Article Text


Tannic acid in plant dust causes airway obstruction
  1. D Taubert,
  2. G Grimberg,
  3. E Schömig
  1. Department of Pharmacology, Medical Hospital of the University of Cologne, D-50931 Cologne, Germany
  1. Correspondence to:
    Dr D Taubert
    Department of Pharmacology, Medical Hospital of the University of Cologne, Gleueler Str 24, D-50931 Cologne, Germany;

Statistics from

Occupational or environmental exposure to plant dusts has been shown to increase the risk of obstructive lung diseases, primarily by non-immunological activity.1 However, the causative agents and the underlying mechanisms have not been established. We have recently suggested that the polyphenolic fraction of hydrolysable tannins in plant derived dusts, with tannic acid (TA) as the main constituent, may contribute to airway constriction.2 Here we present experimental evidence that TA causes acute airway obstruction by non-competitive inhibition of the constitutive endothelial isoform of nitric oxide synthase (eNOS) in the tracheobronchial epithelium, which is reported to provoke airway hyperresponsiveness and bronchoconstriction.3

Organ bath experiments were performed using the trachea and main bronchi of non-sensitised guinea pigs. The tracheobronchial tree was dissected out of CO2 sacrificed guinea pigs of either sex weighing 300–450 g and cut into rings of 3–4 cartilage segments wide. Isometric contractions were recorded as described previously.4 Briefly, individual rings were mounted in organ baths containing 10 ml carbogen aerated Tyrode solution (pH 7.4, 37°C), kept at a preload of 25 mN, left to equilibrate for 60 minutes, and precontracted by addition of 25 μmol/l prostaglandin F to 30–40% of their individual isometric maximum (100%).

NO release was determined in real time by an amperometric microsensor as described elsewhere.4 Briefly, the tracheal and bronchial rings were opened longitudinally and kept in Hepes-Krebs solution (10 ml, pH 7.4, 25°C). The sensor was placed onto the luminal surface at a distance of 200 μm. After 30 minutes of equilibration, individual NO reactivity was assessed by addition of 15 nmol/l substance P.

Tannic acid (penta-o-digalloyl-β-d glucose; Fluka, Seelze, Germany) produced an immediate concentration dependent contraction of the tracheobronchial rings (lasting 30–60 minutes) with a mean EC50 of 0.19 μmol/l (95% CI 0.10 to 0.35) and a maximal response (Emax) of 85.0% (95% CI 76.6 to 93.4). The threshold concentration of TA eliciting a significant contraction (p = 0.05) was 0.7 nmol/l (corresponding to 1.2 mg/m3). The contraction was completely abolished in epithelium denuded rings and by pretreatment with an unselective NOS inhibitor. It was not affected by the presence of an inhibitor of the neuronal and inducible isoforms of NOS (fig 1A), indicating that the TA mediated contraction in non-sensitised guinea pig was entirely due to inhibition of eNOS in the airway epithelium and that TA does not elicit direct effects on tracheobronchial muscle. The contractions were not blunted by addition of the NO synthase substrate l-arginine, which suggests a non-competitive eNOS blockade by TA (fig 1B). This finding also agrees with a biochemical study which reported that TA non-competitively inhibits the eNOS enzyme with a 20–30 fold higher selectivity compared with inhibition of nNOS or iNOS isoforms.5 A possible TA induced formation of spasmogenic prostanoids was ruled out because the cyclo-oxygenase inhibitor indomethacin (10 μmol/l) did not reduce vascular tone (data not shown). In keeping with these findings, the epithelial release of NO caused by receptor independent agents (l-arginine, calcium ionophore A23187) as well as receptor dependent eNOS activating agents (bradykinin) was reduced in tissue pretreated with TA (fig 1C). Moreover, the contraction response to stimulation with bradykinin (1 μmol/l) was significantly enhanced when the rings were pretreated with 0.2 μmol/l TA (the EC50 for contraction). A similar effect was achieved after removal of the epithelium, indicating that the hyperresponsiveness to bradykinin was due to a lack of concomitant bradykinin induced eNOS activation (fig 1D).

Figure 1

 Effects of tannic acid (TA) and barley flour extract on muscle tone and nitric oxide formation of guinea pig tracheobronchial rings. (A) Contraction following treatment with cumulatively increasing concentrations of TA after pre-incubation (15 min) with the non-specific nitric oxide synthase (NOS) inhibitor l-NG-monomethyl arginine (l- NMMA), after removal of the epithelium, and after pre-incubation (15 min) with the inducible NOS isoform specific inhibitor 1400 W. (B) Contraction in response to a single 10 μmol/l dose of TA compared with subsequent treatment with 100 μmol/l l-arginine. (C) NO release in response to 1 μmol/l bradykinin, 100 μmol/l l-arginine, or 10 μmol/l calcium ionophore A23187 in the absence and after pretreatment (15 min) with 0.2 μmol/l TA. (D) Contraction in response to 1 μmol/l bradykinin, after pre-incubation (15 min) with 0.2 μmol/l TA, and after removal of the epithelium. (E) Contraction following treatment with cumulatively increasing concentrations of aqueous barley flour extract (20 g/l) and after pre-incubation (15 min) with l-NMMA. Values represent mean (SE) of five independent experiments each. A p value <0.05 (two tailed t test with Holm correction for multiple comparisons) was considered statistically significant. p values in (A) and (E) indicate the nominal significance levels of TA and barley flour extract induced contractions compared with the respective contractions in the presence of l-NMMA.

We determined concentrations of total hydrolysable tannins in samples of barley flour, oak wood, and green tea dust of 8.7, 11.2 and 5.9 mg/g, respectively, by spectrophctometric dieaction after reaction with polassion iodote, as describe.6 Acute bronchoconstriction in previously unexposed subjects after challenge with grain dust concentrations above 100 mg/m3 may therefore be explained by direct exposure to contractile levels of hydrolysable tannins, while chronic obstructive respiratory symptoms associated with occupational exposure to average inhalable dust concentrations in the range of 5–20 mg/m3 may result from tannin accumulation in the airway epithelium.7 In support of the mechanism of dust induced bronchoconstriction proposed here, we found that an aqueous extract of barley flour caused contraction of guinea pig tracheobronchial rings that was blunted by pretreatment with l-NMMA (fig 1E). The threshold concentration for contraction (p = 0.05) was 58.9 μg/l extract, corresponding to 0.5 mg/m3 hydrolysable tannins.

Other plant phenols (including non-hydrolysable tannins and the hydrolysable tannin monomers gallic acid and ellagic acid) did not alter smooth muscle tone or, conversely, caused relaxation,4 indicating that acute contractions induced by plant dust are specific for the hydrolysable tannin fraction.

In contrast to our findings, Flesch et al8 reported that TA produced endothelium dependent relaxations of rat aortic rings. In our model we could also record epithelial NO release accompanied by diminished contractions, but this effect required TA concentrations >10 mmol/l above any clinical relevance (data not shown) and was probably caused by non-specific activities of TA towards the airway epithelium.

In conclusion, hydrolysable tannins may be aetiologically involved in the development of plant dust induced acute and chronic obstructive airway diseases by impairing the endogenous release of bronchoprotective NO.


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.