/. Embryol. exp. Morph. Vol. 34, 2, pp. 485-495, 1975
Printed in Great Britain
485
A new in vitro system for studying secondary
palate development
By L. BRINKLEY, 1 G. BASEHOAR, 1
A. BRANCH 1 AND J. AVERY 1
From the Department of Oral Biology,
Dental School, The University of Michigan
SUMMARY
An in vitro system was devised which supports palate development in partially dissected
embryonic mouse heads. The heads were suspended in the culture chamber so that they
were not held in a fixed orientation and were constantly surrounded with a fluid medium.
Under these circumstances the developing palate must effect closure without the aid of
gravitational forces. The culture medium was constantly circulated, gassed with 95 % O2,
5 % CO2 using hollow fiber gas permeation devices, and kept at 34 °C. Swiss-Webster
mouse embryos of 12 days 12-18 h (ca. 48 h prior to expected in vivo closure) or 13 days
8-14 h (ca. 24 h prior to closure) were used to test the ability of the system to support
palatal development. Embryonic heads were dissected in one of two ways before culture:
brain and tongue removed, or brain, tongue and mandible removed. After 24 h in culture,
preparations of either age with only the brain and tongue removed had made substantially
greater progress than their counterparts with the brain, tongue and mandible removed.
With only the brain and tongue removed, the palatal shelves were contacting, adhered or
fused in 67 % of the older embryos, whereas most of the embryos of the same age cultured
with the brain, tongue and mandible removed had shelves that were not fully elevated and
still separated by a moderate gap. Thus for maximal progress in the present system, the oral
cavity must be intact except for the tongue.
INTRODUCTION
Secondary palate development in the embryo normally takes place in a wellaerated, fluid environment. The secondary palate is formed by the union of two
shelves of tissue which arise from the maxillary processes and initially grow
down to a position on either side of the tongue. The shelves then move to the
horizontal plane above the tongue, where fusion of the two shelves to each
other at the midline and to the nasal septum on their superior surface will
complete the normal form of the palate. In vitro methods offer promise for
direct experimentation on this process, but their potential has not yet been
realized. Standard organ culture methods used for studies of palatal development involve a considerable departure from the in vivo situation. Explants
1
Authors'" address: Department of Oral Biology, Dental School, The University of Michigan, Ann Arbor, Michigan 48104, U.S.A.
486 L. BRINKLEY, G. BASEHOAR, A. BRANCH AND J. AVERY
ranging from the middle one-third of the embryonic rodent face, containing
the maxillary vault, to pairs of excised palatal shelves have been either placed in
Trowell-type surface organ culture (Trowell, 1954), in which the explant is placed
on a metal grid or other support at the gas-fluid interface (Myers, Petrakis
& Lee, 1967; Pourtois, 1966; Lahti, Antila & Saxen, 1972), or anchored in
a nutrient agar or plasma clot which may be covered with medium (Moriarty,
Weinstein & Gibson, 1963; Reeve, Porter & Lefkowitz, 1966; Chaudhry &
Sian, 1967; Goss, Bodner & Avery, 1970a, b), or simply placed inaPetridish
and submerged in medium (Thompson & Schweisthal, 1969). With these
culture techniques it has not been possible to study palate development in an
intact oral cavity; the explants were held in a fixed orientation and had unequal
access to oxygen and nutrients (Fainstat, 1972).
Culture of whole somite-stage rodent embryos has demonstrated the importance of circulation and gassing to achieve growth in such large preparations
(New, 1967; New & Daniel, 1969; Robkin, Shepard & Tanimura, 1972;
Cockroft, 1973). An in vitro system has therefore been devised which supports
palatal development in partially dissected embryonic mouse heads suspended
in gassed, circulating medium so that they are not in a fixed position, but are
surrounded and partially supported by fluid. This method permits observation
of and experimentation on the development of palatal shelves for at least 24 h,
beginning 48 h before expected in vivo closure. It also allows study of the
effects of various craniofacial structures on palatal development.
MATERIALS AND METHODS
The culture apparatus is shown in Fig. 1A. Individual components and
their operations are described below.
Culture chamber. The culture chamber consisted of a modified polycarbonate
40 ml beaker, with a removable lid so that samples could be added or removed
while the system was running. Stainless-steel wires were attached across one
end of the beaker with Silastic Medical Adhesive Type A (Dow Corning Corp.,
Medical Products Division, Midland, Michigan). Embryonic heads were suspended in the fluid-filled beaker by placing one end of a stainless steel S-shaped
hook through the foramen magnum of the head and the other end over the
wire. Fig. 1B illustrates the chamber and method of hanging the preparations.
The heads were almost totally supported by the medium, and were able to
move freely with the current as the medium was pumped through the chamber.
Hollow fiber devices. The medium was oxygenated by using silicone copolymer
hollow fiber devices (Bio-fiber 5 minitubes, Bio-Rad Laboratories, Richmond,
Calif.). The minitubes contained hollow fibers with a surface area of 50 cm2,
a fiber volume of 0-3 ml and a jacket volume of 2-5 ml. Two tubes were linked
in sequence and medium was pumped through the fibers while a gas mixture of
95 % O2, 5 % CO2 was run through the jacket. Since gas exchange takes place
487
In vitro palate development
(b)
Fig. 1. Schematic drawing of components of the culture apparatus and method of
suspending culture preparations, (a) Culture apparatus. The components are:
culture chamber (A), two hollow fiber devices connected in sequence (B), gas
source (C), tubing pump (D) and tele-thermometer with attached thermistor (E).
Medium is continuously pumped through the hollow fibers and chamber while gas
is run into the jackets surrounding the hollow fibers, (b). Method of suspending
preparations in the culture chamber.
across the fibers, the medium could be highly oxygenated without bubbling,
which could cause denaturation of the serum component.
Pump. Medium was circulated through the system by a variable speed
Masterflex-tubing pump equipped with a 7013 pump head (Cole-Parmer,
Chicago, 111.). The system was connected throughout with Tygon 'sterile'
tubing, which is relatively impermeable to oxygen.
Thermistor. A fine probe (Yellow Springs Instruments, series 500, Yellow
Springs, Ohio) was inserted into the system and attached to a remote reading
tele-thermometer (YSI, Yellow Springs, Ohio). The temperature of the medium
could be monitored throughout the course of the experiment.
The pH, pOa and pCOa of the medium was assessed with a 1L Model 113
blood gas analyzer. A sample of medium was removed at the beginning of an
experiment, after 1 h for equilibration, and at the termination of the experiment.
Incubator. The beaker, hollow fiber devices and almost all connecting tubing
resided in a full-view Plexiglass incubator (Precision Scientific, Chicago, 111.).
The pump was placed outside the incubator to allow maximum cooling of the
pump motor. Connecting tubing entered the incubator through a port designed
by the manufacturer for that purpose.
31
E MB 34
488 L. BRINKLEY, G. BASEHOAR, A. BRANCH AND J. AVERY
Maintenance of the apparatus. The culture chamber and hollow fiber minitubes
were rinsed with sterile saline after each experiment. The chamber was sterilized
with either ethylene oxide or 70% ethanol. The hollow fiber minitubes were
stored in 1-5% formalin (as specified by the manufacturer). Before use the
minitubes were flushed with 3 1 of sterile saline. If the chamber and tubing
had been chemically sterilized, they were also flushed with 3 1 of sterile saline.
The hollow fibers must be continually wet as they break on drying. Chemical
sterilization of the fibers must be used as the fibers will not withstand temperatures or agents used for dry heat, autoclave or gas sterilization. New gassterilized Tygon tubing was used for each experiment. A minimum of 3 days
was allowed between gas sterilization and tubing use.
If contamination occurred during an experiment the fibers were filled with
a pronase solution (2 mg/ml pronase in 0-005 M phosphate buffer, with 1-86
mg/ml EDTA and 0-088 mg/ml L-cysteine HC1 hydrate, pH 6-5) and left
overnight. The fibers were then flushed with sterile saline prior to use.
Culture conditions. The culture medium consisted of 50% supplemented
Eagle's Basal Medium (BME) and 50 % heat-inactivated fetal calf serum with
100 i.u./ml penicillin, and 100 /*g/ml streptomycin. The BME was supplemented
with 1 x MEM amino acids, 2 x MEM vitamins, 0-6 mg/ml glutamine and
50 /Ag/ml sodium ascorbate. Medium, serum and supplements were purchased
from Grand Island Biological Co. (Grand Island, N.Y.).
The medium was circulated at a flow rate of 36 ml/min and gassed with
5 % CO2, 95 % O2 at a flow rate of 50 ml/min, pOa of the medium was 430460 mm, p c02 was 28-30 mm and pH was 7-3. The temperature was maintained at 34 + 0-5 °C, as this gave better results than 37 °C.
Preparation of explants. Swiss-Webster mice at known stages of pregnancy
were obtained from Spartan Research (E. Lansing, Michigan). Estimates of
the time of fertilization of these commercially supplied embryos were probably
accurate to only ± 8 h. The age of each embryo was therefore estimated by the
morphological condition of five parts of the embryo: forefeet, hindfeet, ears,
hair follicles and eyes. Each embryo was assigned a morphological rating
using the system of Griineberg as modified by Walker (Walker & Crain, 1960),
which assigns a numerical value for each developmental stage of these features.
When time of development is given in days and hours, day 0 is the day the
vaginal plug is found and fertilization is assumed to occur at 2 a.m. on the
morning of day 0.
The mice were killed by cervical dislocation, and all subsequent operations
were carried out under an Edgegard laminar flow hood (The Baker Co., Sanford,
Maine) using sterile technique. The abdominal area was flooded with 70%
ethanol, the skin removed and the uterine horns and their contents dissected
and placed in warmed, gassed medium. The embryos were then rapidly removed
from the uterus and their amniotic sacs and placed in fresh, warmed, gassed
medium. Heads of embryos were removed and dissected in one of two ways:
In vitro palate development
489
either the brain and tongue (BT) or the brain, tongue and mandible (BTM)
were removed. A single circumferential cut is made just above the eyes, and
the brain carefully removed to reduce the volume of tissue to be cultured. The
tongue was removed by severing it at the base with microscissors. As a part of
this operation a small vent was cut from the area of the base of the tongue to
the outside to allow greater medium circulation in the oral cavity. If the tongue
and mandible were to be removed, a cut was made at the corner of the mouth
and the entire mandible and attached tongue removed. The BT preparations
had an essentially intact oral cavity with the maxillary-mandibular relationship
as in vivo. The BTM explants were the middle one-third of the mouse face and
were approximately the preparations used for Trowell-type organ culture by
other investigators (Moriarty, 1963; Saxen, 1966; Lahti et al. 1972). In each
litter of mice used, two were examined and fixed immediately for controls and
the remainder divided into two groups, BT and BTM. None of the embryos
showed any spontaneous palatal shelf movement when the tongue was removed.
The palatal regions of cultured and uncultured heads were photographed
with a Zeiss Icarex camera mounted on a Zeiss dissecting microscope. All
tissues were submerged while being photographed. The length of the palatal
shelves and the width of the gap between them was measured using an ocular
micrometer and verified by measurements from photographs. Samples of control
and cultured heads were fixed in Bouin's fixative, dehydrated, embedded in
Paraplast and sectioned at 10 /.im. Sections were stained with hematoxylin
and eosin and examined or photographed with a Zeiss photoscope.
RESULTS
Dissected heads of embryos with morphological ratings (MR) 1-2 (ca. 12
days 12-18 h gestation) and MR 5-6 (ca. 13 day 8-14 h gestation) were cultured
for 24 h. Palatal closure in the Swiss-Webster embryos used for these experiments is usually accomplished by MR 12 (ca. 14 days 8-16 h gestation).
Features typical of each stage of embryo used are shown in Fig. 2. The stages
of palatal development attained after culture are shown in Fig. 3.
MR 1-2. Sixty-three preparations were cultured, 40 with the brain and tongue
(BT) removed, and 23 with the brain, tongue and mandible removed (BTM).
After culture, the state of the palatal shelves of BT preparations ranged from
vertical in the posterior half to fully horizontal along their entire length. No
BT preparation displayed totally vertical shelves at the end of the culture
period. The gap between the shelves was small and some heads showed contact
(10%), adhesion (5%) or fusion (2-5%). The BTM heads showed much less
progress, with 80 % having totally vertical shelves or shelves which were vertical
in the posterior half. No BTM preparation showed fully horizontal shelves at
the end of the culture period. There was usually a gap between the shelves,
with only one preparation showing adhesion at the point of contact.
31-2
4 9 0 L. BRINKLEY, G. BASEHOAR, A. BRANCH AND J. AVERY
Fig. 2. Photographs of the palatal regions of uncultured heads and their littermates cultured 24 h. (A-C) Morphological rating 1-2, litter-mates. (A) Uncultured
preparation, (B) tongue, brain and mandible removed prior to culture, (C) brain
and tongue removed prior to culture. (D-F) Morphological rating 5-6, litter-mates.
(D) Uncultured preparation, (E) brain, tongue and mandible removed prior to
culture, (F) brain and tongue removed prior to culture. Magnification, x 13.
In vitro palate development
Shelf form
Morphological rating
1-2 (ca. 12 day/12 h)
Brain
491
Width/length ratio
100%
and
tongue
removed
(/V=40)
Brain, tongue
and mandible
removed (/V=23)
Morphological rating
5-6 (ca. 13 day/12 h) Brain and
tongue
removed
(JV=31)
Brain, tongue
and mandible
removed (N= 12)
Palate development in dissected preparations cultured 24 h
Fig. 3. Palate development in dissected preparations cultured 24 h. The height of
each block represents the percentage of embryos attaining a given developmental
stage.
MR 5-6. BT preparations of MR 5-6 embryos also made substantially greater
progress than did BTM preparations. Palatal shelves of 31 BT dissections were
all either mainly or fully horizontal at the end of the culture period, and were
very close (33 %), touching (16 %), adhered (12 %) or fused (39 %). In contrast,
the shelves of the 12 BTM dissections ranged from half to fully horizontal
and had gaps between them. No preparations were touching, adhered or fused.
Considering both shelf form and width-to-length ratio, palatal development
in culture was greater in BT preparations than in BTM preparations at either
stage of development.
Explants were also examined histologically. The general condition of the
tissue was good. The interior areas showed some degenerative changes, especially the areas immediately surrounding the cartilaginous portion of the nasal
septum, and the tissue immediately lateral to the nasal septum. Cartilaginous
tissue (e.g. the nasal capsule and Meckel's cartilage) appeared to do well in
culture. BTM preparations showed fewer degenerative changes than did BT
ones (Fig. 4), although BT preparations made better progress.
4 9 2 L. BRINKLEY, G. BASEHOAR, A. BRANCH AND J. AVERY
Fig. 4. Uncultured preparation of morphological rating 1-2; and litter-mates
cultured for 24 h. (A) Uncultured preparation with brain, tongue, and mandible
removed immediately prior to fixation, (B) brain, tongue and mandible removed
prior to culture, (C) brain and tongue removed prior to culture, (p = palatal
shelf, ns = nasal system). Sections are 10 /im thick, stained with hematoxylin and
eosin. Magnification, x 22.
DISCUSSION
These results demonstrate that palatal development can take place in vitro
in embryonic mouse heads with essentially intact oral cavities. Earlier organ
culture systems required that the explant be placed palate upwards, in a static
situation, so that the palatal shelves often fell inward, towards each other
(Reeve et ah 1966); this makes it difficult or impossible to study mechanisms
of shelf movement. The present system is an improvement over previous in
vitro methods used to study palatal development as it more closely simulates
conditions in utero. The palatal shelves are in normal relationship to other
In vitro palate development
493
structures in the oral cavity and must effect closure against gravity while submerged in a dynamic fluid environment. The survival and development of such
large preparations must in part be attributed to the high oxygen tension in the
medium (Cockcroft, 1973) achieved by using the hollow fiber devices. However,
circulation of the medium and lowered culture temperature are also important.
It has been estimated that palatal elevation in mice requires approximately
3 h in vivo (Walker & Fraser, 1956). In culture, the sequence of elevation appears
to occur prematurely, though more gradually. BT preparations of embryos of
MR 1-2 (12 days 12-18 h) have elevated shelves after 24 h in culture. In vivo
elevation will not take place until 14 days 8-16 h. Since none of the embryos
dissected for culture showed spontaneous shelf movement upon tongue removal, any shelf changes or force necessary for elevation must have developed
during the culture period. Information necessary for shelf elevation must
therefore be present or able to be elicited in embryos 48 h before closure.
Shelf fusion occurred more rapidly than in other in vitro systems: shelves of
half the preparations taken at MR 5-6 were adhered or fused after 24 h, while
shelves cultured on a Millipore filter in Trowell-type organ culture require
48-72 h to adhere and fuse (Pourtois, 1966; Myers et al 1967).
Whether or not the oral cavity is essentially intact also influences the fate
of the culture preparations. Previous work on in vitro elevation has used only
preparations with the brain, tongue and mandible removed, and most explants
have been placed in culture within 24 h of expected closure. In the present
system, heads with an oral cavity intact except for the tongue made better
progress than those with the mandible removed. The presence of intact cheek
tissues may stabilize both vertical and horizontal dimensions of the oral cavity
within which the palatal shelves elevate.
Since no blood supply exists in the dissected heads, nutrients, gas and waste
exchange must take place by diffusion; circulation of the medium is therefore
of crucial importance. After 24 h in culture, areas with limited access to the
medium began to show some degenerative changes (Fig. 4). Preparations in
Trowell-type culture require several days for shelf horizontalization or fusion,
and often show extensive degeneration and necrosis (Thompson & Schweisthal,
1969).
In the present system, the events occur over a relatively short period of
time. Problems of severe tissue degeneration are therefore avoided and tissue
vitality is excellent. Thus not only the effects of various craniofacial structures
on elevation but also the cellular activities associated with development and
elevation of the palatal shelves can be investigated.
494 L. BRINKLEY, G. BASEHOAR, A. BRANCH AND J. AVERY
ZUSAMMENFASSUNG
Es wurde ein in vitro System entwickelt, welches die Gaumenentwicklung im teilweise
sezierten embryonalen Mauskopf ermoglicht. Die Praparate werden in eine Kulturkammer
eingetaucht, so dass sie in keiner fixierten Richtung gehalten und standig von fliissigem
Medium umgeben werden. Auf diese Weise muss sich der sich entwickelnde Gaumen schliessen, ohne dass die Gravitationskrafte eine Wirkung haben konnen. Das Kulturmedium
zirkuliert standig und wird mittles eines Hohlfaservergasers mit 95 % O2 und 5 % CO2
begast und auf 34 °C gehalten. Das System wurde mit Swiss-Webster Mausembryonen
zweier Tragzeiten, namlich 12 Tage und 12-18 Stunden (ca. 48 Stunden vor der erwarteten
in vivo Schliessung des Gaumens) und 13 Tage und 8-14 Stunden (ca. 24 Stunden vor
Schliessung) auf seine Fahigkeit, die Schliessung des Gaumens zu ermoglichen, getestet.
Die embryonalen Kopfe wurden auf zwei Weisen seziert: Entweder wurden Gehirn und
Zunge oder Gehirn, Zunge und Mandibula entfernt. Nach 24 stiindiger Kultur waren die
Praparate beider Alter, bei welchen nur Gehirn und Zunge entfernt worden waren bedeutend
weiter fortgeschritten im Vergleich zu ihren Gegenpartnern, bei welchen Gehirn, Zunge und
Mandibula entfernt worden waren. Die Gaemenplatten der alteren Embryonen, denen nur
Gehirn und Zunge entfernt worder waren, beruhrten sich, verklebten oder verschmelzten
in 67 % der Falle nach 24 stiindiger Kultur. Im Gegensatz dazu zeigte ein Grossteil der
Embryonen gleichen Alters nach 24 stiindiger Kultur ohne Gehirn, Zunge und Mandibula
keine vollstandige Elevation der Gaumenplatten, aber wiesen eine ziemliche Splate zwischen
diesen auf. Im vorliegenden System wird eine Mundhohle ohne Zunge fur maximale Entwicklung benotigt.
We wish to express our appreciation to Dow Chemical Co., Midland, Michigan, for their
generosity in donating various hollow fiber devices for use in developing the system.
This work was supported by U.S.P.H.S. Grant DE 02774-07 from the National Institutes
of Health, Bethesda, Maryland.
REFERENCES
CHAUDHRY, A. & SIAN, S. (1967). In vitro study of fusion of palatal shelves in A/Jax mouse
embryos. /. dent. Res. 46, 257-260.
COCKCROFT, D. (1973). Development in culture of rat foetuses explanted at 12-5 and 13-5
days of gestation. /. Embryol. exp. Morph. 29, 473-483.
FAINSTAT, T. (1972). Submerged organ culture: an improved method. In Vitro 7, 300-307.
Goss, A., BODNER, J. & AVERY, J. (1970o). In vitro fusion of cleft palate shelves. Cleft Palate
J. 7, 737-747.
Goss, A., BODNER, J. & AVERY, J. (19706). Fusion capability of rat embryonic oral tissue in
vitro. Archs Oral Biol. 15, 797-804.
LAHTI, A., ANTILA, A. & SAXEN, L. (1972). The effect of hydrocortisone on the closure of the
palatal shelves in two inbred strains of mice in vivo and in vitro. Teratology 9, 37-42.
MORIARTY, T., WEINSTEIN, S. & GIBSON, R. (1963). The development in vitro and in vivo of
fusion of the palatal processes of rat embryos. /. Embryol. exp. Morph. 11, 605-619.
MYERS, G., PETRAKIS, N. & LEE, M. (1967). Cultivation of embryonic rat palate in defined
and semi-defined media. Archs Oral Biol. 12, 565-567.
NEW, D. (1967). Development of explanted rat embryos in circulating medium. / . Embryol.
exp. Morph. 17, 513-525.
NEW, D. & DANIEL, J. (1969). Cultivation of rat embryos explanted at 7-5 to 8-5 days of
gestation. Nature, Lond. 223, 515-516.
POURTOIS, M. (1966). Onset of the acquired potentiality for fusion in the palatal shelves of
rats. J. Embryol. exp. Morph. 16, 171-182.
REEVE, W., PORTER, K. & LEFKOWITZ, W. (1966). In vitro closure of the rat palate. / . dent.
Res. 45, 1375-1380.
In vitro palate development
495
M., SHEPARD, T. & TANIMURA, T. (1972). A new in vitro culture technique for
rat embryos. Teratology 5, 367-376.
SAXEN, L. (1966). The effect of tetracycline on osteogenesis in vitro. J. exp. Zool. 162, 269-294.
THOMPSON, J. & SCHWEISTHAL, M. (1969). Study of closure of the embryonic rat palate in
vitro with the effects of certain chemicals. /. dent. Res. 48, 568-572.
TROWELL, O. (1954). A modified technique for organ culture in vitro. Expl Cell Res. 6,
246-248.
WALKER, B. & FRASER, F. (1956). Closure of the secondary palate in three strains of mice.
/. Embryol. exp. Morph. 4, 176-189.
WALKER, B. & CRAIN, B. (1960). Effects of hypervitaminosis A on palate development in
two strains of mice. Am. J. Anat. 107, 49-58.
ROBKIN,
(Received 4 March 1975, revised 24 April 1975)