Comments
Abstract
In most living cells, the second-messenger roles for adenosine 3′,5′-cyclic monophosphate (cAMP) are short-lived, confined to the intracellular space, and tightly controlled by the binary switch–like actions of Gαs (stimulatory G protein)–activated adenylyl cyclase (cAMP production) and cAMP-specific PDE (cAMP breakdown). Here, by using human airway smooth muscle (HASM) cells in culture as a model, we report that activation of the cell-surface β2AR (β2-adrenoceptor), a Gs-coupled GPCR (G protein–coupled receptor), evokes cAMP egress to the extracellular space. Increased extracellular cAMP levels ([cAMP]e) are long-lived in culture and are induced by receptor-dependent and receptor-independent mechanisms in such a way as to define a universal response class of increased intracellular cAMP levels ([cAMP]i). We find that HASM cells express multiple ATP-binding cassette (ABC) membrane transporters, with ABCC1 (ABC subfamily member C 1) being the most highly enriched transcript mapped to MRPs (multidrug resistance–associated proteins). We show that pharmacological inhibition or downregulation of ABCC1 with siRNA markedly reduces β2AR-evoked cAMP release from HASM cells. Furthermore, inhibition of ABCC1 activity or expression decreases basal tone and increases β-agonist–induced HASM cellular relaxation. These findings identify a previously unrecognized role for ABCC1 in the homeostatic regulation of [cAMP]i in HASM that may be conserved traits of the Gs-GPCRs (Gs-coupled family of GPCRs). Hence, the general features of this activation mechanism may uncover new disease-modifying targets in the treatment of airflow obstruction in asthma. Surprisingly, we find that serum cAMP levels are elevated in a small cohort of patients with asthma as compared with control subjects, which warrants further investigation.
We report that activation of β2ARs (β2-adrenoceptors), common druggable targets for asthmatic bronchospasm, evokes adenosine 3′,5′-cyclic monophosphate (cAMP) release from human airway smooth muscle cells and show that cAMP egress is mediated by ABCC1 (ATP-binding cassette subfamily member C 1) belonging to MRPs (multidrug resistance–associated proteins), implicating new disease-modifying targets in the treatment of airflow obstruction in asthma.
As a cornerstone in the management of asthma or chronic obstructive pulmonary disease, bronchodilators such as β-agonists reverse or prevent the airflow obstruction. In the constricted airways, the principal action of β-agonists is to stimulate β2ARs (β2-adrenoceptors), a class of 7-transmembrane–spanning cell-surface GPCRs (G protein–coupled receptors) expressed on the smooth muscle of human bronchi—the pivotal tissue regulating bronchomotor tone (1–3). Upon activation, β2ARs signal to the Gαs (stimulatory G protein) to activate adenylyl cyclase, which generates a key second messenger, adenosine 3′,5′-cyclic monophosphate (cAMP) (4). Increased intracellular cAMP levels ([cAMP]i) consequently activate PKA (protein kinase A), which in turn modulates multiple downstream targets to promote human airway smooth muscle (HASM) relaxation and reverse airways obstruction (4–6).
Classically, the signal transduction evoked by the Gs-GPCRs (Gs-coupled family of GPCRs) is short-lived, and multiple mechanisms are in place to ensure homeostatic regulation of [cAMP]i. These include PKA-mediated receptor uncoupling, as well as homologous or heterologous desensitization by β-arrestin, inhibition of adenylyl cyclase by the Gαi (inhibitory G protein), and termination of cAMP signaling by PDE (7, 8). In HASM, the enzymatic conversion of cAMP to its inactive adenosine 5′-monophosphate (AMP) state is fast and predominated by PDE4 (PDE isoform 4), which functions to switch off Gαs-activated increased [cAMP]i (8). Accordingly, significant research and drug discovery efforts are focused on the elements of these molecular pathways to not only improve the functional efficacy of β-agonists but also develop new signal splitting mechanisms that are biased toward Gαs and away from β-arrestin signaling in HASM for the treatment of obstructive lung diseases.
Apparently, the binary switch–like actions of Gαs-activated adenylyl cyclase and cAMP-specific PDE are prone to breakdown, perturbing cAMP homeostasis and dysregulating excitation–contraction coupling in HASM shortening (9, 10). Furthermore, Gαi and PDE4 expression and activity are upregulated by cytokines and mediators implicated in asthma, diminishing β-agonist–induced [cAMP]i and HASM relaxation (11, 12). Of note, genetic variants of PDE4D (13, 14) and ADRB2 (15, 16) in patients with asthma and chronic obstructive pulmonary disease have been associated with suboptimal control and/or interindividual differences in the response to β-agonist treatment (17). Interestingly, there is an emergence of pharmacogenetic studies showing an association of the variability in treatment response to asthma therapy with single nucleotide polymorphisms in the ABCC1 (ATP-binding cassette [ABC] subfamily member C 1) gene, which encodes MRP1 (multidrug resistance–associated protein 1) (18–20).
ABCC1/MRP1 (herein referred to as ABCC1), a member of the superfamily of ABC membrane transporters, is ubiquitously expressed and functions to displace out of the cells a wide range of structurally diverse and physiologically important endogenous substrates (21–23). In human lungs, ABCC1 and ABCC4/MRP4 are expressed on the epithelium lining the proximal airways (24–26) and are reported to pump inflammatory mediators, lipid metabolites, and signaling molecules out of the cells (27–32). The molecular signature and biological function of MRPs in HASM remain unclear, however. Here, we find multiple ABCC transcripts to be expressed in HASM cells, with the ABCC1 subtype being the most highly expressed in results from RNA-sequencing (RNA-seq), quantitative PCR (qPCR), and proteomic analyses. We show that β-agonists evoke cAMP egress to the extracellular space and that inhibition of ABCC1 activity and/or expression decreases β2AR-evoked cAMP efflux, resulting in increased [cAMP]i and HASM cell relaxation. This previously unrecognized mechanism of β-agonist–mediated cAMP homeostasis in HASM may provide new therapeutic strategies to promote the functional efficacy of β2ARs and/or mitigate the mechanical endotype of airflow obstruction in asthma.
Unless otherwise stated, all chemicals and drugs were purchased from Sigma-Aldrich, reconstituted in sterile distilled water or DMSO, frozen in aliquots, and serially diluted in the appropriate buffer or media on the day of use. Ham’s F-12 medium, PBS, FBS, 0.05% trypsin and EDTA, and PAGE/Western blotting systems were purchased from Life Technologies. All antibodies were purchased from Cell Signaling Technology. MK-571 was purchased from Abcam. The cAMP-Screen System ELISA was purchased from Applied Biosystems. The synthetic arginine–glycine–aspartic acid–containing peptide was purchased from American Peptide Co.
Human lungs were received from the National Disease Research Interchange and from the International Institute for the Advancement of Medicine. All tissues are obtained from de‐identified donors, and their use does not constitute human subject research as described by the Rutgers Institutional Review Board. Primary HASM cells were derived from the tracheas and were harvested, characterized, and grown in culture as described by us in detail elsewhere (33). In brief, cells were cultured in Ham’s F‐12 medium supplemented with 100 U ⋅ ml−1 penicillin, 0.1 mg ⋅ ml−1 streptomycin, 2.5 mg ⋅ ml−1 amphotericin B, and 10% FBS. Medium was replaced every 48 hours. For all experiments, HASM cells were serum deprived for 24 hours and used within the first four passages to ensure the proper smooth muscle phenotype (2, 3, 33).
siRNAs targeting ABCC1 (M-007308-01-0005) and ABCC4 (M-007313-01-0005) were purchased from Dharmacon. siRNA-mediated knockdown was performed by using a reverse-transfection procedure. Briefly, HASM cells were trypsinized, resuspended in plain media, and incubated with Hi-Perfect Transfection Reagent (catalog number 301705, Qiagen) and siRNAs before plating for a final siRNA concentration of 10 μM. A cell suspension was added to the siRNA mixture and incubated for 30 minutes. The cell suspension and the siRNA mixture were then seeded into cell culture plates according to the experimental design and were incubated for 5 hours. After 5 hours, complete cell culture media (described above) were added to the cell culture plate wells in a 1:1 ratio and were incubated for 18 hours. After 18 hours, media were changed to complete media. Cells were serum deprived for 24 hours before collection. Cells were collected 72 hours after transfection.
RNA was collected and converted to cDNA by using the Superscript IV CellsDirect cDNA Synthesis Kit (Invitrogen) in accordance with the manufacturer protocol. A no-enzyme control was included to ensure fidelity of amplification and was made in parallel with other samples (reverse-transcriptase was replaced with nuclease free water). Real-time PCR was subsequently conducted by using predesigned TaqMan probes for GABARAP (assay identifier: Hs.PT.58.2542712, lot number 291912106), ABCC1 (assay identifier: Hs.PT.58.26515729, lot number 280747473) and ABCC4 (assay identifier: Hs.PT.58.26860542, lot number 280747481) from Integrated DNA Technologies by using a two-step, real-time PCR system (Applied Biosystems).
For each donor sample (n = 6 donors), 1 mg of lysate from donor lung–derived HASM cells was digested with trypsin by following a standard filter-aided sample preparation protocol (34). Fractions of the digests were labeled with TMT10 Plex reagent (Thermo Scientific) by following the manufacturer’s instructions and were combined and dried. The peptides in combined samples were desalted and fractionated by using high-pH reverse-phase HPLC with a C18 Xbridge column (Waters) connected to an Agilent HP1100 system. For the total proteome analysis, 14 fractions (between 27 and 40 min) were chosen, desalted with a stage tip, and analyzed by using nanoscale liquid chromatography coupled to tandem mass spectrometry (Q Exactive HF, Thermo Scientific). Mass spectrometry data were analyzed by using MaxQuant version 1.6.2.6 (MaxPlanck Institute of Biochemistry) and the Andromeda search engine (35). The UniProt database was used to identify proteins and potential contaminants. Proteins with a false discovery rate <1% are reported.
HASM cells were seeded in 4-well chamber slides and cultured for 3–5 days until 60% confluence. Adherent cells were serum deprived overnight and then incubated with 1 μg/ml wheat germ agglutinin–Alexa 594 (Invitrogen) for 10 minutes at 37°C. Cells were washed two times with PBS and fixed with 1% paraformaldehyde in PBS for 10 minutes. Fixed, nonpermeabilized cells were washed five times with PBS for 5 minutes, incubated with 1% BSA/PBS–containing FcR blocking reagent (Order number 130-059-901, MACS Mitenyi Biotec) for 30 minutes at room temperature, and incubated with either 1 μg/ml rabbit anti–human ABCC1 antibody (catalog number 72202, Cell Signaling Technology) or rabbit anti-IgG isotype control (catalog number 3900, Cell Signaling Technology) for 2 hours at room temperature. Subsequently, cells were washed three times with PBS and incubated with 1:500 donkey anti–rabbit IgG–Alexa 488 (code 711-545-152, Jackson ImmunoResearch Labs) for 50 minutes at room temperature. Cells were then permeabilized with 0.01% Triton-X100 in PBS for 5 minutes and incubated with DAPI for 5 minutes. Slides were washed, mounted with coverslips, and imaged under a fluorescent microscope (Nikon-80i, 40× objective lens).
HASM cells were seeded in a 24‐well plate with F-12 medium containing 10% FBS until ∼80–90% confluent and were then serum deprived overnight. Cells were washed with fresh serum-free media before agonist treatments. After the indicated times, the media (infranatant) in each sample well were collected, and the cells were treated with lysis buffer from the cAMP‐Screen System ELISA kit (catalog number 4412182, Applied Biosystems) for 30 minutes at 37°C. For each sample well, 150 μl of the infranatant and 150 μl of the cell lysate were applied for the cAMP ELISA assay to quantitate the extracellular cAMP levels ([cAMP]e) and [cAMP]i, respectively. Experiments were conducted according to the manufacturer’s protocol. Unless otherwise noted, assays were performed in duplicate for each sample.
For measurements of cAMP in patient samples, study participants were recruited from the Yale Center for Asthma and Airways Disease at Yale New Haven Hospital in New Haven, Connecticut, by using an institutional review board–approved protocol. The inclusion and exclusion criteria and the Yale Center for Asthma and Airways Disease phenotyping protocol have been previously described (36). Patients underwent asthma phenotyping that included a medical history, lung function testing, and a blood draw. All participants provided informed consent.
We used magnetic twisting cytometry (MTC) to measure dynamic changes in the cytoskeletal stiffness as a surrogate for agonist-induced, single-cell contractility, as we have validated previously (2, 3). Briefly, an arginine–glycine–aspartic acid–coated ferrimagnetic microbead (4.5 μm in diameter) functionalized to the cytoskeleton through cell-surface integrin receptors was magnetized and twisted by an external magnetic field that varied sinusoidally in time, and forced bead motions were detected with a spatial resolution of ∼5 nm (2, 3). Unless otherwise stated, the ratio of specific torque to lateral bead displacements was computed and expressed as the cell stiffness in units of Pa/nm.
Unless otherwise stated, HASM cells from at least five donors were used. When applicable, the experimental readouts were first normalized to the vehicle control in each donor to obtain the fold change. The fold changes from individual donors were used to obtain group mean graphs. The data are expressed as the mean ± SE. For these studies, GraphPad Prism 6.0 was used for statistical analysis consisting of one-way ANOVA with a Dunnett multiple-comparison test, and the means were considered significantly different when P values were <0.05. For MTC studies on HASM cells derived from multiple donors, we used nested-design analysis to control for random effects from repeated measurements of multiple cells in the same subject (3). To satisfy the normal distribution assumptions associated with the ANOVA, stiffness data were converted to a log scale before analyses. The Tukey-Kramer method was applied for multiple-comparison adjustment by using SAS version 9.2 (SAS Institute Inc.), and two-sided P values <0.05 were considered to indicate significance.
To define the gene expression signature of MRPs in HASM, we surveyed our RNA-seq data from a previously published study that is available in the Gene Expression Omnibus under accession number GSE94335 (37). In this data set that consisted of primary HASM cells derived from 17 White, nonsmoking lung donors (9 who died of fatal asthma and 8 with no chronic illness or medication use; see Table E1 in the data supplement), we detected transcripts for 12 of the 14 known ABCC genes, based on having nonzero normalized read counts. As shown in Figure 1A, levels for the 12 expressed ABCC genes varied in HASM cells, but none differed by asthma status. ABCC1 was the most highly expressed ABCC transcript, with mean normalized counts of 8,102 in HASM cells derived from donors with asthma and 8,136 in HASM cells derived from donors without asthma (Figure 1A) being demonstrated. In both groups, the next most expressed ABCC was ABCC4, with normalized counts of 2,785 and 2,723 in HASM cells derived from donors with and without asthma being demonstrated, respectively, representing ABCC4 levels that are approximately a third of those of ABCC1 (Figure 1A). By using qPCR, we confirmed the relative mRNA expression levels of ABCC1 and ABCC4 in HASM cells derived from eight additional lung donors with and without asthma (Figure 1B and Table E2). In both groups, mRNA expression levels of ABCC1 were approximately fivefold higher than those of ABCC4, and there were no differences by asthma status (Figure 1B).
Figure 1. Human airway smooth muscle (HASM) cells express ABCC (ATP-binding cassette [ABC] subfamily member C) transporter. (A) Expression of ABCC genes in HASM cells as assessed in a publicly available RNA-sequencing data set (GSE94335) (37). RNA-sequencing results are expressed as normalized read counts (read counts were obtained by using HTSeq and normalized by using DESeq2). (B) RT-PCR cross-validation of the two most abundant ABCC genes, ABCC1 and ABCC4, in HASM cells from donors with and without asthma. The characteristics of the lung donors (n = 4 with asthma and n = 4 without asthma) are provided in Table E2 in the data supplement. Results are normalized to endogenous GABARAP to account for template quantity discrepancies between samples and are expressed as relative fold changes to ABCC1 expression levels detected in HASM cells derived from one donor without asthma. Analyses were done by using one-way ANOVA, followed by a Tukey multiple-comparison test. ****P < 0.0001. (C) Tryptic digests from cell lysates were screened by using nanoscale liquid chromatography coupled to tandem mass spectrometry, and the reporter-ion intensities of ABCC isoforms were expressed as the relative percentage of ABCC1 (set at 100%) for each HASM sample derived from a donor without asthma (n = 6). Analyses were done by using one-way ANOVA, followed by a Tukey multiple-comparison test. ***P < 0.001 and ****P < 0.0001. (D) A representative (merged) immunofluorescent image of HASM cells probed with wheat germ agglutinin (red) and anti-ABCC1 antibody (green). Nuclei are identified by the blue color from DAPI fluorescence. The image was acquired by using a Nikon-80i microscope under a 40× objective lens. Individual fluorescent signals, including IgG isotype control for ABCC1 antibody, are shown in Figure E1. Scale bars, 10 µm. Ct = cycle threshold; ns = not significant.
[More] [Minimize]Concordant with a rank order based on the ABCC transcripts detected by using RNA-seq and qPCR, nanoscale liquid chromatography coupled to tandem mass spectrometry enabled detection of a high abundance of ABCC1 protein levels in the tryptic digests of HASM cells (Figure 1C). In this data set that consisted of HASM cells derived from six donors without asthma (Table E3), protein levels of ABCC4, ABCC3, and ABCC10 were ∼33%, 14%, and 2% of the ABCC1 levels; other ABCC members were not detected. Immunohistochemistry studies on nonpermeabilized HASM cells showed a strong fluorescent signal (Alexa 488) for ABCC1 that clustered and localized with wheat germ agglutinin (wheat germ agglutinin–Alexa 594), which stains glycoproteins and glycolipids at the cell membrane (Figure 1D); a weak, diffuse fluorescence (Alexa 488) was detected with IgG isotype control (Figure E1). In contrast, we were unable to verify the cell-surface expression of ABCC4 by using commercially purchased antibodies against ABCC4. These results established that ABCC1, a member of the ABC transporter superfamily, is highly expressed in HASM cells.
In a number of cellular models, ABCC1 and ABCC4 show substrate specificities to a broad array of signaling molecules, including cAMP (21–23, 27–30). These studies prompted us to ask 1) whether HASM cells are able to pump out intracellular cAMP and 2) whether these events are mediated by ABCC1/ABCC4. In initial studies, we stimulated HASM cells with the long-acting β2-agonist formoterol (half-life, ∼8–10 h) and measured [cAMP]i/[cAMP]e by using a highly sensitive ELISA method. For these studies, we used HASM cells derived from donors without asthma and studied them under identical experimental conditions, without PDE inhibitors, in standard culture media for 24 hours. As expected, formoterol rapidly increased [cAMP]i in HASM cells (from basal levels of 6 ± 2 fmol to 432 ± 46 fmol within 5 min) (Figure 2A). Formoterol-evoked [cAMP]i were transient, however, and declined to lower but suprabasal levels within 30 minutes (81 ± 9 fmol); the suprabasal [cAMP]i were maintained over 4–24 hours (in keeping with a typical half-life of formoterol). Furthermore, forskolin that directly activates adenylyl cyclase also evoked a similar time course of intracellular cAMP generation in HASM cells (Figure 2A).
Figure 2. Activation of the Gs-GPCRs (Gs-coupled family of GPCRs [G protein–coupled receptors]) evokes adenosine 3′,5′-cyclic monophosphate (cAMP) efflux. (A and B) Representative intracellular cAMP levels ([cAMP]i) (A) and extracellular cAMP levels ([cAMP]e) (B) detected by using an ELISA. HASM cells were stimulated with formoterol (1µM) or forskolin (10 µM) for 24 hours. For each time point, assays were performed in duplicate. Data are presented as the mean ± SE (data are stacked in relation to each agonist, showing the relative level changes with time). (C and D) Formoterol- and forskolin-induced [cAMP]e from HASM cells derived from five lung donors (4 h). The characteristics of the five lung donors are provided in Table E4. For each donor-derived HASM sample, agonist-induced [cAMP]e were superimposed in relation to basal [cAMP]e. Individual donor–derived cellular responses to isoproterenol (ISO) are shown in Figure E2A.
[More] [Minimize]Strikingly, exposure of HASM cells to formoterol or forskolin culminated in a slow accumulation of cAMP in the extracellular space (Figure 2B). Increased [cAMP]e were substantial at 15 minutes (from basal levels of 32 ± 6 fmol to 169 ± 30 fmol in response to formoterol; from 25 ± 6 fmol to 184 ± 35 in response to forskolin), equieffective to peak [cAMP]i within 30 minutes (498 ± 9 fmol for formoterol; 414 ± 109 fmol for forskolin), and reached a peak by 4–24 hours. Of note, HASM cells showed expected between-donor variations in basal [cAMP]e, but all showed significant increased [cAMP]e in response to β-agonists (Figures 2C and E2A; Table E4). This long-lived increase in [cAMP]e was also induced by forskolin, which activates adenylyl cyclase, and was independent of β2AR activation (Figure 2D).
To test whether the observed β2AR-evoked cAMP efflux is a shared mechanism of the Gs-GPCRs, we next stimulated HASM cells with prostaglandin E2 (PGE2). PGE2 activates EP (PGE2 receptor) subtypes (EP1–4) that display multifunctional signaling. In a previous study, we have shown that HASM cells express a high abundance of EP2 and EP4 transcripts, the receptor subtypes that largely signal to the Gαs to activate adenylyl cyclase: PGE2 is a strong generator of [cAMP]i and a potent inducer of HASM relaxation (38). HASM cells stimulated for 4 hours with PGE2 showed a dose-dependent increase in [cAMP]e (Figure E2B). Of note, PGE2-evoked cAMP egress was appreciably greater than that of formoterol and that of the nonselective β-agonist isoproterenol (Figure E2B). Taken together, these results suggest that cAMP egress is a universal response class of increased [cAMP]i in HASM cells. The findings also support the presence of intracellular cAMP sensors or additional homeostatic regulation of [cAMP]i mediated by the Gs-GPCRs, involving the export of the second messenger from HASM cells.
To explore the roles for ABCC1/ABCC4 in β-agonist–induced cAMP egress, we employed a well-established MRP inhibitor, MK-571 (39, 40). MK-571 decreased isoproterenol-induced cAMP release from HASM cells in a dose-dependent manner (Figure E3). Compared with vehicle-treated cells, HASM cells treated with MK-571 (10 μM) produced markedly lower [cAMP]e in response to isoproterenol (Figures 3A and 3B; Table E5). As shown in Figure 3B, whereas the magnitude of decreases varied between donors for donor-derived HASM cells, MK-571 treatment effectively inhibited β-agonist–induced cAMP efflux.
Figure 3. Pharmacologic inhibition of ABCC1 decreases cAMP egress. (A) HASM cells were first treated for 20 minutes with or without 10 µM MK-571, followed by incubation with 10 µM ISO for an additional 4 hours. Data are presented as the mean ± SE (n = 5 independent samples). Analyses were done by using one-way ANOVA, followed by a Dunnett multiple-comparison test. *P < 0.05. (B) ISO-induced [cAMP]e were measured in HASM cells derived from seven lung donors (Table E5). For each donor-derived HASM sample, ISO-induced [cAMP]e in MK-571–treated cells were normalized to the respective vehicle-treated cells. Data are presented as the mean ± SE. The P value shown is from a paired t test. RFC = relative fold change.
[More] [Minimize]To determine the relative contribution of ABCC1 and ABCC4 in regulating cAMP efflux, we used siRNA-mediated knockdown of ABCC1 in HASM cells (Table E5). Here, we focused on ABCC1 because, unlike human airway epithelial cells (24–26), ABCC4 mRNA and protein levels were markedly lower than those of ABCC1 in HASM cells (Figure 1). Furthermore, consistent with a previous report (30), we were unable to verify the cell-surface expression of ABCC4 in HASM cells. Transfection of HASM cells with siRNA directed against ABCC1 decreased ABCC1 protein levels compared with the respective nontargeting siRNA controls; Western blots showed variable reductions of ABCC1 protein levels in HASM cells derived from seven donors (Figure 4A). Compared with nontargeting siRNA (control)–transfected cells, however, HASM cells transfected with ABCC1 siRNAs showed an ∼40% reduction in cAMP efflux in response to isoproterenol (Figures 4B, 4C, and E4); there were no group differences in basal [cAMP]e. These results suggest that β-agonist–induced accumulation of cAMP in the extracellular space is predominantly mediated by ABCC1 expressed on HASM cells.
Figure 4. siRNA-mediated knockdown of ABCC1 decreases β2AR (β2-adrenoceptor)–evoked cAMP release from HASM cells. HASM cells were transfected with siRNAs directed against ABCC1 and nontargeting (NT) siRNAs. (A) Expression of ABCC1 was detected by using a Western blot (HASM cells from n = 7 donors; Table E5). Tubulin was used as a loading control. (B) [cAMP]e were detected by using an ELISA before and after stimulation with 10 µM ISO (4 h). Data are presented as box-and-whisker plots. Analyses were done by using one-way ANOVA, followed by Tukey multiple-comparison tests. *P < 0.05 and **P < 0.005. (C) Individual donor–derived HASM cellular responses to ISO. Colors indicate HASM cells derived from different donors. Circles indicate cells transfected with NT siRNAs. Squares indicate cells transfected with ABCC1 siRNAs. Dotted lines indicate matching donor lungs. The P value shown is from a paired t test.
[More] [Minimize]Of note, in HASM cells derived from multiple donors, isoproterenol stimulation markedly increased [cAMP]i in MK-571–treated cells compared with untreated cells (Figure 5A). To address the physiological consequences of ABCC1 inhibition, MTC was applied to quantify the stiffness changes of HASM cells in response to isoproterenol (3). For these studies, we used HASM cells derived from three different lung donors (Table E6) and applied mixed-effect models to control for the random effects because of multiple cell measurements from the same donor, as we have done previously (3). Compared with untreated cells, HASM cells treated with MK-571 exhibited significant decreases in baseline stiffness (basal tone) and, moreover, showed further stiffness decreases in response to isoproterenol (Figures 5B and E5A). Interestingly, levels of stiffness after MK-571 treatment were comparable with those induced by isoproterenol in untreated cells; there were no significant differences between the two groups. These data suggest that endogenous activity of ABCC1 may regulate cAMP homeostasis in HASM cells.
Figure 5. Inhibition of ABCC1 activity or expression increases β-agonist–induced [cAMP]i and promotes HASM-cell relaxation. (A and B) Individual HASM cells were treated for 20 minutes with or without MK-571 and then stimulated with the β-agonist ISO for 5 minutes. (A) [cAMP]i were detected by using an ELISA (HASM cells from n = 7 donors). For each donor-derived HASM sample, ISO-induced [cAMP]i in MK-571–treated cells were normalized to the respective untreated cells. The P value indicates the result from a paired t test. (B) Cell stiffness measured by using magnetic twisting cytometry. HASM cells were derived from three different lung donors. Data are presented as the geometric mean ± 95% confidence interval (n = 351–486 individual cell measurements from HASM cells derived from each donor for each condition). A mixed-effect model was applied to test the group differences in stiffness, with multiple comparisons being adjusted by using Tukey-Kramer method. *P < 0.05 and ***P < 0.001. (C) ISO-induced [cAMP]i in HASM cells transfected with ABCC1 siRNAs versus NT siRNAs (HASM cells from n = 7 donors). The P value indicates the result from a paired t test. (D) Magnetic twisting cytometry was applied in HASM cells derived from one donor lung that were transfected with ABCC1, ABCC4, and NT siRNAs. Data are presented as the geometric mean ± 95% confidence interval (n = 465–612 individual cell measurements from three independent experiments with multiple-cell wells transfected with siRNAs). Analyses were done by using one-way ANOVA, followed by Tukey multiple-comparison tests. ****P < 0.0001. To satisfy the normal distribution assumptions associated with the ANOVA, stiffness data were converted to a log scale before analyses. Pa = pascal.
[More] [Minimize]Similar to HASM cells that underwent pharmacological inhibition of ABCC1 with MK-571, HASM cells transfected with ABCC1 siRNAs showed appreciable increases in peak [cAMP]i in response to isoproterenol (Figures 5C and E4C). Of note, ABCC1 siRNA–transfected cells exhibited markedly lower baseline stiffness (basal tone) than cells transfected with nontargeting siRNAs (Figures 5D and E5B). As a negative control, we also transfected HASM cells with siRNA targeting ABCC4. There were no significant differences in baseline stiffness or isoproterenol-induced stiffness decreases in HASM cells transfected with ABCC4 siRNA versus nontargeting siRNA (Figure 5D). Compared with nontargeting or ABCC4 siRNA–transfected cells, however, HASM cells transfected with ABCC1 siRNA showed markedly lower basal tone and isoproterenol-induced cellular relaxation (Figures 5D and E5B). These findings suggest that ABCC1 directly regulates [cAMP]i in HASM and that this previously unrecognized mechanism of cAMP homeostasis may provide new therapeutic strategies to promote the functional efficacy of β-agonists and/or inhibit excitation–contraction coupling in HASM shortening.
Patients with asthma express increased circulating levels of chitinase and the chitinase-like protein YKL-40, effector molecules implicated in immune inflammatory responses of the airways (41, 42). Furthermore, it has been shown that serum YKL-40 levels are positively correlated with airway remodeling, airway hyperresponsiveness, and asthma severity (36, 42). As a proof of concept that [cAMP]e may provide new markers for asthma, we measured the levels of circulating cAMP in a small cohort of patients with (n = 16) and without (n = 8) asthma (as defined by Global Initiative for Asthma criteria). The demographics of the patients with and without asthma are shown in Table E7. As shown in Figure 6, we were able to detect cAMP in patient samples and found increased serum cAMP levels in patients with asthma compared with control subjects. Further investigations are warranted, however, to establish its relationship to asthma severity and determine the clinical utility of extracellular cAMP in health and diseases.
Figure 6. Serum cAMP levels are increased in patients with asthma. An ELISA was applied to detect cAMP levels in blood drawn from a small cohort of patients with and without asthma. The demographics of the 16 patients with asthma and 8 patients without asthma are shown in Table E7. The P value shown is from a Mann-Whitney test.
[More] [Minimize]In 1971, Earl Sutherland won a Nobel Prize in Physiology or Medicine for his work on the “mechanisms of the action of hormones,” which led to the identification of a “heat-stable factor,” cAMP, and established the founding principles on the second-messenger system (43, 44). The second-messenger role of cAMP is widespread and is one of the most studied intracellular signaling cascades of the Gs-GPCRs. Classically, the multifunction cAMP transduction evoked by Gs-GPCRs is short-lived, confined to the intracellular space, and tightly controlled by the binary switch–like actions of adenylyl cyclase (cAMP production) and PDE (cAMP breakdown), with many built-in nodal points for signal amplification, modification, and attenuation.
For example, in this signaling pathway, the production of cAMP from ATP by adenylyl cyclase is triggered or repressed by the activation GPCRs that coupled to Gαs or Gαi, respectively. Once synthesized, cAMP acts on its primary effector(s), including PKA and EPAC (exchange proteins activated by cAMP), which in turn modulate further downstream signal transduction. Signal attenuation occurs through enzymatic degradation of cAMP to AMP by PDE and compartmentalization, desensitization, and/or extrusion of cAMP to the extracellular space. The latter is cell-type dependent, however, and is mainly mediated by MRPs belonging to the ABC membrane transporters (45).
Here, we find that β-agonists, the most widely used bronchodilators for the treatment of asthmatic bronchospasm, evoke cAMP release from HASM cells. We show that cAMP is exported from HASM cells in response to agonists/agents that stimulate a different family/component of the Gs-coupled GPCRs (β2ARs, EPs, and adenylyl cyclase) and that cAMP is long-lived in culture. These general features of the receptor-dependent and receptor-independent mechanisms of cAMP egress suggest a universal response class of increased [cAMP]i mediated by the Gs-GPCRs expressed on HASM cells.
We sought to understand the mechanism(s) for this previously unrecognized β2AR-evoked cAMP secretion from HASM cells. Because ABCC4/MRP4, ABCC5/MRP5, and ABCC11/MRP8 have substrate specificities to cyclic nucleotides (46), we first screened for the expression of 14 known ABCC genes in HASM cells. In our own RNA-seq data set (GSE94335) (37), which profiled primary HASM cells derived from 17 White, nonsmoking lung donors, we detected transcripts of 12 of 14 ABCCs (based on having nonzero normalized read counts), including ABCC4 (encoding MRP4), ABCC5 (encoding MRP5), and ABCC11 (encoding MRP8). We did not detect protein expression of MRP5 and MRP8 in our proteomic data set, however. In addition, consistent with the findings of Huff and colleagues (30), we were unable to verify the cell-surface expression of MRP4 in isolated HASM cells. Strikingly, ABCC1 was highly expressed in both RNA-seq and proteomic data sets. Applying qPCR in HASM cells derived from eight additional donors, we confirmed the relative mRNA levels of ABCC1, which were approximately fivefold higher than those of ABCC4 (second highest in both RNA-seq and proteomics). Furthermore, by using commercially purchased antibodies against ABCC1, we detected cell-surface expression of ABCC1 in isolated HASM cells. These results established that HASM cells manifest restricted expression of MRPs and that ABCC1/MRP1 is highly expressed in HASM cells.
Of note, SNPs in the ABCC1 gene (rs119774) have been associated with variability in montelukast (leukotriene inhibitor) treatment responses in patients with asthma (18, 19). Furthermore, ABCC1 is reported to have substrate specificity to lipid metabolites, including leukotriene C4 and sphingosine-1-phosphate (21–23, 27, 28, 46). We pursued the role for ABCC1 in cAMP secretion from HASM cells. We showed that pharmacological inhibition or downregulation of ABCC1 with siRNA markedly decreased β2AR-evoked cAMP release from HASM cells derived from multiple donors. On the other hand, inhibition of ABCC1 activity and/or expression increased β-agonist–induced [cAMP]i and the associated HASM-cell relaxation. These findings identify a previously unrecognized role for ABCC1 in the homeostatic regulation of [cAMP]i in HASM that may be conserved across the Gs-GPCRs. Hence, the general features of this activation mechanism may uncover new disease-modifying targets in the treatment of airflow obstruction in asthma.
Characterized by airway inflammation and reversible airflow obstruction, asthma is a common debilitating lung disease that afflicts ∼4.5% of the world’s population. Despite advances in treatment, asthma remains poorly controlled in most patients. Recently, for patients with difficult-to-control and severe asthma, the use of biologics that target type 2 (T-helper cell type 2) inflammation has substantially impacted asthma management. Surprisingly, T-helper cell type 2–high (IL-4, IL-5, and IL-13) endotype–based approaches provide therapeutic improvements in only a minority of patients (47). Although multiple factors contribute to asthma pathobiology, the principal mechanism of increased airway resistance to airflow is aberrant shortening of airway smooth muscle (2, 3, 48, 49). The results presented in this study showed new molecular pathways invoked by β-agonists involving the extrusion of intracellular cAMP through ABCC1. Of note, [cAMP]e remained elevated in culture for 24 hours. This is despite finding appreciable transcript levels of ectonucleotidases (i.e., ENTPD1–8, NT5E, ENPP1–6, and ALPL) in our RNA-seq data set (37), suggesting limited negative feedback through adenosinergic signaling mechanisms.
What are the roles for extracellular cAMP in the lungs? Could the levels of cAMP in the lung and/or circulation provide other clinical determinants or mechanisms of airflow obstruction in asthma? Almost 40 years ago, Scavennec and colleagues (50) observed that the cAMP concentration in urine was higher in patients with solid tumors than in healthy volunteers. Therefore, cAMP levels in urine, or even plasma, could eventually serve as an efficacy biomarker to assess the patient’s response to MRP-targeted therapy. On the basis of these previous studies, we asked 1) whether cAMP can be reliably measured in the serum of patients and 2) whether serum cAMP levels can differentiate patients with obstructive lung diseases. Surprisingly, we find that the serum concentration of cAMP is increased in a small cohort of patients with asthma compared with control subjects. This proof-of-concept study now allows us to explore further the clinical utility of serum cAMP levels as potential endotype of obstructive lung diseases and/or markers of β2AR tachyphylaxis in large clinical studies.
| 1. | Barnes PJ. New drugs for asthma. Nat Rev Drug Discov 2004;3: 831–844. |
| 2. | An SS, Fabry B, Trepat X, Wang N, Fredberg JJ. Do biophysical properties of the airway smooth muscle in culture predict airway hyperresponsiveness? Am J Respir Cell Mol Biol 2006;35:55–64. |
| 3. | An SS, Mitzner W, Tang WY, Ahn K, Yoon AR, Huang J, et al. An inflammation-independent contraction mechanophenotype of airway smooth muscle in asthma. J Allergy Clin Immunol 2016;138:294–297, e4. |
| 4. | Billington CK, Penn RB. Signaling and regulation of G protein-coupled receptors in airway smooth muscle. Respir Res 2003;4:2. |
| 5. | Komalavilas P, Penn RB, Flynn CR, Thresher J, Lopes LB, Furnish EJ, et al. The small heat shock-related protein, HSP20, is a cAMP-dependent protein kinase substrate that is involved in airway smooth muscle relaxation. Am J Physiol Lung Cell Mol Physiol 2008;294: L69–L78. |
| 6. | Morgan SJ, Deshpande DA, Tiegs BC, Misior AM, Yan H, Hershfeld AV, et al. β-Agonist-mediated relaxation of airway smooth muscle is protein kinase A-dependent. J Biol Chem 2014;289:23065–23074. |
| 7. | Daaka Y, Luttrell LM, Lefkowitz RJ. Switching of the coupling of the beta2-adrenergic receptor to different G proteins by protein kinase A. Nature 1997;390:88–91. |
| 8. | Fan Chung K. Phosphodiesterase inhibitors in airways disease. Eur J Pharmacol 2006;533:110–117. |
| 9. | Trian T, Burgess JK, Niimi K, Moir LM, Ge Q, Berger P, et al. β2-Agonist induced cAMP is decreased in asthmatic airway smooth muscle due to increased PDE4D. PLoS One 2011;6:e20000. |
| 10. | Ojiaku CA, Cao G, Zhu W, Yoo EJ, Shumyatcher M, Himes BE, et al. TGF-β1 evokes human airway smooth muscle cell shortening and hyperresponsiveness via Smad3. Am J Respir Cell Mol Biol 2018;58:575–584. |
| 11. | Ojiaku CA, Chung E, Parikh V, Williams JK, Schwab A, Fuentes AL, et al. Transforming growth factor-β1 decreases β2-agonist-induced relaxation in human airway smooth muscle. Am J Respir Cell Mol Biol 2019;61:209–218. |
| 12. | Hakonarson H, Herrick DJ, Serrano PG, Grunstein MM. Mechanism of cytokine-induced modulation of beta-adrenoceptor responsiveness in airway smooth muscle. J Clin Invest 1996;97:2593–2600. |
| 13. | Himes BE, Hunninghake GM, Baurley JW, Rafaels NM, Sleiman P, Strachan DP, et al. Genome-wide association analysis identifies PDE4D as an asthma-susceptibility gene. Am J Hum Genet 2009;84:581–593. |
| 14. | Homma S, Sakamoto T, Hegab AE, Saitoh W, Nomura A, Ishii Y, et al. Association of phosphodiesterase 4D gene polymorphisms with chronic obstructive pulmonary disease: relationship to interleukin 13 gene polymorphism. Int J Mol Med 2006;18:933–939. |
| 15. | Liggett SB. β2-Adrenergic receptor pharmacogenetics. Am J Respir Crit Care Med 2000;161:S197–S201. |
| 16. | Hancox RJ, Sears MR, Taylor DR. Polymorphism of the beta2-adrenoceptor and the response to long-term beta2-agonist therapy in asthma. Eur Respir J 1998;11:589–593. |
| 17. | Israel E, Drazen JM, Liggett SB, Boushey HA, Cherniack RM, Chinchilli VM, et al. The effect of polymorphisms of the β2-adrenergic receptor on the response to regular use of albuterol in asthma. Am J Respir Crit Care Med 2000;162:75–80. |
| 18. | Lima JJ, Zhang S, Grant A, Shao L, Tantisira KG, Allayee H, et al. Influence of leukotriene pathway polymorphisms on response to montelukast in asthma. Am J Respir Crit Care Med 2006;173:379–385. |
| 19. | Tantisira KG, Lima J, Sylvia J, Klanderman B, Weiss ST. 5-Lipoxygenase pharmacogenetics in asthma: overlap with Cys-leukotriene receptor antagonist loci. Pharmacogenet Genomics 2009;19:244–247. |
| 20. | García-Menaya JM, Cordobés-Durán C, García-Martín E, Agúndez JAG. Pharmacogenetic factors affecting asthma treatment response: potential implications for drug therapy. Front Pharmacol 2019;10:520. |
| 21. | Deeley RG, Westlake C, Cole SP. Transmembrane transport of endo- and xenobiotics by mammalian ATP-binding cassette multidrug resistance proteins. Physiol Rev 2006;86:849–899. |
| 22. | Prieß M, Göddeke H, Groenhof G, Schäfer LV. Molecular mechanism of ATP hydrolysis in an ABC transporter. ACS Cent Sci 2018;4:1334–1343. |
| 23. | Rees DC, Johnson E, Lewinson O. ABC transporters: the power to change. Nat Rev Mol Cell Biol 2009;10:218–227. |
| 24. | van der Deen M, de Vries EG, Timens W, Scheper RJ, Timmer-Bosscha H, Postma DS. ATP-binding cassette (ABC) transporters in normal and pathological lung. Respir Res 2005;6:59. |
| 25. | Rotoli BM, Barilli A, Visigalli R, Ferrari F, Frati C, Lagrasta CA, et al. Characterization of ABC transporters in EpiAirwayTM, a cellular model of normal human bronchial epithelium. Int J Mol Sci 2020;21:3190. |
| 26. | Aguiar JA, Tamminga A, Lobb B, Huff RD, Nguyen JP, Kim Y, et al. The impact of cigarette smoke exposure, COPD, or asthma status on ABC transporter gene expression in human airway epithelial cells. Sci Rep 2019;9:153. |
| 27. | Mitra P, Oskeritzian CA, Payne SG, Beaven MA, Milstien S, Spiegel S. Role of ABCC1 in export of sphingosine-1-phosphate from mast cells. Proc Natl Acad Sci USA 2006;103:16394–16399. |
| 28. | Wijnholds J, Evers R, van Leusden MR, Mol CA, Zaman GJ, Mayer U, et al. Increased sensitivity to anticancer drugs and decreased inflammatory response in mice lacking the multidrug resistance-associated protein. Nat Med 1997;3:1275–1279. |
| 29. | Gold MJ, Hiebert PR, Park HY, Stefanowicz D, Le A, Starkey MR, et al. Mucosal production of uric acid by airway epithelial cells contributes to particulate matter-induced allergic sensitization. Mucosal Immunol 2016;9:809–820. |
| 30. | Huff RD, Rider CF, Yan D, Newton R, Giembycz MA, Carlsten C, et al. Inhibition of ABCC4 potentiates combination beta agonist and glucocorticoid responses in human airway epithelial cells. J Allergy Clin Immunol 2018;141:1127–1130, e5. |
| 31. | Cheng D, Ren J, Jackson EK. Multidrug resistance protein 4 mediates cAMP efflux from rat preglomerular vascular smooth muscle cells. Clin Exp Pharmacol Physiol 2010;37:205–207. |
| 32. | Wielinga PR, van der Heijden I, Reid G, Beijnen JH, Wijnholds J, Borst P. Characterization of the MRP4- and MRP5-mediated transport of cyclic nucleotides from intact cells. J Biol Chem 2003;278: 17664–17671. |
| 33. | Panettieri RA, Murray RK, DePalo LR, Yadvish PA, Kotlikoff MI. A human airway smooth muscle cell line that retains physiological responsiveness. Am J Physiol 1989;256:C329–C335. |
| 34. | Wiśniewski JR, Zougman A, Nagaraj N, Mann M. Universal sample preparation method for proteome analysis. Nat Methods 2009;6: 359–362. |
| 35. | Cox J, Mann M. MaxQuant enables high peptide identification rates, individualized p.p.b.-range mass accuracies and proteome-wide protein quantification. Nat Biotechnol 2008;26:1367–1372. |
| 36. | Ober C, Tan Z, Sun Y, Possick JD, Pan L, Nicolae R, et al. Effect of variation in CHI3L1 on serum YKL-40 level, risk of asthma, and lung function. N Engl J Med 2008;358:1682–1691. |
| 37. | Kan M, Koziol-White C, Shumyatcher M, Johnson M, Jester W, Panettieri RA Jr, et al. Airway smooth muscle-specific transcriptomic signatures of glucocorticoid exposure. Am J Respir Cell Mol Biol 2019;61: 110–120. |
| 38. | Kilic O, Yoon A, Shah SR, Yong HM, Ruiz-Valls A, Chang H, et al. A microphysiological model of the bronchial airways reveals the interplay of mechanical and biochemical signals in bronchospasm. Nat Biomed Eng 2019;3:532–544. |
| 39. | Tanfin Z, Serrano-Sanchez M, Leiber D. ATP-binding cassette ABCC1 is involved in the release of sphingosine 1-phosphate from rat uterine leiomyoma ELT3 cells and late pregnant rat myometrium. Cell Signal 2011;23:1997–2004. |
| 40. | Matsson P, Pedersen JM, Norinder U, Bergström CA, Artursson P. Identification of novel specific and general inhibitors of the three major human ATP-binding cassette transporters P-gp, BCRP and MRP2 among registered drugs. Pharm Res 2009;26:1816–1831. |
| 41. | Zhu Z, Lee CG, Zheng T, Chupp G, Wang J, Homer RJ, et al. Airway inflammation and remodeling in asthma. Lessons from interleukin 11 and interleukin 13 transgenic mice. Am J Respir Crit Care Med 2001;164:S67–S70. |
| 42. | Chupp GL, Lee CG, Jarjour N, Shim YM, Holm CT, He S, et al. A chitinase-like protein in the lung and circulation of patients with severe asthma. N Engl J Med 2007;357:2016–2027. |
| 43. | Sutherland EW. Studies on the mechanism of hormone action. Science 1972;177:401–408. |
| 44. | Rall TW, Sutherland EW. Formation of a cyclic adenine ribonucleotide by tissue particles. J Biol Chem 1958;232:1065–1076. |
| 45. | Davoren PR, Sutherland EW. The effect of l-epinephrine and other agents on the synthesis and release of adenosine 3′,5′-phosphate by whole pigeon erythrocytes. J Biol Chem 1963;238: 3009–3015. |
| 46. | Sager G, Ravna AW. Cellular efflux of cAMP and cGMP: a question about selectivity. Mini Rev Med Chem 2009;9:1009–1013. |
| 47. | Mavissakalian M, Brady S. The current state of biologic therapies for treatment of refractory asthma. Clin Rev Allergy Immunol 2020;59: 195–207. |
| 48. | An SS, Bai TR, Bates JH, Black JL, Brown RH, Brusasco V, et al. Airway smooth muscle dynamics: a common pathway of airway obstruction in asthma. Eur Respir J 2007;29:834–860. |
| 49. | King GG, Paré PD, Seow CY. The mechanics of exaggerated airway narrowing in asthma: the role of smooth muscle. Respir Physiol 1999;118:1–13. |
| 50. | Scavennec J, Carcassonne Y, Gastaut JA, Blanc A, Cailla HL. Relationship between the levels of cyclic cytidine 3′:5′-monophosphate, cyclic guanosine 3′:5′-monophosphate, and cyclic adenosine 3′:5′-monophosphate in urines and leukocytes and the type of human leukemias. Cancer Res 1981;41: 3222–3227. |
* These authors contributed equally to this work.
‡ These authors contributed equally to this work.
Supported by New Jersey Alliance for Clinical and Translational Science–National Center for Advancing Translational Sciences grant UL1TR0030117, National Heart, Lung, and Blood Institute grant P01HL114471, and National Institutes of Health grant R56HL155937.
Author Contributions: G.C., S.S.A., and R.A.P. designed research. G.C., H.L., J.A.J., and N.K. performed research. J.A.J., M.K., W.J., C.K.-W., B.E.H., G.L.C., S.S.A., and R.A.P. contributed new reagents and analytic tools. G.C., H.L., J.A.J., N.K., C.K.-W., S.S.A., and R.A.P. analyzed data. G.C., S.S.A., and R.A.P. wrote the paper. S.S.A. and R.A.P. directed all studies.
This article has a data supplement, which is accessible from this issue’s table of contents at www.atsjournals.org.
Originally Published in Press as DOI: 10.1165/rcmb.2021-0345OC on October 14, 2021
Author disclosures are available with the text of this article at www.atsjournals.org.
