Regulation of interleukin-1beta and interleukin-1beta inhibitor release by human airway epithelial cells

In asthma, human airway epithelial cells (HAECs) regulate the intensity of mucosal inflammation, in part, by releasing the pro-inflammatory cytokine interleukin (IL)-1beta. However, the IL-1beta inhibitors, IL-1 receptor antagonist (IL-1RA) and soluble IL-1 receptor type II (sIL-1RII), regulate IL-1beta bioactivity. In order to better understand the control of IL-1beta activity in the airway mucosa, the role(s) of tumour necrosis factor (TNF)-alpha, cyclic adenosine monophosphate (cAMP) and cyclic guanosine monophosphate (cGMP) in the release of IL-1beta and its inhibitors by cultured HAECs were examined. HAECs were treated with TNF-alpha (2-200 ng.mL(-1)), dibutyryl cAMP (0.01-1 mM), 8-bromo-cGMP (0.01-1 mM) or vehicle for 24 h, and cytokine levels in the HAEC-conditioned medium were measured by enzyme-linked immunosorbent assay. HAECs produced IL-1beta, IL-1RA and sIL-1RII constitutively, but the inhibitor concentrations greatly exceeded that of IL-1beta (by approximately 100- and approximately 550-fold, respectively). TNF-alpha dose-dependently increased the levels of all IL-1beta cytokine family members. However, over the range of TNF-alpha concentrations studied, IL-1beta concentration increased more than those of its inhibitors. cAMP increased constitutive and TNF-alpha-stimulated IL-1beta release but reduced that of sIL-1RII. In contrast, cGMP had no effect on IL-1beta but reduced IL-1RA and sIL-1RII release. Under basal conditions, the disproportionate release of inhibitors relative to interleukin-1beta by human airway epithelial cells probably prevents interleukin-1beta-mediated biological effects. Tumour necrosis factor-alpha, cyclic adenosine monophosphate and cyclic guanosine monophosphate may potentiate mucosal inflammation by increasing interleukin-1beta levels relative to those of its inhibitors in the airway mucosa.

Nuclear factor-kappa B augments beta(2)-adrenergic receptor expression in human airway epithelial cells

Interleukin (IL)-1 beta increases beta(2)-adrenergic receptor (beta(2)-AR) mRNA and density by protein kinase C (PKC)-dependent mechanisms in human airway epithelial cells. The present study examined the role of several nuclear transcription factors in the PKC-activated upregulation of beta(2)-AR expression. BEAS-2B cells were exposed to the PKC activator phorbol 12-myristate 13-acetate (PMA; 0.1 microM for 2-18 h). PMA had no effect on activator protein (AP)-2 or cAMP response element binding protein DNA binding activity but markedly increased nuclear factor (NF)-kappa B and AP-1 binding as assessed by electrophoretic gel mobility shift assay. PMA also increased the activity of a beta(2)-AR promoter-luciferase reporter construct in transiently transfected cells. These effects were inhibited by the PKC inhibitors Ro-31-8220 and calphostin C. Furthermore, with increasing Ro-31-8220, beta(2)-AR promoter-reporter activity correlated closely with both NF-kappa B and AP-1 activities (r > 0.89 for both). Finally, the selective NF-kappa B inhibitor MG-132 dose dependently reduced NF-kappa B binding and beta(2)-AR promoter activity but increased AP-1 binding. We conclude that PKC-induced upregulation of beta(2)-AR expression in human airway epithelial cells appears to be mediated, at least in part, by increases in NF-kappa B activity.

IL-1beta enhances beta2-adrenergic receptor expression in human airway epithelial cells by activating PKC

Protein kinase C (PKC)-activated signal transduction pathways regulate cell growth and differentiation in many cell types. We have observed that interleukin (IL)-1beta upregulates beta2-adrenergic receptor (beta2-AR) density and beta2-AR mRNA in human airway epithelial cells (e.g., BEAS-2B). We therefore tested the hypothesis that PKC-activated pathways mediate IL-1beta-induced beta-AR upregulation. The role of PKC was assessed from the effects of 1) the PKC activator phorbol 12-myristate 13-acetate (PMA) on beta-AR density, 2) selective PKC inhibitors (calphostin C and Ro-31-8220) on beta-AR density, and 3) IL-1beta treatment on the cellular distribution of PKC isozymes. Recombinant human IL-1beta (0.2 nM for 18 h) increased beta-AR density to 213% of control values (P < 0.001). PMA (1 microM for 18 h) increased beta-AR density to 225% of control values (P < 0.005), whereas Ro-31-8220 and calphostin C inhibited the IL-1beta-induced upregulation of beta-AR in dose-dependent fashion. PKC isozymes detected by Western blotting included alpha, betaII, epsilon, mu, zeta, and lambda/iota. IL-1beta increased PKC-mu immunoreactivity in the membrane fraction and had no effect on the distribution of the other PKC isozymes identified. These data indicate that IL-1beta-induced beta-AR upregulation is mimicked by PKC activators and blocked by PKC inhibitors and appears to involve selective activation of the PKC-mu isozyme. We conclude that signal transduction pathways activated by PKC-mu upregulate beta2-AR expression in human airway epithelial cells.

Chronic effects of inhaled albuterol on beta-adrenoceptor system function in human respiratory cells

The in vivo effects of beta-adrenergic receptor (betaAR) agonists given chronically by metered-dose inhaler (MDI) on the molecular components of the beta-adrenoceptor system expressed by human respiratory cells are poorly understood. This study examined the effects of inhaled albuterol (180 microg four times daily for 7 days) on betaAR function of airway epithelial cells (AECs) and alveolar macrophages (AMs) freshly isolated from 10 normal subjects. Responses were related to beta2AR genotype in codons 16 and 27, regions which affect chronic responses to beta2-agonists. In AEC, betaAR density and adenosine cyclic 3',5'-phosphate (cAMP) production in response to isoproterenol (ISO) were significantly lower in the albuterol versus placebo treatment arm (p < 0.01 for both). Moreover, in AEC, albuterol treatment increased betaAR-kinase (betaARK) protein immunoreactivity. In contrast, in AM, albuterol tended to decrease betaAR density and cAMP production but changes did not achieve statistical significance (p > 0.20 for both) and had no effect on betaARK immunoreactivity. Changes in betaAR density occurred in all subjects but tended to be greater in subjects with the glycine 16 genotype. In cultured cells exposed to equal concentrations of beta-agonist in vitro, the magnitude of betaAR down-regulation (p < 0.05) and cAMP densensitization (p < 0.05) was greater in AEC than AM. These results indicate that albuterol taken by inhalation in a therapeutically relevant dose for 1 week produces betaAR down-regulation, densensitizes the cAMP response of airway epithelial cells to a beta2-adrenergic agonist, and increases betaARK immunoreactivity. Greater densensitization of AEC than AM in response to chronic albuterol inhalation likely reflects cell type-specific responses.

Effects of topical corticosteroids on inflammatory mediator-induced eicosanoid release by human airway epithelial cells

BACKGROUND

Airway epithelial cells are among the first cells to come in contact with aerosolized corticosteroids. However, the relative potencies and time course of action of the several commonly used aerosolized corticosteroids on eicosanoid production by airway epithelial cells are unknown.

OBJECTIVES

This study compared the effects of fluticasone, budesonide, and triamcinolone on eicosanoid output by human airway epithelial cells in vitro. We also determined the spectrum of eicosanoids affected and the mechanism for corticosteroid action.

METHODS

Cultured BEAS-2B airway epithelial cells (a transformed cell line) were exposed to corticosteroids (1 nmol/L to 1 micromol/L) for 2 to 48 hours and then assayed for basal- and bradykinin (BK)-stimulated eicosanoid output. The eicosanoid profile was identified by HPLC in tritiated arachidonic acid prelabelled cells, and PGE2, the major eicosanoid product, was quantitated by RIA. The effect of corticosteroids on the immunoreactivity of key proteins involved in eicosanoid metabolism (ie, cyclooxygenase [COX], phospholipase A2 [PLA2], and Clara cell protein, a PLA2 inhibitor) was determined by Western blotting.

RESULTS

Eicosanoid output was largely confined to prostaglandins with values of 5 +/- 2 and 82 +/- 35 ng PGE2/10(6) cells for basal- and BK stimulation, respectively (n = 8). All 3 corticosteroids inhibited basal- and BK-induced PGE2 output in a dose- and time-dependent manner. Fluticasone and budesonide completely eliminated PGE2 output in nanomolar concentrations in contrast to triamcinolone, which required micromolar concentration. The rank order of potency was: fluticasone = budesonide > triamcinolone. The time course of action for PGE2 inhibition also differed, with budesonide acting more slowly than the other 2 corticosteroids (P = .04). All 3 corticosteroids markedly reduced COX2 with little effect on COX1, cPLA2 (Type IV), or iPLA2 (Type VI) immunoreactivity or their relative distribution in cytosol versus membrane fractions. Clara cell protein immunoreactivity was undetectable in control and corticosteroid-treated cell lysates.

CONCLUSION

These results show that in a human airway epithelial cell line, the 3 inhaled corticosteroids commonly used to treat asthma differ in onsets of action as inhibitors of prostaglandin synthesis and vary considerably in potency. All 3 corticosteroids act mechanistically in similar fashion by inhibiting COX2 synthesis.

IL-1 beta alters beta-adrenergic receptor adenylyl cyclase system function in human airway epithelial cells

Inflammatory cells release a variety of cytokines, including interleukin (IL)-1 beta, into the airway in asthma. This study examined the effects of human IL-1 beta on the function of the beta-adrenergic receptor (beta AR)-adenylyl cyclase (AC) system in BEAS-2B cells, a human airway epithelial cell line. IL-1 beta markedly increased beta AR density (Bmax; P < 0.001) primarily by increasing the percentage of the beta 2AR subtype (from 67 to 91%; P < 0.001). Bmax increased monotonically over time in response to 200 pM IL-1 beta and was approximately 2.5-fold greater than control cells between 36 and 42 h. In contrast, the concentration response of Bmax to IL-1 beta given for 18 h was biphasic. Bmax increased with IL-1 beta concentrations from 2 to 200 pM, but, at > 200 pM, it decreased progressively toward control values. IL-1 beta-induced increases in Bmax with IL-1 beta were associated with approximately threefold increases in beta 2 AR mRNA and were blocked by the protein synthesis inhibitor cycloheximide. Despite the marked increase in Bmax, however, IL-1 beta depressed adenosine 3',5'-cyclic monophosphate (cAMP) responses to isoproterenol and forskolin, a direct activator of AC (P < 0.001 by analysis of variance for both). The inhibitory effect of IL-1 beta on cAMP production appeared to be explained by increases in the activity of an inhibitory GTP binding protein because IL-1 beta treatment increased the activity of a pertussis toxin ADP-ribosylated Gi alpha protein by approximately 2.5-fold; and pretreatment of intact cells with pertussis toxin inhibited the effect of IL-1 beta on cAMP production. These data indicate that IL-1 beta-mediated changes in the beta AR-AC system function in airway epithelial cells are complex and involve expression of receptor protein, GTP binding protein, and possibly AC itself. Increases in IL-1 beta may contribute to abnormalities in airway function in subjects with asthma.

Chronic effects of catecholamines on the beta 2-adrenoreceptor system in cultured human airway epithelial cells

Chronic catecholamine treatment induces beta-adrenergic receptor (beta AR) downregulation, i.e., a loss of total cell receptors. In the human respiratory tract, the mechanism(s) underlying beta AR downregulation remains poorly understood. The present study, therefore, examined the effects of 24 h of exposure to isoproterenol (Iso; 10 nM or 1 microM) on beta AR density and the rate of beta AR degradation, steady-state beta 2-adrenergic receptor (beta 2 AR) mRNA levels, and the content of Gs alpha and Gi alpha proteins in cultured human bronchial epithelial cells (i.e., the BEAS-2B cell line). beta AR density assessed by binding with [125I]iodopindolol decreased in a dose-dependent fashion with 24 h of Iso exposure. With Iso (1 microM), beta AR density decreased by approximately 82%. In contrast, forskolin (100 microM) and dibutyryl adenosine 3',5'-cyclic monophosphate (1 mM), agents that also increase adenosine 3',5'-cyclic monophosphate (cAMP) levels, had no significant effect on beta AR density. Iso exposure also elicited a concomitant decrease in Iso-stimulated cAMP but had no significant effect on the content of the G proteins G alpha i2 and Gs alpha assessed by immunoblotting and toxin-catalyzed ADP ribosylation. Of note, Iso exposure (1 microM) had no effect on steady-state levels of beta 2 AR mRNA measured both by Northern analysis and by reverse transcriptase-polymerase chain reaction. However, beta AR half-life assessed in the presence of the protein synthesis inhibitor cycloheximide was reduced by approximately 60% in Iso-treated cells (i.e., from 37 h in control to 16 h in 1 microM Iso). These results suggest that, in human airway epithelial cells, beta 2 AR downregulation 1) is not primarily driven by intracellular cAMP levels, 2) is not associated with significant decreases in steady-state levels of beta 2 AR mRNA, and 3) is largely posttranslationally regulated by increases in the rate of receptor protein degradation.

Eicosanoid production in rabbit tracheal epithelium by adenine nucleotides: mediation by P2-purinoceptors

Adenosine triphosphate (ATP) acting through epithelial nucleotide receptors exerts multiple physiologic actions on airway mucociliary clearance and caliber. However, the effect of ATP on arachidonate metabolism in the airway remains unknown. In this study, the ability of ATP to regulate eicosanoid production was studied in vitro in full-thickness rabbit tracheal strips and separately in rabbit epithelial explant cultures. In the freshly isolated strips, ATP increased prostaglandin E2 (PGE2) release in a dose-dependent fashion, with an activation threshold at 10 microM ATP and a 3.5-fold increase in PGE2 output at 1 mM ATP. Epithelium removal decreased 1 mM ATP-evoked PGE2 release by 68%. Reverse-phase, high-pressure liquid chromatography (HPLC) of media from 3H-arachidonic acid-incubated epithelial explants exposed to 1 mM ATP demonstrated increased output of the cyclooxygenase products PGE2 and prostaglandin F2a (PGF2a). Other identifiable eicosanoids did not increase. The concentration-response for ATP-induced PGE2 release by explants was similar to that of tracheal strips. PGE2 release by 1 mM ATP was 27% of that elicited by ionomycin (10 microM) and was markedly inhibited by indomethacin (10 microM). Purinoceptor agonist-stimulated PGE2 release by the epithelium yielded a rank order of potency of uridine triphosphate (UTP) > or = ATP > 2-methylthio-ATP (2MeSATP) >> alpha,beta-methyleneadenosine-5'-triphosphate (AMP-CPP) > or = adenosine. These results indicate that ATP, acting primarily through an epithelial P2-purinoceptor similar to the P2a subtype, stimulates eicosanoid metabolism in rabbit airway epithelium via the cyclooxygenase pathway, producing PGE2 as the predominant species.

Relaxation of rabbit tracheal smooth muscle by adenine nucleotides: mediation by P2-purinoceptors

The relaxant action of adenine nucleotides was studied in isolated rabbit trachealis to assess the presence of P2-purinoceptors in the airways, their cellular location, and pharmacologic properties. Strips of tracheal smooth muscle with intact epithelium were incubated in tissue baths and contracted with 1 microM acetylcholine. Over a dose range of 0.1 microM to 1 mM, ATP and ADP were significantly more potent than adenosine in relaxing tracheal smooth muscle. Significant relaxations were also elicited by AMP-PCP, AMP-CPP, and AMP-PNP, three ATP analogs stable to enzymatic hydrolysis to adenosine. In the absence of acetylcholine, neither ATP nor AMP-CPP exerted any contractile effect on the tracheal strips. In tissues selectively denuded of epithelium, ATP-, ADP-, and AMP-PCP-induced relaxations were markedly reduced. ATP-induced relaxation was also inhibited by the P2y-purinoceptor antagonist Reactive Blue 2 (RB2) (50 to 300 microM) and partially reduced by the cyclooxygenase inhibitor indomethacin (10 microM), whereas adenosine-induced relaxation was not significantly affected by these agents. These results suggest that ATP can induce smooth muscle relaxation in acetylcholine-contracted tracheal strips through a distinct P2-purinoceptor. This receptor appears to be located on the epithelium where its relaxant effect is mediated in part by release of one or more cyclooxygenase products. Additional relaxation at high ATP concentrations may occur through enzymatic hydrolysis of ATP to adenosine and interaction at P1-purinoceptors.

Mediation of bradykinin-induced contraction in canine veins via thromboxane/prostaglandin endoperoxide receptor activation

1. Canine jugular and femoral veins were studied to determine the possible importance of thromboxane (TXA2) and prostaglandin endoperoxides (prostaglandin H2, PGH2) in mediating bradykinin(BK)-induced contraction. 2. Isolated vein rings incubated in modified Krebs solution contracted to TXA2/PGH2 analogs SQ26655 and U44069 with potency of contraction exceeding that for BK. The potency ranking for both veins was SQ26655 greater than U44069 greater than BK greater than PGF2 alpha greater than TXB2 much greater than PGD2. 3. The cyclo-oxygenase inhibitors indomethacin (3 x 10(-7) M) and flufenamic acid (10(-5) M) reduced BK contractions without affecting those induced by noradrenaline (NA). 4. TXA2/PGH2 receptor antagonists SQ29548 (10(-8) M) and BM13177 (10(-6) M) strongly inhibited BK-induced tension. The action of antagonists was reversible with negligible influence on NA-elicited contraction. Selective removal of endothelium had no effect on BK-induced contraction or the action of the antagonists. 5. The thromboxane synthase inhibitors dazoxiben (10(-4) M) and CGS 12970 (10(-5) M) had no significant inhibitory effect on BK-induced tension. 6. These results suggest that in canine jugular and femoral vein, the action of BK is largely dependent upon stimulation of the cyclo-oxygenase pathway to produce PGH2 and possibly TXA2, which can activate a smooth muscle TXA2/PGH2 receptor to elicit vasoconstriction.

Myosin light chain phosphorylation and evidence for latchbridge formation in norepinephrine stimulated canine veins

The role of 20000 dalton myosin light chain phosphorylation in mediating venous smooth muscle contraction was studied in isolated preparations of canine jugular and femoral vein. One min 10(-5) M norepinephrine-induced contraction was accompanied by significant increases in phosphorylation (jugular - 21 to 46%; femoral - 19 to 54%) which were reversed within 10 min after agonist washout. During 40 min stimulation, phosphorylation and isometric force redevelopment rates declined to near basal levels while force was maintained. These findings implicate light chain phosphorylation as a prerequisite for initial tension development by crossbridge cycling in venous smooth muscle. However, long term tension can be maintained through a process similar to the latchbridge state in tracheal and arterial smooth muscle.

Ca2+, cAMP, and changes in myosin phosphorylation during contraction of smooth muscle

Phosphorylation of myosin increases rapidly upon stimulation of an arterial smooth muscle. However, peak values are not maintained and phosphorylation declines, while active stress increases monotonically to a sustained steady state. The aim of this study was to determine the reason(s) for the transient change in myosin phosphorylation. Four hypotheses were considered: 1) reduced substrate, i.e., ATP depletion, 2) altered access of either the myosin kinase or phosphatase to the cross bridge, 3) reduced myosin kinase activity secondary to its phosphorylation by adenosine 3',5'-cyclic monophosphate-dependent protein kinase, and 4) reduced myoplasmic [Ca2+] during the contraction. Our results suggest that the most likely explanation is that there are two Ca2+-dependent regulatory processes: 1) myosin phosphorylation and 2) a second, unidentified site allowing stress maintenance with reduced cross-bridge cycling rates. A higher cell Ca2+ concentration appears to be necessary to activate myosin kinase and produce myosin phosphorylation than is needed for force maintenance. We suggest that agonist-induced Ca2+ transients, coupled with the differential Ca2+ sensitivity of the two regulatory systems, may explain the observed transient in myosin phosphorylation during a maintained contraction in smooth muscle.

The role of myosin light chain phosphorylation in regulation of the cross-bridge cycle

Ca2+ binding to myofibrillar regulatory sites can produce conformational changes allowing cross-bridge attachment and cycling. Measurements of smooth muscle actomyosin ATPase activity suggested that Ca2+ might act indirectly to mediate cross-bridge attachment by stimulating myosin light chain phosphorylation. However, the predicted obligatory relationship between developed force and myosin phosphorylation was not always observed in living smooth muscle. The observation that myosin phosphorylation was always tightly correlated with average cross-bridge cycling rates estimated from isotonic shortening velocities suggested that Ca2+ has two regulatory roles. One action is exerted via a Ca2+-binding protein whose identity is unknown in smooth muscle. This regulatory site acts like other Ca2+-binding regulatory proteins in muscle to permit cross-bridge interaction and to determine active stress. The second regulatory role involves stimulation of myosin light chain kinase and light chain phosphorylation. Increasing the level of phosphorylated cross-bridges increases shortening velocities or rate of force development. We suggest that the dephosphorylated cross-bridges are noncycling or slowly cycling in activated smooth muscle. Smooth muscle may be a particularly favorable experimental preparation for demonstrating a general regulatory role of myosin phosphorylation in modulating the kinetics and energetics of muscle contraction.

Role of Ca2+ and myosin light chain phosphorylation in regulation of smooth muscle

The time course of phosphorylation of the 20,000-dalton myosin light chain (LC 20) was determined during contraction and relaxation in K+- and histamine-stimulated medial strips of swine carotid arteries. Resting LC 20 phosphorylation levels of 0.15 mol P/mol LC 20 rapidly increased to peak values of 0.6-0.7 mol P/mol LC 20 after stimulation and then declined significantly, although stress continued to rise to a stable steady-state maximum. LC 20 dephosphorylation after agonist washout preceded the decline in isometric stress. Over the entire contraction-relaxation cycle, phosphorylation was correlated with shortening velocity and not with developed stress. The maximum shortening velocity with no external load (Vo) was directly proportional to LC 20 phosphorylation (r = 0.986). The data indicate that LC 20 phosphorylation is necessary for cross-bridge cycling leading to shortening or stress development but that stress can be maintained by additional mechanisms. We suggest that dephosphorylation of an attached cross bridge in the presence of Ca2+ arrests the cycle, forming an attached, noncycling cross bridge.

Myosin light chain phosphorylation associated with contraction in arterial smooth muscle

The hypothesis that Ca2+ initiates contraction in smooth muscle by activating an endogenous myosin light chain kinase (MLCK) that phosphorylates the 20,000 dalton light chain (LC 20) of myosin was tested in tissues prepared from the media of swine carotid arteries. Unstimulated tissues with low levels of tone exhibited low levels of phosphorylated LC 20. On stimulation with a high-K+ physiological salt solution containing 1.6 mM CaCl2, LC 20 phosphorylation increased to 0.6 mol P/mol LC 20 within 30 s. This increase preceded force development, which required 2-4 min to attain a maximum steady-state value of 3.34 +/- 0.15 (SE) X 10(5) N/m2. These results support the hypothesis, as the stimulus was submaximal for the preparation. However, LC 20 phosphorylation declined significantly from its peak value before steady-state force was attained, reaching near control levels after 10 min of stimulation. The results suggest that Ca2+-stimulated LC 20 phosphorylation is an important physiological control mechanism but that additional factors are involved in the maintenance of tonic isometric force.

Myosin phosphorylation and the cross-bridge cycle in arterial smooth muscle

Phosphorylation of the 20,000-dalton light chain of myosin is closely correlated with cross-bridge cycling in arterial smooth muscle. Evidence is presented that dephosphorylation can produce an attached, noncycling cross-bridge (latch-bridge) which is responsible for the high economy of force maintenance in this tissue.

BHK21 myosin: isolation, biochemical characterization and intracellular localization

Myosin has been isolated from baby hamster kidney cells (BHK21/C13) in high yield and characterized biochemically and immunologically. The subunit composition consists of 2 heavy chains, approximately 200,000 Daltons each, and 2 classes of light chains of approximately 16,000 and 20,000 Daltons. The myosin exhibits ATPase activity in the presence of K+-EDTA or Ca2+ but very little activity with Mg2+-ATP. The Mg2+-ATPase activity is stimulated only about 2-fold by skeletal actin, but a much larger activation is obtained in the presence of a protein kinase isolated from chicken gizzard. The increase in actin activation is accompained by the phosphorylation of the 20,000-Dalton light chain. BHK21 myosin is insoluble at low ionic strength and forms typical biopolar thick filaments. A specific antiserum generated against this protein forms a single precipitin line with the antigen but does not crossreact with either skeletal or smooth muscle myosin. The antiserum also specifically stains stress fibres in BHK21 cells as shown by indirect immunofluorescence.