- Email: [email protected]
Epithelium-derived relaxing factor(s) and bronchial reactivity
THE JOURNAL OF
ALLERGY AND
CLINICAL VOLUME
IMMUNOLOG
83
ostgraduate
5
course
pithelium-derived ronchial reactivity aul M. Vanhoutte,
relaxing factor(s)
an
MD, PhD Rochester, Minn.
In isolated blood vessels, the mechanical or enzymatic removal of the endothelium augments the contractions evoked by a variety of vasoconstrictor agents, because the endothelial cells release a powerful relaxing substance(s) (endothelium-derived relaxing factor(s)). The focal absence of epithelial cells in airways of patients with asthma is well documented. When the luminal surface of canine bronchi is rubbed gently, the only morphologic change observed is the disappearance of the epithelial cell layer. The removal of the epithelium causes an increased sensitivity of the bronchial smooth muscle to acetylcholine without alteration in the maximal responsiveness to the cholinergic transmitter. The augmentation cannot be attributed to reduced enzymatic breakdown of acetylcholine after removal of the epithelial cells, since it is not affected by inhibitors of acetylcholinesterase. It cannot be attributed to disappearance of a dtffusion barrier, since epithelium removal also augments the contractions evoked by electrical stimulation of the cholinergic nerve endings. Removal of the epithelium potentiates the contractions evoked by histamine and S-hydroxytryptamine in a manner similar to that observed for acetylcholine. The potentiating effect of epithelium removal is more pronounced in larger rather than in smaller bronchi. By contrast, the relaxations evoked by P-adrenergic agonists are less pronounced in bronchi without, rather than in bronchi with, epithelium. The influence of the epithelium on isoproterenol-induced relaxation is more pronounced in smaller rather than in larger bronchi. These observations suggest that the bronchial epithelium releases an inhibitory factor that partially counteracts activation of the airway smooth muscle by bronchoconstrictor substances. Preliminary bioassay studies confirm that the bronchial epithelium can release a potent relaxing substance that inhibits not only bronchial but also vascular smooth muscle. In larger bronchi, the epithelium-derived relaxing factor is released under basal conditions; in smaller bronchi, its release can be activated by catecholamines. The augmentation of the response to bronchoconstrictor agents by removal of the epithelium has been observed, in the author’s and other laboratories, in isolated airways from different species. (.I ALLERGYCLIN IMMUNOL 1989;83:85.5-61.)
From the Department of Physiology and Biophysics, Mayo Clinic and Mayo Foundation, Rochester, Minn. Supported in part by National Institutes of Health Grant HL 21584. Reptin? requests: Paul M. Vanhoutte, MD, PhD, Department of Physiology and Biophysics, Mayo Clinic and Mayo Foundation, Rochester, MN 55905.
A dysfunctional respiratory epithelium may explain in part bronchial hyperreactivity in asthma and airway infection. 1,2 in the vascular system the e~dothelial cells can affect the tone of the underlying smooth muscle by releasing relaxing and contacting factor(s).3-8When experiments were designed first to de55
Vanhoutte
. 1. Histologic bronchi removes
section demonstrating epithelial ceils (right).
that gentle rubbing of the iuminal (From reference 12. By permission.)
surface
of canine
Epithelium: orwithout o=with
Histamine,-log FIG. 2. Effects of removal and 5-hydroxytryptamine
M
5-HT,-log
M
of the epithelium on concentration-response curves to histamine (kefti (5-HT, right) in canine bronchi. (From reference 11. By permission.)
termine whether the bronchial epithelium modulates the responsiveness of airway smooth muscle, they demonstrated that removal of the epithelial cells caused a hyperresponsiveness to acetylcholine, histamine, and serotonin9‘” The present article updates earlier brief overviews summarizing these data that are in support of these initial observations and strengthen the concept that the release of relaxing factor(s) from epithelial cells contributes to the regulation of bronchomotor tone.12-‘5
Rubbing of the mucosal surface of canine bronchi removes the epithelial cells without disturbing the underlying layers (Fig. I).” Thus, the responsiveness of isolated bronchi (or tracheas) with or without ep-
ithelium can be compared, allowing the dete~inati~~ of the role of the epithelium in controlling airway smooth muscle. In canine bronchi, removal of the epithelium causes a shift to tbe left of the doseresponse curves to the bronchoconstrictors, histamine, 5-hydroxytryptamine (Fig. 2) and acetylcholine (Fig. ,).I’ These observations in canine bronchi implied that the absence of epithelial cells potentiate the reactivity of the underlying smooth muscle, They have been extended to different species, including cattle,16 guinea pigs,‘7-23 rabbits,24 pigs,25 and humans26 with different bronchoconstrictors, including allergen. Thus, there can be little doubt that removal or absence of the epithelium can lead to exaggerated constriction of airways. However, although the phenom-
VOLUME NLrMBER
Epithelium-derived
83 5
relaxing
factor
Epi~~~lium:
Epithelium:
o=without
--Control
e-with
-Ecothiopate,
ith, 10m7 M
lOO%=
0 Without,
a3
12.62
2,
loo%=?.39
n=6
Acetylcholine,-log
M
FIG. 3. Effects of epithelium removal on responsiveness of canine bronchial rings to acetylcholine in control preparations (dashed lines) and after echothiopate (1Ue7 mol/L) &o/id linesj in canine bronchi. (From reference 11. By permission.)
enon cannot be disputed, the mechanism underlying the phenomenon remains a matter of interpretation. ~pith~iium~derived or reality?
relaxing
factor:
Myth
The hyperresponsiveness evoked by the removal of I Ihe epithelium cannot be explained by a nonspecific increase in the sensitivity of the bronchial smooth muscle. Indeed, the dose-response curve to potassium chloride is not affected (Fig. 4); neither are the maximal responses to acetylcholine, histamine, or serotonin (Figs. 2 and ,).‘I, 16*I7327Also, the removal of the epithelium does not affect the resting membrane potential in canine bronchial smooth muscle and does not affect the concentration-depolarization curve to potassium ioms.28 Enzymatic breakdown within the epithelial cells is not involved since the epithelial cells do not contain acetylcholinesterase (unpublished observations), and the removal of epithelium augments responses to acetylcholine after incubation with inhibitors of acetylcholinesterase.” Inhibition of monoamine oxidase also does not prevent the potentiation of the response to Shydroxytyptamine.’ However, the epithelium may play a role in the inactivation of adenosine.29 The disappearance of a diffusion barrier theoretically could be a logical explanation for the potentiation induced by removal of the epithelium. However, this interpretation appears unlikely in view of the fact that the hyperresponsiveness on removal of the epithelium has been demonstrated in rings of airways where the bronchoconstrictors reach the bronchial smooth muscle from the outside.” Moreover, in canine bronchi, removal of the epithelium also potentiates constric-
=i .-E 2
4
E z s ,’
0
10
20
30 K’,
40
60
120
m
FIG. 4. Contractions evoked by increasing concentrations of potassium chloride (K+) in rings with (O - - - - O) and without (e---e) epithelium prepared from the same canine bronchi. These data are presented as means I SEM. No significant differences were noted between the two groups, implying that the procedure to remove the epithelium has not altered the ability of the bronchial smooth muscle to be activated directly. (From reference 15, By permission.)
tions evoked by electrical stimulation of the cholinergic nerve endings that release acetylcboli~e in the immediate vicinity of the bronchial smooth muscle (Fig. 5). A major reason to reject the hypothesis that the epithelium acts as a passive diffusion barrier to bronchoconstrictor agonists is that removal of the epithelium attenuates relaxations induced by isoproterenol “, 16,27 arachidonic acid,30-33the calcium ionophor; A23187,” and the calcium antagonist, verapami1.24,26 In analogy with the blood vessel wa11,3,6-7the observations on bronchi with and without ep~t~elium are explained best if the epithelial cells were to release an inhibitory factor(s) (epithelium-derived relaxing factor(s)) that controls the tone of the underlying smooth muscle.1’.34Bioassay studies in canine bronchi indicate that indeed epithelial cells release a fa,ctor(s) that relaxes airway smooth muscle (Fig. 6). 12*3” However, such bioassay studies are not possible if the airways are taken from animals with ~ea~wo~s or
37”6 ~rebs-R~~~~r solution at 2ml/min
Epithelium:
++i Direct superfusion without
__
_
__
a Hz
4 Hz
2Hz
37% aerated organ clamber
16 Hz
Bronchus Propranolol, 6x 1Om6
wit
ronchus Mtithoert epithelium
M
Acetylcholine
without
fi -__-~ 2 Hz
JI
JI-
4 Hz
JI-
a Hz
’ 3g
16 Hz
FIG. 5. Comparison of the response to electrical activation of the cholinergic nerve endings of rings, with and without epithelium, of the same canine bronchi. In the rings without epithelium the initial peak response was larger (left) and the delay of the response on prolonged stimulation (fade;right) less pronounced. (From reference 11. By permission.)
infection of the airways (unpublished results). Epitbelial cells (which are highly polarized) could release tbe factor(s) preferentially toward the underlying smooth muscle rather than in the airway lumen, and mucus could inactivate the factor(s). Alternatively, epithelium-derived relaxing factor may have a very short half-life. The nature of the inhibitory mediator released from the airway epithelium is unknown, and more than one factor may be involved. In the guinea pig, but not in the dog and in cattle, products of the metabolism of arachidonic acid may contribute to the inhibitory role of the epithelial cells.“* 16,19,2’ They must contribute to epithelium-dependent relaxations induced by arachidonic acid and platelet-activating factor.23,31-33* 35 Factors released from the epithelium cause relaxation of isolated blood vessels.“, 36 The epitheliumderived factor that relaxes vascular smooth muscle is not a product of cycIooxygenase.36 Caution should be exerted when the factor(s) derived from the epithelium that can be bioassayed with isolated blood vessels is assumed to be the one causing inhibition of airway smooth muscle. Indeed, vascular and bronchial smooth muscle are different in their responsiveness to endogenous-relaxing substances .37 The similar parallel effect of epithelium removal on concentration-contraction curves to acetylcholine, histamine, and 5-hydroxyhyptamine implies that the release of the relaxing factor(s) is continuous rather than caused by activation of specific epithelial
Superfusion mode:
epithelium direct
epitheiium direct
direct
QPf Acetylcholine,
3xlO%
FIG. 6. Bioassay experiment in which a ring of canine bronchus without epithelium (contracted with acetylcholine) is superfused with Krebs-Ringer bicarbonate solution flowing either through a bronchus with epithelium (~7~8 arrow; note the marked relaxation), a stainless steel tube (direct; note the reversal of the relaxation), or a bronchus without epithelium (note the absence of relaxation]. This demonstrates that the bronchial epithelium can reiease a potent inhibitory factor. (From reference 12. By permission.)
receptors. However, the increased responsiveness to P-adrenergic agonists in intact, compared with epithelium-denuded bronchi,“, 27,3x suggests tbat activation of P-adrenoceptors on the epithelial cells (which have a high density of P-adrenoceptors, compared to airway smooth muscle’7, 39,40-42)triggers the active release of epithelium-derived relaxing factor (Fig. 7). A similar conclusion can be reached for arachidonic acid,23,30-32 platelet-activating factorY3” and methylxanthines.43 Heterogeneity In airways of decreasing diameter, the ~otentiatin~ effect of epithelium removal on the response to adrenergic activation is augmented (Fig. 8) .27A similar trend is noted in porcine airways.25 However, with
VOLUME NUMBER
Epithelium-derived
83 5
relaxing
factor
FIG. 7. Proposed interaction between epithelial cells and bronchial smooth muscle. Under basal conditions the epithelium secretes epithelium-derived relaxing factor that acts as a functional antagonist on the bronchial smooth muscle, causing a comparable depression of the response to a variety of bronchoconstrictor agonists. The release of the factor can be activated by stimulation of B-adrenoceptors on the endothelial cells. ACh, acetylcholine; B, p-adrenoceptors; H, histaminic receptor; 5-UT, serotonergic receptor; and M, muscarinic receptor. (From reference 15. By permission.)
2nd order
Epithelium: o.with ozwithout
+9
9
8
7
4
5
6
n=Ei
4th order
3rd order
80 100 120 ,401 9
Isoproterenol,
8
7
6
5
4
- log M
FIG. 8. Comparison in lobar (2nd order), segmental (3rd order), and subsegmental (4th order) bronchi of the same dogs of the effect of epithelium removal on the relaxations evoked by increasing concentrations of isoproterenol. Note that with decreasing diameter of the bronchus, the potentiating effect of the removal is progressively augmented. These studies support the hypothesis that the triggered release of epithelium-derived relaxing factor is less in larger than in smaller airways. (From reference 27. By permission.)
0
vanhautte
2nd order
3rd order
W-iydroxytryptamine,-log FIG. 9. Comparison in lobar (Zndorderj, segmental (3rdorder), subsegmental {4tb order) bronchi of the same dogs of the effect of epithelium removal on the contractions evoked by increasing concentrations of 5-hydroxytryptamine. Note that with decreasing diameter of the bronchus, the potentiating effect of the removal is progressively reduced, These studies support the hypothesis that the basal release of epithelium-derived relaxing factor is greater in larger than in smaller airways. (From reference 27. By permission.)
decreasing diameter, the modulating influence of the epithelium on the response to bronchoconstrictors becomes less prominent (Fig. 9).27 Thus, if the airway epithelium releases a factor (or factors) that promotes relaxation. of bronchial smooth muscle, a marked heterogeneity exists in the release or the effect of the factor(s) along the bronchial tree. Thus, the absence of potent&ion of bronchoconstrictor effects with epithelium removal in a given airway preparation can not be used against the hypothesis that the epithelium can control the underlying bronchial smooth muscle.
If the release of epithelial-derived inhibitory factor(s) occurs in the intact organism, the absence or dysfunction of the epithelium could contribute to abnormal responses of the bronchial wall. Stimuli that induce hyperreactivity of the bronchi, such as exposure to ozone, allergen, or respiratory viruses, cause acute inflammation of the epithelium.” ‘. 44,45-49The major basic protein of eosinophils, which is involved in certain types of inflammation and hyperreactivity of the airways,50 augments the responsiveness of isolated airways if epithelial cells are present, but not in their absence.‘* Thus, malfunction or absence of the epithelium may cause bronchial hyperreactivity. This interpretation is supported by the fact that epithelial destruction is prominent in patients with asthma.51 The facilitatory role of the epithelium on the response to P-adrenergic agonists and methylxanthines could be of particular importance when these therapeutic agents are administered by aerosol.” 1 thank Mrs. H. Hendricksen and Mr. R. Lorenz for preparing the figures, and Ms. K. Kros for the secretarial assistance.
REFEREt’lCES 1. Nadel JA. Bronchial reactivity. Adv intern Med 1983;28:20723. 2. Hogg JC, Eggleston PA. Is asthma an epithelial disease? Am Rev Respir Dis 1984;129:207-8. 3. Furchgott RF. The role of endothelium in responses of vascular smooth muscle to drugs. Ann Rev Pharmacol Toxicol 1984;24:175-97. 4. Rubanyi GM, Vanhoutte PM. Hypoxia releases a vasoconstrictor substance from the canine vascu!ar endothelium. J Physiol 1985;364:45-56. 5. Furchgott RF, Zawadzki JV. The obligatory role of endolhelial cells in the relaxation of arterial smooth muscle by acetyleholine. Nature 1980;288:373-6. 6. Vanhoutte PM, Rubanyi GM, Miller VM, Houston DS. Modulation of vascular smooth muscle contraction by the endothelium. Annu Rev Physiol 1986;48:307-20. 7. Vanhoutte PM. Endothelium and the control of vascular tissue. News in Physiol Sci 1987;2:18-22. 8. Vanhoutte PM. Endothelium-dependent contracrions in arteries and veins. Blood Vessels 1987b;24:141-4. 9. Aarhus LL, Rimele TJ, Vanhoutte PM. Removal of the epithelium causes bronchial supersensitivity Lo acetylcholine and 5-hydroxytryptamine. Fed Proc 1984;43:955. 10. Flavahan NA, Vanhoutte PM. Epitheiial-dependent attenuation of bronchial smooth muscle lone. Fed Proc 1984;43:429. 11. Flavahan NA, Aarhus LL, Rimele TJ, Vanhoutte PM. The respiratory epithelium inhibits bronchial smooth muscle tone. J Appl Physiol 1985;58:834-8. 12. Vanhoutte PM. Airway epithelium and bronchial reactivity. Can J Physiol Phamacol 1987;65:44&-50. 13. Farmer SG. Airway smoorb muscle responsiveness: modulation by the epithelium. Trends Pharmacol Sci 1987$X:5-10. 14. Vanhoutte PM, Flavahan NA. Modulation of cholinergic neuretransmission in the airways. In: Kaliner MA, Bares PJ, eds. The airways: neural control in health and disease. New York: Marcel Dekker, chap 9, 1987:203-16. 15. Vanhoutte PM. Epithelium-derived relaxing factor(s) a& bronchial reactivity. Ann Rev Respir Dis 1988;138:S24-S30. 16. Barnes PJ, CUSSFM, Palmer JB. The effect of airway epitbelium on smooth muscle contractility in bovine trachea. Br J Pharmacol 1985;86:684-91,
VOLUME NUMBER
83 5
17. Goldie RG. Papadimitriou JM, Paterson WJ, Rigby PJ, Self NM, Spina D. Influence of the epithelium on responsiveness of guinea pig isolated trachea to contractile and relaxant agonists. Br .I Pharmacol 1986;87:5-14. 18. Hay DWP, Robinson VA, Fleming WW, Fedan JS. Role of the epithehum in contractile responses of the guinea pig isolated trach.ea. Fed Proc 1985;44:506. 19. Hay DWP, Farmer SG, Raebum D, Robinson VA, Fleming WW, Fedan JS. Airway epithelium modulates the reactivity of guinea pig respiratory smooth muscle. Eur J Pharmacol 1986;129:11-18. 20. Hay DWP, Raebum D, Farmer SG, Fleming WW, Fedan JS. Epithelium modulates the reactivity of ovalbumin-sensitized guinea pig airway smooth muscle. Life Sci 1986;38:2461-8. 21. Holroyde MC. The influence of epithelium on the responsiveness of guinea pig isolated trachea. Br J Pharmacol 1986; 87501-7. 22. Vanhoutte PM, Flavahan NA, Slifman NF, Gleich GJ. Eosinophil granule major basic protein causes hyperreactivity of respiratory muscle in the presence, but not in the absence of, epithelial cells. Proceedings of the International Union of Physiological Sciences, Vancouver, B.C., Canada, July 13-18, 1956. 23. Tschirhart E, Frossard N, Bertrand C, Landry Y. Arachidonic acid metabolites and airway epithelium-dependent relaxant factor. J Phannacol Exp Ther 1987;243:310-316. 24. Raebum D, Hay DWP, Robinson VA, Farmer SG, Fleming WW, Fedan JS. The effect of verapamil is reduced in isolated airway smooth muscle preparations lacking the epithelium. Life Sci 1986;38:809-16. 25. Stuart-Smith K, Vanhoutte PM. Airway epithelium modulates the responsiveness of porcine bronchial smooth muscle. J Appl Physiol 1988;65:721-7. 26. Raebum D, Hay DWP, Farmer SG, Fedan JS. Epithelium removal increases the reactivity of human isolated tracheal muscle to methacholine and reduces the effect of verapamil. Eur J Pharmacol 1986;123:451-3. 27. Stuart-Smith K, Vanhoutte PM. Heterogeneity in the effects of epitheiium removal in the canine bronchial tree. J Appl Physiol 1987;63:2510-15. 28. Gao Y, Vanhoutte PM. Removal of the epithelium causes depoiarization of canine bronchial smooth muscle [Abstract]. Physiologist 1987;30:240. 29. Advenier C, Devillier P, Matran R, Naline E. Influence of epithelium on the responsiveness of guinea pig-isolated trachea to adenosine. Br 3 Pharmacol 1988;93:295-302. 30. Flavahan NA, Danser AJ, Vanhoutte PM. Arachidonic acid and calcium ionophore cause epithelium-dependent relaxation of canine bronchial smooth muscle. Proceedings of the International Union of Physiological Sciences, Vancouver, B.C., Canada, July 13-18, 1986. 31. Stuart-Smith K, Vanhoutte PM. Arachidonic acid evokes epithelium-dependent relaxations of canine bronchi [Abstract]. Pharmacologist 1987;29:365. 32. Nijkamp FP, Folkerts G. Reversal of arachidonic acid-induced guinea pig tracheal relaxation into contraction after epithelium removal. Eur J Pharmacol 1987;131:315-6. 33. Farmer SG, Hay DWP, Raebum D, Fedan JS. Relaxation of guinea pig tracheal smooth muscle to arachidonate is converted
Epithelium-derived
relaxing
factor
to contraction following epithelium removal. Br J Pharmacol 1987;92:231-6. 34. Flavahan NA, Vanhoutte PM. The respiratory epitbelium releases a smooth muscle relaxing factor. Chest 1985;87 (suppl): 189S-90s. 35. Brunelleschi S, Haye-Legrand I, Labat C, Norel X, Benveniste J, Brink C. Platelet-activating factor-acether-induced relaxation of guinea pig airway muscle: role of prostaglandin I& and the epithelium. J Pharmacol Exp Ther 1987;243:356-63. 36. Ilhan M, Sahin I. Tracheal epithelium releases a vascular smooth muscle relaxant factor: demonstration by bioassay, Eur J Pharmacol 1986;131:293-6. 37. Shikano K, Berkowitz BA. Endothelium-derived relaxing factor is a selective relaxant of vascular smooth muscle. J Pharmacol Exp Ther 1987;243(1):55-60. 38. Ruff F, Zander JF, Edoute Y, Santais MC, Flavahan NA, Verbeuren TJ, Vanhoutte PM. Beta,-adrenergic responses to tulobuterol in airway smooth muscle, vascuiar smooth muscle, and adrenergic nerves [in press]. J Pharmacol Exp Ther. 39. Xue Q-F, Maurer R, Engel G. Selective distribution of betaand alpha,-adrenoceptors in rat lung visualized by autoradiography. Arch Int Pharmacodyn Ther 1983;266:308-14. 40. Carstairs JR, Nimmon AJ, Barnes PJ. Autoradiographic localisation of B-adrenoceptors in human lung. Em J Pharmacol 1984;103:189-90. 41. Carstairs JR, Nimmon AJ, Barnes PJ. Autoradiographic visualization of B-adrenoceptor subtypes in human lung. Am Rev Respir Dis 1985;132:541-7. 42. Argarwal DK, Schugel JW, Townley RG. Comparison of beta adrenoceptors in bovine airway epithelium and smooth muscle cells. Biochem Biophys Res Commun 1987;148:178-83, 43. Busk M, Vanhoutte PM. The methylxanthine, S 9795, causes the release of an epithelium-derived relaxing factor [Abstract]. Clin Res 1988;36:59OA. 44. Empey DW, Laitinen LA, Jacobs L, Gold WM, Nadel JA. Mechanisms of bronchial hyperreactivity in normal subjects after upper respiratory tract infection. Am Rev Respir Dis 1976;113:131-9. 4.5. Boushey HA, Holtzmann MJ: Sheller JR, Nadel JA. Bronchial hyperreactivity. Am Rev Respir Dis 1980;121:389-413. 46. Boushey HA, Holtzman MJ. Autonomic regu!ation of airways: parasympathetic system. In: Weiss EB, ed. Bronchial asthma: mechanisms and therapeutics. 2nd ed. Boston: Little, Brown, 1985. 47. Cockcroft DW, Ruffin RE, Hargrave RE. Allergin-induced increase in nonallergic bronchial reactivity. CIin Allergy 1977;7:503-13. 48. Holtzman MJ, Fabbri LM, O’Byme PM et al. Importance of airway inflammation for hyperresponsiveness induced by ozone. Am Rev Respir Dis 1983;127:686-90. 49. Laitinen LA, Elkm RB, Empey DW, et al. Changes in bronchial reactivity after administration of live attenuated influenza virus. Am Rev Respir Dis 1976;113:94. 50. Gleich GJ, Loegering DA. Inununobiology of eosinophils. Annu Rev Immunol 1984;2:429-59. 51, Laitinen LA, Heino M, Laitinen A, Kava T, Haahtela T. Damage of the airway epithelium and bronchial reactivity in patients with asthma. Am Rev Respir Dis 1985;131:599-606.
ALLERGY AND
CLINICAL VOLUME
IMMUNOLOG
83
ostgraduate
5
course
pithelium-derived ronchial reactivity aul M. Vanhoutte,
relaxing factor(s)
an
MD, PhD Rochester, Minn.
In isolated blood vessels, the mechanical or enzymatic removal of the endothelium augments the contractions evoked by a variety of vasoconstrictor agents, because the endothelial cells release a powerful relaxing substance(s) (endothelium-derived relaxing factor(s)). The focal absence of epithelial cells in airways of patients with asthma is well documented. When the luminal surface of canine bronchi is rubbed gently, the only morphologic change observed is the disappearance of the epithelial cell layer. The removal of the epithelium causes an increased sensitivity of the bronchial smooth muscle to acetylcholine without alteration in the maximal responsiveness to the cholinergic transmitter. The augmentation cannot be attributed to reduced enzymatic breakdown of acetylcholine after removal of the epithelial cells, since it is not affected by inhibitors of acetylcholinesterase. It cannot be attributed to disappearance of a dtffusion barrier, since epithelium removal also augments the contractions evoked by electrical stimulation of the cholinergic nerve endings. Removal of the epithelium potentiates the contractions evoked by histamine and S-hydroxytryptamine in a manner similar to that observed for acetylcholine. The potentiating effect of epithelium removal is more pronounced in larger rather than in smaller bronchi. By contrast, the relaxations evoked by P-adrenergic agonists are less pronounced in bronchi without, rather than in bronchi with, epithelium. The influence of the epithelium on isoproterenol-induced relaxation is more pronounced in smaller rather than in larger bronchi. These observations suggest that the bronchial epithelium releases an inhibitory factor that partially counteracts activation of the airway smooth muscle by bronchoconstrictor substances. Preliminary bioassay studies confirm that the bronchial epithelium can release a potent relaxing substance that inhibits not only bronchial but also vascular smooth muscle. In larger bronchi, the epithelium-derived relaxing factor is released under basal conditions; in smaller bronchi, its release can be activated by catecholamines. The augmentation of the response to bronchoconstrictor agents by removal of the epithelium has been observed, in the author’s and other laboratories, in isolated airways from different species. (.I ALLERGYCLIN IMMUNOL 1989;83:85.5-61.)
From the Department of Physiology and Biophysics, Mayo Clinic and Mayo Foundation, Rochester, Minn. Supported in part by National Institutes of Health Grant HL 21584. Reptin? requests: Paul M. Vanhoutte, MD, PhD, Department of Physiology and Biophysics, Mayo Clinic and Mayo Foundation, Rochester, MN 55905.
A dysfunctional respiratory epithelium may explain in part bronchial hyperreactivity in asthma and airway infection. 1,2 in the vascular system the e~dothelial cells can affect the tone of the underlying smooth muscle by releasing relaxing and contacting factor(s).3-8When experiments were designed first to de55
Vanhoutte
. 1. Histologic bronchi removes
section demonstrating epithelial ceils (right).
that gentle rubbing of the iuminal (From reference 12. By permission.)
surface
of canine
Epithelium: orwithout o=with
Histamine,-log FIG. 2. Effects of removal and 5-hydroxytryptamine
M
5-HT,-log
M
of the epithelium on concentration-response curves to histamine (kefti (5-HT, right) in canine bronchi. (From reference 11. By permission.)
termine whether the bronchial epithelium modulates the responsiveness of airway smooth muscle, they demonstrated that removal of the epithelial cells caused a hyperresponsiveness to acetylcholine, histamine, and serotonin9‘” The present article updates earlier brief overviews summarizing these data that are in support of these initial observations and strengthen the concept that the release of relaxing factor(s) from epithelial cells contributes to the regulation of bronchomotor tone.12-‘5
Rubbing of the mucosal surface of canine bronchi removes the epithelial cells without disturbing the underlying layers (Fig. I).” Thus, the responsiveness of isolated bronchi (or tracheas) with or without ep-
ithelium can be compared, allowing the dete~inati~~ of the role of the epithelium in controlling airway smooth muscle. In canine bronchi, removal of the epithelium causes a shift to tbe left of the doseresponse curves to the bronchoconstrictors, histamine, 5-hydroxytryptamine (Fig. 2) and acetylcholine (Fig. ,).I’ These observations in canine bronchi implied that the absence of epithelial cells potentiate the reactivity of the underlying smooth muscle, They have been extended to different species, including cattle,16 guinea pigs,‘7-23 rabbits,24 pigs,25 and humans26 with different bronchoconstrictors, including allergen. Thus, there can be little doubt that removal or absence of the epithelium can lead to exaggerated constriction of airways. However, although the phenom-
VOLUME NLrMBER
Epithelium-derived
83 5
relaxing
factor
Epi~~~lium:
Epithelium:
o=without
--Control
e-with
-Ecothiopate,
ith, 10m7 M
lOO%=
0 Without,
a3
12.62
2,
loo%=?.39
n=6
Acetylcholine,-log
M
FIG. 3. Effects of epithelium removal on responsiveness of canine bronchial rings to acetylcholine in control preparations (dashed lines) and after echothiopate (1Ue7 mol/L) &o/id linesj in canine bronchi. (From reference 11. By permission.)
enon cannot be disputed, the mechanism underlying the phenomenon remains a matter of interpretation. ~pith~iium~derived or reality?
relaxing
factor:
Myth
The hyperresponsiveness evoked by the removal of I Ihe epithelium cannot be explained by a nonspecific increase in the sensitivity of the bronchial smooth muscle. Indeed, the dose-response curve to potassium chloride is not affected (Fig. 4); neither are the maximal responses to acetylcholine, histamine, or serotonin (Figs. 2 and ,).‘I, 16*I7327Also, the removal of the epithelium does not affect the resting membrane potential in canine bronchial smooth muscle and does not affect the concentration-depolarization curve to potassium ioms.28 Enzymatic breakdown within the epithelial cells is not involved since the epithelial cells do not contain acetylcholinesterase (unpublished observations), and the removal of epithelium augments responses to acetylcholine after incubation with inhibitors of acetylcholinesterase.” Inhibition of monoamine oxidase also does not prevent the potentiation of the response to Shydroxytyptamine.’ However, the epithelium may play a role in the inactivation of adenosine.29 The disappearance of a diffusion barrier theoretically could be a logical explanation for the potentiation induced by removal of the epithelium. However, this interpretation appears unlikely in view of the fact that the hyperresponsiveness on removal of the epithelium has been demonstrated in rings of airways where the bronchoconstrictors reach the bronchial smooth muscle from the outside.” Moreover, in canine bronchi, removal of the epithelium also potentiates constric-
=i .-E 2
4
E z s ,’
0
10
20
30 K’,
40
60
120
m
FIG. 4. Contractions evoked by increasing concentrations of potassium chloride (K+) in rings with (O - - - - O) and without (e---e) epithelium prepared from the same canine bronchi. These data are presented as means I SEM. No significant differences were noted between the two groups, implying that the procedure to remove the epithelium has not altered the ability of the bronchial smooth muscle to be activated directly. (From reference 15, By permission.)
tions evoked by electrical stimulation of the cholinergic nerve endings that release acetylcboli~e in the immediate vicinity of the bronchial smooth muscle (Fig. 5). A major reason to reject the hypothesis that the epithelium acts as a passive diffusion barrier to bronchoconstrictor agonists is that removal of the epithelium attenuates relaxations induced by isoproterenol “, 16,27 arachidonic acid,30-33the calcium ionophor; A23187,” and the calcium antagonist, verapami1.24,26 In analogy with the blood vessel wa11,3,6-7the observations on bronchi with and without ep~t~elium are explained best if the epithelial cells were to release an inhibitory factor(s) (epithelium-derived relaxing factor(s)) that controls the tone of the underlying smooth muscle.1’.34Bioassay studies in canine bronchi indicate that indeed epithelial cells release a fa,ctor(s) that relaxes airway smooth muscle (Fig. 6). 12*3” However, such bioassay studies are not possible if the airways are taken from animals with ~ea~wo~s or
37”6 ~rebs-R~~~~r solution at 2ml/min
Epithelium:
++i Direct superfusion without
__
_
__
a Hz
4 Hz
2Hz
37% aerated organ clamber
16 Hz
Bronchus Propranolol, 6x 1Om6
wit
ronchus Mtithoert epithelium
M
Acetylcholine
without
fi -__-~ 2 Hz
JI
JI-
4 Hz
JI-
a Hz
’ 3g
16 Hz
FIG. 5. Comparison of the response to electrical activation of the cholinergic nerve endings of rings, with and without epithelium, of the same canine bronchi. In the rings without epithelium the initial peak response was larger (left) and the delay of the response on prolonged stimulation (fade;right) less pronounced. (From reference 11. By permission.)
infection of the airways (unpublished results). Epitbelial cells (which are highly polarized) could release tbe factor(s) preferentially toward the underlying smooth muscle rather than in the airway lumen, and mucus could inactivate the factor(s). Alternatively, epithelium-derived relaxing factor may have a very short half-life. The nature of the inhibitory mediator released from the airway epithelium is unknown, and more than one factor may be involved. In the guinea pig, but not in the dog and in cattle, products of the metabolism of arachidonic acid may contribute to the inhibitory role of the epithelial cells.“* 16,19,2’ They must contribute to epithelium-dependent relaxations induced by arachidonic acid and platelet-activating factor.23,31-33* 35 Factors released from the epithelium cause relaxation of isolated blood vessels.“, 36 The epitheliumderived factor that relaxes vascular smooth muscle is not a product of cycIooxygenase.36 Caution should be exerted when the factor(s) derived from the epithelium that can be bioassayed with isolated blood vessels is assumed to be the one causing inhibition of airway smooth muscle. Indeed, vascular and bronchial smooth muscle are different in their responsiveness to endogenous-relaxing substances .37 The similar parallel effect of epithelium removal on concentration-contraction curves to acetylcholine, histamine, and 5-hydroxyhyptamine implies that the release of the relaxing factor(s) is continuous rather than caused by activation of specific epithelial
Superfusion mode:
epithelium direct
epitheiium direct
direct
QPf Acetylcholine,
3xlO%
FIG. 6. Bioassay experiment in which a ring of canine bronchus without epithelium (contracted with acetylcholine) is superfused with Krebs-Ringer bicarbonate solution flowing either through a bronchus with epithelium (~7~8 arrow; note the marked relaxation), a stainless steel tube (direct; note the reversal of the relaxation), or a bronchus without epithelium (note the absence of relaxation]. This demonstrates that the bronchial epithelium can reiease a potent inhibitory factor. (From reference 12. By permission.)
receptors. However, the increased responsiveness to P-adrenergic agonists in intact, compared with epithelium-denuded bronchi,“, 27,3x suggests tbat activation of P-adrenoceptors on the epithelial cells (which have a high density of P-adrenoceptors, compared to airway smooth muscle’7, 39,40-42)triggers the active release of epithelium-derived relaxing factor (Fig. 7). A similar conclusion can be reached for arachidonic acid,23,30-32 platelet-activating factorY3” and methylxanthines.43 Heterogeneity In airways of decreasing diameter, the ~otentiatin~ effect of epithelium removal on the response to adrenergic activation is augmented (Fig. 8) .27A similar trend is noted in porcine airways.25 However, with
VOLUME NUMBER
Epithelium-derived
83 5
relaxing
factor
FIG. 7. Proposed interaction between epithelial cells and bronchial smooth muscle. Under basal conditions the epithelium secretes epithelium-derived relaxing factor that acts as a functional antagonist on the bronchial smooth muscle, causing a comparable depression of the response to a variety of bronchoconstrictor agonists. The release of the factor can be activated by stimulation of B-adrenoceptors on the endothelial cells. ACh, acetylcholine; B, p-adrenoceptors; H, histaminic receptor; 5-UT, serotonergic receptor; and M, muscarinic receptor. (From reference 15. By permission.)
2nd order
Epithelium: o.with ozwithout
+9
9
8
7
4
5
6
n=Ei
4th order
3rd order
80 100 120 ,401 9
Isoproterenol,
8
7
6
5
4
- log M
FIG. 8. Comparison in lobar (2nd order), segmental (3rd order), and subsegmental (4th order) bronchi of the same dogs of the effect of epithelium removal on the relaxations evoked by increasing concentrations of isoproterenol. Note that with decreasing diameter of the bronchus, the potentiating effect of the removal is progressively augmented. These studies support the hypothesis that the triggered release of epithelium-derived relaxing factor is less in larger than in smaller airways. (From reference 27. By permission.)
0
vanhautte
2nd order
3rd order
W-iydroxytryptamine,-log FIG. 9. Comparison in lobar (Zndorderj, segmental (3rdorder), subsegmental {4tb order) bronchi of the same dogs of the effect of epithelium removal on the contractions evoked by increasing concentrations of 5-hydroxytryptamine. Note that with decreasing diameter of the bronchus, the potentiating effect of the removal is progressively reduced, These studies support the hypothesis that the basal release of epithelium-derived relaxing factor is greater in larger than in smaller airways. (From reference 27. By permission.)
decreasing diameter, the modulating influence of the epithelium on the response to bronchoconstrictors becomes less prominent (Fig. 9).27 Thus, if the airway epithelium releases a factor (or factors) that promotes relaxation. of bronchial smooth muscle, a marked heterogeneity exists in the release or the effect of the factor(s) along the bronchial tree. Thus, the absence of potent&ion of bronchoconstrictor effects with epithelium removal in a given airway preparation can not be used against the hypothesis that the epithelium can control the underlying bronchial smooth muscle.
If the release of epithelial-derived inhibitory factor(s) occurs in the intact organism, the absence or dysfunction of the epithelium could contribute to abnormal responses of the bronchial wall. Stimuli that induce hyperreactivity of the bronchi, such as exposure to ozone, allergen, or respiratory viruses, cause acute inflammation of the epithelium.” ‘. 44,45-49The major basic protein of eosinophils, which is involved in certain types of inflammation and hyperreactivity of the airways,50 augments the responsiveness of isolated airways if epithelial cells are present, but not in their absence.‘* Thus, malfunction or absence of the epithelium may cause bronchial hyperreactivity. This interpretation is supported by the fact that epithelial destruction is prominent in patients with asthma.51 The facilitatory role of the epithelium on the response to P-adrenergic agonists and methylxanthines could be of particular importance when these therapeutic agents are administered by aerosol.” 1 thank Mrs. H. Hendricksen and Mr. R. Lorenz for preparing the figures, and Ms. K. Kros for the secretarial assistance.
REFEREt’lCES 1. Nadel JA. Bronchial reactivity. Adv intern Med 1983;28:20723. 2. Hogg JC, Eggleston PA. Is asthma an epithelial disease? Am Rev Respir Dis 1984;129:207-8. 3. Furchgott RF. The role of endothelium in responses of vascular smooth muscle to drugs. Ann Rev Pharmacol Toxicol 1984;24:175-97. 4. Rubanyi GM, Vanhoutte PM. Hypoxia releases a vasoconstrictor substance from the canine vascu!ar endothelium. J Physiol 1985;364:45-56. 5. Furchgott RF, Zawadzki JV. The obligatory role of endolhelial cells in the relaxation of arterial smooth muscle by acetyleholine. Nature 1980;288:373-6. 6. Vanhoutte PM, Rubanyi GM, Miller VM, Houston DS. Modulation of vascular smooth muscle contraction by the endothelium. Annu Rev Physiol 1986;48:307-20. 7. Vanhoutte PM. Endothelium and the control of vascular tissue. News in Physiol Sci 1987;2:18-22. 8. Vanhoutte PM. Endothelium-dependent contracrions in arteries and veins. Blood Vessels 1987b;24:141-4. 9. Aarhus LL, Rimele TJ, Vanhoutte PM. Removal of the epithelium causes bronchial supersensitivity Lo acetylcholine and 5-hydroxytryptamine. Fed Proc 1984;43:955. 10. Flavahan NA, Vanhoutte PM. Epitheiial-dependent attenuation of bronchial smooth muscle lone. Fed Proc 1984;43:429. 11. Flavahan NA, Aarhus LL, Rimele TJ, Vanhoutte PM. The respiratory epithelium inhibits bronchial smooth muscle tone. J Appl Physiol 1985;58:834-8. 12. Vanhoutte PM. Airway epithelium and bronchial reactivity. Can J Physiol Phamacol 1987;65:44&-50. 13. Farmer SG. Airway smoorb muscle responsiveness: modulation by the epithelium. Trends Pharmacol Sci 1987$X:5-10. 14. Vanhoutte PM, Flavahan NA. Modulation of cholinergic neuretransmission in the airways. In: Kaliner MA, Bares PJ, eds. The airways: neural control in health and disease. New York: Marcel Dekker, chap 9, 1987:203-16. 15. Vanhoutte PM. Epithelium-derived relaxing factor(s) a& bronchial reactivity. Ann Rev Respir Dis 1988;138:S24-S30. 16. Barnes PJ, CUSSFM, Palmer JB. The effect of airway epitbelium on smooth muscle contractility in bovine trachea. Br J Pharmacol 1985;86:684-91,
VOLUME NUMBER
83 5
17. Goldie RG. Papadimitriou JM, Paterson WJ, Rigby PJ, Self NM, Spina D. Influence of the epithelium on responsiveness of guinea pig isolated trachea to contractile and relaxant agonists. Br .I Pharmacol 1986;87:5-14. 18. Hay DWP, Robinson VA, Fleming WW, Fedan JS. Role of the epithehum in contractile responses of the guinea pig isolated trach.ea. Fed Proc 1985;44:506. 19. Hay DWP, Farmer SG, Raebum D, Robinson VA, Fleming WW, Fedan JS. Airway epithelium modulates the reactivity of guinea pig respiratory smooth muscle. Eur J Pharmacol 1986;129:11-18. 20. Hay DWP, Raebum D, Farmer SG, Fleming WW, Fedan JS. Epithelium modulates the reactivity of ovalbumin-sensitized guinea pig airway smooth muscle. Life Sci 1986;38:2461-8. 21. Holroyde MC. The influence of epithelium on the responsiveness of guinea pig isolated trachea. Br J Pharmacol 1986; 87501-7. 22. Vanhoutte PM, Flavahan NA, Slifman NF, Gleich GJ. Eosinophil granule major basic protein causes hyperreactivity of respiratory muscle in the presence, but not in the absence of, epithelial cells. Proceedings of the International Union of Physiological Sciences, Vancouver, B.C., Canada, July 13-18, 1956. 23. Tschirhart E, Frossard N, Bertrand C, Landry Y. Arachidonic acid metabolites and airway epithelium-dependent relaxant factor. J Phannacol Exp Ther 1987;243:310-316. 24. Raebum D, Hay DWP, Robinson VA, Farmer SG, Fleming WW, Fedan JS. The effect of verapamil is reduced in isolated airway smooth muscle preparations lacking the epithelium. Life Sci 1986;38:809-16. 25. Stuart-Smith K, Vanhoutte PM. Airway epithelium modulates the responsiveness of porcine bronchial smooth muscle. J Appl Physiol 1988;65:721-7. 26. Raebum D, Hay DWP, Farmer SG, Fedan JS. Epithelium removal increases the reactivity of human isolated tracheal muscle to methacholine and reduces the effect of verapamil. Eur J Pharmacol 1986;123:451-3. 27. Stuart-Smith K, Vanhoutte PM. Heterogeneity in the effects of epitheiium removal in the canine bronchial tree. J Appl Physiol 1987;63:2510-15. 28. Gao Y, Vanhoutte PM. Removal of the epithelium causes depoiarization of canine bronchial smooth muscle [Abstract]. Physiologist 1987;30:240. 29. Advenier C, Devillier P, Matran R, Naline E. Influence of epithelium on the responsiveness of guinea pig-isolated trachea to adenosine. Br 3 Pharmacol 1988;93:295-302. 30. Flavahan NA, Danser AJ, Vanhoutte PM. Arachidonic acid and calcium ionophore cause epithelium-dependent relaxation of canine bronchial smooth muscle. Proceedings of the International Union of Physiological Sciences, Vancouver, B.C., Canada, July 13-18, 1986. 31. Stuart-Smith K, Vanhoutte PM. Arachidonic acid evokes epithelium-dependent relaxations of canine bronchi [Abstract]. Pharmacologist 1987;29:365. 32. Nijkamp FP, Folkerts G. Reversal of arachidonic acid-induced guinea pig tracheal relaxation into contraction after epithelium removal. Eur J Pharmacol 1987;131:315-6. 33. Farmer SG, Hay DWP, Raebum D, Fedan JS. Relaxation of guinea pig tracheal smooth muscle to arachidonate is converted
Epithelium-derived
relaxing
factor
to contraction following epithelium removal. Br J Pharmacol 1987;92:231-6. 34. Flavahan NA, Vanhoutte PM. The respiratory epitbelium releases a smooth muscle relaxing factor. Chest 1985;87 (suppl): 189S-90s. 35. Brunelleschi S, Haye-Legrand I, Labat C, Norel X, Benveniste J, Brink C. Platelet-activating factor-acether-induced relaxation of guinea pig airway muscle: role of prostaglandin I& and the epithelium. J Pharmacol Exp Ther 1987;243:356-63. 36. Ilhan M, Sahin I. Tracheal epithelium releases a vascular smooth muscle relaxant factor: demonstration by bioassay, Eur J Pharmacol 1986;131:293-6. 37. Shikano K, Berkowitz BA. Endothelium-derived relaxing factor is a selective relaxant of vascular smooth muscle. J Pharmacol Exp Ther 1987;243(1):55-60. 38. Ruff F, Zander JF, Edoute Y, Santais MC, Flavahan NA, Verbeuren TJ, Vanhoutte PM. Beta,-adrenergic responses to tulobuterol in airway smooth muscle, vascuiar smooth muscle, and adrenergic nerves [in press]. J Pharmacol Exp Ther. 39. Xue Q-F, Maurer R, Engel G. Selective distribution of betaand alpha,-adrenoceptors in rat lung visualized by autoradiography. Arch Int Pharmacodyn Ther 1983;266:308-14. 40. Carstairs JR, Nimmon AJ, Barnes PJ. Autoradiographic localisation of B-adrenoceptors in human lung. Em J Pharmacol 1984;103:189-90. 41. Carstairs JR, Nimmon AJ, Barnes PJ. Autoradiographic visualization of B-adrenoceptor subtypes in human lung. Am Rev Respir Dis 1985;132:541-7. 42. Argarwal DK, Schugel JW, Townley RG. Comparison of beta adrenoceptors in bovine airway epithelium and smooth muscle cells. Biochem Biophys Res Commun 1987;148:178-83, 43. Busk M, Vanhoutte PM. The methylxanthine, S 9795, causes the release of an epithelium-derived relaxing factor [Abstract]. Clin Res 1988;36:59OA. 44. Empey DW, Laitinen LA, Jacobs L, Gold WM, Nadel JA. Mechanisms of bronchial hyperreactivity in normal subjects after upper respiratory tract infection. Am Rev Respir Dis 1976;113:131-9. 4.5. Boushey HA, Holtzmann MJ: Sheller JR, Nadel JA. Bronchial hyperreactivity. Am Rev Respir Dis 1980;121:389-413. 46. Boushey HA, Holtzman MJ. Autonomic regu!ation of airways: parasympathetic system. In: Weiss EB, ed. Bronchial asthma: mechanisms and therapeutics. 2nd ed. Boston: Little, Brown, 1985. 47. Cockcroft DW, Ruffin RE, Hargrave RE. Allergin-induced increase in nonallergic bronchial reactivity. CIin Allergy 1977;7:503-13. 48. Holtzman MJ, Fabbri LM, O’Byme PM et al. Importance of airway inflammation for hyperresponsiveness induced by ozone. Am Rev Respir Dis 1983;127:686-90. 49. Laitinen LA, Elkm RB, Empey DW, et al. Changes in bronchial reactivity after administration of live attenuated influenza virus. Am Rev Respir Dis 1976;113:94. 50. Gleich GJ, Loegering DA. Inununobiology of eosinophils. Annu Rev Immunol 1984;2:429-59. 51, Laitinen LA, Heino M, Laitinen A, Kava T, Haahtela T. Damage of the airway epithelium and bronchial reactivity in patients with asthma. Am Rev Respir Dis 1985;131:599-606.