Subpopulations of chondrocytes from different zones of pig articular
cartilage
Isolation, growth and proteoglycan synthesis in culture
MARTIN SICZKOWSKP
Department of Anatomy and Developmental Biology, University College and Middlesex School of Medicine, Windeyer Building,
Cleveland Street, London W1P 6DB, UK
and FIONA M. WATT
Keratinocyte Laboratory, Imperial Cancer Research Fund, PO Box 123, Lincoln's Inn Fields, London WC2A 3PX, UK
* Present address: Leukaemia Research Fund Centre, Institute of Cancer Research, Chester Beatty Laboratories, Fulhatn Road,
London SW3 6JB, UK
Summary
Articular cartilage varies in ultrastructure and
composition with distance from the articular surface.
We have cultured chondrocytes from different zones
of pig articular cartilage and investigated whether
there are intrinsic differences in their behaviour that
might account for the variation observed in intact
tissue. On isolation, cells from the upper third of the
cartilage were smaller than those of the lower third,
but this difference was not maintained in culture.
Upper zone cells attached and spread more slowly
than lower zone cells; morphological differences
between the two populations could be seen for
several weeks. The growth rates of the two populations were similar, but upper zone cells reached a
lower confluent density. Levels of protein synthesis
were similar for both populations, but upper zone
cells deposited less proteoglycan in the cell layer. On
isolation, the percentage of upper zone cells that
stained positive with MZ15, a monoclonal antibody to
keratan sulphate, was smaller than the percentage of
lower zone cells, but this difference was lost after
several days in culture. Nevertheless, the keratan
sulphate content of proteoglycan synthesised by
lower zone chondrocytes at high density was greater
than of that synthesised by upper zone cells. The
proportion of nonaggregating proteoglycan was
greater in upper than lower zone cartilage and this
difference was also observed in long-term cultures.
Proteoglycans were further characterised by composite and polyacrylamide gel electrophoresis and by
immunoblotting; differences detected in cartilage
extracts were not, however, maintained in culture;
instead, the small proteoglycans synthesised by both
upper and lower zone cells varied with plating
density. Finally, alkaline phosphatase, a marker of
hypertrophic, calcifying cartilage, was only expressed in lower zone cultures. We conclude that
some of the observed heterogeneity of articular
cartilage reflects intrinsic differences between the
cells of different zones, whereas some may reflect the
response of chondrocytes to different environmental
conditions.
Introduction
zone, the cells are larger than in the other zones; they are
round and may be arranged in vertical columns. Finally,
between the deep zone and bone there is a layer of
hypertrophic cartilage.
In addition to variation in cell morphology and density
with distance from the articular surface, there is also
variation in the biochemical composition and physical
properties of the extracellular matrix. Thus, the packing
and orientation of collagen fibres change with depth from
fine, tangentially orientated in the superficial zone to
thick and vertically orientated in the deep zone (Ghadially, 1978; Poole et al. 1982). The amount of proteoglycan, its size, glycosaminoglycan composition and aggregation properties also change with distance from the
Articular cartilage covers the ends of bones where they
meet at joints and absorbs forces that are brought to bear
on the joints. It consists of a single cell type, the
chondrocyte, that secretes an extensive extracellular
matrix, the most-abundant components of which are type
II collagen and large proteoglycan molecules aggregated
with hyaluronic acid.
Several distinct zones of cartilage can be recognised by
histology (Palfrey and Davies, 1966). The superficial zone,
furthest away from the bone, is a thin zone of elongated
chondrocytes. The transitional, or middle, zone consists of
randomly oriented cells that are oval or round. In the deep
Journal of Cell Science 97, 349-360 (1990)
Printed in Great Britain © The Company of Biologists Limited 1990
Key words: chondrocytes, cartilage, proteoglycan.
349
articular surface (Maroudas et al. 1969; Franzen et al.
1981; Bayliss et al. 1983; Ratcliffe et al. 1984; Kuijer et al.
1986; Manicourt et al. 1988). Two proteins that are
specifically expressed in hypertrophic, calcifying, cartilage are type X collagen (Kielty et al. 1985; Schmid and
Linsenmayer, 1985;- Kwan et al. 1986) and alkaline
phosphatase (Fortuna et al. 1980; Hsu et al. 1985).
An important question that has been largely unexplored
is the extent to which the differences between different
cartilage zones reflect intrinsic differences between the
cells of those zones and the extent to which they are due to
environmental regulation. One approach to answering
this question was taken by Zanetti et al. (1985), who
showed that a monoclonal antibody that recognizes
keratan sulphate, MZ15 (Zanetti et al. 1985; Mehmet et al.
1986), stained the upper third of pig articular cartilage
more weakly than the deeper zones, and that the
proportion of cells isolated from the upper zone with
surface keratan sulphate was less than the proportion of
cells from the deeper zone. With time in culture, the
proportion of upper zone cells that stained positive with
MZ15 increased, so that the two subpopulations could no
longer be distinguished on the basis of MZ15 staining.
This suggested that keratan sulphate expression can be
modulated by the environment. However, differences in
the morphology of cells from different zones did persist
with time in culture, indicating that intrinsic differences
between chondrocyte subpopulations might exist (Zanetti
et al. 1985).
In this paper, we have carried out more extensive
studies on the properties of subpopulations of pig articular
chondrocytes in culture, which have enabled us to identify
both inherent and environmentally regulated differences
between the cells of different zones.
Materials and methods
Isolation of chondrocytes from different cartilage zones
The method used to isolate chondrocytes was essentially that
described by Zanetti et al. (1985). Cartilage was dissected from the
metacarpo-phalangeal joints of trotters from bacon pigs, rinsed
twice in calcium/magnesium-free Dulbecco's phosphate-buffered
saline (PBS) containing 400 units ml" 1 penicillin/streptomycin
and 200 units ml" 1 nystatin and stored at 4°C overnight in PBS
containing
200 units ml" 1
penicillin/streptomycin
and
67 units ml" 1 nystatin. The cartilage was finely chopped and the
chondrocytes were released from their extracellular matrix by
sequential digestion at 37 °C with 0.05% hyaluronidase (Boehringer-Mannheim) in PBS for 20min, 0.1% Pronase (Sigma) in
PBS for l h and 0.25% collagenase (Boehringer-Mannheim) in
Dulbecco's modified Eagle's medium (DMEM) containing 1 %
foetal calf serum (FCS) for 1-3 h. If the cartilage was not
completely digested by this stage, a further incubation with
collagenase was carried out. The small numbers of cells released
by hyaluronidase and Pronase were discarded. Cells obtained
from the collagenase digests were pooled and passed through a
sterile 60 fan aperture nylon screen (Nitex) to remove any
undigested cartilage fragments.
2xl0 5 unitsl 1 penicillin/streptomycin and 10% FCS (Newman
and Watt, 1988; Watt and Dudhia, 1988). The same batch of FCS
was used for any one experiment.
Cells were seeded on tissue culture plastic at high (1.25 xlO 5
cells per cm2) or low (1.25x 104 cells per cm2) density, as described
previously (Watt, 1988; Watt and Dudhia, 1988). For some
experiments cells were suspended in medium made viscous by
addition of methylcellulose (Dow Chemical Corp.) to a final
concentration of 1.2% (Stoker, 1968). Cells were suspended in
methylcellulose at a concentration of 5xlO 5 or 5xl0 4 ml" 1 ,
corresponding to the same cell/medium ratio as in high- and lowdensity cultures on tissue culture plastic.
Determination of cell size
Cells were fixed in 3.7% formaldehyde for 8min at room
temperature and resuspended in PBS. The cells were photographed under bright field using a Zeiss Photomicroscope III and
cell diameter determined from the photographs. More than 800
cells were scored for each determination.
Proteoglycan synthesis
Cultures were incubated with S^Ciml" 1 [35SJsulphate (Amersham, carrier-free, specific activity 25-40 Cimg" 1 ) for 24 h. The
medium was recovered and mixed with an equal volume of 8 M
guanidinium chloride, 0.1M sodium acetate, pH6.8, containing
the protease inhibitors 10 mM phenylmethylsulphonyl fluoride,
10 mM EDTA, 10 mM benzamidine hydrochloride and 50 HM
ra-caproic acid (Oegema et al. 1975; Pearson and Mason, 1977). The
cell layer was extracted for fluorimetric determination of DNA
content (see below) and then mixed with an equal volume of the
8 M guanidinium chloride extraction buffer, for proteoglycan
analysis.
After incubation for 48 h at 4°C with rotation, [35S]sulphate
incorporation into proteoglycans in the medium or cell extracts
was determined by precipitation on Whatman 3 MM filter paper
with 1 % cetyl pyridinium chloride, as described previously
(Wasteson et al. 1973; Newman and Watt, 1988), and counted in a
scintillation counter. All samples were analysed in duplicate and
the radioactivity was corrected to allow for decay.
Protein synthesis
Cultures were labelled with lO^Ciml" 1 [35S]methionine (Amersham, specific activity llSeCimmol" 1 ) for 24h. Cells and
medium were extracted as described above for proteoglycan
analysis and precipitated, in the presence of excess BSA (bovine
serum albumin; as carrier), onto Whatman GF/C glass microfibre
filters with 10% trichloroacetic acid, TCA (4°C). Filters were
washed twice with 10% TCA, then with absolute ethanol, airdried, and counted in a scintillation counter. Each sample was
assayed in duplicate and the radioactivity extrapolated back to
the activity date.
Chondrocytes were isolated from different depths of cartilage
by sequentially dissecting away the upper, middle and lower third
of the tissue. Cells were then isolated separately from the upper
and lower thirds and the middle third was discarded.
For every chondrocyte isolation, cartilage from several trotters
(6-24) was pooled. The yield of cells from unfractionated cartilage
was typically 107 per gram wet weight.
Fluorimetric determination of DNA content
Cell number was measured by fluorimetric determination of DNA
content (Karsten and Wollenberger, 1972; Newman and Watt,
1988). The cell layer was rinsed twice in PBS; incubated for
20 min at 37 °C in 20 fig ml" 1 Pronase (DNase-free, Sigma) in PBS;
sonicated for 20 s on ice; and stored at -20°C until assayed. For
the assay, samples or diluted DNA standards were mixed with
0.08 ml Pronase, 0.08 ml RNase (Sigma, DNase-free, 125 /(gml"1),
and PBS to give a volume of 0.8 ml. After 1 h incubation at 37 °C,
0.2 ml ethidium bromide (25^gml" 1 in PBS) was added. Fluorescence was determined in a fluorimeter with an excitation
wavelength of 360 nm and an emission wavelength of 580 nm,
while stirring the sample. All samples were read within 5-60 min
of adding the ethidium bromide. Standard curves of fluorescence
versus known amounts of DNA or numbers of cells were
constructed.
Culture conditions
The culture medium consisted of 3 parts of DMEM to 1 part of
Ham's F-12 medium, supplemented with 24.3 mg I" 1 adenine,
Immunofluorescence
Cell suspensions were fixed in 3.7 % formaldehyde in PBS at room
temperature for 8-10 min, then stained with MZ15, a mouse
350
M. Siczkowski and F. M. Watt
monoclonal antibody recognising keratan sulphate (Zanetti et al.
1985; Mehmet et al. 1986). The second antibody, fluoresceinated
rabbit anti-mouse immunoglobulin, was purchased from Miles
Scientific. Antibody incubations were for l h at room temperature. Controls were carried out in which second antibody alone
was used. Preparations were mounted in Gelvatol and examined
with a Zeiss Photomicroscope III.
Determination of [35S]sulphate incorporation into
keratan sulphate
Extracts of proteoglycan in guanidinium hydrochloride were
dialysed against 40mM Tris-HCl, pH7.3, for 48h with four
changes of dialysis buffer. The total amount of incorporated
[35S]sulphate was determined by cetyl pyridinium chloride
precipitation, as described above. The amount of [35S]sulphate
present in keratan sulphate side-chains was determined by using
chondroitinase ABC to remove the chondroitin sulphate sidechains, leaving chrondroitin sulphate stubs in addition to intact
keratan sulphate side-chains. The validity of this method was
confirmed by a second method in which chondroitinase ABC
treatment was followed by papain digestion to release individual
keratan sulphate chains with a small core protein peptide
attached. The amount of [35S]sulphate in the keratan sulphate
chains was determined by applying a 0.5 ml sample to a 5 ml
Sephadex G50 column (Pharmacia), collecting 10x0.5 ml fractions and counting radioactivity in the void volume peak.
Samples were digested with chondroitinase ABC for 5 h at 37 °C
with 0.05 unit enzyme (Miles Scientific) per 0.5 ml dialysed
medium in 40 mM Tris-HCl buffer, pH 7.3. In the second method,
the sample was then digested for 20 h at 60 "C with 2mgml~
papain (Sigma, twice recrystallised) in 0.1M sodium acetate,
0.02 M EDTA, 0.2 M cysteine hydrochloride, 4 M sodium chloride,
pH5.5(Oldberge*aZ. 1977).
Sepharose CL2B gel filtration chromatography
The method was based on that described by Mitchell and
Hardingham (1981). Two gel nitration columns of approximate
size 100 cm length by 0.6 cm diameter were packed with
Sepharose CL2B (Pharmacia). One column, to be run under
associative conditions, was equilibrated with 0.5 M sodium
acetate, pH6.8. This column was used to determine what
proportion of proteoglycans was capable of forming aggregates
with hyaluronic acid. The second column was set up under
dissociative conditions in 0.5 M sodium acetate, pH6.8, containing
2 M guanidinium hydrochloride (Fluka), to determine proteoglycan monomer size.
Sepharose CL2B has a fractionation range of 7xlO 4 to
4xlO 7 Mr. Molecules larger than 4xlO 7 Mr are excluded from the
Sepharose beads and elute in the void volume (Vo), while
molecules smaller than 7xlO4Afr enter the column matrix and
elute at the total volume (Vt) of the column. The Vo and Vt of each
column were determined using high molecular weight Dextran
Blue (Pharmacia) and Phenol Red solutions, respectively.
Proteoglycan extracts in guanidinium chloride were dialysed
against 0.5 M sodium acetate, pH 6.8, for 48 h with four changes of
dialysis buffer. The sample was then divided into two. One half
was mixed with an equal volume of 0.5 M sodium acetate, pH 6.8,
containing 8 M guanidinium chloride and rotated overnight at
4°C before application to the dissociative column. A 10 fig portion
of hyaluronic acid was added to the second aliquot and the sample
left overnight at 4°C to aggregate before application to the
associative column. Samples were applied to each column in a
total volume of no greater than 0.7 ml; buffer was pumped out of
the bottom of each column at a speed of 3-4 ml per h and fractions
of 0.7 ml were collected. The distribution coefficient (ifav) and
relative molecular mass (Mr) of each species resolved on the
column were determined as described by Ohno et al. (1986).
Agarose-polyacrylamide composite gel electrophoresis
The method is based on that of McDevitt and Muir (1971), as
modified by Carney et al. (1986), with the further modification
that a sheet of Gelbond (for agarose gels; Pharmacia) was added to
the glass plate to facilitate handling of the gel. The Gelbond was
omitted when gels were to be immunoblotted.
The gels were 2 mm thick and usually consisted of 1.2%
polyacrylamide and 0.6 % agarose, although for some experiments
the acrylamide concentration was increased to 2.4 %. The running
buffer was 4 M urea, 10 mM Tris-HCl, 0.25 mM sodium sulphate,
lmM EDTA, pH6.8. Gels were run at 60 V for 5min, followed by
120 V for 1 h, or until the dye front had migrated 3 cm. Samples
were usually dialysed extensively against distilled water, freezedried, taken up in sample buffer of 8 M urea, 10 % sucrose, 10 mM
Tris-HCl, 0.25mM sodium sulphate, lmM EDTA, 0.001%
Bromophenol Blue, pH6.8, and left at 4°C overnight. In some
experiments the cell layer was extracted directly in sample buffer.
Gels were stained with 0.2 % Toluidine Blue in 0.1 M acetic acid
for 20min; destained in 3% acetic acid for 90min and left in
distilled water until clear. For fluorography, gels were impregnated with 0.4 % 2,5-diphenyloxazole in absolute ethanol for 16 h,
washed in distilled water for 1 h, then dried down under vacuum
and exposed to preflashed Fuji X-ray film at —70°C.
In some experiments composite gels were lightly stained with
Toluidine Blue and destained to visualise the proteoglycan bands.
Individual bands were cut out of the gel with a scalpel and
electroeluted using an ISCO electroelution apparatus with
composite gel running buffer at 3 W overnight. The electroeluted
bands were rerun on composite gels to confirm the presence of
single bands.
Polyacrylamide gel electrophoresis
Linear gradient (4 % to 20 %) polyacrylamide gels were prepared
using the buffer system of Laemmli (1970). Electrophoresis was at
a constant voltage of 60 V while the sample was in the stacking
gel and at 120 V thereafter. Gels were stained with Coomassie
Blue or prepared for fluorography, as described by Bonner and
Laskey (1974).
Immunoblotting
Blotting of composite gels or polyacrylamide gels was carried out
essentially as described by Towbin et al. (1979), except that the
transfer buffer contained 0.01 % SDS, to aid transfer of large
molecules to nitrocellulose (Erickson et al. 1982). Transfer was
carried out at 50 V, 4°C, overnight. The transfer buffer was then
replaced with the same buffer without SDS and transfer
continued for a further 2 h.
Nitrocellulose was stained for protein using AuroDye Forte
(Janssen), according to the manufacturer's instructions. Antigens
were visualised using AuroProbe BL plus second antibody and
silver enhancer IntenSE II (Janssen), also according to the
manufacturer's instructions (Brada and Roth, 1984; Moeremans
et al. 1984). When the first antibody recognised chondroitin
sulphate stubs, blots were incubated with 0.025 unit ml" 1
chondroitinase ABC (Miles Scientific) in blocking buffer for
60min at 37 °C, with shaking after the blocking step.
The following mouse monoclonal antibodies were used: l-B-5
(recognising nonsulphated chondroitin sulphate), 2-B-6 (chondroitin 4-sulphate), 3-B-3 (chondroitin 6-sulphate), 5-D-4 (keratan sulphate), all generous gifts from Bruce Caterson (Caterson et
al. 1983; Couchman et al. 1984; Caterson et al. 1985; Caterson et
al. 1986). In addition, MZ15 (keratan sulphate), another mouse
monoclonal, was used (Zanetti et al. 1985; Mehmet et al. 1986).
Rabbit antiserum to the proteoglycan hyaluronic acid binding
region was a gift from T. Hardingham (Ratcliffe and Hardingham,
1983).
Histochemical staining for alkaline phosphatase
Cultures were fixed in 70 % ethanol at 4°C for 5 min, rinsed in
50 mM Tris-base, pHIO, and stained for 20 min in 50 mM
Tris-base, pHIO, containing 0.1% a-naphthyl acid phosphate
(sodium salt), 0.1% magnesium chloride and 0.1% fast blue RR
salt, which had been mixed and filtered immediately before use
(Kiernan, 1981). After staining, the samples were rinsed in
distilled water and mounted in glycerol. Controls consisted of
adding the alkaline phosphatase inhibitor levamisole at 4 mM to
the incubation mixture (Reynolds and Dew, 1977) or omitting the
substrate, tv-naphthyl acid phosphate.
Chondrocytes from different cartilage zones
351
Results
Cell isolation
Chondrocytes were isolated from two regions of cartilage,
designated the upper and lower zones. The upper zone
(100-200 ;Um thick) corresponded to the histologically
recognised superficial zone plus some of the transitional
zone; the lower zone (200-350/tm thick) corresponded to
the deep zone and occasionally a small amount of the
hypertrophic zone. The middle zone that was discarded
contained some transitional and, mainly, deep zone
cartilage (Palfrey and Davies, 1966). In order to obtain a
uniform sample of cartilage from each zone only cartilage
from the metacarpus of the metacarpo-phalangeal joint
was used, excluding the ridges. The cell yield (per gram
wet weight) from upper zone cartilage was higher than
from lower zone, as would be expected since there are more
cells per unit volume in the upper zone.
The diameters of chondrocytes from each zone were
measured on isolation and after growth in culture. Freshly
isolated upper zone cells had a modal diameter of 10 [im,
range 8-12 /an. Lower zone cells had the same modal
diameter on isolation, but there appeared to be an
additional size class with a modal diameter of about 15 /im;
there were also a few large cells of 17-20 ,um. After 32 days
in culture at high or low density the modal diameter of
both populations had increased to 17 //m and the range in
sizes had increased to 14-22 ^an; thus the two subpopulations could no longer be distinguished on the basis of cell
size distribution.
Cell morphology in culture
As previously reported (Zanetti et al. 1985), upper zone
cells adhered and spread more slowly in culture than lower
zone cells. After 5 days upper zone cells that had attached
at high or low seeding density were still rounded (Fig. 1 A).
In contrast, cells from lower zone cartilage attached and
spread within 5 days (Fig. IB). Less than 5 % of cells in
each population failed to attach. The initial failure of
upper zone cells to spread in culture was not due to cell
damage during isolation, because reducing the collagen-
Fig. 1. Morphology of cells grown in culture at high density for different lengths of time. (A,B) 5 days; (C,D) 14 days; (E,F) 32
days. (A,C,E) Upper zone; (B,D,F) lower zone. Bar: A-D, 200 ^m; E,F, 114 fim.
352
M. Siczkowski and F. M. Watt
E
U
104
12000-
8000-
D
10000-
c
o
«
60008000-
1
8
C
a
6000-
j
4000-
40002000-
1
r*
2000-
020
10
15
Days
20
25
30
30
Fig. 2. Replicate wells of 24-well (2.4 cm2) plates were seeded at high (A,C,E,G) or low (B,D,F,H) density and assayed for: A,B, cell
number per well; points are means of duplicate determinations from 2 wells, which typically varied by less than 5 % in A and less
than 8% in B. C-F, [35S]sulphate incorporation per 104 cells; C,D, total counts incorporated into medium and cell layer; E,F, ratio
of counts incorporated into medium versus cell layer; points are means of counts from 2 wells. G,H, [36S]methionine incorporation
(medium plus cell layer) per 104 cells; one well assayed per point. (•) Upper zone; ( • ) lower zone cells.
ase digestion time to 1.5 h and including 15 % FCS in the
collagenase solution had no effect on spreading.
Cells were grown at high or low density. When cultured
at high density chondrocytes continue to express type II
collagen and large aggregating proteoglycans for several
weeks, whereas at low density the differentiated phenoChondrocytes from different cartilage zones
353
type of the cells is less stable (Watt, 1988; Watt and
Dudhia, 1988). After 14 days at high density, differences in
cell morphology were still apparent. Upper zone cultures
consisted of small clusters of rounded cells separated by
flattened cells (Fig. 1C), while lower zone cells were more
uniformly spread and polygonal (Fig. ID). By 32 days at
high density, upper zone cultures no longer contained
clusters of rounded cells, but were spread, mainly as a
monolayer, with only a few rounded cells (Fig. IE). In
lower zone cultures most of the cells were rounded and the
cultures appeared to be multilayered with large amounts
of refractile matrix between adjacent cells (Fig. IF).
At low density, differences between the two subpopulations were less pronounced, both types of culture
consisting mainly of flattened cells at 14 days. At 32 days
the cells were still mainly flattened, but nodules of
rounded cells were evident. Nodules covered a greater
proportion of the lower zone than the upper zone cultures
(data not shown).
Growth of chondrocytes from upper and lower zone
cartilage
Chondrocytes isolated from upper and lower zone cartilage
were plated at high or low density in 24-well multiwell
dishes, and every 2-3 days for 28 days cell number per
well was determined by DNA fluorimetry. At high density
the average increase in cell number was less than 2-fold
(Fig. 2A), while at low density cell number increased by
more than 10-fold (Fig. 2B). At high density, cell number
per well in upper zone cultures was consistently lower
than in lower zone cultures. At low density, the growth
curves for the two subpopulations were initially the same,
but after day 6 the number of cells per dish in the upper
zone cultures was consistently less than in lower zone
cultures.
Proteoglycan synthesis
Proteoglycan synthesis was measured by incorporation of
[35S]sulphate into glycosaminoglycans. Upper and lower
zone cells were grown at high or low density and
incorporation of label into the medium and cell layer was
measured at intervals for 28 days (Fig. 2C,D). At high
density, incorporation was similar in upper and lower zone
cultures for the first 12 days, but thereafter incorporation
was greater in lower zone cultures. At low density,
incorporation was initially greater in upper zone cultures,
but from day 15 onwards incorporation was similar in both
zones. The level of [35S]sulphate incorporation was
initially greater in high density than low density cultures,
as observed previously (Watt, 1988).
More pronounced differences between upper and lower
zone cells were observed when the ratio of [35S]sulphate
counts incorporated into the medium versus the cell layer
was plotted (Fig. 2E,F). The ratio was consistently higher
in upper than in lower zone cultures, a difference that was
more pronounced at high (Fig. 2E) than at low (Fig. 2F)
density.
In all of these experiments cells from full-depth
cartilage were included for comparison and consistently
gave levels of incorporation intermediate between upper
and lower zone cells.
Protein synthesis
The differences in proteoglycan synthesis did not reflect
differences in total protein synthesis. There were no major
differences in total [35S]methionine incorporation between
70
Fig. 3. Percentage of cells stained with MZ15 after seeding at: (A,C) high density; (B,D) low density. (A,B) Cells grown on tissue
culture plastic; (C,D) cells grown in suspension in methylcellulose. In A,B at least 250 cells were counted per time-point and at
least 400 in C,D. Results in A,B are means of duplicate dishes. (•) Upper zone; (•) lower zone cells.
354
M. Siczkowski and F. M. Watt
upper and lower zone cultures maintained in suspension
in methylcellulose for up to 7 days. In each case the
proportion of positively stained upper zone cells rapidly
increased to the same level as in lower zone suspensions,
subsequently remaining constant at high density, but
falling at low density.
The keratan sulphate content of proteoglycan synthesised by upper and lower zone cells in culture is shown in
Fig. 4. [35S]sulphate-labelled proteoglycan extracted from
the culture medium was digested with chondroitinase
ABC and the amount of incorporated sulphate was
determined by cetyl pyridinium chloride precipitation.
The majority of incorporated [35S]sulphate remaining
after digestion would be in keratan sulphate, although a
small proportion would be in the chondroitin sulphate
stubs. In high-density cultures the keratan sulphate
content of the proteoglycan was consistently greater in
lower than upper zone cultures (Fig. 4A); this could reflect
an increase in the number of keratan sulphate chains, in
chain length or in sulphation. In low-density cultures,
however, the keratan sulphate content of proteoglycan
synthesised by both zones decreased with time and there
was no significant difference between the two cell
populations (Fig. 4B).
Fig. 4. [35S]sulphate incorporation into keratan sulphate,
expressed as a percentage of total [35S]sulphate incorporation.
Cells were grown at: (A) high density; (B) low density.
(•) Upper zone cells; ( • ) lower zone cells.
upper, lower and full-depth cells, at high or low density
(Fig. 2G,H and data not shown). In all situations there was
a decrease in incorporation during the first two weeks,
followed by a transient increase during the third week.
Keratan sulphate expression
The keratan sulphate content of cartilage proteoglycan
increases with depth (Stockwell and Scott, 1967; Kempson
et al. 1973; Venn, 1978). Zanetti et al. (1985) reported that
the proportion of cells from upper zone cartilage that
stained positively with MZ15, a monoclonal antibody to
keratan sulphate, was less than the proportion of cells
from deep cartilage when freshly isolated or incubated
overnight in suspension, but that with prolonged cultivation at high density the difference disappeared. In order
to extend these observations, we measured both the
percentage of MZ15-positive cells and the keratan sulphate content of proteoglycan synthesised at different
times in culture (Figs 3, 4).
In high-density lower zone cultures the percentage of
MZ15-positive cells fell from approximately 55 % to 42 %
within 2 days and declined only slightly thereafter. In
high-density upper zone cultures the number of keratan
sulphate-positive cells rose from 16% to 30% within 2
days and remained between 30 and 38% thereafter
(Fig. 3A). In low density cultures the number of keratan
sulphate-positive cells initially rose in the upper zone
cultures and fell in the lower zone cultures to reach about
33 % by day 7. In upper zone cultures this fell to 23 % by 28
days, whereas in lower zone cultures, the proportion of
positively stained cells remained approximately constant
at 35% (Fig. 3B).
Fig. 3C,D shows the percentage of MZ15-positive cells in
Sepharose CL2B profiles of'3S'S-labelled proteoglycans
extracted from different cartilage zones and from
cultures of upper and lower zone chondrocytes
Cartilage slices from different depths of cartilage were
labelled for 24 h with lO^Ciml" 1 [35S]sulphate in culture
medium containing 10 % FCS. The medium was removed,
the cartilage slices rinsed twice in PBS and extracted for
48 h in 4 M guanidinium hydrochloride buffer. After
extensive dialysis the samples were run on Sepharose
CL2B columns under associative and dissociative conditions.
Under associative conditions most of the proteoglycan
from both zones formed aggregates with hyaluronic acid,
but the proportion of nonaggregating proteoglycan was
greater in upper zone than lower zone cartilage (30% of
incorporated label compared with 14%) (Fig. 5A,B). Under
dissociative conditions the proteoglycan that failed to
aggregate under associative conditions appeared as a
shoulder to the main monomer peak and was more
pronounced in upper than lower zone cartilage (data not
shown). From the elution positions of the main proteoglycan species and the less-abundant species, molecular
weights
of
1.68±0.26xl0 6 M r
(# av =0.22)
and
5
4.4±O.9xlO Mr (.Kav=0.57), respectively, were calculated
(Mr values are means of 5 determinations ± standard
deviation).
[35S]sulphate-labelled proteoglycans secreted into the
medium by cultures of upper and lower zone chondrocytes
were also analysed by Sepharose CL2B chromatography.
After 4 days at low (Fig. 5C,D) or high density (data not
shown) there was a single aggregating species with a Kav
of approximately 0.21 (1.71xlO 6 M r ). Thus the low Mr
proteoglycan that was a feature of upper zone cartilage
profiles (Fig. 5A) was not observed after 4 days in culture.
With time in low density culture, changes in the
proteoglycans synthesised were observed. At day 12 and
for the remainder of the culture period a small nonaggregating proteoglycan was synthesised by both types of
culture (Fig. 5E,F, and results not shown). This eluted
from Sepharose CL2B columns with a Kav of 0.56,
corresponding to a Mr of about 4.5 xlO 5 . At 42 days the
majority of the proteoglycan still aggregated with hyaluChondrocytes from different cartilage zones
355
20000
10
20
30
Elution volume (ml)
10
20
30
Elution volume (ml)
40
40
25000
25000
2000015000100005000-
10
20
30
10
20
30
Elution volume (ml)
40
Elution volume (ml)
40
20000
F
2000015000-
il
1000050001
010
20
30
Elution volume (ml)
40
10
20
30
Elution volume (ml)
40
Fig. 5. Sepharose CL2B chromatography under associative conditions. [35S]sulphate-labelled proteoglycans were extracted from:
(A) upper zone (B) lower zone cartilage or from the medium of upper zone (C,E) or lower zone (D,F) chondrocytes, that had been
cultured for 4 days (C,D) or 42 days (E,F) at low density. Vo and Vt are indicated by arrows (left and right, respectively).
ronic acid, but upper zone cultures contained a higher
proportion of the non-aggregating proteoglycan than
lower zone cultures (27 % of total incorporated [35S]sulphate versus 16%). Thus with time in culture, the column
profiles came to resemble those of proteoglycans synthesised in cartilage slices (Fig. 5A,B,E,F).
Composite gel electrophoresis of proteoglycans from
different cartilage zones
Proteoglycan extracted from the different cartilage zones
was analysed by composite gel electrophoresis. Toluidine
Blue staining snowed that there was a major band of high
molecular weight in both samples and an additional minor
species of slightly faster mobility in extracts of upper, but
not lower, zone cartilage. A third, faint, band, migrating
close to the dye front, was present in both cartilage
extracts, but was more abundant in upper zone extracts
(Fig. 6A).
356
M. Siczkowski and F. M. Watt
The relative levels of the two high molecular weight
proteoglycans varied between preparations, but the faster
mobility band was always less abundant. Both bands were
present in extracts of freshly isolated upper zone cartilage,
in cartilage slices cultured for 24 h and in medium in
which the slices had been incubated (Fig. 6B).
Mobility on composite gels is a function not only of size,
but of glycosaminoglycan composition (McDevitt and
Muir, 1971; Carney et al. 1986) and so molecular weights
cannot be assigned solely on the basis of mobility.
However, if the two high molecular weight bands are both
labelled with [35S]sulphate, it is likely that they elute in
the major peak on dissociative Sepharose CL2B columns
CK"av=0.22), whereas the low molecular weight band would
correspond to the material that elutes with a Kav of 0.57.
Immunoblotting of cartilage proteoglycans
Immunoblotting was used to characterise the two high
1
B
2
3
2
3
c l
i-
M
5
6
*•
7
8
Fig. 6. Composite gel electrophoresis of cartilage
proteoglycans. (A) Toluidine Blue staining; proteoglycan was
extracted from: 1, upper zone; 2, lower zone; 3, full-depth
cartilage. Arrows mark the bands discussed in the text.
(B) Toluidine Blue-stained gel of upper zone proteoglycans
extracted from: 1, cartilage maintained in culture for 24 h;
2, freshly isolated cartilage; 3, culture medium in which
cartilage had been incubated for 24 h. (C) Immunoblotting of
proteoglycan from full-depth cartilage. The two high molecular
weight bands (see A,B) were dissected from the gel,
electroeluted and then re-run in separate tracks of a second
gel. In each case the slower migrating band is on the left-hand
side of each pair of tracks: 1, stained with AuroDye; 2, MZ15;
3, 5-D-4; 4, 3-B-3 after chondroitinase ABC predigestion;
5, 3-B-3 without chondroitinase digestion; 6, 2-B-6 with
chondroitinase ABC predigestion; 7, l-B-5 with chondroitinase
ABC predigestion; 8, rabbit antiserum to the hyaluronic acidbinding region.
molecular weight proteoglycans resolved by composite gel
electrophoresis. Since the two bands migrated very close
together they were cut out of the gels, electroeluted and rerun prior to transfer to nitrocellulose. Staining the
nitrocellulose for protein showed that the individual high
molecular weight proteoglycans migrated to the same
relative positions as in the unfractionated sample. The
lower band was slightly contaminated with the upper
band, but could be clearly distinguished by its position
relative to the upper band on the same gel (Fig. 6C, 1).
Antibodies MZ15 and 5-D-4, which recognise keratan
sulphate, reacted with both bands, although the upper
band appeared to stain more intensely (Fig. 6C, 2,3). When
the proteoglycans on nitrocellulose were digested with
chondroitinase ABC, 3-B-3 (recognising chondroitin 6sulphate) reacted strongly with both bands (Fig. 6C, 4);
however, if the digestion step was omitted no reactivity
was observed (Fig. 6C, 5). After chondroitinase digestion,
2-B-6 (recognising chondroitin 4-sulphate) stained the
upper band more strongly than the lower band (Fig. 6C, 6)
while l-B-5 (recognising non-sulphated chondroitin sulphate) reacted very weakly with both bands (Fig. 6C, 7).
Rabbit antiserum against the hyaluronic acid-binding
region reacted more strongly with the upper band than the
lower band (Fig. 6C, 8). The results show that both
proteoglycans contained keratan sulphate and chondroitin
sulphate side-chains, but that the lower band contained
relatively less keratan sulphate, chondroitin 4-sulphate
and hyaluronic acid-binding region than the upper band.
Electrophoretic analysis of proteoglycans synthesised in
culture
Medium from [35S]sulphate-labelled upper and lower zone
cultures was analysed by composite gel electrophoresis at
intervals of from 2 to 27 days (Fig. 7A and results not
shown). No significant differences between upper and
lower zone cultures were observed. In high-density
cultures some of the samples resolved into two bands that
migrated close together (Fig. 7A), while at low density
only one band could be detected (results not shown).
In order to compare the small non-aggregating proteoglycans (Kav=0.56) detected after prolonged culture at low
density with the smallest proteoglycan synthesised in
organ culture of cartilage slices, [35S]sulphate labelled cell
and cartilage extracts were subjected to electrophoresis on
gradient gels of 4% to 20% acrylamide (Fig. 7B). The
major high molecular weight proteoglycans CKav=0.22) do
not enter this type of gel. As shown in Fig. 7B, extracts of
upper and lower zone cartilage contained only a single
sulphated species that had an Mr of approximately
300 xlO 3 . Cultured
chondrocytes synthesised
a
300xl0 3 M r band when grown at high density (Fig. 7B).
However, in low-density cultures, an additional band of
125xlO3Afr was observed (Fig. 7B,C).
Alkaline phosphatase
Alkaline phosphatase is a marker of calcifying cartilage.
32-day- and 42-day-old high- and low-density upper and
lower zone cultures were stained for alkaline phosphatase.
Upper zone cultures at both densities were virtually
unstained, while lower zone cultures at both densities
stained intensely (Fig. 8 and results not shown).
Discussion
In this report we have compared the behaviour of upper
and lower zone chondrocytes in culture at high and low
density. On isolation upper zone cells were smaller than
lower zone cells, as reported previously (Zanetti et al.
1985), but with time in culture this difference disappeared.
Again, as reported previously, upper zone cells initially
showed less attachment and spreading (Zanetti et al.
1985). Morphological differences between the two cell
populations persisted for several weeks and were more
pronounced in cultures seeded at high than those seeded at
low density: the lower zone cells deposited a more
extensive extracellular matrix than the upper zone cells.
The growth rates for both populations were similar, but
upper zone cultures reached a lower final density.
[ S]methionine incorporation into total protein was
Chondrocytes from different cartilage zones
357
B
2
3
4
5
1 23 4 5 6 7 8
LUJ
Fig. 7. Electrophoresis of [35S]sulphate-labelled proteoglycans. (A) Composite gels of proteoglycans secreted into the medium by
high-density cultures of (top row) upper or (bottom row) lower zone chondrocytes. Cells were in culture for: 1, 9 days; 2, 16 days;
3, 25 days. (B,C) [35S]sulphate-labelled proteoglycans resolved on 4 % to 20 % polyacrylamide gels. Equal numbers of counts were
loaded per track. Arrowheads indicate Mr (xlO~3) markers: 214, 111, 68, 45, 24 and 18. (B) 1,2. Extracts of chondrocytes grown at
low density for 70 days: 1, upper zone; 2, lower zone; 3-5. Cartilage extracts: 3, upper zone; 4, lower zone; 5, full depth. (C) Medium
from cultures of full-depth chondrocytes grown at low density (1-4) or high density (5-8) for 7 days (1,5), 20 days (2,6), 28 days
(3,7), 37 days (4,8).
similar for both populations, but upper zone cells deposited
less proteoglycan in the cell layer. The latter observation
fits well with the in vivo data showing that superficial
cartilage contains a lower concentration of proteoglycan
than deep cartilage (Maroudas et al. 1969; Franzen et al.
1981; Bayliss et al. 1983; Ratcliffe et al. 1984). Our results
are consistent with the observations of Aydelotte and coworkers (Aydelotte and Kuettner, 1988; Aydelotte et al.
1988) on chondrocytes from different zones of bovine
articular cartilage: they found that superficial zone cells
were smaller on isolation than deep zone cells and that
deep zone cells produced more proteoglycan and retained a
greater proportion in the extracellular matrix.
The keratan sulphate content of proteoglycan synthesised by lower zone chondrocytes in high density culture
was greater than that of upper zone cells, consistent with
the observation that the keratan sulphate content of
cartilage increases with depth (Stockwell and Scott, 1967;
Kempson et al. 1973; Venn, 1978). At low density the
keratan sulphate content of proteoglycan synthesised by
both subpopulations was not significantly different and
declined with time. It is interesting that the two
populations could be distinguished in this way, when they
could not be distinguished by MZ15 staining (Zanetti et al.
1985; and Fig. 3); this is presumably because cell surface
staining with the antibody can give an indication of the
presence or absence of keratan sulphate, but not of relative
abundance.
358
M. Siczkowski and F. M. Watt
The proteoglycans of upper and lower zone cartilage and
of chondrocyte cultures were compared in more detail. The
proportion of proteoglycan that did not aggregate with
hyaluronic acid was greater in upper than lower zone
cartilage; this difference was lost in early cultures, but
reappeared in long-term cultures. Aydelotte et al. (1988)
have also reported that superficial bovine articular
chondrocytes in culture synthesise less aggregating
proteoglycan than deep cells.
Composite gel electrophoresis resolved two high molecular weight proteoglycans in cartilage, the less abundant of
which was unique to the upper zone. However, in culture
two bands, when resolved, were present in both upper and
lower zone proteoglycans. The cartilage bands were
further studied by immunoblotting: we established that
both bands contained chondroitin sulphate and keratan
sulphate side-chains. The less-abundant band stained
more weakly for keratan sulphate, chondroitin 4-sulphate
and hyaluronic acid-binding region. One explanation for
these results is that the lower band is generated from the
upper band by limited proteolysis (Hascall et al. 1983;
Campbell et al. 1984), but further experiments are
required to test this possibility. A number of other studies
have revealed proteoglycan heterogeneity, by using
composite gel electrophoresis. Changes in gel band
patterns as a function of cartilage depth (Franzen et al.
1981; Bayliss et al. 1983), anatomical site (Stanescu et al.
1980) and age (Bayliss and Ali, 1978) have all been
differences between the cells of different zones, whereas
some probably reflects the response of chondrocytes to
local differences in their environment.
This project was initiated at the Kennedy Institute of
Rheumatology and supported by a grant to Fiona Watt from the
Arthritis and Rheumatism Council. We are grateful to Paul
Newman, Tim Hardingham and Mike Bayliss for advice and to
Lewis Wolpert for providing laboratory space for Martin Siczkowski. We thank Wendy Senior for expert typing of the
manuscript.
8A
References
AYDELOTTE, M. B., GEEENHILL, R. R. AND KUETTNER, K. E. (1988).
Differences between sub-populations of cultured bovine articular
chondrocytes. II. Proteoglycan metabolism. Conn. Tiss. Res. 18,
223-234.
AYDELOTTE, M. B. AND KUETTNER, K. E. (1988). Differences between subpopulations of cultured bovine articular chondrocytes. I. Morphology
and cartilage matrix production. Conn. Tiss. Res. 18, 205-222.
BAYLISS, M. T. AND ALI, S. Y. (1978). Age-related changes in the
composition and structure of human articular-cartilage proteoglycan.
Biochem. J. 176, 683-693.
BAYLISS, M. T., VENN, M., MAROUDAS, A. AND ALI, S. Y. (1983).
Structure of proteoglycans from different layers of human articular
cartilage. Biochem. J. 209, 387-400.
BONNER, W. M. AND LASKEY, R. A. (1974). A film detection method for
tritium-labelled proteins and nucleic acids in polyacrylamide gels.
Eur. J. Biochem. 46, 83-88.
BRADA, D. AND ROTH, J. (1984). 'Golden Blot' - detection of polyclonal
and monoclonal antibodies bound to antigens on nitrocellulose by
Protein A-gold complexes. Analyt. Biochem. 142, 79-83.
Fig. 8. Cultures stained for alkaline phosphatase activity.
Cells were grown for 42 days at high density. (A) Upper zone;
(B) lower zone. Bar, 200 fim.
BRUCKNER, P., HORLER, I., MENDLER, M., HOUZE, Y., WINTERHALTER, K.
H., EICH-BENDER, S. G. AND SPYCHER, M. A. (1989). Induction and
prevention of chondrocyte hypertrophy in culture. J. Cell Biol. 109,
2537-2545.
CAMPBELL, M. A., HANDLEY, C. J., HASCALL, V. C, CAMPBELL, R. A. AND
reported and there is also variation between animal
species.
Small, non-aggregating proteoglycans were analysed by
gradient polyacrylamide gel electrophoresis. In cartilage
and in cultures seeded at high density, a single diffuse
band of 300 x 103 Mr was observed, but in low density
cultures an additional band of 125 x 103 MT was found. The
differentiated phenotype of chondrocytes is less stable in
low-density cultures (Watt and Dudhia, 1988; Watt, 1988)
and this may be the explanation for the appearance of the
125 x 103 Mr band. Further studies are necessary to
establish the composition of both the 300 x 103 Mr and the
125 xlO 3 M r proteoglycans.
Chondrocytes in culture can undergo hypertrophy and
calcification (see, for example, Tacchetti et al. 1987), a
process that can be regulated by growth factors or other
serum components (Bruckner et al. 1989; Kato and
Iwamoto, 1990). We examined expression of one marker of
calcification, alkaline phosphatase, in upper and lower
zone cultures and found that it was only expressed by
lower zone cells, in keeping with their relative proximity,
in vivo, to the underlying bone.
In conclusion, some of the differences between upper and
lower cartilage zones, such as cell size and percentage of
MZ15-positive cells, were lost in culture, whereas others,
such as the keratan sulphate content of proteoglycan,
proportion of proteoglycan deposited in the cell layer and
proportion of nonaggregating proteoglycan synthesised,
were retained. In addition, differences between upper and
lower zone cultures, which did not have a strict in vivo
correlate, including cell morphology and confluent density, were observed. We therefore conclude that some of the
heterogeneity of articular cartilage reflects intrinsic
LOWTHER, D. A. (1984). Turnover of proteoglycans in cultures of
bovine articular cartilage. Archs Biochem. Biophys. 234, 275—289.
CARNEY, S. L., BAYLISS, M. T., COLLIER, J. M. AND MUIR, H. (1986).
Electrophoresis of 35S-labelled proteoglycans on polyacrylamideagarose composite gels and their visualisation by fluorography.
Analyt. Biochem. 156, 38-44.
CATERSON, B., CALABRO, T., DONOHUE, P. J. AND JAHNKE, M. R. (1986).
Monoclonal antibodies against cartilage proteoglycan and link
protein. In Articular Cartilage Biochemistry (ed. Kuettner, K.), pp.
59-73. Raven Press, New York.
CATERSON, B., CHRISTNER, J. E. AND BAKER, J. R. (1983). Identification
of a monoclonal antibody that specifically recognizes corneal and
skeletal keratan sulphate. J. biol. Chem. 258, 8848-8854.
CATERSON, B., CHRISTNER, J. E., BAKER, J. R. AND COUCHMAN, J. R.
(1985). Production and characterization of monoclonal antibodies
directed against connective tissue proteoglycans. Fedn Proc. Fedn Am.
Socs exp. Biol. 44, 386-393.
COUCHMAN, J. R., CATERSON, B., CHRISTNER, J. E. AND BAKER, J. R.
(1984). Mapping by monoclonal antibody detection of
glycosaminoglycans in connective tissue. Nature 307, 650—652.
ERICKSON, P. F., MINIER, L. N. AND LASHER, R. S. (1982). Quantitative
electrophoretic transfer of polypeptides from SDS polyacrylamide gels
to nitrocellulose sheets: a method for their re-use in
immunoautoradiographic detection of antigens. J. Immun. Meth. 51,
241-249.
FORTUNA, R., ANDERSON, H. C, CARTY, R. P. AND SAJDERA, S. W. (1980).
Enzymatic characterization of the matrix vesicle alkaline phosphatase
isolated from bovine fetal epiphyseal cartilage. Calc. Tiss. Int. 30,
217-225.
FRANZ£N, A., INEROT, S., HEJDERUP, S-O. AND HEINEGARD, D. (1981).
Variations in the composition of bovine hip articular cartilage with
distance from the articular surface. Biochem. J. 195, 535-543.
GHADIALLY, F. N. (1978). Fine structure of joints. In The Joints and
Synovial Fluid, vol. 1 (ed. L. Sokoloff), pp. 105-176. Academic Press,
NY, London.
HASCALL, V. C , MORALES, T. I., HASCALL, G. K., HANDLEY, C. J. AND
MCQUILLAN, D. J. (1983). Biosynthesis and turnover of proteoglycans
in organ culture of bovine articular cartilage. J. Rheum. (Suppl.) 11,
45-52.
Hsu, H. H. T., MUNOZ, P. A., BARR, J., OPPLIGER, I., MORRIS, D. C,
VAANANEN, H. K., TARKENTON, N. AND ANDERSON, H. C. (1985).
Purification and partial characterization of alkaline phosphatase of
Chondrocytes from different cartilage zones
359
matrix vesicles from fetal bovine epiphyseal cartilage. J. biol. Chem.
260, 1826-1831.
KARSTEN, U. AND WOLLENBEROER, A. (1972). Determination of DNA and
RNA in homogenized cells and tissues by surface fluorometry. Analyt.
Biochem. 46, 135-148.
KATO, Y. AND IWAMOTO, M. (1990). Fibroblast growth factor is an
inhibitor of chondrocyte terminal differentiation. J. biol. Chem. 265,
5903-5909.
KEMPSON, G. E., Mum, H., POLLARD, C. AND TUKE, M. (1973). The
tensile properties of the cartilage of human femoral condyles related
to the content of collagen and glycosaminoglycans. Biochim. biophys.
Ada 297, 456-472.
KIELTY, C. M., KWAN, A. P. L., HOLMES, D. F., SCHOR, S. L. AND GRANT,
M. E. (1985). Type X collagen, a product of hypertrophic chondrocytes.
Biochem. J. 227, 545-554.
KIERNAN, J. A. (1981). Histological and Histochemical Methods: Theory
and Practice. Pergamon Press, Oxford.
KUIJER, R., VAN DE STADT, R. J., VAN KAMPEN, G. P. J., DE KONING, M.
H. M. T., VAN DE VOORDE-VISSERS, E. AND VAN DER KORST, J. K.
(1986). Heterogeneity of proteoglycans extracted before and after
collagenase treatment of human articular cartilage. Arthritis Rheum.
29, 1248-1255.
KWAN, A. P. L., FREEMONT, A. J. AND GRANT, M. E. (1986).
Immunoperoxidase localization of type X collagen in chick tibiae.
Biosci. Rep. 6, 155-162.
LAEMMLI, U. K. (1970). Cleavage of structural proteins during the
assembly of the head of bacteriophage T4. Nature 227, 680-685.
MANICOURT, D. H., PITA, J. C , MODEVITT, C. A. AND HOWELL, D. S.
(1988). Superficial and deeper layers of dog normal articular cartilage.
Role of hyaluronate and link protein in determining the
sedimentation coefficients distribution of the nondissociatively
extracted proteoglycans. J. biol. Chem. 263, 13121-13129.
MAROUDAS, A., MUIR, H. AND WINGHAM, J. (1969). The correlation of
fixed negative charge with glycosaminoglycan content of human
articular cartilage. Biochim. biophys. Ada 177, 492-500.
MCDEVITT, C. A. AND MUIR, H. (1971). Gel electrophoresis of
proteoglycans and glycosaminoglycans on large pore composite
polyacrylamide agarose gels. Analyt. Biochem. 44, 612-622.
MEHMET, H., SCUDDER, P., TANG, P. W., HOUNSELL, E. F., CATERSON, B.
AND FEIZI, T. (1986). The antigenic determinants recognized by three
monoclonal antibodies to keratan sulphate involve hepta or larger
oligosaccharides of the poly (N-acetyllactosamine) series. Bur. J.
Biochem. 157, 385-391.
MITCHELL, D. AND HARDINGHAM, T. E. (1981). The effects of
cycloheximide on the biosynthesis and secretion of proteoglycans by
chondrocytes in culture. Biochem. J. 196, 521-529.
MOEREMANS, M., DANEELS, G., VAN DUCK, A., LANGANGER, G. AND DE
MEY, J. (1984). Sensitive visualization of antigen-antibody reactions
in dot and blot immune overlay assays with immunogold and
immunogold/silver staining. J. immun. Meth. 74, 353-360.
NEWMAN, P. AND WATT, F. M. (1988). Influence of cytochalasin Dinduced changes in cell shape on proteoglycan synthesis by cultured
articular chondrocytes. Expl Cell Res. 178, 199-210.
OEGEMA, T. R., HASOALL, V. C. AND DZIEWIATKOWSKI, D. D. (1975).
Isolation and characterization of proteoglycans from the Swarm rat
chondrosarcoma. J. biol. Chem. 250, 6151-6159.
OHNO, H., BLACKWELL, J., JAMIESON, A. M., CARRINO, D. A. AND
PALFREY, A. J. AND DA VIES, D. V. (1966). The fine structure of
chondrocytes. J. Anat. 100, 213-226.
PEARSON, J. P. AND MASON, R. M. (1977). The stability of bovine nasal
cartilage proteoglycans during isolation and storage. Biochim.
biophys. Ada 498, 176-188.
POOLE, A. R., PIDOUX, I., REINER, A. AND ROSENBERG, L. (1982). An
immunoelectron microscope study of the organization of proteoglycan
monomer, link protein, and collagen in the matrix of articular
cartilage. J. Cell Biol. 93, 921-937.
RATCLIFFE, A., FRYER, P. R. AND HARDINGHAM, T. E. (1984). The
distribution of aggregating proteoglycans in articular cartilage:
Comparison of quantitative immunoelectron microscopy with
radioimmunoassay and biochemical analysis. J. Histochem. Cytochem.
32, 193-201.
RATCLIFFE, A. AND HARDINGHAM, T. (1983). Cartilage proteoglycan
binding region and link protein. Radioimmunoassays and the
detection of masked determinants in aggregates. Biochem. J. 213,
371-378.
REYNOLDS, J. J. AND DEW, G. W. (1977). Comparison of the inhibition of
avian and mammalian bone alkaline phosphatases by levamisole and
compound. R8231. Experientia 33, 154-155.
SCHMID, T. M. AND LINSENMAYER, T. F. (1985). Immunohistochemical
localization of short chain cartilage collagen (type X) in avian tissues.
J. Cell Biol. 100, 598-605.
STANESCU, V., MAROTEAUX, P. AND SOBCZAK, E. (1980). Proteoglycan
populations of baboon (Papio papio) cartilages from different
anatomical sites. Gel electrophoretic analysis of dissociated
proteoglycans and of fractions obtained by density gradient
centrifugation. Biochim. biophys. Ada 629, 371-381.
STOCKWELL, R. A. AND SCOTT, J. E. (1967). Distribution of acid
glycosaminoglycans in human articular cartilage. Nature 215,
1376-1378.
STOKER, M. (1968). Abortive transformation by polyoma virus. Nature
218, 234-238.
TACCHETTI, C , QUATRO, R., NITSCH, L., HARTMANN, D. J. AND
CANCEDDA, R. (1987). In vitro morphogenesis of chick embryo
hypertrophic cartilage. J. Cell Biol. 105, 999-1006.
TOWBIN, H., STAEHELIN, T. AND GORDON, J. (1979). Electrophoretic
transfer of proteins from polyacrylamide gels to nitrocellulose sheets:
procedure and some applications. Proc. natn. Acad. Sci. U.S.A. 76,
4350-4354.
VENN, M. F. (1978). Variation of chemical composition with age in
human femoral head cartilage. Ann. rheum. Dis. 37, 168-174.
WASTESON, A., UTHNE, K. AND WESTERMARK, B. (1973). A novel assay
for the biosynthesis of sulphated polysaccharide and its application to
studies on the effects of somatomedin on cultured cells. Biochem. J.
136, 1069-1074.
WATT, F. M. (1988). Effect of seeding density on stability of the
differentiated phenotype of pig articular chondrocytes in culture. J.
Cell Sci. 89, 373-378.
WATT, F. M. AND DUDHIA, J. (1988). Prolonged expression of
differentiated phenotype by chondrocytes cultured at low density on a
composite substrate of collagen and agarose that restricts cell
spreading. Differentiation 38, 140-147.
ZANETTI, M., RATCLIFFE, A. AND WATT, F. M. (1985). Two subpopulations
of differentiated chondrocytes identified with a monoclonal antibody to
keratan sulfate. J. Cell Biol. 101, 53-59.
CAPLAN, A. I. (1986). Calibration of the relative molecular mass of
proteoglycan subunit by column chromatography on Sepharose CL2B.
Biochem. J. 235, 553-557.
OLDBERG, A., HOOK, M., OBRINK, B., PERTOFT, H. AND RUBIN, K. (1977).
Structure and metabolism of rat liver heparan sulphate. Biochem. J.
164, 75-81.
360
M. Siczkowski and F. M. Watt
(Received 8 May 1990 - Accepted 22 June 1990)