Search tips
Search criteria 


Logo of ajrcmbIssue Featuring ArticlePublisher's Version of ArticleSubmissionsAmerican Thoracic SocietyAmerican Thoracic SocietyAmerican Journal of Respiratory Cell and Molecular Biology
Am J Respir Cell Mol Biol. 2006 October; 35(4): 496–502.
Published online 2006 May 18. doi:  10.1165/rcmb.2005-0290OC
PMCID: PMC2065849

Adenosine A1 Receptors Mediate Mobilization of Calcium in Human Bronchial Smooth Muscle Cells


Adenosine stimulates contraction of airway smooth muscle, but the mechanism is widely considered indirect, depending on release of contractile agonists from mast cells and nerves. The goal was to determine whether adenosine, by itself, directly regulates calcium signaling in human bronchial smooth muscle cells (HBSMC). Primary cultures of HBSMC from normal subjects were loaded with fura 2-AM, and cytosolic calcium concentrations ([Ca2+]i) were determined ratiometrically by imaging single cells. The nonselective adenosine receptor agonist, 5′-N-ethylcarboxamidoadenosine (NECA), and the adenosine A1 receptor agonist, N6-cyclopentyladenosine (CPA), both stimulated rapid, transient increases in [Ca2+]i. In contrast, there were no calcium responses to 2-p-(2-carboxyethyl)phenethylamino-5′-N-ethylcarboxamido-adenosine (100 nM) or N6-(3-iodobenzyl)-adenosine-5′-N-methyluronamide (100 nM), selective agonists at adenosine A2A receptors and adenosine A3 receptors, respectively. Calcium responses to NECA and CPA were inhibited by 8-cyclopentyl-1,3-dipropylxanthine, an adenosine A1 receptor antagonist, and by pertussis toxin (PTX). In other experiments, NECA stimulated calcium transients in the absence of extracellular calcium, but not when cells were preincubated in cyclopiazonic acid or thapsigargin to empty intracellular calcium stores. Calcium responses were attenuated by xestospongin C and 2-aminoethoxydiphenylborane, inhibitors of inositol trisphosphate (IP3) receptors, and by U73122, an inhibitor of phospholipase C. It was concluded that stimulation of adenosine A1 receptors on HBSMC rapidly mobilizes intracellular calcium stores by a mechanism dependent on PTX-sensitive G proteins, and IP3 signaling. These findings suggest that, in addition to its well-established indirect effects on HBSMC, adenosine also has direct effects on contractile signaling pathways.

Keywords: adenosine A1 receptor, calcium, cAMP, human bronchial smooth muscle, insulin

Adenosine stimulates airway smooth muscle contraction in normal tissues from humans and animals, and this response is upregulated in tissues from individuals with asthma and in models of asthma (15). Most evidence suggests that adenosine causes airway smooth muscle contraction by stimulating mast cells and nerves to release mediators that secondarily contract the airway smooth muscle cell (1, 3, 68). However, a few studies have isolated human airway smooth muscle cells in culture and shown that adenosine has some direct effects on this cell type and that adenosine A1, A2A, and A2B receptors are all expressed on normal human airway smooth muscle cells. For example, A2B receptors are coupled to the release of cytokines and vascular endothelial growth factor and to stimulation of cAMP formation (911). Also, an adenosine A1 receptor agonist inhibited forskolin-stimulated cAMP production and chronic activation of the adenosine A1 receptor has been linked to sensitization of adenylyl cyclase (9). Interestingly, studies in an allergic rabbit model showed that activation of adenosine A1 receptors caused contraction in airway smooth muscle tissue devoid of mast cells and suggested that A1 receptors are directly coupled to calcium mobilization and contraction (12, 13). Evidence that adenosine A1 receptors directly regulate calcium in airway smooth muscle is potentially important because individuals with asthma are hyperresponsive to adenosine (14), and a recent report demonstrated that this receptor subtype is present in biopsies of asthmatic human bronchial smooth muscle tissue (15).

Cultured cells are the main, widely used model for studying isolated airway smooth muscle without interference from other cell types. Although all four adenosine receptors (A1, A2A, A2B, and A3) have been shown to stimulate calcium signaling in other cell types (16), the coupling of cultured airway smooth muscle cell adenosine receptors to calcium signaling has not been previously demonstrated. For the current study we hypothesized that adenosine directly stimulates one or more adenosine receptor subtypes coupled to calcium signaling in airway smooth muscle cells. Using cultured bronchial smooth muscle cells from normal human donors, the goal was to identify the adenosine receptor subtype(s), calcium source(s), and signaling pathways mediating these calcium responses.


Human Bronchial Smooth Muscle Cell Culture

Human bronchial smooth muscle cells (HBSMC) from three different human donors were obtained at passage 3 from Cambrex Bioproducts (Walkersville, MD). These cultured cells were from normal human donors with no history of asthma. The cells were primary, nontransformed cultures that exhibited fluorescent staining that was positive for smooth muscle actin and negative for Factor VIII. Cells were seeded and cultured in smooth muscle growth media (SMGM2; Cambrex) consisting of MCDB-based smooth muscle basal medium (Cambrex) supplemented with 5% fetal bovine serum, 5 μg/ml insulin, 2 ng/ml fibroblast growth factor, 0.5 ng/ml epidermal growth factor, 50 μg/ml gentamicin, and 50 ng/ml amphotericin B. Cells were maintained at 37°C in a humidified atmosphere (5% CO2), fed every 48 h, and passaged when 80–90% confluent.

Cell Preparation

For calcium imaging, HBSMC (passages 4–6) were seeded in 96-well special optic plates with thin (0.005-in) bottoms (Corning Inc., Corning, NY) at a density of 3,500 cells/cm2. Except for experiments described in Figures 6 and and7,7, all cells were grown to 50% confluence in SMGM2 then serum-starved by incubation in modified arresting medium (MAM) consisting of DMEM (2 mM l-glutamine, 100 U/ml penicillin, 100 μg/ml streptomycin) and supplemented with HEPES (25 mM), NaOH (10 mM), insulin (5.7 μg/ml), and transferrin (5 μg/ml) (17). After 24 h, cells were washed with a modified Krebs-Ringer-Henseleit buffer (KRH) (18) containing (in mM): 115 NaCl, 5 KCL, 1 KH2PO4, 1 MgSO4, 25 HEPES, 15 glucose, and 2 CaCl2, and incubated in 2.0 μM fura 2-AM for 60 min at room temperature. Then cells were washed twice with KRH and preincubated in KRH containing adenosine deaminase (ADA) (1 U/ml) to eliminate any endogenous adenosine released by the cells (9, 19). After 30 min, agonists, antagonists, and/or inhibitors were added to culture wells as specified.

Figure 6.
Calcium responses to NECA are modulated by insulin. HBSMC were seeded and grown to 50% confluence in SMGM2 as described in Materials and Methods. Medium was replaced with MAM (which contains insulin), DMEM, DMEM with HEPES (25 mM), DMEM with insulin (5.7 ...
Figure 7.
Stimulation of cAMP by NECA is increased by DPCPX. HBSMC were serum-starved for 24 h by incubation in either (A) MAM (insulin-containing media with frequent calcium responses, see Figure 6) or (B) DMEM (non–insulin-containing media with infrequent ...


Fura 2 was excited by computer-controlled 337- and 380-nm ultraviolet light generated by a nitrogen laser and a nitrogen laser-pumped dye laser, respectively (Laser Science, Franklin, MA). Each laser alternately fired pulses (3 ns) at 30 Hz, and the pulses were directed at the cells through a ×40 objective lens (Nikon, Melville, NY). The fluorescent signals emitted by fura 2 were passed back through the objective to a 455-nm dichroic mirror, a 475-nm barrier filter (Omega Optics, Brattleboro, VT), an image intensifier (Xybion Electronic Systems, San Diego, CA), and captured by a Philips-based frame transfer charge coupled device (CCD) camera (CCTV, New York, NY). The analog signals from the camera were digitized with outputs to a personal computer with software by Recognition Technology, Inc. (Framingham, MA). As described previously (20), background from a cell-free region was subtracted before data acquisition and then an 11 × 11 pixel area was selected over the cell. The fluorescence stimulated by alternating pulses of 337- and 380-nm light were recorded and their ratios plotted. Ratios were converted to calcium concentrations: [Ca2+]i = Kd · β · (R − Rmin)/(Rmax − R), where Rmax and Rmin are the fluorescence ratios measured in high and zero calcium, respectively; β is the ratio of fluorescence stimulated by 380 nm light in zero versus high calcium; and Kd (224 nM) is the equilibrium dissociation constant describing calcium binding to fura 2 (21).

Cyclic AMP

Confluent HBSMC (passages 4–6) in 24-well plates were serum-starved for 24 h by incubation in either MAM or DMEM. Then cultures were washed with KRH and preincubated in 1,3-dipropyl-8-cyclopentylxanthine (DPCPX) (100 nM) or vehicle (KRH) for 30 min. Cells were then stimulated for 30 s with either vehicle or NECA (50 μM), and the reaction was terminated by addition of 0.1 N HCl containing 0.2% Triton X-100. An aliquot of this cell extract was used for determination of protein concentration by Bradford assay (Bio-Rad, Hercules, CA) and another aliquot of the same sample was used for determination of cAMP concentration by competitive immunoassay (R&D Systems, Minneapolis, MN).

Data Analysis

Data were expressed as mean ± SEM. Means were compared by Student's t test, and multiple comparisons between means were analyzed by ANOVA with Newman-Keuls post hoc follow-up testing. For assessing the effects of different culture media on calcium responses, proportions of cells showing any calcium responses to NECA were compared by chi-square testing and Fisher's exact test with Bonferroni correction for multiple comparisons. GraphPad Prism Software (San Diego, CA) was used for analyses and P < 0.05 was considered significant.


Fura 2-AM and pluronic F-127 were obtained from Molecular Probes (Eugene, OR). Insulin and transferrin were obtained from Invitrogen (Carlsbad, CA). Other reagents were obtained from Sigma (St. Louis, MO).


For HBSMC, 5′-N-ethylcarboxamidoadenosine (NECA), a nonselective adenosine receptor agonist that activates all four adenosine receptor subtypes, caused a rapid, transient increase in [Ca2+]i that was similar in magnitude to transients elicited by histamine (Figure 1A). Approximately 85% of cells responded to NECA with a calcium transient. Calcium responses to NECA were observed at concentrations ranging from 10−8–10−4 M, and the peak magnitude of the calcium transient was concentration dependent, with maximum responses occurring at ~ 1 μM NECA (Figure 1B). Responses of similar magnitude were observed in cells from all three donors.

Figure 1.
NECA stimulates concentration-dependent calcium responses in HBSMC. Cells were prepared as described in Materials and Methods. (A) Representative traces show F340/F380 ratio as a measure of [Ca2+]i in single cells exposed to NECA (10 μM) ...

The potent and selective adenosine A1 receptor agonist, N6-cyclopentyladenosine (CPA), also caused rapid, transient increases in [Ca2+]i (Figure 2A). Responses were dose–dependent, with maximal responses occurring at 0.1–10 μM (Figure 2B) and, as with NECA, ~ 85% of the cells responded to CPA. Similar responses were observed in cells from all three donors. In experiments comparing CPA to other selective adenosine receptor agonists, five of six cells responded to CPA (100 nM), and mean [Ca2+]i increased from a basal level of 66 ± 12 nM to a maximum of 501 ± 110 nM, (P < 0.01, n = 6) (Figure 3A). In contrast, neither 2-p-(2-carboxyethyl)phenethylamino-5′-N-ethylcarboxamidoadenosine (CGS-21680) (100 nM), a selective agonist at the adenosine A2A receptor, nor N6-(3-iodobenzyl)-adenosine-5′-N-methyluronamide (IB-MECA) (100 nM), a selective agonist at the adenosine A3 receptor, stimulated calcium responses in any cell (Figure 3A).

Figure 2.
CPA stimulates concentration-dependent calcium responses. The adenosine A1 receptor agonist, CPA, was added to HBSMC and responses recorded. (A) A representative trace shows the F340/F380 ratio in a single cell as a measure of [Ca2+]i in response ...
Figure 3.
Adenosine A1 receptor signaling stimulates calcium responses in HBSMC. (A) Agonists (100 nM) selective for adenosine receptor subtypes were added to HBSMC and calcium responses recorded. (B) CPA (1 μM) or NECA (10 μM) was added to HBSMC ...

The selective adenosine A1 receptor antagonist, DPCPX, inhibited calcium responses to both CPA and NECA (Figure 3B). In eight control cells stimulated with 1 μM CPA (seven responders), [Ca2+]i increased from 62 ± 10 nM to 783 ± 267 nM (P < 0.01). In contrast, in the presence of DPCPX (100 nM), responses to CPA were attenuated in eight of eight cells tested. Specifically, when CPA was applied to the cells, [Ca2+]i increased from a basal value of 55 ± 12 nM to a peak value of only 105 ± 38 nM (NS, n = 8). Similarly, when cells were stimulated by NECA (10 μM) instead of CPA, [Ca2+]i increased from 77 ± 8 nM to 1,084 ± 208 nM (P < 0.001, n = 10) in the absence of DPCPX, but from 72 ± 8 nM to only 215 ± 89 nM in the presence of DPCPX (100 nM) (NS, n = 9). DPCPX (100 nM) alone had no effect on [Ca2+]i.

Pretreatment of HBSMC with pertussis toxin (PTX) (200 ng/ml × 24 h), an inhibitor of Gi/Go proteins, attenuated calcium transients in response to both CPA and NECA (Figure 3C). In control cells, [Ca2+]i increased from 65 ± 20 nM to 535 ± 98 nM (P < 0.001, n = 7) in response to CPA (1 μM). In contrast, in cells pretreated with PTX, basal [Ca2+]i was 64 ± 19 nM and was only 69 ± 19 nM in response to CPA (NS, n = 7). Similarly, in control cells, [Ca2+]i increased from 48 ± 9 nM to 715 ± 209 nM (P < 0.001, n = 5) in response to NECA (10 μM), but in cells pretreated with PTX, basal [Ca2+]i was 48 ± 7 nM and was only 46 ± 5 nM in response to NECA (NS, n = 5). In contrast, the peak magnitude of calcium transients in response to histamine (10 μM) was the same with, or without, pretreatment with PTX. Specifically, in response to histamine, [Ca2+]i increased from 63 ± 14 nM to 677 ± 109 nM (n = 8) and from 97 ± 18 nM to 690 ± 95 nM (n = 8), without and with pretreatment with PTX, respectively.

To determine the source of calcium supporting these increases in [Ca2+]i in response to adenosine receptor agonists, transients in response to NECA (10 μM) were compared and found to be similar in the presence and absence of extracellular calcium (Figure 4). In the presence of extracellular calcium, 8 of 10 cells responded (peak increase in [Ca2+]i = 990 ± 237 nM, n = 10). Similarly, in the absence of extracellular calcium, six of seven cells responded (peak increase in [Ca2+]i = 755 ± 173 nM, n = 7). In other experiments, in the presence of extracellular calcium, HBSMC were treated with cyclopiazonic acid (30 μM) or thapsigargin (0.3 μM) for 20 min to deplete intracellular calcium stores by inhibiting sarco-endoplasmic reticulum calcium ATPase (SERCA) (20, 22). No responses to NECA were observed after pretreatment with either cyclopiazonic acid (n = 6) or thapsigargin (n = 6) (Figure 4).

Figure 4.
Calcium transients in response to NECA depend on mobilization of calcium from intracellular stores. NECA (10 μM) was added to HBSMC in the presence (I) or absence (II) of extracellular calcium, or after pretreatment with cyclopiazonic acid (30 ...

In the presence of 2-aminoethoxydiphenylborane (2-APB) and xestospongin C, cell-permeable inhibitors of the inositol trisphosphate (IP3) receptor (IP3R), calcium responses to NECA were inhibited (Figure 5). In five control cells, [Ca2+]i increased by 1,019 ± 335 nM in response to NECA (10 μM), but in the presence of xestospongin C the maximum increase in [Ca2+]i was significantly less (312 ± 197 nM, P < 0.05, n = 5). Similarly, in the presence of 2-APB, [Ca2+]i did not increase in response to NECA (n = 4). The presence of a phospholipase C (PLC) inhibitor, 1-[6-[((17β)-3-methoxyestra-1,3,5[10]-trien-17-yl)amino]hexyl]-1H-pyrrole-2,5-dione (U73122), also significantly inhibited calcium responses to NECA (Figure 5). In the presence of U73122, [Ca2+]i was increased by only 4 ± 4 nM (NS, n = 5) in response to NECA.

Figure 5.
Stimulation of calcium responses by NECA is inhibited by U73122, xestospongin C, and 2-APB. Cells were stimulated with NECA (10 μM) in the absence or presence of U73122 (5 μM × 60 min), xestospongin C (xest C) (20 μM × ...

At the outset of these projects it appeared that calcium responses to NECA were more likely to occur when cells were preincubated for 24 h in a serum-free modified arresting medium (MAM) and exposed to ADA for 30 min before experiments. For this reason, experiments described above were performed under these culture conditions. We then performed additional experiments to determine what components of these conditions were important for eliciting responses to NECA. To assess whether the presence of ADA was important for responses to NECA, we exposed cells to ADA (1 U/ml) for 30 min or 24 h and found that, for both incubation times, the proportion of responding cells was similar to cells not exposed to ADA. After 30 min in ADA or vehicle, calcium responses to NECA (10 μM) were observed in 10 of 11 and 6 of 8 cells, respectively. Similarly, after 24 h in ADA or vehicle, 6 of 8 and 7 of 8 cells responded, respectively. Adding ADA alone had no effect on [Ca2+]i. Based on these results, it appeared that addition of ADA did not play a role in optimizing responses to NECA.

Because MAM is composed of DMEM supplemented with insulin, transferrin, and HEPES, in separate experiments we investigated which component(s) in MAM favored calcium responses to NECA. Calcium responses to NECA (10 μM) were assessed after 24 h incubation in media containing various combinations of MAM components (Figure 6). There was a broad range of calcium responses in all five media tested (Figure 6A), but the fraction of cells responding to NECA with increased [Ca2+]i was significantly greater when cells were preincubated in media containing insulin (i.e., MAM or DMEM plus insulin) (P < 0.005) (Figure 6B). In all of the media tested, and for responding and nonresponding cells, there were no differences in basal [Ca2+]i and no morphologic differences by phase contrast microscopy. In other experiments we were unable to increase the proportion of cells responding to NECA (10 μM) in DMEM (no insulin) by including ADA (1 U/ml) in the DMEM. After 24 h in DMEM with or without ADA, 6 of 12 and 5 of 12 cells responded, respectively.

In many cell types, adenosine A1 receptors are negatively coupled to adenylyl cyclase via Gi and A2 receptors are coupled to stimulation of adenylyl cyclase via Gs (16). The nonselective adenosine receptor agonist, NECA, activates both receptor subtypes. For cells preincubated 24 h in MAM (contains insulin), the overall effect of NECA (50 μM) was a cAMP increase from 3.7 ± 0.2 to 24.0 ± 1.5 pmoles/mg protein (P < 0.001, n = 8−9) (Figure 7A). After preincubation for 30 min in DPCPX (100 nM), a selective adenosine A1 receptor antagonist, stimulation of cAMP levels by NECA was increased to 31.0 ± 2.9 pmoles/mg protein, a value significantly greater than that achieved in the absence of DPCPX (P < 0.04, n = 9). The finding is consistent with the presence of A1 receptors on these cells and suggests that when the nonselective agonist, NECA, is applied by itself, there is a mixed effect on cAMP levels because NECA activates both A1 and A2 receptors. In contrast, for cells preincubated 24 h in DMEM (no insulin), NECA still increased cAMP from 4.7 ± 0.4 to 28.0 ± 2.4 pmoles/mg protein, but this response was not enhanced by A1 receptor antagonism with DPCPX (27.9 ± 3.7 pmoles/mg protein, NS) (Figure 7B).


The main finding of this study is that adenosine receptor agonists stimulate large, transient increases in [Ca2+]i (calcium transients) in normal HBSMC. These calcium responses to adenosine agonists are mediated by adenosine A1 receptors that are coupled to PTX-sensitive G proteins and IP3 signaling pathways to mobilize calcium from intracellular stores. These findings are novel because calcium transients in direct response to adenosine receptor agonists have not been demonstrated previously in cultures of human airway smooth muscle cells. Because calcium is an important regulator of cell contraction, these findings suggest the possibility that, in addition to the well-established, indirect effects that it has on airway smooth muscle function, adenosine may also have direct effects that regulate bronchomotor tone in the airways.

In the current study, for HBSMC, calcium transients in response to adenosine were mediated by the adenosine A1 receptor. Several findings support this. First, calcium responses were elicited by CPA, a selective adenosine A1 agonist, and NECA, which activates all four of the known adenosine receptor subtypes (A1, A2A, A2B, and A3). Responses to both of these agonists were concentration dependent and, consistent with responses mediated by adenosine A1 receptors, were detected at concentrations as low as 10 nM (16, 23). Second, additional evidence that adenosine A1 receptors mediated the responses was the finding that agonists selective for adenosine A2A and A3 receptors did not stimulate calcium transients in these cells. The lack of response to an A3 agonist is significant because a study of rat tracheal smooth muscle cells previously showed that A3 receptor activation enhances calcium transients in response to the purine nucleotide, ATP (24). Third, also consistent with the presence of A1 receptors on these cells, the nonselective adenosine agonist, NECA, increased cAMP levels more when an A1 receptor antagonist was present.

Although highly selective agonists for the adenosine A2B receptor are not available (16, 25), three separate findings all support the conclusion that the adenosine A1 receptor, not A2B, is the receptor subtype coupled to calcium transients in HBSMC. First, the low concentrations (100 nM) of adenosine agonists selective for the A1, A2A, and A3 receptors in this study do not significantly activate A2B receptors (11, 16). Second, a low concentration (100 nM) of the selective adenosine A1 receptor antagonist, DPCPX, inhibited calcium responses to both CPA and NECA. Finally, adenosine A1 receptors, but not adenosine A2B receptors, are known to be coupled to PTX-sensitive G proteins in other cell types (13, 16, 2527). In the current study, NECA- and CPA-stimulated calcium transients were inhibited by PTX. This finding was not a nonspecific effect of PTX on calcium signaling in general because mobilization of calcium by a different agonist, histamine, was unaffected by the same preincubation in PTX. Therefore, in aggregate, all these findings are most consistent with adenosine A1 receptors, not adenosine A2B receptors, mediating the generation of calcium transients in HBSMC.

To begin to assess the mechanism underlying adenosine A1 receptor–stimulated increases in [Ca2+]i, the source of the calcium ions supporting the transient was evaluated. Because NECA stimulated calcium transients in calcium-free media, it was concluded that intracellular stores were the main source of calcium for transients. Consistent with this, depleting SR calcium by preincubating the cells in thapsigargin or cyclopiazonic acid, both inhibitors of SERCA, abolished subsequent calcium transients in response to NECA. Therefore, adenosine A1 receptors mobilize calcium from the SR. Because responses to adenosine agonists also were attenuated by an inhibitor of PLC and by inhibitors of IP3R, it was further concluded that PLC activation, IP3 generation, and calcium release through IP3R are the signaling events coupling adenosine A1 receptor activation to the release of calcium ions from the SR. That adenosine A1 receptors could couple to PLC through a PTX-sensitive G protein is possible because studies in a variety of cells and tissues have demonstrated calcium mobilization stimulated by adenosine A1 receptor coupling to Gi (13, 16, 26, 27), probably involving the activation of PLC-β by G protein βγ subunits (2629). It is acknowledged, however, that our findings do not exclude an alternative possibility that A1 receptor activation stimulates the release of an unidentified mediator from HBSMC that secondarily stimulates PLC activation and calcium mobilization.

Finding that adenosine A1 receptors are coupled to calcium in HBSMC suggests that this receptor subtype could play a role in regulating smooth muscle cell functions in the airways. Previous studies reporting direct effects of adenosine on cell signaling in human airway smooth muscle have identified A2B as the predominant functionally coupled adenosine receptor subtype (911). The adenosine A2B receptor is a low affinity receptor that is coupled to Gs and, possibly, Gq, but not to Gi or Go (16, 25) and activation of the A2B receptor in human airway smooth muscle stimulates release of cytokines and vascular endothelial growth factor (10, 11) as well as cAMP accumulation (9). Consistent with there being both A1 and A2B receptors on HBSMC, we found that the non-selective adenosine receptor agonist, NECA, increased cAMP levels more when a specific adenosine A1 antagonist was present in the media. Although this finding constitutes additional evidence that A1 receptors are present on HBSMC, it is acknowledged that our data do not necessarily indicate that A1 receptors are coupled to Gi and adenylyl cyclase directly because cAMP levels could have been regulated by other mechanisms, including activation of phosphodiesterase and calcium signaling. However, another study of human airway smooth muscle cells has shown that, in the presence of a phosphodiesterase inhibitor, forskolin-stimulated cAMP accumulation was inhibited by the adenosine A1 agonist, CPA (9). In aggregate, our findings with cAMP do not define the mechanism of A1 coupling but are consistent with prior evidence showing that both adenosine A1 and A2B receptors are expressed by human airway smooth muscle cells (911).

Several reasons may explain why calcium transients in response to adenosine were observed in the present study, but not in previous studies of airway smooth muscle cells from human and rat trachea (8, 22, 30). Perhaps most important, it is possible that the bronchial origin of the cells used in the present study was an important difference compared with earlier studies that cultured tracheal cells (9, 24, 30). In support of this possibility, in an allergic rabbit model, smooth muscle from the peripheral airways contracted more in response to adenosine than smooth muscle from the trachea (31). Second, a feature of the current study was that changes in [Ca2+]i were measured in single cells rather than in a population of cells (24, 30). Studies that employed simultaneous measurement of [Ca2+]i in multiple cells might not have detected significant changes if calcium responses were occurring in a subpopulation of cells. Third, our inclusion of insulin in the culture medium 24 h before experiments was important for establishing optimal conditions for A1 receptor coupling to both calcium signaling and inhibition of cAMP. However, because a subpopulation of cells responded to NECA even in media without insulin, the presence of insulin does not appear to be an absolute requirement.

Adenosine A1 receptor activation has been linked to insulin sensitivity in muscle tissue and adipocytes (32, 33) but we are not aware of prior studies showing that insulin treatment enhances responses that are coupled to A1 receptors. The reason adenosine-stimulated calcium responses are more frequent among HBSMC incubated in media containing insulin is not known. Our data do suggest the effect of insulin-containing media also extends to other cellular functions coupled to A1 receptors, specifically A1 receptor inhibition of cAMP. Because media containing insulin optimized both calcium responses and inhibition of cAMP, we speculate that a common, proximal point in these two different signaling pathways is being modulated by insulin, for example at the adenosine receptor or G protein levels. However, it is acknowledged that our data do not address directly the mechanism underlying effects of these media on A1 receptor responses and do not exclude effects downstream of G proteins. We also considered the possibility that adenosine release into the culture medium might be regulated by the presence of insulin and that this could contribute to our findings. However, against this, incubating cultures in different media (with and without insulin) for up to 24 h in ADA did not affect calcium responses to A1 receptor stimulation.

In vivo and in vitro studies in normal humans and animals have shown airway smooth muscle contraction or bronchoconstriction in response to adenosine (15). However, contractile responses to adenosine in mixed cell-type preparations are widely believed to be indirect and dependent on mediator release from mast cells and nerves (18). Our findings suggest that caution is warranted when making an assumption that adenosine-stimulated airway smooth muscle contraction is always indirect and mediated only by adenosine acting first on cells other than smooth muscle. Although our cells are derived from normal subjects, we speculate that our findings may be relevant in asthma because it is well-established that responses to adenosine are increased in asthma (1, 14), and a recent study has identified adenosine A1 receptors in the airway smooth muscle of subjects with asthma (15).

In summary, for single HBSMC, adenosine A1 receptors are coupled to calcium signaling, a second messenger pathway important to the regulation of bronchomotor tone. The calcium responses are mediated by adenosine A1 receptors coupled to PTX-sensitive G proteins and depend on the activation of PLC and the mobilization of calcium from the SR. These findings indicate that adenosine can directly regulate calcium mobilization in airway smooth muscle cells and that HBSMC are a potentially useful model for studying these direct effects of adenosine.


This study was supported by a grant from the National Institutes of Health (HL-54143).

Originally Published in Press as DOI: 10.1165/rcmb.2005-0290OC on May 18, 2006

Conflict of Interest Statement: Neither author has a financial relationship with a commercial entity that has an interest in the subject of this manuscript.


1. Bjorck T, Gustafsson LE, Dallen SE. Isolated bronchi from asthmatics are hyperresponsive to adenosine which apparently acts indirectly by liberation of leukotrienes and histamine. Am Rev Respir Dis 1992;145: 1087–1091. [PubMed]
2. Finney MJB, Karlsson J-A, Persson CGA. Effects of bronchoconstrictors and bronchodilators on a novel human small airway preparation. Br J Pharmacol 1985;85:29–36. [PMC free article] [PubMed]
3. Meade C, Mierau J, Indepencia L, Ensinger HA. In vivo role of the adenosine A3 receptor: N6–2-(4-aminophenyl) ethyladenosine induces bronchospasm in BDE rats by a neurally mediated mechanism involving cells resembling mast cell. J Pharmacol Exp Ther 1996;297:1148–1156. [PubMed]
4. Fan M, Qin W, Mustafa SJ. Characterization of adenosine receptors involved in adenosine-induced bronchoconstriction in an allergic mouse model. Am J Physiol 2003;284:L1012–L1019. [PubMed]
5. Ghai G, Zimmerman MB, Hopkins MF. Evidence for A1 and A2 adenosine receptors in guinea pig trachea. Life Sci 1987;41:1215–1224. [PubMed]
6. Peachell PT, Columbo M, Kagey-Sobotka A, Lichtenstein LM, Marone G. Adenosine potentiates mediator release from human lung mast cells. Am Rev Respir Dis 1988;138:1143–1151. [PubMed]
7. Polosa R, Phillips GD, Rajakulasingam K, Holgate ST. The effect of inhaled ipratropium bromide alone and in combination with oral terfenadine on bronchoconstriction provoked by adenosine 5′-monophosphate in asthma. J Allergy Clin Immunol 1991;87:939–947. [PubMed]
8. Church MK, Holgate ST. Adenosine-induced bronchoconstriction and its inhibition by nedocromil sodium. J Allergy Clin Immunol 1993;92: 190–194. [PubMed]
9. Mundell SJ, Olah ME, Panettieri RA, Benovic JL, Penn RB. Regulation of G protein-coupled receptor-adenylyl cyclase responsiveness in human airway smooth muscle by exogenous and autocrine adenosine. Am J Respir Cell Mol Biol 2001;24:155–163. [PubMed]
10. Feoktistov I, Ryzhov S, Zhong H, Goldstein AE, Matafonov A, Zeng D, Biaffioni I. Hypoxia modulates adenosine receptors in human endothelial and smooth muscle cells toward an A2B angiogenic phenotype. Hypertension 2004;44:1–6. [PubMed]
11. Zhong H, Belardinelli L, Maa T, Feoktistov I, Biaggioni I, Zeng D. A2B adenosine receptors increase cytokine release by bronchial smooth muscle cells. Am J Respir Cell Mol Biol 2004;30:118–125. [PubMed]
12. Ali S, Metzger JW, Mustafa JS. Simultaneous measurement of cyclopentyladenosine-induced contraction and intracellular calcium in bronchial rings from allergic rabbits and its antagonism. J Pharmacol Exp Ther 1996;278:639–644. [PubMed]
13. Abebe W, Mustafa J. A1 adenosine receptor-mediated Ins(1,4,5)P3 generation in allergic rabbit airway smooth muscle. Am J Physiol 1998;275: L990–L997. [PubMed]
14. Cushley MJ, Tattersfield AE, Holgate ST. Inhaled adenosine and guanosine on airway resistance in normal and asthmatic subjects. Br J Clin Pharmacol 1983;5:161–165. [PMC free article] [PubMed]
15. Brown RA, Clarke GW, Hurle MJ, Denyer JC, Savage TJ, Ratoff JC, Ledbetter CL, Murdoch RD, O'Connor BJ, Page CP, et al. Immunolocalisation of adenosine A1 receptors in asthmatic bronchial biopsy sections [abstract]. Proc Am Thoracic Soc 2005;2:A805.
16. Fredholm BB, Ijzerman AP, Jacobson KA, Klotz K-N, Linden J. International Union of Pharmacology. XXV. Nomenclature and classification of adenosine receptors. Pharmacol Rev 2001;53:527–552. [PubMed]
17. Deshpande DA, Walseth TF, Panettieri RA, Kannan MS. CD38/cyclic ADP-ribose-mediated Ca2+ signaling contributes to airway smooth muscle hyper-responsiveness. FASEB J 2003;17:452–454. [PubMed]
18. Wang YX, Fleischmann BK, Kotlikoff MI. Modulation of maxi-K+ channels by voltage-dependent Ca2+ channels and methacholine in single airway myocytes. Am J Physiol 1997;272:C1151–C1159. [PubMed]
19. Ethier MF, Hickler RB, Dobson JG. Aging increases adenosine and inosine release by human fibroblast cultures. Mech Ageing Dev 1989; 50:159–168. [PubMed]
20. Ethier MF, Yamaguchi H, Madison JM. Effects of cyclopiazonic acid on cytosolic calcium in bovine airway smooth muscle cells. Am J Physiol 2001;281:L126–L133. [PubMed]
21. Grynkiewicz G, Poenie M, Tsien RY. A new generation of Ca2+ indicators with greatly improved fluorescence properties. J Biol Chem 1985; 260:3440–3450. [PubMed]
22. Ethier MF, Madison JM. LY294002, but not wortmannin, increases intracellular calcium and inhibits calcium transients in bovine and human airway smooth muscle cells. Cell Calcium 2002;32:31–38. [PubMed]
23. Van Galen PJM, van Bergen AH, Gallo-Rodriquez C, Melman N, Olah ME, Ijzerman AP, Stiles GL, Jacobson KA. A binding site model and structure-activity relationships for rat A3 adenosine receptor. Mol Pharmacol 1994;45:1101–1111. [PMC free article] [PubMed]
24. Michoud M-C, Tao FC, Pradhan AA, Martin JG. Mechanisms of the potentiation by adenosine of adenosine triphosphate-induced calcium release in tracheal smooth-muscle cells. Am J Respir Cell Mol Biol 1999;21:30–36. [PubMed]
25. Feoktistov I, Biaggioni I. Adenosine A2B receptors. Pharmacol Rev 1997;49:381–402. [PubMed]
26. Dickenson JM, Hill SJ. Involvement of βγ subunits in coupling the adenosine A1 receptor to phospholipase C in transfected CHO cells. Eur J Pharmacol 1998;355:85–93. [PubMed]
27. Tomura H, Itoh H, Sho K, Sato K, Nagao M, Ui M, Kondo Y, Okajima F. βγ subunits of pertussis toxin-sensitive G proteins mediate A1 adenosine receptor agonist-induced activation of phospholipase C in collaboration with thyrotropin. J Biol Chem 1997;272:23130–23137. [PubMed]
28. Exton JH. Regulation of phosphoinositide phospholipases by hormones, neurotransmitters, and other agonists linked to G proteins. Annu Rev Pharmacol Toxicol 1996;36:481–509. [PubMed]
29. Rhee SG, Bae YS. Regulation of phosphoinositide-specific phospholipase C isozymes. J Biol Chem 1997;272:15045–15048. [PubMed]
30. Michoud M-C, Tolloczko B, Martin JG. Effects of purine nucleotides and nucleoside on cytosolic calcium levels in rat tracheal smooth muscle cells. Am J Respir Cell Mol Biol 1997;16:199–205. [PubMed]
31. Ali S, Mustafa SJ, Metzger WJ. Adenosine-induced bronchoconstriction and contraction of airway smooth muscle from allergic rabbits with late phase airway obstruction: evidence for an inducible A1 receptor. J Pharmacol Exp Ther 1994;268:1328–1334. [PubMed]
32. Green A. Adenosine receptor down-regulation and insulin resistance following prolonged incubation of adipocytes with an A1 adenosine receptor agonist. J Biol Chem 1987;262:15702–15707. [PubMed]
33. Challiss RA, Richards SJ, Budohoski L. Characterization of the adenosine receptor modulating insulin action in rat skeletal muscle. Eur J Pharmacol 1992;226:121–128. [PubMed]

Articles from American Journal of Respiratory Cell and Molecular Biology are provided here courtesy of American Thoracic Society