/. Embryol. exp. Morph. Vol. 39, pp. 71-77, 1977
Printed in Great Britain
71
Intercellular contacts at the
epithelial-mesenchymal interface of the developing
rat submandibular gland in vitro
By LESLIE S. CUTLER 1
From the Department of Oral Biology, School of Dental Medicine,
University of Connecticut Health Center
SUMMARY
An ultrastructural study of the development of the rat submandibular gland (SMG)
anlage in vitro was undertaken to determine if epithelial-mesenchymal and epithelial-nerve
contacts were integral events in the differentiation of the gland in vitro as they are in vivo.
SMG rudiments were removed at the stalk-bulb stage (15 days in utero) and cultured for 6
days on a millipore filter in supplemented McCoy's 5A media. Rudiments were taken at
daily intervals, fixed and processed for electron microscopy. The overall development of the
explanted rudiments closely paralleled their maturation in vivo although cultured glands
lagged 24-36 h behind their normal counterparts. Direct epithelial-mesenchymal contacts
were seen after the morphogenetic patterning of the gland had been established but prior
to functional differentiation of the rudiment. Epithelial-nerve contacts were not seen although
healthy axons were seen in the stroma throughout the culture period. The study indicates
that epithelial-nerve contacts are probably not required for morphogenesis of cytodifferentiation of the rat SMG. However, direct epithelial-mesenchymal contacts appear to be an
integral part of the developmental sequence of the rat SMG.
INTRODUCTION
In an earlier report (Cutler & Chaudhry, 1973 a) direct epithelial-mesenchymal and epithelial-nerve contacts were observed at the epithelial-mesenchymal
interface in the developing rat and submandibular gland (SMG). These contacts were seen during the period of rapid branching and budding of the early
glandular anlage and were only observed along the epithelial-mesenchymal
interface of the end-buds from the initial 4-12 branches. It was noted that these
contacts appeared after the morphogenetic branching pattern of the rudiment
had been established but prior to the onset of secretion and structural maturation in the gland. Thus, it was hypothesized that these contacts might play a
role in the functional differentiation and continued morphodifferentiation of
the gland subsequent to the establishment of the morphogenetic branching
pattern.
A recent report by Coughlin (1975 a) indicated that during the phase of rapid
1
Author's address: Department of Oral Biology, University of Connecticut Health Center,
School of Dental Medicine, Farmington, Connecticut 06032, U.S.A.
72
L. S. CUTLER
morphogenesis of the mouse submandibular gland epithelial-mesenchymal
contacts but not epithelial-nerve occur in vivo. On the other hand, Coughlin
(1975a) showed that epithelial-nerve contacts occur during morphogenesis of
the mouse SMG in vitro. The function of these epithelial-nerve contacts in
morphogenesis was questioned since normal morphogenesis occurred in the
total absence of the neural tissue (Coughlin, 1975 b).
An ultrastructural study of the development of the rat submandibular gland
anlage in vitro was undertaken to determine if epithelial-mesenchymal and
epithelial-nerve contacts could be observed and if so were such contacts important in morphogenesis and/or functional differentiation of the rudiment
cells.
MATERIALS AND METHODS
Holtzman strain, Sprague-Dawley rats were bred under the controlled conditions reported previously (Cutler & Chaudhry, 1973a). On the 15th day of
gestation fetuses were aseptically delivered by sterile laparotomy. The submandibular glands were quickly excised, washed twice in Hanks' balanced salt
solution and then transferred to a Falcon organ culture assembly and cultured
for 6 days on Millipore filter in supplemented McCoy's 5A media as described
previously (Cutler & Chaudhry, 1973 b). Rudiments were taken at daily intervals and fixed in 3 % phosphate buffered glutaraldehyde, then post-fixed in 1 %
osmium tetroxide, stained en bloc with 0-25% aqueous uranyl acetate and
routinely processed and embedded for electron microscopy (Cutler & Chaudhry,
1973 a, b, c). For orientation, 1 jLtm thick sections were cut and then stained with
1 % methylene blue and examined by light microscopy. Thin sections (60-90 jtcm)
were double stained with methanolic uranyl acetate (Stempak & Ward, 1964)
and lead citrate (Reynolds, 1963). Thin sections were examined on a Siemens
Elmiskop 1 or a Zeiss EM 10 electron microscope.
RESULTS
At the time of explantation the SMG rudiments were composed of a stalk and
solid terminal bulb (Fig. 1) which occasionally showed cleft formation indicative
of the initiation of branching. One day after explantation the rudiments were
actively branching and budding and by 2 days in vitro they consisted of multiple
branches and end-buds (Fig. 2). The branching and budding activity of the
explants decreased on the third day in culture but lumen formation was seen in
the main stalk and many of the end-buds. Occasionally, secretory granules were
seen in the end-bud cells bordering on the newly formed lumina (Fig. 3). By 6
days in culture the rudiments were richly arborized and there was a marked
accumulation of secretory product and ductal differentiation was well under
way (Fig. 4). Thus, while the overall development of the explanted rudiments
closely paralleled their in vivo maturation the cultures lagged 24-36 h behind
normal.
Epithelial-mesenchymal contacts in vitro
73
Direct epithelial-mesenchymal contacts were observed between the epithelium of the end-buds of the initial 6-15 branches and the surrounding mesenchyme. These contacts were seen predominantly on the second day and occasionally very early on the third day of the culture period. Contacts of this nature
were not seen prior or subsequent to this time point.
Most of the contact zones were formed by the junction of a mesenchymal cell
process with the plasma membrane of an end-bud cell (Fig. 5). Infrequently an
epithelial cell gave off an extension which contacted a neighboring mesenchymal
cell (Fig. 5). The predominant configuration of these junctions was that of an
intermediate contact zone. The opposing cell membranes were separated by a
gap of about 15 nm which frequently contained an amorphous material. The
opposed membranes were occasionally thickened but often showed no observable changes. A band of electron-dense material was seen in the cytoplasm
beneath the contact zone.
Contact zones in which a mesenchymal cell projection penetrated through
the basal lamina and extended into the end-bud in the intercellular space
between two end-bud cells (Fig. 6) were also seen. Areas of apparent fusion of
opposed epithelial and mesenchymal cell membranes were only rarely seen.
No epithelial-nerve contacts were observed although axons were seen in the
stroma throughout the culture period.
DISCUSSION
The epithelial-mesenchymal contacts observed in the current study appeared
at about the same stage of development as those observed in vivo (Cutler &
Chaudhry, 1973a). In both systems the contacts were seen during the period of
rapid branching and budding of the rudiment which followed the establishment of the morphogenetic pattern of the gland. Similarly, the contacts preceded the onset of glandular secretory activity, morphodifferentiation of myoepithelium and the maturation of specialized ductal components (Cutler &
Chaudhry, 1974, 1973a, b). In both cases, the contacts were seen only
along the epithelial-mesenchymal interface of the developing end-buds and the
morphology of the contact zone was remarkably similar.
Despite the continued vitality of nerves in the stroma throughout the culture
period, no epithelial-nerve contacts were seen. This is in contrast to observations made in vivo (Cutler & Chaudhry, 1973a). These observations are consistent with and extend those of Coughlin (1975a, b). Since glandular morphogenesis and functional differentiation proceeds in vitro in the absence of direct
epithelial-nerve interactions the role, if any, of such contacts in the cytodifferentiation of the SMG rudiment is probably minimal. However, direct
epithelial-mesenchymal interactions are always seen and their role in cytodirTerentiation may be significant.
The classic transfilter experiments of Grobstein (1956) and his associates
74
L. S. CUTLER
wmmmm
Epithelial-mesenchymal contacts in vitro
75
(Grobstein & Dalton, 1957; Koch & Grobstein, 1963; Kallman & Grobstein,
1965, 1966) suggested the presence of diffusible molecules which lead to 'embryonic induction' in the absence of direct epithelial-mesenchymal contacts.
However, reports by Nordling, Mirttinen, Wartiovaara & Saxen (1971) and
Saxen (1972) have indicated that the length of time required for induction to
occur between interacting cell populations in a transfilter situation tends to rule
out diffusion of molecules as a mechanism for information transfer. They
postulated that inductive information transmission might be mediated by direct
cellular contact or extracellular materials which were bound to the cell periphery. Both of these hypotheses require that the plasma membranes of cells
from the interacting tissues be brought into relatively close apposition. Reports
by Wartiovaara, Nordling, Lehtonen & Saxen (1974), Lehtonen (1975) and
Saxen et al. (1976) indicate that direct epithelial-mesenchymal contacts are
essential for the induction of kidney tubule morphogenesis. Recent studies have
added embryonic lung (Bluemink, van Maurik & Lawson, 1976) and tooth
germs (Slavkin & Bringas, 1976) to the growing list of tissues in which transient
epithelial-mesenchymal contacts are seen. In both cases, these contacts seem
more closely associated with functional differentiation of cells rather than
morphogenesis of the structure, a situation analogous to that in the developing
salivary gland.
The present study indicates that epithelial-mesenchymal contacts are an
integral part of the developmental sequence of the rat submandibular gland
both in vivo and in vitro while epithelial-nerve contacts are only seen in vivo
and are probably not of major importance in the cytodifferentiation of the
SMG. Moreover, the appearance of the epithelial-mesenchymal contacts after
FIGURES 1-6
Fig. 1. .1 /tm thick Epon section of a typical SMG rudiment at the time of explantation. Note that the rudiment is composed of a stalk (S) and a terminal bulb (TB).
x 350.
Fig. 2. 1 fim thick Epon section of a 1-day explant. Note the numerous end-buds
(EB). xl75.
Fig. 3. 1 /tm thick Epon section of an end-bud from a 3-day explant. Note the
dense secretory granules (arrows) around the small lumen, x 225.
Fig. 4. 1 (im thick section of SMG rudiment after 6 days in culture. Note that the
ducts and end-buds show lumina. x 80.
Fig. 5. Electron micrograph showing two epithelial-mesenchymal junctions
(arrows). Junction A shows thickening of the opposed membranes and is made by a
mesenchymal cell extension (ME) contacting the basal plasmalemma of an endbud cell. Junction B is made by an epithelial extension (EE) from an end-bud cell
which contacts the mesenchymal cell extension (ME) injunction A. Section from an
end-bud of a 2-day explant. x 75000.
Fig. 6. Electron micrograph showing a mesenchymal cell extension (ME) penetrating through the basal lamina (BL) and extending into the bud-end (SMG) in the
intercellular space between end-bud cells. Section from 2-day explant. x 32000.
76
L. S. CUTLER
the establishment of the morphogeaetic branching pattern and prior to the
functional differentiation (production of cell specific secretory protein) of the
rudiments suggests a potential role for these contacts in the 'induction' of
cytodifferentiation rather than morphodifferentiation in the rat SMG. Bernfield, Banerjee & Cohn (1972) have previously shown that deposition of proteoglycan is a central phenomenon in the development of the classic arborized
salivary gland structure.
It is possible that the contact zones observed in epithelial-mesenchymal
interactions in the salivary gland and other tissues might act as sites of transfer
for 'informational molecules'. Such molecules could be normal or specialized
constituents of the cell surface. Contacts such as these might also lead to focal
depolarization of the membranes allowing divalent cations such as Ca 2+ to
serve as informational molecules. In any case, the efficiency of such interactions
would be greatly facilitated by such anatomical arrangements. The molecules
would not have to travel (diffuse) over large distances, thus reducing their dilution or loss in the extracellular space. The rate of information transfer would be
enhanced since the interactions would occur over a short distance and finally,
such an arrangement could bring 'molecular inducers' into close proximity
with specific receptor sites on the opposing cell membrane.
The author wishes to acknowledge the valuable technical assistance of Mrs Constance
Christian and the secretarial help of Mrs Carol Manna.
This project was supported in part by N. I. H. grant 5R3DEO3933.
REFERENCES
M. R., BANERJEE, S. D. & COHN, R. H. (1972). Dependence of salivary epithelial
morphology upon acid mucopolysaccharide-protein (proteoglycan) at the epithelial surface. /. Cell Biol. 52, 674-689.
BLUEMINK, J. G., VAN MAURIK, P. & LAWSON, K. A. (1976). Intimate contact at the epithelialmesenchymal interface in embryonic mouse lung. / . Ultrastruct. Res. 55, 257-270.
COUGHLIN, M. D. (1975fl). Early development of parasympathetic nerves in the mouse submandibular gland. Devi Biol. 43, 123-139.
COUGHLIN, M. D. (19756). Target organ stimulation of parasympathetic nerve growth in the
developing mouse submandibular gland. Devi Biol. 43, 140-158.
CUTLER, L. S. & CHAUDHRY, A. P. (1974). Cytodifferentiation of the acinar cells of the rat
submandibular gland. Devi Biol. 41, 31-41.
CUTLER, L. S. & CHAUDHRY, A. P. (1973 a). Intercellular contact at the epithelial-mesenchymal interface during the prenatal development of the rat submandibular gland. Devi Biol.
33, 229-240.
CUTLER, L. S. & CHAUDHRY, A. P. (19736). Differentiation of the myoepithelial cells of the
rat submandibular gland in vivo and in vitro: an ultrastructural study. /. Morph. 140,
343-354.
CUTLER, L. S. & CHAUDHRY, A. P. (1973 c). Release and restoration of the secretory granules
in the convoluted granular tubules of the rat submandibular gland. Anat. Rec. 176, 405420.
GROBSTEIN, C. (1956). Trans-filter induction of tubules in mouse metanephrogenic mesenchyme. Expl Cell Res. 10, 424-440.
GROBSTEIN, C. & DALTON, A. J. (1957). Kidney tubule induction in mouse metanephrogenic
mesenchyme without cytoplasmic contact. /. exp. Zool. 135, 57-74.
BERNFIELD,
Epithelial-mesenchymal contacts in vitro
77
F. & GROBSTEIN, C. (1965). Source of collagen at the epitheliomesenchymal interfaces during inductive interaction. Devi Biol. 1, 169-183.
KALLMAN, F. & GROBSTEIN, C. (1966). Localization of glucosamine-incorporating materials
at epithelial surfaces during salivary epithelio-mesenchymal interaction in vitro. Devi
Biol. 14, 52-67.
KOCH, W. E. & GROBSTEIN, C. (1963). Transmission of labeled materials during embryonic
induction in vitro. Devi Biol. 7, 303-323.
LEHTONEN, E. (1975). Epithelio-mesenchymal interface during mouse kidney tubule induction in vivo. J. Embryol. exp. Morph. 34, 695-705.
NORDLING, S., MIRTTINEN, H., WARTIOVAARA, J. & SAXEN, L. (1971). Transmission and spread
of embryonic induction. ]. Temporal relationship in transfilter induction of kidney tubules
in vitro. J. Embryol. exp. Morph. 26, 231-252.
REYNOLDS, E. S. (1963). The use of lead citrate at high pH as an electron-opaque strain in
electron microscopy. /. Cell Biol. 17, 208-213.
SAXEN, L. (1972). Interactive mechanisms in morphogenesis. In Tissue Interactions in Carcinogenesis (ed. D. Tarin) pp. 49-80. New York: Academic Press.
KALLMAN,
SAXEN, L., LEHTONEN, E., KARKINEN-JAASKELAINEN, M., NORDLING, S. & WARTIOVAARA, J.
(1976). Are morphogenetic tissue interactions mediated by transmissible signal substances or through cell contacts? Nature, Lond. 259, 662-663.
SLAVKIN, H. C. & BRINGAS, P. (1976). Epithelial-mesenchymal interactions during odontogenesis. IV. Morphologic evidence for direct heterotypic cell-cell contacts. Devi Biol. 50,
428-442.
STEMPAK, J. G. & WARD, R. T. (1964). An improved staining method for electron microscopy.
/. Cell Biol. 22, 697-701.
WARTIOVAARA, J., NORDLING, S., LEHTONEN, E. & SAXEN, L. (1974). Transfilter induction of
kidney tubules: correlation with cytoplasmic penetration into nucleopore filters. /. Embryol. exp. Morph. 31, 667-682.
{Received JO August 1976)
EMB 39