The incidence of peripheral pulmonary adeno-carcinoma has increased in recent years. Clara cellhas been known as target for carcinogens andsource of pulmonary tumors. One of the pre-sumed roles of the bronchiolar Clara cell is thesecretion of pulmonary surfactant into the bron-chiolar lumen. To establish the secretory mor-phology of Clara cell, a well-defined secretoryagonist, isoproterenol (500 mg/kg) and the antago-nist, propranolol (20 mg/kg), were administered intofive-week old mice. The secretory response wasexamined at 1 hour and 4 hours after injection.Ultrastructural morphometry was used to quanti-tate the secretory response by measuring area ofapical cap of the Clara cells. Isoproterenol causeda significant increase in area of apical cap ofClara cells 1 and 4 hours after injection (p＜0.0001), while pretreatment with propranolol pre-vented this effect at 4 hours. Propranolol alone sig-nificantly decreased the area of Clara cells (p＜0.0001). Clara cells secretory granules disap-peared 1 hour after propranolol plus isoproterenoladministration, and the granules reappeared at 4hours. The accelerated secretion of Clara cells byisoproterenol provides evidence of their secretorymechanism controlled by beta-adrenergic ago-nists. The study has confirmed the secretory role

of Clara cells. The secretion is both apocrine andmerocrine type.

Key words: Clara cell, isoproterenol, lung adeno-carcinoma, beta-antagonist

Introduction

Clara cells are the principal epithelial cell phenotypepresent in the distal airway in many species1 and arethe primary cellular site of xenobiotic metabolism by thecytochrome p-450 (CYP) monooxygenase system inthe lung.2,3 These cells represent the predominantsecretory cell within distal conducting airways ofmammals and exhibit functional alterations withchronic pulmonary diseases in human. Moreover,84% of adenocarcinomas of human lung are the non-ciliated bronchiolar cell type4, and 85% of peripheraladenocarcinomas of human lungs has Clara cellgranule.5 The functional roles of Clara cell secretionsthat are induced by exposure to various environmentalagents are poorly understood. One of the presumedfunctions of Clara cells6 and type II pneumocytes7 is tosynthesize and secrete pulmonary surfactant, which isparticularly important during birth when lung expansionis required. The intracellular mechanism involved inmediating the secretion of Clara cells by beta-adrener-gic receptor (β-AR) agonist is via cyclic 3’, 5’ adenosinemonophosphate-dependent protein kinase A.Activation of these receptors mediates mucus secre-tion, surfactant secretion, and airway smooth musclecell relaxation.8

Kolliker9 first described the cell types in the terminal

Original Article

Ultrastructure of Clara cells stimulated by isoproterenol

Gopi Aryal1, Yuji Kimula2 and Morio Koike1

1) Department of Human Pathology, Digestive and Metabolic Disease, TokyoMedical and Dental University, Graduate School, Tokyo, Japan2) Anatomical Pathology Service, Ome Municipal General Hospital, Tokyo, Japan

J Med Dent Sci 2003; 50: 195–202

Corresponding Author: Morio Koike, MD, PhD, Department ofHuman Pathology, Digestive and Metabolic Disease, Tokyo Medicaland Dental University, Graduate School, 1-5-45 Yushima, Bunkyo-ku,Tokyo 113-8519, JapanTel: (+81)-3-5803-5172, Fax: (+81)-3-5803-0123E-mail: koike.pth1@tmd.ac.jpReceived April 30; Accepted June 13, 2003

bronchioles in the dog, and a more detailed study inman and in the rabbit was conducted by Clara.10 Clarasuggested that the non-ciliated epithelial cells of thedistal bronchioles were secretory cells releasing anon-mucoid substance by an apocrine mechanism. Theapocrine mechanism was later supported by otherresearchers.11,12 On the other hand, Yoneda13 sug-gested the process of exocytosis or merocrine type ofsecretion, while others illustrated the secretion asbeing of both apocrine and merocrine types.14,15 Themechanism of secretion of Clara cells is still controver-sial.

Previous studies have shown that isoproterenol,one of the β-AR agonists, stimulates secretion by ratClara cells.16,17 However, the morphological changesafter stimulating with well-defined secretory agonisthave not been demonstrated. The purpose of thisstudy is to provide the ultrastructural evidence ofsecretory activity of Clara cells by inducing an acceler-ated secretion with isoproterenol stimulation.

Materials and Methods

Mice and experimental designFive-week old male C57BL/6J mice, weighing 14-

20 g, were purchased from Nippon Crea Inc Co(Tokyo, Japan). The animals were randomly assignedto 7 groups (A-G) consisting of 5 animals in each.

Group A (iso1) received a single intraperitonealinjection of 500 mg isoproterenol hydrochloride(Sigma, St Louis, USA) per kilogram body weight,equivalent to 20% of LD50 in mice18 and was sacrificedafter 1 hour. Group B (iso4) received the same regimenas the first group and was sacrificed after 4 hours.

Group C (pro+iso1) received a single subcutaneousinjection of 20 mg of propranolol hydrochloride(Sigma, St Louis, USA) per kilogram body weight,equivalent to 20% of LD50 in mice19, followed by 500mg/kg body weight of isoproterenol after 10 minutes.The animals were sacrificed after 1 hour. Group D(pro+iso4) received the same regimen as group C andwas sacrificed after 4 hours.

Group E (pro1) received a single subcutaneousdose of 20 mg per kilogram body weight of propranololand was sacrificed after 1 hour. Group F (pro4)received the same regimen as that of E, and was sac-rificed after 4 hours. Control animals (Group G) wereinstilled with saline alone and sacrificed 1 hour afterinjection. Animals were rapidly killed by cervical dislo-cation and exsanguinations.

The trachea was cannulated and the lungs werefixed by intratracheal instillation of 1% gluteraldehydesolution buffered with 0.1M phosphates at pH 7.2. Thetrachea, heart, and lung were rapidly removed from thethoracic cavity and immersed in the same fixative for atleast 48 hours. The left lobe of the lung wasprocessed for scanning electron microscopy and rightlower lobe for transmission electron microscopy. Theright upper, middle, and cardiac lobes wereprocessed for histological examination. Animal experi-mentation was performed in accordance with theguidelines established by the Animal ExperimentalCommittee of the Tokyo Medical and Dental University.

Scanning Electron MicroscopySpecimens from the left lung lobe were dehydrated in

a graded ethanol series of 70%, 95%, and 100% fol-lowed by n-pentyl acetate. The lung lobe underwentcritical point drying system using liquid carbon dioxide,JCPD-5 (JEOL, Japan). After sputtering of gold parti-cles for 8 minutes using an ion-sputtering device,JFC-1100 (JEOL, Japan), the tissue was examined witha scanning electron microscope, JSM-T20 (JEOL,Japan).

Transmission Electron MicroscopyThe right lower lung lobe was cut into blocks mea-

suring about 2-3 cubic millimeters and post-fixed with2% osmium tetraoxide buffered in 0.1M phosphates atpH 7.2 for 2 hours. After dehydration by gradedethanol series, the tissues were embedded in Epon-epoxy resins. Ultrathin sections cut with an ultramicro-tome, Ultracut E (Leica, Germany) were doublystained with uranyl acetate and lead citrate, andexamined with a transmission electron microscope,TEM-100SX (JEOL, Japan).

Morphometry The area of Clara cells was measured using an

image analyzing software, Scion Image (ScionCorporation, MD, USA) from the scanning electronmicrographs of the terminal bronchioles of the mice.Clara cells were identifiable by their typical domeshaped apical cytoplasm. A line was drawn along theborder of apical cap of each Clara cell to be assessedusing computer mouse. The area of apical cap of Claracells was automatically determined. Total of 4550Clara cells were measured, ranging from 424-760cells in each group. The images were counted ran-domly in order to decrease bias based on drugs or timeintervals after injection. Clara cells lying obliquely or

G. ARYAL, Y. KIMULA and M. KOIKE J Med Dent Sci196

cells area of either less than 8 or more than 80 squaremicrometers were omitted from counting. We pre-sumed that the Clara cells with area less than 8square micrometers possibly represents detachedapical blebs where as surface area with more than 80square micrometers represents the apical cytoplasmyet to be detached during secretion.

Statistical analysisFor each parameter measured or calculated, the val-

ues for individual animals averaged per experimentalgroup and standard error of group mean (S. E.) calcu-lated. An analysis of variance was used for multiplegroup comparisons and an unpaired Student’s t-test tocompare means of two groups. Differences betweentwo groups were considered significant at p value lessthan 0.05. Computation was performed using StatView Software (SAS Institute, 100 SAS Campus DriveCary, USA)

Results

Histologically, the Clara cells of isoproterenol-treatedmice showed large volume of apical cytoplasm pro-truding into the bronchiolar lumen. The cytoplasmwas intensely eosinophilic. Dome-shaped projection ofClara cells into the bronchiolar lumen or alreadydetached apical blebs were observed inside bronchio-lar lumen in all experimental groups without significant

changes in bronchial or bronchiolar wall. It was not pos-sible to differentiate between β-AR-treated anduntreated groups.

Isoproterenol caused significant increase in thearea of apical cap of Clara cells both at 1 hour (p＜0.0001, Fig. 1) and 4 hours after injection (p＜0.0001,Fig. 2). Although pretreatment of propranolol (10 min-utes before isoproterenol-treatment) had no effect onisoproterenol-stimulated secretion after 1 hour (Fig. 1),propranolol-pretreatment prevented isoproterenoleffect at 4 hours (p＜0.0001, Fig. 2). Propranolol alonecaused a significant decrease in the area of apical capof Clara cells both at 1 hour (p＜0.0001, Fig. 1) and 4hours after injection (p＜0.0001, Fig. 2).

Dome-like projection of the apical cap of Clara cellsin control group showed irregular surface (Fig. 3a). Thesurface appeared rather smooth (Fig. 3b) in isopro-terenol-treated mice at 4 hours.

The Clara cells in control mice appeared ascuboidal cells showing irregular shaped nuclei at thebasal part of the cytoplasm (Fig. 4). Round to rodshaped mitochondria were scattered throughout thecells. There were abundant cisterns of rough andsmooth endoplasmic reticulum (rER, sER) . The apicalcytoplasm of Clara cells showed various sized electrondense secretory granules mainly collected beneath theterminal web of the cells. After 1 hour of isoproterenoltreatment, the Clara cell showed protrusion andpinching-off of cytoplasmic buds containing the densesecretory granules (Fig. 5). The pinching-off process

197CLARA CELLS STIMULATED BY ISOPROTERENOL

Fig. 1. Effect of isoproterenol, propranolol, and pretreated propra-nolol on the response to isoproterenol at 1 hour after injection.Mean±S.E are given and the numbers within the blocks indicate thenumber of Clara cells. Iso (1) and Pro+Iso (1) mean sacrificing hourof animals after injection of respective drugs. NS= not significant.

Fig. 2. Effect of isoproterenol, propranolol, pretreated propranolol onthe response to isoproterenol 4 hours after injection. Mean±S.E aregiven and the numbers within the blocks indicate the number of Claracells. Iso (4) and Pro+Iso (4) mean sacrificing hour of animals afterinjection of respective drugs.

G. ARYAL, Y. KIMULA and M. KOIKE J Med Dent Sci198

Fig. 3. Dome-like projection of Clara cells with rough and smooth surfaces.(a) Apical protrusion of Clara cells in control mouse showing rough surface. (b) Clara cells 4 hours after Isoproterenol administration showingsmooth surface. Each scale bar=5 µm.

(a) (b)

Fig. 4. Clara cells in control mice contain electrondense secretory granules (arrowhead) in apicalcytoplasm, sER, and dimorphic mitochondria.Uranyl acetate and lead citrate stain. Each scalebar=1 µm.

Fig. 5. Clara cells 1 hour after isoproterenol admin-istration. Note protrusion and pinching-off of cyto-plasmic buds containing dense secretory granules(long arrow) of Clara cells. Uranyl acetate and leadcitrate stain. Each scale bar=1 µm.

199CLARA CELLS STIMULATED BY ISOPROTERENOL

Fig. 6. (a) Clara cells 4 hours after isoproterenol administration. The Clara cells show electron dense secretory granules inclose proximity to the plasma membrane (arrowhead). Uranyl acetate and lead citrate stain. Each scale bar=1 µm. (b) High mag-nification of the electron dense secretory granule in figure 6a. Each scale bar=1 µm.

Fig. 7. Clara cells 1 hour after propranolol and iso-proterenol administration. Pretreated propranololcauses the discharge of Clara granules. Uranylacetate and lead citrate stain. Each scale bar=1 µm.

Fig. 8. Clara cells 4 hours after propranolol andisoproterenol administration. Reappearance ofClara granules (arrowhead) in apical cytoplasm isshown. Uranyl acetate and lead citrate stain. Eachscale bar=1 µm.

ba

was approximately three times more frequent than inthe group at 4 hours after isoproterenol treatment. At 4hours after isoproterenol treatment, the close proximityof Clara granules to the plasma membrane of the cellswere more predominant than in the group at 1 hourafter isoproterenol treatment. Intracytoplasmic degen-eration was often noted (Fig. 6). Pretreatment of pro-pranolol caused disappearance of the Clara granulesfrom cytoplasm on isoproterenol-stimulated secretionafter 1 hour (Fig. 7), whereas after 4 hours, the Claragranules reappeared in the apical cytoplasm (Fig. 8).

Discussion

There are more than seven types of bronchialepithelial cells in human airway. They are the basal,Kultschitzsky, type I alveolar cells, type II alveolarcells, ciliated cells, goblet cells, and Clara cells. Eachtype of epithelial cell may be associated with intraep-ithelial nerves, including sensory and motor, someadrenergic, and some cholinergic nerves.20 There ishigh density of β2-receptors on human airway epithelialcells.21,22 Despite the fact that β-AR agonists are sowidely used for chronic airway diseases, the function ofthese receptors is not clear. Schuller HM23 showed thatisoproterenol stimulates the growth of normal cells aswell as nitrosamine-stimulated adenocarcinoma cellswith features of Clara cells, whereas propranololappears to inhibit the growth-stimulating effect. On theother hand, β-AR agonists are known to be one of thestimulators of Clara cell secretion.17 Clara cellsecretes a variety of lipid molecules and proteins thathave surfactant properties and may play an importantrole in preventing airway closure.

In this experiment, we have shown the increase inarea of apical cap of Clara cells after stimulation with β-AR agents. Our analysis rests on the assumption thatthese cells are indeed secretory cells. The increase inthe area of apical cap is a result of secretion.15 Theassumption is based on the previous reports describingthe ultrastructural characteristics of cells, which makeand store protein destined to be secreted.24,25 Thesereports have shown that secretory cells, irrespective ofthe organs or tissue of origin, have ample rER, a well-developed Golgi apparatus, and inclusion bodies thatserve as storage granules. The Clara cell exhibits theultrastructural appearances, which include well-devel-oped sER and rER, large number of mitochondria, andpresence of membrane bound electron dense granules,indicating that it is metabolically active secretory

cells.26,27 The biochemical nature of secretory granulesis not clear, but they possess a crystalline fine structure28,are proteolytically sensitive26, and can be shown toincorporate radiolabeled amino acids.29

Yoneda30 suggested, on the basis of cytochemicalevidence, that this secretory granule protein may con-tain lipid-degrading activity, which is released into thebronchiolar lumen and catabolizes the surfactantmaterial.

Massaro et al16 found that after 1 hour of isopro-terenol treatment in rat produced a decrease in the vol-ume density of Clara cell secretory granules and theeffect was blocked by propranolol. The increase in iso-proterenol-stimulated secretion after 1 hour in ourexperiment is in accord with their findings. However,our findings are clearly at odds with the effect of pro-pranolol having no influence on isoproterenol-stimulat-ed secretion at 1 hour. It is considered that propranololhas long lag-time and a low steady-state flow after sub-cutaneous injection. The use of mice as experimentalanimals in our study, instead of rats might have con-tributed for contrasting result. The disappearance of theClara granules in the face of an unchanged rate ofsecretion could represent the initial effect of propranololcausing decreased synthesis of secretory product.However, propranolol prevented isoproterenol-stimu-lated secretion 4 hours after injection. Massaro et al17,in their continuing investigation of Clara cell, showedthat the increase in tidal volume of lung stimulatesClara cell secretion that cannot be blocked by propra-nolol 1 hour after in vitro perfusion and ventilation. Inpresent study, the decrease in surface area of Claracells 4 hours after injection of propranolol could be dueto the blockage of endogenous secretion of Claracells by β-antagonists.

Previous studies on Clara cells were centered pri-marily on their morphology and included discussion ofmechanism of secretion based on histologic observa-tion. Investigators had no consensus regarding themode of secretion of Clara cells. It is difficult to decidethe exact nature of secretion of Clara cells from thisexperiment also. However, transmission electronmicrographs at 1 hour after isoproterenol treatmentshowed pinching-off process involving the loss ofsmall amount of cytoplasm in most instances. At 4hours after isoproterenol treatment, release of Claragranules before the extrusion of cap like bodies wasevident in most cases. The apocrine secretion mostlyappeared after 1 hour whereas merocrine type ofsecretion appeared after 4 hours of isoproterenoltreatment. Our findings demonstrate that Clara cells

G. ARYAL, Y. KIMULA and M. KOIKE J Med Dent Sci200

release their secretory granules by both apocrine andmerocrine fashion simultaneously as suggested byStinson SF14 and Peao MN et al.15

It has been recently reported that the growth of ade-nocarcinoma of breast31 and pancreatic ductal cancer32

are also regulated by β-AR. Similarly, β-AR-mediatedmitogenic pathway has been identified in peripheraladenocarcinoma of the lung with features of Clara cells.Agonists of β-AR strongly stimulated cell proliferation,whereas antagonists were potent inhibitors of thiseffect.33 Most of human pulmonary adenocarcinomahas features of Clara cells4, which may be related to thelocalization of CYP containing mixed-function oxy-dase system. On the other hand, CYP2D6 isresponsible for the metabolism of propranolol.34

Whether this translates into the differentialresponse of pretreated propranolol at 1 hour and 4hours of isoproterenol-stimulated secretion is unclear.Further study is needed to clarify the role of CYP and β-AR in the Clara cells and terminal bronchiolar epitheli-um.

Acknowledgements

We thank Mr. T. Haryu and Mr. S. Aoki for their skill-ful technical assistance. We are obliged to Dr. CraigBolles, University of Wisconsin Hospital and Clinic,USA, for his kind proofreading of this manuscript.

References1. Plopper CG, Hyde DM, Buckpitt AR: Clara cells. In The lung:

Scientific Foundation. RG Crystal and JB West, editors.Raven Press, Newyork, 1991,215-228.

2. Plopper CG, Cranz DL, Kemp L, et al: Immunohistochemicaldemonstration of cytochrome p-450 monooxygenase inClara cells throught the tracheobronchial airways of the rabbit.Exp.Lung Res 1987;13:59-68.

3. Devereux TR, Domin BA, Philpot RM: Xenobiotic metabolismby isolated pulmonary cells. Pharmacol Ther 1989;41:243-256.

4. Kimula Y: A histochemical and ultrastructural study of adeno-carcinoma of the lung. Am J Surg Pathol 1978;2:253-264.

5. Ogata T, Endo K: Clara cell granules of peripheral lung can-cers: Cancer 1984;54:1635-1644.

6. Niden AH: Bronchiolar and large alveolar cell in pulmonaryphospholipid metabolism. Science 1967;158:1323-1324.

7. Kikkawa Y, Yoneda K, Smith F, et al: The type II epithelial cellsof the lung. II. Chemical composition and phospholipid syn-thesis. Lab Invest 1975;32:295-302

8. Nijkamp FP: β-adrenergic receptors in the lung: An introduc-tion. Life Sciences 1993;52:2073-2082.

9. Koelliker A: Zur Kenntnis des Baues der Lunge desMenschen: Verh physik Med Ges Wuerzburg 1881;16:1-24.

10. Clara M: Zur Histobiologie des Bronchialepithels. Z.Mikrosk.Anat. Forsch 1937;41:321-347.

11. Hung KS, Chapman AL, Hung SH, Menon CD: Clara cell sur-face of the rat: scanning and transmission electron micro-scopic study. Experientia 1982;38:959-961.

12. Smith P, Heath D, Moosavi H: The Clara cell. Thorax 1974;29:147-163.

13. Yoneda K: Pilocarpine stimulation of the bronchiolar Clara cellsecretion. Lab Invest 1977;37:447-452.

14. Stinson SF, Loosli CG: Ultrastructural evidence concerning themode of secretion of electron-dense granules by Clara cells. JAnat 1978;127:291-298.

15. Peao MN, Aguas AP, de Sa CM, Grande NR: Anatomy ofClara cell secretion: surface changes observed by scanningelectron microscopy .J Anat 1993;182: 377-388.

16. Massaro GD, Paris M, Thet LA: In vivo regulation of secretionof bronchiolar Clara cells in rats. J Clin Invest 1979;63:167-172.

17. Massaro GD, Fischman CM, Chiang MJ, et al: Regulation ofSecretion in Clara cells: studies using the isolated perfused ratlung. J Clin Invest 1981;67:345-351.

18. Goldenthal EI: A compilation of LD50 values in newborn andadult animals. Toxicol Appl Pharmacol 1971;18:185-207.

19. Martin M, Linee P: Toxicity data. Eur. J. Med.Chem;1974;9:563.

20. Reid L, Jones R: Bronchial mucosal cells. Fed Proc 1979;38:191-196.

21. Carstairs JR, Nimmo AJ, Barnes PJ: Autoradiographic visual-ization of beta-adrenoceptors subtype in human lung. Am RevRespir Dis 1985;132:541-547.

22. Davis PB, Silski CL, Kercomar P, Infeld M: Beta-adrenergicreceptors on human tracheal epithelial cells in primary culture.Am J Physiol 1990;258:C71-C76.

23. Schuller HM and Cole B: Regulation of cell proliferation bybeta-adrenergic receptors in a human lung adenocarcinomacell line. Carcinogenesis 1989;10:1753-1755.

24. Schramm M: Secretion of enzymes and other macromole-cules. Ann. Rev. Biochemn 1967,36:307-320.

25. Palade G: Intracellular aspects of the process of protein syn-thesis. Science (Wash. D.C) 1975;189:347-358.

26. Cutz E, Conen PE: Ultrastructure and cytochemistry of Claracells. Am J Pathol 1971;62:127-141.

27. Van Scott MR, Hester S, Boucher RC: Ion transport by rabbitnonciliated bronchiolar epithelial cells (Clara cells) in culture.Proc Natl Acad Sci U S A 1987;84:5496-5500.

28. Kuhn III C, Callaway LA, Askin FB: The formation of granulesin the bronchiolar Clara cells of the rat. I. Electronmicroscopy. J Ultrastruct Res 1974;49:387-400.

29. Petrik P, Collet AJ: Quantitative electron microscopic autora-diography of in vivo incorporation of 3H-choline, 3H-acetate and3H-galactose in non-ciliated bronchiolar (Clara) cells of mice.Am J Anat 1974;139:519-534.

30. Yoneda K: Ultrastructural localization of phospholipases in theClara cell of the rat. Bronchiole.Am J Pathol 1978;93:745-752.

31. Cakir Y, Plummer III H, Tithof PK, Schuller HM: Beta-adren-ergic and arachidonic acid-mediated growth regulation ofhuman breast cancer cell lines. Int J Oncol 2002;21:153-157.

32. Weddle DL, Tithoff P, Williams M, Schuller HM: Beta-adrener-gic growth regulation of human cancer cell lines derived frompancreatic ductal carcinomas. Carcinogenesis 2001;22:473-479.

201CLARA CELLS STIMULATED BY ISOPROTERENOL

33. Park PG, Merryman J, Orloff M, Schuller HM: Beta-adrenergicmitogenic signal transduction in peripheral lung adenocarci-noma: implications for individuals with preexisting chronic lungdisease. Cancer Res 1995;55:3504-3508.

34. Masubuchi Y, Hosokawa S, Horie T et al. Cytochrome p450isozymes involved in propranolol metabolism in human livermicrosomes. The role of CYP2D6 as ring-hydroxylase andCYP1A2 as N-desisopropylase. Drug Metab Dispos1994;22:909-915.

G. ARYAL, Y. KIMULA and M. KOIKE J Med Dent Sci202