Pulmonary Pharmacology (1990) 3 198-202©1990 Longman Group UK Ltd

The Effect of Epithelium Removal on Human Bronchial Smooth MuscleResponsiveness to Acetylcholine and Histamine

D. A. Knight, J . A. Adcock, M . J. Phillips, P . J. Thompson

University Department of Medicine, Queen Elizabeth II Medical Centre, Nedlands, WA 6009, Australia

SUMMARY. The respiratory epithelium produces a variety of inflammatory mediators which may be influencingthe bronchial hyperreactivity observed in patients with asthma . Animal studies have demonstrated that removal ofthe epithelium from tracheal and bronchial smooth muscle causes enhanced responses to cholinergic agonists andhistamine (Hist). In this study the effect of epithelium removal on human bronchial smooth muscle response toacetylcholine (ACh) and Hist was assessed. Bronchial smooth muscle was obtained fresh from the operating theatrefrom 12 patients undergoing thoracotomy . Cumulative concentration effect curves (CCEC) for Hist and ACh weregenerated for epithelium intact and epithelium denuded muscle strips . All CCEC's were performed in duplicate andall denuded strips were obtained from the same airway immediately adjacent to the intact strip . The mean (±SEM) maximum response for Hist for the intact strip was 8 .6 ± 1.1 (grams/gram wet weight) and 12 .0±1.4(grams/gram wet weight) for the denuded strip (p < 0.05). For ACh the values were 9 .3 ± 1.3 (g/g wet weight and14.3 ± 1.8 (g/g wet weight), respectively (p< 0.05). The pD 2 (- log ECsa) for ACh was increased two-fold followingepithelium removal (p<0.05). For Hist there was a similar increase in pD 2 but this did not reach statisticalsignificance. Thus removal of the epithelium from human isolated bronchial smooth muscle appears to modulateresponsiveness to ACh and Hist. This enhanced responsiveness consequent to epithelium loss may prove importantwith respect to the development of worsening asthma .

INTRODUCTION

The aetiology and pathogenesis of asthma is poorlyunderstood. Many theories involve the induction ofbronchial hyperreactivity through the interaction ofinflammatory cells and specific mediators. Recently,the respiratory epithelium has been demonstrated togenerate a variety of mediators including lipoxygen-ase and cycloxygenase products of arachadonic acidmetabolism .' As such the respiratory epithelium maybe contributing to the maintenance of resting bron-chial tone, and epithelium damage may be influencingairway smooth muscle hyperreactivity . )-3

Mechanical removal of the epithelium has beenshown to increase airway smooth muscle responsive-ness to contractile agonists in isolated airway prepar-ations from a variety of animal species . These includethe dog, guinea-pig, rabbit and cattle ."- 'o In themajority of these species, enhanced responsiveness tohistamine (Hist) and to cholinergic agonists, such asmethacholine, bethanacol and acetylcholine (ACh)has been demonstrated. Consequently, epithelial de-rived inhibitory factors have been posutlated whichare thought to be released from epithelial cells andresponsible for modulating responsiveness of the

Address correspondence to : Dr P . J . Thompson, UniversityDepartment of Medicine, Queen Elizabeth II Medical Centre,Nedlands, WA 6009, Australia .

198

0952-0600/90/0003-0198/$10 .00

underlying airway smooth muscle . 4 .6 .11 However,there is considerable species variation in both quanti-tative and qualitative responses to agonists followingepithelium removal and thus it seems important toinvestigate the role of the respiratory epithelium inhuman airways .Raeburn et al .' 2 demonstrated enhanced airway

smooth muscle sensitivity to methacholine followingepithelium removal in five post mortem trachea .Subsequently, similar findings have been reported byAizawa et al." using human bronchial smoothmuscle, but there appears to be little other datarelating to human tissue .

In the present study, the effect of epithelial loss onhuman bronchial smooth muscle responses to bothACh and Hist was investigated .

METHODS

Intact bronchial segments were obtained from 12patients undergoing surgery for resection of lungcarcinoma. A total of 48 bronchial smooth musclestrips were prepared from macroscopically normalairways with an internal diameter of 4- 10 mm . Allbronchi were placed in ice-cold Krebs Henseleit solu-tion (Composition (mM); NaCl 121, KCI 5 .4, MgSO,1 .2, NaH2PO4 1 .2, NaHCO3 15, CaCl2 2.5, glucose11 .5, previously gassed with 5% C0 2/95% 02) within

20 min of surgical resection . From the resected air-ways of each patient, bronchial muscle strips, 3-4 mmwide, were prepared from either fresh specimens or fromspecimens stored overnight in Krebs Henseleit solu-tion at 4°C .

Four bronchial smooth muscle strips preparedfrom the same specimen were tested simultaneously,two of which were denuded of epithelium by gentlerubbing with a cotton swab . Each muscle strip wasmounted in a 20 ml organ bath containing KrebsHenseleit solution at 37°C, gassed continuously with5% C02/95% 0 2 . The bronchial muscle strips wereadjusted to a previously determined optimal restingtension of I g. The preparations were allowed toequilibrate for 60-90 min, the bathing solutionbeing exchanged every 15 min . Changes in isometrictension were measured by a Grass FTO3C transducercoupled to a Rikadenki L50 multi-channel pen re-corder.

Following the equilibration period, a submaximalconcentration of ACh (10 -5 M) was added to theorgan bath to assess tissue viability and reproducibil-ity of responses . A cumulative concentration-effectcurve (CCEC) was then generated for ACh and,following washout, the tissue was allowed to return tothe resting level of tension . The procedure was thenrepeated for Hist. A second ACh CCEC was gener-ated at the end of the experiment as a measure oftissue viability . For all CCECs the concentrationrange of agonist was between 10 -6 and 10 -3 M.Assessment of the preservation of length-tension rela-tionship in the epithelium denuded preparations wasevaluated using a standard concentration of AChusing resting tensions of 0 .5, 1 .0 and 2 .0 g.

At the completion of each experiment bronchialspecimens were fixed in formyl saline and glutaral-dehyde for subsequent light and electron microscopyexamination .

The data generated was expressed as agonist po-tency, pD 2 (pD 2 = - Log EC50) and maximal tensiongenerated, Emax (g/g wet weight) .

Statistical analyses utilised the computer programeSPSS. Simple descriptive statistics were performedusing the meaned data from each patient . Statisticalsignificance was assessed by using ANOVA and sub-sequent Student's t-test for paired or unpaired datawith a minimum significance level of 0 .05 .

ACh and Hist were obtained from Sigma ChemicalCo (St Louis, Missouri) .

RESULTS

A total of 24 pairs of muscle strips (with/withoutepithelium) were studied from 12 patients .

In all denuded preparations scanning electron andlight microscopy demonstrated the epithelial layer ofthe bronchus had been successfully removed without

obvious damage to underlying mucosa or smoothmuscle .

Removal of the epithelium significantly increasedresponsiveness to ACh at all concentrations (Fig . 1)such that Emax for epithelium rubbed preparationswas enhanced approximately 50% above that ob-tained with the epithelium intact preparations(p < 0 .05; n = 12). The potency (pD 2 ) of ACh wasenhanced two-fold following epithelium removal(p < 0.05; n = 12) (Table 1) .

Responses of epithelial stripped preparations toHist are shown in Figure 2 . Emax was enhanced to asimilar degree to that observed with ACh (p < 0 .05 ;n = 12). The pD 2 of Hist, although nearly two-foldgreater in epithelial denuded preparations was notstatistically significantly different to epithelial intactpreparations (p > 0 .1; n = 12) (Table 1) .The pD 2 value for Hist in the epithelium intact

preparations was significantly greater than thatachieved with ACh (p < 0 .05) (Table 1). For all otherparameters no statistically significant difference wasobserved.

Epithelium and Human Airway Response 199

ACh

Hist(+)EPI

(-)EPI

(+)EPI

(-)EPI

Log ACh (M)

Fig. 1-Cumulative concentration effect curves for acetylcholine(ACh) for epithelium intact bronchial smooth muscle strips (0)and denuded strips (I) n = 12. Results expressed as mean ± SEM .

Table 1 . Effects of epithelium removal on human bronchialsmooth muscle responses to acetylcholine and histamine .

pD2

4.61±0.1

4.91±0.1#

5.07±0.06*

5.14±0.11Emax 9.31±1 .28 14.26±1.81#

8.62±1.08

11 .97±1 .43#

Emax and pD2 values for epithelium intact and epithelium strippedbronchial smooth muscle derived from cumulative concentrationeffect curves to acetylcholine (ACh) and histamine (Hist) . Emaxexpressed as gram active tension developed per gram wet weighttissue . (n=12) pD2 = - log EC50 .#=p<0.05 compared with epithelium intact bronchi .* = p < 0.05 compared to ACh induced responses .

200 Pulmonary Pharmacology

Fig . 2-Cumulative concentration effect curves for histamine(Hist) for epithelium intact bronchial smooth muscle strips (•)and denuded strips (•) n = 12 . Results expressed as mean ± SEM .

Responses to the third ACh CCEC following theHist curve were not statistically significantly differentto the initial ACh CCEC (p>0.1), indicating thatthere were no changes in muscle responsiveness dur-ing the course of the experiment .

Length/tension relationships were similar for epi-thelium intact and epithelium denuded bronchi sup-porting the microscopy data that there was no appar-ent damage to underlying muscle following epitheliumremoval .

All patients received the same anaesthetic agentsand none received any medications known to affectairway responsiveness prior to surgery .

DISCUSSION

The removal of the epithelium from freshly obtainedhuman bronchial smooth muscle produced enhancedcontractile responsiveness for both ACh and Hist .The mechanisms involved are not clear . There was noapparent physical damage to underlying smoothmuscle and functionally length/tension relationshipsappeared to be preserved . It seems likely the increasedresponsiveness was due to loss of inhibitory factorsgenerated by the epithelium .

The release of inhibitory factors from airway epi-thelium has been proposed for a variety of animalspecies. However, responses to contractile agonistsfollowing removal of the epithelium have been shownto differ between species. In most animal studiesepithelium removal produces enhanced sensitivity tomuscarinic agonists and to Hist without alteration inEmax .4-8 However, in contrast, bovine trachealsmooth muscle demonstrates both an increase in

Emax and sensitivity following removal of the epithel-ium.' Similar species variation is also seen withrelaxant agonists . Airway smooth muscle responses toisoprenaline following epithelium removal vary fromenhanced sensitivity in bovine airways to diminutionof relaxation in the dog . 4 '9'"

Studies involving the modulatory effect of the epi-thelium in human airways are limited . Raeburn etal." using methacholine in post mortem trachealsmooth muscle, demonstrated a 2 .2-fold increase inthe methacholine pD2 value but no change in Emaxfollowing epithelium removal . The effect of epitheliumremoval on Hist responses was not examined . In theonly other reported study, Aizawa et al .' 3 assessed theeffect of epithelium removal on responses to ACh,Hist and PgF2a using six human isolated bronchialring preparations for each agonist . The results fromthis study were similar to those of Raeburn et al ., 12demonstrating an increase in the pD 2 values for AChas well as for Hist without any significant increase inEmax .

The results we obtained with human bronchi from12 patients, are similar with respect to sensitivity toACh showing also a two-fold increase in pD 2 afterepithelium removal . However, while there was atendency for an enhanced Hist pD 2 in epitheliumdenuded preparations this was not statistically signifi-cant. The lack of significant change in Hist sensitivitymay reflect a problem of statistical power as the datafrom the epithelium denuded preparations appearedto have considerable variability for the Hist response .

With respect to Emax, our results contrast withthose of both Raeburn et al .' 2 and Aizawa et al . 13Removal of epithelium causes a significant increase inEmax for both ACh and Hist and the reasons for thisdisparity are unclear. However, regional differences inairway responsiveness may exist . Most investigatorsusing animal models have, like Raeburn et al .,'2 onlystudied tracheal smooth muscle . More recently, sev-eral investigators have demonstrated considerableheterogeneity of smooth muscle responses to contrac-tile agonists in canine airways . 14-16 The more distalthe airway the less influence the epithelium appearedto have on contractile responses . A similar differencemay exist between human trachea and smaller air-ways. This emphasises the need to pursue regionalstudies in human airways particularly since asthmainvolves both central and distal airway obstruction .Additionally, it has been suggested the contractileproperties of post-motem tissue may differ from post-operative specimens" and whether this has contrib-uted to the disparity between the results of our studyand that of Raeburn et al ., 12 is uncertain .

Regional differences and the use of post mortemtissue do not explain the different results obtained byAizawa et al .' 3 They, as in our study, used humanisolated bronchial smooth muscle obtained from post-operative tissue and were still unable to demonstrate a

significant change in Emax following epithelium re-moval. However, Aizawa et al." used bronchial ringpreparations, as opposed to bronchial strips, and thismay have led to a geometric constraint on the musclereaching its true Emax . In closed ring preparationsthe preload tension is applied obliquely to the major-ity of contracting muscle fibres and additionally non-muscular tissue is more liable to restrict muscle fibreshortening. Similar differences in response when usingdifferent preparations have been suggested in animalstudies ." In addition, neither Raeburn et al . norAizawa et al . normalized the data for muscle weightor cross sectional area and thus random differences inmuscle mass may have introduced greater variabilityin the assessment of maximal tension .

In animal studies the proposed epithelium depen-dent inhibitory factors have yet to be isolated andcharacterised although recent studies using bioassaysystems have supported a `transferable' factor existingin the epithelium cell layer." The nature of the inhibi-tory factor(s) again appears to be species dependent .Flavahan et al . (1985) and Butler et al . (1987) have bothsuggested a cycloxygenase product of arachadonic acidmetabolism is at least partially involved 4,10 and incanine trachea it has been suggested this is PGE 2 . 2°However, in bovine trachea indomethacin did notinfluence the epithelial relaxant response.' Alterna-tively the `inhibitory factor(s)' may not act directly toinhibit smooth muscle contraction but rather exerttheir effects indirectly . Such mechanisms could involveenhanced agonist metabolism . Frossard et al ., sug-gested tachykinins, such as Substance P are activelydegraded by metalloendopeptidases resident in theepithelium of the guinea-pig isolated trachea . 21 Similarfindings have been reported in human isolated bron-chi . 22 However, Flavahan et al.' failed to demonstratean enhanced reponse to ACh following inhibition ofacetylcholinesterase in the canine trachea . Thus,whether or not enzymatic degradation of ACh and Histwithin the respiratory epithelium of human airwayscontributes to enhanced responsiveness to contractileagonists awaits further investigation .

Further characteristion of the modulatory functionof the epithelium on human airway smooth muscleresponsiveness is required . This is particularly import-ant since during episodes of acute asthma there isgeneralised epithelial desquamation which may becontributing significantly to the induction of airwayhyperreactivity through loss of inhibitory factors .

Acknowledgements

The authors thank the pathologists and cardiothora-cic surgeons at Royal Perth, Sir Charles Gairdner andMount Hospitals for their help in performing thesestudies and are also indebted to the National Healthand Medical Research Council of Australia for theirfinancial support .

References

1 . Nadel J A. Some epithelial metabolic factors affectingairway smooth muscle . Am Rev Resp Dis 1988 ; 138 :S22-S23 .

2. Laitinen L A, Heino M, Laitinen A, Kava T,Haahtela T. Damage of the epithelium and bronchialreactivity in patients with asthma . Am Rev Resp Dis1985 ; 131 : 599-606 .

3 . Cuss F M, Barnes P J. Epithelial mediators. Am RevResp Dis 1987; S32-S35 .

4. Flavahan N A, Aarhus L L, Rimele J J, VanhoutteP M. Respiratory epithelium inhibits bronchialsmooth muscle tone . J Appl Physiol 1985 ; 58 :834-838 .

5 . Hay D W P, Farmer S G, Raeburn D, MuccitelliR M, Wilson K A, Fedan J S . Differential effects ofepithelium removal on the responsiveness of guineapig tracheal smooth muscle to bronchoconstrictors . BrJ Pharmac 1987; 92: 381-388 .

6 . Hay D W P, Muccitelli R M, Horstmeyer D L, WilsonK A, Raeburn D . Demonstration of the release of anepithelium derived inhibitory factor from a novelpreparation of guinea pig trachea . Eur J Pharmac1987;136:247-250 .

7 . Goldie R G, Papadimitriou J M, Paterson J W, RigbyP J, Self H M, Spina D . The influence of theepithelium on responsiveness of guinea pig isolatedtrachea to contractile and relaxant agonists . BrJ Pharmac 1986 ; 87 : 5-14 .

8 . Raeburn D, Hay D W P, Robinson V A, Farmer S G,Flemming W W, Fedan J S. The effect of Verapamil isreduced in isolated airway preparations lacking theepithelium . Life Sci 1986 ; 38 : 809-816.

9. Barnes P J, Cuss P J, Palmer J B . The effect of airwayepithelium on smooth muscle contractility of bovinetrachea . Br J Pharmac 1985 ; 86 : 685-691 .

10. Butler G B, Adler K B, Evans J N, Morgan P W,Szarek J L . Modulation of rabbit airway smoothmuscle responsiveness by respiratory epithelium . AmRev Resp Dis 1987 ; 137 : 1099-1104.

11 . Vanhoutte P M. Epithelium derived relaxant factor(s)and bronchial reactivity . Am Rev Resp Dis 1988 ; 138 :s24-s30.

12 . Raeburn D, Hay D W P, Farmer S G, Fedan J S .Epithelium removal increases the reactivity of isolatedhuman tracheal muscle to methacholine and reducesthe effect of verapamil . Eur J Pharmac 1986 ; 123 :451-453 .

13 . Aizawa H, Miyazaki N, Shigematsu N, Tomooka M .A possible role of airway epithelium in modulatinghyperresponsiveness . Br J Pharm 1988; 93: 139-145 .

14 . Shioya T, Solway J, Munoz N M, Mack M, Leff A R .Distribution of airway contractile responses within themajor diameter bronchi during exogenousbronchoconstriction . Am Rev Resp Dis 1987; 135 :1105-1111 .

15 . Stuart-Smith K, Vanhoutte P M . Heterogeneity in theeffects of epithelium removal in the canine bronchialtree . J Appl Physiol 1987 ; 63 : 2510-2515 .

16. Hay D W P, Raeburn D, Fedan J S . Regionaldifferences in reactivity and in the influence of theepithelium on canine intrapulmonary bronchialsmooth muscle responsiveness . Eur J Pharm 1987 ; 141 :363-370 .

17 . De Jongste J C, Mons H, Bonta I L, Kerrebijn K F .In vitro responses to airways from an asthmaticpatient . Eur J Respir Dis 1987 ; 71 : 23-29 .

18 . Armour C L, Diment L M, Black J L . Relationshipbetween smooth muscle volume and contractileresponse in airway tissue. Isometric versus isotonicmeasurement. J Pharm Exp Ther 1988 ; 245 : 687-691 .

19 . Fernandes L B, Paterson J W, Goldie R G . Co-axialbioassay of a smooth muscle relaxant factor releasedfrom guinea pig tracheal epithelium . Br J Pharmac1989 ; 96 : 117-124 .

20 . Barnett K, Jacoby D B, Nadel J A, Lazarus S C. The

Epithelium and Human Airway Response 20 1

20 2 Pulmonary Pharmacology

effects of epithelial cell supernatant on contraction of

22. Naline E, DeVillier P, Drapeau G et al .isolated canine tracheal smooth muscle. Am Rev Res

Characterization of neurokinin effects and receptorDis 1988 ; 138 : 780-783 .

selectivity in human isolated bronchi . Am Rev Resp21 . Frossard N, Rhoden K J, Barnes P J . Influence of the

Dis 1988; 140 : 679-686.epithelium on guinea pig airway responses totachykinins : role of endopeptidase and

Date received : 13 November 1989cyclooxygenase . J Pharm Exp Ther 1989; 248 :

Date revised : 7 December 1989292-298 .

Date accepted : 4 April 1990