamine (13, 41). It could be argued that bradykinin acts simultaneously on excitatory and inhibitory receptors on the ASM (the latter coupled through adenylate cyclase), thus confounding the overall net response: however, the same is true of histamine and the isoprostanes. Another interpretation is that bradykinin receptors do not couple to the Rho/ROCK signaling pathway, as is the case for cholinergic agonists (24) and histamine (18); the involvement of that pathway in ASM responses to serotonin and bradykinin has not yet been described.
The finding that isoprostanes profoundly augment ASM responsiveness to threshold and even subthreshold cholinergic agonist is highly clinically relevant: the airways in vivo are frequently exposed to such concentrations of acetylcholine and rarely experience concentrations approaching the maximally effective range. Mitchell and Sparrow (28, 29) showed that concentrations of agonists that are submaximal with respect to tension development in excised strips are nonetheless sufficient to evoke complete airway narrowing in intact bronchial segments. Concurrently, isoprostanes have been shown to accumulate in airway tissues and fluids of patients with asthma (4, 5, 31, 43), cystic fibrosis (10, 11, 33), chronic obstructive pulmonary disorder (6, 26, 30, 35), or pulmonary hypertension (12) or in normal people after the inhalation of ozone (15), cigarette smoke (3, 9, 30, 34), or allergen (32). In fact, we previously proposed (19) that isoprostanes may play an important role within the pathophysiology underlying some of the diseases referred to above: the data presented in the present study support this hypothesis.
Many laboratories are examining the possible roles of various cytokines and interleukins in the mechanisms underlying AHR in asthma. For example, interleukin-13 accumulates in the airways in asthma and augments ASM responsiveness, apparently via an action on Ca2+ homeostasis (13, 41); the same can be said for tumor necrosis factor (TNF)-α (2) and interleukin-1β (44). Our data suggest that the actions of these cytokines on Rho/ROCK signaling should also be investigated. Indeed, several recent reports present indirect data (sensitivity to the selective ROCK inhibitor Y-27632 or increased expression of Rho protein) that TNF-α stimulates Rho/ROCK signaling in ASM (17, 37, 38). It remains to be shown directly whether or not the activities of these signaling molecules are increased by TNF-α and/or by the other cytokines referred to above. Other future studies should examine the coupling mechanisms linking activation of the cytokine receptors and Rho/ROCK. G protein-coupled receptors—such as those for cholinergic agonists, histamine, serotonin, and bradykinin—stimulate Rho/ROCK through activation of G12,13. However, whether the latter can also couple to cytokine receptors is entirely unclear.
In conclusion, we have shown that 15-E2t-IsoP augments cholinergic responsiveness in ASM by enhancing Ca2+ handling and/or Rho/ROCK signaling. This novel finding may contribute to a better understanding of the mechanisms underlying AHR and asthma.
These studies were supported by operating funds provided by the Canadian Institutes of Health Research, the Ontario Thoracic Society, and GlaxoSmithKline (Canada).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2006 the American Physiological Society