Thorax 1993;48:63-67

Effect of bumetanide on toluene diisocyanateinduced contractions in guinea pig airways

Cristina E Mapp, Anna Boniotti, Alberto Papi, Carlo A Maggi, Antonino Di Stefano,Marina Saetta, Adalberto Ciaccia, Leonardo M Fabbri

AbstractBackground The loop diuretic frusem-ide has been shown to inhibit the bron-choconstrictor response to exercise,inhaled allergen, distilled water, aden-osine, and sodium metabisulphite.Toluene diisocyanate contracts smoothmuscle by activating capsaicin sensitivenerves and causes asthma that sharesmany features with allergen inducedasthma.Methods The study was designed toassess the effect of two loop diuretics,bumetanide (10 and 100 PM) andfrusemide (100 PM), on smooth musclecontraction induced by toluenediisocyanate (0'03-1000 PM) in guinea pigairways with and, in the case ofbumetan-ide, without epithelium. The effect ofbumetanide on the response to acetyl-choline, neurokinin A, and electrical fieldstimulation in guinea pig bronchialsmooth muscle rings was also examined.Results Bumetanide (10 and 100 PM)had no effect on toluene diisocyanateinduced contraction whether airwayepithelium was present or not. Frusemide(100 pM) caused no significant inhibitionof toluene diisocyanate induced contrac-tion (mean reduction on the entire curve25%). Bumetanide inhibited non-adren-ergic, non-cholinergic contractioninduced by electrical field stimulation ofbronchi pretreated with atropine (1 PM)and indomethacin (5 PM) and this inhibi-tion was inversely related to thefrequency of stimulation, suggesting thatbumetanide may be inhibiting transmit-ter release at the prejunctional level.Bumetanide and frusemide did notinhibit the responses to exogeneousacetylcholine (0 1 PM) or neurokinin A(1 nM).Conclusions Bumetanide and frusemidein doses that are known to inhibit non-adrenergic, non-cholinergic contractiondue to electrical field stimulation failed toinhibit the response to toluene diisocyan-ate in guinea pig airways.

(Thorax 1993;48:63-67)

Bumetanide and frusemide are high ceilingdiuretics and both inhibit the Na/K/Cl cotran-sporter in renal tubules. Bumetanide is morepotent than frusemide but the two compoundshave a similar effect on Na/K/Cl cotransport in

various mammalian tissues.' 2Frusemide, when administered by nebuliser

but not orally, has been shown to protectasthmatic patients against non-specific andspecific bronchoconstrictor stimuli.' Themechanism of action of frusemide in vivo hasnot been determined. Because frusemide butnot bumetanide inhibits adenosine 5'-mono-phosphate and sodium metabisuphite inducedbronchoconstriction in asthmatic persons, itwas recently suggested that the protective effectof frusemide may be independent of the Na/K/Cl cotransport.9 The effect of high ceilingdiuretics such as frusemide and bumetanidecould be due to inhibition of the Na/K/Clcotransport in airway smooth muscle, thusreducing contractility,'0 or it could be due toinhibition of chloride secretion into the bron-chial lumen by blocking Na/Cl cotransport onthe basolateral membrane of epithelial cells." 12Other possible mechanisms include inhibitionof epithelial cell swelling and oedema inducedby osmotic stimuli,'3 inhibition of mediatorrelease from inflammatory cells,'4 15 and an effecton the responsiveness of sensory nerves."There is evidence to suggest that loop

diuretics inhibit both cholinergic andexcitatory non-adrenergic, non-cholinergicneurotransmission in guinea pig airwayswithout a direct effect on airways smooth mus-cle. This effect may be due to inhibition of theNa/K/Cl cotransport. As we have shown thattoluene diisocyanate contracts smooth muscleby activating capsaicin sensitive sensorynerves'M20 we have studied the effects ofbumetanide and frusemide on the contractionof guinea pigs airway smooth muscle in vitroinduced by toluene diisocyanate. Toluenediisocyanate, a widely used industrial chemical,is a potent airway sensitiser and able to induceoccupational asthma.2' We also examined theeffects ofbumetanide on airway smooth musclecontraction in guinea pig bronchial rings invitro induced by acetylcholine, neurokinin A,and electrical field stimulation.

MethodsPREPARATION OF TISSUEThe experimental procedure and specificprotocols were approved by the committee onanimal care of the University of Padova.Male Hartley outbred guinea pigs (Rodentia

Laboratories, Torre Pallavicina, Bergamo)weighing 300-400 g were anaesthetised withpentobarbital sodium (50 mg/kg intraperi-toneally). The lungs were removed rapidly andimmersed in oxygenated Krebs-Henseleitsolution containing 118 3 mmol/l NaCl,

Institute ofOccupationalMedicine, Universityof Padova, Padova,ItalyCristina E MappA Di StefanoM SaettaPharmacologyDepartment,A MenariniPharmaceuticals,Firenze, ItalyC A MaggiInstitute of Infectiousand PulmonaryDiseases, Section ofExperimentalPathophysiology,University of Ferrara,ItalyA BoniottiA PapiA CiacciaL M FabbriReprint requests to:Dr Cristina E Mapp,Istituto di Medicina delLavoro,Universita degli studi diPadova,Via J Facciolati 71,35127 Padova, Italy.Accepted 16 March 1992

63

on Novem

ber 14, 2021 by guest. Protected by copyright.

http://thorax.bmj.com

/T

horax: first published as 10.1136/thx.48.1.63 on 1 January 1993. Dow

nloaded from

Mapp, Boniotti, Papi, Maggi, Di Stefano, Saetta, Ciaccia, Fabbri

4-7 mmol/l KCl, 1-2 mmol/l MgSO4,1 2 mmol/l KH2PO4, 25-0 mmol/l NaHCO3,2-5 mmol/l CaCl2, and 11 1 mmol/l D (+)-glucose). The main bronchi were dissected freeof loose connective tissue and were prepared intwo rings. The rings were mounted in glasschambers filled with 15 ml of Krebs-Henseleitsolution that was maintained at 37°C andgassed with a mixture of 95% 02 and 5% Co2,which produced a pH of 7 4. The bronchialrings were connected to force displacementtransducers (Grass FTO3) for continuousrecording of isometric tension. The rings wereallowed to equilibrate for 90 minutes whileresting tension was adjusted to 5 mN. Duringequilibration the medium was changed every20 minutes. Contractions were expressed as apercentage of the active tension obtained inresponse to acetylcholine (1 mM). Bronchialrings dissected free of epithelial and connectivetissue and bronchial rings with intact epi-thelium were studied. The epithelium wasremoved by rubbing the luminal surface gentlywith gauze, cut into strips, and tied at one sideto a suture; this was used to guide the gauzeinside the bronchial rings. Removal of theepithelium was confirmed histologically. At theend of the experiment control rings and ringswithout epithelium were fixed in 4% formalde-hyde in 0 1 M phosphate buffer at pH 7-2. Afterfixation, samples were dehydrated with ethanol,passed through xylene, and embedded inparaffin. Sections 6 ,um thick were cut perpen-dicular to the minor internal diameter of thebronchi and tissue blocks were oriented forlight microscopical analysis. Four sections atintervals of 100 gim were stained withhaematoxylin in eosin. The sections werecoded and examined without knowledge of thephysiological data. Light microscopical(Jenamed 30G0040) measurements wereobtained with an eyepiece graticule at a magni-fication of 160. The length of the mucosawithout epithelium was measured and expres-sed as a percentage of the total length ofbasement membrane in the section. The finalresult is the mean of all the measurementsperformed in each specimen.

DRUGSAcetylcholine and frusemide were obtainedfrom Sigma Chemical Co (St Louis, MO,USA) and neurokinin A from Peninsula (StHelens, England). Bumetanide was given byDr F Gaetani (Sigma-Tau).Toluene diisocyanate consisted of an 80: 20

mixture of the 2,4 and 2,6 isomers and wasobtained from Montedison (Porto Marghera,VE, Italy). It was dissolved in dimethyl sul-phoxide and prepared fresh before each study.The maximum final concentration of dimethylsulphoxide in the organ bath was 0 3%.Bumetanide was dissolved in 100% ethanol. Astock solution of frusemide was made up in10 mM sodium hydroxide solution and dilutedwith distilled water to the appropriate concen-tration. Addition of the solutions used todissolve the drugs was shown to have no effecton resting tension.

EXPERIMENTAL PROTOCOLSToluene diisocyanateThe effect of bumetanide and frusemide on theairway response to increasing concentrations oftoluene diisocyanate (003-1000 mM) wasassessed. After an equilibration period abaseline response to acetylcholine wasobtained. The rings were then rinsed. Parallelexperiments were carried out on paired bron-chial rings from the same animal; bumetanide(10 jiM or 100 MuM, contact time 30 minutes) orfrusemide (100 uM, contact time 30 minutes)was added to one of the two bronchial rings ofeach animal and nothing was added to thesecond ring. The response to increasing con-centrations of toluene diisocyanate was thenmeasured in both rings. Each successive con-centration of toluene diisocyanate was addedonly after the previous response had reached aconstant value. The effect of bumetanide wasstudied on rings with and without epithelium;the effect offrusemide was studied on rings withepithelium.

Neurokinin A, acetylcholine, and electrical fieldstimulationThe effect of bumetanide (10 jpM) on the res-ponses to neurokinin A, acetylcholine(0-1 jIM), and electrical field stimulation wasalso examined. In these experiments atropine(1 jMM) was added to antagonise the cholinergiccomponent of the response to electrical fieldstimulation and indomethacin (5 ,uM) to min-imise the spontaneous variations in tone of thetissue. Under these conditions a reproduciblecontractile response to electrical field stimula-tion was obtained for several hours.Experiments started after a 90 minute equi-libration period. The preparations underwentfield stimulation (1-10 Hz, 60 V, 0 5 ms for10 s) by means of two wire platinum electrodesplaced at the top and the bottom of the organbath and connected to a Grass SI 1 stimulator.A frequency-response study (1-10 Hz) wascarried out on the bronchial rings withoutepithelium until a reproducible response wasobtained. Ethanol (the solvent for bumetanide)or bumetanide (10 MuM) was added to the bathand a further frequency-response curve wascarried out 30 minutes later. The response wasexpressed as a percentage of the maximalcontraction obtained with 40 mM KCl.

ANALYSISValues are given as mean (SE). The effects ofbumetanide and frusemide on contractioninduced by toluene diisocyanate and of bume-tanide on contraction induced by acetylcholine,neurokinin A, and electrical field stimulationwere compared by two tailed Student's t testfor paired data and by analysis of variance(ANOVA) with repeated measures (BMDPStatistical Software, Inc, Los Angeles, CA).Values of p < 0-05 were considered significant.

ResultsTOLUENE DIISOCYANATEToluene diisocyanate caused a concentration

64

on Novem

ber 14, 2021 by guest. Protected by copyright.

http://thorax.bmj.com

/T

horax: first published as 10.1136/thx.48.1.63 on 1 January 1993. Dow

nloaded from

Effect ofbumetanide on toluene diisocyanate induced contractions in guinea pig airways

dependent contraction of bronchial smoothmuscle in bronchi with epithelium. Bumetan-ide (10 and 100 MM) did not alter the toluenediisocyanate induced contractions (fig 1).Toluene diisocyanate also caused contrac-

tion in bronchial rings without epithelium, andthis did not differ significantly between controlrings and rings pretreated with bumetanide (10and 100 uM; fig 2). Histological analysis con-

firmed that 94-1 (4 6)% of the epithelium hadbeen removed.Frusemide (100 MM) did not alter toluene

diisocyanate induced contraction significantly(fig 3). At a concentration of300 pM oftoluenediisocyanate there was a 39% reduction in ringspretreated with frusemide (95% confidenceinterval (95% CI) 78%-14% of response to1 mM acetylcholine). The mean values foundin control rings and in frusemide treated ringswere 64-5 (14-3)% (95% CI 97%-32% of theresponse to 1 mM acetylcholine) and 39-1(6 7)% (95% CI 54%-24% of the response to1 mM acetylcholine) respectively. The tablegives the results for both drugs.

ACETYLCHOLINE AND NEUROKININ ABumetanide (10 MM, contact time 30 minutes)did not alter the response to 01 MM acetyl-choline. The contractions were 43-6 (4-8)% incontrol rings and 45-0 (8-8)% in bumetanidetreated rings (n = 6, NS).Bumetanide pretreatment (10 MM, contact

time 30 minutes) did not alter the response to1 nM neurokinin A. Contractions were 23-4(2 8)% in control rings and 22-8 (4-2)% inbumetanide treated rings (n = 6, NS).Frusemide (100 MM, contact time 30 min-

utes) did not alter the response to 01 pM

120

2E

0

o

0

a)40-

100

80

1--I-1 1601F

40

20 [

01

003 01 03 1

1201[ B

E

0

coC

0

0.oQ)

100 [80 [

60

40 [

3 10 30 100 300 1000

I_6

20F

0

0-03 0.1 0 3 1 3 10 30 100 300 1000Toluene diisocyanate (,uM)

Figure 2 Effect of toluene diisocyanate on guinea pigmain bronchi with the epithelium removed in the absence(open circles) and presence (closed circles) of bumetanide(10 /lM panel A; 100 iM panel B). Each point is themean (SE) of 10 experiments. Ach-acetylcholine.

Figure I Effect of toluenediisocyanate on guinea pigmain bronchi withepithelium intact in theabsence (open circles) andpresence (closed circles) ofbumetanide (10 MM panelA; 100 uM panel B).Each point is the mean(SE) of six experiments.Ach-acetylcholine.

loo1

E 80

0

< 600

a)Co 400

0 20L-

I)O

0-03 0-1 0-3 1 3 10 30 100 300 1000Toluene diisocyanate (,uM)

A

100r

E 80

0 600

a) 400

a) 20

)1

acetylcholine. The contractions were 37-6(7-2)% in control rings and 415 (71)% infrusemide treated rings (n = 6, NS).Frusemide pretreatment (100 MM, contacttime 30 minutes) did not alter the response to1 nM neurokinin A. The contractions were

17-7 (2-4)% in control rings and 20-9 (1-6)% infrusemide treated rings (n = 6, NS).

Electrical field stimulationBumetanide (10 MM) caused a significantinhibition of non-cholinergic, non-adrenergiccontraction induced by electrical field stimula-

80F

60F

40

E

0

0)

4-

Co

0.Co0)a)cn

20p

0L

.- .-

003 0-1 03 1 3 10 30 100 3001000Toluene diisocyanate (,uM)

B

a

a .

0-03 0-1 0 3 1 3 10 30 100 300 1000Toluene diisocyanate (,uM)

Figure 3 Effect of toluene diisocyanate on guinea pigmain bronchi in the absence (open circles) and presence(closed circles) offrusemide (100 M1M). Each point is themean (SE) of 10 experiments. Ach-acetylcholine.

| l t 65

)F

0

loor

6-

41--i r-l-

x 11-100' JL

T///I

on Novem

ber 14, 2021 by guest. Protected by copyright.

http://thorax.bmj.com

/T

horax: first published as 10.1136/thx.48.1.63 on 1 January 1993. Dow

nloaded from

Mapp, Boniotti, Papi, Maggi, Di Stefano, Saetta, Ciaccia, Fabbri

Results of analysis of variance (ANO VA) with repeated measures

Drug Preparation F Value p

Bumetanide (10 ,uM) Isolated guinea pig bronchi with the 0 03 NSepithelium intact

Bumetanide (100 MM) Isolated guinea pig bronchi with the 0 52 NSepithelium intact

Bumetanide (10 MiM) Isolated guinea pig bronchi with the 0.19 NSepithelium removed

Bumetanide (100 MM) Isolated guinea pig bronchi with the 0 16 NSepithelium removed

Frusemide (100 pM) Isolated guinea pig bronchi with the 1-48 NSepithelium intact

tion. Under control conditions, field stimula-tion of the bronchial rings caused 9 (2)%, 15(3)%, and 47 (6)% of maximum contractionrespectively at 2, 5, and 10 Hz. Bumetanide(10 MM) caused an 89% inhibition of theresponse at 2 Hz (p < 0 01), 75% inhibition at

5 Hz (p < 0 01), and 38% inhibition at 10 Hz(p < 0-05) (fig 4). The vehicle for bumetanidehad no significant effect on the response to fieldstimulation in bronchial smooth muscle rings.After pretreatment with ethanol, contractionswere 8 (2)%, 14 5 (3)%, and 46 (8)% ofmaximum contraction at 2, 5, and 10 Hz andthese values did not differ from those obtainedin control conditions.

DiscussionIn the present study neither bumetanide (10and 100 ,uM) nor frusemide (100 ,uM) had a

significant effect on the contractile responseproduced by toluene diisocyanate in guinea pigairways. It is possible that the lack of a sig-nificant change after frusemide treatment was

due to the variability of the response. At a

concentration (10 MM) that has been shown to

inhibit Na/K/Cl cotransport bumetanide in-hibited the neurally mediated airways smoothmuscle contraction induced by electrical fieldstimulation. Because bumetanide had no sig-nificant effect on the contractile response to

exogenous acetylcholine or neurokinin A, our

findings suggest that its effect on electrical fieldstimulation is due to an effect on airway nerves

rather than a direct effect on contractility of

airway smooth muscle. Previous work supportsthis conclusion. Frusemide in bovine airways

Figure 4 Effect ofbumetadine (10 iM) on

the non-cholinergicresponse (atropine(I1pM) in the bath) toelectricalfield stimulation(60 V, 0 5 ms pulse width,10 s train duration) in

guinea pig isolated bronchiwith the epitheliumremoved. Results are mean

(SE) (n = 4).*p <005; **p < 001.

60 r

a,

c

In0)Cu

E

E

M Bumetanide

Control

40

20

01L2 5

Hz10

without epithelium failed to inhibit contractioninduced by exogenous histamine, potassiumchloride, or hyperosmolar saline, and was alsowithout effect on hypertonic saline inducedcontraction in bovine and human airways withepithelium intact. The conclusions from thisstudy, by Knox and Ajao,22 were that inhibitionof Na/K/Cl cotransport does not alter airwaysmooth muscle contractility and that theprotective effect of frusemide on bronchocon-strictor stimuli in vivo is unlikely to be due to adirect effect on airway smooth muscle. Wefound a similar lack of response with bumetan-ide on contractions induced by acetylcholineand neurokinin A on bronchial tissue withintact epithelium, so this would explain the lackof effect of loop diuretics in preparations withintact epithelium is not confined to onestimulus.

Inhaled frusemide has a protective effect on awide range of bronchoconstrictor stimuli invivo. It also inhibits cough induced by inhala-tion of solutions with low chloride content inhumans but not that induced by capsaicin.23The mechanism of action of frusemide on thesedifferent stimuli is not understood. Our resultsand the findings of previous studies suggestthat, in human, bovine, and guinea pig airwaysat least, loop diuretics at concentrations thatinhibit Na/K/Cl cotransport do not alter air-way smooth muscle contractility.Bumetanide inhibited contraction induced

by electrical field stimulation. Electrical fieldstimulation in isolated bronchi of guinea pigsproduces a reproducible constrictor responsethat is thought to be due to antidromic activa-tion of capsaicin sensitive nerves, the responsebeing largely mediated by endogenous tachy-kinin like peptides.2"27 Previous work hasshown that bumetanide and frusemide inhibitboth cholinergic and non-cholinergic excita-tory nerves in guinea pig airways in vitro"7 andone possible mechanism of action of thesediuretics could be inhibition of the Na/K/Clcotransport protein in airway nerves.

In the guinea pig isolated bronchus,frusemide and bumetanide have recently beenshown to inhibit non-cholinergic, capsaicinsensitive bronchial contraction produced byelectrical field stimulation without affecting theresponse to substance P, thus suggesting aprejunctional inhibitory effect.28 Our findingssupport this conclusion and they also suggestthat prejunctional inhibition of tachykininrelease by bumetanide does not involvemediator release from airway epithelium orgeneration of cyclooxygenase products as theinhibitory effect was evident in epitheliumdeprived bronchi and in the presence ofindomethacin. Two other points emerge fromour study in relation to the data of Elwood etal.28 Firstly, the lack of effect of bumetanide onneurokinin A induced contraction suggests thatneurokinin A is acting at the postjunctionallevel. This is important because neurokinin Arather than substance P is the main mediator ofthe tachykininergic response in guinea pigbronchi.2930 Secondly, bumetanide inhibitionof the non-cholinergic contraction was inver-sely related to the frequency of stimulation.

66

on Novem

ber 14, 2021 by guest. Protected by copyright.

http://thorax.bmj.com

/T

horax: first published as 10.1136/thx.48.1.63 on 1 January 1993. Dow

nloaded from

Effect of bumetanide on toluene diisocyanate induced contractions in guinea pig airways

This pattern would be expected from a sub-stance that inhibited transmitter release at theprejunctional level.The present study showed no significant

inhibition of toluene diisocyanate induced con-tractions in guinea pig airways by bumetanideor frusemide. We have previously shown thattoluene diisocyanate activates the sensorynerves in guinea pig airways by releasingtachykinins and that neutral endopeptidasemay play a part in modulating the response."8We have also shown that the action of toluenediisocyanate on sensory nerves is likely to beindirect, and that toluene diisocyanate maycause the generation of a prostanoid that inturn activates capsaicin sensitive primaryafferents via a ruthenium red sensitive mechan-ism.203' It was proposed recently that stimulisuch as toluene diisocyanate or an increase in[H+I can release an endogenous capsaicin likesubstance that acts either on the capsaicin/resiniferatoxin receptor or at a distinct receptorthat shares a common membrane transductionmechanism/ion channel with the capsaicin/resinferatoxin receptor."At least two independent modes of transmit-

ter secretion are known to exist in peripheralterminals of capsaicin sensitive primaryafferents. The first is activated by capsaicin andinvolves a calcium dependent transmittersecreted directly from the sensory nerve ter-minals; the second is activated by electrical fieldstimulation and involves a propagated actionpotential that invades the sensory nerve ter-minal with consequent depolarisation.27 Thelack of effect of bumetanide and frusemide ontoluene diisocyanate induced contraction atconcentrations that inhibit electrical fieldstimulation induced contraction supports theexistence of at least two distinct modes oftransmitter release in peripheral terminals ofcapsaicin sensitive primary afferent nerves.

We are grateful to Luigi Zedda for drawing the figures and to DrLicia Tolin for typing and editing the manuscript.This work was supported in part by grants from the National

Research Council (89.02961.04, 89.02664.04, FATMA90.01508.CT04), the Italian Ministry of Education, RegioneVeneto, the European Community of Coal and Steel, and theCNR-ENEL Project Interactions of Energy Systems withHuman Health and Environment, Rome.

1 Cohen MR, Hirsch E, Vergona R, Ryan J, Kolis SJ,Schwartz A. Comparative diuretic and tissue distributionstudy of bumetanide and furosemide in the dog. J Phar-macol Exp Ther 1976;197:697-702.

2 Mudge GH, Weiner I. High-ceiling diuretics. In: GoodmanGilman A, Godman LS, Gilman A, eds. The pharma-cological basis of therapeutics. New York: MacMillan,1990;896-900.

3 Bianco S, Pieroni MG, Refini M, Rottoli L, Sestini P.Protective effect of inhaled frusemide on allergen-inducedearly and late asthmatic reactions. N Engl J Med1989;321:1069-73.

4 Bianco S, Vaghi A, Robuschi M, Pasargiklian M. Preventionof exercise-induced bronchoconstriction by inhaledfrusemide. Lancet 1988;ii:252-5.

5 Robuschi M, Vaghi A, Gambaro G, Spagnotto S, Bianco S.Inhaled frusemide (F) is highly effective in preventingultrasonically nebulized water (UHN20) bronchocon-striction. Am Rev Respir Dis 1988;137:412 (abstract).

6 Vaghi A, Robuschi M, Bemi F, Bianco S. Effect of inhaledfrusemide (F) on bronchial response to histamine (H) inasthma. Eur Respir J 1988;1(suppl 2):4069 (abstract).

7 Nichol GM, AltonEWFW, Nix A, Geddes DM, Chung KF,Barnes PJ. Effect of inhaled frusemide on metabisulphiteand methacholine induced bronchoconstriction and nasalpotentials in asthmatic subjects. Thorax 1989;44:851P(abstract).

8 Bianco S, Pieroni MG, Refini M, Sturman A, Robuschi M.Inhaled frusemide (F) prevents allergen induced bron-choconstriction in atopic asthmatics [abstract]. Am RevRespir Dis 1988;137:27.

9 O'Connor BJ, Chung KF, Chen-Worsdell YM, Fuller RW,Barnes PJ. Effect ofinhaled furosemide and bumetanide onadenosine 5'-monophosphate- and sodium metabilsulfite-induced bronchoconstriction in asthmatic subjects. AmRev Respir Dis 1991;143:1329-33.

10 Deth RC. Influence offrusemide on rubidium 86 uptake andalpha-adrenergic responsiveness of aterial smooth muscle.Blood Vessels 1987;24:321-33.

11 Widdicombe JH. Ion transport by tracheal cells in culture.Clin Chest Med 1986;7:299-305.

12 Welsh MJ. Inhibition of chloride secretion of frusemide incanine tracheal epithelium. J Membr Biol 1983;71:219-26.

13 Lloyd EL. Frusemide and prevention ofbumetanide. Lancet1988;2:564 (letter).

14 Belcher NG, Rees PJ, Clark TJH, Lee TH. A comparison ofthe refractory periods induced by hypertonic airwaychallenge and exercise in bronchial asthma. Am Rev RespirDis 1987;135:822-5.

15 Hook WA, Siraganian RP. Influence of anions, cations, andosmolarity on anti-IgE induced histamine release fromhuman basophils. Immunology 1981;3:723-31.

16 Barnes PJ. Asthma as an axon reflex. Lancet 1986;1:242-5.17 Elwood W, Lotvall JO, Barnes PJ, Chung KF. Loop

diuretics inhibit cholinergic and noncholinergic nerves inguinea pig airways. Am Rev Respir Dis 1991;143:1340-4.

18 Mapp CE, Boschetto P, Dal Vecchio L, Maestrelli P, FabbriLM. Occupational asthma due to isocyanates. Eur Respir J1988;1 :273-9.

19 Mapp CE, Chitano P, Fabbri LM, Patacchini R, SanticioliP, Geppetti P, Maggi CA. Evidence that toluenediisocyanate activates the efferent function of capsaicin-sensitive primary afferents. Eur J Pharmacol1990;180:1 13-8.

20 Mapp CE, Chitano P, Fabbri LM, Patacchini R, Maggi CA.Pharmacological modulation ofthe contractile response totoluene diisocyanate in the rat isolated urinary bladder.Br J Pharmacol 1990;100:886-8.

21 Mapp CE, Graf PD, Boniotti A, Nadel JA. Toluenediisocyanate contracts guinea pig bronchial smoothmuscle by activating capsaicin-sensitive sensory nerves.J Pharmacol Exp Ther 1991;256:1082-5.

22 Knox AJ, Ajao P. Effect of frusemide on airway smoothmuscle contractility in vitro. Thorax 1990;45:856-9.

23 Ventresca PG, Nichol GM, Barbes PJ, Chung KF. Inhaledfrusemide inhibits cough induced by low chloride contentsolutions but not by capsaicin. Am Rev Respir Dis1990;142:143-6.

24 Szolcsanyi J, Bartho L. Capsaicin-sensitive non-cholinergicexcitatory innervation of the guinea-pig tracheobronchialsmooth muscle. Neurosci Lett 1982;34:247-50.

25 Bartho L, Amann R, Saria S, Szolcsanyi J, Lembeck F.Peripheral effect of opioid drugs on capsaicin-sensitiveneurones of the guinea-pig bronchus and rabbit ear.Naunyn-Schmiedebergs Arch Pharmacol 1987;336:316-21.

26 Lundberg JM, Saria A, Brodin E, Rosell S, Folkers K. Asubstance P antagonist inhibits vagally-mediated increasein vascular permeability and bronchial smooth musclecontraction in the guinea-pig. Proc Natl Acad Sci USA1983;80: 1 120-3.

27 Maggi CA, Patacchini R, Giuliani S, Santicioli P, Meli A.Evidence for two independent modes of activation of the"efferent" function of capsaicin-sensitive nerves. Eur JPharmacol 1988;156:367-73.

28 Elwood W, Barnes PJ, Chung KF. Inhibition of neurallymediated contraction ofguinea-pig airway smooth muscleby frusemide and bumetanide in vitro. Br J Pharmacol1990;100:473P.

29 Maggi CA, Patacchini R, Baroldi P, Theodorsson E, Meli A.Immunoblockade by a specific tachykinin antiserum ofthenon-cholinergic contractile responses in the guinea-pigisolated bronchus. J Auton Pharmacol 1990;10:173-9.

30 Maggi CA, Patacchini R, Rovero P, Santicioli P. Tachykininreceptors and noncholinergic bronchoconstriction in theguinea-pig isolated bronchi. Am Rev Respir Dis1991;144:363-7.

31 Mapp CE, Graf PD, Boniotti A, Nadel JA. Bronchialsmooth muscle responses evoked by toluene diisocyanateare inhibited by ruthenium red and by indomethacin. EarJPharmacol 1991;200:73-6.

32 Maggi CA. Capsaicin and primary afferents neurones: frombasic science to human therapy? JAuton Pharmacol1991;11:177-212.

67

on Novem

ber 14, 2021 by guest. Protected by copyright.

http://thorax.bmj.com

/T

horax: first published as 10.1136/thx.48.1.63 on 1 January 1993. Dow

nloaded from