INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

INHALATION OF COTTON MILL dust by some textile workers results in the development of the occupational lung disease byssinosis, a disease characterized by symptoms of dyspnea, coughing, wheezing, and chest tightness that begin several hours after exposure to mill dust (1). These symptoms are accompanied by an across-shift reduction in lung ventilatory capacity due to reversible bronchoconstriction and are worse on Mondays or after prolonged absences from the mill (1, 22, 23, 27). Although the etiology of byssinosis is not known, endotoxin and tannin, isolated from cotton bracts, have been implicated as important etiologic agents (20, 21).

Endotoxin has no direct effect on the airway epithelium (8). In contrast, tannin inhibits net chloride (Cl) secretion in the airway epithelium (8). This inhibition demonstrates specificity for the apical membrane and is dose dependent and reversible. Signal transduction pathways involved in Cl secretion in the airway epithelium and affected by tannin include protein kinase C activity, nonmetabolized arachidonic acid release, and intracellular calcium release (5, 6). In addition, tannin has significant effects on the adenylyl cyclase--adrenergic receptor pathway of Cl secretion. Increasing concentrations of tannin inhibit basal and epinephrine-stimulated adenosine 3',5'-cyclic monophosphate (cAMP) accumulation in part by decreasing maximum binding (B_max) without affecting the dissociation constant (K_d) (7). In addition, when the -adrenergic receptor is bypassed by forskolin, tannin noncompetitively and reversibly inhibits forskolin-stimulated adenylyl cyclase activity in a dose-dependent manner (4). Thus tannin has profound effects on the -adrenergic receptor and on cAMP.

We have hypothesized that the unusually long tannin polymer, through an affinity for cysteine residues, alters the tertiary configuration of the -adrenergic receptor and thus affects the binding of -agonists to the receptor and the coupling between the receptor and its stimulatory G protein. To test this hypothesis, we examined the effects of changes in tannin polymer length and the effects of N-acetylcysteine, which interacts with cysteine residues, on isoproterenol-stimulated cAMP accumulation. We also examined the effect of tannin and polymer length on the coupling between the -adrenergic receptor and its stimulatory G protein. In these experiments, we demonstrate that tannin reversibly decreases cAMP levels in response to isoproterenol, that N-acetylcysteine inhibits the effects of tannin on cAMP accumulation, that tannin uncouples the -adrenergic receptor from its stimulatory G protein, and that polymer length is important for isoproterenol-stimulated cAMP accumulation and for receptor-stimulatory G protein uncoupling.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Bovine tracheae were obtained from a local slaughterhouse and placed in cold Hanks' buffered saline solution (HBSS). Cell suspensions were prepared by scoring, stripping, and cutting the bovine tracheal epithelium into small pieces using sharp dissection as previously described (4, 7). Cells were isolated by gently stirring the strips at room temperature for 2 h in 50 ml of 50% Dulbecco's modified Eagle's medium-50% Ham's F-12 medium (DMEM-F12, BioWhittaker) with 5% fetal calf serum containing dithiothreitol (5 mM; Sigma Chemical, St. Louis, MO), deoxyribonuclease I (100 mg/ml, Sigma), and 0.1% protease type XIV (Sigma). Cells were centrifuged, resuspended in media, and allowed to rest for 1 h at 37°C to remove any contaminating fibroblasts. Cells were then plated onto collagen-coated plastic culture dishes at 250,000 cells/cm² and grown in culture medium consisting of DMEM-F12 supplemented with 5% fetal calf serum and (per ml) 80 µg gentamicin, 2.5 µg Fungizone, 100 U penicillin, and 100 µg streptomycin. After 3-4 days in culture, the culture medium was replaced with HBSS containing 20 mM N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES, pH 7.4) and various combinations of different compounds as described below.

Membrane fragments were prepared using airway epithelial cells scraped from the surface of bovine tracheae and placed in DMEM-F12 overnight (13). The suspension was centrifuged, and the pellet was washed twice with HBSS. The pellet was resuspended in lysis buffer [10 mM tris(hydroxymethyl)aminomethane hydrochloride (Tris · HCl), 5 mM MgCl₂, 2 mM dithiothreitol, and 10 mM phenylmethylsulfonyl fluoride, pH 7.4] and homogenized on ice using a Dounce homogenizer. After a low-speed spin (300 g) for 8 min at 4°C, the pellet was resuspended and rehomogenized. The rehomogenized pellet was combined with the previous supernatant and centrifuged for 8 min at 4°C at 2,000 g. The postnuclear fraction (supernatant) was resuspended and spun for 10 min at 4°C at 45,000 g, and the resulting pellet, consisting of both apical and basolateral membrane fragments, was used as the crude membrane fragment preparation.

Binding studies were performed at 30°C for 30 min using [³H]dihydroalprenolol ([³H]DHA, 100 Ci/mmol, NEN) at concentrations between 10¹⁰ and 10⁷ M and crude membrane fragments. Membrane fragments were separated from buffer by vacuum filtration. The pellet was washed twice with 4 ml of buffer solution at 4°C to optimally separate free and bound [³H]DHA and to reduce nonspecific binding. The radioactivity in the pellet was counted using liquid scintillation spectrophotometry (Beckman LS1801). Nonspecific binding was determined by displacement of [³H]DHA binding with 10⁶ M propranolol. Data were analyzed as a function of free ligand concentration using an iterative nonlinear curve-fitting program (Ligand) and by Scatchard and Hill analysis to derive K_d and B_max, with the latter expressed in terms of membrane protein content (15).

The effect of tannin on cell surface number was more precisely determined using [³H]CGP-12177 (38 Ci/mmol; Amersham, Arlington Heights, IL). Bovine tracheal epithelial (BTE) cells in culture (~4 × 10⁵ cells) were exposed to 5-25 µg/ml tannin for 4 h and then incubated in 1 ml of DMEM containing 25 mM HEPES and 30 µg/ml bovine serum albumin (pH 7.4, Sigma) at 4°C for 3 h in the presence of a saturating concentration (1 nM) of [³H]CGP-12177. Cell surface receptor density was calculated from one-point analysis in which 10⁶ M propranolol was used to assess nonspecific binding. Results from tannin experiments were compared with similar experiments using isoproterenol (10⁵ M for 3 h), which is known to cause a rapid decrease in cell surface receptor number (18).

In other experiments, membrane fragments (~500 mg of protein) were diluted in a 50 mM Tris · HCl (pH 7.5)-120 mM NaCl-5 mM KCl-3 mM MgCl₂ (binding buffer) solution and incubated with [³H]DHA (2.5 nM) and 16 concentrations of ()-isoproterenol at 30°C for 30 min in the presence or absence of the nonhydrolyzable GTP analog Gpp(NH)p (5'-guanylylimidodiphosphate; 50 mM). The binding incubations were terminated by the addition of 4 ml of ice-cold buffer solution and poured over Whatman GF/C glass fiber filters under vacuum. The filters were washed once with 4 ml of cold buffer solution and counted in a liquid scintillation counter.

cAMP was measured using a radioimmunoassay kit (Amersham). The cells were then treated with 1 N NaOH to dissolve cellular protein, which was measured according to the method of Lowry et al. (15), using bovine serum albumin as the standard. cAMP levels were calculated as picomoles cAMP per milligram protein.

In some experiments, cells in culture were incubated for 6 h at 37°C with either 10 or 30 mM N-acetylcysteine (Sigma). The N-acetylcysteine was dissolved in cell media (DMEM-F12), and the pH was adjusted with NaOH before addition to the culture well. cAMP levels were measured under various conditions in the presence and absence of tannin.

Condensed tannins were isolated from the 1985 crop of bracts from Acala SJ-5 cotton grown in Texas utilizing sequential Amicon ultrafiltration and a modification of the procedure of Taylor et al. (26) as previously described (8). Stock solutions of the high-molecular-mass tannin [YM-10 retentate, molecular mass >10,000 Da] were prepared daily, immediately before use, by dissolving the tannin at a concentration of 19.2 mg/ml in water. This represented the tannin concentration in the cotton bract extract used in the original study of Cloutier and Rohrbach (8). Tannin concentrations are reported as micrograms per milliliter. In some experiments, the effects of stock solutions of low-molecular-mass tannin (YM-2 retentate, molecular mass 1,000-5,000 Da) on cAMP accumulation were examined. In other experiments, high-molecular-mass tannin was prepared in water and allowed to sit at room temperature for 72 h. This exposure to room air results in tannin oxidation and decreases polymer length (molecular mass 500-7,500 Da) (19).

Data were analyzed using analysis of variance and Student's t-test unless otherwise specified (2).

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Effects on cAMP accumulation. In BTE cells grown in culture, cAMP accumulation after a 10-min exposure to isoproterenol (10⁵ M) was inhibited in cells exposed to tannin (25 µg/ml) for increasing times (5 min to 8 h). Inhibition began within 10 min and reached a maximum of 52 ± 5% (mean ± SE, n = 6) at 60 min (Fig. 1). After an 8-h exposure to 25 µg/ml tannin, there was further inhibition (82 ± 9%, P < 0.001) in isoproterenol-stimulated cAMP accumulation. Inhibition was dose dependent. Five micrograms per milliliter tannin inhibited isoproterenol-stimulated cAMP accumulation by 17 ± 4% after 10 min and by 33 ± 9% (n = 6, P < 0.001) after 8 h of exposure. The inhibition of isoproterenol-stimulated cAMP accumulation was reversible after 8 h of exposure to 25 µg/ml tannin. Reversibility began within 4 h and approached baseline by 24 h.

View larger version (12K): [in this window] [in a new window]   Fig. 1.   Effect of time (5 min to 8 h) of exposure to 25 µg/ml tannin on cAMP accumulation after a 10-min exposure to isoproterenol (105 M). Tannin was removed after 8 h, cells were extensively washed, and recovery was monitored for 24 h. Data are expressed as a percentage of control cAMP accumulation after exposure to isoproterenol in absence of tannin. * Significantly different from control value, P < 0.001.

View this table: [in this window] [in a new window]   Table 1.   Effect of low-molecular-mass tannin on isoproterenol (106 M)-stimulated cAMP accumulation

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Effect of N-acetylcysteine [NAC; 10 (A) or 30 (B) mM for 6 h] on isoproterenol (Iso)-stimulated (105 M) cAMP accumulation in presence and absence of tannin (25 µg/ml). Iso alone increased cAMP levels compared with control levels (P < 0.001), but there was no difference in degree of stimulation of cAMP between Iso alone and Iso + NAC. Tannin inhibition of Iso-stimulated cAMP was blocked in cells pretreated with both concentrations of NAC. NS, not significant.

Effects on -adrenergic receptor. -Adrenergic receptor density was determined using saturation binding with [³H]DHA and BTE membrane fragments at 30°C. In preliminary experiments, we demonstrated that binding equilibrium occurred by 30 min and that between 0.1 and 5 nM DHA, specific binding was >75% of total binding. In control membrane fragments, the K_d for [³H]DHA was 0.41 ± 0.03 nM, with a B_max of 252 ± 32 fmol/mg protein (n = 3). In membranes incubated with tannin (25 µg/ml for 30 min), there was no change in K_d (0.26 ± 0.06 nM), whereas B_max decreased to 162 ± 36 fmol/mg protein (P < 0.05), a 36% decrease in receptor number (Fig. 3). Using [³H]CGP-12177, we determined that cell surface receptor density decreased 55% after tannin (25 µg/ml for 4 h) as shown in Table 2. The decrease in cell surface receptor density was tannin dose dependent. Isoproterenol (10⁵ M) exposure produced a 72% decrease in cell surface receptor density after a 3-h exposure.

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Scatchard analysis of [3H]dihydroalprenolol ([3H]DHA) binding in control () bovine tracheal epithelial (BTE) cell membranes and in BTE membranes pretreated with tannin (25 µg/ml; ). Dissociation constant was unchanged at 0.41 ± 0.03 vs. 0.26 ± 0.06 nM (control vs. tannin pretreatment, respectively), whereas maximum binding decreased from 252 ± 32 to 161 ± 36 fmol/mg protein (n = 3, P < 0.05).

10

3

View larger version (18K): [in this window] [in a new window]   Fig. 4.   Effect of Gpp(NH)p on DHA displacement by Iso. Experiments were performed in absence (A) and presence (B) of tannin (25 µg/ml). In absence of tannin, 2 affinity sites were noted and Gpp(NH)p shifted curve to right, compatible with an increase in low-affinity sites. Tannin alone shifted curve to right. Gpp(NH)p had no further effect on shape or position of displacement curve in presence of tannin.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

These experiments demonstrate that tannin blunts the response to isoproterenol in bovine tracheal airway epithelial cells and that polymer length and cysteine residues appear to be important for these effects. Tannin also decreases -adrenergic receptor density and uncouples the receptor from its stimulatory G protein, effects due to polymer length. The tannin present in cotton mill dust is a polymer of monoflavanoid subunits (19). Compared with other plant tannins, cotton bract tannin has a longer polymer length (~9.4 monomer units) and an unusual monoflavanoid composition (procyanidin and prodelphinidin in a ratio of 2:3) (3). Cyanidins react strongly with cysteine residues (11), and of the 15 cysteine residues present in the -adrenergic receptor, 4 are in the putative extracellular domain (14). These four cysteines form disulfide bonds, which provide stability to the receptor, result in high-affinity ligand binding (17), contribute to the pharmacological specificity of agonists and antagonists, promote coupling of the -adrenergic receptor to stimulatory G proteins, and play a major role in agonist-induced stimulation of adenylyl cyclase (14, 17, 25). We have hypothesized that cotton bract tannin interacts with extracellular cysteine residues and that this interaction affects ligand binding, uncouples the receptor from its stimulatory G protein, and inhibits cAMP accumulation. Results from experiments in which polymer length was changed either by oxidation or through ultrafiltration and results from experiments with N-acetylcysteine support this hypothesis.

N-acetylcysteine had no effect on basal cAMP accumulation and no effect on isoproterenol-stimulated cAMP accumulation. The increase in basal cAMP levels approached but did not attain statistical significance. Higher concentrations of N-acetylcysteine were not examined. Both the low and the high N-acetylcysteine concentrations were equally effective in inhibiting the effects of tannin, irrespective of their ability to stimulate basal cAMP accumulation. The mechanism for this inhibition is not known, although it is possible that the interaction of N-acetylcysteine with cysteine residues blocks the interaction of tannin with these same cysteine residues. N-acetylcysteine has been shown to activate a non-cystic fibrosis transmembrane conductance regulator Cl conductance in airway epithelial cells (12), but the mechanism is unknown. Data from our laboratory suggest that N-acetylcysteine at low concentrations (10 mM), however, has no effect on Cl secretion (unpublished data); this is compatible with our current data.

Tannin also decreased -adrenergic receptor number. DHA measures both intracellular and surface receptors, whereas CGP-12177 measures surface receptors only. DHA binding studies suggested a decrease in -adrenergic receptors after tannin exposure. This decrease is secondary, at least in part, to a decrease in cell surface receptor density, which is dose dependent. The effect of tannin on DHA binding has been previously observed using epinephrine and whole cell airway epithelial cell preparations, and studies revealed a 38% decrease in receptor binding (7). Receptor number was, however, considerably higher, probably secondary to the decreased specificity of epinephrine to nonreceptor acceptor sites and to oxidation and enzyme metabolism (25).

Epithelial cells release relaxing and contracting factors that play a role in modulating airway smooth muscle tone (16). Through its effects on the -adrenergic receptor, tannin could alter the balance between relaxing and contracting factors and contribute to the across-shift Monday changes in pulmonary function that occur with acute exposure to cotton dust. The diminishing symptoms that occur during the remaining weekdays are compatible with rapid desensitization of -adrenergic receptors with -adrenergic receptor-stimulatory G protein uncoupling. Partial recovery of receptor number and resensitization during weeknights away from the mill could occur and contribute to some continued symptoms with daily exposure. Baseline pulmonary function would not be altered, but the responsiveness with repeated exposure would be affected. A state of airway hyporesponsiveness after repeated exposure to cotton dust has been observed in mill workers with byssinosis (10). Such a mechanism of rapid desensitization has also been proposed for patients with asthma taking -adrenergic drugs, except in asthma the stimulus is rarely continuous (9). Tannin exposure, however, is one of the few instances in which exposures are repetitive and frequent and occur over long periods of time.

In summary, tannin desensitizes the airway epithelium to -agonists, decreases cell surface -adrenergic receptor number, and uncouples the -adrenergic receptor from its stimulatory G protein. These effects are related to polymer length and possibly to the interaction of tannin with cysteine residues. These effects result in Cl secretion inhibition, which could alter secondary water transport in the airway, mucus secretion, and mucociliary transport. These effects could result in the pathological pulmonary findings in patients with byssinosis. Decreases in mucociliary transport would also result in secretion retention and exposure of the epithelium to other substances in cotton dust, such as endotoxin, for longer than normal periods of time. Thus tannin may directly and indirectly contribute to the pathogenesis of byssinosis.