J. Embryo!, exp. Morph. Vol. 69, pp. 193-213, 1982
Printed in Great Britain © Company of Biologists Limited 1982
193
The effects of chlorcyclizine-induced
alterations of glycosaminoglycans on mouse
palatal shelf elevation in vivo and in vitro
LINDA L. BRINKLEY1 AND MARY M. VICKERMAN1
From the Department of Anatomy and Cell Biology,
Medical School, University of Michigan
SUMMARY
To define whether glycosaminoglycans play a role in palatal shelf movement, we studied
the morphology and elevation behaviour of chlorcyclizine-treated mouse palatal shelves.
Chlorcyclizine treatment was used because this agent enhances degradation of the palatal
glycosaminoglycans, hyaluronate and chondroitin sulphates, with little or no effect on their
synthesis. Use of in vitro and in vivo experiments enabled us to control the complicating
effects of other factors on elevation.
Drug-administration resulted in a reduction in shelf size, as measured by cross-sectional
surface area, in the posterior two-thirds of the palatal shelf. In vivo shelf reorientation was
also inhibited. When elevation behaviour was observed in vitro, pronounced regional variation was noted. The anterior third of the shelf was able to reorient, the posterior two-thirds
was not. This region also showed distinct histological changes as compared to control$.
Mesenchymal cells were rounded with prominent nuclei and nucleoli and were more densely
packed than in controls.
These results suggest that for at least the posterior two-thirds of the palatal shelf, the
intrinsic reorientation ability may in large part be linked to the acquisition of a specific
temporal and spatial distribution of hyaluronate and possibly other matrix components.
INTRODUCTION
The secondary palate is formed from two wedge-shaped masses of tissue, the
palatal shelves, that initially grow down vertically from the maxillary processes
to occupy positions on either side of the tongue. Subsequently the shelves
undergo a reorientation or elevation to a horizontal position above the tongue
where they meet in the midline, adhere and fuse to one another to form the
secondary palate. The palatal shelves are not passively displaced, but are
generally acknowledged to play a direct and active role in their reorientation
(Walker & Fraser, 1956). The internal 'shelf force' underlying their movement
has been the subject of considerable speculation and observation, but little
direct experimentation.
1
Authors' address: Department of Anatomy and Cell Biology, Medical School, University
of Michigan, Ann Arbor, Michigan 48109, U.S.A.
194
L. L. BRINKLEY AND M. M. VICKERMAN
We have confirmed the idea that a reorienting force clearly resides within the
shelf itself by demonstrating that palatal shelves in severed heads with the brain
and tongue removed are able to elevate (Brinkley, Basehoar, Branch & Avery,
1975). Further, parts of shelves that have been lesioned transversely are able to
reorient (Brinkley & Vickerman, 1979). Thus it seems that the shelves themselves are in some way responsible for generating the internal force that results
in a form change.
Examination of the shelves' composition may help to generate hypotheses
about the factors necessary for reorientation. The palatal shelves are composed
of a core of mesenchymal cells surrounded by extracellular matrix and covered
by an epithelium. The extracellular matrix of shelves just prior to reorientation
has been shown to contain collagen, fibronectin and two glycosaminoglycans
(GAGs): hyaluronate (HA) and chondroitin sulphates. HA is the predominant
molecule present at this time, comprising about 60 % of the extracellular GAGs
(Pratt, Goggins, Wilk & King, 1973).
The presence, molecular state and interactions of HA with other extracellular
molecules have been correlated with morphogenetic changes in a variety of
embryonic systems (Toole, Underhill, Mikuni-Takagaki & Orkin, 1980). A
number of unique physicochemical attributes of this macromolecule could
provide the extracellular matrix with morphogenetically important properties.
HA molecules occupy large domains and can enmesh with themselves and other
extracellular molecules to form networks which can sequester water and are
responsive to changes in ionic strength and tissue compression (Laurent, 1970;
Comper & Laurent, 1978). Glycosaminoglycans, particularly HA, are thought
to play a role in palatal shelf movement, but little direct evidence for this
suggestion exists (Lazzaro, 1940; Walker, 1961; Wilk, King & Pratt, 1978;
Ferguson, 1978; Brinkley, 1980).
Study of the effects of disruptive agents can often clarify basic mechanisms.
One such example is the potential effect of chlorcyclizine (CHLR) on palatal
shelf morphology and behaviour. CHLR is an antihistaminic benzhydrylpiperazine compound which has been shown to enhance the degradation of HA and
chondroitin sulphates with little effect on their synthesis. Both chlorcyclizine
administered by gavage in vivo and exposure of isolated palatal shelves to its
metabolite norchlorcyclizine (NORCHLR) in vitro causes the molecules to be
degraded into smaller-molecular-weight pieces with lower charge densities
(Wilk, et al. 1978). In vivo treatment of pregnant mice with this compound
results in both cleft palate and mandibular reductions in their offspring (King,
1963; King, Weaver & Narrod; 1965; Wilk et al. 1978). Cleft palate could be a
secondary result of the mandibular reduction, since a reduced mandible could
cause the tongue to become wedged between the palatal shelves, thus providing
a physical barrier to shelf elevation (Diewert & Pratt, 1979). Another possibility
is that CHLR treatment is a direct cause of clefting through its effect on the
GAGs within the shelves themselves (Wilk et al. 1978).
Chlorcydizine and palate elevation in vivo and in vitro
195
Anterior
Posterior
Soft
Fig. 1. Schematic drawings of the method used to delimit the palatal shelves for
cross-sectional area measurements. For anterior and mid presumptive hard palate,
a horizontal line in register with the inferior surface of the nasal septum was drawn
(broken line). Boundary lines for each shelf were determined by drawing a line from
the edge of the oral cavity to the point of intersection of the horizontal line with the
nasal shelf surface. Posterior presumptive hard palate and soft palate were delimited
by drawing a line from the medial edge of the tooth germ, if present, or edge of the
oral cavity if no tooth germ was present, which was also tangential to the curve of
the roof of the oral cavity. The first point of intersection of this line with the roof
of the oral cavity was then used as the most superior point of the shelf. Representative cross-sectional areas which were measured by this method are stippled.
To distinguish between these effects both in vivo and in vitro techniques must
be used. We have previously shown that normal palatal shelves are able to
elevate, adhere and fuse in our in vitro system (Brinkley et al. 1975; Lewis,
Thibault, Pratt & Brinkley, 1980). We can use this culture system to determine
directly the effects of CHLR treatment on the shelves without the complicating
effects of the tongue. Comparison of in vivo and in vitro behaviour of the palatal
shelves of CHLR-treated embryos should contribute to a better understanding
of the possible role glycosaminoglycans play in shelf elevation.
MATERIALS AND METHODS
Animals. A random-bred CD-I mouse strain was used for all experiments.
Animals were maintained in quarters with controlled light cycles and fed Purina
mouse chow and water ad libitum. Fertilization was assumed to occur between
midnight and 2 a.m. of the morning the vaginal plug was found.
Chlorcydizine administration. Sequential doses (250 mg/kg) of chlorcydizine
hydrochloride (CHLR) (Burroughs-Wellcome) were administered to pregnant
mice via intragastric tube on gestational days 10-5, 11-5 and 12-5, as these three
days were found to be the sensitive period for CHLR induction of cleft palate.
Each dose was delivered in a total volume of 0-5 ml sterile water. Controls Were
given intragastric feedings of equal volumes of sterile water.
196
L. L. BRINKLEY AND M. M. VICKERMAN
Animals were sacrificed on days 13-5, 14-25, 14-5 or 15-25, times which were
approximately 1, 2 and 3 days after the final dose of CHLR. Palate closure in
these mice normally occurs by day 14-5.
The drug treatment resulted in an 85 % survival rate of the treated mice. Half
of these mice had litters with 100% survival, a quarter had litters with a 10%
resorption incidence and a quarter had litters with about 50% resorption. In all
litters, the drug dosage given resulted in 100 % cleft palate in offspring of treated
mothers despite the fact that the treatment ended two days prior to palate
closure.
Preparation of explants and culture technique. Control and CHLR-treated
pregnant females were killed by cervical dislocation at day 13-5 or 14-25.
Foetuses were removed under sterile conditions, measured for crown-rump
length and assigned a morphological rating of the developmental state of fore
and hindlimbs, ears, eyelids and hair follicles using the system of Walker &
Crain (1960). Inter- and intralitter variability was minimized by excluding any
individuals that differed from the group in either of these two parameters.
The brain and tongue were removed and then a small vent was cut in the
floor of the posterior oral cavity to allow for better circulation of culture
medium. The mandible was left in place, as described previously (Brinkley et al.
1975). Several of the heads were then fixed in phosphate-buffered formalin to
serve as time zero, unincubated controls. The remainder were hung in a gassed,
circulating culture chamber that we developed (Lewis et al. 1980).
Specimens with unelevated palatal shelves were incubated in either standard
medium or standard medium containing norchlorcyclizine (NORCHLR), the
metabolite of CHLR. In vivo CHLR was administered by gavage, as the animal
converts the compound to its metabolite NORCHLR. For in vitro studies
NORCHLR itself was used. This procedure provided for comparison of the
elevation behaviour of the palatal shelves under recovery conditions provided
by standard medium, or in the continued presence of the active drug.
Standard medium consisted of BGJb medium (Gibco) with 10 % foetal bovine
serum (KC Biologicals), supplemented with 8 mM glutamine and containing
50 /*g/ml gentamicin (Schering Corp.). When norchlorcyclizine (NORCHLR)
was added to the medium, a final concentration of 10 /*g/ml was used. Cultures
containing NORCHLR were always kept in the dark because of the compound's
light-sensitivity.
In all cultures the medium was constantly circulated, gassed with 95% O 2 ,
5% CO 2 using silicone copolymer hollow fibre devices, and maintained at
34 °C. Specific details of the procedure are described in Brinkley et al. (1975)
and Lewis et al. (1980). Specimens were incubated 18 h, a time span previously
shown to permit elevation and adhesion in day-13 palates (Brinkley, Basehoar
& Avery, 1978).
Histological technique. Specimens to be examined histologically were fixed for
24 h in phosphate-buffered formalin, dehydrated through an alcohol series and
Chlorcyclizine and palate elevation in vivo and in vitro
197
Table 1. Chlorcyclizine effects on growth and morphology
Gestational age and treatment
Day 13-5
Day 14-25
Day 14-5
Day 15-25
Control {n = 12)
CHLR (n = 37)
Control in = 24)
CHLR (n = 20)
Control in = 59)
CHLR (n = 26)
Control in = 27)
CHLR(/z = 11)
Crown-rump
length (mm)*
Morphological
rating
10-4 ±0-2
9-8±0-3f
12-4±0-5
10-6 + 0-5t
12-7 ±0-6
11-1+Olf
14-6 ±0-4
14-2 ±0-4
4-5
2-3
14-15
6-9
14-17
8-9
24-25
24-25
* Mean ± standard deviation.
f Different from age-matched control values, P < 0005.
embedded in glycol methacryiate. Sections, 3 fim thick, were cut and stained
with toluidine blue.
Evaluation techniques
Palatal Shelf Index (PSI). The degree of elevation of palatal shelves was
assessed using the palatal shelf index (PSI). The PSI is a five-level scale to assess
deviation of the palatal shelves from the horizontal (Wee, Wolfson & Zimmerman, 1976). A grid with five lines from vertical (1) to horizontal (5) was placed
in the ocular of a dissecting microscope. Then, with the grid in place, freehand
razor-blade sections were made transversely through the palate of each fixed
specimen at four levels along the rostral-caudal axis. The slice was exposed to
a drop of toluidine blue to enhance contrast. Then the shelf position was
assessed by placing the horizontal line in register with the inferior surface of the
nasal septum or cranial base and recording which of the five angular lines most
closely corresponded to the position of the nasal surface of each shelf. This grid
was also used to assess stained, mounted sections of palates.
Cross-Sectional Area (CSA) measurements. Sections of anterior, mid and
posterior presumptive hard palate and presumptive soft palate were selected
from each specimen. With the aid of a microscope equipped with a drawing
tube, both shelves were traced. Anatomical landmarks visible in each region
which were used to assign a given section to a palate region were those of
Diewert (1978). The landmarks were: (1) anterior hard palate: nasal septum,
vomeronasal organs, nasal cavity and palatal shelves; (2) mid hard palate: nasal
septum posterior to vomeronasal organ, nasal cavity and palatal shelves; (3)
posterior hard palate: posterior aspects of nasal cavity or immediately posterior
to it, maxillary and mandibular toothbuds; (4) soft palate: cranial structures,
the cranial base and palatal shelves. The shelf area measured for each region
described is shown in Fig. 1.
Day 14-25
Day 13-5
Controls
CHLR
CHLR
Controls
2-7 ±0-7
50±0*
2-5 ±0-6
4-6±0-5*
4-3±0-5*
50±0
4-2±0-7*,$
4-5±0-8
4-8±0-7
Anterior
2-6 ±0-5
4-1 ±0-7*
2-9 ±0-2
3-6±O-7*
3-5±O-5f
50±0
3-3±O-5*,t
4-2±0-9
41 ±1-0
Mid
* Different from age-matched, time zero, controls, P < 001.
t Different from age-matched, time zero, CHLRs, P < 001.
i Different from day-13-5 time zero, CHLRs, P < 005.
Time zero (n = 14)
Cultured, standard medium (n = 20)
Time zero (n = 12)
Cultured, standard medium (n = 16)
Cultured, NORCHLR medium (n = 10)
Time zero (n = 8)
Time zero (n = 14)
Cultured, standard medium (n = 14)
Cultured, NORCHLR medium (« = 8)
Gestational age and treatment
2-3 ±0-6
4-0±0-7*
1-9 ±0-6
2-2±0-6
2-2±0-6
50±0
2-7±0-5*,}
3-9±l-0f
4-1 ± l-0f
Posterior
1-4 ±0-5
3-7±0-6*
1-7 ±0-4
2-0±0-4
l-0±0-2
50±0
2-l±0-5*,:
3-8±l-2f
4 0 ± l-2f
Soft
Palatal shelf index of shelf regions (mean ± standard deviation)
Table 2. Elevation behaviour of control and CHLR-treated palatal shelves in vitro
o
o
m
i—i
<
D
S
®
r
CO
Day 14-5
Day 15-25
Day 14-25
Day 13-5
61-7 ±10-9
59-4 + 22-7
110-9±33-7t
97-7±35-4f
167-4±49-3J,§
119-0±26-8§
Anterior
97-9+10-7
94-9 ±17-8
194-2 ±80-0f
127-8 +42-5*, f
251-5 + 58-3J
199-2 ±31-6§
Mid
156-2 + 25-3
114-9 ±280*
182-3 ±38-9
148-3 + 44-9*, t
202-9+ 30-4f
242-3 ±400§
Posterior
* Different from age-matched controls, P < 0-05.
t Different from day-13-5 treatment-matched groups, P < 0-01.
% Different from day-13-5 and day 14-25 treatment-matched groups, P < 0004.
§ Different from day-14-5 controls, P < 0-005.
Controls (« = 8)
CHLR in = 12)
Controls (n = 8)
CHLR in = 14)
Controls (« = 10)
CHLR in = 12)
Gestational age and treatment
110-6 ±44-4
78-1 ±34-1*
174-9 ±35-8f
122-7 +36-8*, f
181-3 ±47-2f
182-6 ±40-4
Soft
Cross-sectional area of shelf regions i/im2 x 103) (mean ± standard deviation)
Table 3. Changes in shelf cross-sectional area over the time ofm vivo palate closure
•
3
a
OS
3
p
I—>•
o
s
late elevatior
200
L. L. BRINKLEY AND M. M. VICKERMAN
The anterior and mid regions each represent about 20% of overall shelf
length, while the posterior and soft regions comprise about 30% each (Brinkley
& Vickerman, 1979).
Sampling and statistical analysis. Because of possible individual variability in
all evaluation parameters samples were taken by randomly selecting two or
more foetuses from each of 4-12 litters. Crown-rump length, PSI and CSA data
were analysed by a univariate analysis of variance. Only differences found to be
significant with P < 0-05 were accepted.
RESULTS
In vivo
Growth and morphology. Beginning on day 13-5, and continuing to day 14-5,
the day of usual palate closure, GHLR-treated individuals were smaller as
judged by crown-rump length and were less well developed as evidenced by
lower morphological ratings (Table 1). By day 15-25, almost three days after
the last drug treatment, CHLR-treated foetuses had apparently recovered and
attained the same crown-rump length and morphological rating as control
individuals. However, at all ages examined, CHLR-treated foetuses had both
cleft palate and reduced mandibles.
Palatal shelf elevation. On day 13-5 no difference in shelf position was found
between control and CHLR palates (Table 2). By day 14-25 palates of control
foetuses had closed whereas only the anterior, and to a lesser degree the posterior
of the CHLR-treated foetuses showed changes in shelf positions. The anterior
had essentially reoriented, but despite some change in the position of the
posterior shelves, it was insufficient to achieve elevation of this portion.
Changes in shelf CSA. Cross-sectional area of control and CHLR-treated
shelves was measured at several times over the span of expected palate closure
(Table 3). On day 13-5, one day prior to palate closure, a comparison of CSA
values of control and CHLR shelves shows that for this parameter the posterior
and soft palate regions are significantly affected by CHLR treatment. These two
regions were reduced to 74% and 71 % of control values, respectively.
After palate elevation, day 14-25, controls showed significant CSA increases
in anterior (80 %), mid (98 %) and soft (58 %) palate regions over that seen at
day 13-5. But no significant change in the posterior region was seen. By the time
adhesion had taken place (day 14-5), the anterior and mid shelf showed an
additional increase in CSA.
During the same period, day 13-5 to day 14-25, only the anterior region of
CHLR-treated palates had elevated (Table 2). The CSA values of control and
CHLR shelves were similar in the anterior, but the CHLR values were significantly lower than those of controls in the remainder of the palate. CHLRtreated mid palate achieved only 66% of control surface area, whereas posterior
*
t
t
§
Different
Different
Different
Different
from
from
from
from
156-3 ±25-3
131-7±18-3
114-9±280t
169-4 ± 32-3*, t
131-2±22-8§
94-9 ±17-8
121-3 ±32-4*,t
97-2±37-0§
59-4 ±22-7
75-0 ±22-5
66-1 ± 20-8
Posterior
97-9 ±10-7
151-1 ±36-6*
Mid
61-7 ± 10-9
84-9 ± 16-3
Anterior
the treatment-matched time zero group, P < 002.
controls, time zero, P < 0-02.
controls, cultured in standard medium, P < 0-03.
CHLR, cultured in standard medium, P < 001.
Control
Time zero (n = 8)
Cultured, standard medium (n = 12)
CHLR
Time zero (n = 12)
Cultured, standard medium (n = 22)
Cultured, NORCHLR medium (n = 20)
In vivo treatment and culture conditions
78-l±34-lf
114-8 ± 30-8*, t
83-3 ±32-5
110-6 ±44-4
129-5 ±31-8
Soft
Cross-sectional area of shelf regions (/*m2 x 103) (mean ± standard deviation)
Table 4. In vitro changes in cross-sectional area of day-13-5 palatal shelves
s.
5^
a"
202
L. L. BRINKLEY AND M. M. VICKERMAN
Table 5. In vitro changes in cross-sectional area of CHLR-treated
day-14-25 palatal shelves
Cross-sectional area of shelf regions Cam2 x 103)
(mean ± standard deviation)
Treatment
Time zero (n = 14)
Cultured, standard
medium (n = 14)
Anterior
Mid
Posterior
Soft
97-7 ±35-4
127-8 ±42-5
148-3 ±44-9
122-7 ±36-8
119-3±30-6
118-8±43-3*
186-7 + 25-1*
127-3±16-4
* Different from time zero, P < 0003.
and soft palate had CSAs which were 8 1 % and 70% of control values, respectively.
Since no in vivo elevation of CHLR shelves had taken place by day 14-25, an
additional day of in vivo development was allowed to determine what changes
if any, would occur in shelf position and area. By day 15-25, three days after
the last in vivo CHLR treatment, 5 of 12 individuals (42%) had shelves which
were elevated, but all had large gaps between them. No statistically significant
difference in CSA was found between elevated and unelevated day-15-25 shelves.
Thus, for purposes of comparison with controls, the mean of all values was used.
No significant increase in CSA over that seen at day 14-25 was observed for
the anterior, but the remaining shelf regions had increased. The cross-sectional
areas achieved by the anterior, mid and soft palate regions of day-15-25 CHLR
shelves were approximately those of day-14-25 controls, that had all elevated.
However, the posterior region of day-15-25 CHLR shelves attained areas
greater than those seen in either day-14-25 or -14-5 controls.
In vitro
Palatal shelf elevation. Palates of day-13-5 control specimens were able to
elevate along the entire length of the shelf, and adhere in the posterior twothirds of the shelf in 67% of the cases. When day-13-5 CHLR individuals were
incubated in standard medium, elevation was achieved in the anterior and to
some degree the mid regions, but no change in shelf position was observed in
either the posterior or soft region. CHLR specimens cultured in medium containing NORCHLR exhibited the same regional pattern of shelf-elevation
behaviour (Table 2).
Day-14-25 CHLR-treated palates were cultured to determine whether the
posterior and soft regions of the shelves would now be able to elevate in vitro.
By this age these regions had achieved CSAs similar to day-13-5 controls.
CHLR-treated palates were indeed able to elevate in both regions whether
cultured in standard medium or NORCHLR medium (Table 2).
+ + + +/0
+ + +/0
+ + +/0
In vivo
Controls
CHLR
In vitro
Controls
CHLR, standard medium
CHLR, NORCHLOR medium
+ +/0
0/+ +
Mid
+ + +/0
0/+ +
+ + + +/0
Posterior
0/0
+ +/0
0/+ +
0/+ + +
Soft
PSI: No change = 0; significant increase to final PSI of: 2-25-3-0 = + ; 3-0-3-75 = + + ; 3-75-4-25 = + + + ; 4-25-5-0 = + + + +
CSA: No change = 0; significant increase of: 0-25 % = + ; 25-50 % = + + ; 50-75 % = + + + ; 75-100 % = + + + + .
Anterior
Treatment
Shelf region (Palatal Shelf Index/cross-sectional area)
Table 6. Summary of changes in shelf Palatal Shelf Index and cross-sectional area during palate closure
3
a.
I
o
,1
204
L. L. BRINKLEY AND M. M. VICKERMAN
Fig. 2. Day-13-5 control (A and B) and CHLR-treated (C, D and E) shelves before
and after incubation. (A) Time-zero controls, prior to incubation. (B) After culture,
shelves have elevated and adhered. Areas of lowered cell density are obvious (-*).
(C) Unincubated, time-zero CHLR-treated shelves. (D) Incubated in standard
medium. Note enlargement of vascular spaces (*). (E) Incubated in NORCHLR
medium. Vascular spaces are not enlarged.
Changes in shelf CSA. Only the mid region of control day-13-5 palates increased significantly in surface area during the in vitro culture period (Table 4).
CHLR specimens cultured in standard medium showed significant CSA increases in the mid, posterior and soft palate regions. Despite this increase,
neither the mid nor soft palate of those specimens achieved CSA values attained
by the same regions of cultured controls. The CSA increase found in the
posterior brought the CHLR shelves to a size similar to that of uncultured
control shelves that showed no CSA change during their culture period.
CHLR specimens cultured in NORCHLR-containing medium increased in
Chlorcyclizine and palate elevation in vivo and in vitro
205
Fig. 3. Comparison of mesenchymal cells of day-13-5 control (A and B), and
CHLR-treated (C, D and E) shelves. (A) Mesenchymal cells of time zero, unincubated shelves. (B) After in vitro elevation and adhesion. (C) Mesenchymal
cells of time zero, unincubated CHLR shelves. (D) After incubation in standard
medium. (E) Cells of shelves incubated in NORCHLR medium. Although densely
packed, several mitotic figures were observed in each field (->).
surface area only in the posterior, achieving a size similar to controls, but
smaller than that of CHLR shelves cultured in standard medium.
When day-14-25 CHLR-treated palates were cultured in standard mediuiti,
the CSA of these shelves increased significantly in the posterior regions (Table
5), and achieved a CSA similar to the posterior region of day-14-25 controls by
the end of the culture period (Table 3).
Summary ofPSI and CSA changes in vivo and in vitro. Table 6 contains a
tabulation of the significant changes observed in both elevation behaviour arid
shelf surface area in vivo and in vitro. In vivo, shelf elevation and expansion in
cross-sectional area of controls occurred concurrently in the anterior, mid afld
206
L. L. BRINKLEY AND M. M. VICKERMAN
Fig. 4. Comparison of day-14-25 control (A) and CHLR-treated (B and C) shelves.
(A) Control shelves. (B) Unincubated CHLR shelves. (C) CHLR shelves elevated
in vitro, o, Areas of osteogenesis; t, tooth germ.
soft shelf regions, but elevation without expansion was seen in the posterior
shelf. In vitro, only mid shelf of control specimens expanded during elevation.
Thus, in normal palates, elevation of the anterior, posterior and soft palate
regions does not appear to be linked to overall shelf expansion as measured by
changes in CSA.
CHLR treatment effectively blocked shelf elevation in vivo in the mid and soft
palate and retarded it to a large degree in the posterior region. However,
increases in CSA of these regions was observed. In vitro behaviour of CHLR
shelves reinforced the findings from controls that changes in overall shelf CSA
are not necessarily linked to elevation in the anterior, posterior and soft palate
regions. This may not, however, be the case for the mid-shelf region. It showed
a pronounced divergence of behaviour in vitro as compared to in vivo. In vitro
mid region was able both to elevate and to expand to some degree in standard
medium, whereas in vivo only CSA expansion is observed.
The presence of NORCHLR in the culture medium permitted the same degree
of shelf reorientation seen in standard medium, but effectively blocked or
severely reduced CSA expansion in all shelf regions.
It is clear from both in vivo and in vitro studies that CHLR treatment rendered
the caudal two-thirds of the palate, the posterior and soft palatal regions, unable
to elevate. These were also the regions whose day-13-5 CSA was significantly
reduced as compared to control shelves (Table 3).
Chlorcyclizine and palate elevation in vivo and in vitro
207
Fig. 5. Comparison of mesenchymal cells of day-14-25 control (A and B) and
CHLR-treated (C and D) shelves. (A) Medial region of adhered control shelves.
(B) Control mesenchymal cells still show similar morphology and density to that
seen at day 13-5. (C) Medial region of adhered CHLR shelves. (D) Cells of CHLR
shelves are now more similar to controls in morphology, but are still smaller.
Histological effects of chlorcyclizine
Day 13-5. The mid, posterior and soft palate regions of the shelf exhibited
histological changes after CHLR treatment. Examined at low magnification,
control shelves (Fig. 2 A) had less densely packed mesenchymal cells than do
CHLR-treated specimens of the same age (Fig. 2C). Local differences in cell
packing were also apparent. An area of lower cell density medial to the tooth
germ was observed in control shelves, which was less extensive in the CHLRtreated specimen. After in vitro elevation this area of lowered cell density wfts
pronounced in control specimen (Fig. 2B), less so in CHLR shelves cultured in
standard medium (Fig. 2D), and NORCHLR-cultured specimens showed little
208
L. L. BRINKLEY AND M. M. VICKERMAN
change in this area (Fig. 2E). CHLR-standard-medium-cultured specimens also
showed an area of decreased cell density on the nasal side of the shelf, which
was not seen in controls or NORCHLR-cultured individuals. It was also consistently noted that the lumens of the blood vessels of CHLR-standard-cultured
shelves were considerably expanded when compared to control animals or to
NORCHLR-cultured specimen.
A qualitative comparison of cell density, size and morphology at higher
magnifications revealed substantial differences between uncultured control
tissue (Fig. 3 A) and that of uncultured CHLR-treated individuals (Fig. 3C).
Mesenchymal cells of control shelves appeared less densely packed, smaller,
more elongate or stellate, and with relatively indistinct nuclei. In contrast, cells
of uncultured CHLR-treated tissue (Fig. 3 C) were more densely packed, larger,
more rounded and exhibited distinct nuclei with prominent nucleoli. After
culture (Fig. 3B), cells of control tissue appeared even more elongate and in fact
both cells and nuclei exhibited wavy perimeters. Culture of CHLR-treated
specimens in standard medium (Fig. 3D) resulted in shelves with mesenchymal
cells which appear to be smaller than those of uncultured CHLR individuals,
but which had a similar rounded shape with prominent nuclei and nucleoli.
However, they remained larger and more densely packed as compared to incubated controls (Fig. 3B). Those cultured in NORCHLR medium (Fig. 3E)
exhibited extremely dense mesenchymal cell packing, and cells which were about
the same size as those seen in uncultured CHLR specimens. However, the
nuclei were lighter staining with several prominent nucleoli; little cytoplasm was
visible around them. Although densely packed with cells, several mitotic figures
were observed in each field.
Day 14-25. All control specimens were elevated and adhered (Fig. 4 A) with
visible epithelial breakdown and areas of osteogenesis. CHLR individuals
(Fig. 4B), however, showed no elevation, retarded tooth-germ development and
less-developed areas of osteogenesis. If such treated individuals were cultured
in standard medium (Fig. 4C), 70% elevated and adhered and ossification
proceeded.
High-power views of control shelves (Fig. 5 A, B) revealed mesenchymal cell
morphologies similar to those seen in day-13-5 controls. Whereas mesenchymal
cells of CHLR shelves which had elevated in vitro (Fig. 5 C, D), remained more
densely packed than controls and still exhibited the prominent nuclei and
nucleoli observed in day-13-5 CHLR specimen. Cells of day-14-25 CHLR
shelves (Fig. 5D) also appeared smaller than those of controls (Fig. 5B).
DISCUSSION
Nature and distribution of CHLR effect
Direct effect on shelves. Our experiments demonstrate that CHLR administration results in changes in shelf size and loss of the inherent ability to elevate.
Chlorcyclizine and palate elevation in vivo and in vitro
209
Other investigators have demonstrated direct effects of CHLR on GAGs in the
palatal shelves, but have also shown an effect on mandibular growth (Wilk et al.
1978; King, 1963). Therefore, in our studies either GAG effects or mandibular
reduction could be key elements in the observed clefting. The latter alternative
can be eliminated. We have shown that the elevation ability of the palatal
shelves is directly affected by CHLR treatment using in vitro techniques, wherein
the tongue is removed and the behaviour of the shelves monitored directly.
Distribution of effects. CHLR effects on shelves were also observed to be
regional, with the severity of response to drug treatment graded in an anteriorposterior manner. A similar gradation in in vitro elevation behaviour has been
previously observed (Brinkley & Vickerman, 1979). The anterior shelf appeared
to be unaffected by CHLR treatment. The loss of elevation ability and reduction
of shelf CSA were evident in the mid to soft regions. The greatest effects were
found in the posterior and soft areas.
There are two possible explanations for a lack of effect in the anterior. Either
formation and elevation of the anterior palate are not dependent on production
of an extensive extracellular matrix, and thus are unaffected by CHLR treat*
ment, or the anterior recovers more rapidly and to a much greater extent than
other regions. The former explanation seems most likely, not only because of the
consistent lack of any drug effect on the CSA and elevation changes in this area>
but also because the anterior is the most highly cellular region of the palate,
with little extracellular space visible (Brinkley, 1980). Likewise, the pronounced
effects observed in the posterior portion of the shelves suggest that these area$
may be richer in the two CHLR-affected GAGs: HA and the chondroitifi
sulphates.
Shelf expansion and elevation ability
One of the oldest proposed mechanisms for the motive force underlying
palatal shelf movement is a rapid increase in shelf volume due to increased
intercellular volume in the connective tissue (Lazzaro, 1940; Walker, 1961). In
vivo, shelf expansion appears to occur over the course of rat palate closure when
observed in frozen sections (Diewert & Tait, 1979). Our in vivo findings of shelf
cross-sectional area expansion are in agreement with those of Diewert & Tait
for all shelf regions except the posterior, where we observed no significant shelf
expansion. It is interesting that both the frozen tissue used by them and the
fixed tissue used in the present study reveal a fairly similar pattern of in vivo
expansion despite the tissue shrinkage associated with fixation.
Comparison of both in vivo and in vitro studies permits us to determine
whether shelf expansion is actually necessary for shelf elevation. Our studies
demonstrate that overt shelf expansion, as measured by change in shelf crosssectional surface area, is not required for shelf reorientation in the anterior,
posterior and soft palate, given the degree of maturity of a day-13-5 palatal shelf.
In the mid palate though, elevation and expansion appear to be linked to some
210
L. L. BRINKLEY AND M. M. VICKERMAN
degree. Present results also suggest that the in vivo condition is somehow related
to shelf expansion. Given the physical properties of an HA-rich extracellular
network it seems possible that in vivo shelf expansion is a response to the external compression of the shelves caused by their wedged position between the
tongue and mandible. This expansion may be necessary for initiating the movement of the shelves around the tongue or for stimulating the tongue to move
out of the way of the elevating shelves.
GAGs and shelf elevation
HA is the predominant GAG species present in the shelves around the time
of elevation (Pratt et ah 1973). Thus CHLR effects on shelf morphology and
elevation behaviour must in large part be attributable to degradation of this
molecule into smaller pieces, resultant changes in its ionic responsiveness and/or
altered interactions with other extracellular molecules. To understand how these
changes might affect shelf elevation it is necessary to examine some of the
properties of HA.
HA in extracellular matrices. The hyaluronate content of the extracellular
matrix is known to be important in the morphogenesis of that tissue (Toole et ah
1977; Toole et ah 1980). HA is a very large, highly charged polyanion with a
large molecular domain that tends to sequester water (Laurent, 1970). Its
molecular weight, length and the presence of other molecules including proteins
or cations can influence the viscosity of a hyaluronate solution (Balazs & Gibbs,
1970; Comper & Laurent, 1978; Morriss, Rees & Welsh, 1980). Hyaluronate
often enmeshes with sulphated proteoglycans and collagen to form a gel-fibre
network characteristic of extracellular matrices (Bernfield, 1981). Cellular
behaviour is known to be directly influenced by its association with the extracellular matrix. For instance, anchorage of cells to a substratum determines cell
shape and mobility and can thus alter cell metabolism (Bernfield, 1981).
Indirectly, the gel-fibre matrix could simply contribute to a complex framework
whose shape is the primary determinant of tissue form (Nakamura & Manasek,
1978).
These networks not only have specific physical configurations due to their
molecular components, but also have unusual physiological properties. Such
networks are capable of generating a 'swelling pressure' that is composed of an
osmotic contribution from the hyaluronate portion's tendency to sequester
water and an elastic contribution from the contractility of the fibres. If the
equilibrium between osmotic and elastic components is disturbed the network
will expand or shrink, depending on which components are affected (Comper &
Laurent, 1978). Morphogenetic changes could then be brought about in tissues
as a result of these physicochemical properties. Physical disturbance such as
external compression of a tissue containing the network, or chemical changes
such as local or systemic variations in the ionic environment, could alter the
configuration of the network and in turn that of surrounding tissue. Such
Chlorcyclizine and palate elevation in vivo and in vitro
211
remodelling could passively displace or translocate cells to areas of least resistance (Solursh, Fisher, Meier & Singley, 1979; Brinkley, 1980), or provide
spaces for actual cell migration (Pratt, Larsen & Johnston, 1975; Tosney, 1978;
Erickson, Tosney & Weston, 1980).
Consequences of altered HA. If one major component, for example HA, is
abnormal, it is reasonable to expect an alteration in extracellular matrix organization. Such a disturbance could affect cell attachment to the matrix as well as
the physicochemical properties of the matrix itself. If the cells were unable to
attach properly a rounded rather than stellate appearance could result, and a
change in cellular metabolism might occur which could preclude active cell
migration. An altered gel-fibre network in the extracellular matrix might also be
unable to undergo conformational changes necessary to displace cells passively
and create local expansions and compressions.
Both the elevation behaviour of CHLR-treated shelves and the histological
appearance of their mesenchymal compartment provide strong support for the
idea that HA plays a central role in shelf elevation through its spatial distribution'and macromolecular interactions in an extracellular network. The regional
nature of CHLR effects on elevation suggests that there is an underlying spatial
or temporal sequence of either HA and chondroitin sulphate production in the
shelves or of differential recovery from the effects of CHLR. Histologically the
mesenchymal cells of CHLR shelves appear large and rounded with prominent
nuclei and nucleoli which may reflect direct metabolic alterations due to the
CHLR. Their morphology might also be the result of an inability to adhere
properly to the surrounding abnormal extracellular matrix. Another histological
finding that may be related to the degradation of hyaluronate is that, following
in vitro culture in standard medium, the vascular spaces of CHLR-treated
shelves are enlarged as compared to controls. This result is comparable to that
of Singley & Solursh (1981), who found similar vascular inflation after degradation of hyaluronate by local injection of Streptomyces hyaluronidase in chick
wing buds.
Culture in NORCHLR medium, however, did not elicit this response, but did
result in a noticeable increase'in the number of dividing cells. It seems possible
that further exposure to the drug may have resulted in sufficient additional
alterations in the gel portion of the matrix both to change its osmotic properties,
precluding vascular enlargement, and to alter the relationship of cells to matrix
which may have resulted in mitotic stimulation.
Possible role of HA in shelf elevation. Our studies do not rule out the possibility
that some subtle regional shelf expansion may indeed play a role in shelf elevation, but do preclude overall shelf swelling as the motive force. Alteration of the
shape of a structure such as a shelf, with no change in surface area, could be
accomplished by local expansions compensated for by other local compressions
Both the present results and changes previously observed in cell patterning oyer
the course of shelf elevation (Brinkley, 1980), suggest that the internal compo-
212
L. L. BRINKLEY AND M. M. VICKERMAN
nents must be rearranged but not expanded. Given the wide range of properties
of gel-fibre matrices a highly patterned distribution of an HA-rich extracellular
matrix with specific local variations in the nature of the matrix could be the
underlying physical basis for such local tissue expansions and contractions. The
development of 'internal shelf force', i.e. the intrinsic ability of shelves to
elevate, may in large part be the acquisition of a specific temporal and spatial
distribution of HA and other matrix components.
This research was supported by NIH-NIDR grant DE02774 to L.L.B.
REFERENCES
BALAZS, E. A. & GIBBS, D. A. (1970). The rheological properties and biological function
of hyaluronic acid. In Chemistry and Molecular Biology of the Intercellular Matrix.
New York.
BERNFIELD, M. (1981). Organization and remodelling of the extracellular matrix in morphogenesis. In Morphogenesis and Pattern Formation (ed. T. G. Connelly, L. L. Brinkley &
B. M. Carlson), pp. 139-162. New York: Raven Press.
BRINKLEY, L. L. (1980). In vitro studies of palatal shelf elevation. In Current Research
Trends in Prenatal Craniofacial Development (ed. R. M. Pratt & R. L. Christiansen), pp.
203-220. New York: Elsevier/North-Holland.
BRINKLEY, L., BASEHOAR, G., BRANCH, A. & AVERY, J. (1975). A new in vitro system for
studying secondary palatal development. / . Embryol. exp. Morph. 34, 485-495.
BRINKLEY, L., BASEHOAR, G. & AVERY, J. (1978). Effects of craniofacial structures on mouse
palatal closure in vitro. J. dent. Res. 57, 402-411.
BRINKLEY, L. L. & VICKERMAN, M. M. (1979). Elevation of lesioned palatal shelves in vitro.
J. Embryol. exp. Morph. 54, 229-240.
COMPER, W. D. & LAURENT, T. C. (1978). Physiological function of connective tissue polysaccharides. Physiol. Rev. 58, 255-315.
DIEWERT, V. M. (1978). A quantitative coronal plane evaluation of craniofacial growth and
spatial relations during secondary palate development in the rat. Archs oral Biol. 23,
607-629.
DIEWERT, V. M. & PRATT, R. M. (1979). Selective inhibition of mandibular growth and
induction of cleft palate by diazo-oxo-norleucine (DON) in the rat. Teratology 20, 37-52.
DIEWERT, V. M. & TAIT, B. (1979). Palatal process movement in the rat as demonstrated in
frozen sections. / . Anat. 128, 609-618.
ERICKSON, C. A., TOSNEY, K. W. & WESTON, J. A. (1980). Analysis of migratory behavior of
neural crest andfibroblasticcells in embryonic tissues. Devi Biol. 11, 142-156.
FERGUSON, M. W. J. (1978). Palatal elevation in the Wistar rat fetus. / . Anat. 125, 555-577.
KING, C. T. G. (1963). Teratogenic effects of meclizine hydrochloride in the rat. Science 141,
353-355.
KING, C. T. G., WEAVER, S. A. & NARROD, S. A. (1965). Antihistamines and teratogenicity
in the rat. /. Pharmacol, exp. Ther. 147, 391-398.
T. C. (1970). Structure of hyaluronic acid. In Chemistry and Molecular Biology of
Intercellular Matrix, vol. 2 (ed. E. A. Balazs), pp. 703-732. New York: Academic Press.
LAZZARO, C. (1940). Sul meccanismo di chiusura del palato secondario. Monitore zool. Ital.
51, 249-273.
LEWIS, C. A., THIBAULT, L., PRATT, R. M. & BRINKLEY, L. L. (1980). An improved culture
system for secondary palatal elevation. In Vitro 16(6), 453-460.
MORRISS, E. R., REES, D. A. & WELSH, E. J. (1980). Conformation and dynamic interactions
in hyaluronate solutions. / . molec. Biol. 138, 383-400.
NAKAMURA, A. & MANASEK, F. J. (1978). Experimental studies of the shape and structure of
isolated cardiac jelly. J. Embryol. exp. Morph. 43, 167-183.
LAURENT,
Chlorcyclizine and palate elevation in vivo and in vitro
213
R. M., GOGGINS, J. R., WILK, A. L. & KING, C. T. G. (1973). Acid mucopolysao
charide synthesis in the secondary palate of the developing rat at the time of rotation and
fusion. Devi Biol. 32, 230-237.
PRATT, R. M., LARSEN, M. A. & JOHNSTON, M. C. (1975). Migration of cranial neural crest
cells in a cell-free hyaluronate-rich matrix. Devi Biol. 44, 298-305.
SINGLEY, C. T. & SOLURSH, M. (1981). The spatial distribution of hyaluronic acid and
mesenchymal condensation in the embryonic chick wing. Devi Biol. 84, 102-120.
SOLURSH, M., FISHER, M., MEIER, S. & SINGLEY, C. T. (1979). The role of extracellular
matrix in the formation of the sclerotome. J. Embryol. exp. Morph. 54, 75-98.
TOOLE, B. P., OKAYAMA, M., ORKIN, R. W., YOSHIMURA, M., MUTO, M. & KAJI, A. (1977),
Developmental roles of hyaluronate and chondroitin sulfate proteoglycans. In Cell and
Tissue Interactions (ed. J. W. Lash & M. M. Burger). New York: Raven Press.
TOOLE, B., UNDERHILL, C. B., MIKUNI-TAKAGAKI, Y. & ORKIN, R. (1980). Hyaluronate in
morphogenesis. In Current Research Trends in Prenatal Craniofacial Development (ed>
R. M. Pratt & R. L. Christiansen), pp. 263-275. New York: Elsevier/North-Holland.
TOSNEY, K. W. (1978). The early migration of neural crest cells in the trunk region of the
avian embryo: an electron microscopic study. Devi Biol. 62, 317-333.
WALKER, B. E. (1961). The association of mucopolysaccharides with morphogenesis of the
secondary palate and other structures in mouse embryos. /. Embryol. exp. Morph. 9, 22-31.
WALKER, B. E. & FRASER, C. F. (1956). Closure of the secondary palate in three strains of
mice. /. Embryol. exp. Morph. 4, 176-189.
WALKER, B. E. & CRAIN, B. (1960). Effects of hypervitaminosis A on palate development in
two strains of mice. Am. J. Anat. 107, 49-58.
WEE, E. L., WOLFSON, L. G. & ZIMMERMAN, E. F. (1976). Palate shelf movement in mouse
embryo culture: evidence for skeletal and smooth muscle contractility. Devi Biol. 48,
91-103.
WILK, A. L., KING, C. T. G. & PRATT, R. M. (1978). Chlorcyclizine induction of cleft palate
in the rat: degradation of palatal glycosaminoglycans. Teratology 18, 199-210.
PRATT,
(Received 9 September 1981, revised 7 December 1981)