/. Embryol. exp. Morph. 93, 29-49 (1986)
29
Printed in Great Britain © The Company of Biologists Limited 1986
Identification of glycosaminoglycans in the
chondrocranium of the chick embryo before and at the
onset of chondrogenesis
CHARLES D. GOLDSTEIN*, JOSEPH J. JANKIEWICZ* AND
M A R Y E . DESMOND
Department of Biology, Villanova University, Villanova, PA 19085, USA
SUMMARY
It appears that hyaluronate is associated with cell migration and the chondroitin sulphates
with differentiation during morphogenesis of the chick embryo. The aim of this study was to see
if such a correlation could be made for chondrocranium morphogenesis. Specifically, the
purpose of this study was (1) to determine the proportion of extracellular matrix (ECM) to cell
area and total head mesenchymal area during chondrocranium morphogenesis; and (2) to
identify the location, types, and relative amounts of glycosaminoglycans (GAG) being synthesized in the presumptive chondrocranium at the onset of chondrogenesis and prior to this
time.
Morphometric analyses were made on median and parasagittal sections of heads of stage-24
and -33 embryos in order to determine relative contributions of cells and ECM to the total area
of head mesenchyme at these stages. Presumptive chondrocrania (heads minus eyes) of these
stage embryos were also analysed histochemically and biochemically in order to identify the
GAGs present in the ECM. Sections of whole heads were stained with alcian blue at low and
high pH as well as digested prior to staining with hyaluronidase (Streptomyces and testicular).
Identification of GAGs was done by pulse labelling embryos with [3H]glucosamine, digesting
homogenates with hyaluronidase (Streptomyces or testicular), precipitating the undigested
GAGs with cetylpyridinium chloride and counting the dissolved precipitates using scintillation
spectrophotometry. The types and relative amounts of GAGs present in the presumptive
chondrocranium were determined by comparing the amount of radioactivity in the precipitates
of the non-digested GAG with the counts in the precipitates of the predigested GAGs.
This study reports that chondrogenesis begins in the presumptive chondrocranium of the chick
embryo at stage 33 and that the area of the head mesenchyme increases 60-fold between stages
24 and 33. Little change in cell density and individual cell area as well as in the relative
proportion of total area allocated to cells and ECM occurs. GAGs are localized exclusively in
the presumptive chondrocranium. These GAGs are restricted to the ventral half of the
presumptive chondrocranium. Within this region, the GAGs are further localized to the
presumptive facial area, perichordal region, ethmoid, sphenoid and periotic regions.
The types of GAG being synthesized in the head mesenchyme of both stage-24 and -33
embryos are hyaluronate, the chondroitins and unidentified sulphated GAGs (dermatan,
keratan, heparin and heparan sulphate). At stage 24, hyaluronate, chondroitin and the
unidentified sulphated GAGs constitute about 33 % each of the GAG being synthesized. At
stage 33, the level of hyaluronate synthesis drops to 2%, the chondroitins to 24% and the
* Investigation performed in partial fulfilment of the requirements for the degree of M.S. in
Biology from Villanova University.
Key words: hyaluronate, chondroitin sulphates, glycosaminoglycans, cell differentiation, cell
migration, skull morphogenesis, neurocranium.
30
C. D. GOLDSTEIN, J. J. JANKIEWICZ AND M. E. DESMOND
unidentified sulphated GAGs increase to 74 %. There is an 18-5-fold decrease in the percentage
of hyaluronate, a 1-5-fold decrease in the amount of chondroitins and a 2-7-fold increase
in the percentage of unidentified sulphated GAGs being synthesized as chondrocranium
morphogenesis proceeds.
INTRODUCTION
Studies of morphogenesis of the cornea, kidney, presumptive spinal column,
and limb suggest that there is an accumulation of hyaluronate along the pathways
of migrating mesenchymal cells, followed by a reduction in the amount of
hyaluronate and a rise in sulphated GAG concurrent with differentiation of the
migrated cells (Fig. 1). This pattern suggests that glycosaminoglycans (GAG)
present in the immediate environment may play a role in these developmental
sequences.
Predominant macromolecules in the extracellular matrix (ECM) include
glycoproteins and proteoglycans (Manasek, 1975; Hay, 1981). Glycosaminoglycans are the carbohydrate moiety of proteoglycans and are composed of longchained, polyanionic, repeating disaccharide units covalently attached to a core
protein that exists in conjunction with counterions, normally Na + (Manasek,
1975). There are eight types of GAGs of which six are sulphated. Based
on numerous developmental biology studies, these GAGs can be divided into
Migration
Differentiation
Hyaluronate &
o
U
Chondroitin sulphate ex.
o
n—""
Hyaluronidase cv
/
XT'.
/
J&
Early
Late
Morphogenesis
Fig. 1. A schema depicting changes in GAG synthesis and hyaluronidase activity
associated with cell migration and differentiation during morphogenesis. Cell
migration appears to be correlated with a high concentration of hyaluronate whereas
cell differentiation occurs with a decrease in hyaluronate and concomitant increase in
chondroitin sulphate. An increase in hyaluronidase activity parallels a decrease in
hyaluronate. This diagram is a composite based on data primarily generated by the
laboratory of Toole (1981).
Glycosaminoglycans in the embryonic chick
three functional groups: (1) hyaluronate, one of the non-sulphated GAGs;
(2) chondroitin and chondroitin sulphate; and (3) dermatan sulphate, keratan
sulphate, heparin, and heparan sulphate (Toole, 1976).
To date, most of what is known about the morphogenetic role of GAGs is their
function in the formation of the chick embryo cornea. It appears that the corneal
endothelial cells produce hyaluronate that causes the cornea stroma to swell
providing space into which mesenchymal cells can migrate (Toole & Trelstad,
1971; Meier & Hay, 1973). Cessation of cell migration and the initiation of
differentiation of these mesenchymal cells into corneal fibroblasts occurs at the
same time as do a significant decrease in hyaluronate coupled with a rise in
hyaluronidase activity plus an increase in the amount of chondroitin sulphates
(Toole & Trelstad, 1971; Meier & Hay, 1973). The newly differentiated corneal
fibroblasts appear to secrete the chondroitin sulphate (Meier & Hay, 1973).
A similar pattern of cell migration associated with a hyaluronate-rich environment and cell differentiation associated with a low hyaluronate, high chondroitin
sulphate environment has been reported for the morphogenesis of the metanephric kidney (Belsky & Toole, 1983); for the regenerating newt limb (Toole &
Gross, 1971); and for the developing vertebral column from sclerotomal cells
(Kvist & Finnegan, 1970a; Toole, 1976; Derby, 1978). Other studies that support
the idea that hyaluronate creates cell-free spaces into which mesenchymal cells
may migrate have been done by Solursh (1976) on chick gastrulae.
Not only has hyaluronate been implicated in cell migration but also in the
prevention of cell differentiation. Cell contact is thought necessary for differentiation of mesenchymal cells of the limb and it is suggested that hyaluronate may
inhibit distal limb differentiation by preventing cell contact (Kosher & Savage,
1981).
Another mesenchymal area that involves cell migration prior to cell differentiation during morphogenesis is the presumptive chondrocranium. Several
studies have focused on the migration of neural crest into the skull region
(Johnston, 1966; Noden, 1975; Pratt, Larsen & Johnston, 1975; and LeLievre,
1978) and some studies have noted the biochemical milieu of the ECM of the
developing skull (Hall, 1968). No study has documented the GAG content of the
ECM in the presumptive chondrocranium at a time preceding chondrogenesis
compared to the time chondrogenesis begins. Furthermore, no study has
documented the GAG content of the chondrocranium during neural crest
migration. Nor has any study shown the proportion of ECM to cell area and total
head mesenchymal area during chondrocranium morphogenesis.
Although we did not determine the amount of GAG synthesis during neural
crest migration in this study, we did determine (1) the proportion of ECM to cell
area and total head mesenchymal area during chondrocranium morphogenesis and
(2) the location, types, and relative amounts of GAG being synthesized in the
presumptive chondrocranium at the onset of chondrogenesis and prior to this time
(Fig- 2).
31
32
C. D. GOLDSTEIN, J. J. JANKIEWICZ AND M. E. DESMOND
Morphometric analyses were made on median and parasagittal sections of skulls
of stage-24 and -33 embryos in order to determine relative contributions of cells
and ECM to the total area of head mesenchyme at these stages. Presumptive
se
Fig. 2. A,B. Photomicrographs using bright-field optics of median sections of heads of
stage-24 and -33 chick embryos. At stage 24 (A) the head mesenchyme (presumptive
chondrocranium) is relatively undifferentiated, consisting of mesenchymal cells and
extracellular matrix (ECM). Likewise, the presumptive skin ectoderm and neuroepithelium both consist of one layer of cells. There arefivemajor brain vesicles at this
stage and the ventricles are the more prominent component of the brain. By stage 33
(B) many morphogenetic changes have occurred in the head, particularly in the head
mesenchyme. The area of the presumptive skull has increased many fold but the cell
density and cell size have not. The mesenchymal cells have aggregated into discrete
foci. The cells within these foci have begun differentiating into chondrocytes as
evidenced by the deposition of alcian blue material and blueish coloration upon
trichrome staining. The foci of chondrogenesis are restricted to the ventral half of the
head which will become face and the ethmoid, sphenoid, temporal and periotic regions
of the chondrocranium. ne, neuroepithelium; bv, brain ventricle; hm, head mesenchyme; se, skin ectoderm. Magnification: xl7 (stage 24) and x9 (stage 33).
Glycosaminoglycans in the embryonic chick
33
chondrocrania (heads minus eyes) of these stage embryos were also analysed
histochemically and biochemically in order to identify the GAGs present in the
ECM. This study reports that chondrogenesis begins in the presumptive
chondrocranium of the chick embryo at stage 33 (Hamburger & Hamilton, 1951)
and that the area of the head mesenchyme increases 60-fold between stages 24 and
33. Cell density and individual cell area as well as the relative proportion of total
area allocated to cells and ECM exhibit little change between these two stages.
GAGs are localized exclusively in the presumptive chondrocranium (head
mesenchyme plus basement membrane of the neuroepithelium). These GAGs are
not uniformly distributed throughout the head mesenchyme but are restricted to
the ventral half of the presumptive skull. Within the ventral half, the GAG is
further localized to specific areas such as the presumptive facial area, perichordal
region, and presumptive ethmoid, sphenoid and periotic regions.
The types of GAG being synthesized in the presumptive chondrocranium of
both stage-24 and -33 embryos are hyaluronate, the chondroitins and unidentified
sulphated GAG (dermatan sulphate, keratan sulphate, heparin and heparan
sulphate). Hyaluronate, chondroitin and the unidentified sulphated GAG
constitute about 33 % each of the GAG being synthesized in the stage-24
presumptive skull. At stage 33, the level of hyaluronate synthesis drops to 2 %, the
chondroitins to 24% and the unidentified sulphated GAG increases to 74%.
There is an 18-5-fold decrease in the percentage of hyaluronate, a 1-5-fold
decrease in the amount of chondroitins and a 2-7-fold increase in the percentage of
unidentified sulphated GAG being synthesized as skull morphogenesis proceeds.
MATERIALS AND METHODS
(A) Culture of embryos
(1) In ovo
White Leghorn chick embryos (stage 0 after Hamburger & Hamilton, 1951) were incubated in
a forced air humidified incubator maintained at 37-5°C and 99-9% humidity. The incubator
rotated the eggs every hour to prevent embryos from sticking to the shell. Embryos were
cultured in ovo for 2 days then either were transferred to in vitro culture or continued incubation
in ovo for a total of 10 days.
(2) In vitro cultures
Stage-12 (2-day-old) embryos were cultured outside the shell via the method of Auerbach,
Kubai, Knighton & Folkman (1974). The fertilized egg was cracked open and the yolk plus a
small amount of white was placed in a 60x15 mm sterile glass Petri dish. A maximum of three
such cultures were then put into a sterile plastic 150x25 mm Petri dish, sterile water was poured
into the dish to ensure humidity, then the dish was covered. The cultures were incubated at
37-5°C, 99-5 % humidity in a BOD type incubator for the appropriate time necessary for the
embryos to develop to stage 24 (4 days).
(B) Morphometric analyses
Mesenchymal areas, total number of cells within this area and individual cell areas were
measured using the SMI Unicomp. The area of mesenchymal cells was calculated based on the
formula for the area of an ellipsoid, jia • b, where a is the long axis and b is the short axis of the
34
C. D. GOLDSTEIN, J. J. JANKIEWICZ AND M. E. DESMOND
Table 1. Summary of enzyme specificity for GAG1
Enzyme
Streptomyces hyaluronidase
Testicular hyaluronidase
Substrate
HA2
HA, CH, CH-S
Non-degraded GAG
(precipitate)
CH, CH-S3, DS4, KS5, HS6, Hep7
DS, KS, HS, Hep
*GAG, glycosaminoglycan.
HA, hyaluronic acid.
3
CH, chondroitin; CH-S, chondroitin sulphate.
4
DS, dermatan sulphate.
5
KS, keratan sulphate.
6
HS, heparan sulphate.
7
Hep, heparin.
2
ellipse. Areas were measured for the median section plus the two sections on either side of the
median section for three different embryos at stages 24 and 33 (Fig. 2). Cell densities were
measured for nine different areas of each of the sections used for area measurements and the
average used.
(C) Histochemical techniques
(1) Enzymatic treatment and alcian blue staining
Embryos of stages 24 and 33 were excised from the yolk. Both vitelline and amniotic
membranes were removed before the embryos were fixed with formalin containing 4%
cetylpyridinium chloride (CPC) for one week (Williams & Jackson, 1956). The embryos were
then dehydrated and cleared using a series of ethanol and butanol solutions. Processing included
exposing the embryos sequentially to 70% ethanol, 70% ENBA, 80% ENBA, 95 % ENBA
plus eosin, 100 % ENBA, and 1:1100 % ENBA:paraplast. ENBA is a mixture of ethanol and nbutyl alcohol where 95 % ENBA contains 5 parts 95 % ethanol: 95 parts n-butyl alcohol and
100% ENBA is 5 parts absolute ethanol: 95 parts n-butyl alcohol. Times of exposure of the
tissue to these solutions was 1 h for all solutions except those containing paraplast which were for
2h each (Humason, 1972).
Serial sections of stage-24 and -33 heads were made with an AO microtome at 5 jum section
thickness. Sections of five heads at each stage were wet mounted onto precleaned slides and
placed on a slide warmer to dry overnight. The sections were then hydrated by placing the slides
in xylene followed by a series of ethanol solutions for 5min each, and finally distilled water
(Humason, 1972). After hydration, the sections were treated with the following enzymes:
Streptomyces hyaluronidase, lOOTRUml"1, and testicular hyaluronidase type IV, 500 units ml" 1
(both enzymes were obtained from Sigma Chemical Co.). Streptomyces and testicular hyaluronidases were dissolved in a 0-02M-sodium acetate, 0-lM-sodium chloride solution, pH5-0
(Table 1). Three to four drops of the appropriate enzyme solutions were placed directly atop the
hydrated sections and the slides were incubated in a humidified chamber at 37-5°C for 4h.
Controls were prepared by placing buffer only atop sections and then subjecting the slides to the
same conditions as the enzyme-treated slides. After 4h, the slides were rinsed three times in
double-distilled water and placed in a 3 % glacial acetic acid solution for 3min. The sections
were then stained with 1 % alcian blue 8GX at pH2-5 or 1-0 for 2h, dehydrated with ethanol,
cleared and mounted with Permount. Both enzyme-treated and control sections were stained at
the low and high pH in order to distinguish between sulphated and non-sulphated GAG (Lev &
Spicer, 1964) (Table 2).
(2) Colorimetric and histochemical assays for hyaluronidase activity
Both a quantitative colorimetric assay and a less quantitative histochemical assay were used to
assess the activity of the hyaluronidases prior to using them to treat embryo sections. The
Glycosaminoglycans in the embryonic chick
35
colorimetric method used was originally devised by Aminoff, Morgan & Morgan (1952), refined
by Reissig, Strominger & Leloir (1955) and further refined by Hatae & Makita (1975).
Four materials were used in this assay: a buffer (0-8M-potassium borate, pH9-l); a chromagen (p-dimethylaminobenzaldehyde (DMAB); the substrate (hyaluronate); and the enzyme
{Streptomyces hyaluronidase). The chromagen was made by placing 10 g of DMAB in 100 ml of
glacial acetic acid that contained 12-5 % (v/v) of 10 N-HCI. This stock solution was diluted to 1:9
(v/v) with glacial acetic acid prior to use in the assay. The substrate, hyaluronate, was made in
the concentrations of SOmgml"1, 40mgml~1, SOmgrnP1, 20mgml~1, and lOmgml"1 of
hyaluronate in 1-0ml distilled water. All concentrations of substrate were done in triplicate.
The enzyme, Streptomyces hyaluronidase, had a concentration of 40TRU per 1-0 ml of distilled
water.
The assay was performed in the following manner: to 200 jul of solutions of the various
hyaluronate concentrations was added 50jUl (2TRU) of hyaluronidase solution. All tubes were
incubated at 60°C for 2 h, after which time the tubes were placed in a boiling water bath for 3 min
and then cooled to room temperature with running tap water. Next, 20 ml of diluted DMAB
solution was added to each tube with immediate shaking. The tubes were left standing at 37°C
for 20 min to allow for full colour development; the tubes were then cooled again to room
temperature with running tap water. Using a Bausch and Lomb Spectronic 21 and quartz
cuvettes, all tubes were read at 585 nm. A linear regression analysis was done with O.D. values
and corresponding concentrations in order to construct a standard curve. This colorimetric assay
was done for every new batch of enzyme.
The routine assay for determining hyaluronidase activity was to treat hydrated sections of
human umbilical cord with the enzyme. Umbilical cord was chosen because of its high content of
hyaluronate (Wharton's Jelly). Following incubation at conditions identical to those cited
previously for the embryo sections, the umbilical sections were stained with alcian blue at
pH2-5. The intensity of blue was compared between treated and control sections. A 50 to 75 %
loss in alcian blue staining was used as the criterion that the hyaluronidase was active and could
be used for the embryonic sections.
(D) Identification of GAG using selective enzymatic degradation procedure
(1) Incorporation of [3H]glucosamine
A precursor of all GAGs, glucosamine hydrochloride, D-(1,6-H(N)) - specific activity
of 38-9Ci minor 1 , NEN)), (Kim & Conrad, 1976), was applied directly atop stage-24 and -33
embryos. Stage-24 embryos were in whole embryo culture at the time of pulse labelling. Since it
is difficult for large numbers of embryos to survive to stage 33 in whole embryo cultures, these
older embryos were left in their shells and a small rectangular opening
was made through which the isotope was applied. Stage-24 embryos received 0-5 ml
[3H]glucosamine (6-25JuCi0-5ml~1) and stage-33 embryos were pulsed with 1-0 ml [3H]glucosamine {\2-5fiC\m\~l). After 4h of incubation (37-5°C) in the presence of the isotope, the
embryos were excised from the yolk and rinsed with chick saline solution. Excess yolk and the
vitelline membranes were removed, followed by extirpation of the eyes. The heads were
transected from the embryo axis at the posterior border of the metencephalon, pooled, and
blotted as dry as possible on filter paper to prevent excess water being included in the weight of
each pool. The number of heads and pool weight of the stages were kept consistent, with a pool
of stage-24 heads averaging 13 heads weighing about 200 mg. Because the heads of the older
Table 2. Summary of differential staining of GAG1 by alcian blue
Conditions
pH 1 • 0
pH2-5
glycosaminoglycan.
Alcian-blue-positive
sulphated GAG only
all GAG
36
C. D. GOLDSTEIN, J. J. JANKIEWICZ AND M. E. DESMOND
Tissue (200 mg)
Homogenize (2 ml volume)
0-2 M-Tris buffer
1-9 ml homogenate
0 1 ml for Lowry assay
protease digestion
24hat37°C
boil
cool
Hydrolyse
NaOH
I
Dilute in
HOH
J
i'
Colour reaction
with phenol
Counts
St. hyase
T. hyase
Boil
Boil
Buffer (Na Ac)
Boil
Add carrier
(HA, Ch-S)
Add carrier
(HA, Ch-S)
Add carrier
(Ha, Ch-S)
Add CPC
Add CPC
Add CPC
Discard
Counts
Discard Counts
Discard
Fig. 3. A synopsis of the biochemical protocol used to identify hyaluronate,
chondroitin sulphate and the rest of the sulphated GAGs. The tissue homogenate
consisted of heads minus eyes.
embryos were so large, only one head was used per assay so that it was not a pool. Individual
heads of these stage-33 embryos averaged 364 mg. All weighed samples were frozen at -60°C.
(2) Selective enzyme degradation of GAG
Each sample (pool or individual) of chick embryo heads was thawed and homogenized in
1-0 ml of 0-2 M-Tris (pH7-8) buffer using a Brinkman Instruments polytron for 5 s at the
maximum setting (Solursh, 1976). After homogenization, the polytron was rinsed with 1-0 ml of
buffer and the rinse was added to the homogenate to bring the total volume of each test pool to
2-0 ml. A 0-1 ml aliquot was taken from the homogenate and put into a test tube containing
0-9 ml of 1 N-NaOH and stored at 4°C for future protein determination (Fig. 3).
To ensure as complete digestion of GAG as possible, linker and core proteins were
enzymatically digested using a non-specific bacterial protease made from Streptomyces griseus
(Sigma). The concentration of protease used for digestion was 1-Omgml"1 of a solid with 5-8
units of proteolytic activity per mg. Protease of the same concentration was again added to the
homogenate 12 h after the first treatment. All tubes were incubated at 37-5°C and continually
shaken for a total of 24 h at which time the tubes were placed in a boiling water bath for 10 min to
stop enzymatic activity.
Glycosaminoglycans in the embryonic chick
37
The enzymes used for the degradation of GAG were Streptomyces hyaluronidase (500 i.u.) and
bovine testicular hyaluronidase (3000 i.u.) (Table 1). Selective enzyme degradation of GAG was
performed by taking three 0-5 ml aliquots from the homogenate and placing them into 15 ml
glass centrifuge tubes. Two aliquots were enzymatically treated: one with Streptomyces
hyaluronidase and the other with bovine testicular hyaluronidase. A 1-0 ml quantity of 0-02 Msodium acetate, 0-lM-sodium chloride buffer (pH5-0) with enzyme was added to bring the
volume to 1-5 ml in each tube. To the third aliquot (control), 1-0ml of buffer only was added
(Toole & Gross, 1971; Solursh, 1976). A 10 ml volume was used to optimize the pH to enhance
optimal activity of the hyaluronidases (Ohya & Kaneko, 1970; Toole, 1973). Preliminary
experimentation showed that two volumes of acetate buffer (pH5-0) to one volume of Tris
buffer (pH7-8) properly corrected the pH. The final pH was approximately 5-5. After 4h of
incubation (37-5 °C) the tubes were placed in a boiling water bath for lOmin to arrest enzyme
activity.
After letting the tubes cool to room temperature (20°C), 1-0 mg each of carrier hyaluronate
and chondroitin sulphate was added to the tubes. This was done to increase the yield of
precipitate of non-digested GAG, specifically hyaluronate and the chondroitin sulphates.
Precipitation of undigested GAG was done by adding 0-5 ml of a 4-0% CPC solution to give
each tube afinalvolume of 2-0 ml and a CPC concentration of 1-0 %. The precipitates were then
centrifuged (Sorvall RC-5) for 30min at 12 000 g. The pellet was washed three times by
suspending it in 95 % ethanol and then centrifuging as before. After the third rinse, the pellet
was suspended in 1-0 ml of methanol. The 1-0 ml of methanol with the precipitate was dissolved
in 10ml of Aquasol (New England Nuclear). Scintillation vials were counted twice for lOmin
using a liquid scintillation counter (Packard C-2425).
(3) Determination of protein
Protein content per pool of homogenated chick embryo heads was determined by using the
colorimetric assay of Lowry, Rosebrough, Farr & Randall (1951). A standard curve was plotted
by preparing a standard protein solution containing crystalline bovine serum albumin at a
concentration of 0-05 g 100ml"1. From this standard solution, aliquots of 0-10, 0-20, 0-25, 0-30,
0-40, 0-50, and 0-70 ml were taken and put into separate test tubes. Each amount was measured
out in triplicate. Double-distilled water was added to all the tubes to bring the volume up to
1-0 ml.
Solutions of 2-0% sodium carbonate (w/v) in 0-lN-NaOH and 0-5% copper sulphate
pentahydrate (w/v) in 1-0 % potassium tartrate (w/v) were made. These solutions were mixed
together 50:1 respectively. Then 5-0 ml of this freshly mixed reagent was added to every tube
that contained the dilutions of the standard proteins. The tubes were mixed with a vortex mixer
and let stand to react at room temperature (20°C) for lOmin. After that time, 0-5ml of 1-0NFolin-Ciocalteau phenol reagent was added to every tube. The tubes were shaken and placed in
an incubator (37°C) for at least 30 min to allow for full colour development. The tubes were then
read using a spectrophotometer at 500 nm. After the O.D. values were recorded, a standard
curve was plotted using linear regression.
The unknown samples, saved after homogenization, were placed in a 90°C water bath for
30 min. Then a 0-1 ml aliquot was taken from each unknown test sample and placed in tubes
containing 0-9 ml of double-distilled water. Experimental samples were run through the assay
the same way as the ones in determining the standard curve. From the O.D. values recorded for
the test samples, an accurate determination of the amount of protein in each homogenated pool
of chick embryo heads was made by referring to the standard curve and allowing for the dilution
factor of the original homogenate concentration.
RESULTS
(A) Morphometric and histological findings
Histological examination of comparable sagittal sections of embryonic heads
from stage-24 and -33 chicks showed at stage 24, the skin ectoderm was lifted away
38
C. D. GOLDSTEIN, J. J. JANKIEWICZ AND M. E. DESMOND
from the neuroepithelium throughout the entire head due to the thickness of the
head mesenchyme. Foci of chondrogenesis were first apparent in the head
mesenchyme at stage 33 based on assessment of analogous sections from stages 24
to 33. Morphometric analysis of the head mesenchyme region for these two stages
showed a 60-fold enlargement of the mesenchymal area between these two stages
while cell density and cell area of individual cells remain unchanged (Table 3)
suggesting that the increase in mesenchymal area is primarily due to the increase in
ECM. Measurements of the ECM support the idea that most of the increase in
mesenchymal area is due to an increase in ECM. Roughly 65 % of the total
mesenchymal area is ECM whereas 35 % is cellular (Table 3). These increases in
ECM are primarily in the ventral half of the head corresponding to the
presumptive facial, ethmoid, sphenoid and periotic regions.
(B) Histochemical findings
(1) Localization and identification of GAG in stage-24 presumptive skulls
Sections stained with alcian blue and examined with bright-field optics showed
distinct presence or absence of blue material in specific regions. Most notable was
that GAGs were present only in the head mesenchyme (presumptive chondrocranium) and basement membrane of the neuroepithelium and not either the
neuroepithelium proper or skin ectoderm. Dark-field optics were employed since
photomicrographs utilizing appropriate filters with bright-field optics did not
clearly show alcian blue material as distinct from unstained regions, especially
in comparing controls with enzyme-treated sections. In order to maximize
localization of alcian blue material in sections of whole heads, low magnification
was used and thus resolution was sacrificed. Nonetheless, dark-field optics
consistently enhanced the presence of alcian blue material when compared to
bright-field optics of the same material. The bright-field image was the kind of
alcian blue result previously published by one of us (Schoenwolf & Desmond,
1984). The photographs do show distinct differences between alcian blue material
differentially stained and on sections pretreated with enzymes (Figs 4, 5). The
Table 3. Proportion of total mesenchyme area of the presumptive skull allocated to cell
and ECM1 areas for stage-24 and -33 chick embryos
XT
2
Stage
24
£
/-> u
No. of
Cell
3
cells density4
5492
Total areas (um 2 )
TJ
Ind.
cell area5
0-005 75-12+18-58
Mesenchyme
1078886
Cell only
412559
1L :
% ECM only
0-38
666327
%
0-62
33 239805 0-004 85-39+16-95 64987821
20476949 0-32 44510872 0-68
ECM = extracellular matrix.
2
Stages according to Hamburger & Hamilton (1951).
3
Number of cells in the total mesenchymal area.
4
Total number of cells per total mesenchymal area.
5
Calculations based on area of an ellipsoid (mib, where a - the major axis and b = the minor
axis).
1
Glycosaminoglycans in the embryonic chick
39
neuroepithelium appears brighter than any other material in the head with darkfield optics. This degree of brightness was never blue under bright field nor
magenta under dark field, and thus was never interpreted to be due to the presence
of alcian blue material. On the other hand, the head mesenchyme was blue under
bright field, magenta under dark field and less bright compared to the neuroepithelium on photographs with dark-field optics.
In the head mesenchyme there is a striking difference in the intensity of alcian
blue material of stage-24 embryos stained at pHl-0 and 2-5. Sections from the
same embryo exhibited almost no alcian blue material at pHl-0 in contrast to a
high intensity of material at pH2-5 (Fig. 4A,B). The stained regions were most
prominent in the ventral half of the head mesenchyme in the region of the
mesencephalon and prosencephalon and in the perinotochordal area. These
regions would correspond to the future face as well as the ethmoid, presphenoid,
basisphenoid, alisphenoid, squamosal and periotic portions of the future chondrocranium. The dorsal half of the skull in the same brain regions was faintly stained.
Stage-24 sections stained with alcian blue after Streptomyces hyaluronidase
treatment exhibited much less alcian blue material in the ventral region compared
to the control section stained at the same pH (2-5) (Fig. 4A,C). Likewise, similar
sections treated with testicular hyaluronidase showed identical staining with alcian
blue as those sections treated with Streptomyces hyaluronidase (Fig. 4C,D). Both
results of the differential pH staining and enzyme treatments suggest that the
majority of GAGs present in the skull of stage-24 embryos is hyaluronate and the
chondroitins.
(2) Localization and identification of GAG in stage-33 presumptive skulls
Alcian-blue-stained sections of stage-33 presumptive skulls like the stage-24
sections exhibited a positive reaction only in the ventral region of the presumptive
skull at pH2-5. The staining was not uniformly distributed throughout the ventral
region, but was evident in specific areas, such as the perichordal region,
presumptive facial area and the presumptive ethmoid, sphenoid, temporal and
periotic regions of the chondrocranium (Fig. 5A). Furthermore, the alcian blue
material was condensed into specific foci (Fig. 5A).
Sections treated with both Streptomyces and testicular hyaluronidase prior to
alcian blue staining exhibited no reduction in alcian-blue-positive material when
compared to the control sections (Fig. 5A,B,C). This clearly shows that the GAGs
contributing to alcian blue staining are other than hyaluronate or the chondroitin
sulphates.
(C) Biochemical findings: identification of GAG via selective enzyme degradation
The histochemical findings localized all GAGs in the heads of both age groups to
the presumptive chondrocranium, i.e. head mesenchyme plus neuroepithelial
basement membrane. This means that all of the [3H]glucosamine that was
incorporated into the head at the time of the pulse-labelling experiments was
utilized by GAGs in only the presumptive chondrocranium region. For stage-24
40
C. D. GOLDSTEIN, J. J. JANKIEWICZ AND M. E. DESMOND
embryos, the presumptive chondrocranium exhibits approximately equal amounts
of synthesis of hyaluronate, chondroitins and all other GAGs (Table 4). Based on
the c.p.m. of CPC precipitates, hyaluronate synthesis accounts for 37-34 % of the
GAG, the chondroitins for 35-33%, while 27-33% is one or more of all other
GAGs, i.e. dermatan sulphate, keratan sulphate, heparan sulphate and/or
heparin (Table 5). The average number of heads per pool for this biochemical
Fig. 4. A-D. Photomicrographs using dark-field optics of median and near median
sections of whole heads from stage-24 chick embryos. Sections A,B were stained with
alcian blue at pH2-5 (A) and pHl-0 (B). Section C was treated with Streptomyces
hyaluronidase prior to alcian blue staining at pH2-5, and section D was pretreated with
testicular hyaluronidase. Alcian blue material within the head and upper neck mesenchyme is most notable in Fig. 4A indicated by arrows. Lack of alcian-blue-stained
material in B,C,D suggest that the majority of GAGs stained with alcian blue is
hyaluronate and the chondroitins. Magnification: x4.
Glycosaminoglycans in the embryonic chick
Fig. 5. A-C. Dark-field photomicrographs of sections of stage-33 chick embryo
heads. Section A was stained with alcian blue at pH2-5; B,C were stained with alcian
blue at pH2-5 after pretreatment with Streptomyces hyaluronidase (B) and testicular
hyaluronidase (C). Foci of chondrogenesis are identified within the arrows. The fact
that the staining profile is the same for all three treatments suggests that the
predominant GAGs present in the head mesenchyme at this time are GAGs other than
hyaluronate and the chondroitins. The arrow indicates the area containing the alcian
blue material. Magnification: x4.
41
42
C. D. GOLDSTEIN, J. J. JANKIEWICZ AND M. E. DESMOND
Table 4. Percentage of GAG1 in pooled heads (stage 24) and individual heads
(stage 33) of chick embryos
Stage2
Sample
24
A
B
C
X
33
A
B
C
D
X
% HA3
34
42
36
37
0
1
7
0
2
% Ch; Ch-S4
38
32
36
35
20
24
22
31
24
%DS;Ks;HS;Hep 5
28
26
28
27
80
75
72
69
74
, glycosaminoglycan.
Staging after Hamburger & Hamilton (1951).
3
HA, hyaluronic acid.
4
Ch,
chondroitin; Ch-S, chondroitin sulphate.
5
DS, dermatan sulphate; KS, keratan sulphate; HS, heparan sulphate; Hep, heparin.
2
assay was 13, the average weight of each pool was 210 mg and the average amount
of protein was 9-86 mg (Table 5).
Synthesis of GAG in the presumptive chondrocrania of stage-33 embryos
consists of 2 % hyaluronate, 24-25 % chondroitins and 73-75 % dermatan sulphate,
keratan sulphate, heparin and heparan sulphate (Tables 3,4). The average weight
per stage-33 head was 364 mg and the average amount of protein per head was
22-18mg.
A comparison of the relative amounts of GAG being synthesized at the two
developmental stages, 24 and 33, based on the amount of GAG per mg protein
indicates there is an 18-5-fold decrease in the percentage of hyaluronate, only a
1-5-fold decrease in the percentage of chondroitins and a 2-7-fold increase in the
percentage of unidentified sulphated GAG as skull morphogenesis proceeds
(Fig. 6).
DISCUSSION
The histochemical and biochemical data of this study must be considered
together in order to make any correlation of GAG synthesis with chondrocranium
morphogenesis. Since it was impossible to isolate the presumptive chondrocranium tissue from the brain and skin ectoderm, it was necessary to show that all
of the GAG in the head was restricted to the presumptive chondrocranium. The
alcian blue staining clearly demonstrates this, to be the case for the heads from
embryos at stages 24 and 33. Similarly, it was impossible to separate the facial area
from the neurocranium proper of the presumptive skull for the biochemical assay.
Based on the histochemistry, the alcian blue material appeared to be distributed
equally between the presumptive face and neurocranium.
In order to determine the relative proportions of GAG present in the
chondrocranium during development, it was necessary to quantify the amount of
Sample
A
B
C
X
A
B
C
D
X
Weight
(mg)
203
210
217
210
351
334
312
460
364
Protein
(mg)
9-00
9-86
10-72
9-86
18-78
16-78
19-94
33-20
22-18
% protein
4-4
4-7
4-9
4-7
5-4
5-0
6-4
7-2
6-0
HA 3
(c.p.m.)
910
870
1021
934
0
9
834
0
211
2
GAG, glycosaminoglycan.
Staging after Hamburger & Hamilton (1951).
3
HA, hyaluronic acid.
4
CH, chondroitin; CH-S, chondroitin sulphate.
5
DS, dermatan sulphate; KS, keratan sulphate; HS, heparan sulphate; Hep, heparin.
1
33
Stage
24
2
CH; Ch-S4
(c.p.m.)
1027
675
1030
911
69
179
2623
19
723
DS; KS; HS; Hep 5
(c.p.m.)
740
534
817
697
278
564
8501
54
2349
Table 5. CPM of GAG1 per mg protein for heads of chick embryos at stages 24 and 33
All GAG
(c.p.m.)
2677
2079
2868
2541
374
752
11958
73
3289
§
B"
Q.
c£
3.
©
!§.
§
£
§'
S*
|.
44
C. D. GOLDSTEIN, J. J. JANKIEWICZ AND M. E. DESMOND
40-i
HA
CH
Ch-S
Stage-24 skull
HA
CH
Ch-S
Stage-33 skull
Fig. 6. This bar graph represents the percentages of hyaluronate (HA) and the
chondroitins, i.e. chondroitin (CH) and chondroitin sulphate (CH-S) being synthesized by the head mesenchyme at two developmental periods. Stages 24 and 33
represent times prior to and at the onset of chondrogenesis of the head mesenchyme.
alcian blue material in sections using microspectrophotometry or densitometry, or
to use an entirely different approach. We chose to use the pulse-labelling method
followed by selective enzyme degradation. The major drawback of this pulselabelling method is that one is limited to measuring the amount of GAG being
synthesized at the time of adding the label with no knowledge of the rates of
degradation of GAGs during the time periods of labelling. In other words, this
methodology does not provide identification of the total amount of a specific GAG
present. Consequently, correlating GAG type with specific morphogenetic events
must be limited to correlating the synthesis of specific GAGs with morphogenesis.
Whether it is the synthesis or total amount of GAG type that may influence a
morphogenetic event remains to be examined. Thus far, developmental biologists
have limited morphogenetic correlations to synthesis of a specific GAG (Toole,
1981). Likewise, this study correlates skull morphogenesis with synthesis of
specific GAG types and has" the same limitations.
The histochemical findings that no GAG was detected in the neuroepithelium of
the heads examined in this study contradict the report of Toole (1976), which cited
an increase of GAG in the brain as the brain developed. It is not clear from that
report by Toole whether the brains were isolated from the head mesenchyme. So
Glycosaminoglycans in the embryonic chick
45
based on our study the question remains whether this reported increase in GAG
synthesis was really for the head not the brain.
Localization of GAG in the basal lamina of the neuroepithelium parallels the
finding of others who have shown that the neuroepithelium secretes GAG (Hay &
Meier, 1974). The absence of GAG in the basal lamina of the skin ectoderm differs
from what is reported by Fisher & Solursh (1977), for younger embryos. Such an
absence may indicate that the GAG present in the basal lamina of the ectoderm
has migrated into the ECM of the head mesenchyme as shown in other developing
systems such as the teeth (Thesleff & Pratt, 1980) and heart (Markwald &
Funderburg, 1983). Although this study did not identify specifically the GAG
located in the basal lamina, the differential staining with alcian blue suggests a high
concentration of sulphated GAG in the basement membrane. Fisher & Solursh
(1977) showed that the predominant GAG in the basal lamina of the ectoderm is
hyaluronic acid. Others showed that chondroitin sulphate is most prominent in the
neuroepithelium (Hay & Meier, 1974; Trelsted, Hayashi & Toole, 1974). These
differences in type of GAG being synthesized by the two different epithelia could
suggest that the roles of the two GAG are different.
It may be that the chondroitin sulphate from the neuroepithelium serves as a
template onto which chondrogenesis and/or osteogenesis shape the skull. In other
words, the chondroitin sulphates might serve as an epithelial-mesenchymal
interface transferring the structural nuances of the brain onto the developing skull.
Such an epithelial-mesenchymal interface could explain the highly refined
morphological similarities between the brain and skull. The GAG interface could
be instructive like an inductor or merely serve as a template. Schowing (1968) used
selective deletion of the prosencephalon and mesencephalon to show that the
frontal bone derived in part from the prosencephalic crest received inductive
signals from both the prosencephalon and mesencephalon.
The histochemistry results show that most of the GAGs are located in the
ventral half of the presumptive chondrocranium at the embryonic stages studied.
This half of the chondrocranium consists primarily of endochondral bones, i.e.
bones formed via cartilage model precursors. The correlation of different
proportions of sulphated to non-sulphated GAGs in the ventral skull over time
may suggest that GAGs function as environmental regulators of endochondral
bone formation. Hall (1968, 1970) has reported that formation of secondary
cartilage at sites of articulation or attachment of muscle, ligaments or tendons
occurs in response to a deposition of GAG. On the other hand, the correlation
may have no causal effect on bone morphogenesis whatsoever but simply
represent the GAGs present in the head mesenchyme during the period that skull
morphogenesis occurs.
The chondroitin sulphates present in the ECM once chondrogenesis has begun
are probably secreted by the chondrocytes as has been shown for other developmental systems (Lash, Holtzer & Whitehouse, 1960; Franco-Browder, DeRydt &
Dorfman, 1963; Kvist & Finnegan, 1970a,b; O'Hare, 1973; Fisher & Solursh, 1977,
1979; Solursh, 1976; Solursh & Morriss, 1977; Morriss & Solursh, 1978). Such a
46
C. D. GOLDSTEIN, J. J. JANKIEWICZ AND M. E. DESMOND
chondroitin-sulphate-rich matrix appears to be necessary for maintaining chondrocytes in the differentiated state (Holtzer, 1964; Lash, 1968). Not only is
chondroitin sulphate necessary for maintenance of chondrocytes in the differentiated state but very low concentrations of hyaluronate or even a tetrasaccharide
product of hyaluronidase digestion totally prevents chondrogenesis (Toole,
Jackson & Gross, 1972; Toole, 1973).
The fact that there is a higher proportion of chondroitin sulphate being
synthesized prior to than at the onset of chondrogenesis seems somewhat contradictory and could indicate several things. As stated above, it appears that the
drop in hyaluronate synthesis serves more as a stimulator of chondrogenesis than
does the increase in synthesis of chondroitin. Furthermore, it may be that the
proportional amounts of these GAGs present in the particular site of differentiation are more critical to regulating differentiation than is the proportional
synthesis of these GAGs. This study only measured the actual synthesis at the two
times. Finally, chondrogenesis may be initiated and/or maintained by one of the
other GAGs not identified in this study such as keratan sulphate. Keratan sulphate
chains are usually present with and covalently linked to the same protein as
chondroitin sulphate (Hay, 1981).
Hyaluronate from the basal laminae of both of the epithelia could increase the
extracellular space. Such an increase in extracellular space is in agreement with
other studies reported in the literature of developmental systems at a time of
undifferentiated cell migration. Mesenchymal cell migration in the presumptive
skull at stage 24 occurs at the same time that cell migration is occurring in the
corneal stroma and from the sclerotome (Kvist & Finnegan, 1970a; Toole &
Trelsted, 1971). Hyaluronate is the major GAG present at this time and probably
initiates cell migration because of its ability to create cell-free spaces and/or
prevent cell-cell interaction. It is thought that hyaluronate which is highly
polyanionic due to the presence of numerous carboxyl moieties creates cell-free
spaces by entrapping water. As a result a space highly disproportionate to the
concentration of hyaluronate is formed.
Additionally, there is evidence that cells may have GAG receptors on their
plasma membranes that facilitate cell-to-cell surface binding (Toole, 1981). GAG
receptor binding may occur via plasma membrane proteins that covalently bind to
the GAGs or by a reaction between the GAG and cell surface receptor sites. An
area rich in hyaluronate would effectively bind up all hyaluronate receptors,
negating any chance for cell-to-cell contact. Therefore, cells would be more free to
move. Studies have shown that cell lines in suspension aggregate when a moderate
amount of hyaluronate is present. When hyaluronidase or excess hyaluronate is
added, aggregation of cells does not occur (Underhill & Toole, 1981). Hyaluronate and chondroitin sulphate are also implicated as weakening the initial
cell-substratum attachments of cells grown in vitro (Culp, Murray & Rollins,
1979). These investigators report that heparan sulphate on the cell surface, along
with fibronectin, initiates binding to the substratum. A weakening of such
adhesion promotes movement. Just how these in vitro findings relate to cell
Glycosaminoglycans in the embryonic chick
47
migrations in vivo is not known but hyaluronate synthesis is elevated during
migration in many morphogenetic systems as discussed in the Introduction.
However, a very direct test of the effect of hyaluronate on neural crest migration in
vivo was reported by Anderson & Meier (1982). They treated embryos with
Streptomyces hyaluronidase in concentrations that significantly reduced normal
amounts of hyaluronate in the ECM and prevented neural tube fusion but not
neural crest migration.
This work was supported by a Villanova University Summer Grant to M.E.D. and a grant
from Sigma Xi to C.D.G. We acknowledge Optical Apparatus, Ardmore, PA for their generous
loan of a dark-field base for use with the Nikon SMZ-10, Mary Pacheco for doing the
morphometric measurements and Dr Russell Gardner for his critical evaluation of the
biochemical methodology.
REFERENCES
AMINOFF, D., MORGAN, W. T. J.
& MORGAN, W. M. (1952). Studies in immunochemistry. II. The
action of dilute alkali on the n-acetyl hexosamines and the specific blood group mucoids.
Biochem. J. 51, 379-389.
ANDERSON, C. & MEIER, S. (1982). Effects of hyaluronidase treatment on the distribution of
cranial neural crest cells in the chick embryo. /. exp. Zool. 221, 329-335.
AUERBACH, R., KUBAI, L., KNIGHTON, D. & FOLKMAN, J. (1974). A simple procedure for the
longterm cultivation of chicken embryos. DevlBiol. 41, 391-394.
BELSKY, E. & TOOLE, B. P. (1983). Hyaluronate regulation in the mesonephros and metanephros
of avian embryos. Cell Differ. 12, 61—66.
CULP, L. A., MURRAY, B. A. & ROLLINS, B. J. (1979). Fibronectin and proteoglycans as
determinants of cell substratum adhesion. /. supramol. Struct. 11, 401-427.
DERBY, M. A. (1978). Analysis of glycosaminoglycans within the extracellular environments
encountered by migrating neural crest cells. Devi Biol. 66, 321-336.
FISHER, M. & SOLURSH, M. (1977). Glycosaminoglycan localization and role in maintenance of
tissue spaces in the early chick embryo. J. Embryol. exp. Morph. 42,195-207.
FISHER, M. & SOLURSH, M. (1979). The influence of local environment on the organization of
mesenchyme cells. /. Embryol. exp. Morph. 49, 295-309.
FRANCO-BROWDER, S., DERYDT, J. & DORFMAN, A. (1963). The identification of a sulfated
mucopolysaccharide in chick embryos, stages 11-23. Proc. natn. Acad. Sci. U.S.A. 49,
643-647.
HALL, B. (1968). Histochemical aspects of the differentiation of adventitious cartilage on the
membrane bones of the embryonic chick. Histochem. 16, 206-220.
HALL, B. (1970). Cellular differentiation in skeletal tissues. Biol. Rev. Camb. Phil. Soc. 45,
455-484.
HAMBURGER, V. & HAMILTON, H. (1951). A series of normal stages in the development of the
chick embryo. /. Morph. 88, 49-92.
HATAE, Y. & MAKTTA, A. (1975). Colorimetric determination of hyaluronate degraded by
Streptomyces hyaluronidase. Anal. Biochem. 64, 30-36.
HAY, E. D. (1981). Collagen and embryonic development. In Cell Biology of the Extracellular
Matrix (ed. E. D. Hay), chap. 12, pp. 379-405. New York: Plenum Press.
HAY, E. D. & MEIER, S. (1974). Glycosaminoglycan synthesis by embryonic inductors: neural
tube, notochord, and lens. /. Cell Biol. 62, 889-898.
HOLTZER, H. (1964). Control of chondrogenesis in the embryo. Biophys. J. 4 Supplement,
239-250.
HUMASON, G. (1972). Animal Tissue Techniques, 3rd edn, pp. 393-405. San Francisco: Freeman
Press.
48
C. D . G O L D S T E I N , J. J. J A N K I E W I C Z AND M. E .
DESMOND
M. (1966). A radioautographic study of the migration and fate of cranial neural crest
cells in the chick embryo. Anat. Rec. 156,143-156.
KIM, J. J. & CONRAD, H. E. (1976). Kinetics of mucopolysaccharides and glycoprotein synthesis
by chick embryo chondrocytes. /. biol. Chem. 251, 6210-6217.
KOSHER, R. A. & SAVAGE, M. P. (1981). Glycosaminoglycan synthesis by the apical ectodermal
ridge of chick limb. Nature, Lond. 291, 231-232.
KVIST, T. N. & FINNEGAN, C. V. (1970a). The distribution of glycosaminoglycans in the axial
region of the developing chick embryo. I. Histochemical analysis. J. exp. Zool. 175, 221-240.
KVIST, T. N. & FINNEGAN, C. V. (1970ft). The distribution of glycosaminoglycans in the axial
region of the developing chick embryo. II. Biochemical analysis. /. exp. Zool. 175, 241-258.
LASH, J. (1968). Chondrogenesis: genotypic and phenotypic expression. /. Cell Physiol. 72,
35-46.
LASH, J., HOLTZER, H. & WHITEHOUSE, M. (1960). In vitro studies on chondrogenesis: the uptake
of radioactive sulfate during cartilage induction. Devi Biol. 2, 76-89.
LELIEVRE, C. (1978). Participation of neural crest-derived cells in the genesis of the skull in birds.
/. Embryol. exp. Morph. 47, 17-37.
LEV, R. & SPICER, S. S. (1964). Specific staining of sulfate groups with alcian blue at low pH.
J. Histochem. Cytochem. 12, 309.
LOWRY, O., ROSENBROUGH, M., FARR, A. & RANDALL, R. (1951). Protein measurement with the
Folin reagent. /. biol. Chem. 193, 265.
MANASEK, F. (1975). The extracellular matrix: a dynamic component of the developing chick
embryo. Curr. Top. devl Biol. 10, 35-102.
MARKWALD, R. R. & FUNDERBURG, F. M. (1983). Use of 6-diazo-5-oxo-L-norleucine to study
interaction between myocardial glycoconjugate secretion and endothelial activation in the
early embryonic chick heart. Devi Biol. 99, 395-407.
MEIER, S. & HAY, E. (1973). Synthesis of sulfated glycosaminoglycans by embryonic corneal
epithelium. Devi Biol. 35, 318-331.
MORRISS, G. & SOLURSH, M. (1978). Regional differences in mesenchymal cell morphology and
glycosaminoglycans in early neural-fold-stage rat embryos. /. Embryol. exp. Morph. 46,
37-52.
NODEN, D. (1975). An analysis of the migratory behavior of avian cephalic neural crest cells.
Devi Biol. 42, 106-130.
O'HARE, M. J. (1973). A histochemical study of sulfated glycosaminoglycans associated with the
somites of the chick embryo. /. Embryol. exp. Morph. 29, 197-208.
OHYA & KANEKO (1970). Novel hyaluronidase from Streptomyces. Biochim. biophys. Ada 10,
607-609.
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.
REISSIG, J. L., STROMINGER, J. L. & LELOIR, L. F. (1955). A modified colorimetric method for the
estimation of N-acetylamino sugars. /. biol. Chem. 217, 959-966.
SCHOENWOLF, G. C. & DESMOND, M. E. (1984). Description studies of occlusion and reopening of
the spinal canal of the early chick embryo. Anat. Rec. 209, 251-263.
SCHOWING, J. (1968). Influence inductrice de l'encephale embryonnaire sur le developpement du
crane chez le poulet. III. Mise evidence du role inducteur de l'encephale dans l'osteogenese
du crane embryonnaire du poulet. J. Embryol. exp. Morph. 19, 83-94.
SOLURSH, M. (1976). Glycosaminoglycan synthesis in the chick gastrula. Devi Biol. 50, 525-530.
SOLURSH, M., FISHER, M. & SINGLEY, C. T. (1979). The synthesis of hyaluronic acid by ectoderm
during early organogenesis in the chick embryo. Differentiation 14, 77-85.
SOLURSH, M. & MORRISS, G. M. (1977). Glycosaminoglycan synthesis in rat embryos during the
formation of the primary mesenchymne and neural folds. Devi Biol. 57, 75-86.
THESLEFF, I. & PRATT, R. M. (1980). Tunicamycin-induced alterations in basement membrane
formation during odontoblast differentiation. Devi Biol. 80,175-185.
TOOLE, B. P. (1973). Hyaluronate and hyaluronidase in morphogenesis and differentiation. Am.
Zool. 13,1061-1065.
JOHNSTON,
Glycosaminoglycans in the embryonic chick
49
B. P. (1976). Morphogenetic role of glycosaminoglycans (acid muco-polysaccharides) in
brain and other tissues. In Neuronal Recognition (ed. S. H. Barondes), pp. 275-329. New
York: Plenum Publishing Corporation.
TOOLE, B. P. (1981). Glycosaminoglycans in morphogenesis. In Cell Biology of the Extracellular
Matrix (ed. E. D. Hay), chap. 9, pp. 259-294. New York: Plenum Press.
TOOLE, B. P. & GROSS, J. (1971). The extracellular matrix of the regenerating newt limb:
synthesis and removal of hyaluronate prior to differentiation. Devi Biol. 25, 51-11.
TOOLE, B. P., JACKSON, G. & GROSS, J. (1972). Hyaluronate in morphogenesis: inhibition of
chondrogenesis in vitro. Proc. natn. Acad. Sci. U.S.A. 69, 1384-1386.
TOOLE, B. P. & TRELSTAD, R. L. (1971). Hyaluronate production and removal during corneal
development in the chick. Devi Biol. 26, 28-35.
TRELSTAD , R. L., HAYASHI , K. & TOOLE , B. P. (1974). Epithelial collagens and glycosaminoglycans
in the embryonic cornea. Macro-molecular order and morphogenesis in the basement
membranes. J. Cell Biol. 62, 815-830.
UNDERHILL, C. B. & TOOLE, B. P. (1981). Receptors for hyaluronate on the surfaces of parent and
virus-transformed cell lines. Binding and aggregation studies. Expl Cell Res. 131, 419-423.
WILLIAMS, G. & JACKSON, D. S. (1956). Two organicfixativesfor acid mucopolysaccharides. Stain
Technol. 31, 189-191.
TOOLE,
{Accepted 3 October 1985)