The Role of the Nonciliated Bronchiolar Epithelial (Clara) Cell as the Progenitor Cell during Bronchiolar Epithelial Differentiation in the Perinatal Rabbit Lung c. G. Plopper, s. J. Nishio, J. L. Alley, P. Kass, and D. M. Hyde Department of Veterinary Anatomy and Cell Biology, School or Veterinary Medicine, University of California, Davis, California
Although it is well established that the nonciliated bronchiolar epithelial (Clara) cell serves as the progenitor for itself and ciliated cells in the adult lung following bronchiolar epithelial injury, the nature of this relationship during development has not been well characterized. To define the pattern of proliferation and differentiation of bronchiolar ciliated and nonciliated cells, lungs of fetuses and offspring from timemated New Zealand White rabbits, ranging in age from 24 days of gestation to 25 wk postnatal (PN), were fixed by airway infusion and embedded for simultaneous light and transmission electron microscopy. Three categories of cells could be distinguished in terminal bronchioles: nonciliated cells with abundant glycogen and variable numbers of organelles; nonciliated cells with little glycogen, large numbers of polyribosomes, and variable numbers of basal bodies; and ciliated cells with cilia of varying height. Together, both types of nonciliated cells were 100% of the epithelium at 24 and 27 days gestation age (DGA). At 30 days DGA, they were 85% of the population; at all postnatal ages, they ranged from 75 to 81% of the total population. Nonciliated cells with polyribosomes and basal bodies were 10 to 20% of the total nonciliated cell population between 24 DGA and 1 wk PN and not found thereafter. Ciliated cells were not observed in animals younger than 30 DGA. Labeling indices of bronchiolar epithelium in fetuses of pregnant rabbits injected with tritiated thymidine, as determined by autoradiography, were 57 cells per thousand at28 DGA (1 h postinjection [PI]), 76 at 29 DGA (24 h PI), and 114 at 30 DGA (48 h PI). Almost all (98 %) of the labeled bronchiolar cells were nonciliated cells with glycogen and no basal bodies at 1 h PI. Nonciliated cells with basal bodies were 2 to 3 % of the labeled cells at all time points. Ciliated cells were unlabeled until 24 h PI (2 %) and increased to 7 % at 48 h PI. We conclude that: (1) bronchiolar ciliated cell differentiation occurs over a short period of time immediately before and after parturition; (2) differentiation of ciliated cells from nonciliated cells involves a transitional cell in which glycogen is lost and polyribosomes are synthesized before the synthesis of basal bodies and cilia; (3) nonciliated bronchiolar cells serve as the progenitors both for themselves and for ciliated cells during the normal process of bronchiolar epithelial differentiation in the fetus.
Among the functions proposed for the nonciliated epithelial cell in distal bronchioles of adult mammals is the role of progenitor for itself and ciliated cells. Evidence for this derives primarily from experimental pathologic studies. As is true for other epithelial populations lining the respiratory airspaces, the mitotic activity of bronchiolar epithelium in steady-state conditions is extremely low (see references I (Received in original form llugust 12, 1991 and in revised form July 13, 1992) Address correspondence to: Charles G. Plopper, Ph.D., Professor ofVeterinary Anatomy and Cell Biology, School of Veterinary Medicine, University of California, Davis, CA 95616-8732. Abbreviations: agranular endoplasmic reticulum, AER; days gestational age, DGA; days postnatal age, DPN; postinjection, PI; postnatal, PN. Am. J. Respir. Cell Mol. BioI. Vol. 7. pp. 606-613, 1992
through 3 for review). For species with poorly developed respiratory bronchioles, such as the rat, hamster, and mouse, the most definitive study of bronchiolar epithelial kinetics was performed in rats exposed acutely to nitrogen dioxide (4). All of the cells in injured bronchiolar epithelium that took up [3H]thymidine I h after injection were nonciliated cells in one of three categories. With time, there was clear transition to other types of nonciliated cells and eventually to ciliated cells. The frequency of labeled nonciliated cells also increased steadily to 50 % by day 6 postinjection. Other studies of rodents exposed to NO z, 0 3 , or high concentrations of O, have confirmed that the majority of the labeled bronchiolar cells after injury are nonciliated and that ciliated as well as Clara cells represent a larger portion of the labeled cells at later time periods (5-8). In species with extensive respiratory bronchioles, such as macaque monkeys and cats,
Plopper, Nishio, Alley et al.: Role of Clara Cells in Bronchiolar Epithelial Differentiation in·the Lung
experimental pathologic studies indicate that the nonciliated bronchiolar (or Clara) cell is also the bronchiolar epithelial progenitor for these species (9, 10). Whether the relationship between ciliated and nonciliated cells exists in steady-state conditions and especially in differentiating bronchiolar epithelial populations during lung development has not been established. Previous studies of differentiating bronchiolar epithelium in rat (11), rabbit (12, 13), and rhesus monkey (14, 15) suggest a progenitor role for the nonciliated cell population. In all three species, ciliated cells appear in a previously nonciliated bronchiolar epithelial population during perinatal lung development. The proportion of ciliated cells within the bronchiolar population appears to increase rapidly and at the expense of the nonciliated cell population. However, as emphasized in a recent review of pulmonary cell kinetics (16), the pattern and time course of events resulting in the shift from a completely nonciliated population to a mixture of ciliated and nonciliated cells has not been defined, nor has the kinetic relationship between nonciliated and ciliated cells been characterized during development. The purpose of the present study was twofold: (1) to identify the progenitor cell population for bronchiolar epithelium during maturation in the fetal lung and (2) to characterize the pattern of differentiation from a nonciliated cell population into a mixed population of nonciliated and ciliated cells.
Materials and Methods Male New Zealand White rabbits, free of chronic respiratory disease, as judged by gross examination and histopathology, were used in this study. Table 1 summarizes the ages and numbers of animals used to calculate population densities and proportions of bronchiolar epithelium in rabbit lungs. Postnatal animals were obtained through the National Institute of Environmental Health Sciences from a colony free of respiratory disease. They were anesthetized with sodium pentobarbital (1 mg/kg body weight). The trachea was cannulated, and the animals were killed by exsanguination. The
TABLE 1
Age, number of animals, number of terminal bronchioles, and number of cells counted for estimated cell density and proportions of ciliated and nonciliated bronchiolar epithelium Animals
Bronchioles
Cells
Age
(n)
(n)
(n)
24 DGA 27 DGA 30 DGA 1-2 DPN 3-4 DPN I wk 2 wk 3 wk 4 wk 5 wk 15-25 wk
5 3 4 4 3 3 4 3 3 3
45 41 60 65 79 34 48 21 34 41 53
634 440 680 742 880 440 600 360 249 350 533
7
Definition ofabbreviations: DGA natal age.
607
thorax was opened by an incision in the diaphragm, and the lung was infused via the trachea by glutaraldehyde/paraformaldehyde in cacodylate buffer (17, 18) adjusted to pH 7.4 and 550 mOsm (at 30 em water pressure). From 2 to 3 h later, the fixed lungs and trachea were removed from the chest and stored in the same fixative until processing. Prenatal animals were obtained by laparotomy from timemated pregnant rabbits anesthetized with acepromazine (1 mg/kg body weight), Rompun (5 mg/kg body weight), and ketamine (50 mg/kg body weight) (intramuscularly). Table 1 summarizes the fetal ages (days gestational age [DGA] or days postnatal age [DPN]) and number of animals used for estimating bronchiolar epithelium densities and proportions. Table 2 summarizes the fetal age and number of animals used for autoradiographic studies of ['Hlthymidine uptake. After removal of the fetus from the uterus, the mouth and nostrils of the sedated fetus were occluded to prevent breathing, the trachea was cannulated, and the thorax was opened by an incision in the diaphragm. For the 27 to 30 DGA fetuses, the trachea and lungs were fixed by infusion at 40 em water pressure with the same fixative used for postnatal animals. The tracheas of the 24 DGA fetuses were ligated, and the lungs and trachea were removed and fixed by immersion in the same fixative used for other age groups. Gonads were removed from all animals that were 1 wk of postnatal age and younger. Sex was determined from histologic sections of the gonadal tissue. Only males were included in this study. For autoradiographic study of ['Hlthymidine uptake, time-mated New Zealand White rabbits were received from the breeding colony on day 21 of pregnancy. The rabbits were housed in a barrier-maintained facility with nonrecycled air and given food and water ad libitum. On day 28 of pregnancy, maternal rabbits were injected intravenously at 9:00 a.m. with [3H]thymidine (1 mCi/kg body weight; specific activity, 80.1 Ci/mmol; New England Nuclear, Boston, MA). Fetuses were removed by laparotomy from anesthetized mothers at 1, 6, 24, and 48 h postinjection (PI). Samples of fixed lungs for light and electron microscopy were selected from the right caudal lobe. The lobe was sliced into 2- to 4-mm-thick slices. A dual-viewing Wild M-8 dissecting microscope was used to select slices that contained terminal bronchioles. Selected pieces (at least five per lung) were postfixed for 2 h in 1% osmium tetroxide. The blocks were embedded in Araldite 502 by a process that allows selection of specific areas from large tissue faces (17). The l-um sections were produced with glass knives using a
TABLE 2
Age, number of animals, number of terminal bronchioles, and number of cells counted for autoradiography Fetal Age
28 28 29 30
DGA DGA DGA DGA
Animals (n)
2 4 5 4
Time Postinjection (h)
Bronchioles (n)
Cells (n)
1 6 24 48
59 164 147 132
1,505 4,039 5,318 4,611
= days gestational age; DPN = days postDefinition of abbreviation: DGA = days gestational age.
608
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 7 1992
Sorvall JB-4 microtome. Sections from animals receiving pH]thymidine were coated with Ilford L-4 emulsion (diluted 1:1) and incubated for 9 wk. Slides were developed with Kodak D-19. All sections were stained with toluidine blue. For both light and electron microscopy, only areas in which terminal bronchiole-alveolar duct junctions were present were evaluated. The number of terminal bronchioles evaluated for estimations of population densities and proportions is summarized in Table 1. Areas containing terminal bronchiole-alveolar duct junctions were removed from the large blocks, remounted, and the junctional areas confirmed using 0.5-J.tm sections made with a Sorvall MT-5000. For proportional counting, a minimum of seven terminal bronchioles were sampled per animal. For autoradiographic evaluation of tritiated thymidine uptake, a minimum of 25 bronchioles were evaluated per animal. Thin sections (30 to 50 nm) were produced with a diamond knife on a Sorvall MT-5000 ultramicrotome, stained with uranyl acetate and lead citrate, and examined with a Zeiss EM-1O at 60 kYo Nuclear size of ciliated andnonciliated bronchiolar epithelial cells was estimated from serial sections of bronchiolar epithelium for three age groups, 30 DGA, 4 wk PN, and 17 wk PN, as described in detail previously (12). Evaluations
r-
of the number of cells per unit length of basement membrane, the number of cells per unit area of basement membrane, and the number of cells per unit volume of bronchiolar epithelium were not significantly different between 30 DGA and 4 wk PN. However, an error of 15.4% was observed at 17 wk. Because there were no statistical differences in the estimation, the number of nuclei per unit length of basal lamina was used to estimate the numerical density of epithelial cells in terminal bronchioles and the percentage of ciliated and nonciliated cells in the bronchioles by light microscopy on l-am sections. Values for adults (ranging in age from 15 to 25 wk) were corrected for overestimation and underestimation. The percentage of the nonciliated cell population that was defined as preciliated cells was estimated by differential counting of epithelial cells in all of the terminal bronchioles evaluated for the ages 24 DGA through 2 wk PN by transmission electron microscopy of adjacent sections. For calculating labeling indices (number of labeled cells per 1,000 cells) based on tritiated thymidine uptake, a minimum of 25 terminal bronchioles were counted per animal. A minimum of750 cells were also counted per animal. Table 2 summarizes the number of cells counted per time period PI. The slides of bronchioles containing labeled cells were photographed in their entirety at an original magnification of 400x and then mapped with contiguous 4 X 5 inch prints. Each labeled cell was identified on the map and then characterized on serial sections by electron microscopy from the same bronchiole (Figure 1). The labeling index is reported as the number of labeled nuclei per 1,000 cells counted by light microscopy. The number of cells undergoing mitosis was also counted, and the mitotic index is reported as number of mitotic nuclei per 1,000 cells (n was considered to be the number of animals). The proportion of the labeled population represented by each cell type is reported as the percentage of all labeled cells.
Results
Figure 1. Serial l-um and ultrathin sections of labeled and unlabeled bronchiolar epithelial cells. The animal was injected with tritiated thymidine, and uptake of labeled measure identified in cells by autoradiography on l-um sections. The labeled cells identified histologically (in brackets) (A) (bar = 100 /tm) were then characterized using transmission electron micrographs of serial sections of the same bronchioles (B) (bar = 5 /tm).
Bronchiolar cells were in approximately equal abundance (12 cells/IOO J.tm) at all ages except at the earliest fetal age examined (24 DGA) (Figure 2A). The nonciliated population as a whole was 100 % of the cell population at 24 and 27 DGA, decreased to approximately 90% at 30 DGA, and decreased to approximately 75 to 80% in animals 2 wk and older (Figure 2B). In fetuses 24 to 27 DGA, two categories of nonciliated cells could be identified. By light microscopy, the most abundant category was cuboidal in appearance with a rounded nucleus and a dome-shaped apical projection that was filled with a heavily stained material. Electron microscopic evaluation of these cells during the early prenatal and postnatal time period showed that their cytoplasm was filled with glycogen (Figure 3A). These cells had variable amounts of agranular endoplasmic reticulum (AER) in the borders of the cytoplasm on the lateral and apical surfaces. The amount of AER increased with age. The central portions of some of the large areas of glycogen also contained AER. Mitochondria and granular endoplasmic reticulum, along with Golgi apparatus, were observed in small amounts, both apically and basally to the nucleus (Figure 3A). Nonciliated cells varied in abundance from 90% (24 DGA) to 66% (30 DGA) of the bronchiolar population in the perinatal period. They
Plopper, Nishio, Alley et al.: Role of Clara Cells in Bronchiolar Epithelial Differentiation in the Lung
A "-0e
609
20
15
>-
C:::i.
ClIo
"0 0
-~
cau~ '': Gi
10
CIIu
E""
~
.....
5
O...L..------------------Figure 2. Changes in abundance of ciliated and nonciliated cells in terminal bronchioles of prenatal and postnatal rabbits. Birth is indicated by the arrows. (A) Numerical density of epithelium in terminal bronchioles in both prenatal and postnatal animals as number per 100 !-tm of basal lamina. (B) The relative proportion (percentage) of cells within the bronchiolar epithelium. Solid squares = nonciliated cells; open triangles = ciliated cells; solid circles = nonciliated cells with polyribosomes.
....
« -c -c
on on
o0
ClCl 00
~
~ g+~..,.
on
""
>- >-
~
00
""
~
N
~..,
on
on
""
"":B ..,~
~ •
~
:B
~
_ ..,
:;
«
It)
Age
B 100
~
Gi
80
u
-
c;
.2....... 60
OC
Nonciliated cells Preciliated cells
CII
-B g' c::
cf
I
Ciliated cells
40
,/f",1---i-'-c---------------------r 20
o...L..---.(r-oi~--....L..-.,.-
« «
c(
U)
U)
~
''f ~
e e e ;;-;;- :B 0
0
~
were 66% (adult) to 78% (30 DGAand 14 DPN) of the population in older animals. A small proportion of the nonciliated cell population at all ages between 24 DGA and 1 wk had a flatter apical surface and was free of densely staining cytoplasmic content but did not have cilia. These cells had the ultrastructural features of preciliated cells undergoing ciliogenesis (Figures 3B and 3C). The cytoplasm was nearly free of glycogen. In some cases (Figure 3B), large numbers of polyribosomes were found distributed throughout the cytoplasm. In the apical cytoplasm, there were areas of condensing microtubule-like structures. In other cells (Figure 3C), large numbers of polyribosomes were present but, in addition, condensing basal bodies were irregularly arranged
.... N
~..,
...- .....- .....---------__.
. "".. "".. .. "" i :B
~ N
:B
..,~
~ •
~
..,
:>
-c
It)
Age
along the apical portion of the cell. Cells containing these ultrastructural features composed approximately 10% of the cell population in 24 DGA animals, were as high as 18% of the population immediately before birth, and decreased to 0% of the population at 2 wk PN (Figure 2B). Ciliated cells were clearly identified ultrastructurally by their extensive cilia, abundance of mitochondria in the apical cytoplasm, the presence of a Golgi apparatus near the nucleus and strands of GER in a cytoplasm nearly free of glycogen (Figure 3D). Ciliated cells were not observed in 24 and 27 DGA fetuses. They composed 15% or more of the population in animals 30 DGA and older (Figure 2B). For autoradiographic evaluation of bronchiolar kinetics,
Figure 3. Ultrastructural comparison of the features of nonciliated and ciliated epithelial cells in the bronchiolar epithelium of perinatal rabbits (27 DGA). L = airway lumen; Nu = nucleus. (A) The majority ofthe nonciliated cells contained abundant glycogen (Gly), mitochondria (M), and small amounts of agranular endoplasmic reticulum (arrowheads) and rough endoplasmic reticulum (solid arrows). (Bar = 1 I-'m). (B) Many of the nonciliated cells contained an abundance of polyribosomes (open arrows), areas of condensing microtubules, and small amounts of smooth endoplasmic reticulum (arrowheads) and rough endoplasmic reticulum (solid arrows). Asterisks indicate condensing basal bodies. (Bar = 1 I-'m.) (C) Many of the nonciliated cells resembled those in panel B but also included basal bodies or forming centrioles near the apical surface (asterisks). Go = Golgi apparatus. (Bar = 1 I-'m.) (D) During this time, a small percentage, of the cells resembled ciliated cells with mature cilia attached to basal bodies (asterisks). The remainder of the cell cytoplasm was primarily
free of glycogen and contained a variety of organelles including mitochondria (M) and small amounts of rough endoplasmic reticulum (solid arrows). (Bar = 1 I-'m.)
Plopper, Nishio, Alley et al.: Role of Clara Cells in Bronchiolar Epithelial Differentiation in the Lung
A
611
140
120
100 )(-
C1lCll '0= CC1l
80
0)0 Co =0 C1l ....
60
.-
(,,)
.0--
CIS~
...J
40
20
Figure 4. Kinetics of bronchiolar epithelium immediately before birth (28 DGA to 30 DGA). (A) Labeling index of bronchiolar epithelium in terminal bronchioles of fetal rabbits (as number of labeled nuclei per 1,000 cells). Times shown are following injection of tritiated thymidine. (B) Percentage of labeled cells based on ultrastructural features. Solid squares = nonciliated cells resembling Figure 3A; solid circles = nonciliated cells resembling Figures 3B and 3C; open triangles = ciliated cells resembling Figure 3D. Times shown are following injection of tritiated thymidine.
48 (30dga)
!!!. Qj o
90
'0 C1l
Qj
--------
.0_
--
CIS~
~
-~ 0
..
--A--
Labeled Nonciliated cells Labeled Preciliated cells Labeled Ciliated cells
C C1l
o
C1l
C.
10
------
0
6 (Fetal age)
----------------
(2Bdga)
(2Bdga)
24 (29dga)
--f 48 (3Odga)
Time after injection (h)
we chose the time period of most rapid transition and differentiation of ciliated cells based on cell density and proportion profiles: 28 to 30 DGA. One hour after administration of tritiated thymidine to the mother, almost 6 % of the bronchiolar epithelial cells in fetal rabbits were labeled (Figure 4A). Almost all, approximately 98%, of the labeled cells
were glycogen-filled nonciliated cells (Figure 4B). The remainder (approximately 2 %) of the labeled cells were nonciliated cells identified above as glycogen-free preciliated cells (Figure 4B). These labeled cells contained many polyribosomes, few condensing microtubular areas, and no basal bodies. The labeling of the total bronchiolar epithelial popu-
612
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 71992
lation doubled by 48 h PI (Figure 4A). Nonciliated glycogenfilled cells still composed almost 93 % of the labeled cells at 48 h PI (Figure 4B). Preciliated cells made up < 2 % of labeled cells at 48 h PI (Figure 4B). Labeled ciliated cells were not observed at 1 and 6 h PI (Figure 4B). At 24 h PI, < 3% of the labeled cells were ciliated cells. This number was more than doubled by 48 h PI. The mitotic index was 3.32 cells/l,OOO cells at 1 h PI (28 DGA), 1.73 cells/l,OOO at 6 h PI (28 DGA), 2.63/1,000 at 29 DGA, and 1.73cells/1,000 at 30 DGA. All the cells with mitotic figures were nonciliated cells.
Discussion The purpose of this study was to characterize the kinetic relationship between nonciliated and ciliated cells in the bronchiolar epithelium during the period of active differentiation of the Clara cell. The relative densities of these two cell types differ between species like the rabbit (12) and the rat (2), and our study indicates that differentiation also varies. In the rabbit, the majority of ciliated cell differentiation occurs prenatally and the steady-state condition found in the adult is almost established by birth. For the rat, in contrast, the steady-state conditions are not present immediately after birth and there is a 3- to 4-fold increase in the density of ciliated cells postnatally that continues for at least 60 DPN (11). In the rabbit, the intermediate form, the preciliated cell, is present for a relatively limited period prenatally and post. natally. Whether the intermediate (preciliated) form is found in rat bronchioles during development has not been reported. Ciliogenesis in the bronchiole of the rabbit does not follow the same pattern of events as in rat bronchi (19). In the rabbit, differentiating ciliated cells do not have fibrogranular accumulations or deuterosomes (19). Whether all the forms we observed in the rabbit are present in the rat bronchiolar epithelium either prenatally or postnatally has not been established by previous studies (11). Differentiation of Clara cells in the rabbit requires 3 to 4 wk, whereas ciliated cell differentiation is complete by 1 wk PN, well before nonciliated cells contain ultrastructural features characteristic of mature Clara cells. In contrast, in the rat, ciliogenesis is active over a significant period of postnatal age long after the differentiation of the Clara cell has occurred (11). This pattern of differentiation in bronchiolar epithelium in the rabbit is similar to that observed in respiratory bronchioles of the rhesus monkey, with the exception that it begins much earlier in the monkey and the bronchiolar epithelial population actually differentiates into four different cell types instead of just two (14, 15). Clara cells also appear to be the progenitor for ciliated cells in the trachea of the hamster during development (20, 21), with Clara cell differentiation occurring much later than that of ciliated cells. When compared with the kinetic activity of bronchiolar epithelium in the steady.state in the adult, the mitotic activity of bronchiolar epithelium in the perinatal period is extremely high. To the best of our knowledge, these are the highest labeling and mitotic indices reported for bronchiolar epithelium at any age (see reference 16 for summary ofliterature). The labeling index is more than twice as high as that observed in l-rno old rats (5.7 versus 1.1 %) or postnatal mice
(0.52%) (22). High labeling indices have also been reported in the parenchyma in fetal mice (1). Labeling indices for bronchioles of adults (0.2% or less) indicate that kinetic activity is well below the prenatal animal (16). Labeling indices in the range observed by us in the late fetal period have been observed in the adult only with acute bronchiolar injury, such as that resulting from exposure to ozone (7, 9) or nitrogen dioxide (5). Injury from diesel exhaust (10), or benzo(a)pyrene (23), elevates cell turnover minimally. Our study shows a clear progenitor relationship between nonciliated bronchiolar epithelial cells and ciliated bronchiolar cells during the period oflung maturation when these two cell types are undergoing cytodifferentiation. Our 2-day pulse label study, applied during the most active phase of transition from a strictly nonciliated to a mixed ciliated and nonciliated cell population, indicates that some portion (about 5 %) of the nonciliated cell population is clearly the progenitor for both the ciliated and the rest of the nonciliated population. Mitotic figures were observed only in nonciliated cells. The nonciliated cells containing large amounts of glycogen and few organelles appear to be the origin of differentiated ciliated cells. They have the highest labeling index at any time during the 2-day period and represent almost all (98 %) the labeled cells. The population of preciliated cells represents a small proportion of the labeled cells and shows most of the kinetic characteristics of a cell population in transition. The labeling index for this cell population indicates that it is in steady state throughout the 2-day period. This is in contrast to nonciliated cells, where the labeling index almost doubles during the 2-day period, and to ciliated cells, where the labeling index doubles at 24 h and again at 48 h PI. The shifts in labeling we observed in the fetus match the pattern reported by Evans and co-workers (5) for bronchiolar epithelial repair following acute injury in adults. Our observations are compatible with two patterns of differentiation: (1) the labeled cells represent a small population of undifferentiated cells whose daughters differentiate into both the slowly differentiating Clara cell population and the rapidly differentiating ciliated cell population or (2) a subpopulation of the undifferentiated Clara cell population that will differentiate later into Clara cells after the stimulus for active mitosis is over, and some of whose daughter cells will differentiate into ciliated cells. We believe that the latter is more likely because of our inability to differentiate between the labeled and unlabeled nonciliated cells on the basis of ultrastructural features and because the labeled nonciliated cells have none of the features associated with purported stem cells in other tissues (24-27). The former pattern should not be completely ruled out, however, since these cells may exist in the adult as a very small population of undifferentiated cells stimulated to divide only under extreme conditions. If this is the case, they are in such small numbers that they have not been detected in healthy adults by morphometric studies (see reference 28 for review). One of the characteristics of stem cells is their lack of some differentiated functions found in their differentiated offspring (24-27). We have observed a small number of nonciliated bronchiolar cells in adult rabbits that lacked antigens for cytochrome P-450 monooxygenases (29) and may represent such a population. The failure to detect such a cell population in kinetic
Plopper, Nishio, Alley et al.: Role of Clara Cells in Bronchiolar Epithelial Differentiation in the Lung
studies in adults following injury may be accounted for by the fact that none of the studies involved an injurant that targeted the nonciliated population (4-10, 23). Acknowledgments: This study was supported by Grant HL43032 from the National Institutes of Health.
References I. Kauffman, S. L. 1980. Cell proliferation in the mammalian lung. Int. Rev. Exp. Pathol. 22: 131-191. 2. Massaro, G. C. 1989. Nonciliated bronchiolar epithelial (Clara) cells. In Lung Cell Biology, D. Massaro, editor. Marcel Dekker, New York. 81-114. 3. Plopper, C. G., and D. L. Dungworth. 1987. Structure, function, cell injury and cell renewal of bronchiolar and alveolar epithelium. In Lung Carcinomas. E. M. McDowell, editor. Churchill Livingstone, London. 94-128. 4. Evans, M. J., L. J. Cabral-Anderson, and G. Freeman. 1978. Role of the Clara cell in renewal of the bronchiolar epithelium. Lab. Invest. 38: 648-655. 5. Evans, M. J., L. V. Johnson, R. J. Stephens, and G. Freeman. 1976. Renewal of the terminal bronchiolar epithelium in the rat following exposure to NO, or 0 3 . Lab. Invest. 35:246-257. 6. Evans, M. J., R. J. Stephens, L. J. Cabral-Anderson, and G. Freeman. 1972. Cell renewal in the lungs of rats exposed to low levels of NO,. Arch. Environ. Health 24:180-188. 7. Lum, H., L. W. Schwartz, D. L. Dungworth, and W. S. Tyler. 1978. A comparative study of cell renewal after exposure to ozone or oxygen. Response to terminal bronchiolar epithelium in the rat. Am. Rev. Respir. Dis. 118:335-345. 8. Shami, S. G., L. A. Thibodeau, A. R. Kennedy, and J. B. Little. 1982. Recovery from ozone-induced injury in the lungs of the Syrian Golden hamster. Exp. Mol. Pathol. 36:57-71. 9. Castleman, W. L., D. L. Dungworth, L. W. Schwartz, and W. S. Tyler. 1980. Acute respiratory bronchiolitis: an ultrastructural and autoradiographical study of epithelial cell injury and renewal in rhesus monkeys exposed to ozone. Am. J. Pathol. 98:811-827. 10. Hyde, D. M., C. G. Plopper, A. J. Weir, R. D. Murnane, D. L. Warren, and J. A. Last. 1985. Peribronchiolar fibrosis in lungs of cats chronically exposed to diesel exhaust. Lab. Invest. 52: 195-206. II. Massaro, G., and D. Massaro. 1986. Development of bronchiolar epithelium in rats. Am. J. Physiol. 250:R783-R788. 12. Hyde, D. M., C. G. Plopper, P. H. Kass, andJ. L. Alley. 1983. Estimation of cell numbers and volumes of bronchiolar epithelium during rabbit lung maturation. Am. J. Anat. 167:359-370: 13. Plopper, C. G., J. L. Alley, C. J. Serabjit-Singh, and R. M. Philpot. 1983. Cytodifferentiation of the nonciliated bronchiolar epithelial (Clara) cell during rabbit lung maturation: an ultrastructural and morphometric study.
613
Am. J. Anat. 167:329-357. 14. Tyler, N. K., D. M. Hyde, A. G. Hendrickx, and C. G. Plopper. 1988. Morphogenesis of the respiratory bronchiole during fetal lung development in the rhesus monkey. Am. J. Anat. 182:215-223. 15. Tyler, N. K., D. M. Hyde, A. G. Hendrickx, and C. G. Plopper. 1989. Cytodifferentiation of two epithelial populations of the respiratory bronchiole during fetal lung development in the rhesus monkey. Anat. Rec. 225:297-309. 16. Shami, S. G., andM. J. Evans. 1991. Kinetics of pulmonary cells. In Comparative Biology of the Lung. R. A. Parent, editor. CRC Press, Boca Raton, FL. 145-156. 17. Plopper, C. G. 1990. Structural methods for studying bronchiolar epithelial cells. In Models of Lung Disease: Microscopy and Structural Methods. 1. Gil, editor. Marcel Dekker, New York. 537-599. 18. Dungworth, D. L., W. S. Tyler, and C. G. Plopper. 1985. Morphologic methods for gross and microscopic pathology. In Toxicology of Inhaled Materials. H. P. Witschi andJ. D. Brain, editors. Springer-Verlag, New York. 229-258. 19. Sorokin, S. P. 1968. Reconstructions of centriole formation and ciliogenesis in mammalian lungs. J. Cell Sci. 3:297-320. 20. McDowell, E. M., C. NewKirk, and B. Coleman. 1985. Development of hamster tracheal epithelium. 1. A quantitative-morphologic study in the fetus. Anat. Rec. 213:429-447. 21. McDowell, E. M., C. NewKirk, and B. Coleman. 1985. Development of hamster tracheal epithelium. II. Cell proliferation in the fetus. Anat. Rec. 213:448-456. 22. Bowden, D. H. 1983. Cell turnover in the lung. Am. Rev. Respir. Dis.
128:546-S48. 23. Shami, S. G., R. K. Wolff, F. F. Hahn, A. L. Brooks, and W. C. Griffith. 1985. Early cytokinetic and morphological response of rat lungs to inhaled benzo(a)pyrene, gallium oxide, and SO,. Environ. Res. 37: 12-25. 24. Daniels, C. P., 1. L. Ponting, and T. M. Dexter. 1989. Growth and development of haemopoietic cells: a deterministic process? Hamatol. Bluttransfus. 32: 172-177. 25. Potten, C. S., andM. Loeffler. 1990. Stem cells: attributes, cycles, spirals, pitfalls and uncertainties. Lessons for and from the crypt. Development 110:1001-1020. 26. Cotsarelis, G., S. Cheng, G. Dong, T. Sun, and R. M. Lavker. 1989. Existence of slow-cycling limbal epithelial basal cells that can be preferentially stimulated to proliferate: implications on epithelial stem cells. Cell 57: 201-209. 27. Cotsarelis, G., T. Sun, and R. Lauker. 1990. Label-retaining cells reside in the bulge area of pilosebaceous unit: implications for follicular stem cells, hair cycle, and skin carcinogenesis. Cell 61:1328-1337. 28. Plopper, C. G., D. M. Hyde, and A. R. Buckpitt. 1991. Clara cells. In The Lung: Scientific Foundations. R. G. Crystal and J. B. West, editors. Raven Press, New York. 215-226. 29. Serabjit-Singh, G. S., S. J. Nishio, R. M. Philpot, and C. G. Plopper. 1988. The distribution of cytochrome P-450 monooxygenase in cells of the rabbit lung: an ultrastructural immunocytochemical characterization. Mol. Pharmacol. 33:279-289.
Although it is well established that the nonciliated bronchiolar epithelial (Clara) cell serves as the progenitor for itself and ciliated cells in the adult lung following bronchiolar epithelial injury, the nature of this relationship during development has not been well characterized. To define the pattern of proliferation and differentiation of bronchiolar ciliated and nonciliated cells, lungs of fetuses and offspring from timemated New Zealand White rabbits, ranging in age from 24 days of gestation to 25 wk postnatal (PN), were fixed by airway infusion and embedded for simultaneous light and transmission electron microscopy. Three categories of cells could be distinguished in terminal bronchioles: nonciliated cells with abundant glycogen and variable numbers of organelles; nonciliated cells with little glycogen, large numbers of polyribosomes, and variable numbers of basal bodies; and ciliated cells with cilia of varying height. Together, both types of nonciliated cells were 100% of the epithelium at 24 and 27 days gestation age (DGA). At 30 days DGA, they were 85% of the population; at all postnatal ages, they ranged from 75 to 81% of the total population. Nonciliated cells with polyribosomes and basal bodies were 10 to 20% of the total nonciliated cell population between 24 DGA and 1 wk PN and not found thereafter. Ciliated cells were not observed in animals younger than 30 DGA. Labeling indices of bronchiolar epithelium in fetuses of pregnant rabbits injected with tritiated thymidine, as determined by autoradiography, were 57 cells per thousand at28 DGA (1 h postinjection [PI]), 76 at 29 DGA (24 h PI), and 114 at 30 DGA (48 h PI). Almost all (98 %) of the labeled bronchiolar cells were nonciliated cells with glycogen and no basal bodies at 1 h PI. Nonciliated cells with basal bodies were 2 to 3 % of the labeled cells at all time points. Ciliated cells were unlabeled until 24 h PI (2 %) and increased to 7 % at 48 h PI. We conclude that: (1) bronchiolar ciliated cell differentiation occurs over a short period of time immediately before and after parturition; (2) differentiation of ciliated cells from nonciliated cells involves a transitional cell in which glycogen is lost and polyribosomes are synthesized before the synthesis of basal bodies and cilia; (3) nonciliated bronchiolar cells serve as the progenitors both for themselves and for ciliated cells during the normal process of bronchiolar epithelial differentiation in the fetus.
Among the functions proposed for the nonciliated epithelial cell in distal bronchioles of adult mammals is the role of progenitor for itself and ciliated cells. Evidence for this derives primarily from experimental pathologic studies. As is true for other epithelial populations lining the respiratory airspaces, the mitotic activity of bronchiolar epithelium in steady-state conditions is extremely low (see references I (Received in original form llugust 12, 1991 and in revised form July 13, 1992) Address correspondence to: Charles G. Plopper, Ph.D., Professor ofVeterinary Anatomy and Cell Biology, School of Veterinary Medicine, University of California, Davis, CA 95616-8732. Abbreviations: agranular endoplasmic reticulum, AER; days gestational age, DGA; days postnatal age, DPN; postinjection, PI; postnatal, PN. Am. J. Respir. Cell Mol. BioI. Vol. 7. pp. 606-613, 1992
through 3 for review). For species with poorly developed respiratory bronchioles, such as the rat, hamster, and mouse, the most definitive study of bronchiolar epithelial kinetics was performed in rats exposed acutely to nitrogen dioxide (4). All of the cells in injured bronchiolar epithelium that took up [3H]thymidine I h after injection were nonciliated cells in one of three categories. With time, there was clear transition to other types of nonciliated cells and eventually to ciliated cells. The frequency of labeled nonciliated cells also increased steadily to 50 % by day 6 postinjection. Other studies of rodents exposed to NO z, 0 3 , or high concentrations of O, have confirmed that the majority of the labeled bronchiolar cells after injury are nonciliated and that ciliated as well as Clara cells represent a larger portion of the labeled cells at later time periods (5-8). In species with extensive respiratory bronchioles, such as macaque monkeys and cats,
Plopper, Nishio, Alley et al.: Role of Clara Cells in Bronchiolar Epithelial Differentiation in·the Lung
experimental pathologic studies indicate that the nonciliated bronchiolar (or Clara) cell is also the bronchiolar epithelial progenitor for these species (9, 10). Whether the relationship between ciliated and nonciliated cells exists in steady-state conditions and especially in differentiating bronchiolar epithelial populations during lung development has not been established. Previous studies of differentiating bronchiolar epithelium in rat (11), rabbit (12, 13), and rhesus monkey (14, 15) suggest a progenitor role for the nonciliated cell population. In all three species, ciliated cells appear in a previously nonciliated bronchiolar epithelial population during perinatal lung development. The proportion of ciliated cells within the bronchiolar population appears to increase rapidly and at the expense of the nonciliated cell population. However, as emphasized in a recent review of pulmonary cell kinetics (16), the pattern and time course of events resulting in the shift from a completely nonciliated population to a mixture of ciliated and nonciliated cells has not been defined, nor has the kinetic relationship between nonciliated and ciliated cells been characterized during development. The purpose of the present study was twofold: (1) to identify the progenitor cell population for bronchiolar epithelium during maturation in the fetal lung and (2) to characterize the pattern of differentiation from a nonciliated cell population into a mixed population of nonciliated and ciliated cells.
Materials and Methods Male New Zealand White rabbits, free of chronic respiratory disease, as judged by gross examination and histopathology, were used in this study. Table 1 summarizes the ages and numbers of animals used to calculate population densities and proportions of bronchiolar epithelium in rabbit lungs. Postnatal animals were obtained through the National Institute of Environmental Health Sciences from a colony free of respiratory disease. They were anesthetized with sodium pentobarbital (1 mg/kg body weight). The trachea was cannulated, and the animals were killed by exsanguination. The
TABLE 1
Age, number of animals, number of terminal bronchioles, and number of cells counted for estimated cell density and proportions of ciliated and nonciliated bronchiolar epithelium Animals
Bronchioles
Cells
Age
(n)
(n)
(n)
24 DGA 27 DGA 30 DGA 1-2 DPN 3-4 DPN I wk 2 wk 3 wk 4 wk 5 wk 15-25 wk
5 3 4 4 3 3 4 3 3 3
45 41 60 65 79 34 48 21 34 41 53
634 440 680 742 880 440 600 360 249 350 533
7
Definition ofabbreviations: DGA natal age.
607
thorax was opened by an incision in the diaphragm, and the lung was infused via the trachea by glutaraldehyde/paraformaldehyde in cacodylate buffer (17, 18) adjusted to pH 7.4 and 550 mOsm (at 30 em water pressure). From 2 to 3 h later, the fixed lungs and trachea were removed from the chest and stored in the same fixative until processing. Prenatal animals were obtained by laparotomy from timemated pregnant rabbits anesthetized with acepromazine (1 mg/kg body weight), Rompun (5 mg/kg body weight), and ketamine (50 mg/kg body weight) (intramuscularly). Table 1 summarizes the fetal ages (days gestational age [DGA] or days postnatal age [DPN]) and number of animals used for estimating bronchiolar epithelium densities and proportions. Table 2 summarizes the fetal age and number of animals used for autoradiographic studies of ['Hlthymidine uptake. After removal of the fetus from the uterus, the mouth and nostrils of the sedated fetus were occluded to prevent breathing, the trachea was cannulated, and the thorax was opened by an incision in the diaphragm. For the 27 to 30 DGA fetuses, the trachea and lungs were fixed by infusion at 40 em water pressure with the same fixative used for postnatal animals. The tracheas of the 24 DGA fetuses were ligated, and the lungs and trachea were removed and fixed by immersion in the same fixative used for other age groups. Gonads were removed from all animals that were 1 wk of postnatal age and younger. Sex was determined from histologic sections of the gonadal tissue. Only males were included in this study. For autoradiographic study of ['Hlthymidine uptake, time-mated New Zealand White rabbits were received from the breeding colony on day 21 of pregnancy. The rabbits were housed in a barrier-maintained facility with nonrecycled air and given food and water ad libitum. On day 28 of pregnancy, maternal rabbits were injected intravenously at 9:00 a.m. with [3H]thymidine (1 mCi/kg body weight; specific activity, 80.1 Ci/mmol; New England Nuclear, Boston, MA). Fetuses were removed by laparotomy from anesthetized mothers at 1, 6, 24, and 48 h postinjection (PI). Samples of fixed lungs for light and electron microscopy were selected from the right caudal lobe. The lobe was sliced into 2- to 4-mm-thick slices. A dual-viewing Wild M-8 dissecting microscope was used to select slices that contained terminal bronchioles. Selected pieces (at least five per lung) were postfixed for 2 h in 1% osmium tetroxide. The blocks were embedded in Araldite 502 by a process that allows selection of specific areas from large tissue faces (17). The l-um sections were produced with glass knives using a
TABLE 2
Age, number of animals, number of terminal bronchioles, and number of cells counted for autoradiography Fetal Age
28 28 29 30
DGA DGA DGA DGA
Animals (n)
2 4 5 4
Time Postinjection (h)
Bronchioles (n)
Cells (n)
1 6 24 48
59 164 147 132
1,505 4,039 5,318 4,611
= days gestational age; DPN = days postDefinition of abbreviation: DGA = days gestational age.
608
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 7 1992
Sorvall JB-4 microtome. Sections from animals receiving pH]thymidine were coated with Ilford L-4 emulsion (diluted 1:1) and incubated for 9 wk. Slides were developed with Kodak D-19. All sections were stained with toluidine blue. For both light and electron microscopy, only areas in which terminal bronchiole-alveolar duct junctions were present were evaluated. The number of terminal bronchioles evaluated for estimations of population densities and proportions is summarized in Table 1. Areas containing terminal bronchiole-alveolar duct junctions were removed from the large blocks, remounted, and the junctional areas confirmed using 0.5-J.tm sections made with a Sorvall MT-5000. For proportional counting, a minimum of seven terminal bronchioles were sampled per animal. For autoradiographic evaluation of tritiated thymidine uptake, a minimum of 25 bronchioles were evaluated per animal. Thin sections (30 to 50 nm) were produced with a diamond knife on a Sorvall MT-5000 ultramicrotome, stained with uranyl acetate and lead citrate, and examined with a Zeiss EM-1O at 60 kYo Nuclear size of ciliated andnonciliated bronchiolar epithelial cells was estimated from serial sections of bronchiolar epithelium for three age groups, 30 DGA, 4 wk PN, and 17 wk PN, as described in detail previously (12). Evaluations
r-
of the number of cells per unit length of basement membrane, the number of cells per unit area of basement membrane, and the number of cells per unit volume of bronchiolar epithelium were not significantly different between 30 DGA and 4 wk PN. However, an error of 15.4% was observed at 17 wk. Because there were no statistical differences in the estimation, the number of nuclei per unit length of basal lamina was used to estimate the numerical density of epithelial cells in terminal bronchioles and the percentage of ciliated and nonciliated cells in the bronchioles by light microscopy on l-am sections. Values for adults (ranging in age from 15 to 25 wk) were corrected for overestimation and underestimation. The percentage of the nonciliated cell population that was defined as preciliated cells was estimated by differential counting of epithelial cells in all of the terminal bronchioles evaluated for the ages 24 DGA through 2 wk PN by transmission electron microscopy of adjacent sections. For calculating labeling indices (number of labeled cells per 1,000 cells) based on tritiated thymidine uptake, a minimum of 25 terminal bronchioles were counted per animal. A minimum of750 cells were also counted per animal. Table 2 summarizes the number of cells counted per time period PI. The slides of bronchioles containing labeled cells were photographed in their entirety at an original magnification of 400x and then mapped with contiguous 4 X 5 inch prints. Each labeled cell was identified on the map and then characterized on serial sections by electron microscopy from the same bronchiole (Figure 1). The labeling index is reported as the number of labeled nuclei per 1,000 cells counted by light microscopy. The number of cells undergoing mitosis was also counted, and the mitotic index is reported as number of mitotic nuclei per 1,000 cells (n was considered to be the number of animals). The proportion of the labeled population represented by each cell type is reported as the percentage of all labeled cells.
Results
Figure 1. Serial l-um and ultrathin sections of labeled and unlabeled bronchiolar epithelial cells. The animal was injected with tritiated thymidine, and uptake of labeled measure identified in cells by autoradiography on l-um sections. The labeled cells identified histologically (in brackets) (A) (bar = 100 /tm) were then characterized using transmission electron micrographs of serial sections of the same bronchioles (B) (bar = 5 /tm).
Bronchiolar cells were in approximately equal abundance (12 cells/IOO J.tm) at all ages except at the earliest fetal age examined (24 DGA) (Figure 2A). The nonciliated population as a whole was 100 % of the cell population at 24 and 27 DGA, decreased to approximately 90% at 30 DGA, and decreased to approximately 75 to 80% in animals 2 wk and older (Figure 2B). In fetuses 24 to 27 DGA, two categories of nonciliated cells could be identified. By light microscopy, the most abundant category was cuboidal in appearance with a rounded nucleus and a dome-shaped apical projection that was filled with a heavily stained material. Electron microscopic evaluation of these cells during the early prenatal and postnatal time period showed that their cytoplasm was filled with glycogen (Figure 3A). These cells had variable amounts of agranular endoplasmic reticulum (AER) in the borders of the cytoplasm on the lateral and apical surfaces. The amount of AER increased with age. The central portions of some of the large areas of glycogen also contained AER. Mitochondria and granular endoplasmic reticulum, along with Golgi apparatus, were observed in small amounts, both apically and basally to the nucleus (Figure 3A). Nonciliated cells varied in abundance from 90% (24 DGA) to 66% (30 DGA) of the bronchiolar population in the perinatal period. They
Plopper, Nishio, Alley et al.: Role of Clara Cells in Bronchiolar Epithelial Differentiation in the Lung
A "-0e
609
20
15
>-
C:::i.
ClIo
"0 0
-~
cau~ '': Gi
10
CIIu
E""
~
.....
5
O...L..------------------Figure 2. Changes in abundance of ciliated and nonciliated cells in terminal bronchioles of prenatal and postnatal rabbits. Birth is indicated by the arrows. (A) Numerical density of epithelium in terminal bronchioles in both prenatal and postnatal animals as number per 100 !-tm of basal lamina. (B) The relative proportion (percentage) of cells within the bronchiolar epithelium. Solid squares = nonciliated cells; open triangles = ciliated cells; solid circles = nonciliated cells with polyribosomes.
....
« -c -c
on on
o0
ClCl 00
~
~ g+~..,.
on
""
>- >-
~
00
""
~
N
~..,
on
on
""
"":B ..,~
~ •
~
:B
~
_ ..,
:;
«
It)
Age
B 100
~
Gi
80
u
-
c;
.2....... 60
OC
Nonciliated cells Preciliated cells
CII
-B g' c::
cf
I
Ciliated cells
40
,/f",1---i-'-c---------------------r 20
o...L..---.(r-oi~--....L..-.,.-
« «
c(
U)
U)
~
''f ~
e e e ;;-;;- :B 0
0
~
were 66% (adult) to 78% (30 DGAand 14 DPN) of the population in older animals. A small proportion of the nonciliated cell population at all ages between 24 DGA and 1 wk had a flatter apical surface and was free of densely staining cytoplasmic content but did not have cilia. These cells had the ultrastructural features of preciliated cells undergoing ciliogenesis (Figures 3B and 3C). The cytoplasm was nearly free of glycogen. In some cases (Figure 3B), large numbers of polyribosomes were found distributed throughout the cytoplasm. In the apical cytoplasm, there were areas of condensing microtubule-like structures. In other cells (Figure 3C), large numbers of polyribosomes were present but, in addition, condensing basal bodies were irregularly arranged
.... N
~..,
...- .....- .....---------__.
. "".. "".. .. "" i :B
~ N
:B
..,~
~ •
~
..,
:>
-c
It)
Age
along the apical portion of the cell. Cells containing these ultrastructural features composed approximately 10% of the cell population in 24 DGA animals, were as high as 18% of the population immediately before birth, and decreased to 0% of the population at 2 wk PN (Figure 2B). Ciliated cells were clearly identified ultrastructurally by their extensive cilia, abundance of mitochondria in the apical cytoplasm, the presence of a Golgi apparatus near the nucleus and strands of GER in a cytoplasm nearly free of glycogen (Figure 3D). Ciliated cells were not observed in 24 and 27 DGA fetuses. They composed 15% or more of the population in animals 30 DGA and older (Figure 2B). For autoradiographic evaluation of bronchiolar kinetics,
Figure 3. Ultrastructural comparison of the features of nonciliated and ciliated epithelial cells in the bronchiolar epithelium of perinatal rabbits (27 DGA). L = airway lumen; Nu = nucleus. (A) The majority ofthe nonciliated cells contained abundant glycogen (Gly), mitochondria (M), and small amounts of agranular endoplasmic reticulum (arrowheads) and rough endoplasmic reticulum (solid arrows). (Bar = 1 I-'m). (B) Many of the nonciliated cells contained an abundance of polyribosomes (open arrows), areas of condensing microtubules, and small amounts of smooth endoplasmic reticulum (arrowheads) and rough endoplasmic reticulum (solid arrows). Asterisks indicate condensing basal bodies. (Bar = 1 I-'m.) (C) Many of the nonciliated cells resembled those in panel B but also included basal bodies or forming centrioles near the apical surface (asterisks). Go = Golgi apparatus. (Bar = 1 I-'m.) (D) During this time, a small percentage, of the cells resembled ciliated cells with mature cilia attached to basal bodies (asterisks). The remainder of the cell cytoplasm was primarily
free of glycogen and contained a variety of organelles including mitochondria (M) and small amounts of rough endoplasmic reticulum (solid arrows). (Bar = 1 I-'m.)
Plopper, Nishio, Alley et al.: Role of Clara Cells in Bronchiolar Epithelial Differentiation in the Lung
A
611
140
120
100 )(-
C1lCll '0= CC1l
80
0)0 Co =0 C1l ....
60
.-
(,,)
.0--
CIS~
...J
40
20
Figure 4. Kinetics of bronchiolar epithelium immediately before birth (28 DGA to 30 DGA). (A) Labeling index of bronchiolar epithelium in terminal bronchioles of fetal rabbits (as number of labeled nuclei per 1,000 cells). Times shown are following injection of tritiated thymidine. (B) Percentage of labeled cells based on ultrastructural features. Solid squares = nonciliated cells resembling Figure 3A; solid circles = nonciliated cells resembling Figures 3B and 3C; open triangles = ciliated cells resembling Figure 3D. Times shown are following injection of tritiated thymidine.
48 (30dga)
!!!. Qj o
90
'0 C1l
Qj
--------
.0_
--
CIS~
~
-~ 0
..
--A--
Labeled Nonciliated cells Labeled Preciliated cells Labeled Ciliated cells
C C1l
o
C1l
C.
10
------
0
6 (Fetal age)
----------------
(2Bdga)
(2Bdga)
24 (29dga)
--f 48 (3Odga)
Time after injection (h)
we chose the time period of most rapid transition and differentiation of ciliated cells based on cell density and proportion profiles: 28 to 30 DGA. One hour after administration of tritiated thymidine to the mother, almost 6 % of the bronchiolar epithelial cells in fetal rabbits were labeled (Figure 4A). Almost all, approximately 98%, of the labeled cells
were glycogen-filled nonciliated cells (Figure 4B). The remainder (approximately 2 %) of the labeled cells were nonciliated cells identified above as glycogen-free preciliated cells (Figure 4B). These labeled cells contained many polyribosomes, few condensing microtubular areas, and no basal bodies. The labeling of the total bronchiolar epithelial popu-
612
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 71992
lation doubled by 48 h PI (Figure 4A). Nonciliated glycogenfilled cells still composed almost 93 % of the labeled cells at 48 h PI (Figure 4B). Preciliated cells made up < 2 % of labeled cells at 48 h PI (Figure 4B). Labeled ciliated cells were not observed at 1 and 6 h PI (Figure 4B). At 24 h PI, < 3% of the labeled cells were ciliated cells. This number was more than doubled by 48 h PI. The mitotic index was 3.32 cells/l,OOO cells at 1 h PI (28 DGA), 1.73 cells/l,OOO at 6 h PI (28 DGA), 2.63/1,000 at 29 DGA, and 1.73cells/1,000 at 30 DGA. All the cells with mitotic figures were nonciliated cells.
Discussion The purpose of this study was to characterize the kinetic relationship between nonciliated and ciliated cells in the bronchiolar epithelium during the period of active differentiation of the Clara cell. The relative densities of these two cell types differ between species like the rabbit (12) and the rat (2), and our study indicates that differentiation also varies. In the rabbit, the majority of ciliated cell differentiation occurs prenatally and the steady-state condition found in the adult is almost established by birth. For the rat, in contrast, the steady-state conditions are not present immediately after birth and there is a 3- to 4-fold increase in the density of ciliated cells postnatally that continues for at least 60 DPN (11). In the rabbit, the intermediate form, the preciliated cell, is present for a relatively limited period prenatally and post. natally. Whether the intermediate (preciliated) form is found in rat bronchioles during development has not been reported. Ciliogenesis in the bronchiole of the rabbit does not follow the same pattern of events as in rat bronchi (19). In the rabbit, differentiating ciliated cells do not have fibrogranular accumulations or deuterosomes (19). Whether all the forms we observed in the rabbit are present in the rat bronchiolar epithelium either prenatally or postnatally has not been established by previous studies (11). Differentiation of Clara cells in the rabbit requires 3 to 4 wk, whereas ciliated cell differentiation is complete by 1 wk PN, well before nonciliated cells contain ultrastructural features characteristic of mature Clara cells. In contrast, in the rat, ciliogenesis is active over a significant period of postnatal age long after the differentiation of the Clara cell has occurred (11). This pattern of differentiation in bronchiolar epithelium in the rabbit is similar to that observed in respiratory bronchioles of the rhesus monkey, with the exception that it begins much earlier in the monkey and the bronchiolar epithelial population actually differentiates into four different cell types instead of just two (14, 15). Clara cells also appear to be the progenitor for ciliated cells in the trachea of the hamster during development (20, 21), with Clara cell differentiation occurring much later than that of ciliated cells. When compared with the kinetic activity of bronchiolar epithelium in the steady.state in the adult, the mitotic activity of bronchiolar epithelium in the perinatal period is extremely high. To the best of our knowledge, these are the highest labeling and mitotic indices reported for bronchiolar epithelium at any age (see reference 16 for summary ofliterature). The labeling index is more than twice as high as that observed in l-rno old rats (5.7 versus 1.1 %) or postnatal mice
(0.52%) (22). High labeling indices have also been reported in the parenchyma in fetal mice (1). Labeling indices for bronchioles of adults (0.2% or less) indicate that kinetic activity is well below the prenatal animal (16). Labeling indices in the range observed by us in the late fetal period have been observed in the adult only with acute bronchiolar injury, such as that resulting from exposure to ozone (7, 9) or nitrogen dioxide (5). Injury from diesel exhaust (10), or benzo(a)pyrene (23), elevates cell turnover minimally. Our study shows a clear progenitor relationship between nonciliated bronchiolar epithelial cells and ciliated bronchiolar cells during the period oflung maturation when these two cell types are undergoing cytodifferentiation. Our 2-day pulse label study, applied during the most active phase of transition from a strictly nonciliated to a mixed ciliated and nonciliated cell population, indicates that some portion (about 5 %) of the nonciliated cell population is clearly the progenitor for both the ciliated and the rest of the nonciliated population. Mitotic figures were observed only in nonciliated cells. The nonciliated cells containing large amounts of glycogen and few organelles appear to be the origin of differentiated ciliated cells. They have the highest labeling index at any time during the 2-day period and represent almost all (98 %) the labeled cells. The population of preciliated cells represents a small proportion of the labeled cells and shows most of the kinetic characteristics of a cell population in transition. The labeling index for this cell population indicates that it is in steady state throughout the 2-day period. This is in contrast to nonciliated cells, where the labeling index almost doubles during the 2-day period, and to ciliated cells, where the labeling index doubles at 24 h and again at 48 h PI. The shifts in labeling we observed in the fetus match the pattern reported by Evans and co-workers (5) for bronchiolar epithelial repair following acute injury in adults. Our observations are compatible with two patterns of differentiation: (1) the labeled cells represent a small population of undifferentiated cells whose daughters differentiate into both the slowly differentiating Clara cell population and the rapidly differentiating ciliated cell population or (2) a subpopulation of the undifferentiated Clara cell population that will differentiate later into Clara cells after the stimulus for active mitosis is over, and some of whose daughter cells will differentiate into ciliated cells. We believe that the latter is more likely because of our inability to differentiate between the labeled and unlabeled nonciliated cells on the basis of ultrastructural features and because the labeled nonciliated cells have none of the features associated with purported stem cells in other tissues (24-27). The former pattern should not be completely ruled out, however, since these cells may exist in the adult as a very small population of undifferentiated cells stimulated to divide only under extreme conditions. If this is the case, they are in such small numbers that they have not been detected in healthy adults by morphometric studies (see reference 28 for review). One of the characteristics of stem cells is their lack of some differentiated functions found in their differentiated offspring (24-27). We have observed a small number of nonciliated bronchiolar cells in adult rabbits that lacked antigens for cytochrome P-450 monooxygenases (29) and may represent such a population. The failure to detect such a cell population in kinetic
Plopper, Nishio, Alley et al.: Role of Clara Cells in Bronchiolar Epithelial Differentiation in the Lung
studies in adults following injury may be accounted for by the fact that none of the studies involved an injurant that targeted the nonciliated population (4-10, 23). Acknowledgments: This study was supported by Grant HL43032 from the National Institutes of Health.
References I. Kauffman, S. L. 1980. Cell proliferation in the mammalian lung. Int. Rev. Exp. Pathol. 22: 131-191. 2. Massaro, G. C. 1989. Nonciliated bronchiolar epithelial (Clara) cells. In Lung Cell Biology, D. Massaro, editor. Marcel Dekker, New York. 81-114. 3. Plopper, C. G., and D. L. Dungworth. 1987. Structure, function, cell injury and cell renewal of bronchiolar and alveolar epithelium. In Lung Carcinomas. E. M. McDowell, editor. Churchill Livingstone, London. 94-128. 4. Evans, M. J., L. J. Cabral-Anderson, and G. Freeman. 1978. Role of the Clara cell in renewal of the bronchiolar epithelium. Lab. Invest. 38: 648-655. 5. Evans, M. J., L. V. Johnson, R. J. Stephens, and G. Freeman. 1976. Renewal of the terminal bronchiolar epithelium in the rat following exposure to NO, or 0 3 . Lab. Invest. 35:246-257. 6. Evans, M. J., R. J. Stephens, L. J. Cabral-Anderson, and G. Freeman. 1972. Cell renewal in the lungs of rats exposed to low levels of NO,. Arch. Environ. Health 24:180-188. 7. Lum, H., L. W. Schwartz, D. L. Dungworth, and W. S. Tyler. 1978. A comparative study of cell renewal after exposure to ozone or oxygen. Response to terminal bronchiolar epithelium in the rat. Am. Rev. Respir. Dis. 118:335-345. 8. Shami, S. G., L. A. Thibodeau, A. R. Kennedy, and J. B. Little. 1982. Recovery from ozone-induced injury in the lungs of the Syrian Golden hamster. Exp. Mol. Pathol. 36:57-71. 9. Castleman, W. L., D. L. Dungworth, L. W. Schwartz, and W. S. Tyler. 1980. Acute respiratory bronchiolitis: an ultrastructural and autoradiographical study of epithelial cell injury and renewal in rhesus monkeys exposed to ozone. Am. J. Pathol. 98:811-827. 10. Hyde, D. M., C. G. Plopper, A. J. Weir, R. D. Murnane, D. L. Warren, and J. A. Last. 1985. Peribronchiolar fibrosis in lungs of cats chronically exposed to diesel exhaust. Lab. Invest. 52: 195-206. II. Massaro, G., and D. Massaro. 1986. Development of bronchiolar epithelium in rats. Am. J. Physiol. 250:R783-R788. 12. Hyde, D. M., C. G. Plopper, P. H. Kass, andJ. L. Alley. 1983. Estimation of cell numbers and volumes of bronchiolar epithelium during rabbit lung maturation. Am. J. Anat. 167:359-370: 13. Plopper, C. G., J. L. Alley, C. J. Serabjit-Singh, and R. M. Philpot. 1983. Cytodifferentiation of the nonciliated bronchiolar epithelial (Clara) cell during rabbit lung maturation: an ultrastructural and morphometric study.
613
Am. J. Anat. 167:329-357. 14. Tyler, N. K., D. M. Hyde, A. G. Hendrickx, and C. G. Plopper. 1988. Morphogenesis of the respiratory bronchiole during fetal lung development in the rhesus monkey. Am. J. Anat. 182:215-223. 15. Tyler, N. K., D. M. Hyde, A. G. Hendrickx, and C. G. Plopper. 1989. Cytodifferentiation of two epithelial populations of the respiratory bronchiole during fetal lung development in the rhesus monkey. Anat. Rec. 225:297-309. 16. Shami, S. G., andM. J. Evans. 1991. Kinetics of pulmonary cells. In Comparative Biology of the Lung. R. A. Parent, editor. CRC Press, Boca Raton, FL. 145-156. 17. Plopper, C. G. 1990. Structural methods for studying bronchiolar epithelial cells. In Models of Lung Disease: Microscopy and Structural Methods. 1. Gil, editor. Marcel Dekker, New York. 537-599. 18. Dungworth, D. L., W. S. Tyler, and C. G. Plopper. 1985. Morphologic methods for gross and microscopic pathology. In Toxicology of Inhaled Materials. H. P. Witschi andJ. D. Brain, editors. Springer-Verlag, New York. 229-258. 19. Sorokin, S. P. 1968. Reconstructions of centriole formation and ciliogenesis in mammalian lungs. J. Cell Sci. 3:297-320. 20. McDowell, E. M., C. NewKirk, and B. Coleman. 1985. Development of hamster tracheal epithelium. 1. A quantitative-morphologic study in the fetus. Anat. Rec. 213:429-447. 21. McDowell, E. M., C. NewKirk, and B. Coleman. 1985. Development of hamster tracheal epithelium. II. Cell proliferation in the fetus. Anat. Rec. 213:448-456. 22. Bowden, D. H. 1983. Cell turnover in the lung. Am. Rev. Respir. Dis.
128:546-S48. 23. Shami, S. G., R. K. Wolff, F. F. Hahn, A. L. Brooks, and W. C. Griffith. 1985. Early cytokinetic and morphological response of rat lungs to inhaled benzo(a)pyrene, gallium oxide, and SO,. Environ. Res. 37: 12-25. 24. Daniels, C. P., 1. L. Ponting, and T. M. Dexter. 1989. Growth and development of haemopoietic cells: a deterministic process? Hamatol. Bluttransfus. 32: 172-177. 25. Potten, C. S., andM. Loeffler. 1990. Stem cells: attributes, cycles, spirals, pitfalls and uncertainties. Lessons for and from the crypt. Development 110:1001-1020. 26. Cotsarelis, G., S. Cheng, G. Dong, T. Sun, and R. M. Lavker. 1989. Existence of slow-cycling limbal epithelial basal cells that can be preferentially stimulated to proliferate: implications on epithelial stem cells. Cell 57: 201-209. 27. Cotsarelis, G., T. Sun, and R. Lauker. 1990. Label-retaining cells reside in the bulge area of pilosebaceous unit: implications for follicular stem cells, hair cycle, and skin carcinogenesis. Cell 61:1328-1337. 28. Plopper, C. G., D. M. Hyde, and A. R. Buckpitt. 1991. Clara cells. In The Lung: Scientific Foundations. R. G. Crystal and J. B. West, editors. Raven Press, New York. 215-226. 29. Serabjit-Singh, G. S., S. J. Nishio, R. M. Philpot, and C. G. Plopper. 1988. The distribution of cytochrome P-450 monooxygenase in cells of the rabbit lung: an ultrastructural immunocytochemical characterization. Mol. Pharmacol. 33:279-289.
