- Email: [email protected]
Relationships between the nuclear membrane, nuclear pore complexes, and organelles in the type II pneumocyte
Tissue & Cell, 1995 27 (6) 613-619 © 1995 Pearson Professional Ltd.
Relationships between the nuclear membrane, nuclear pore complexes, and organelles in the type II pneumocyte M.L. Miller, A. Andringa, L. Hastings
Abstract. Functional relationships among organelles of the type II cell are suggested based upon the proximity of organelles to specialized areas of the plasma- and nuclear membranes. In a two-dimensional morphometric analysis of the profiles of organelles in type II cells of the ferret and rat (and beagle dog), lamellar bodies were more likely to be located near the nuclear membrane than at the alveolar space (where exocytosis occurs). The size of lamellar body profiles was not correlated with distance from the nuclear membrane; however, large profiles were nearer the apical membrane, and smaller ones nearer the basement membrane. Profiles of highly branched mitochondria were 10 times more frequently associated with nuclear pore complexes than with the inter-pore nuclear membrane. Forty percent of all mitochondrial profiles were within 0.25 #m of the nucleus, 5% were within 0.02 pm and half of these appeared to touch the filaments of the nuclear pore complexes. The size of mitochondrial profiles was not correlated with distribution. In the ferret and rat, 8.6% and 2.5% respectively, of the nuclear pore complexes were associated with mitochondria. Sebaceous cells, from control mice, demonstrated a spatial distribution of granules which was size dependent but unrelated to the nuclear membrane. Keywords: Nuclearmembrane, mitochondria, nuclear pore complexes, lamellar bodies, spatial relations
Introduction The type II pneumocyte has been shown by numerous methods to secrete surfactant which reduces surface tension in the alveolus. The secretion granules may be composed of newly produced surfactant as well as endocytically derived components (Walker et al., 1985; Williams, 1987; Williams et al., 1984). Morphometric studies have established the mean volume and numerical densities for organelles within the type II cell in a variety of species and under various experimental conditions (Crapo et al., 1978; Fram et al., 1985; Kliewer et al., Department of Environmental Health, College of Medicine, University of Cincinnati, 3223 Eden Avenue, Cincinnati, OH 45267-0056, USA. Received 17 March 1995 Accepted 26 June 1995 Correspondence to: M. L. Miller.
1985; Massaro et al., 1975; Miller et al., 1985; 1986a,b; Shimura et al., 1984; 1985; Young et al., 1985; 1986; 1991 ). Several organelles are directly involved in surfactant production including, lamellar, composite and multivesicular bodies, and cisternal bodies (Miller et al., 1982). There has been some documentation on the spatial distribution of these structures (Young et al., 1985; 1991), but rarely have the nuclear membrane and nuclear pore complexes been examined in relation to intracellular organelles (Miller et al., 1986a; 1995). Nuclear pores are octagonal ring-like structures between 50-90 nm in diameter which punctuate the nuclear membrane at the 'crossroads' between the nucleus and cytoplasm, controlling the import and export of constituents (Hanover, 1992). In fact, most nucleo-cytoplasmic exchanges occur through the nuclear pore complex (Miller et al., 1991). The number of 613
614
M I L L E R ET AL.
nuclear pores varies in different cell types, during different phases of the cell cycle, states of activity, and even at different sites on the surface of the same nucleus (Miller et al., 1991). Circumstances controlling the number of pores and their arrangement in the nuclear membrane are not known (Hanover, 1992; Miller et al., 1991). Transcriptionally active cells have the greatest number of nuclear pores. The number of nuclear pores doubles just before S phase of the cell cycle. Miller et al., (1991) also suggest that the number of nuclear pore complexes may be determined by the 'tissue and cell-specific gene expression'. It has been suggested that cytoskeletal components may be in continuity with nuclear pore filaments (Kessel, 1988; Miller et al., 1991; Miller et al., 1995). What Hanover (1992) calls cytoplasmic anchoring is likely to be a factor in nuclear transport and it appears to require energy in the form of ATP. Mitochondria lie near the nucleus in the developmental period, as well as at some specific times in the cell cycle of oocytes (Kessel, 1988). Morphologic evidence to suggest interactions between lamellar bodies and other surfactant related organelles and the nuclear membrane has been reported (Miller et al., 1986a), but measures of frequency or distance to support these observations are lacking. In addition, a new observation, the association of mitochondria with the filaments of the nuclear pore complex, has not been reported. This study describes the size and spatial distribution of profiles of lamellar bodies and mitochondria within the type II cell, and documents the extent and nature of the contacts of these organelles with the nuclear membrane and nuclear pores complexes, and contrasts this with data obtained on sebaceous cells.
Materials and methods Seven male ferrets (400 g), three male Sprague Dawley rats (300-500 g) and 10 female SENCAR mice were quarantined for at least one week. Animal care and treatment procedures were as previously (Baxter et al., 1988; Miller et al., 1982) and conform to AAALAC standards. Rats and ferrets were euthanized with pentobarbital, chests opened, tracheas dissected and lungs fixed by instillation of fluid (20 cm of fluid pressure) for 20 rain (Miller et al., 1982). The fixative contained 2% paraformaldehyde, 2.5% glutaraldehyde, 0.10% dextrose, and 0.002% calcium chloride in 0.032 M Millonig's buffer (pH 7.4, 350 mosm). The trachea was tied, the lungs removed and immersed in fresh fixative for at least 24 h and cut into 1 mm 3 cubes. Some tissue was fixed directly in osmium tetroxide in Palade's fluid (Miller et al., 1982). Blocks of tissue from middleright mid-clavicular parenchymal lung from each animal were taken. Mice were sacrificed by cervical dislocation, their interscapular skin fixed by immersion in the same
fixative. The skin was oriented in the blocks to provide sections perpendicular to the epidermis. Epon 812 embedded blocks of lung from eight female beagle dogs, at least 22 months old, were kindly supplied by Dr R. J. Stephens. The methods of tissue preparation have been published elsewhere (Stephens et al., 1973), but include both direct osmium fixation, and fixation with buffered glutaraldehyde and postfixation in osmium tetroxide. All tissues were then rinsed in buffer, postfixed for 2 h in 1% osmium tetroxide, dehydrated in a series of ascending ethanol concentrations, passed through propylene oxide, and embedded in Spurr resin. Blocks were thin sectioned and stained with uranyl acetate and lead citrate for transmission electron microscopy (60 kV). Ten to 15 type II cells from each animal were photographed at 5000 x and enlarged to 20000x (Fig. 1). Magnification at the microscope was calibrated with a Ladd Research Industries Carbon Grating. Type II cells containing a profile of nucleus, lamellar body, mitochondrion, and plasmalemma bordering the alveolar space, basement membrane and an adjacent epithelial cell were selected consecutively. It was determined that about 8% of the type II cells identified were excluded by this selection procedure, i.e. those without nuclear profiles, resulting in only a small bias (see Results) in the sampling procedure. Since randomness was not assumed for the orientation of the type II cell in the alveolus or the organelles within the cells, random orientation of specimens for sectioning was required. No adjustment was made for shrinkage. Data were obtained by measuring the shortest distance between an organelle and the plasma membrane (apical at the lumenal interface; basal - adjacent to basement membrane; lateral - adjacent to another epithelial cell; and nuclear membrane - either 'interpore' or at the nuclear pore complex). This method of data acquisition was modified from that described for a threedimensional analysis of type II cells (Young et al., 1985; 1991) and Clara cells (Young et al., 1986) with a computer-based program which calculated the shortest distance between lamellar bodies and the apical plasmalemma and the Golgi. Although the use of twodimensional techniques underestimates the actual distances between organelles, significant relationships were found. The perimeter and area of each type II or sebaceous cell and each organellar profile were measured using SigmaScan (Jandel Scientific). The prevalence of mitochondria at the nuclear pore was determined by counting the number of profiles of mitochondria in contact with the filaments of the nuclear pore complex, relative to the total number of nuclear pores. A point on the nucleus 4.0 gm from the pore was used to determine the contacts between mitochondria and control-equivalent for the length of the nuclear pores on the nuclear membrane. In addition, all profiles
ORGANELLES/NUCLEAR MEMBRANE 615
Fig. 1 (a) Type II cells contain lamellar bodies which are frequently situated within 0.02 gin of the nuclear membrane (arrowheads). The points of very close contact are primarily at the interpore region of the outer nuclear membrane. Dog, aldehyde fixation, x 21000. (b) Mitochondria are frequently associated with filaments which are a part of the nuclear pore complex (arrowhead). Cristae within the mitochondrion are parallel to the filaments of the pore. Ferret, aldehyde fixation, × 39 000. (c) Mitochondria are visibly drawn towards the pore complex where contact with the filament is visualized (arrowhead). Ferret, aldehyde fixation, x 44 000. appearing adjacent to the nuclear m e m b r a n e were counted (200-300 each for ferret and rat, 1400 for dogs). N o migration o f sebaceous cell granules for exocytosis occurs in the sebaceous cell since the cell itself becomes the sebum. In addition, the entire plasmalemma o f cells lying above the basement m e m b r a n e contacts other sebaceous cells, and was used as a 'negative control' for the effect o f specializations o f the plasmalemma on spatial locations o f the organelles. P h o t o g r a p h i c m o n tages o f cell profiles contained a nucleus (not pyknotic), and at least one granule and mitochondrion. The data were analyzed separately for each species. Statistical analyses were m a d e using the General Linear
Model (Statistical Analysis System ( P C SAS)), and Scheffe and Tukey options were used for post hoc comparisons. D a t a were considered significant when P~<0.05.
Results Type II cell The ratio o f profiles o f nuclear and cytoplasmic area was nearly the same in rats and ferrets. As the nuclear area increased so did the cytoplasmic area (ferret and rat, P<~O.O05, P~
616
MILLERET AL.
n u m b e r o f profiles o f l a m e l l a r bodies a n d m i t o c h o n d r i a a n d the p e r c e n t o f the c y t o p l a s m which l a m e l l a r bodies c o m p r i s e d was similar in the ferret a n d r a t ( T a b l e 1). These values are similar to previous r e p o r t s ( M a s s a r o et al., 1975; M i l l e r et al., 1982; 1986; Y o u n g et al., 1985; 1991). Preselected type II cells (containing profiles o f l a m e l l a r b o d y , nucleus, m i t o c h o n d r i o n , a n d m e m b r a n e s at the lumen, b a s e m e n t m e m b r a n e a n d an epithelial cell) ( T a b l e 2) were n o t statistically different f r o m cells selected on a r a n d o m basis. T h e a m o u n t o f p l a s m a l e m m a specialized at each o f these sites was n e a r l y equal, yet, m o r e t h a n t w o - t h i r d s o f the l a m e l l a r b o d i e s were l o c a t e d within 0.25 ~tm o f the nuclear m e m b r a n e ( T a b l e 3). H a l f o f all the l a m e l l a r b o d i e s were l o c a t e d close to the i n t e r p o r e a r e a o f the n u c l e a r m e m b r a n e , a n d a b o u t 12% a n d 8% a p p e a r e d to coalesce with it in the ferret a n d the r a t respectively. T h e highest p e r c e n t a g e o f l a m e l l a r b o d y c o n t a c t with the n u c l e a r m e m b r a n e was seen in the d o g (13.9%). A s the a r e a o f profiles o f l a m e l l a r bodies increased, distance to the m e m b r a n e at the a l v e o l a r space decreased
Rat type II cell
Ferret type II cell
Mouse sebaceous cell
n = 20 35.9_+2.9 gm2. 27.1 -+ 1.3 gm n = 20 10.2-+1.1 pan2 14.5 ± 1.5 [ira 25%
n = 25 43.0_+2.7 27.6 _+1.0 n = 25 11.7-+1.1 15.6 _+0.9 27%
n = 14 181.9_+22.3 53.3 _+3.6 n = 14 34.6_+4.4 20.8 _+1.6 19%
n = 208
n = 266
n = 261
0.52_+0,03 gm2 15%
0.57_+0.05 15%
0.74_+0.08 13%
n = 14l 0.28_+0.02 Jam2 5.4%
n = 222 0.21 _+0.02 4.3%
n = 148 0.21 _+0.03 2%
*mean_+SEM; +does not include microvilli. Table 2
Contributions to the plasmalemma Alveolar space Cell/cell membrane Basal plasmalemma (Nuclear membrane)*
Rat type II cell
Ferret type II cell
Mouse sebaceous cell
Lamellar bodies < 0 . 0 2 ttm from:
Nuclear membrane Alveolar space Cell/cell membrane Basement membrane
8%
7%
1
2
0
0
2
t
45%
38%
10
13
4
3
26
14
0.4%
0.8
Lamellar bodies < 0.25 ~m from:
Nuclear membrane Alveolar space Cell/cell membrane Basement membrane
6%
8
Mitochondria < 0 . 0 2 ~m from:
Table 1
Cell area Cell perimeter Nucleus Area Perimeter + Volume density Secretory granules Area Volume density Mitochondria Area Volume density
Table 3
Rat type II cell
Ferret type II cell
Mouse sebaceous cell
44%
52%
-
8 48
9 39
-
(54)
(52)
100 (39)
*The total plasma membrane consisted of membrane at the alveolar space, the cell/cell contacts and the basement membrane; the nuclear membrane was equivalent to 54% of the total distance of the plasmalemma.
Nuclear membrane Alveolar space Cell/cell membrane Basement membrane
5%
4%
0.07%
0
0.5
0
0
3
0.5
40%
27%
9
9
-
6
8
20
33
19
1.4 -
Mitochondria < 0.25 I~m from:
Nuclear membrane Alveolar space Cell/cell membrane Basement membrane
10%
-
in b o t h the ferret a n d r a t (P ~<0.0003, P~<0.0015 respectively). In ferrets, there was no c o r r e l a t i o n between a r e a o f l a m e l l a r bodies a n d distance to the b a s e m e n t m e m brane. In rats, however, larger l a m e l l a r bodies were significantly f a r t h e r f r o m the b a s e m e n t m e m b r a n e (P~<0.004) a n d smaller l a m e l l a r bodies were l o c a t e d basally. W h e t h e r these smaller l a m e l l a r bodies were c o m p o s i t e was n o t determined. A r e a o f the l a m e l l a r b o d y profiles was n o t related to the distance to the nuclear m e m b r a n e in either ferrets or rats t h o u g h there was a t r e n d for larger l a m e l l a r bodies to be close to the nuclear m e m b r a n e in the ferret. D a r k multivesicular bodies were l o c a t e d n e a r nuclei (P~<0.004) while light multivesicular bodies showed no such distribution. D i s t r i b u t i o n o f profiles o f multivesicular bodies within the cell d i d n o t a p p e a r to be related to o r g a n e l l a r area, t h o u g h s a m p l e size was small.
ORGANELLES/NUCLEARMEMBRANE 617
Mitochondrial profiles were slightly smaller in the type II cells of the ferret than the rat, though the number of profiles as well as the volume density were similar for these highly branched organelles (Table 1). About half as many were seen in the dog. Profiles of mitochondria were frequently found within 0.25 gm of the nuclear membrane in rat, ferret, mouse (Table 3) and in dog. Considering the branching nature of mitochondria, this would include the majority if not all mitochondria with at least some segment at the nuclear membrane. At least 5.7% and 3.6% of the profiles in the rat and the ferret respectively are within 0.02 gm of the nucleus. On occasion, a single profile of a mitochondrion contacted as many as three adjacent nuclear pore complexes. In addition, 8.6%_+1.6, and 2.5%+0.009 (S.E.M.) and 14.5%_+4.0 of nuclear pores appeared to be in direct contact with mitochondria in the ferret, rat and beagle respectively. At points on the nuclear membrane 4 gm from each nuclear pore complex, only about 2% of the mitochondria were in contact with the nuclear membrane. While the length (in microns) occupied by nuclear pore complexes in the type II cell is a small fraction of the whole nuclear membrane, about half of all mitochondrial contact with the nuclear membrane occurred there. Larger mitochondrial profiles appeared to be polarized towards the nucleus in the ferret, while larger profiles were closer to the junctional complexes between epithelial cells in the rat. In contrast, less than 1% of the profiles of lamellar bodies were associated with the nuclear pores and none with the junctional complexes of the epithelial cells. Subjectively, convoluted nuclei seemed to have many pore complexes associated with mitochondria, and single mitochondrial profiles were associated with more than one pore complex. The contours of mitochondria appear to be physically directed by their contacts with pore complexes. Sebaceous
cell
Unlike the type II cell, there was no correlation between nuclear and cell area of the sebaceous cell (Table 1). In addition, though the area of lamellar body profiles in the type II cell were not associated with distance from the nuclear membrane, in the sebaceous cell, there was positive correlation between the size of the profiles of secretory granules and distance to the nucleus (P~<0.0003), and as profile area increased, the distance to the periphery of the cell decreased (P~<0.005). Fewer profiles of mitochondria were seen at the nuclear membrane in the sebaceous cell than in the type II cell, but greater numbers of mitochondria were present at the junctional complexes between epithelial cells (Tables 3 and 4). The volume density of the mitochondria was about 2% of the cytoplasm. The mean distance to the nucleus of the secretory granules and the mitochondria in the sebaceous cell was greater than in
Table 4
Rat typeII cell
Ferret typeII cell
Mouse sebaceous cell
0.6±0.08gm*
0.8±0.1
2.2±0.3
2.1±0.3
2.6±0.2
-
2.4±0.2
2.2±0.1
1.5±0.2
0.9±0.1
1.5±0.2
-
0.6±0.06gm
1.0±0.1
2.2±0.2
2.0±0.2
2.2±0.2
-
2.1±0.2
2.0±0.1
1.5±0.3
0.9±0.1
1.6±0.2
-
Lamellar body mean distance to:
Nuclear membrane Alveolar space Cell/cell membrane Basement membrane Mitochondrial mean distance to:
Nuclear membrane Alveolar space Cell/cell membrane Basement membrane *Mean_ SEM.
the type II cell (relating to relative sizes of the two cell types). Area of mitochondrial profiles was not associated with distance to the nucleus or the periphery of the cell.
Discussion Significant relationships between the lamellar body and interpore nuclear membrane, and the mitochondrion and nuclear pore complex have been documented using a morphometric analysis of two-dimensional images. These associations were seen using either an aldehyde or a direct osmium fixation, with instillation or immersion fixation and were apparent in at least four species, the ferret, rat, dog and mouse and in the guinea pig (unpublished observations). Profiles of mitochondria are frequently found near nuclear pore complexes in many tissues and cells such as brain, keratinocytes, cultured olfactory cells, ciliated cells of the trachea, and liver, to name a few (Miller unpublished observations). Explanations can easily be made for both types of interactions, however, the data are admittedly descriptive. Associations between mitochondria and nuclear pore complexes appear to be ubiquitous in mammalian cells. Highest numbers of these occur in cells which have little condensed chromatin, large nuclei, and have high metabolic rates (e.g. pyramidal cells in brain. Miller, unpublished observations). It is not surprising that mitochondria would be situated adjacent to nuclear pores, particularly since trafficking of ribonucleic acid and proteins into and out of the nucleus is critical for cell function, and energy requirements are proposed for some forms of trans-pore transport, such as the uptake
618
MILLERET AL.
of molecules into the nucleus during gene activation. Kessel (1988) has suggested that more complex molecules, larger than 64 kd, such as ribonucleoproteins, may require activation of the ATPase in order to gain passage through the pore complex. Conversely, mitochondria are associated less frequently with the interpore region of the nuclear membrane than the nuclear pores. In view of the fact that mitochondria are highly branched structures, the contacts of a single profile of a mitochondrion represent a far greater percent of contact in the mitochondrial population. Mitochondria are associated with the nuclear membrane (no specific reference was made to the nuclear pore complex) during stage 53 of regeneration of the hind limbs of the frog (Ramadan et al., 1987). Variations in the incidence of mitochondrialnuclear pore complexes will undoubtedly relate to such phenomena as dissociation and reassembly of the nuclear membrane, cell cycle, and metabolic activity. Harris (1978) reported that patterns in the placement of nuclear pore complexes in the nuclear membrane could be correlated with the arrangement of organelles within plant cells. Unlike the evidence provided by Kessel (1988) and our data in mammalian cells, the nuclear pores in plants appeared to be associated with only smooth vesicles. Young et al. (1985; 1991) in their three-dimensional reconstructions of the type II cell found only a small portion of the plasmalemma at the lumen of the alveolus. A non-random distribution of several surfactant-related organelles was also seen by Young et al., (1985) and Fram et al., (1985) in the type II cell. In this study we found that 44% of the plasmalemma was lumenal, and that large lamellar bodies appeared to be located near the air border (Young et al., 1991), but that an even larger number of lamellar body profiles was near the nuclear membrane. When area of lamellar body profiles and distance to the membranes were modelled, more frequent associations were seen between lamellar bodies and the nuclear membrane than lamellar body profiles and the alveolar membrane, where a known functional relationship exists. Movement of organelles through the
type II cell cytoplasm has been suggested many times, but was addressed specifically by Walker et al. (1985) who hypothesized that in the type II cell surfactant protein moves via the multivesicular bodies, 'from the alveolar space to the lamellar bodies'. It has been shown that endocytic vesicles derived from the cell membrane can fuse with the nuclear membrane (Kessel, 1988). Data obtained from the present study demonstrated that various types of multivesicular bodies, as well as lamellar bodies, were located in specific relationship to the interpore portion of the nucleus. This is perhaps a site for the exchange of surfactant related components, such as proteins, from the cisternae of the nuclear membrane (Miller et al., 1986a), The secretory granules in the sebaceous cell displayed a cytoplasmic distribution related to size, smaller granules being located closest to the center of the cell, which likely correlates with production and storage of the sebum. The absence of contacts between the sebum granules and the interpore membrane, where no interaction was predicted, is in contrast to that seen in the type II cell. In summary, detection of significant differences in the spatial organization of organelles in the cytoplasm can be achieved by modelling distance and area measurements from profiles of two-dimensional images. Data from this study demonstrate contacts between: (1) the lamellar body and the interpore region of the nuclear membrane, which may be an important site for the exchange of components important in the production of mature surfactant; and (2) the mitochondrion and the filaments of the nuclear pore complex, which may provide a source of energy for trans-nuclear pore transport. ACKNOWLEDGEMENTS The authors gratefully acknowledge those who run the animal care facilities of this department and the secretarial assistance o f N . Knapp. We are also appreciative of the comments of Drs S. Young, M. Radike, and G. Leikauf. This study was funded in part by grants ES00159, and NIEHS 04099, 1 P30 ESO6096~
REFERENCES Baxter, C.S., Andringa, A., Chalfin, K., and Miller, M.L. 1988. Comparative histomorphometricchanges in SENCAR mouse epidermis in response to multiple treatments with completeand stage-specific tumor promoting agents. Carcinogen., 10, 1855-1861. Crapo, J.D., Peters-Golden,M., Marsh-Shalin, J. and Shelbutne, J.S. 1978. Pathologicchanges in the lungs of oxygen-adapted rats. A morphometricanalysis. Lab, Invest., 39, 640-653. Fram, E.K., Kliewer, M., and Young, S.L. 1985. Three-dimensional reconstruction and quantitative analysis of the type lI cells demonstrate selectivityof lamellar body secretion. Clin. Res., 33, 464A. Hanover, J.A. 1992. The nuclear pore: at the crossroads. FASEB J., 6, 2288-2295.
Harris, N. 1978. Nuclear pore distribution and relation to adjacent cytoplasmic organelles in cotyledoncells of developing Vicia faba. Planta, 141, 121-128. Kessel, R.G. 1988. The contribution of the nuclear envelopeto eukaryotic cell complexity:architectureand functional roles. CRC Critical Rev. Anat. Cell Biol., 1, 327-433. Kliewer, M., Fram, D.K., Brody, A.R. and Young, S.L. 1985. Secretion of surfactant by rat alveolar type II cells: morphometric analysis and three-dimensionalreconstruction. Exp. Lung. Res., 9, 351-361. Massaro, G.D., Gail, D.B. and Massaro, D. 1975. Lung oxygen consumption and mitochondria of alveolar epithelial and endothelial cells. J. Appl. Physiol., 38, 588-592. Miller, M., Park, M.K. and Hanover, J.A. 1991. Nuclear pore
ORGANELLES/NUCLEAR MEMBRANE
complex: structure, function and regulation. Path. Rev., 71, 909-949. Miller, M.L., Andringa, A., Adams, W. and Radike, M. 1986a. Intracisternal protein in the type II pneumocyte of the ferret, guinea pig and mongrel dog. J. Ultrastruct. Res., 95, 131-141. Miller, M.L., Andringa, A., Evans, J.E. and Hastings, L. 1995. Microvillar cells of the olfactory epithelium: morphology and regeneration following exposure to toxic compounds. Brain Res., 669, 1-9. Miller, M.L., Andringa, A., Manson, J.M, Radike, M.J., Clark Jr L.C., Warshawsky, D. and Vinegar, A. 1985. An ultrastructural survey of desmosomal-mitochondrial complexes in the liver of several species of laboratory mammal. Ohio J. Sci., 85, 74 85. Miller, M.L., Andringa, A., Vinegar, A., Adams, W.D., Cibulas, Jr. W, and Brooks, S.M. 1986b. Morphology of tracheal and bronchial epithelium and type II cells of the peripheral lung of the guinea pig after inhalation of toluene diisocyanate vapors. ]Exp. Lung. Res., 11, 145-163. Miller, M.L., Andringa, A., and Vinegar, A. 1982. Ultrastructure and morphometry of the alveolar type II cell of the ferret. J. Ultrastruct. Res., 79, 85 91. Ramadan, A.A., Michael, M.I., Khadre, S.E.M., Aziz, F.K., and Hamed, S.S. 1987. Mitochondrial changes during dedifferentiation and redifferentiation of regenerating hind limbs in larval and metamorphic stages of Bufa reguaris Reuss. Folia. Morphol., 35, 86-95. Shimura, S., Aoki, T., Tomioka, M., Shindoh, Y., and Takishima, T. 1985. Concentrically arranged endoplasmic reticulum containing some lamellae (bar-like structure) in alveolar type II cells of rat lung. J. Ultrastruct. Res., 93, 116-128.
619
Shimura, S., Maeda, S., and Takishima, T. 1984. Giant lamellar bodies in alveolar type II cells of rats exposed to a low concentration of ozone. Resp., 46, 303-309. Shimura, S., Aoki, T., Sato, S. and Takishima, T. 1980. A cytoplasmic bar-like structure seen in type II alveolar cells of rat lungs. Tohoku. J. Exp. Med., 131, 409410. Stephens, R.J., Freeman, G., Stara, J.F. and Coffin, D.L. 1973. Cytologic changes in dog lungs induced by chronic exposure to ozone. Am. J. Path., 73, 711 726. Walker, S.R., Williams, M.C. and Benson, B. 1985. Immunohistochemical localization of the mouse surfactant apoproteins in type II cells, Clara cells and alveolar macrophages of rat lung. J. Histochem. Cytochem., 34, 1137-1148. Williams, M.C. 1987. Vesicles within vesicles: what role do multivesicular bodies play in alveolar type II cells? Am. Rev. Resp. Dis., 135, 744-746. Williams, M.C., Sueishi, K. and Benson, B.J. 1984. Localization of a surfactant apoprotein in multivesicular bodies of type II cells. Prog. Resp. Res. 18, 101-105. Young, S.L., Fram, E.K., and Craig, B.L. 1985. Three-dimensional reconstruction and quantitative analysis of rat lung type II cells: a computer-based study. Am. J. Anat., 174, 1-14. Young, S.L., Fram, E.K. and Randell, S.H. 1986. Quantitative three-dimensional reconstruction and carbohydrate cytochemistry of rat nonciliated bronchiolar (Clara) cells. Am. Rev. Resp. Dis., 133,899-907. Young, S.L., Fram, E.K., Spain, C i . , and Larson, E.W. 1991. Development of type II pneumocytes in rat lung. Am. J. Physiol. (Lung Cell Morph. Physiol.), 260, Ll13 122.
Relationships between the nuclear membrane, nuclear pore complexes, and organelles in the type II pneumocyte M.L. Miller, A. Andringa, L. Hastings
Abstract. Functional relationships among organelles of the type II cell are suggested based upon the proximity of organelles to specialized areas of the plasma- and nuclear membranes. In a two-dimensional morphometric analysis of the profiles of organelles in type II cells of the ferret and rat (and beagle dog), lamellar bodies were more likely to be located near the nuclear membrane than at the alveolar space (where exocytosis occurs). The size of lamellar body profiles was not correlated with distance from the nuclear membrane; however, large profiles were nearer the apical membrane, and smaller ones nearer the basement membrane. Profiles of highly branched mitochondria were 10 times more frequently associated with nuclear pore complexes than with the inter-pore nuclear membrane. Forty percent of all mitochondrial profiles were within 0.25 #m of the nucleus, 5% were within 0.02 pm and half of these appeared to touch the filaments of the nuclear pore complexes. The size of mitochondrial profiles was not correlated with distribution. In the ferret and rat, 8.6% and 2.5% respectively, of the nuclear pore complexes were associated with mitochondria. Sebaceous cells, from control mice, demonstrated a spatial distribution of granules which was size dependent but unrelated to the nuclear membrane. Keywords: Nuclearmembrane, mitochondria, nuclear pore complexes, lamellar bodies, spatial relations
Introduction The type II pneumocyte has been shown by numerous methods to secrete surfactant which reduces surface tension in the alveolus. The secretion granules may be composed of newly produced surfactant as well as endocytically derived components (Walker et al., 1985; Williams, 1987; Williams et al., 1984). Morphometric studies have established the mean volume and numerical densities for organelles within the type II cell in a variety of species and under various experimental conditions (Crapo et al., 1978; Fram et al., 1985; Kliewer et al., Department of Environmental Health, College of Medicine, University of Cincinnati, 3223 Eden Avenue, Cincinnati, OH 45267-0056, USA. Received 17 March 1995 Accepted 26 June 1995 Correspondence to: M. L. Miller.
1985; Massaro et al., 1975; Miller et al., 1985; 1986a,b; Shimura et al., 1984; 1985; Young et al., 1985; 1986; 1991 ). Several organelles are directly involved in surfactant production including, lamellar, composite and multivesicular bodies, and cisternal bodies (Miller et al., 1982). There has been some documentation on the spatial distribution of these structures (Young et al., 1985; 1991), but rarely have the nuclear membrane and nuclear pore complexes been examined in relation to intracellular organelles (Miller et al., 1986a; 1995). Nuclear pores are octagonal ring-like structures between 50-90 nm in diameter which punctuate the nuclear membrane at the 'crossroads' between the nucleus and cytoplasm, controlling the import and export of constituents (Hanover, 1992). In fact, most nucleo-cytoplasmic exchanges occur through the nuclear pore complex (Miller et al., 1991). The number of 613
614
M I L L E R ET AL.
nuclear pores varies in different cell types, during different phases of the cell cycle, states of activity, and even at different sites on the surface of the same nucleus (Miller et al., 1991). Circumstances controlling the number of pores and their arrangement in the nuclear membrane are not known (Hanover, 1992; Miller et al., 1991). Transcriptionally active cells have the greatest number of nuclear pores. The number of nuclear pores doubles just before S phase of the cell cycle. Miller et al., (1991) also suggest that the number of nuclear pore complexes may be determined by the 'tissue and cell-specific gene expression'. It has been suggested that cytoskeletal components may be in continuity with nuclear pore filaments (Kessel, 1988; Miller et al., 1991; Miller et al., 1995). What Hanover (1992) calls cytoplasmic anchoring is likely to be a factor in nuclear transport and it appears to require energy in the form of ATP. Mitochondria lie near the nucleus in the developmental period, as well as at some specific times in the cell cycle of oocytes (Kessel, 1988). Morphologic evidence to suggest interactions between lamellar bodies and other surfactant related organelles and the nuclear membrane has been reported (Miller et al., 1986a), but measures of frequency or distance to support these observations are lacking. In addition, a new observation, the association of mitochondria with the filaments of the nuclear pore complex, has not been reported. This study describes the size and spatial distribution of profiles of lamellar bodies and mitochondria within the type II cell, and documents the extent and nature of the contacts of these organelles with the nuclear membrane and nuclear pores complexes, and contrasts this with data obtained on sebaceous cells.
Materials and methods Seven male ferrets (400 g), three male Sprague Dawley rats (300-500 g) and 10 female SENCAR mice were quarantined for at least one week. Animal care and treatment procedures were as previously (Baxter et al., 1988; Miller et al., 1982) and conform to AAALAC standards. Rats and ferrets were euthanized with pentobarbital, chests opened, tracheas dissected and lungs fixed by instillation of fluid (20 cm of fluid pressure) for 20 rain (Miller et al., 1982). The fixative contained 2% paraformaldehyde, 2.5% glutaraldehyde, 0.10% dextrose, and 0.002% calcium chloride in 0.032 M Millonig's buffer (pH 7.4, 350 mosm). The trachea was tied, the lungs removed and immersed in fresh fixative for at least 24 h and cut into 1 mm 3 cubes. Some tissue was fixed directly in osmium tetroxide in Palade's fluid (Miller et al., 1982). Blocks of tissue from middleright mid-clavicular parenchymal lung from each animal were taken. Mice were sacrificed by cervical dislocation, their interscapular skin fixed by immersion in the same
fixative. The skin was oriented in the blocks to provide sections perpendicular to the epidermis. Epon 812 embedded blocks of lung from eight female beagle dogs, at least 22 months old, were kindly supplied by Dr R. J. Stephens. The methods of tissue preparation have been published elsewhere (Stephens et al., 1973), but include both direct osmium fixation, and fixation with buffered glutaraldehyde and postfixation in osmium tetroxide. All tissues were then rinsed in buffer, postfixed for 2 h in 1% osmium tetroxide, dehydrated in a series of ascending ethanol concentrations, passed through propylene oxide, and embedded in Spurr resin. Blocks were thin sectioned and stained with uranyl acetate and lead citrate for transmission electron microscopy (60 kV). Ten to 15 type II cells from each animal were photographed at 5000 x and enlarged to 20000x (Fig. 1). Magnification at the microscope was calibrated with a Ladd Research Industries Carbon Grating. Type II cells containing a profile of nucleus, lamellar body, mitochondrion, and plasmalemma bordering the alveolar space, basement membrane and an adjacent epithelial cell were selected consecutively. It was determined that about 8% of the type II cells identified were excluded by this selection procedure, i.e. those without nuclear profiles, resulting in only a small bias (see Results) in the sampling procedure. Since randomness was not assumed for the orientation of the type II cell in the alveolus or the organelles within the cells, random orientation of specimens for sectioning was required. No adjustment was made for shrinkage. Data were obtained by measuring the shortest distance between an organelle and the plasma membrane (apical at the lumenal interface; basal - adjacent to basement membrane; lateral - adjacent to another epithelial cell; and nuclear membrane - either 'interpore' or at the nuclear pore complex). This method of data acquisition was modified from that described for a threedimensional analysis of type II cells (Young et al., 1985; 1991) and Clara cells (Young et al., 1986) with a computer-based program which calculated the shortest distance between lamellar bodies and the apical plasmalemma and the Golgi. Although the use of twodimensional techniques underestimates the actual distances between organelles, significant relationships were found. The perimeter and area of each type II or sebaceous cell and each organellar profile were measured using SigmaScan (Jandel Scientific). The prevalence of mitochondria at the nuclear pore was determined by counting the number of profiles of mitochondria in contact with the filaments of the nuclear pore complex, relative to the total number of nuclear pores. A point on the nucleus 4.0 gm from the pore was used to determine the contacts between mitochondria and control-equivalent for the length of the nuclear pores on the nuclear membrane. In addition, all profiles
ORGANELLES/NUCLEAR MEMBRANE 615
Fig. 1 (a) Type II cells contain lamellar bodies which are frequently situated within 0.02 gin of the nuclear membrane (arrowheads). The points of very close contact are primarily at the interpore region of the outer nuclear membrane. Dog, aldehyde fixation, x 21000. (b) Mitochondria are frequently associated with filaments which are a part of the nuclear pore complex (arrowhead). Cristae within the mitochondrion are parallel to the filaments of the pore. Ferret, aldehyde fixation, × 39 000. (c) Mitochondria are visibly drawn towards the pore complex where contact with the filament is visualized (arrowhead). Ferret, aldehyde fixation, x 44 000. appearing adjacent to the nuclear m e m b r a n e were counted (200-300 each for ferret and rat, 1400 for dogs). N o migration o f sebaceous cell granules for exocytosis occurs in the sebaceous cell since the cell itself becomes the sebum. In addition, the entire plasmalemma o f cells lying above the basement m e m b r a n e contacts other sebaceous cells, and was used as a 'negative control' for the effect o f specializations o f the plasmalemma on spatial locations o f the organelles. P h o t o g r a p h i c m o n tages o f cell profiles contained a nucleus (not pyknotic), and at least one granule and mitochondrion. The data were analyzed separately for each species. Statistical analyses were m a d e using the General Linear
Model (Statistical Analysis System ( P C SAS)), and Scheffe and Tukey options were used for post hoc comparisons. D a t a were considered significant when P~<0.05.
Results Type II cell The ratio o f profiles o f nuclear and cytoplasmic area was nearly the same in rats and ferrets. As the nuclear area increased so did the cytoplasmic area (ferret and rat, P<~O.O05, P~
616
MILLERET AL.
n u m b e r o f profiles o f l a m e l l a r bodies a n d m i t o c h o n d r i a a n d the p e r c e n t o f the c y t o p l a s m which l a m e l l a r bodies c o m p r i s e d was similar in the ferret a n d r a t ( T a b l e 1). These values are similar to previous r e p o r t s ( M a s s a r o et al., 1975; M i l l e r et al., 1982; 1986; Y o u n g et al., 1985; 1991). Preselected type II cells (containing profiles o f l a m e l l a r b o d y , nucleus, m i t o c h o n d r i o n , a n d m e m b r a n e s at the lumen, b a s e m e n t m e m b r a n e a n d an epithelial cell) ( T a b l e 2) were n o t statistically different f r o m cells selected on a r a n d o m basis. T h e a m o u n t o f p l a s m a l e m m a specialized at each o f these sites was n e a r l y equal, yet, m o r e t h a n t w o - t h i r d s o f the l a m e l l a r b o d i e s were l o c a t e d within 0.25 ~tm o f the nuclear m e m b r a n e ( T a b l e 3). H a l f o f all the l a m e l l a r b o d i e s were l o c a t e d close to the i n t e r p o r e a r e a o f the n u c l e a r m e m b r a n e , a n d a b o u t 12% a n d 8% a p p e a r e d to coalesce with it in the ferret a n d the r a t respectively. T h e highest p e r c e n t a g e o f l a m e l l a r b o d y c o n t a c t with the n u c l e a r m e m b r a n e was seen in the d o g (13.9%). A s the a r e a o f profiles o f l a m e l l a r bodies increased, distance to the m e m b r a n e at the a l v e o l a r space decreased
Rat type II cell
Ferret type II cell
Mouse sebaceous cell
n = 20 35.9_+2.9 gm2. 27.1 -+ 1.3 gm n = 20 10.2-+1.1 pan2 14.5 ± 1.5 [ira 25%
n = 25 43.0_+2.7 27.6 _+1.0 n = 25 11.7-+1.1 15.6 _+0.9 27%
n = 14 181.9_+22.3 53.3 _+3.6 n = 14 34.6_+4.4 20.8 _+1.6 19%
n = 208
n = 266
n = 261
0.52_+0,03 gm2 15%
0.57_+0.05 15%
0.74_+0.08 13%
n = 14l 0.28_+0.02 Jam2 5.4%
n = 222 0.21 _+0.02 4.3%
n = 148 0.21 _+0.03 2%
*mean_+SEM; +does not include microvilli. Table 2
Contributions to the plasmalemma Alveolar space Cell/cell membrane Basal plasmalemma (Nuclear membrane)*
Rat type II cell
Ferret type II cell
Mouse sebaceous cell
Lamellar bodies < 0 . 0 2 ttm from:
Nuclear membrane Alveolar space Cell/cell membrane Basement membrane
8%
7%
1
2
0
0
2
t
45%
38%
10
13
4
3
26
14
0.4%
0.8
Lamellar bodies < 0.25 ~m from:
Nuclear membrane Alveolar space Cell/cell membrane Basement membrane
6%
8
Mitochondria < 0 . 0 2 ~m from:
Table 1
Cell area Cell perimeter Nucleus Area Perimeter + Volume density Secretory granules Area Volume density Mitochondria Area Volume density
Table 3
Rat type II cell
Ferret type II cell
Mouse sebaceous cell
44%
52%
-
8 48
9 39
-
(54)
(52)
100 (39)
*The total plasma membrane consisted of membrane at the alveolar space, the cell/cell contacts and the basement membrane; the nuclear membrane was equivalent to 54% of the total distance of the plasmalemma.
Nuclear membrane Alveolar space Cell/cell membrane Basement membrane
5%
4%
0.07%
0
0.5
0
0
3
0.5
40%
27%
9
9
-
6
8
20
33
19
1.4 -
Mitochondria < 0.25 I~m from:
Nuclear membrane Alveolar space Cell/cell membrane Basement membrane
10%
-
in b o t h the ferret a n d r a t (P ~<0.0003, P~<0.0015 respectively). In ferrets, there was no c o r r e l a t i o n between a r e a o f l a m e l l a r bodies a n d distance to the b a s e m e n t m e m brane. In rats, however, larger l a m e l l a r bodies were significantly f a r t h e r f r o m the b a s e m e n t m e m b r a n e (P~<0.004) a n d smaller l a m e l l a r bodies were l o c a t e d basally. W h e t h e r these smaller l a m e l l a r bodies were c o m p o s i t e was n o t determined. A r e a o f the l a m e l l a r b o d y profiles was n o t related to the distance to the nuclear m e m b r a n e in either ferrets or rats t h o u g h there was a t r e n d for larger l a m e l l a r bodies to be close to the nuclear m e m b r a n e in the ferret. D a r k multivesicular bodies were l o c a t e d n e a r nuclei (P~<0.004) while light multivesicular bodies showed no such distribution. D i s t r i b u t i o n o f profiles o f multivesicular bodies within the cell d i d n o t a p p e a r to be related to o r g a n e l l a r area, t h o u g h s a m p l e size was small.
ORGANELLES/NUCLEARMEMBRANE 617
Mitochondrial profiles were slightly smaller in the type II cells of the ferret than the rat, though the number of profiles as well as the volume density were similar for these highly branched organelles (Table 1). About half as many were seen in the dog. Profiles of mitochondria were frequently found within 0.25 gm of the nuclear membrane in rat, ferret, mouse (Table 3) and in dog. Considering the branching nature of mitochondria, this would include the majority if not all mitochondria with at least some segment at the nuclear membrane. At least 5.7% and 3.6% of the profiles in the rat and the ferret respectively are within 0.02 gm of the nucleus. On occasion, a single profile of a mitochondrion contacted as many as three adjacent nuclear pore complexes. In addition, 8.6%_+1.6, and 2.5%+0.009 (S.E.M.) and 14.5%_+4.0 of nuclear pores appeared to be in direct contact with mitochondria in the ferret, rat and beagle respectively. At points on the nuclear membrane 4 gm from each nuclear pore complex, only about 2% of the mitochondria were in contact with the nuclear membrane. While the length (in microns) occupied by nuclear pore complexes in the type II cell is a small fraction of the whole nuclear membrane, about half of all mitochondrial contact with the nuclear membrane occurred there. Larger mitochondrial profiles appeared to be polarized towards the nucleus in the ferret, while larger profiles were closer to the junctional complexes between epithelial cells in the rat. In contrast, less than 1% of the profiles of lamellar bodies were associated with the nuclear pores and none with the junctional complexes of the epithelial cells. Subjectively, convoluted nuclei seemed to have many pore complexes associated with mitochondria, and single mitochondrial profiles were associated with more than one pore complex. The contours of mitochondria appear to be physically directed by their contacts with pore complexes. Sebaceous
cell
Unlike the type II cell, there was no correlation between nuclear and cell area of the sebaceous cell (Table 1). In addition, though the area of lamellar body profiles in the type II cell were not associated with distance from the nuclear membrane, in the sebaceous cell, there was positive correlation between the size of the profiles of secretory granules and distance to the nucleus (P~<0.0003), and as profile area increased, the distance to the periphery of the cell decreased (P~<0.005). Fewer profiles of mitochondria were seen at the nuclear membrane in the sebaceous cell than in the type II cell, but greater numbers of mitochondria were present at the junctional complexes between epithelial cells (Tables 3 and 4). The volume density of the mitochondria was about 2% of the cytoplasm. The mean distance to the nucleus of the secretory granules and the mitochondria in the sebaceous cell was greater than in
Table 4
Rat typeII cell
Ferret typeII cell
Mouse sebaceous cell
0.6±0.08gm*
0.8±0.1
2.2±0.3
2.1±0.3
2.6±0.2
-
2.4±0.2
2.2±0.1
1.5±0.2
0.9±0.1
1.5±0.2
-
0.6±0.06gm
1.0±0.1
2.2±0.2
2.0±0.2
2.2±0.2
-
2.1±0.2
2.0±0.1
1.5±0.3
0.9±0.1
1.6±0.2
-
Lamellar body mean distance to:
Nuclear membrane Alveolar space Cell/cell membrane Basement membrane Mitochondrial mean distance to:
Nuclear membrane Alveolar space Cell/cell membrane Basement membrane *Mean_ SEM.
the type II cell (relating to relative sizes of the two cell types). Area of mitochondrial profiles was not associated with distance to the nucleus or the periphery of the cell.
Discussion Significant relationships between the lamellar body and interpore nuclear membrane, and the mitochondrion and nuclear pore complex have been documented using a morphometric analysis of two-dimensional images. These associations were seen using either an aldehyde or a direct osmium fixation, with instillation or immersion fixation and were apparent in at least four species, the ferret, rat, dog and mouse and in the guinea pig (unpublished observations). Profiles of mitochondria are frequently found near nuclear pore complexes in many tissues and cells such as brain, keratinocytes, cultured olfactory cells, ciliated cells of the trachea, and liver, to name a few (Miller unpublished observations). Explanations can easily be made for both types of interactions, however, the data are admittedly descriptive. Associations between mitochondria and nuclear pore complexes appear to be ubiquitous in mammalian cells. Highest numbers of these occur in cells which have little condensed chromatin, large nuclei, and have high metabolic rates (e.g. pyramidal cells in brain. Miller, unpublished observations). It is not surprising that mitochondria would be situated adjacent to nuclear pores, particularly since trafficking of ribonucleic acid and proteins into and out of the nucleus is critical for cell function, and energy requirements are proposed for some forms of trans-pore transport, such as the uptake
618
MILLERET AL.
of molecules into the nucleus during gene activation. Kessel (1988) has suggested that more complex molecules, larger than 64 kd, such as ribonucleoproteins, may require activation of the ATPase in order to gain passage through the pore complex. Conversely, mitochondria are associated less frequently with the interpore region of the nuclear membrane than the nuclear pores. In view of the fact that mitochondria are highly branched structures, the contacts of a single profile of a mitochondrion represent a far greater percent of contact in the mitochondrial population. Mitochondria are associated with the nuclear membrane (no specific reference was made to the nuclear pore complex) during stage 53 of regeneration of the hind limbs of the frog (Ramadan et al., 1987). Variations in the incidence of mitochondrialnuclear pore complexes will undoubtedly relate to such phenomena as dissociation and reassembly of the nuclear membrane, cell cycle, and metabolic activity. Harris (1978) reported that patterns in the placement of nuclear pore complexes in the nuclear membrane could be correlated with the arrangement of organelles within plant cells. Unlike the evidence provided by Kessel (1988) and our data in mammalian cells, the nuclear pores in plants appeared to be associated with only smooth vesicles. Young et al. (1985; 1991) in their three-dimensional reconstructions of the type II cell found only a small portion of the plasmalemma at the lumen of the alveolus. A non-random distribution of several surfactant-related organelles was also seen by Young et al., (1985) and Fram et al., (1985) in the type II cell. In this study we found that 44% of the plasmalemma was lumenal, and that large lamellar bodies appeared to be located near the air border (Young et al., 1991), but that an even larger number of lamellar body profiles was near the nuclear membrane. When area of lamellar body profiles and distance to the membranes were modelled, more frequent associations were seen between lamellar bodies and the nuclear membrane than lamellar body profiles and the alveolar membrane, where a known functional relationship exists. Movement of organelles through the
type II cell cytoplasm has been suggested many times, but was addressed specifically by Walker et al. (1985) who hypothesized that in the type II cell surfactant protein moves via the multivesicular bodies, 'from the alveolar space to the lamellar bodies'. It has been shown that endocytic vesicles derived from the cell membrane can fuse with the nuclear membrane (Kessel, 1988). Data obtained from the present study demonstrated that various types of multivesicular bodies, as well as lamellar bodies, were located in specific relationship to the interpore portion of the nucleus. This is perhaps a site for the exchange of surfactant related components, such as proteins, from the cisternae of the nuclear membrane (Miller et al., 1986a), The secretory granules in the sebaceous cell displayed a cytoplasmic distribution related to size, smaller granules being located closest to the center of the cell, which likely correlates with production and storage of the sebum. The absence of contacts between the sebum granules and the interpore membrane, where no interaction was predicted, is in contrast to that seen in the type II cell. In summary, detection of significant differences in the spatial organization of organelles in the cytoplasm can be achieved by modelling distance and area measurements from profiles of two-dimensional images. Data from this study demonstrate contacts between: (1) the lamellar body and the interpore region of the nuclear membrane, which may be an important site for the exchange of components important in the production of mature surfactant; and (2) the mitochondrion and the filaments of the nuclear pore complex, which may provide a source of energy for trans-nuclear pore transport. ACKNOWLEDGEMENTS The authors gratefully acknowledge those who run the animal care facilities of this department and the secretarial assistance o f N . Knapp. We are also appreciative of the comments of Drs S. Young, M. Radike, and G. Leikauf. This study was funded in part by grants ES00159, and NIEHS 04099, 1 P30 ESO6096~
REFERENCES Baxter, C.S., Andringa, A., Chalfin, K., and Miller, M.L. 1988. Comparative histomorphometricchanges in SENCAR mouse epidermis in response to multiple treatments with completeand stage-specific tumor promoting agents. Carcinogen., 10, 1855-1861. Crapo, J.D., Peters-Golden,M., Marsh-Shalin, J. and Shelbutne, J.S. 1978. Pathologicchanges in the lungs of oxygen-adapted rats. A morphometricanalysis. Lab, Invest., 39, 640-653. Fram, E.K., Kliewer, M., and Young, S.L. 1985. Three-dimensional reconstruction and quantitative analysis of the type lI cells demonstrate selectivityof lamellar body secretion. Clin. Res., 33, 464A. Hanover, J.A. 1992. The nuclear pore: at the crossroads. FASEB J., 6, 2288-2295.
Harris, N. 1978. Nuclear pore distribution and relation to adjacent cytoplasmic organelles in cotyledoncells of developing Vicia faba. Planta, 141, 121-128. Kessel, R.G. 1988. The contribution of the nuclear envelopeto eukaryotic cell complexity:architectureand functional roles. CRC Critical Rev. Anat. Cell Biol., 1, 327-433. Kliewer, M., Fram, D.K., Brody, A.R. and Young, S.L. 1985. Secretion of surfactant by rat alveolar type II cells: morphometric analysis and three-dimensionalreconstruction. Exp. Lung. Res., 9, 351-361. Massaro, G.D., Gail, D.B. and Massaro, D. 1975. Lung oxygen consumption and mitochondria of alveolar epithelial and endothelial cells. J. Appl. Physiol., 38, 588-592. Miller, M., Park, M.K. and Hanover, J.A. 1991. Nuclear pore
ORGANELLES/NUCLEAR MEMBRANE
complex: structure, function and regulation. Path. Rev., 71, 909-949. Miller, M.L., Andringa, A., Adams, W. and Radike, M. 1986a. Intracisternal protein in the type II pneumocyte of the ferret, guinea pig and mongrel dog. J. Ultrastruct. Res., 95, 131-141. Miller, M.L., Andringa, A., Evans, J.E. and Hastings, L. 1995. Microvillar cells of the olfactory epithelium: morphology and regeneration following exposure to toxic compounds. Brain Res., 669, 1-9. Miller, M.L., Andringa, A., Manson, J.M, Radike, M.J., Clark Jr L.C., Warshawsky, D. and Vinegar, A. 1985. An ultrastructural survey of desmosomal-mitochondrial complexes in the liver of several species of laboratory mammal. Ohio J. Sci., 85, 74 85. Miller, M.L., Andringa, A., Vinegar, A., Adams, W.D., Cibulas, Jr. W, and Brooks, S.M. 1986b. Morphology of tracheal and bronchial epithelium and type II cells of the peripheral lung of the guinea pig after inhalation of toluene diisocyanate vapors. ]Exp. Lung. Res., 11, 145-163. Miller, M.L., Andringa, A., and Vinegar, A. 1982. Ultrastructure and morphometry of the alveolar type II cell of the ferret. J. Ultrastruct. Res., 79, 85 91. Ramadan, A.A., Michael, M.I., Khadre, S.E.M., Aziz, F.K., and Hamed, S.S. 1987. Mitochondrial changes during dedifferentiation and redifferentiation of regenerating hind limbs in larval and metamorphic stages of Bufa reguaris Reuss. Folia. Morphol., 35, 86-95. Shimura, S., Aoki, T., Tomioka, M., Shindoh, Y., and Takishima, T. 1985. Concentrically arranged endoplasmic reticulum containing some lamellae (bar-like structure) in alveolar type II cells of rat lung. J. Ultrastruct. Res., 93, 116-128.
619
Shimura, S., Maeda, S., and Takishima, T. 1984. Giant lamellar bodies in alveolar type II cells of rats exposed to a low concentration of ozone. Resp., 46, 303-309. Shimura, S., Aoki, T., Sato, S. and Takishima, T. 1980. A cytoplasmic bar-like structure seen in type II alveolar cells of rat lungs. Tohoku. J. Exp. Med., 131, 409410. Stephens, R.J., Freeman, G., Stara, J.F. and Coffin, D.L. 1973. Cytologic changes in dog lungs induced by chronic exposure to ozone. Am. J. Path., 73, 711 726. Walker, S.R., Williams, M.C. and Benson, B. 1985. Immunohistochemical localization of the mouse surfactant apoproteins in type II cells, Clara cells and alveolar macrophages of rat lung. J. Histochem. Cytochem., 34, 1137-1148. Williams, M.C. 1987. Vesicles within vesicles: what role do multivesicular bodies play in alveolar type II cells? Am. Rev. Resp. Dis., 135, 744-746. Williams, M.C., Sueishi, K. and Benson, B.J. 1984. Localization of a surfactant apoprotein in multivesicular bodies of type II cells. Prog. Resp. Res. 18, 101-105. Young, S.L., Fram, E.K., and Craig, B.L. 1985. Three-dimensional reconstruction and quantitative analysis of rat lung type II cells: a computer-based study. Am. J. Anat., 174, 1-14. Young, S.L., Fram, E.K. and Randell, S.H. 1986. Quantitative three-dimensional reconstruction and carbohydrate cytochemistry of rat nonciliated bronchiolar (Clara) cells. Am. Rev. Resp. Dis., 133,899-907. Young, S.L., Fram, E.K., Spain, C i . , and Larson, E.W. 1991. Development of type II pneumocytes in rat lung. Am. J. Physiol. (Lung Cell Morph. Physiol.), 260, Ll13 122.