Human Reproduction, Vol. 15, (Suppl. 3), pp. 182-188, 2000
Junctional barrier complexes undergo major alterations
during the plasma membrane transformation of uterine
epithelial cells
Christopher R.Murphy
Department of Anatomy and Histology, The University of Sydney, NSW, Australia
Address for correspondence: Department of Anatomy and Histology, The University of Sydney, NSW
2006, Australia. E-mail: [email protected]
Junctions in the plasma membrane of uterine
epithelial cells as well as between these cells and
their extracellular environment are examined
in this review to see if a synthetic appreciation
of their role can be gained from the disparate
evidence presently available. Major changes in
most junctional components are noted during
early pregnancy and the role of progesterone
and oestrogen in promoting these changes is
examined. In particular it is noted that while
tight junctions become deeper and morphologically 'tighter' towards the time of implantation, other basolateral junctional structures as
well as their cytoskeletal associations are absent.
These junctional alterations are part of the
'plasma membrane transformation' of early
pregnancy and allow the conclusion that while
paracellular permeability is reduced by the time
of blastocyst attachment, the epithelial cells are
paradoxically less firmly attached to each other,
and to their extracellular environment.
Key words: barrier function/epithelial cell/junction/
pregnancy/uterus
Introduction
Epithelial cells have special structural features and
among these are the junctions they form with each
other and with their extracellular environment.
These junctions contribute in various ways to the
barrier function of epithelia — the capacity of
this tissue to separate different environments. The
182
uterine epithelium has the usual complement of
junctional types but, unlike other epithelia, this
epithelium has roles and sensitivities not shared
by these other epithelia. Unique among epithelia,
the uterine epithelium is cyclically altered by
ovarian hormones which regulate its capacity to
permit blastocyst implantation, and it is then periodically invaded from its apical surface by the
implanting blastocyst. The uterine epithelium thus
has, in addition to the usual epithelial barrier
function, the function of regulating the early events
of uterine receptivity for implantation. The effects
of the ovarian hormones on the uterine epithelium
are many but those on the plasma membrane
have been described as 'the plasma membrane
transformation' (Murphy, 1993, 1995, 1998;
Murphy and Shaw, 1994) to encapsulate the concept of a process of change involving all membrane
domains during early pregnancy.
The epithelial cell junctions are key players in
epithelial events during early pregnancy and reports
on different junctional components have been
available for some years. Recently, however, new
data on cytoskeletal and molecular components
of the junctions of uterine epithelial cells have
highlighted the extent of junctional involvement
in the processes of change during early pregnancy
in particular. This review draws together our knowledge of junctional processes in uterine epithelial
cells and highlights the inter-relatedness of junctional involvement in the plasma membrane transformation of early pregnancy and uterine function
more generally.
© European Society of Human Reproduction & Embryology
Junctions of uterine epithelial cells
The tight junction (zonula occludens)
The lateral plasma membranes of epithelial cells
face the relatively constant environment of another
epithelial cell and thus — unlike the apical and
basal portions of a plasma membrane which face
different external and internal environments
respectively — are generally similar to each other.
This is also true of the uterine epithelium, notwithstanding the reproductive specialization of
these epithelial cells.
The tight junction is the most apical in the
junctional complex and was originally described
in a thin-section electron microscopic study of
several epithelia (Farquhar and Palade, 1963);
these workers used the term zonulae occludentes
(occluding junctions) and surmised their role in
preventing or reducing paracellular movement of
molecules. Because of its capacity to expose laterally extended regions of plasma membrane, freezefracture is uniquely suited for the study of tight
junctions and has been extensively employed to
examine the rows of closely packed integral membrane proteins (IMP) which appear as strands in
replicas and constitute the freeze-fracture appearance of these junctions (Claude and Goodenough,
1973; Staehelin, 1974; Cereijido, 1992).
In uterine epithelial cells, tight junctions were
first studied in rat uterine epithelial cells during
early pregnancy (Murphy et al., 1982a); it was
found that on day 1 these junctions consisted
almost entirely of strands running parallel to the
apical surface with few vertical strands connecting
the horizontally oriented ones. By the time of
blastocyst attachment, however, the junctions had
become over three times deeper, at up to 3 (im
down the lateral plasma membrane, and also much
more complex in structure with many interconnections between the strands, which by then consisted
of arrays of branching, anastomosing structures.
These changes in structure are predominantly under
the control of progesterone. Earlier, the capacity
of ovarian hormones injected into ovariectomized
rats to generate different patterns of tight-junctional
organization had been studied (Murphy et al.,
1981) and it was found that oestrogen produced a
tight junction consisting largely of the parallel
strand type whereas progesterone alone resulted in
the much more complex pattern of many intercon-
nections and a greater depth down the lateral
membrane. Progesterone and oestrogen together
produced effects not discernibly different from
progesterone alone.
An increase in depth and geometrical complexity
of tight junctions has similarly been described
during early pregnancy in rabbits together with the
formation of isolated strands of macular tight
junctions below the main complex (Winterhager
and Kuhnel, 1982); an increase in complexity but
not depth around the time of implantation in pigs
was also interpreted as having been progesteroneinduced (Johnson et al., 1988). In women, tight
junctions are geometrically more complex earlier
than later in the menstrual cycle (Murphy et al.,
1982b, 1992). It is thus remarkable that in all
the species where sufficient evidence exists, it
points to a progesterone-induced increase in tightjunctional complexity. Tight junctions are well
known as regulators of paracellular flow with
more strands and particularly a more complex
geometry, generally indicating decreased paracellular permeability (Claude and Goodenough, 1973;
Gonzalez-Mariscal, 1992). It would seem likely
that the increasing complexity and morphological
'tightness' seen during early pregnancy is a
reflection of the need to preserve the luminal
contents specially developed by epithelial secretion
for blastocyst development, from dilution or loss
by unwanted flow into or out of the lumen (Murphy
et al., 1982a).
The adherens junction (zonula adherens)
The adherens junction occurs immediately beneath
the tight junction and may occupy 1-2 fim of
lateral plasma membrane (Farquhar and Palade,
1963). Of the various components of epithelial
junctions, the adherens component has probably
attracted least attention, perhaps because it is not
as ultrastructurally conspicuous as the others and
freeze-fracture preparations are able to add little
understanding to its structure — unlike the spectacular contributions this technique made to understanding of tight junctions (Staehelin and Hull,
1978).
Adherens junctions are usually seen as comparatively indistinct densities immediately below the
tight junction and special staining preparations
may be useful in highlighting this area.
183
C.R.Murphy
Based on a previous study (Karnovsky, 1971),
Hyland and Murphy (unpublished observations)
developed and extended a method employing the
addition of potassium ferrocyanide to the osmium
fixation protocol which gave significantly improved
resolution of these difficult-to-resolve junctions.
This technique allowed the observation that, on
day 1 of pregnancy, there were prominent densities
immediately beneath the tight junction region
which extended some 0.75-1 fim down the lateral
plasma membrane. By day 3 of pregnancy, while
some cytoplasmic density remained, it had become
difficult to clearly discern this region. By the time
of the completed 'plasma membrane transformation' on day 6 of pregnancy however, observable
cytoplasmic density which distinguished this region
of the plasma membrane from other adjacent
regions could not be discerned.
These ultrastructural observations on the
adherens junction suggest that, as the tight junction
extends further down the lateral plasma membrane,
the adherens junction immediately beneath it is
displaced and lost as a distinct membranous junctional structure. Consistent with these direct observations, light microscopical studies using
immunohistochemistry of E-cadherin, which is a
molecular component of the adherens junction,
show a loss of these molecules from the lateral
plasma membrane in mice as early pregnancy
advances towards the time of attachment (Potter
et al., 1996). Studies on rabbits (Denker, 1994)
and on human uterine epithelial cells (Getsios
et al., 1998) have also reported down-regulation
of different cadherins during early pregnancy. Collectively, these observations suggest that uterine
epithelial cells become less adherent to each other
as early pregnancy progresses towards the time of
blastocyst attachment.
The terminal web
A key component of the adherens junction on its
cytoplasmic side is the terminal web — a layer of
actin filaments which inserts into the lateral plasma
membrane principally at the level of the adherens
junction. Given the lateral downwards spread of
the tight junction and the virtual disappearance of
the adherens junction during early pregnancy, the
interesting question of the fate of this important
184
cytoplasmic-junctional structure is raised and some
major change might be expected.
To visualize and understand the involvement of
membrane skeletal elements associated with the
plasma membrane region, a new high-resolution
approach was called for; a method to detect these
elements in rat uterine epithelial cells at the ultrastructural level has been adapted and developed
(Luxford and Murphy, 1992a,b). This approach
employs mild membrane permeablization using the
detergent Triton, followed by decoration of actin
microfilaments with myosih subfragment 1
(Mooseker and Tilney, 1975). At the electron
microscopic level, it allowed the relationship
between actin filaments in the submembranous
cytoplasm and terminal web and the plasma membrane to be followed (Luxford and Murphy,
1992a,b). On day 1 of pregnancy a prominent
terminal web was visible and surface microvilli
contained bundled microfilaments descending into
this terminal web. Other filaments were associated
with the membrane in a less regular fashion. As
pregnancy progressed, however, and microvilli
gradually became less regular, the terminal web
progressively disassociated into disconnected short
clumps of filaments until by day 6 any form of
organized terminal web was no longer present. At
this time of uterine receptivity in rats, the irregular
flattened projections of the apical plasma membrane contained either isolated bundles or meshworked microfilaments with few laterally arranged
filaments connecting the bundles. Isolated, unconnected clusters of actin filaments were also evident
on day 6 in the apical cytoplasm just beneath the
plasma membrane. It was also found that as the
terminal web was lost, cellular vesicles and other
organelles which previously were excluded from
the submembranous cytoplasm, presumably by the
dense terminal web of filaments, were able to
approach the apical plasma membrane (Luxford
and Murphy, 1992a). In a continuation of these
studies, it was shown (Luxford and Murphy, 1992b)
that oestrogen and progesterone had quite different
effects on the actin microfilament organization and
that neither, acting alone, could restore a terminal
web in ovariectomized rats. However, the two
hormones acting together in the receptivity-producing combination (Psychoyos, 1973) produced a
Junctions of uterine epithelial cells
pattern of microfilament organization in ovariectomized rats indistinguishable from that described
above on day 6. Thus the loss of the terminal web
is under maternal ovarian hormonal control and its
dissolution at the time of uterine receptivity does
not rely on the presence of a blastocyst.
The loss of the terminal web in any epithelial
cell would seem to be a highly unusual event and
the resulting cell might be thought of as a 'house
without a ceiling' if, on the same analogy, the
apical plasma membrane is thought of as the 'roof
of the cell. Such an event is likely to have
considerable significance for the many changes
seen in the various regions of the plasma membrane
during early pregnancy, and disassociation of actin
terminal web microfilaments could contribute to
the apical loss of microvilli which is a key component of the plasma membrane transformation that
seems to be essential for receptivity (Murphy,
1993, 1995, 1998). An important agent in these
processes may be the characteristic large apical
vesicles of uterine epithelial cells which are under
progesterone control in these cells (Ljungkvist,
1971). Previously excluded from the subapical
membranous cytoplasm — presumably by the
dense actin terminal web — these vesicles, which
have cholesterol-rich membranes, approach and
fuse with the apical membrane around the receptive
phase for attachment as the terminal web is lost
(Parr, 1982; Murphy and Martin, 1987). Such
alterations in vesicular traffic could contribute to
the well-known increase in apical plasma membrane cholesterol by the time of attachment and
may also contribute to the rearrangements and
altered expression of the many molecules involved
in the transformation of this membrane during
early pregnancy (Murphy and Martin, 1987;
Hyland et al, 1998; Isaacs and Murphy, 1998)
As the terminal web is lost, several actin-binding
molecules undergo significant reorganization. In
particular, cc-actinin and gelsolin display a prominently altered distribution being lightly distributed
all around the plasma membrane on day 1 of
pregnancy but becoming strongly concentrated
apically by day 6 (Terry et al, 1996). The apical
focus of these actin-capping molecules may contribute to the formation of the 'bundles' of actin
seen near the apical plasma membrane on day 6
(Luxford and Murphy, 1992a). Of particular interest
in this context is a recent study (Orchard et al.,
1999) which found that as the terminal web and
adherens junction are lost, plakoglobin, a key
component of desmosomal-type junctions, became
restricted to the apical quarter of the lateral plasma
membrane whereas previously it had occurred all
down the lateral plasma membrane. This suggests
that the loss of these junctional components allows
this molecule to focus along the apical-most portion
of the lateral plasma membrane where it might
contribute to junction formation with the penetrating shafts of trophoblast which are known to
transiently share junctions with uterine epithelial
cells in some species (Schlafke and Enders, 1975;
Enders et al, 1995).
The basal plasma membrane and basal
junctions
The basal plasma membrane and its associated
basal lamina (lamina densa) have in general
received less attention than other regions of the
plasma membrane of uterine epithelial cells,
although one topic of considerable interest has
been the degree to which the basal lamina and
attendant material is degraded and penetrated by
trophoblast or stromal cells or both (Blankenship
and Given, 1995).
Several workers have found that the basal plasma
membrane and basal lamina increase in tortuosity
during early pregnancy (Guillomot et al., 1981;
Blankenship and Given, 1992; Keys and King,
1992); in addition, in rabbits (Marx et al, 1990)
and in women (Dockery et al, 1998) the thickness
of the basal lamina increases during the reproductive cycle. To appreciate the influence of ovarian
hormones on the changing nature of the basal
plasma membrane and basal lamina over the full
period of early pregnancy up to the time of
attachment in the rat, one study (Shion and Murphy,
1995) has compared the basal region of uterine
epithelial cells during early pregnancy as well as
those from rats treated with various regimes of
exogenous ovarian hormones. In agreement with
most other studies on simple epithelia, ultrastructurally obvious hemidesmosomes were not reported
(Shion and Murphy, 1995). The basal plasma
membrane on day 1, however, was seen to be
185
C.R.Murphy
Figure 1. (A) Uterine luminal epithelial cell on day 1 of pregnancy. Microvilli (MV) are long and numerous and actin
microfilaments descend from inside them to join the actin filaments of the terminal web (TW). A prominent adherens junction
(AJ) is evident just below the tight junction (TJ). Basally, focal contacts (FC) along the relatively smoothly undulating basal
plasma membrane are seen just above the basal lamina (BL). (B) Uterine luminal epithelial cell on day 6 of pregnancy. The
apical plasma membrane is now highly flattened and microvilli are no longer present. The adherens junction and terminal web
are also no longer present but the tight junction (TJ) now extends three to four times further down the lateral plasma
membrane than on day 1. Focal contacts are absent from the basal plasma membrane which is now much more highly folded,
as is the basal lamina (BL) which is also thicker than on day 1.
relatively smooth in outline with the attendant
lamina densa averaging 13 nm in thickness and
with frequent thickenings between the two which
were referred to as 'immature hemidesmosomes'.
Such structures might now more appropriately be
referred to as 'focal adhesions' (Voori, 1998). By
day 6 of pregnancy, however, this study found that
the 'immature hemidesmosomes' were no longer
present but that the basal plasma membrane and
lamina densa had become highly folded. The
lamina densa was found on day 6 to have significantly increased in thickness to 23 nm although it
still ran parallel to the basal plasma membrane in
all regions observed. Shion and Murphy (1995)
also found that while oestrogen alone induced
'focal contacts' in the basal plasma membrane, it
had little other effect on basal lamina thickness.
Progesterone acting alone, however, was responsible for most of the increase in basal folding and
thickness of the lamina densa as well as loss of
the 'immature hemidesmosomes' (focal contacts).
This finding on the effect of progesterone is consistent with studies on users of long-term progesterone
contraceptives which have also been reported to
induce increased tortuosity in the basal region of
uterine epithelial cells (Pakarinen et al., 1998).
186
The increased tortuosity of the basal plasma
membrane and increased thickness of the basal
lamina reported in several studies are perhaps
indicative of the tissue remodelling which occurs
during early pregnancy and the thicker basal lamina
may be a device to slow the blastocyst in its
journey into the stroma. On the other hand, the
loss of 'focal contacts' suggests that perhaps uterine
epithelial cells are less adherent to the basal lamina
by the time of blastocyst attachment and this is
consistent with observations that the epithelium is
more easily removed from the stroma at this time
(Tachi et al, 1970; Schlafke and Enders, 1975;
Murphy, 1993). At the molecular level and also
consistent with this view, it has recently been
shown (Hyland et al, 1998) that at least some
cadherins — molecules important in adhesion —
are lost from the basal surface of uterine epithelial
cells in rats by day 6 of pregnancy.
Changes in membrane junctional structures of
uterine epithelial cells during 'the plasma membrane transformation' of early pregnancy are summarized in Figure 1.
Conclusion
It is clear that during early pregnancy, there are
major shifts in the organization of most if not all
Junctions of uterine epithelial cells
junctional structures in the plasma membrane of
uterine epithelial cells as well as in many of their
associated cytoskeletal structures. These shifts in
organization emphasize the diversity and extent of
'the plasma membrane transformation' of uterine
epithelial cells.
An interesting conclusion, based on this evidence
of changes to uterine junctional structures in
response to hormones, is that while uterine epithelial cells continue to be a barrier to molecular
movement between the cells, they are much less
of a barrier to physical trauma — whether that be
a blastocyst from above or, in the case of abnormal
uterine bleeding, blood from below.
Acknowledgements
The author is grateful to the Australian Research Council
who supported aspects of the work described here
and to Serono Australia Ltd who have supported its
continuation.
References
Blankenship, T.N. and Given, R.L. (1992) Penetration
of the uterine epithelial basement membrane during
blastocyst implantation in the mouse. Anat. Rec, 233,
196-204.
Blankenship, T.N. and Given, R.L. (1995) Loss of
laminin and type IV collagen in uterine luminal
epithelial basement membrane during blastocyst
implantation in the mouse. Anat. Rec, 243, 27-36.
Cereijido, M. (ed.) (1992) Tight Junctions. CRC Press,
Boca Raton, 367 pp.
Claude, P. and Goodenough, D.A. (1973) Fracture faces
of zonulae occludentes from 'tight' and 'leaky'
epithelia. J. Cell Biol., 58, 390-400.
Denker, H.-W. (1994) Endometrial receptivity: cell
biological aspects of an unusual epithelium. A review.
Ann. Anat., 176, 53-60.
Dockery, P., Khalid, J., Sarani, S.A. etal. (1998) Changes
in basement membrane thickness in the human
endometrium during the luteal phase of the menstrual
cycle. Hum. Reprod. Update, 4, 486-495.
Enders, A.C., Liu, I.K.M., Mead, R.A. and Welsh, A.O.
(1995) Active and passive morphological interactions
of trophoblast and endometrium during early
implantation. In Dey, S.K. (ed.), Molecular and
Cellular Aspects of Preimplantation Processes.
Springer Verlag, New York, pp. 168-182.
Farquhar, M.G. and Palade, G.E. (1963) Junctional
complexes in various epithelia. /. Cell Biol., 17,
375^112.
Getsios, S., Chen, G.T.C., Stephenson, M.D. etal. (1998)
Regulated expression of cadherin-6 and cadherin-11
in the glandular epithelium and stromal cells of the
human endometrium. Dev. Dynam., 211, 238-247.
Gonzalez-Mariscal, L. (1992) The relationship between
structure and function of tight junctions. In Cereijido,
M. (ed.), Tight Junctions. CRC Press, Boca Raton,
pp. 67-76.
Guillomot, M., Flechon, J.E. and Wintenberger-Torres,
S. (1981) Conceptus attachment in the ewe: an
ultrastructural study. Placenta, 2, 169-182.
Hyland, R.A., Shaw, T.J., Png, F.Y. and Murphy, C.R.
(1998) Pan-cadherin concentrates apically in uterine
epithelial cells during uterine closure in the rat. Acta
Histochem., 100, 75-81.
Isaacs, J. and Murphy, C.R. (1998) Expression of the
carbohydrate antigen CD 15 in rat uterine epithelial
cells during the early stages pregnancy. Eur. J.
Morphol. 36, 49-56.
Johnson, S.A., Morgan, G. and Wooding, F.B.P. (1988)
Alterations in uterine epithelial tight junction structure
during the oestrous cycle and implantation in the pig.
/. Reprod. Fertil., 83, 915-922.
Karnovsky, M.J. (1971) Use of ferrocyanide-reduced
osmium tetroxide in electron microscopy. Proc. 14th
Annu. Meet. Am. Soc. Cell Biol., pp. 146 (Abstract
294).
Keys, J.L. and King, G.J. (1992) Effects of topical and
systemic estrogen on morphology of porcine uterine
luminal epithelia. Biol. Reprod., 46, 1165-1175.
Ljungkvist, I. (1971) Attachment reaction of rat uterine
luminal epithelium. The effect of progesterone on the
morphology of the uterine glands and the luminal
epithelium of the spayed, virgin rat. Acta Soc. Med.
Upsal., 76, 110-126.
Luxford, K.A. and Murphy, C.R. (1992a) Reorganization
of the apical cytoskeleton of uterine epithelial cells
during early pregnancy in the rat: a study with myosin
subfragment 1. Biol. Cell, 74, 195-202.
Luxford, K.A. and Murphy, C.R. (1992b) Changes in
the apical microfilaments of rat uterine epithelial cells
in response to estradiol and progesterone. Anat. Rec,
233, 521-526.
Mooseker, M.S. and Tilney, L.G. (1975) Organization
of an actin filament-membrane complex: filament
polarity and membrane attachment in the microvilli
of intestinal epithelial cells. /. Cell Biol., 67, 725-743.
Marx, M., Winterhager, E. and Denker, H.-W (1990)
Penetration of the basal lamina by processes of the
uterine epithelial cells during implantation in the
rabbit. Trophoblast Res., 4, 417^30.
Murphy, C.R. (1993) The Plasma Membrane of Uterine
Epithelial Cells: Structure and Histochemistry. Gustav
Fischer, Stuttgart and New York, 68 pp.
Murphy, C.R. (1995) The cytoskeleton of uterine
epithelial cells: a new player in uterine receptivity and
the plasma membrane transformation. Hum. Reprod.
Update, 1, 567-580.
Murphy, C.R. (1998) Commonality within diversity: the
plasma membrane transformation of uterine epithelial
187
C.R.Murphy
cells during early placentation. J. Assist. Reprod.
Genet., 15, 179-183.
Murphy, C.R. and Martin, B. (1987) Digitonin
cytochemistry reveals cholesterol-rich vesicles in
uterine epithelial cells. Acta Histochem., 81,143-147.
Murphy, C.R. and Shaw, T.J. (1994) Plasma membrane
transformation: a common response of uterine
epithelial cells during the peri-implantation period.
CellBiol. Int., 18, 1115-1128.
Murphy, C.R., Swift, J.G., Mukherjee, T.M. and Rogers,
A.W. (1981) Effects of ovarian hormones on cell
membranes in the rat uterus. II. Freeze-fracture studies
on tight junctions of the lateral plasma membrane of
the luminal epithelium. Cell Biophys., 3, 57-69.
Murphy, C.R., Rogers, P.A.W., Hosie, M.J. et al. (1992)
Tight junctions of human uterine epithelial cells
change during the menstrual cycle: a morphometric
study. Acta Anat., 144, 36-38.
Murphy, C.R., Swift, J.G., Mukherjee, T.M. and Rogers,
A.W. (1982a) The structure of tight junctions between
uterine luminal epithelial cells at different stages of
pregnancy in the rat. Cell Tiss. Res., 223, 281-286.
Murphy, C.R., Swift, J.G., Need, J.A. et al. (1982b) A
freeze-fracture electron microscopic study of tight
junctions of epithelial cells in the human uterus. Anat.
EmbryoL, 163, 367-370.
Orchard, M.D., Shaw, T.J. and Murphy, C.R. (1999)
Junctional plaque proteins shift to the apical surface
of uterine epithelial cells during early pregnancy in
the rat. Acta Histochem., 101, 147-156.
Pakarinen, P.I., Lahteenmaki, P., Lehtonen, E. and Reima,
I. (1998) The ultrastructure of human endometrium is
altered
by
administration
of
intrauterine
levonorgestrel. Hum. Reprod., 13, 1846-1853.
Parr, M.B. (1982) Apical vesicles in the rat uterine
epithelium during early pregnancy: a morphometric
study. Biol. Reprod., 26, 915-924.
Potter, S.W., Gaza, G. and Morris, J.E. (1996) Estradiol
induces E-cadherin degradation in mouse uterine
epithelium during the estrous cycle and early
pregnancy. J Cell. Physiol., 169, 1-14.
Psychoyos, A. (1973) Hormonal control of
ovoimplantation. Vitam. Horm., 31, 201-256.
Terry, V., Shaw, T.J., Shorey, CD. and Murphy, C.R.
(1996) Actin binding proteins undergo major
alterations
during
the
plasma
membrane
transformation in uterine epithelial cells. Anat. Rec,
246, 71-77.
Schlafke, S. and Enders, A.C. (1975) Cellular basis of
interaction between trophoblast and uterus at
implantation. Biol. Reprod., 12, 41-65.
Shion, Y-L. and Murphy, C.R. (1995) The basal plasma
membrane and lamina densa of uterine epithelial cells
are both altered during early pregnancy and by ovarian
hormones in the rat. Eur. J. Morphol., 33, 257-264.
Staehelin, L.A. (1974) Structure and function of
intercellular junctions. Int. Rev. Cytol., 39, 191-283.
188
Staehelin, L.A. and Hull, B.E. (1978) Junctions between
living cells. Scient. Am., 238, 140-152.
Tachi, S., Tachi, C. and Lindner, H.R. (1970)
Ultrastructural features of blastocyst attachment and
trophoblastic invasion in the rat. J. Reprod. Fertil.,
21, 37-56.
Voori, K. (1998) Integrin signalling: tyrosine
phosphorylation events in focal adhesions. J. Memb.
Biol 165, 191-199.
Winterhager, E. and Kuhnel, W. (1982) Alterations in
intercellular junctions of the uterine epithelium during
the preimplantation phase in the rabbit. Cell Tiss.
Res., 224, 517-526.