THE ANATOMICAL RECORD 238:360-367 (1994)

Basal Cells In Human Bronchial Epithelium FRANK BALDWIN

Department of Pathology, University of Manitoba, Winnipeg, Manitoba, Canada R3W OW3

ABSTRACT The morphology and distribution of basal cells were inves- tigated in adult human bronchial epithelium.

In both smokers and non-smokers typical basal cells were more numer- ous than atypical basal cells, which were distinguished by their spindle- shaped nucleus, polar cytoplasm, and processes that extended along the basal lamina. The nucleus of the atypical basal cell was consistently closer to the muco-ciliary surface than was the nucleus of the typical basal cell.

In cross-sections of bronchi the numbers of typical basal cells/mm epi- thelium were greatest in large airways; in smaller bronchi (i.e., generations 7-16) the frequency declined. The numbers of atypical basal cells/mm epi- thelium were similar throughout the bronchial tree.

Throughout the bronchial tree both types of basal cells contributed to the maintenance of epithelial cohesion by providing desmosomal attachment for columnar cells. The importance of typical basal cells in this role was indicated by their greater numbers, which collectively presented a large surface for epithelial cell attachment by desmosomes. The surface pre- sented by the cell body and processes of each atypical basal cell for attach- ment of columnar cells by desmosomes was extensive.

If the basal cell is the most important progenitor of bronchial epithelium and the cell at risk in the development of lung cancer, the presence of more basal cells in upper airways, where many lung cancers originate, may be significant. The basal cell populations in larger bronchi are perhaps the greatest concentration of cells with proliferative capability and potential for neoplastic transformation in the human bronchial tree. 0 1994 Wiley-Liss, Inc.

Key words: Human, Lung, Bronchi, Epithelium, Light microscopy, Elec- tron microscopy, Morphometry

Much of what is known about the cells of bronchial epithelium stems from studies of young laboratory an- imals; much less is known about the detailed morphol- ogy of adult human airways. In common with the air- ways of other mammals, adult human bronchi are lined by a pseudostratified epithelium in which the prepon- derant elements are ciliated and non-ciliated columnar cells, goblet cells, and basal cells. Morphometric anal- ysis of adult human lung (Baldwin et al., 1991) has established the precise thickness of bronchial epithe- lium, the sizes of its different cells, and the distances of their nuclei from potentially harmful influences a t the muco-ciliary surface. Analysis of variance of these measurements showed that there were no significant differences when smokers and non-smokers with differ- ent pack-year histories were compared (Baldwin et al., 1991).

The basal cell, usually characterized as a small py- ramidal element attached to the basal lamina (Lentz, 1971; Jeffery and Reid, 1973; Rhodin 1974; McDowell and Beals, 1986) has been widely held to be the pro- genitor of bronchial epithelium (Ayers and Jeffery, 1982; Bindrieter et al., 1968; Boren and Paradise, 1978; Donnelly et al., 1982; Evans et al., 1986; Inayama et

0 1994 WILEY-LISS, INC.

al., 1988, 1989; Lane and Gordon, 1974). However, an- imal studies show that whilst basal cells possess greater proliferative activity, secretory cells also con- tribute to cell renewal (Bindrieter e t al., 1968; Boren and Paradise, 1978; Breuer e t al., 1990) and replicate in response to acute injury (Evans et al., 1986; Keenan et al., 1982a,b).

The role of the basal cell in airway epithelium might not be solely progenitorial; animal studies suggest that basal cells in the trachea are important in the mainte- nance of epithelial cohesion by providing attachment surfaces for columnar cells (Evans and Plopper, 1988; Evans et al., 1989; Evans and Moller, 1991). During a study of adult human bronchial epithelium it was ev- ident that there were two different morphological types of basal cell (Baldwin et al., 1991). The present study was undertaken to broaden understanding of the na- ture and occurrence of basal cell populations in the adult human bronchial tree.

Received August 25, 1992, accepted April 16, 1993

BASAL CELLS IN HUMAN BRONCHIAL EPITHELIUM 361

MATERIALS AND METHODS For light microscopy and morphometry, material

was obtained from 29 patients undergoing lobectomy or pneumonectomy for the following conditions-primary carcinoma of the lung, 24; metastatic carcinoma, 3; hamartoma, 1; and non-specific inflammatory change, 1. There were 26 smokers and 3 non-smokers in the group. Material from a further four patients with pri- mary carcinoma of the lung were obtained for electron microscopy and light microscopy. Three of these pa- tients were smokers. The time which elapsed from sur- gical removal of tissue to fixation was less than 2 hours.

TISSUE PREPARATION Light Microscopy

The bronchus of 29 specimens was cannulated and the lungs perfused with 10% phosphate buffered for- malin, pH 7, delivered a t a constant pressure of 2.4515 kPa (25 cm of water) for 36 hours (Weiss and Tweedale, 1966). Using a rotary meat slicer, 0.5 cm slices were cut to obtain the maximum number of airways in cross section. Blocks containing these airways were embed- ded in glycol methacrylate as described previously (Baldwin et al., 1991). Sections of 1.5 km thickness were stained with toluidine blue-basic fuschin.

Blocks containing bronchi from the lungs fixed by perfusion and from an additional four lungs fixed by immersion in 10% buffered formalin and in methacarn (Puchtler et al., 1970) were embedded in paraffin. Cross-sections, 5 pm thick, of these bronchi, which in- cluded bronchial glands, were immunolabeled using a muscle a-actin specific monoclonal antibody (MA 931)l diluted 1:800. This antibody recognises only skeletal, cardiac and smooth muscle cells and is non-reactive with other mesenchymal and all epithelial cells except for myoepithelium. Sites of antibody labeling were de- tected by alkaline phosphatase anti-alkaline phos- phatase2 or avidin-biotin peroxidase3 immunostaining. Sections for negative controls were from the same blocks as test sections; skeletal muscle sections, as well as smooth muscle around bronchi and arteries in the lung test sections, were used as positive controls.

Electron Microscopy One lung was perfused with 4% phosphate buffered

gluteraldehyde pH 6.8, using the same constant pres- sure technique described previously (Weiss and Tweedale, 1966). Samples from three additional lungs were fixed by immersion in buffered gluteraldehyde.

Epon-Araldite sections for light microscopy were stained with toluidine blue; ultrathin sections, stained with uranyl acetate and lead citrate, were examined in a Zeiss 10CR electron microscope.

Measurement of Bronchi The diameter of bronchi was computed from the cir-

cumferential measurement of the projected image of the lumen using a MOP-Videoplan Image Analysis

'Enzo Diagnostics Inc. 325 Hudson Street, New York, NY. 'Dakopatts ais, 42 Produktionsvej, DK-2600 Glostrup, Denmark. 3Biomeda Corp. P.O. Box 8045, Foster City, CA,

Fig. 1. Measurements of typical (A) and atypical basal cells (B) and adjacent epithelial cells: a) height of epithelial cell adjacent to basal cell; b) epithelial surface to basal cell nucleus; c) thickness of basal cell nucleus.

System and Standard D-Circle Measurement pro- gram.4

Measurement of Bronchial Epithelium The measurement system used was a Quantimet

Q-10 Image Analyzer,5 interfaced to a Hipad digitizing tablet.6 The image measured was produced from an Olympus BH27 binocular microscope using oil-immer- sion optics and an MTI-65 television camera. The im- age was displayed on the monitor, together with a red cross-hair cursor. Measurements were made by moving the cross-hair on the screen to start and end points of the distance to be calculated and pressing the cursor button once a t each point. The line length was then calculated and stored in a one-dimensional array for subsequent down-loading to a disk file for statistical analysis.

Measurements of typical epithelium were taken from cross-sections of methacrylate embedded bronchi. The measurements were, the height of the epithelial cell adjacent to the basal cell, the distance from the surface of the epithelium to the basal cell nucleus and the thickness of the basal cell nucleus (Fig. 1). The mea- surements were grouped according to bronchial diam- eter and bronchial generation (Weibel, 1963). One way analysis of variance and modified T tests were per- formed on an Amdahl4700 computer using a Biomed- ical Data Package Version 5.2.8

Cell Enumeration In 1.5 pm cross-sections of methacrylate embedded

bronchi, classified by their computed diameter and bronchial generation, typical and atypical basal cells were counted using light microscopy. Typical basal cells (Fig. 2) were identified by their triangular or po- lygonal shape, high nuclear-to-cytoplasmic ratio, and location of the cells on the basement membrane. Atyp- ical basal cells (Fig. 2) were recognised by their elon- gated shape, spindle-shaped nucleus, polar cytoplasm, and the location of the cells on the basement mem-

4Kontron Bildanalyse GMBH, Breslauer Strasse 2, Eching, Mu-

'Cambridge Instruments Ltd. Rustat Road, Cambridge, England. 'Houston Instruments, 8500iT Cameron Road, Austin, TX. 'Olympus Corporation Medical Scientific, 4 Nevada Drive, Lake

'BMDP Statistical Software Inc. 1440 Sepulveda Boulevard, Los

nich, Germany.

Success, New York, NY.

Angeles, CA.

362 F. BALDWIN

Fig. 2. Cross-section of bronchial epithelium showing the elongated shape of the atypical basal cell (A) and the triangular shape of the typical basal cell (B) which has a high nucleus to cytoplasmic ratio. Toluidine blue-basic fuchsin stain. Glycol methacrylate section. x 476.

brane. From these data, of the basal cell populations in each cross-section of bronchus, were calculated the number of basal cells of each type per millimeter length of epithelium (cells/mm epithelium) and the percentage of the total basal cell population contrib- uted by each cell type.

Statistical Methods Using the Quantimet-Q10 instrument and standard

statistical methods, means and standard errors were computed for epithelial measurements (Fig. 1) and for basal cell populations in cross-sections of bronchi clas- sified by their diameter and bronchial generations.

RESULTS Epithelial Measurements

Two types of basal cells (Fig. 2) were present in all generations of the bronchial tree (Table 1); there were no significant differences in epithelial distances when non-smokers and smokers with different pack-year his- tories were compared. Thus the results presented are composite data from smokers and non-smokers.

In all bronchial generations the distance from the epithelial surface to the nucleus of the basal cell was less for atypical than typical basal cells (Table 1). For both basal cell types the distances were least in small bronchi (i.e., generations 7-16), which analysis of vari- ance between individual bronchial generations showed were significant between generation 1 and generations 7-16 for typical (P = 0.0044) and atypical (P = 0.0061) basal cells. Comparisons of measurements for all gen- erations (Table 1) showed that there was a decrease in nucleus to epithelial surface distances for typical basal cells with successive generations (P = 0.0033). For atypical basal cells, analysis of variance showed signif- icant differences for basal cell nucleus-epithelial sur- face measurements between generations 1 and 3 (P = 0.0487), generation 1 and generations 5-7 (P =

0.0464) and generation 4 and generations 7-16 (P = 0.0449) as well as generation 1 and generations 7-16 (P = 0.0044). When all columns were compared (Table 1) there was not a successive decline in atypical basal cell nucleus-epithelial surface distances for bronchial generations of decreasing diameter (P = 0.0686). With successive bronchial generations there was a decrease in the height of epithelial cells adjacent to typical (P = 0.0052) and atypical basal cells (P = 0.03961, but the height of columnar cells was consistently less when they were adjacent to atypical basal cells.

The elongated nucleus of the atypical basal cell was much thinner than the pyramidal shaped nucleus of the typical basal cell (Table 1); the different dimen- sions of the nucleus for each basal cell type were main- tained throughout the bronchial tree.

Cell Enumeration In bronchi from smokers (Table 2) and non-smokers

(Table 3), typical basal cells were much more numerous than the atypical basal cell type. No significant trends were detected in the analysis of these figures, the per- centage of typical/atypical basal cells bore no clear re- lationship to smoking history or to bronchial genera- tion.

Calculations of basal cells/mm epithelium derived from 1.5 pm cross-sections of bronchi (Tables 2, 3) showed a greater number of typical basal cells in bron- chial generations 2-7 from both smokers (Fig. 3) and non-smokers (Fig. 4). There were fewer atypical basal cells/mm epithelium in generation one bronchi than in subsequent generations of large bronchi. In smaller bronchi (i.e., generations 7-16) the frequency of typical basal cells decreased, but the frequency of atypical basal cells/mm epithelium was fairly constant through- out the bronchial tree and similar for smokers and non- smokers.

Morphology The features of the typical basal cell (Figs. 2,5) were

consistent with established morphological characteris- tics (Lentz, 1971; Jeffery and Reid, 1973; Rhodin, 1974; McDowell and Beals, 1986). It had a high nuclear-to- cytoplasmic ratio, relatively few organelles and ran- domly distributed bundles of filaments, some of which terminated at junctions. Hemidesmisomes attached the broad surface of the cell to the basal lamina and des- mosomes were seen at interfaces with columnar cells.

The atypical basal cell (Figs. 2, 6) was readily dis- tinguished by its spindle-shaped nucleus and predom- inantly polar cytoplasm (Fig. 7). Processes coursed along the basal lamina beneath and between the cyto- plasm of columnar cells (Fig. 8). The cytoplasm (Fig. 7) was of moderate density and the organelles included small oval mitochondria, some rough endoplasmic re- ticulum, and variable quantities of ribosomes, which were frequently abundant. Golgi profiles and lyso- somes were present. Lipid droplets and lipofuscin were found in the polar cytoplasm. Some small cytoplasmic vesicles were present, and there was evidence of pi- nocytosis at the basal surface of the cell (Fig. 9). Bun- dles of filaments, approximately 10 nm in diameter, were aligned with the longitudinal axis of the cell (Fig. 7, 9) but were also distributed randomly. In some cells the bulk of the cytoplasm was packed with filaments. Hemidesmosomes attached the cell to the basal lamina and desmosomes were present between atypical basal cells and columnar cells (Fig. 6) and where the pro- cesses of atypical cells were in contact with one another (Fig. 10) or where the perikaryal cytoplasm of basal cells was apposed. Cytoplasmic interdigitations were also often present where basal cells contacted one an- other. Columnar cells were attached to one another by desmosomes (Fig. 8) but there were no hemidesmo- somes, nor other recognizable means of attachment be- tween columnar cells and the basal lamina.

BASAL CELLS IN HUMAN BRONCHIAL EPITHELIUM

TABLE 1. Epithelial measurements (pm)'

363

Bronchial Generation' 1 (n = 31) 2 (n = 9) 3 (n = 39) 4 (n = 70) 5-7 (n = 207) 7-16 (n = 88) P values"

0.26-2 4.51-5.50 2.01-4.50 Bronchial diameter (mm) > 9.01 7.01-9 5.51-7 Typical basal cell

Epithelial surface to basal cell nucleus 26.65 -t 2.363 25.83 * 3.96 24.08 i 2.22 22.02 i 1.29 24.59 f 1.76 17.46 * 0.93 0.0033

Thickness of basal cell nucleus 4.71 f 0.28 4.87 f 0.44 4.54 f 0.21 4.58 f 0.20 4.69 -t 0.20 4.26 f 0.18 0.5415 Height of adjacent epithelial cell 33.35 2 2.73 32.62 ? 4.59 30.28 2 2.45 27.91 -t 1.55 31.03 f 2.00 23.15 i 1.13 0.0052

Epithelial surface to basal cell

Thickness of basal cell nucleus 3.11 i 0.15 2.96 f 0.36 2.81 i 0.15 4.23 f 1.56 2.83 -+ 0.05 2.79 i 0.09 0.6681 Height of adjacent epithelial cell 27.72 f 2.61 25.12 f 5.19 20.90 f 1.43 21.96 i 1.22 21.90 f 2.09 19.59 F 1.27 0.0396

'Composite data: smokers and non-smokers. Comparisons between non-smokers and smokers with different pack-year histories showed no significant differences (Baldwin et al., 1991). 'Weibel (1963). 3 N ~ . represent means f standard error. *P values are for comparing all columns.

Atypical basal cell

nucleus 23.34 2 2.40 21.33 5.07 17.19 i 1.36 20.11 -c 1.84 18.15 f 1.08 15.98 +- 1.19 0.0686

TABLE 2. Basal cell populations in cross-sections of bronchial generations: smokers

Bronchial generation' l(n = 25) 2(n = 7) 3(n = 38) 4(n = 62) 5-7(n = 196) 7-16(n = 70)

Bronchial diameter (mm) > 9.01 7.01-9 5.51-7 4.51-5.50 2.01-4.50 0.26-2 Basal cells

Typical 495.2 i 97.5' 377.9 t 116.7 349.6 i 61.8 355.4 i 67.5 253.1 2 37.1 79.3 i 9.3 Atypical 79.0 2 18.2 87.8 2 38.8 73.6 2 11.2 61.4 t 9.7 53.8 k 7 23.9 t 4.3

% Typical 86.2 81.1 82.6 85.3 82.5 76.8 % Atypical 13.8 18.9 17.4 14.7 17.5 23.2

Basal cells

'Weibel (1963). 'Nos. represent means i standard error.

TABLE 3. Basal cell populations in cross-sections of bronchial generations: non-smokers

Bronchial generation' Bronchial diameter (mm) l (n = 6) 2(n = 2) 3(n = 1) 4(n = 8) 5-7(n = 11) 7-16(n = 18)

Basal cells Bronchial diameter (mm) > 9.01 7.01-9 5.51-7 4.51-5.50 2.01-4.50 0.26-2

Typical 4163 496 389 316.5 * 88.52 195.7 t 22.8 57.3 2 3.7 Atypical 161.7 23 68 50.9 t 38.2 46.8 i 11.7 8.5 t 6.4

% Typical 72 95.6 85.1 86.1 80.7 87.2 % Atypical 28 4.4 14.9 13.9 19.3 12.8

Basal cells

'Weibel (1963). 'Nos. represent means i standard error. 3Absence of standard error is due to insufficient No. of cases.

The morphological differences between the two basal cell types were maintained throughout the bronchial tree. Junctions on the more extensive surface of the atypical basal cell appeared to be more common than those on the typical basal cell. There was no obvious variation in the number ofjunctions borne by each type of basal cell in the different generations of the bron- chial tree.

In routinely stained light microscope preparations of bronchial epithelium the atypical basal cell bore some resemblance to the contractile myoepithelial cell. Elec-

tron microscopy showed that the typical myoepithelial cell of bronchial gland shared a number of features with the atypical basal cell, including lipid droplets, lipofuscin, and pinocytosis vesicles, and cytoplasmic processes running along the basal lamina. However, parallel 6 nm filaments and nodal aggregations (Fig. 11) and the terminations of filaments a t sub-plasmale- mmal densities were restricted to myoepithelial cells. Junctions between myoepithelial cells and adjacent ep- ithelial cells were not seen. The cytoplasm of myoepi- thelial cells from bronchial gland stained positively for

F. BALDWIN

populations of the two cell types are constant despite smoking history.

In human bronchial epithelium there are a variety of non-epithelial cells. Their known morphological and immunostaining characteristics identify them as mac- rophages, lymphocytes, neuro-endocrine cells, and an- tigen-processing dendritic cells; basal cells share none of the characteristics of these cell types. Whilst these non-epithelial elements are frequently adjacent to the basement membrane, none have the elongated form and spindle shaped nucleus which characterises the atypical basal cell. By reason of its light microscopic appearance and location, the atypical basal cell in bronchial epithelium is reminiscent of the myoepithe- lial cell (Tandler, 1965) but similarities are only super-

Basal Cells A typical + atypical

,-’-’

40 - 35 -

5 30- g 25- .- - m

0 ; 20-

d 10-

B 2 15- - m

5 -

35 -

5 30 - m g 25-

.- -

0 ’ , I I I I I

1 2 3 4 5 - 7 7 - 1 6 Bronchial Generations

Fig. 3. Basal cell distribution by bronchial generation. Derived from numbers of basal cells in 1.5 pm cross-sections of bronchi. Smokers.

Basal Cells rtypical + atypical

“1

0 5; 1 2 3 4 5 - 7 7 - 1 6

Bronchial Generations

Fig. 4. Basal cell distribution by bronchial generation. Derived from numbers of basal cells in 1.5 pm cross-sections of bronchi. Non-smok- ers.

muscle specific a-actin (Fig. 12) but basal cells in bron- chial epithelium did not contain a-actin (Fig. 13). Staining results were the same despite the use of dif- ferent fixatives and different immunostaining tech- niques.

DISCUSSION The present study shows there are two distinct pop-

ulations of basal cells which are present throughout the adult human bronchial tree. The typical basal cell is the most numerous and meets established morpho- logical criteria for this cell type. The atypical basal cell is distinguished by its elongated shape, spindle shaped nucleus and the polar distribution of cytoplasm, which extends into processes. Both basal cell types are firmly attached to the basement membrane by junctions. The distinct morphology of each basal cell type is main- tained throughout the bronchial tree. Smoking history has no effect on their morphology and in general the

- - ficial. The presence of muscle-specific a-actin and my- ofilaments with nodal aggregations which indicate the contractile nature of myoepithelial cells, for example in bronchial glands, are absent from atypical basal cells ruling out a contractile function for them.

Although by light microscopy the columnar cell in bronchial epithelium appears to sit directly on the basement membrane, in fact, only parts of its cyto- plasm contact this underlying substratum and there are no junctions to provide attachment. Columnar cells are fastened to basal cells by desmosomes. In human bronchial epithelium the more numerous typical basal cell is important in providing attachment for columnar cells, as described for trachea in a number of species (Evans and Plopper, 1988; Evans et al., 1989). How- ever, in human bronchial epithelium the flattened atypical basal cell, by reason of its extensive surface for desmosomal attachment of columnar cells, is also im- portant in the maintenance of epithelial cohesion. The firm attachment of basal cells to the basement mem- brane and to one another by desmosomes is most likely the reason that basal cells remain in situ when other epithelial cells are removed by mechanical forces, for example, during dissection of unfixed tissue. Studies of the trachea of different animal species (Evans and Plopper, 1988; Evans et al., 1989) suggest that the greatest numbers of basal cells, which provide the most extensive surface for epithelial attachment, occur where the epithelium is tallest. However, in human lung there is not a distinct correlation between the number of basal cells/mm epithelium (Figs. 3, 4) and the height of epithelium (Table 1).

Whilst the atypical basal cell plays a role in the maintenance of epithelial cohesion, its other functions await clarification. Its progenitorial role and its precise relationship to the typical basal cell are not yet deter- mined and the development of atypical basal cells as a consequence of airborne influences is untested. The spindle-shaped nucleus of the atypical basal cell is con- sistently closer to the epithelial surface of the airway than is the nucleus of the typical basal cell. This shorter distance might place the atypical basal cell at greater risk to harmful influences at the muco-ciliary surface but such effects would likely be only marginal because all basal cells in bronchial epithelium are well within range of such powerful carcinogenic agents as a emitters, whose depth of penetration is only short (Baldwin et al., 1991).

The basal cell has long been thought to be the pro- genitor of the human bronchial epithelium and it has

Fig. 5. Typical basal cell. Organelles include a few mitochondria and filaments, a centriole and polyribosomes are evident but endoplasmic reticulum and Golgi are poorly developed. A desmosome (D) is present. Bar = 1 pm.

shape of the cell is characteristically elongated and the cytoplasm is organelle rich. Bar = 1 pm.

Fig. 7. Polar cytoplasm of atypical basal cell showing small mito- chondria (M), endoplasmic reticulum (Er), polyribosomes (P), Golgi (G), lysosome (Ly), and lipid (L). Filaments (arrows) are mostly ori- ented on the longitudinal axis of the cell. Bar = 1 pm.

Fig. 6. Atypical basal cell attached to the basal lamina by hemides- mosomes (arrows) and to a columnar cell by a desmosome (D). The

366 F. BALDWIN

Fig. 8. Process of atypical basal cell (A). Columnar cell (C), desmo- Fig. 11. Myoepithelial cell from bronchial gland. Mitochondrion, rough endoplasmic reticulum, polyribosomes, and Golgi are perinu- clear. Filaments and nodal aggregations (F), pinocytosis vesicles (ar- rows). Bar = 1 pm.

some (D), basal lamina (arrow). Bar = 1 pm.

Fig. 9. Pinocytosis vesicles (arrows) on basal surface of atypical

Fig. 10. Desmosome at contact between processes of atypical basal

basal cell. Bar = 0.5 pm.

cells. Bar = 0.5 pm.

been widely held that proliferation of basal cells is of paramount importance in the development of lung can- cer (Auerbach, 1980). Animal evidence points to prolif- eration by basal cells as well as secretory and non- secretory cells but the proliferative potential of cell

populations in the human bronchial tree has yet to be elucidated. If basal cells are indeed the most important proliferative element and the cell type which is also prominent in the development of human lung neopla- sia, the greater concentration of basal cells in the epi-

BASAL CELLS IN HUMAN BRONCHIAL EPITHELIUM 367

Fig. 12. Myoepithelial cells in bronchial gland, showing positive immunostaining for muscle specific a-actin (arrows). Alkaline phosphatase anti-alkaline phosphatase technique. x 476.

Fig. 13. Bronchial epithelium showing absence of immunostaining for muscle specific a-actin in atyp- ical basal cells (arrows). Avidin-biotin peroxidase technique. X 476.

thelium of larger airways may have some bearing on the development of cancer in the upper airways where a t least half, and perhaps the majority, of all lung can- cers arise (Spencer, 1985; Dunnill, 1987). The more nu- merous basal cells in the epithelium of larger airways might be the greatest concentration of targets for the induction of neoplastic transformation. The basal cells in larger bronchi might also be at greater risk because the presence of potentially harmful airborne influ- ences is prolonged because of the resistance to mass airflow which is inherent in the upper airways of the human lung (Pedley et al., 1970; Weibel, 1984).

ACKNOWLEDGMENTS This research was supported by the Atomic Energy

Control Board of Canada.

LITERATURE CITED Auerbach, 0. 1980 Natural history of carcinoma of the lung. In: Pul-

monary Diseases and Disorders. A.P. Fishman, ed. McGraw-Hill, New York, p. 1388.

Ayers, M.M., and P.K. Jeffery 1982 Cell division and differentiation in bronchial epithelium. In: Cellular Biolom of the Lung. G. Bon- signore and 1.- Cumming, eds. Plenum, New York, pp. j3-59.

Baldwin, F., A. Hovey, T. McEwen, R. OConnor, H. Unruh, and D.H. Bowden 1991 Surface to nuclear distances in human bronchial epithelium: relationship to penetration by Rn daughters. Health

~~

Phys., 60t155-162. Bindreiter, M., J. Shuppler, and L. Stockinger 1968 Zellproliferation

und Differenzierung im Trachealepithel der Ratte. Exp. Cell Res., 50r377.

Boren, H.G., and L.J. Paradise 1978 Cytokinetics of the lung. In: Pathogenesis and Therapy of Lung Cancer. C.C. Harris, ed. Dekker, New York, pp. 369-418.

Breuer, R., G. Zajicek, T.G. Christensen, E.C. Lucey, and G.L. Snider 1990 Cell kinetics of normal adult hamster bronchial epithelium in the steady state. Am. J . Respir. Cell Mol. Biol., 251-58.

Donnelly, G.M., D.G. Haack, and C.S. Heird 1982 Trachea epithelium: cell kinetics and differentiation in normal rat tissue. Cell Tissue Kinet., I5t119-130.

Dunnill, M.S. 1987 Carcinoma of the bronchus and lung. In: Pulmo- nary Pathology. Churchill Livingstone, Edinburgh, pp. 333-401.

Evans, M.J., S.G. Shami, L.J. Cabral-Anderson, and N.P. Dekker 1986 Role of non-ciliated cells in renewal of the bronchial epithe- lium of rats exposed to NO,. Am. J. Pathol., 123t126-133.

Evans, M.J., and P.C. Moller 1991 Biology of airway basal cells. Exp. Lung Res., 17.513-531.

Evans, M.J., and C.G. Plopper 1988 The role of basal cells in adhesion of columnar epithelium to airway basement membrane. Am. Rev. Respir. Dis., 138:481-483.

Evans, M.J., R.A. Cox, S.G. Shami, B. Wilson, and C.G. Plopper 1989 The role of basal cells in attachment of columnar cells to the basal lamina of the trachea. Am. J . Respir. Cell Mol. Biol., 1t463-469.

Inayama, Y., G.E. Hook, A.R. Brody, G.S. Cameron, A.M. Jetten, L.B. Gilmore, T. Gray, and P. Nettesheim 1988 The differentiation potential of tracheal basal cells. Lab. Invest., 58t706-717.

Inayama, Y., G.E. Hook, A.R. Brody, A.M. Jetten, T. Gray, J . Mahler, and P. Nettesheim 1989 In vitro and in vivo growth and differ- entiation of clones of basal cells. Am. J . Pathol., 134.539-548.

Jeffery, P.K., and L.M. Reid 1973 The ultrastructure of airway epi- thelium and submucosal gland during development. In: Develop- ment of the Lung. W.A. Hodson, ed. Dekker, New York, pp. 87- 134.

Keenan, K.P., J.W. Combs, and E.M. McDowell1982a Regeneration of hamster tracheal epithelium after mechanical injury. I. Focal le- sions: quantitative morphological study of cell proliferation. Vir- chows Arch. B Cell Pathol., 41:193-214.

Keenan, K.P., J.W. Combs, and E.M. McDowell1982b Regeneration of hamster tracheal epithelium after mechanical injury. 11. Multi- focal lesions: stathmokinetic and autoradiographic studies of cell proliferation. Virchows Arch B Cell Pathol., 41t215-229.

Lane, B.P., and R. Gordon 1974 Regeneration of rat tracheal epithe- lium after mechanical injury. I. The relationship between mitotic activity and cellular differentiation. Proc. SOC. Exp. Biol. Med., 145:1139-1144.

Lentz, T.L. 1971 Cell Fine Structure. Saunders, Philadelphia, pp. 100-101.

McDowell, E.M., and T.F. Beals 1986 The bronchial airways. In: Bi- opsy Pathology of the Bronchi. Chapman and Hall, London, pp. 49-76.

Pedley, T.J., R.C. Schroter, and M.F. Sudlow 1970 The prediction of pressure drop and variation of resistance within the human bron- chial airways. Respir. Physiol., 9t387-405.

Puchtler, H., Waldrop, F.S., Meloan, S.N., Terry, M.S. and H.M. Con- ner 1970 Methacarn (methanol-carnoy) fixation. Histochemie, 21:97-116.

Rhodin, J.A.G. 1974 Respiratory system. In: Histology. Oxford, New York, pp. 608-645.

Spencer, H. 1985 Carcinoma of the lung. In: Pathology of the Lung. Pergamon, Oxford, pp. 837-927.

Tandler, B. 1965 Ultrastructure of the human submaxillary gland. 111. Myoepithelium. Z. Zellforsch., 68t852-863.

Weibel, E.R. 1963 Morphometry of the Human Lung. Academic, New York.

Weibel, E.W. 1984 Airways and blood vessels. In: The Pathway for Oxygen. Harvard, Cambridge, pp. 272-301.

Weiss, D.L., and D.N. Tweedale 1966 Inflation fixation of lungs: use of a simple inexpensive apparatus. Am. Rev. Respir. Dis., 94: 629-631.