Imidazobenzodiazepine PI320 Relaxes Mouse Peripheral Airways by Inhibiting Calcium Mobilization

Abstract

Asthma is a common respiratory disease characterized, in part, by excessive airway smooth muscle (ASM) contraction (airway hyperresponsiveness). Various GABA_AR (γ-aminobutyric acid type A receptor) activators, including benzodiazepines, relax ASM. The GABA_AR is a ligand-operated Cl⁻ channel best known for its role in inhibitory neurotransmission in the central nervous system. Although ASM cells express GABA_ARs, affording a seemingly logical site of action, the mechanism(s) by which GABA_AR ligands relax ASM remains unclear. PI320, a novel imidazobenzodiazepine designed for tissue selectivity, is a promising asthma drug candidate. Here, we show that PI320 alleviates methacholine (MCh)-induced bronchoconstriction in vivo and relaxes peripheral airways preconstricted with MCh ex vivo using the forced oscillation technique and precision-cut lung slice experiments, respectively. Surprisingly, the peripheral airway relaxation demonstrated in precision-cut lung slices does not appear to be GABA_AR-dependent, as it is not inhibited by the GABA_AR antagonist picrotoxin or the benzodiazepine antagonist flumazenil. Furthermore, we demonstrate here that PI320 inhibits MCh-induced airway constriction in the absence of external Ca², suggesting that PI320-mediated relaxation is not mediated by inhibition of Ca²⁺ influx in ASM. However, PI320 does inhibit MCh-induced intracellular Ca²⁺ oscillations in peripheral ASM, a key mediator of contraction that is dependent on sarcoplasmic reticulum Ca²⁺ mobilization. Furthermore, PI320 inhibits peripheral airway constriction induced by experimentally increasing the intracellular concentration of inositol triphosphate (IP₃). These novel data suggest that PI320 relaxes murine peripheral airways by inhibiting intracellular Ca²⁺ mobilization in ASM, likely by inhibiting Ca²⁺ release through IP₃Rs (IP₃ receptors).

Asthma, a common cause of dyspnea, is a chronic airway disease characterized by excessive airway smooth muscle (ASM) contraction (airway hyperresponsiveness), inflammation, and increased mucus production. Despite current therapeutic approaches (e.g., β-agonists, corticosteroids, leukotriene modifiers, and immunomodulators), asthma remains a considerable source of morbidity and mortality (1). In the United States, the Centers for Disease Control and Prevention report 8.4% of adults suffer from asthma, leading to 5.8 million office visits and 1.6 million emergency department visits a year (2), underlining a persistent need for new therapeutic approaches.

We have previously demonstrated that GABA_AR (γ-aminobutyric acid type A receptor) activators, including benzodiazepines, directly relax ASM in several experimental paradigms (3–9). The GABA_AR is a ligand-operated, heteropentameric Cl⁻ channel best known for its role in inhibitory neurotransmission in the central nervous system (CNS) (10), hence the common clinical use of benzodiazepines as anxiolytics and sedatives. However, the GABA_AR is also expressed in several locations outside the CNS (11–14), including ASM (7, 12). This led us to hypothesize that benzodiazepines directly relax ASM by activating GABA_ARs on the surface of ASM cells, altering membrane potential, and inhibiting extracellular Ca²⁺ influx via membrane potential–sensitive pathways (5, 7, 8). However, this proposed mechanism has not been firmly established, and recent reports have suggested that one mechanism of benzodiazepine-mediated ASM relaxation may be independent of the GABA_AR (15, 16).

PI320, an asthma drug candidate, is a novel imidazobenzodiazepine (positive allosteric modulator of the GABA_A receptor) designed for tissue selectivity and favorable pharmacokinetic and pharmacodynamic properties. We previously demonstrated that inhaled PI320 is highly absorbed into the lungs of mice and quickly relaxes guinea pig tracheal rings in ex vivo organ bath experiments (17). Here we present evidence from precision-cut lung slice (PCLS) experiments suggesting that PI320 ameliorates bronchoconstriction in mice by relaxing peripheral ASM through an additional mechanism independent of GABA_AR stimulation. Rather, PI320-mediated peripheral airway relaxation appears to depend on the restriction of intracellular Ca²⁺ mobilization in ASM, likely by direct or indirect inhibition of sarcoplasmic reticulum (SR) Ca²⁺ release via IP₃R (inositol triphosphate receptor).

Methods

Animals

All animal experimental protocols were approved by the Columbia University Medical Center Institutional Animal Care and Use Committee. Eight- to 10-week-old A/J mice (Jackson Laboratories) were used for in vivo respiratory mechanics studies because of their increased airway responsiveness to inhaled methacholine (MCh). C57BL/6J mice were used for peripheral airway luminal area studies, and transgenic mice globally and constitutively expressing the fluorescent intracellular Ca²⁺ indicator GCaMP6f (C57BL/6J background, Jackson Laboratories) were used for Ca²⁺ oscillation studies in PCLS.

PI320

PI320, (R)-8-bromo-6-(2-fluorophenyl)-N-(2,5,8,11,14-pentaoxahexadecan-16-yl)-4-methyl-4H-benzo[f]imidazo[1,5-a][1,4]diazepine-3-carboxamide, an imidazobenzodiazepine and positive allosteric modulator of the GABA_AR, was provided by Pantherics Incorporated (17). Purity (>99%) was confirmed by high-performance liquid chromatography. Identity was determined by ¹H-nuclear magnetic resonance, ¹³C-nuclear magnetic resonance, and high-resolution mass spectrometry.

In Vivo Mouse Respiratory Mechanics Analyses

In vivo airway resistance and compliance were assessed in live A/J mice using a FlexiVent with an FX1 module (SciReq) and an in-line nebulizer (Aerogen). The mice were anesthetized with pentobarbital (50 mg/kg i.p.; Henry Schein), paralyzed with succinylcholine (10 mg/kg i.p., Sigma-Aldrich), and mechanically ventilated via a 20-gauge cannula placed into the trachea via tracheostomy (tidal volume, 10 mg/kg; 150 breaths/min; positive end expiratory pressure, 3 cm H₂O). The mice received nebulized 2 mM PI320 (10 s nebulization, 50% duty cycle; ∼25 μl delivered) or vehicle (water with 0.4% Tween-20, Sigma-Aldrich) 10 minutes before measuring lung resistances (conducting airway resistance and respiratory system resistance) and compliance during a graded, nebulized MCh (Sigma-Aldrich) challenge (0–50 mg/ml). Electrocardiogram and temperature monitoring were performed throughout the experiment. Resistance and compliance values for each mouse at each MCh dose represent an average of three measurements using the forced oscillation technique.

Mouse PCLS Preparation

PCLSs were prepared from mouse lungs as previously described (18, 19). Briefly, mice were killed with pentobarbital (Fatal Plus; Vortech Pharmaceuticals), their thoracic cavities opened, and their lungs inflated with warm 2% agarose in Hanks’ balanced salt solution (HBSS; Gibco) supplemented with 20 mM HEPES and adjusted to pH = 7.40 (sHBSS). After allowing the agarose to cool at 4°C for 20 minutes, the lungs were removed and maintained in ice-cold sHBSS for 15 minutes. The lung lobes were cut to form a base near the entrance of major bronchi, attached to a tissue holder with two drops of the 2% agarose solution, and then embedded with 6% gelatin in sHBSS to form a block. The block was cut into serial 130-μm sections with a vibratome (VF-300; Precisionary Instruments). Lung slices containing peripheral airways were incubated overnight in low-glucose Dulbecco’s modified Eagle medium supplemented with 1× antibiotic solution containing l-glutamine, penicillin, and streptomycin (Thermo Fisher Scientific) at 37°C and 10% CO₂.

Assessments of Peripheral Airway Constriction in PCLSs with Phase-Contrast Microscopy

PCLSs containing airways with a lumen diameter of 100–300 μm with active (beating) ciliated epithelial cells were used. PCLSs were mounted in a custom-made perfusion chamber on the stage of an inverted microscope (Nikon TMD) and were visualized using a 10× objective. The PCLSs were continuously suprafused with sHBSS and compounds of interest dissolved in 0.1% DMSO (PI320; MCh, picrotoxin, and caffeine: Sigma-Aldrich; flumazenil, muscimol, and 2-APB: Tocris). All experiments were performed at room temperature. To assess changes in airway cross-sectional luminal area, digital images were recorded at 0.5 Hz using a charge-coupled device camera (KP-M1A; Hitachi), frame grabber (Picolo; Euresys), and image-acquisition software (Video Savant; IO Industries). The cross-sectional lumen area of airways was determined in each image using a custom-built script in Video Savant to automatically recognize and count the airway lumen area. Finally, airway area changes were normalized to the initial area and plotted over time. For experiments conducted without Ca²⁺ in the suprafusion buffer, Ca²⁺ was excluded from the sHBSS, and 1 mM EGTA (Sigma-Aldrich) was added and the pH adjusted to 7.40.

Assessments of Intracellular Ca²⁺ Concentrations in PCLSs with Confocal Microscopy

PCLSs were prepared from mice expressing the genetically encoded intracellular Ca²⁺ probe GCaMP6f and incubated overnight at 37°C and 10% CO₂ as indicated above. PCLSs were mounted in a custom-made perfusion chamber and transferred to the stage of a custom-made, video-rate confocal microscope, described earlier (19, 20). The PCLSs were illuminated with a 488-nm laser beam, and the fluorescence emission (510–530 nm) was collected with a photomultiplier tube (PMT R3896; Hamamatsu) and frame grabber (Alta-AN; BitFlow). Fluorescence images were recorded at 15 Hz using Video Savant. The changes in fluorescence intensity in single ASM cells were analyzed by selecting regions of interest (ROI) ranging from 25 to 49 pixels. Average fluorescence intensities of an ROI were obtained from each image using a custom-written script in Video Savant that allowed the tracking of the ROI within an ASM cell as it moved with contraction. Final fluorescence values were expressed as a fluorescence ratio (F/F₀) normalized to the initial fluorescence (F₀).

Flash Photolysis of Caged-IP₃

To experimentally increase the intracellular concentration of the second messenger IP₃ in the cytosol of ASM cells within the peripheral airways, we used flash photolysis of caged-IP₃, as described earlier (19). Briefly, PCLSs were incubated with 2 μM caged-IP₃-PM (Enzo Life Sciences) for 45 minutes at 30°C in sHBSS containing 0.1% Pluronic F-127 (Sigma-Aldrich) and subsequently for 30 minutes at 30°C in sHBSS alone to allow de-esterification of the caged-IP₃ inside the cells. A mercury arc lamp was used as the source for the ultraviolet (UV) light; it was filtered with a band-pass filter (330 nm) and focused into the PCLS using a biconvex lens (focal distance, 200 mm) on the phase-contrast microscope. The illumination iris was adjusted to irradiate a circular region on the PCLS using an iris diaphragm at a conjugate image plane in the microscope. Flash duration was controlled using a Uniblitz shutter (Vincent Associates) with electronic timing control.

Statistical Analyses

For in vivo airway mechanics experiments, two-way ANOVAs with repeated measures and Bonferroni posttest comparisons were used. For PCLS airway area comparisons, either Student’s t test or two-way ANOVAs with repeated measures were used as indicated in each case after testing normality using the Kolmogorov Smirnov test. Analyses were performed using Prism 9 software (GraphPad). Data are presented as mean ± SEM, and P < 0.05 was considered significant.

Results

PI320 Ameliorates Bronchoconstriction in Vivo

In vivo mouse respiratory mechanics were assessed by the forced oscillatory technique using a flexiVent. Anesthetized, paralyzed A/J mice received 25 μl of 2 mM PI320 or vehicle by nebulization via a tracheotomy 10 minutes before a graded, inhaled MCh challenge. PI320 significantly inhibited MCh-induced increases in resistance of conducting airways measured using the low-frequency forced oscillation technique (Figure 1, left) and dynamic resistance of whole respiratory system (Figure 1, center) compared with vehicle control. Similarly, nebulized PI320 significantly inhibited MCh-induced decreases in lung compliance (Figure 1, right).

Figure 1.

Inhaled PI320 ameliorates methacholine (MCh)-induced bronchoconstriction in vivo. Nebulized PI320 significantly inhibits changes in (left) conducting airway resistance (R_N), (center) respiratory system resistance (R_RS), and (right) compliance in live, anesthetized mice exposed to an inhaled, graded MCh challenge (two-way ANOVA with repeated measures, P < 0.05). Asterisks indicate significant difference between PI320 and vehicle at the indicated MCh concentration by post hoc comparisons: *P < 0.05, **P < 0.01, and ***P < 0.001. Data points are means ± SEM of n = 9 mice.

PI320 Relaxes Peripheral Airways Constricted with MCh in PCLSs

To investigate whether PI320 relaxes peripheral airways, we prepared mouse PCLSs and used video phase-contrast microscopy to monitor the changes in airway cross-sectional luminal area during suprafusion of the PCLS with MCh and PI320 (Figure 2 and Video E1 in the data supplement). Addition of 0.3 μM MCh induces a fast reduction in airway lumen reaching 53.5 ± 4.2% of the area compared with before stimulation, and this area is maintained approximately constant during MCh suprafusion (Figures 2A and 2B). The addition of 30 μM PI320, in the continuous presence of MCh, provokes a fast and strong airway relaxation, increasing the lumen area by 44.3 ± 3.0% after 8 minutes of PI320 addition. The subsequent removal of PI320 by suprafusing the solution with MCh alone reversed the PI320 effect, causing a progressive airway reconstriction to reach a lumen area similar to the that before PI320 addition. Finally, removal of MCh causes airway relaxation, and its lumen area progressively reached the initial area before MCh exposure. To further characterize the relaxing effect of PI320, the airways preconstricted with MCh were exposed to different concentrations of PI320 in separate experiments, and the airway relaxation was quantified (Figure 2C). PI320 relaxes the airways from 1 μM to 300 μM with an apparent half-maximal inhibitory concentration of 44.5 μM, and the maximal airway relaxation was 82.5%. These results indicate that PI320 induces a strong, fast, reversible, and concentration-dependent relaxation of MCh-constricted peripheral airways.

Figure 2.

PI320-induced peripheral airway relaxation in precision-cut lung slice (PCLS). (A) Representative phase-contrast microscope images, obtained at the times indicated under the trace in B, of a peripheral airway in a PCLS before stimulation (resting), contracted in response to 0.3 μM MCh, relaxed after 30 μM PI320 addition, and recontracted after PI320 washout in the continuous presence of MCh. (B) Changes in cross-sectional luminal area over time, of the peripheral airway shown in A, in response to the suprafusion of the PCLS with MCh, PI320, or supplemented Hanks’ balanced salt solution (sHBSS) alone (not labeled) at the times indicated by the lines on top of the trace. (C) Concentration-response curve of PI320-induced airway relaxation. Airway relaxation was calculated as the increase in airway lumen area after 8 minutes of suprafusion with PI320, divided by the airway contraction induced by MCh just before PI320 addition. Data were obtained from separate experiments similar to that shown in A and B for each PI320 concentration. Half-maximal inhibitory concentration was 44.5 μM. Data points are means ± SEM of n = 12 slices from six mice. A time-lapse video of the data shown in A and B, showing MCh-induced airway contraction and subsequent relaxation induced by the addition of PI320, is included as Video E1.

PI320-induced Peripheral Airway Relaxation Is Not Inhibited by GABA_AR Antagonists

Because PI320 is an imidazobenzodiazepine and binds GABA_ARs, we first investigated whether its relaxing activity in the peripheral airways is antagonized by the GABA_AR antagonist picrotoxin (100 μM) or the benzodiazepine antagonist flumazenil (20 μM). Figures 3A–3D show that there are no statistically significant differences in the PI320-induced relaxation of airway preconstricted with MCh and exposed to either 0.1% DMSO (vehicle), picrotoxin, or flumazenil. In addition, 100 μM muscimol, a potent GABA_AR agonist, does not induce significant relaxation of peripheral airways preconstricted with MCh at any of the concentrations tested (Figures 3E and 3F). These results suggest that classic GABA_AR agonists like muscimol do not relax murine peripheral airways and that the relaxing effects of PI320 are not wholly mediated by GABA_AR activation in this model.

Figure 3.

GABA_AR (γ-aminobutyric acid type A receptor) antagonists do not inhibit PI320-induced peripheral airway relaxation, and the GABA_AR agonist muscimol does not relax peripheral airways in PCLSs. (A–C) Representative traces showing the effects of preexposure to either 0.1% vehicle (DMSO, A), 100 μM picrotoxin (PTX, B), or 20 μM flumazenil (FLU, C) on airway relaxation induced by 30 μM PI320 of airways precontracted with 0.3 μM MCh. Drugs were suprafused at the times indicated by the lines above the traces. (D) Summary of the effects of picrotoxin and flumazenil on the PI320-induced airway relaxation calculated as in Figure 2. Bars represent means ± SEM of n = 12 slices from three mice. ns = not significant, one-way ANOVA. (E) Representative trace showing the effects on airway luminal area of muscimol (100 μM) addition to airways precontracted with 0.3 μM MCh. (F) Concentration-response effects of muscimol (1–100 μM) on airways precontracted with MCh. Airway relaxation was obtained from separate experiments similar to that shown in Figure 2C and calculated as in Figure 1C for PI320-induced airway relaxation. Data points are means ± SEM of n = 4 slices from two mice and were not significantly different from the airway relaxation induced by 0.1% DMSO (P < 0.05, one-way ANOVA). ns = not significant.

PI320 Inhibits MCh-induced Peripheral Airway Constriction in the Absence of Extracellular Ca²⁺

To further investigate the mechanism of PI320-induced airway relaxation, we first studied whether PI320 relaxes the airways by inhibiting Ca²⁺ influx from the extracellular medium. To test this hypothesis, we compared MCh-induced airway constriction in the absence of extracellular Ca²⁺ in PCLSs exposed to either vehicle or PI320. Figure 4 shows that the airway constriction is sustained in the presence of extracellular Ca²⁺, whereas it is transient in the absence of extracellular Ca²⁺. Importantly, PI320 exposure inhibits the transient MCh-induced constriction in the absence of extracellular Ca²⁺. These results indicate that, even in the absence of extracellular Ca²⁺, PI320 inhibits MCh-induced constriction and suggest that the mechanism of PI320-induced peripheral airway relaxation is independent of inhibition of Ca²⁺ influx from the extracellular medium in this murine model of peripheral airways.

Figure 4.

PI320 inhibits the transient peripheral airway constriction induced by MCh in the absence of extracellular Ca²⁺ in PCLSs. (A and B) Representative traces showing airway constrictions induced by 0.3 μM MCh in the presence (first exposure) and the absence (second exposure) of extracellular Ca²⁺ (zero Ca²⁺ sHBSS supplemented with 1 mM EGTA) plus either 0.1% DMSO (A) or 30 μM PI320 (B). (C) Summary of the peak MCh-induced airway constriction in absence of extracellular Ca²⁺ with respect to the sustained MCh-induced constriction in presence of extracellular Ca²⁺, obtained from experiments like those shown in A and B. Bars represent means ± SEM of n = 6 slices from two mice. ***P < 0.001, unpaired t test.

PI320 Inhibits MCh-induced Intracellular Ca²⁺ Oscillations in ASM Cells in PCLSs

Next, we tested whether PI320 inhibited the mobilization of intracellular Ca²⁺ and the generation of Ca²⁺ oscillations, which play a fundamental role in the regulation of peripheral airway constriction. To this end, we prepared PCLS from a transgenic mouse that expresses the intracellular Ca²⁺ sensor GCaMP6f and used a video-rate confocal microscope to assess the changes in intracellular Ca²⁺ concentrations in the smooth muscle cells of the wall of the peripheral airways. Figure 5 and Video E2 show that PI320 inhibits the MCh-induced Ca²⁺ oscillations in ASM cells and that this inhibition is reversible after PI320 washout. These results suggest that PI320-induced airway relaxation is mediated, at least in part, by the inhibition of intracellular Ca²⁺ mobilization and the frequency of Ca²⁺ oscillations.

Figure 5.

PI320 inhibits MCh-induced Ca²⁺ oscillations in the smooth muscle cells of peripheral airways in PCLSs. (A) Fluorescence image of part of a peripheral airway in a lung slice showing airway smooth muscle cells (ASMCs) and airway epithelial cells (AEPCs) lining the airway lumen. (B) Representative trace showing the changes in intracellular Ca²⁺ concentration of a single ASMC over time showing the Ca²⁺ oscillations induced by 0.3 μM MCh, inhibited by the addition of 30 μM PI320, and recovered after PI320 withdrawal in the continuous presence of MCh. (C) Summary of the frequency of MCh-induced Ca²⁺ oscillations before, during, and after PI320 addition. The Ca²⁺ oscillation frequency was calculated by counting the fluorescence peaks of individual ASMCs during 2 minutes before PI320 addition, 3 minutes after PI320 addition, and 3 minutes after PI320 withdrawal in experiments similar to that shown in B. Bars represent the means ± SEM of n = 15 ASMCs from two mice. *P < 0.05 and **P < 0.01, one-way ANOVA with repeated measurements and Tukey’s post hoc test. A video of the data presented in A and B, showing the Ca²⁺ oscillations induced by MCh and inhibited by the addition of PI320 is included as Video E2.

PI320 Inhibits IP₃-induced Peripheral Airway Constriction

Because IP₃Rs and ryanodine receptors are major intracellular Ca²⁺ release channels mediating intracellular Ca²⁺ mobilization in ASM cells, we tested whether PI320 also inhibits airway constriction after direct activation of these channels by uncaging intracellular IP₃ or by caffeine, respectively. First, we loaded PCLSs with membrane-permeable caged-IP₃ and used a video-rate phase-contrast microscope equipped with a shutter-controlled UV-light illuminator to uncage IP₃ in the ASM cells of peripheral airways using photolysis while assessing the changes in airway lumen area. Figures 6A–6E show that PI320 as well as 2-APB (an IP₃R blocker) significantly inhibit the airway constriction induced by IP3 uncaging with the UV light flash, with respect to that in presence of vehicle. In contrast, PI320 does not inhibit caffeine-induced airway constriction (Figures 6F–6H). Altogether, these results suggest that PI320-induced airway relaxation is mediated by the inhibition of Ca²⁺ release through the IP₃R and the inhibition of intracellular Ca²⁺ mobilization in ASM cells.

Figure 6.

PI320 inhibits inositol triphosphate (IP₃)-induced peripheral airway constriction in PCLS. (A) Phase-contrast microscope image of a peripheral airway in a lung slice and a broken line enclosing the area illuminated with the ultraviolet (UV) light to uncage the intracellular IP₃. (B–D) Representative traces showing the changes in airway luminal area in response to 20 mM caffeine and to IP₃ uncaging with a 900 millisecond UV light flash in the presence of 0.1% DMSO (vehicle, B), 50 μM 2-APB (IP₃R [IP3 receptor] antagonist, C), or 30 μM PI320 (D), suprafused at the times indicated on the top of the graphs. (E) Summary of the peak airway constrictions induced by IP₃ uncaging, as a percentage of the peak constriction induced by the previous stimulation with caffeine, in experiments similar to those shown in B–D. Bars represent the means ± SEM of n = 7 slices from two mice. *P < 0.05, one-way ANOVA with repeated measurements and Tukey’s post hoc test. (F and G) Representative traces showing caffeine-induced airway constriction in presence and absence of 0.1% DMSO or 30 μM PI320. (H) Summary of the effect of DMSO and PI320 on caffeine-induced airway constriction. Bars represent the means ± SEM of n = 7 slices from two mice; ns, unpaired t test. A video showing the transient airway constriction in response to caffeine and the UV flash, of the experiment shown in B, is included as Video E3.

Discussion

Previous reports have demonstrated that several GABA_AR activators, including benzodiazepines, directly relax ASM and alleviate airway hyperresponsiveness (3, 5, 7, 8, 16). PI320 is a novel imidazobenzodiazepine and positive allosteric modulator of the GABA_AR developed from the parent compound MIDD0301 (an imidazobenzodiazepine with limited blood–brain barrier penetration and hence no apparent CNS effects [3, 8]) by introducing a polyethylene glycol amide function. This modification improves the compound’s duration of action, aqueous solubility, and lung absorbance after inhalation (17). Given these properties, PI320 is a particularly promising candidate for an asthma therapeutic. In this report, we present data demonstrating that nebulized PI320 ameliorates airway responsiveness in vivo in mice as assessed by the forced oscillation technique (FlexiVent; Figure 1). Similarly, we demonstrate that PI320 dose-dependently relaxes mouse peripheral airways preconstricted with MCh in PCLS experiments (Figure 2).

Airway constriction is induced by a number of physiologic mediators (e.g., acetylcholine, histamine, leukotrienes, neuropeptides), which activate GPCRs (G protein–coupled receptors) on the surface of ASM cells. Activation of these receptors leads to the dissociation of associated G_q protein α subunits and the activation of PLC-β (phospholipase C-β). PLC-β subsequently hydrolyses phosphatidylinositol 4,5-bisphosphate in the inner layer of the plasma membrane to IP₃ and diacylglycerol. IP₃ is then free to diffuse in the cytosol, where it activates the IP₃ receptor in the SR membrane (21). This leads to rapid oscillations (0.07–0.4 Hz) in cytosolic Ca²⁺ concentrations due to cyclic opening and closing of the IP₃R Ca²⁺ pore in the setting of sustained IP₃ concentrations and ongoing SERCA (sarco/endoplasmic reticulum Ca²⁺-ATPase) activity (22). These oscillations are necessary for ASM contraction under physiologic conditions, with oscillation frequency relating directly to contractility (18, 23). To maintain these oscillations, and hence maintain airway constriction, Ca²⁺ must also influx from the extracellular space (largely via the STIM1/Oria1 complex in a process known as store-operated Ca²⁺ entry) (24).

We have previously hypothesized that the mechanism of GABA_AR ligand-induced upper airway relaxation entails an alteration in ASM cell membrane potential resulting from GABA_AR-mediated Cl⁻ flux (5, 7, 8). We postulated that this alteration then limits extracellular Ca²⁺ influx via membrane potential–sensitive pathways such as voltage-gated Ca²⁺ channels and/or store-operated Ca²⁺ entry (which is influenced by membrane potential [25, 26]), thereby decreasing intracellular Ca²⁺ concentration and limiting contraction. In fact, we recently published a report demonstrating that PI320-mediated relaxation of acetylcholine-constricted guinea pig tracheal rings (i.e., upper airways) was partially inhibited by the GABA_AR antagonist gabazine (17).

However, in the current report we present data from mouse peripheral (lower) airway experiments suggesting the presence of a mechanism of PI320-mediated relaxation in this tissue that is independent of the GABA_AR. Specifically, the classic GABA_AR antagonist picrotoxin and the benzodiazepine antagonist flumazenil do not inhibit PI320-mediated relaxation of MCh-constricted murine peripheral airways (Figure 3). Furthermore, the mechanism of relaxation does not appear to be dependent, at least not wholly, on inhibition of extracellular Ca²⁺ influx, as PI320 inhibited MCh-induced airway constriction even in the absence of external Ca²⁺ (Figure 4).

The complexity of these findings suggests that multiple mechanisms of benzodiazepine-mediated airway relaxation may exist and that the manifestation of each is dependent on experimental conditions. For example, the GABA_AR dependence of PI320-mediated airway relaxation may differ between species. As stated above, we previously demonstrated GABA_AR-dependent PI320-mediated relaxation of guinea pig tracheal rings (17), but we did not detect GABA_AR dependence in mice peripheral airways in the present study. Notably, we previously demonstrated that GABA_ARs are indeed expressed in the ASM of murine tracheal and intrapulmonary ASM (27) and showed that tracheal rings from transgenic mice lacking expression of the GABA_AR α4 subunit demonstrate diminished relaxation in response to benzodiazepines compared with wild-type mice (7), suggesting GABA_AR dependence in this setting. This raises the possibility that differing mechanisms of PI320-mediated relaxation are relevant to upper (tracheal) and lower (peripheral) ASM, with a GABA_AR-dependent mechanism predominantly applying to the ASM of upper airways (4, 5, 7, 28–30). This hypothesis is consistent with previous studies suggesting that membrane potential may only play an important role in the contractile state of upper airway smooth muscle, as voltage-gated Ca²⁺ channel blockers strongly relax smooth muscle from upper airways (e.g., tracheal ASM) in organ bath experiments but do not significantly relax peripheral airways in PCLS experiments at clinically relevant concentrations (24, 31, 32).

The current study also demonstrates that PI320 decreases MCh-induced Ca²⁺ oscillation frequency in peripheral ASM cells (Figure 5). Furthermore, PI320 inhibits the constriction of PCLS peripheral airways induced by experimentally increasing intracellular IP₃ using caged-IP₃ (Figure 6), suggesting its mechanistic site of action is downstream of IP₃ generation after the activation of G_q-coupled GPCRs by procontractile mediators such as MCh. PI320 did not inhibit caffeine-mediated airway constriction, suggesting it has no effect on Ca²⁺ mobilization from the SR mediated by ryanodine receptors. Taken together, these data suggest that PI320 is inhibiting Ca²⁺ mobilization from the SR, likely by inhibiting IP₃R function.

We cannot say at present whether the proposed PI320-induced inhibition of IP₃R function is the result of a direct or indirect PI320 effect. The IP₃R function is regulated by several ligands, interacting proteins, and intracellular signals, including second messengers such as cAMP (33). It is interesting to note that β-agonists, mainstays of asthma therapy, reduce Ca²⁺ oscillation frequency (and hence relax ASM) by inhibiting the IP₃R, at least in part. This occurs via activation of receptor-associated G_Sα subunits, the generation of cAMP, and likely PKA (protein kinase A)-mediated phosphorylation of the IP₃R (although the nature of PKA-mediated IP₃R regulation seems dependent on cell type and experimental preparation) (22, 34–36). Moreover, a recent study has demonstrated that certain benzodiazepines activate the proton-sensing OGR1 or GPR68 (ovarian cancer G protein–coupled receptor) on the surface of ASM cells. This GPCR has promiscuous signaling properties, with certain benzodiazepines biasing signaling via select canonical G protein pathways. The benzodiazepine sulazepam, for example, selectively activates G_S signaling (16). It is possible that PI320 is eliciting similar signaling events via OGR1 or another G_S-coupled receptor. However, in preliminary studies not presented here, PI320 did not increase intracellular cAMP concentrations in cultured human tracheal ASM cells or fresh tracheal ASM tissue. The mechanism by which PI320 inhibits SR Ca²⁺ mobilization will be the focus of future studies. Interestingly, previous studies have demonstrated that PI320 is highly permeable to the cell membranes (17), affording the possibility of direct IP₃R inhibition.

In addition to ASM, GABA_AR is expressed on airway epithelium. The effect of PI320 on epithelial function was not formally evaluated in the present study. However, no gross change in ciliary beating was noted, and a previous study of MIDD0301, the parent compound of PI320, demonstrated that it had no effect on epithelial mucous metaplasia in a murine asthma model (3). This is of interest, as epithelial GABA_AR function has been linked to the regulation of mucous production (11). Given the epithelium is intact in PCLS experiments, it is possible that the relaxant effect of PI320 is mediated by epithelial cells. However, if this is indeed the case, the data presented in the current study suggest this mechanism would be GABA_AR-independent. GABA_B receptors, which are G protein–coupled receptors, are also expressed on ASM (37) and epithelial cells (38). However, benzodiazepines are not known to activate these receptors.

In summary, we demonstrate that PI320, a novel imidazobenzodiazepine and asthma therapeutic candidate, relaxes mouse airways in vivo and ex vivo in clinically relevant experimental paradigms. We further show that murine peripheral airway relaxation is independent of the GABA_AR and alterations in ASM cell extracellular Ca²⁺ influx, in apparent contrast to previous studies in upper airway samples from guinea pigs. However, PI320 inhibits MCh-induced Ca²⁺ oscillations in ASM cells, a process dependent on SR Ca²⁺ release via IP₃Rs, and it inhibits the constriction of peripheral airways stimulated by experimentally increasing the intracellular concentration of IP₃. These data suggest that PI320 relaxes ASM in murine peripheral airways by inhibiting Ca²⁺ mobilization from the SR, likely by direct or indirect IP₃R inhibition. This represents a novel mechanism of PI320-mediated ASM relaxation and suggests multiple context-dependent mechanisms exist.