centrioles in the mammary epithelium of the rat

J. Cell Sci. 72, 185-193 (1984)
185
Printed in Great Britain © The Company of Biologists 1984
CENTRIOLES IN THE MAMMARY EPITHELIUM OF
THE RAT
DANIEL P. DYLEWSKI AND THOMAS W. KEENAN
Department of Biochemistry and Nutrition, Virginia Polytechnic Institute and State
University, Blacksburg, VA 24061, U.SA.
SUMMARY
Through serial thin-section analysis of rat mammary epithelial cells, the number of centrioles per
cell and their intracellular location were determined. In all developmental stages (e.g. virginal,
pregnant, lactation, involution), each epithelial cell contained a single centriole that was located in
the apical region. Centrioles were 200-220 (x = 210) nm in transverse section, and exhibited the
typical 9 x 3 'pinwheel' configuration of microtubules. In longitudinal section, centrioles were
330—380 (x — 360) nm in length. Each centriole was surrounded by a homogeneous pericentriolar
matrix. During mitosis in pregnant animals, centrioles were paired at the nuclear poles and oriented
at right angles (90°) to each other. At the completion of mitosis a single diplosome (pair of
centrioles) was associated with each interphase nucleus. Because all postmitotic cells contained only
a single centriole, it was assumed that one of the two diplosomal centrioles had disintegrated. There
appeared to be a correlation between centriole location and cell polarity. When centrioles were
located near the apical plasma membrane, epithelial cells exhibited polarity. However, when
centrioles were associated with the nuclear poles during mitosis, epithelial cells were typically
apolar. These observations suggest that centrioles may function as determinants in cell polarity.
INTRODUCTION
Centrioles are ubiquitous organdies found in most living cells with the exception
of angiosperms, higher gymnosperms, higher fungi and several other groups of organisms (Wheatley, 1982). Since their discovery by van Beneden (1876), microscopic
investigations have led to a thorough understanding of centriole structure. Centriole
function, on the other hand, has remained poorly understood and is the subject of
continuing controversies (for review, see Wheatley, 1982). The intracellular position
of centrioles has led some investigators to believe that they are the primary determinants of cell polarity (Mazia, 1978).
Bargmann & Knoop (1959) were the first to describe the fine structure of the
mammary epithelial cell. Since their work, the morphology, morphometry and intracellular distribution of all organelles and cytoplasmic structures have been determined
with the exception of centrioles (for review, see Hollmann, 1974; Keenan, Morr6 &
Huang, 1974). Centrioles of mammary epithelial cells have never been described and
their intracellular location during cell development is unknown. Ultrastructural investigations of centrioles are needed, however, because they: (1) may lead to a better
understanding of structure-function relationships in the mammary epithelial cell;
Key words: centriole, mammary epithelium, cell polarity.
7
CEL72
186
D. P. Dylewski and T. W. Keenan
and (2) may help to resolve the possible role of centrioles as determinants in the
establishment of cell polarity.
The purpose of this study was to describe the ultrastructural morphology of
centrioles and determine their number and intracellular location in rat mammary
epithelial cells during development and differentiation.
MATERIALS
AND METHODS
Twenty-one primiparous female Sprague-Dawley rats between 3 and 4 months of age were killed
by cervical dislocation. Tissues samples were taken from 12-week-old virgin animals, from animals
at 1, 7, 14 and 21 days pregnant, at day 7 of lactation, and at day 4 of involution. Three animals were
used for each stage. Portions of mammary tissue were dissected from each animal and placed
immediately into large volumes of 0-5 % glutaraldehyde in 0-1 M-cacodylate buffer (pH 7) at room
temperature and cut into pieces of 1 mm . Specimens were then fixed by the method of Warchol,
Herbert & Rennels (1974), which was developed for preservation of microtubules. Tissue samples
were immersed in 2-5 % glutaraldehyde in 0-1 M-cacodylate buffer (pH 7) at room temperature for
1 h, transferred to a cold room at4°C for 1 h, then placed on crushed ice for lOmin. Specimens were
postfixed in solutions (2:1, 1:2, v/v) of 2-5 % glutaraldehyde in 0-1 M-cacodylate buffer, and 1 %
osmium tetroxide in 0-1 M-cacodylate buffer (pH 7) for 30min on crushed ice. Finally, tissue
samples were immersed in 1% osmium tetroxide in 0-lM-buffer for 2h on crushed ice, rinsed
briefly in cold buffer, dehydrated in a graded series of ethanol, soaked in acetone, and embedded
in a mixture of Epon-Araldite (Poolswat, 1973). Serial thin sections were collected on Formvarcoated copper slot grids, and stained with uranyl acetate (Watson, 1958) and lead citrate (Venable
& Coggelshell, 1965). Sections were examined in a Philips EM300 electron microscope operated at
60 kV. All linear dimensions were based on measurements of at least 25 examples.
RESULTS AND DISCUSSION
During the development of the rat mammary epithelium, centrosome morphology
remained unchanged (Figs 1-5,7-10). Each centrosome was composed of a cylindrical
centriole and surrounding percentriolar matrix (Figs 2 - 5 , 8—10). Centrioles in transverse section were 200-220 (x = 210) nm in diameter and exhibited the typical 9 x 3
'pinwheel' configuration of microtubules (Figs 4, 8). In longitudinal section,
centrioles were 330-380 (x = 360) nm in length (Figs 2, 3, 9, 10). The pericentriolar
matrix was homogeneous and electron-dense (Figs 3, 5, 8—10).
Through serial thin-section analysis, it was determined that 32 of 33 non-mitotic
epithelial cells contained a single centriole, which was located in the apical region of
Figs 1 to 4. Centrioles in virgin (Figs 1, 2) and 14-day pregnant (Figs 3, 4) animals.
Fig. 1. Survey micrograph of epithelial cell showing centriole (arrow) in longitudinal
section. Note that the centriole is located in the apical region of the cell; n, nucleus.
X10000; bar, 2-0^m.
Fig. 2. High magnification of centriole shown in Fig. 1. The centriole (arrow) is
embedded in a homogeneous pericentriolar matrix. X41 000; bar, 0 - 25^m.
Fig. 3. Longitudinal section of centriole (arrow) near apical plasma membrane
(arrowheads). Note that the long axis of the centriole is almost parallel to the plasma
membrane; alveolar lumen (/), and cytoplasmic lipid droplet (eld). X47 000; bar, 0-5 /.tm.
Fig. 4. Centriole (arrow) in transverse section showing typical 9 x 3 pinwheel configuration of microtubules. The centriole is located near the apical plasma membrane
(arrowheads); alveolar lumen (/). X92000; bar, 0-1 fan.
Centrioles in mammary gland
Figs 1-4
188
D. P. Dylewski and T. W. Keenan
the cell (Figs 1-4, 7-10). In one cell two centrioles were located near the apical plasma
membrane (not shown in figures). The centrioles in cells from virgin rats were
positioned equidistant between the nucleus and the apical plasma membrane (Figs 1,
2). In all other developmental stages, however, centrioles were located within 520 nm
of the apical plasma membrane (Figs 3, 4, 7—10). The orientation of the centriole to
the apical plasma membrane was variable, except during pregnancy when the long axis
of the centriole was nearly parallel to the surface of the membrane (Figs 3, 4, 7, 8).
During the first two weeks of pregnancy, mitotic figures were observed frequently
within epithelial cells (Figs 5,6). The mitotic spindle was composed of a diplosome
at each nuclear pole, and numerous chromosomal microtubules (Fig. 5). Each
diplosome consisted of two centrioles that were oriented at right angles (90°) to each
other (Fig. 5). At the completion of mitosis a single diplosome was associated with
each interphase nucleus (not shown in figures). Because most postmitotic cells (count
33) contained only a single centriole it was assumed that one of the two diplosomal
centrioles had disintegrated. The postmitotic disintegration of centrioles is a welldocumented phenomenon known to occur in both plant and animal cells (for review,
see Wheatley, 1982).
In the premitotic cells of virgin animals, centrioles were located in the apical region
and not at the nuclear poles (Figs 1,2). During mitosis in pregnant animals paired
centrioles were positioned at each spindle pole in close association with numerous
spindle microtubules (Fig. 5). In postmitotic cells of pregnant, lactating and involuting animals, centrioles were positioned again in the apical region and not at the nuclear
poles (Figs 7-10). This intracellular repositioning of centrioles during epithelial cell
development appears to be triggered by the process of nuclear division, and could be
accomplished by migration or disintegration and reassembly of centrioles. Two lines
of evidence support the possibility of centriole migration. First, centrioles were observed along the presumed migratory pathway between apical and nuclear positions.
Second, centrioles in stages of disintegration and reassembly were never observed.
Although centriole migration appeared to be a naturally occurring phenomenon of
epithelial cell development, the mechanisms for control and coordination by the cell
and the role of microtubules in this process remain unknown.
The migration of centrioles has been documented during the process of differentiation in many plant and animal cells (for review, see Wheatley, 1982). Perhaps the best
example is in the olfactory sense organs of mammals (Heist & Mulvaney, 1968;
Mulvaney & Heist, 1971), in which nuclear-associated centrioles of the columnar
epithelium migrate apically for a distance of five times that of the diameter of the rrlain
Figs 5, 6. Mitotic spindle apparatus of 7-day pregnant animals.
Fig. 5. Polar region of mitotic figure showing paired centrioles (arrows) arranged at
right angles (90°) to each other; microtubules (arrowheads), and lateral plasma membranes (pm). X41000; bar, 0-5/xm.
Fig. 6. Survey micrograph of epithelial cell showing metaphase chromatin plate (large
arrows) in oblique section. Note that the cell does not exhibit polarity. Secretory vesicles
(small arrows) are distributed throughout the apical (a), medial (m) and basal (b) regions
of the cell, and mitochrondria (mt) show an enrichment at the cell periphery. X9000; bar,
2-0/an.
Centrioles in mammary gland
0
Figs 5 and 6
189
D. P. Dylewski and T. W. Keenan
8
Figs 7 and 8
Centrioles in mammary gland
mlg
y-.
/
•
rv -...
Figs 9, 10. Centrioles in mammary epithelial cells of 7-day lactating (Fig. 9), and 4-day
involuting (Fig. 10) animals.
Fig. 9. Centriole (arrow) in longitudinal section and pericentriolar matrix: apical plasma membrane (arrowheads), portion of milk lipid globule (mlg), and casein micelles (me)
in alveolar lumen (/). X50000; bar, 0 5 ^m.
Fig. 10. Longitudinal section of centriole (arrow) located near apical plasma membrane. Note that the alveolar lumen (/) contains a finely fibrillar and granular material.
X41000; bar, 0-
cell body, and eventually form sensory cilia. In mammary epithelial cells centrioles
migrate half the length of the cell body (x = 8-0/im). Examples of the intercellular
migration of centrioles are not uncommon. This capacity was demonstrated in the
nurse cells of the oocyte in Drosophila, in which centrioles passed from the former to
Figs 7, 8. Centrioles in epithelial cells of 21-day pregnant animals.
Fig. 7. Survey micrograph of epithelial cell showing polar arrangement of organelles
and secretory vesicles; basal region (b), rough endoplasmic reticulum (er); medial region
(m), Golgi apparatus (G), nucleus (n); apical region (a), secretory vesicles (small arrows),
and centriole (large arrow). Mitochondria and cytoplasmic lipid droplets (eld) are
distributed evenly throughout the cell. Note that the centriole (large arrow) is located near
the apical plasma membrane (arrowheads); alveolar lumen (/). X8000; bar, 2-0/im.
Fig. 8. Centriole (arrow) in transverse section near apical plasma membrane
(arrowheads); alveolar lumen (/). X41000; bar, 0-5/mi.
192
D. P. Dylewski and T. W. Keenan
the latter during egg development (Mahawald & Strassheim, 1970; Mahowald, Caulton, Edwards & Floyd, 1979).
The determination of polarity is of enormous importance in all cells, particularly
epithelial cells, in which the movement of secretory products is unidirectional.
Although numerous theories have been proposed concerning the role of centrioles in
determining cell polarity (Bornens, 1978; Albrecht-Buehler, 1977, 1979), there is no
conclusive evidence to support such hypotheses. From our observations we can state
with certainty that mammary epithelial cells exhibit polarity in all developmental
stages when the centriole is located in the apical region. The polarized arrangement of
epithelial cell organelles and secretory products wasfirstdetectable on day 6 or day 7 of
pregnancy (total 22-day gestation period) and became more obvious in later developmental stages when cellular organelles increased in size and complexity. Each polarized
cell could be divided into three regions (e.g. apical, medial and basal) based on the
composition of cytoplasmic structures (Fig. 7). Rough endoplasmic reticulum was the
predominant organelle in the basal region of the cell (Fig. 7). The nucleus and Golgi
apparatus were usually present in the medial region (Fig. 7). Secretory vesicles, a single
centriole and frequently a portion of the Golgi apparatus were found in the apical region
(Fig. 7). Mitochondria and cytoplasmic lipid droplets were distributed throughout the
cell (Fig. 7). Epithelial cell polarity was most obvious during lactation when organelle
development was complete, and secretion of milk serum and lipid was at a maximum.
During mitosis epithelial cells were typically apolar with respect to the intracellular
distribution of organelles and secretory products (Fig. 6). In cells containing mitotic
figures condensed chromatin was found in the central region and all organelles and
cytoplasmic structure were observed on the periphery of the cell. The presence of
secretory vesicles in the basal region of the cell was the clearest indication that the cell
lacked polarity (Fig. 6).
In conclusion, apically positioned centrioles were found in polarized cells, and
nuclear-associated centrioles were found in apolar cells. Such indirect evidence
might lead one to speculate that the intracellular position of the centriole is the
primary determinant in cell polarity. However, other interpretations of centriole
function and cell polarity are possible. Clearly, the centriole will remain an enigma
and its function uncertain until they are studied in greater detail at the cellular and
molecular levels.
This research was supported by grants GM31244 from the National Institute of General Medical
Science and PCM82-41913 from the National Science Foundation.
REFERENCES
ALBRECHT-BUEHLER, G. (1977). Phagokinetic tracks of 3T3 cells: parallels between the orientation
of track segments and of cellular structures which contain actin or tubulin. Cell 12, 333-339.
ALBRECHT-BUEHLER, G. (1979). The orientation of centrioles in migrating 3T3 cells. Expl Cell Res.
120, 111-118.
BARGMANN, W. & KNOOP, A. (1959). Uber die Morphologie der Milchsekretion. Licht-und
elecktronenmikroskopische Studien an der Mitchdriise der Ratte. Z. Zcllforsch. mikrosk. Anat.
49, 344-388.
Centrioles
in mammary
gland
193
BORNENS, M. (1978). Sur le r61e et l'origine du centriole. C. hebd. Seanc. Acad. Sci., Paris 287,
1417-1420.
HEIST, H. E. & MULVANEY, B. D. (1968). Centriole migration. J. Ultrastruct. Res. 24, 86-101.
HOLLMANN, K. H. (1974). Cytology and fine structure of the mammary gland. In Lactation,
vol. 1 (ed. B. L. Larson & V. R. Smith), pp. 3-95. New York: Academic Press.
KEENAN, T . W., MORRE, D. J. & HUANG, C. M. (1974). Membranes of the mammary gland. In
Lactation, vol. 2 (ed. B. L. Larson, V. R. Smith), pp. 191-233. New York: Academic Press.
MAHOWALD, A. P., CAULTON, J. H., EDWARDS, M. K. & FLOYD, A. D. (1979). Loss of centrioles
and polyploidization in follicle cells of Drosophila melanogaster. Expl Cell Res. 118, 404-410.
MAHOWALD, A. P. & STRASSHEIM, J.. M. (1970). Intracellular migration of centrioles in the
germanum of Drosophila melanogaster. An electron microscopic study..7. CellBiol. 45, 306-320.
MAZIA, D. (1978). Origin of twoness in cell reproduction. \r\Cell Reproduction: In Honor of Daniel
Mazia (ed. E. R. Dirksen, D. M. Prescott & G. C. Fox), pp. 1—14. New York: Academic Press.
MULVANEY, B. D. & HEIST, H. E. (1971). Centriole migration during regeneration and normal
development of olfactory epithelium. J. Ultrastruct. Res. 35, 274-281.
POOLSWAT, S. S. (1973). The hardness of epoxy embedding compounds for ultrathin sectioning.
Proc. Ann. Meet. Electron Microsc. Soc. Am., vol. 31 ( ed. J. D. Davis), pp. 364—365. San
Francisco: San Francisco Press, Inc.
VAN BENEDEN, E. (1876). Contribution a l'histoire de la vfsicule germinative et du premier embryonnaire. Bull. Acad. r. Med. Belg. 42, 35-97.
VENABLE, J. H. & COGGESHELL, R. (1965). A simplified lead citrate stain for use in electron
microscopy. J. CellBiol. 25, 407-408.
WARCHOL, j . B., HERBERT, D. C. & RENNELS, E. G. (1974). An improved fixation procedure for
microtubules and microfilaments in cells of the anterior pituitary. Am.J. Anat. 141, 427-432.
WATSON, M. L. (1958). Staining of tissue sections for electron microscopy with heavy metals.
J. biophys. biochem. Cytol. 4, 475-478.
WHEATLEY, D. N. (1982). The Centriole: A Central Enigma of Cell Biology. New York: Elsevier
Biomedical Press.
(Received 25 April 1984 -Accepted 26 June 1984)