/. Embryol. exp. Morph. Vol. 27, 2, pp. 413-430, 1972
41 3
Printed in Great Britain
Cytodifferentiation of the mouse
secondary palate in vitro: morphological,
biochemical, and histochemical aspects
By VICTOR IDOYAG A-VARGAS,1
CARLOS E. NASJLETI2 AND JULIO M. AZCURRA 3
From the Mental Health Research Institute, University of Michigan
and Veterans Administration Hospital, Ann Arbor, Michigan
SUMMARY
Previous investigations had indicated that the epithelial cells at the medial surface of the
palate differentiate prior to fusion. An increase in adhesiveness and signs of partial autolysis
were reported to be part of the differentiation. It was not determined, however, whether or
not the autolytic processes may result in the total lysis of the cells, even if fusion is prevented.
This problem was approached in the present investigation by culturing shelves in isolation,
thus preventing fusion. The morphological properties of the presumptive fusing cells were then
observed during and after the actual time of fusion. Differentiation was assessed in vivo at
equivalent developmental ages. In addition, the role of the lysosomes in the process of fusion
was investigated. This was done by measuring the activities of two lysosomal enzymes,
glucosaminidase and arylsulfatase, before, during, and after fusion in vivo and at the equivalent times in vitro. Glucosaminidase activity was also localized histochemically.
The in vitro and in vivo differentiation were similar. The cells of the mesenchyme produced
intramembranous bone and loose mesenchyme. The nasal surface differentiated into a pseudostratified columnar ciliated epithelium, while the oral surface became stratified squamous
epithelium. At the medial surface, however, the morphological changes during fusion in vivo
were different from those observed at the equivalent time in vitro. Few, if any, of the presumptive fusing cells forming a bi-layered or multilayered epithelium degenerated in the
explants, while many cells forming the epithelial seam in vivo underwent complete degeneration. In addition, after fusion, the medial surface of the palatal explants was covered by a
stratified squamous and/or pseudostratified columnar ciliated epithelia. In vivo, the mesenchyme merged as a consequence of the total destruction of the seam. The specific activity of
glucosaminidase increased progressively, reaching a peak three days after fusion. In all cases
no statistically significant differences were found between the in vivo and in vitro values.
Arylsulfatase specific activity remained the same both in vivo and in vitro. Histochemically
glucosaminidase activity was localized in the cells undergoing degeneration during fusion. It
is concluded that fusion is a necessary condition for the completion of the autolysis of the
presumptive fusing cells. The cells may alter their developmental fate, forming a stratified
squamous and/or pseudostratified columnar epithelia, if fusion is prevented.
1
Author's address: Oak Ridge National Laboratories, Oak Ridge, Tennessee, 38830
U.S.A.
2
Author's address: Veterans Administration Hospital, Ann Arbor, Michigan, 48105
U.S.A.
3
Author's address: Instituto de Anatomia Gral y Embriologia, Facultad de Ciencias
Medicas, Buenos Aires, Argentina.
414
V. IDOYAGA-VARGAS AND OTHERS
INTRODUCTION
One of the main events during the morphogenesis of the oro-facial region is
the closure of the secondary palate. This consists of the process of elevation and
fusion of the palatine shelves, both of which have been imputed to be involved
in the production of cleft palate (Fraser, 1967). Thus, it becomes important to
understand the mechanisms underlying shelf elevation and shelf fusion.
Conspicuous among these mechanisms are the changes in the epithelial cells
at the medial surface of the palatine shelves. These cells acquire the potentiality
to fuse during their development (Pourtois, 1966), a potentiality which is
expressed as an increased adhesiveness (Pourtois, 1968) and signs of partial
autolysis of the cells before the actual time of fusion (Mato, Aikawa & Katahira,
1967; Angelici & Pourtois, 1968). It was not determined, however, whether or
not the autolytic processes may result in the total lysis of the cells even if fusion
is prevented. The dissociation of the processes of fusion and degeneration may
help to decide which alternative is valid. If palate closure is prevented and the
cells undergo complete degeneration, then epithelial autolysis is independent of
fusion. The degeneration of cells during development is accompanied by an
increase in lysosomal activity (Scheib-Pfleger & Wattiaux, 1962; Weber, 1963;
Salzgeber & Weber, 1966). Thus, an increase in lysosomal activity may be
expected if the presumptive fusing cells degenerate. However, if by preventing
fusion cells do not degenerate, then palatine closure and epithelial seam formation is a necessary condition for total cell lysis.
In the present investigation an approach was made to the problem of whether
the cells degenerate regardless of fusion. Palatine shelves were cultured in
isolation and the activity of two lysosomal enzymes was measured. These
enzymes were /?-2-acetamido-2-deoxy-D-glucoside acetamidoglucohydrolase
(Glucosaminidase; E.C.: 3.2.1.30) and arylsulfatesulphohydrolase (Arylsulfatase; E . C : 3.1.6.1). In previous studies in vitro (Pourtois, 1966; Vargas,
1967), the palatine shelves were explanted with the medial surface in contact to
permit analysis of the fusion process. In the present work the shelves were
explanted without permitting contact with each other; fusion was thus prevented
and conditions established for the observation of the presumptive fusing cells
before and during the equivalent time of fusion in vivo. Moreover, by appropriately extending the culture period, the cells of the oral and nasal surfaces
differentiated so as to allow a comparison with the cells at the fusing edge. This
provided information on the fate of the cells at the medial surface after fusion.
MATERIALS AND METHODS
Organ culture
Pregnant mice of the G.P. white (N.I.H.) Swiss strain were used in this study.
The time of pregnancy was determined by the plug method, and the final
Cytodijferentiation of mouse palate in vitro
415
selection of the embryos was made using the morphological criterion of Walker
& Fraser (1956). Anesthetized pregnant mice were killed by cervical dislocation.
The embryos were immediately removed and the palatine shelves dissected out
in balanced salt solution (B.S.S.) Hanks's medium, as described previously
(Vargas, 1967). Isolated palatine shelves were explanted, nasal or oral surface
up, on a medium of the following composition:
Solution A:
NCTC-109 (Evans et al 1956) 22-5 vol.
Fetal bovine serum
5-0 vol.
Sodium bicarbonate, 7-5 %
0-25 vol.
The above-mentioned chemicals were obtained from the Microbiological
Associates.
Solution B:
lonagar 2, Colab. Laboratories, Inc. 0-4 g.
Bi-distilled water
22-5 vol.
To obtain the final medium, solution B was autoclaved, allowed to cool and
mixed with solution A.
Palatine shelves were dissected from 13^ and 14^ (prior to fusion) days postconception (P.C.) and were cultured from 1 to 4 days according to the five
different schedules described below. The controls consisted of histological serial
sections of embryonic heads aged 12^ to 18^ days P.C. In the experimental group,
one palatine shelf was fixed in Bouin's immediately after dissection and the
homotypic shelf was explanted. The culture schedules were as follows:
Series I. Explants dissected at 13^ days P.C. and cultured for one day; total
developmental age was 14^ days.
Series II. Explants dissected at 13^ and 14i days P.C. and cultured for two
days and one day, respectively; total developmental age was 15^ days.
Series III. Explants dissected at 13^ and 14^ days P.C. and cultured for three
and two days, respectively; total developmental age was 16^ days.
Series IV. Explants dissected at 13^ and 14^ days P.C. and cultured for four
and three days, respectively; total developmental age was 17^ days.
Series V. Explants dissected at 14^ days P.C. and cultured for four days.
Three experiments were carried out for each age group for a total of 192 explants.
Biochemistry
The in vivo experiments utilized embryos of 13^, 14^ and 17^ days P.C. In
vitro experiments utilized those of 12| days P.C. which were cultured for 24, 48
and 120 hours.
The fused palates were dissected out from the maxillary arch and the nasal
septum and separated into individual shelves which were pooled and suspended
in 0-25 M sucrose. The suspension was homogenized vigorously up to 20 times
416
V. IDOYAGA-VARGAS AND OTHERS
by means of a mechanically driven pestle rotating in a glass Potter-Elvejhem
homogenizer at about 1300 rev/min.
For the assay of glucosaminidase, the tubes contained between 400 and 550 fig
of palate homogenate proteins, 0-15 ml of 4-8 IHM /?-nitrophenyl-2-acetamido-2deoxy-/?-glucopyranoside (Pierce Chemicals, Rockford, Illinois) and 0-68 ml of
phosphate-citrate buffer (Elving, Markowitz & Rosenthal, 1956), pH4-4.
(Proteins were determined according to Lowry, Rosebrough, Farr & Randall,
1951.) In the blanks, the tissue homogenate was replaced by 0-1 ml of 0-25 M
sucrose. The tubes were incubated for one hour at 37 °C and the enzymic
reaction was stopped by the addition of 1 ml of cold 4 % (w/v) trichloroacetic
acid. The tubes were centrifuged for 10 min at 3000 rev/min, and 0-7 ml of the
supernatant fluid was mixed with 0-7 ml of a 5:1 (v/v) mixture of glycine carbonate, pH 10-5 and 0-5 N NaOH. The optical density was determined at 412 nm.
In the arylsulfatase assay, 0-65 ml of 1 M pyridine buffer, pH 5-0 and 0-15 ml
of j9-nitrocatechol sulphate dipotassium salt (Sigma Co., St Louis, Mo.) were
added to tubes containing 400-500 jug of palate proteins. The final molarity of
sucrose in the tubes was 0-25 M and the final volume was 0-93 ml. In the blanks,
the tissue homogenate was replaced by 0-1 ml of 0-25 M sucrose. The tubes were
incubated for 1 h at 37 °C. The reaction was stopped by the addition of 0-25 ml
of cold 8 % (w/v) trichloroacetic acid. The tubes were centrifuged and 0-9 ml
of the supernatant were transferred to tubes containing 0-9 ml of 5 N NaOH.
The optical density was detected at 550 nm.
Histochemistry
Heads of embryos ranging from 1 3 | to 14| days P.C. were fixed in formol
calcium and processed for the histochemical detection of N-acetyl-/?-D-glucosaminidase (Hayashi, 1965).
RESULTS
Table 1 summarizes the temporal development of the morphological changes.
Epithelia
Histogenesis in vivo
The epithelium covering the oral, nasal and medial surfaces of the palatine
shelves was composed, before elevation, of a basal layer of cubic cells and an
outer layer of squamous cells. The cells of the basal layers were actively dividing
in all surfaces. In the oral surface, the cells differentiated to form a stratified
squamous epithelium, while in the nasal surface they became stratified columnar.
The columnar epithelium then produced typical cilia. The epithelial cells of the
opposed medial surfaces came into contact at about 14^ days. The seam thus
formed contained many degenerated cells. The seam subsequently became discontinuous and the mesenchyme of both primordia merged.
Stratified columnar
Pseudostratified
columnar ciliated
Pseudostratified
columnar ciliated
Stratified cuboidal
Stratified cuboidal
Stratified squamous
without stratum
corneum
14+.
15+.
through 18+.
Stratified columnar
ciliated
Pseudostratified
columnar ciliated
Pseudostratified
columnar ciliated
Stratified squamous
Stratified squamous
Stratified squamous
with stratum
corneum (s.c.)
15+.; II*
16+; III*
17+;IV*
18i;V*
(
—
—
—
Intramembranous
Intramembranous
Loose
Loose
Intramembranous
—
Intramembranous
Loose
Loose
Loose
Fused
Double or multilayered
Stratified squamous
and/or stratified
columnar ciliated
Stratified squamous
and/or stratified
columnar ciliated
Stratified squamous
with s.c. and/or
columnar ciliated
Loose
Loose
Loose
Loose
Bone
Intramembranous
Mesenchyme
Connective tissue
Double layered
Double or multilayered
Fused
Fused
Medial surface
* Experimental series number.
Stratified columnar
Stratified cuboidal
In vitro
14+;I*
16+.
Double layered
Double layered
Nasal surface
Double layered
Stratified cuboidal
Oral surface
Epithelium
m
13*
In vivo
Developmental age
(days)
Table 1. Histodifferentiation of the secondary palate
5'
o
i-t
<
a"
Sf
of mo
418
V. IDOYAGA-YARGAS AND OTHERS
Cytodifferentiation of mouse palate in vitro
419
Mesenchyme
The mesenchyme of the palate, before elevation, showed densely packed cells
with relatively scant intercellular substance. During fusion, however, the neighboring cells of the medial area became further separated while the cells of the
dorsolateral portion differentiated into osteocytes. The osteocytes subsequently
underwent ossification which progressively extended toward the midline. The
histogenetic changes in the palate can be followed in Fig. 1 A, C, E, G.
P . , ,.
Histogenesis in vitro
The changes observed in the cells of the oral surface were similar to those
observed in vivo, though they occurred more rapidly. The cells differentiated
into a stratified squamous epithelium, forming a stratum corneum. The cells of
the nasal surface differentiated following the same morphological and temporal
patterns seen in vivo producing a pseudostratified columnar ciliated epithelium.
The histogenesis of the medial surface was followed throughout the culture
periods. In explants of Series I the mesenchyme was covered with epithelium
containing two or more cell layers. However, portions of the palatine shelves of
Series I and the remaining series were not covered by epithelium. The nuclei
of the epithelial cells covering the mesenchyme showed the heterochromatin
homogeneously distributed. Some condensation of the heterochromatic
material was noticed near the nuclear edge. Other nuclei were seen undergoing
mitosis. The cytoplasm of the cells was devoid of vacuoles. Nevertheless, in some
cells cytoplasmic granules with intense basophilia were occasionally found. The
presence of the granules was not confined to the cells of the medial surface.
Karyopycnosis, or signs of cytoplasmic degeneration, was seldom observed in
Series I through V.
FIGURE 1
(A) Cross-section of a 13^ palatine process, showing a loose mesenchyme and a
bi-layered epithelum. x 800.
(B) Oral surface of a Series IV explant, belonging to the same experiment from which
the specimen in (A) was taken. Notice the presence of a stratified squamous
epithelium, x 1250.
(C) Oral surface of a 17^-day-old mouse palate. The epithelium is stratified squamous.
x 1250.
(D) Nasal surface of the same explant of (B). A pseudostratified columnar ciliated
epithelium can be observed, x 1250.
(E) Nasal surface of the same palate of (C), showing a respiratory epithelium,
x 1250.
(F) Intramembranous bone developed in the same explant of (B) and (D). x 1250.
(G) Intramembranous bone present in a region of the same palate of (C) and (E).
x 1250.
420
V. IDOYAGA-VARGAS AND OTHERS
t I
•
Cy to differentiation of mouse palate in vitro
421
The double or multilayered epithelium was replaced in Series II by a stratified
squamous or a pseudostratified columnar ciliated epithelium. Both joined in
many cases at the medial surface. The appearance of the stratum corneum in
Series IV was the last change noticed at the unfused edge.
Mesenchyme
The mesenchymal cells followed the same morphological patterns of differentiation as in vivo. In the dorsolateral areas osteocytic activity was evident in
the explants of Series II. The progression toward the midline was achieved as
in vivo. The rest of the mesenchymal cells proliferated and became progressively
separated from each other. Histogenesis in vitro is illustrated in Fig. IB, D, F;
Fig. 2A; Fig. 3A.
An analysis of Table 1 obtained the following conclusions by comparing the
in vitro and in vivo results, (a) In vitro, the cells of the oral surface differentiated
faster; unlike in vivo, they formed a stratum corneum. (b) The nasal surface
differentiated in a manner similar to that in vivo, (c) At the time when fusion
would normally occur, the epithelial cells at the medial surface of the explants
did not degenerate. In vivo, however, many cells underwent complete autolysis.
(d) After fusion, cornified and/or respiratory epithelia covered the mesenchyme
at the unfused edge. A similar situation was observed in a shelf that was not
fused in vivo (Fig. 3B, C). (e) The mesenchyme underwent similar patterns of
differentiation in vivo and in vitro.
Biochemical observations in vivo and in vitro
In order to achieve the optimal assay conditions for glucosaminidase, the
velocity of hydrolysis as a function of time of incubation and of enzyme concentration was tested. Fig. 4 shows that enzyme activity was linear up to 2 h
incubation, and Fig. 5 illustrates the linearity of the reaction as a function of
different amounts of palate protein. The velocity of the glucosaminidase reaction
appeared to be maximum at pH 4-4 (Fig. 6). An identical pH optimum was
obtained by Shuter, Robins & Freeman (1970) for the same enzyme in the
central nervous system of the rat.
Fig. 7 illustrates the saturation curve of the enzyme with its substrate; an
apparent Km value of 0-96 m.M was calculated from this curve. Idoyaga-Vargas
and Sellinger (1971) had reported a Km value of 0-96 mM for glucosaminidase of
FIGURE 2
(A) Cross-section of a Series I explant showing the presumptive fusing area. Notice
the bi-layered epithelium containing cells devoid of vacuoles or nuclear debris,
x 800.
(B) Area of fusion from a 14^-day-old mouse palate. Arrows indicate vacuolization
and signs of kariolysis in cells of the epithelial seam formed as consequence of fusion,
x 1250.
27
EMB 27
422
V. IDOYAGA-VARGAS AND OTHERS
423
Cy to differentiationof mouse palate in vitro
1-2
10
0-6
-
4
0-5
0-8
6 0-6
6
0-4
0-4
-
^0-3
^
^
d
-
0-2
I
0
0-4
a 0-3
01
o
u
1
120
/
/
1
1
1
1
1
-
0
^ T l
1 1
137 274 411 548
//g of palate protein
1-4
, 1-2
.^^^
^_*
'—°
:f
-/
'10
o 0-8
0-6 •7
0-4 f
0-2
-i
/
f
01
1-8 ~ 7
1-6
i
_
5
0-2
1
//r\ \\
i/
~6
_
d
0-2
I
15 30 45 60
90
Incubation time (min)
06
0-5
I
_
1
1
30 3.4 3-8 4-2 4-6 5 0 5-4 5-8
pH
0
i
i
i
i
i
i
2
4
6
8
10
12
1
HIM, [S]
Figs. 4-7. The kinetics of glucosaminidase activity is illustrated.
(See explanation in the text.)
glial cells isolated from the cerebral cortex of the rat. Table 2 shows the results
obtained with glucosaminidase and arylsulfatase as a function of age. The
specific activity of glucosaminidase increased progressively up to a value of
8-68 in vivo and 12-67 in vitro. The difference between the ages was under all
conditions statistically significant, with P-values ranging between 001 and
0-005. The age dependency of arylsulfatase specific activity values was slightly
different. However, these differences were not statistically significant except
between the in vitro 48- and 120-h values.
Examination of the ratios of the specific activity values of glucosaminidase
FIGURE 3
(A) Presumptive fusing area from a Series IV explant. Pseudostratified columnar
ciliated and stratified squamous epithelia joined without transition, x 800.
(B) Cross-section of a 17^-day-old mouse embryo showing the contact between the
nasal septum and the secondary palatine processes. Notice that the left side is fused,
while the right is unfused. Stratified squamous epithelia cover the lower portion of the
nasal septum and the oral and medial surfaces of the palate. Arrow indicates where
the respiratory epithelium is present in the nasal surface, x 310.
(C) The region indicated by the arrow in (B) is seen at a higher magnification.
Stratified squamous epithelium meets with pseudostratified ciliated epithelium
without transition. Arrow indicates the presence of cilia, x 1250.
424
V. IDOYAGA-VARGAS AND OTHERS
Table 2. Specific activities of glucosaminidase and arylsulfatase
in palatal total homogenate as function of age
Enzyme
In vivo
days
Glucosaminidase
Arylsulfatase
2-7110-33
1-73 ±006
14^ days
17^-days
3-86 ±0-26
2-10 ±0-24
8-68 ±0-70
2-32 ±0-37
In vitro
24 h
Glucosaminidase
Arylsulfatase
3-69 ±0-84
2-93 ±0-63
120 h
48 h
6-26 ± 1 07 12-67 ± 2 1 0
3-35 ±0-85 5-56±0-74
Values are expressed as total units of enzyme activity per mg of palate protein/h. Each
value is given with the corresponding standard errors of means.
Fusion
100
In vitro
80
o e
In viro
60
40
20
13-5
14-5
17-5
Total developmental age (days)
Fig. 8. Glucosaminidase specific activity from palatal total homogenates, in vivo
and in vitro, plotted as a function of total developmental age. The specific activity
is expressed as micromoles of ^-nitrophenol liberated/mg protein/h.
and arylsulfatase at the different age points revealed an upward trend with
increasing age, suggesting a preferential increase of glucosaminidase activity
with age.
The relationship between the activity of glucosaminidase and the process of
fusion is depicted in Fig. 8. There was an increase in the specific activity of the
hydrolase both in vivo and in vitro during fusion. The highest specific activity
was noted, however, three days after closure of the palate.
Cytodifferentiation of mouse palate in vitro
425
Histochemical observation in vivo
When 13-i-day-old embryo heads were sectioned for the histochemical
detection of glucosaminidase activity in the region of the secondary palate,
virtually no activity was observed. However, at 14^ days a reaction was noticed
in the area of fusion. Fig. 9 A illustrates the localization of the enzyme activity
in the triangular epithelial area formed as a consequence of fusion. The cells that
are being expelled from the seam also appear to show enzyme activity (Fig. 9B).
However, the fusion zone was not the only area within the palate where activity
could be detected, since the lateral portions, presumably containing macrophages, also stained (Fig. 9C).
The above histochemical evidence strongly suggests that glucosaminidase
activity and hence lysosomes are closely related to cellular degeneration
occurring during fusion of the secondary palate.
DISCUSSION
Differentiation of embryonic tissues into stratified squamous epithelium
(Soriano, Saxen, Vainio & Toivonen, 1964), respiratory epithelium (Aydelotte,
1963), and bone without a previous cartilage template (Fell & Robinson, 1930),
were all demonstrated to occur in vitro. In the present investigation, cells of the
explanted palatine shelves differentiated, forming cornified and respiratory
epithelia, as well as intramembranous bone. The attention, however, was focused
on the differentiation of the epithelial cells at the medial surface. Pourtois (1968)
has described a stratification of the outer layer of the epithelium occurring in the
marginal region of the palatine shelves. Analogous transformations were
observed in some of our explants. This change may be attributed to an increased
adhesiveness of the peridermal cells at the medial surface. Moreover, epitheliomesenchymal recombinations supported the conclusion that the differential
adhesiveness is induced by the underlying mesenchyme (Pourtois, 1969). However, the adhesivity of the peridermal cells maybe a general property shared by
other types of epithelia at a specific time of their cytomorphosis. This has been
indicated by the in vitro fusion of epithelial surfaces of different embryonic
origins (Vargas, 1968; Pourtois, 1969; Goss, Bodner & Avery, 1970; Bodner,
Goss & Avery, 1970).
Greulich (1964) reported that the basal cells of the epidermis give rise to two
daughter cells. These may retain the characteristics of basal cells for an indeterminate period of time after cell division. For each cell division, however, another
basal cell migrates and differentiates, to be subsequently eliminated after degeneration. In this way the steady state of the epithelium is maintained. If this
applies to the palatine epithelium, one may be able to find cells degenerating
before fusion on the nasal, oral and medial surfaces. Such degenerative cells have
been reported in these three surfaces (Sweeney & Shapiro, 1970). Thus, in order
426
V. IDOYAGA-VARGAS AND OTHERS
Cytodifferentiation of mouse palate in vitro
427
to critically test the assumption that complete epithelial autolysis is a necessary
consequence of differentiation before palate closure, it must be ascertained
whether there is an increase in cell degeneration at the medial surface of the
palate before fusion.
The literature reveals studies of cells undergoing degeneration before fusion
(Smiley, 1970; Sweeney & Shapiro, 1970), as well as evidence of lysosomal
activity in the palatine epithelium (Waterhouse & Squier, 1966; Andersen &
Matthiessen, 1966; Mato, Aikawa & Katahira, 1966, 1967; Angelici and
Pourtois, 1968; Farbman, 1969; Hayward, 1969). However, a quantitative
demonstration of an increased number of degenerating cells at the fusing edge
has not been made. Other embryonic systems in which cell degeneration is
accompanied by increased lysosomal activity have been described (Weber, 1963).
Therefore, lysosomal activity may serve in certain circumstances as an indication
of cell degeneration. If cell degeneration is occurring at a higher rate before than
during fusion, then a simultaneous increase in lysosomal activity should be
expected. However, in the present work, the specific activity of the lysosomal
enzyme, glucosaminidase, was lower before than during fusion and was highest
after fusion. The finding of a higher specific activity during fusion than prior
to it suggests the degeneration of a higher number of cells during the fusion
process per se. That the enzyme is involved in cell degeneration is indicated by its
histochemical localization in the degenerating cells at the zone of fusion during
the closure of the palate. Some activity, however, was also evident in the epithelial cells throughout the surfaces and in the mesenchymal cells. The increase in
glucosaminidase activity may not indicate only degeneration at the medial
surface during fusion. The enzyme may be involved in other hydrolytic processes
taking place in the palate. For increase in glucosaminidase specific activity in
palate explants, during the equivalent time of fusion in vivo, is not accompanied by
morphological signs of cell degeneration at the medial surface. The occurrence
of these processes, that are apparently independent of fusion, may account in
part for the increase in specific activity of the enzyme in the explants during the
actual closure. In addition, both in vivo and in vitro, the specific activity of
glucosaminidase reached the highest value when morphological differentiation
was achieved. The cells of the oral mucosa lose organelles such as mitochondria
FIGURE 9
(A) Cross-section of the fusing area from a 14^-day-old mouse embryo. Observe
the localization of glucosaminidase activity in the midline. The reaction product is
present in the triangular area and in a group of cells below it. They represent the
remnants of the epithelial seam formed during fusion, x 800.
(B) Cross-section of the fused area from a 14i-day-old mouse embryo. In the midline cells containing the reaction product are being expelled from the palate, x 800.
(C) Mesenchymal cells of the dorsolateral portions of the palate from a 14^-day-old
embryo show localization of glucosaminidase activity, x 800.
428
Y. IDOYAGA-VARGAS AND OTHERS
during keratinization (Rogers, 1964). It is possible that the lysis of organelles is
mediated by lysosomes. Thus, glucosaminidase, a lysosomal enzyme located in
palatal oral epithelium, may be involved in the lysis of organelles. The increase
in glucosaminidase activity observed when differentiation of the oral mucosa
occurs would then partly reflect the higher lysosomal activity in the cells undergoing keratinization.
During fusion a morphological comparison was made between the presumptive
and actual fusing areas. This was done under three different conditions, in vitro,
in vivo and both. In the first, isolated rat palatine shelves, cultured up to the
equivalent time of fusion in vivo, showed no evidence of spontaneous epithelial
breakdown. However, when cultured with the medial surface or a Millipore
filter in contact the epithelium was destroyed (Shapiro, personal communication). The second revealed similar results when electron microscopical examination was made of the same respective areas in vivo (Mato et al. 1966, 1967;
Farbman, 1968, 1969; De Angelis & Nalbandian, 1968; Hayward, 1969; Smiley,
1970). Finally in the third, the epithelial cells of the seam in vivo were compared
with the presumptive fusing cells in vitro. This comparison, made in the present
studies, showed that while many cells undergo degeneration in vivo, few if any
presented clear signs of degeneration in vitro. These morphological observations
are compatible with the assumption that fusion is a necessary condition for
degeneration to occur. Although partial autolysis may indicate differentiation of
the cells at the medial surface (Angelici & Pourtois, 1968), fusion appears
indispensable for their complete lysis. Consequently, if fusion is prevented, the
fate of the cells can be altered. The presumptive fusing cells may develop
different morphological characteristics or become part of the respiratory and/or
Malpighian-like epithelia. The joining of both epithelial types without transition
supports the latter alternative. If this is the case, it may not be an isolated
developmental event. For an analogous situation develops during the morphogenesis of the wing bud in the chick embryo. Cells of the posterior necrotic zone
are genetically determined to die. A 'death clock' is set at stage 17. However, the
cells can be diverted to a new developmental fate if they are grafted before stage
22 to the dorsal side of the wing bud. There they may develop cartilage or form
extra feathers (Saunders, 1966).
In conclusion, the morphological, biochemical and histochemical findings of
the present investigation are consistent with the concept that fusion is a necessary
condition for the complete autolysis of prospective fusing cells. The cells may
alter their developmental fate to form Malpighian- or respiratory-like epithelia if
fusion is prevented.
The authors wish to thank Dr O. Z. Sellinger for reading the manuscript and making
useful comments, and Mrs P. Petiet for her technical assistance. The critical discussions of
Dr R. Narbaitz in the early part of this work are also gratefully acknowledged. This work was
supported in part by Grant MH-07417, and one of us (V. I.-V.) was a postdoctoral trainee of
the U.S. Public Health Service.
Cytodifferentiation of mouse palate in vitro
429
REFERENCES
ANDERSEN, J. & MATTHIESSEN, M. E. (1966). The histiocyte in human foetal tissues: its
morphology, chemistry, origin, function and fate. Z. Zellforsch. microsk. Anat. 72, 193-211.
ANGELICA, D. & POURTOIS, M. (1968). The role of acid phosphatase in the fusion of the
secondary palate. /. Embryol. exp. Morph. 20, 15-23.
AYDELOTTE, M. B. (1963). The effects of Vitamin A and citral on epithelial differentiation in
vitro. 1. The chick tracheal epithelium. /. Embryol. exp. Morph. 11, 279-291.
BODNER, J. W., Goss, A. N. & AVERY, J. K. (1970). Fusion capability of rat embryonic extraoral tissue in vitro. Archs oral Biol. 15, 861-868.
DEANGELIS, V. & NALBANDIAN, J. (1968). Ultrastructure of mouse and rat palatal processes
prior to and during secondary palate formation. Archs oral Biol. 13, 601-608.
ELVING, J. P., MARKOWITZ, J. M. & ROSENTHAL, I. (1956). Preparation of buffers of constant
ionic strength. Analyt. Chem. 28, 1179-1180.
EVANS, V. J., BRIANT, F. C, FIORAMONTI, M. C, MCQUILKIN, W. T., SANDFORD, K. K. &
EARLE, W. R. (1956). Studies on nutrient media for tissue cells in vitro. 1. A protein-free
chemically defined medium for cultivation of strain L cells. Cancer Res. 16, 77-94.
A. I. (1968). Electron microscope study of palate fusion in mouse embryo. Devi
Biol. 18, 93-116.
FARBMAN, A. I. (1969). The epithelium-connective interface during closure of the secondary
palate in rodent embryos. /. dent. Res. 48, 617-624.
FELL, H. B. & ROBINSON (1930). The development and phosphatase activity in vivo and in vitro
of the manidibular skeletal tissue of the embryonic fowl. Biochem. J. 24, 1905-1921.
ERASER, F. C. (1967). Cleft lip and cleft palate. Science, N.Y. 158, 1603-1606.
Goss, A. N., BODNER, J. W. & AVERY, J. K. (1970). Fusion capability of rat embryonic oral
epithelium in vitro. Archs oral Biol. 15, 797-804.
GREULICH, R. C. (1964). Aspects of cell individuality in the renewal of stratified squamous
epithelia. In The Epidermis (ed. W. Montagna & W. C. Lobitz, jun.), pp. 117-135. New
York: Academic Press.
HAYASHI, M. (1965). Histochemical demonstration of N-acetyl-/?-glucosaminidase employing
Naphthol AS-BI N-acetyl-/?-glucosaminide as substrate. /. Histochem. Cytochem. 13,
355-360.
HAYWARD, A. F. (1969). Ultrastructural changes in the epithelium during fusion of the
palatine processes in rat. Archs oral Biol. 14, 661-678.
IDOYAGA-VARGAS, V., & SELLINGER, O. Z. (1971). Lysosomal glycosidases in isolated nerve
cell bodies (NCB) and glial cells (GC). Trans. Amer. Soc. Neurochem. 2, 84.
LOWRY, O. H., ROSEBROUGH, N. J., FARR, A. L. & RANDALL, R. J. (1951). Protein measurement with the Folin phenol reagent. /. biol. Chem. 193, 265-275.
MATO, M., AIKAWA, E. & KATAHIRA, M. (1966). Appearance of various types of lysosomes
in the epithelium covering lateral palatine shelves during secondary palate formation.
Gunma J. Med. Sci. 15, 46-56.
MATO, M., AIKAWA, E. & KATAHIRA, M. (1967). Alteration offinestructure of the epithelium
on the lateral palatine shelves during the secondary palate formation. Gunma J. Med. Sci.
16, 79-99.
POURTOIS, M. (1966). Onset of the acquired potentiality for fusion in the palatal shelves of rats.
/. Embryol. exp. Morph. 16, 171-182.
POURTOIS, M. (1968). La fusion des cretes palatines et son alteration par quelques agents
teratogenes. Archs Biol. (Liege), 78, 1-75.
POURTOIS, M. (1969). Study on the fusion of palatal processes in vitro and in vivo. J. dent. Res.
48, 1135.
ROGERS, G. E. (.1964). Structural and biochemical features of the hair follicle. In The Epidermis
(ed. W. Montagna & W. C. Lobitz, jun.), pp. 179-236. New York: Academic Press.
SALZGEBER, B. & WEBER, R. (1966). La regression du mesonephros chez l'embryon de poulet.
Etude des activites de la phosphatase acide et des cathepsines. Analyse biochimique,
histochimique et observations au microscope electronique. /. Embryol. exp. Morph 15,
397-419.
FARBMAN,
27-2
430
V. IDOYAGA-VARGAS AND OTHERS
SAUNDERS, J. W. (1966). Death in embryonic systems. Science, N.Y. 154, 604-612.
SCHEIB-PFLEGER, D. & WATTIAUX, R. (1962). Etude des hydrolases acides des canaux de Muller
de l'embryon de poulet. I. Activites totales et solubles des canaux d'embryons de 8 a 10
jours d'incubation. Devi Biol. 5, 205-217.
SHUTER, E., ROBINS, E. & FREEMAN, M. L. (1970). /?-Hexosaminidases in the nervous system:
the quantitative histochemistry of /?-glucosaminidase and /#-galactosaminidase in the
cerebellar cortex and subjacent white matter. /. Histochem. Cytochem. 18, 271-277.
SMILEY, G. R. (1970). Fine structure of mouse embyronic palatal epithelium prior to and after
midline fusion. Archs oral Biol. 15, 287-296.
SORIANO, L., SAXEN, L., VAINIO, T. & TOIVONEN, S. (1964). The development of the oesophageal
and tracheobronchial epithelia in vitro. Ada anat. 57, 105-114.
SWEENEY, L. R. & SHAPIRO, B. L. (1970). Histogenesis of Swiss white mouse secondary palate
from the nine and one half days to fifteen and one half days in utero. I. Epithelial-mesenchymal relationships: light and electron microscopy. /. Morph. 130, 435-450.
VARGAS, V. I. (1967). Palatal fusion in vitro in the mouse. Archs oral Biol. 12, 1283-1288.
VARGAS, V. I. (1968). Fusion of the palatine shelves with heterotypic explants in the mouse.
Archs oral Biol. 13, 845-848.
WALKER, B. E. & FRASER, F. C. (1956). Closure of the secondary palate in three strains of
mice. /. Embryo!, exp. Morph. 4, 176-189.
WATERHOUSE, J. P. & SQUIER, C. A. (1966). Form and function of lysosomes in oral epithelium.
Nature, Lond. 211, 420-421.
WEBER, R. (1963). Behavior and Properties of Acid Hydrolases in Regressing Tails of Tadpoles
during Spontaneous and Induced Metamorphosis in vitro. In Lysosomes (Ciba Fdn Symp.),
pp. 282-310. London: Churchill.
(Manuscript received 23 June 1971)