Proteoglycan synthesis by sternal chondrocytes was studied in the presence of excess vitamin A (10 i.u./ml). Proteoglycans synthesized by the treated cells were smaller, and had larger amounts of chondroitinase ABC-resistant materials than control cells. Vitamin A-pretreated cells, when provided with normal feeding medium, failed to revert back to normal morphology and synthetic processes. Chrondrocyte cultures prelabelled with [35S]- sulphate, when maintained in the presence of excess of vitamin A, showed: (1) increased release of labelled proteoglycans into the medium, and (2) increased (19%) degradation of the proteoglycans. The proteoglycans synthesized by the vitamin A-treated chondrocytes are also incapable of binding with exogenous large molecular weight hyaluronic acid. Thus, high levels of vitamin A modulate the differentiation of chondrocytes by altering cellular synthetic processes.

Vitamin A has a well-documented effect upon cartilage, both in vivo and in vitro. Vitamin A in excess causes resorption of cartilage (Wolbach, 1947), loss of extracellular glycosaminoglycans (Fell & Mellanby, 1952; Goodman, Smith, Hembry & Dingle, 1974), and increases the synthesis and release of lyzosomal enzymes (Lucy, Dingle & Fell, 1961; Fell & Dingle, 1963).

Recently, Solursh & Meier (1973) have reported that vitamin A in excess inhibits mucopolysaccharide synthesis and enhances the degradation of cartilage matrix in chondrocyte cultures. In an earlier investigation we have reported that vitamin A in excess depresses the synthesis and increases the rate of degradation of mucopolysaccharide, the greater effect being on the synthesis of mucopolysaccharide (Vasan & Lash, 1975). Studying the effect of various levels of retinoic acid on chondrocytes, Shapiro & Poon (1976) concluded that the effect could best be explained in terms of modification of mucopolysaccharide synthesis, rather than accelerated degradation. It has been reported that while the proteoglycans synthesized by vitamin A-treated limb mesenchymal cell culture were smaller, the mannosylation of a specific fraction of glycopeptide was stimulated (Pennypacker, Lewis & Hassell, 1978). Vitamin A-treated cells also retained a 240000 dalton surface protein and retarded the appearance of an 80000 dalton surface protein (Lewis, Pratt, Pennypacker & Hassell, 1978). In this report, using sternal chondrocytes, the effect of excess vitamin A on the synthesis of proteoglycan molecules has been extended further.

Chondrocytes were obtained from 13-day-old embryonic chick sterna by the previously described method (Vasan & Lash, 1975). The freshly liberated chondrocytes were plated at an initial density of 5 x104 cells per 60 mm Falcon tissue culture dish, containing 5 ml of the F12X nutrient medium (Marzullo & Lash, 1970). The medium was supplemented with 7% foetal calf serum (GIBCO) which contained 1% penicillin-streptomycin and 1% kanamycin.

Hypervitaminosis A chondrocytes were supplied with F12X feeding medium containing 10 i.u./ml vitamin A (retinyl acetate) (kindly supplied by Dr. Stanley Shapiro, Roche Research Center, Hoffman-La Roche, Nutley, N.J.). Vitamin A was added at zero time, just before plating. The possibility of initial damage was excluded by the fact that after 24 h of culture both the control and the experimental showed similar plating efficiencies (51 ±3%) (Vasan & Lash, 1975). Similar results were also obtained by Kochhar, Dingle & Lucy (1968) using vitamin A-treated fibroblasts. The cultures were fed daily by replacing half the medium with fresh medium. All cultures were maintained at 38°C in a humidified atmosphere of 95% air/5% CO2 for the experimental periods.

In some experiments, cultures were pre-labelled to study the effect of vitamin A by exposing 4-day cultures to [35S]sulphate (Na2[35S]O4-carrier free) overnight (16 h) and then washed three times with normal feeding medium, before adding the medium containing vitamin A (10 i.u./ml). The media collected after 72 h were frozen at -70°C until further analyses were made.

Other experiments were conducted to investigate whether the effect of vitamin A treatment is reversible. For this, cultures were grown in a medium containing vitamin A (10 i.u./ml) for four days. Then the cell layer was washed five times with Sims balanced salt solution and the feeding medium before replacing with normal feeding medium, and cultured an additional 72 h. Vitamin A-treated cultures were also trypsinized (as suggested by Dr P. F. Goetinck, Storrs, Conn.), replated at the initial density and provided with normal medium for 5 days.

In all the experiments the cultures were exposed to media containing 10 μCi/ ml of Na2[35S]O4 (carrier free - New England Nuclear, Boston, Ma.).

Proteoglycan extraction and analysis

The medium was removed, the cultures were washed twice with 2 ml of saline, and the washes were pooled with the medium. Proteoglycan was precipitated using cold ethanol (final concentration 70%) containing 1·3% potassium acetate (Kimata, Okayama, Oohira & Suzuki, 1974). The precipitate was dis-solved in 4·0 M guanidinium chloride (GuHCl) containing proteolytic inhibitors (Oegema, Hascall, Dziewiatkowski, 1975) to solubilize the proteoglycans. The insoluble residue was removed by centrifugation and the extract was dialysed against 0·5 M-NaCl before being subjected to column chromatography.

The proteoglycan from the cell layer was extracted using 4·0 M-GUHC1 solution buffered with 0·05 M sodium acetate, pH 5·8, containing proteolytic inhibitors, at 5°C for 24 h. The insoluble material was removed by centrifugation and the extract was subjected to various analyses after 48 h of dialysis.

Characterization of [35S]sulphate-labelled glycosaminoglycans

The relative amounts of chondroitin 4-sulphate and chondroitin 6-sulphate synthesized by cell cultures incubated in the presence of [35S]sulphate were determined by the enzymatic method of Saito, Yamagata & Suzuki (1968) using chondroitinase ABC and AC (Miles Laboratories, Inc.).

Molecular sieve chromatography of proteoglycans

Column chromatography was performed using controlled-pore glass beads (CPG-10-2500; Electronucleonics, Fairfield, N.J.) according to the procedure described earlier (Vasan & Lash, 1979). The elution salt, 0·5 M-NaCl contained 0·02% sodium azide. The eluate was collected in 1·0 ml fractions and 35S. incorporation measured by liquid scintillation counter. In some experiments an aliquot was counted and the remainder used for other analyses.

Preparation of A1–D1 proteoglycan monomers

Proteoglycan monomers (A1-D1) were obtained by the method of Hascall & Heinegard (1974). Proteoglycan extracted with 4·0 M-GUHC1 was first dialysed against 100 volumes of 0·5 M-GUHC1 buffer, pH 5·8, (containing proteolytic inhibitors) for 24 h at 4°C. An aggregate (A1) fraction was then prepared with associative caesium chloride density gradient centrifugation (initial density 1·65 g/ml, 40000 rev./min, 15°C, 48 h in a fixed angle rotor). The A1 fraction was subsequently partioned in a dissociative gradient (initial density 1·52 g/ml, 40000 rev./min, 15°C, 48 h, fixed angle rotor) to separate monomeric proteoglycans (A1–D1). This monomer was subjected to extensive dialysis and lyophilized.

Interaction of A1–D1 Fractions with hyaluronic acid

[35S]sulphate-labelled proteoglycan monomer (A1–D1 obtained as described above) in 4·0 M-GUHC1 was mixed with 1 mg of unlabelled proteoglycan monomer obtained from 14-day embryonic chick sterna. To this, hyaluronic acid (1 mg/ml in 4·0 M-GUHC1) was added. (High-molecular-weight cockscomb hyaluronic acid was kindly provided by Dr E. A. Balaz, New York, U.S.A.). The mixtures, as well as controls (A1–D1 alone) without hyaluronic acid, were set at room temperature for 2 h. They were dialysed against 0·5 M-NaCl at 4°C (with two changes of dialysing solution, solvent conditions which facilitate the interaction of proteoglycans monomer with hyaluronic acid (Hardingham & Muir, 1972; Vasan & Lash, 1978, 1979). The samples were chromatographed on a CPG-10-2500 Column and the eluted fractions analysed for radioactivity.

Morphology and matrix accumulation

Chondrocytes supplied with normal medium exhibited the typical polygonal morphology and accumulated precocious amounts of hyaline matrix after 48 h of culture (Fig. 1 a). Chondrocytes exposed to vitamin A became flattened and stellate within the first 24 h of culture (Fig. 1 b) and failed to accumulate extracellular matrix. After 4 days of vitamin A exposure, chondrocytes failed to return to normal morphology even after 48 h of feeding with control medium (Fig. 1 c). There was also an inhibition of [35S]sulphate incorporation into cetylpyridinium chloride-precipitable glycosaminoglycans and an increase in the release of glycosaminoglycans into the medium due to vitamin A exposure. Trypsinization and replating of vitamin A-treated cells, provided with normal feeding medium neither changed the cellular morphology to normal polygonal shape nor resulted in the accumulation of hyaline matrix (Fig. 1 d).

Fig. 1

Phase-contrast micrographs showing the morphology of chondrocytes grown (a) 6 days in F12X medium; (b) F12X supplied with 10 i.u./ml of vitamin A, (c) and cultures pretreated with vitamin A for 4 days and again cultured in normal F12X medium for over 48 h. Control chondrocytes show a typical polygonal morphology with surrounding hyaline matrix (a) while the vitamin A-treated culture (b) becomes fibroblastic with very little matrix. Vitamin A-treated cultures when provided with normal feeding medium (c) or trypsin treated and then provided with normal feeding medium (d) did not reverse the cellular morphology to normal polygonal shape, × 240.

Fig. 1

Phase-contrast micrographs showing the morphology of chondrocytes grown (a) 6 days in F12X medium; (b) F12X supplied with 10 i.u./ml of vitamin A, (c) and cultures pretreated with vitamin A for 4 days and again cultured in normal F12X medium for over 48 h. Control chondrocytes show a typical polygonal morphology with surrounding hyaline matrix (a) while the vitamin A-treated culture (b) becomes fibroblastic with very little matrix. Vitamin A-treated cultures when provided with normal feeding medium (c) or trypsin treated and then provided with normal feeding medium (d) did not reverse the cellular morphology to normal polygonal shape, × 240.

Molecular sieve chromatography of proteoglycans

To determine more accurately the molecular size distribution during chondrogenic expression, molecular sieve chromatography was used to characterize the proteoglycans. Using the technique of separation by means of controlled pore glass beads (Vasan & Lash, 1978, 1979; Lash, Ovadia & Vasan, 1978) it is now possible to resolve proteoglycans into an excluded large aggregate fraction (PG-A), an included intermediate size fraction (PG-1) and a monomer fraction eluted close to the total volume (PG-M). Proteoglycans extracted from the control culture showed the presence of large proteoglycan aggregate (PG-A) and an intermediate size population (PG-I) included in the column which eluted close to the void volume (Fig. 2). On the contrary proteoglycans extracted from vitamin A-treated culture resolved into only one included peak eluted close to the total volume (PG-M) (Fig. 2).

Fig. 2

Molecular sieve chromatographic profile of proteoglycans on a CPG-10-2500 column. Proteoglycans from control (solid line) and vitamin A-treated (dotted line) cell layer were extracted with 4·0 M-GUHC1 as described in the methods and chromatographed. The bars on the top of the graph showing proteoglycan aggregates (PG-A), proteoglycan intermediates (PG-I) and proteoglycan monomers (PG-M) are the characteristic of any mature embryonic cartilage.

Fig. 2

Molecular sieve chromatographic profile of proteoglycans on a CPG-10-2500 column. Proteoglycans from control (solid line) and vitamin A-treated (dotted line) cell layer were extracted with 4·0 M-GUHC1 as described in the methods and chromatographed. The bars on the top of the graph showing proteoglycan aggregates (PG-A), proteoglycan intermediates (PG-I) and proteoglycan monomers (PG-M) are the characteristic of any mature embryonic cartilage.

The proteoglycans isolated by molecular sieve chromatography were subjected to chondroitinases AC and MBC digestion and the resultant disaccharides were separated on a descending paper chromatography. Table 1 represents the various amounts of [35S]sulphated acid mucopolysaccharide in these fractions. The small molecular size fraction (PG-M) obtained from vitamin A-treated culture was composed mainly of chondroitin 4-sulphate and enzyme-resistant material. A major proportion of this enzyme-resistant material was reported as heparan sulfate (Pennypacker et al. 1978). Proteoglycan fractions (PG-A and PG-I) obtained from control cultures showed a higher proportion of chondroitin 4- and 6-sulphate while the enzyme-resistant material was low (Table 1).

Table 1

Relative amounts of [35S]sulphated acid mucopolysaccharides present in the various proteoglycan fractions resolved by CPG-10-2500 column chromatography

Relative amounts of [35S]sulphated acid mucopolysaccharides present in the various proteoglycan fractions resolved by CPG-10-2500 column chromatography
Relative amounts of [35S]sulphated acid mucopolysaccharides present in the various proteoglycan fractions resolved by CPG-10-2500 column chromatography

Additional experiments were carried out to determine whether the alterations noticed in proteoglycan synthesis due to exposure of the chondrocytes to vitamin A could be reversed. In the 72 h period with normal medium there was no reversal of phenotypic expression observed (Fig. 1c). At the same time the type of proteoglycan synthesized, after 24, 48, and 72 h feeding with normal medium did not result in any appreciable change in the distribution of molecular sizes (Fig. 3).

Fig. 3

Elution pattern of proteoglycans by CPG-10-2500 Column. Chondrocytes were grown in medium containing 10 i.u./ml vitamin A for 4 days. Then the cell layer was washed five times with normal medium and the chondrocyte recovery was studied. There was no morphological recovery (Fig. 1 c). Cultures were exposed to radioactive sulphate for 16 h and the proteoglycans in the cell layer analysed after 24 h (△), 48h(□) and 72 h (◯).

Fig. 3

Elution pattern of proteoglycans by CPG-10-2500 Column. Chondrocytes were grown in medium containing 10 i.u./ml vitamin A for 4 days. Then the cell layer was washed five times with normal medium and the chondrocyte recovery was studied. There was no morphological recovery (Fig. 1 c). Cultures were exposed to radioactive sulphate for 16 h and the proteoglycans in the cell layer analysed after 24 h (△), 48h(□) and 72 h (◯).

Investigation was further extended to probe whether the small molecular size proteoglycans observed due to vitamin A treatment were the result of degradation of chondrocyte proteoglycans or a defect in the synthetic mechanism. Normally grown, prelabelled (35S) chondrocytes were exposed to vitamin A-containing medium for 72 h and the proteoglycan in the medium was analyzed. In addition to the increase in the release of pre-labelled proteoglycans, vitamin A also caused a certain amount of degradation (19%), resulting in the small molecules eluted close to the total volume (Fig. 4).

Fig. 4

Cultures were grown in normal medium, prelabelled with radioactive sulphate and then maintained in the presence of 10 i.u./ml of vitamin A for 72 h. The proteoglycans in the medium were then subjected to potassium acetate-ethanol precipitation. The elution profile on pattern of control (solid line) and vitamin A-treated (dotted line) are shown in this figure.

Fig. 4

Cultures were grown in normal medium, prelabelled with radioactive sulphate and then maintained in the presence of 10 i.u./ml of vitamin A for 72 h. The proteoglycans in the medium were then subjected to potassium acetate-ethanol precipitation. The elution profile on pattern of control (solid line) and vitamin A-treated (dotted line) are shown in this figure.

The decrease in the size of proteoglycans synthesized by the vitamin A-treated chondrocyte culture could be due to the lack of ability to form large aggregates. One of the reasons for such failure could be due to the failure of the synthesis of proper core protein which possesses the important hyaluronic acidbinding region. To test this hypothesis, the monomeric fraction (A1–D1) obtained after CsCl-GuHCl gradient centrifugation was mixed with cockscomb hyaluronic acid and subjected to column chromatography. Proteoglycan monomer (A1–D1) obtained from vitamin A-treated chondrocyte culture failed to show appreciable affinity for hyaluronic acid while the control A1–D1 fractions showed hyaluronic acid-binding capacity (Fig. 5).

Fig. 5

CPG-10-2500 Column Chromatography of proteoglycan monomers (A1–D1) from control (dotted line) and vitamin A treated (dotted line with open circle (◯)). Upon the addition of hyaluronic acid, there is significant interaction, and larger aggregates are formed in control culture (solid line) while there is no change in the vitamin A-treated culture (solid line with solid circle (●)).

Fig. 5

CPG-10-2500 Column Chromatography of proteoglycan monomers (A1–D1) from control (dotted line) and vitamin A treated (dotted line with open circle (◯)). Upon the addition of hyaluronic acid, there is significant interaction, and larger aggregates are formed in control culture (solid line) while there is no change in the vitamin A-treated culture (solid line with solid circle (●)).

Inhibition of chondrogenic differentiation using excess of vitamin A has been well documented (Solursh & Meier, 1973; Vasan & Lash, 1975; Pennypacker et al. 1978; Lewis et al. 1978). Treated cells in cultures did not develop cartilage nodules with predominant quantities of hyaline matrix. Instead of polygonal morphology (Fig. 1 a) they became flattened and stellate (Fig. 1 b). Vitamin A-pretreated cultures when postcultured with normal feeding medium for 72 h, or trypsinized and cultured in normal medium, did not show any change in morphology (Fig. 1 c, d). When well differentiated colonies of chondrocytes were picked from the primary cultures and subcultured in the presence of vitamin A, there were alterations in the morphological and biochemical synthetic mechanisms as seen in the present study (work in progress) (Solursh & Meier, 1973). Hence the observations made in this report may not be due to the selection of chondrocytes by the normal medium and non-chondrocytes by the Vitamin A medium. Recently, Jetten, Jetten & Sherman (1979) showed that retinoic acid causes phenotypic changes in embryonal carcinoma cells (an undifferentiated stem cell of teratocarcinomas). The observed phenotypic alteration induced by retinoic acid is maintained, even several generations after retinoic acid is removed from the culture medium. The absence of a cartilagenous matrix is mainly due to the inhibition in the synthesis and not due to an extraordinary degradation (Vasan & Lash, 1975). The alteration in the cellular morphology seen in the treated culture could be due to the retention of high molecular weight glycoprotein involved in cell-cell adhesion and, therefore, the maintenance of close cell contacts (Lewis et al. 1978; Hassell, Pennypacker, Yamada & Pratt, 1978).

Vitamin A-treated chondrocyte cultures failed to elaborate large aggregated proteoglycan molecules. Chromatographic separation of proteoglycan extracted from the treated chondrocytes showed radioactive sulphate incorporation into small size molecules (Fig. 2). This is similar to the proteoglycan molecule isolated and fractionated from limb mesenchymal cells (Pennypacker et al. 1978), stage-18 and -24 limb buds (Vasan & Lash, 1979), tendon fibroblast and myoblast cultures (Vasan, unpublished observations) and virus transformed somite cultures (Yoshimura et al. 1980). Chondrocyte cultures pretreated with Vitamin A, then supplied with normal medium failed to show any appreciable changes in their phenotypic (Fig. 1 c) or genotypic expression up to 72 h (Fig. 3).

Small-molecular-size proteoglycan obtained after vitamin A treatment resembled that of the mesenchymal or non-chondrogenic cells. This may be the result of changes in synthetic processes or increased degradation. Vitamin A causes an enhancement in the release of proteoglycan into the medium (Vasan & Lash, 1975; Pennypacker et al. 1978). When prelabelled (35S) chondrocytes were exposed to vitamin A for 72 h, the released 35S-labelled proteoglycan showed some small-size proteoglycans eluted close to the total volume (Fig. 4). At least 19% of the small-size proteoglycans could be due to the degradation of the molecules by the lyzosomal enzymes (Lucy et al. 1961; Fell & Dingle, 1963; Vasan & Lash, 1975). This is in contrast to the recent report of Pennypacker et al. (1978) where a 48 h exposure to vitamin A of prelabelled culture did not show any degradation. This discrepancy may be possible because Pennypacker et al. (1978) used only the A1 fraction for column chromatography, and the low-molecular-weight molecules (A1–D2) were still in the low-bouyant-density fractions that were not included for chromatography. In the present study, total proteoglycan was used without partitioning by gradient centrifugation.

Proteoglycans synthesized by mesenchymal tissues (stage-18 and -24 limb buds) were predominantly small molecular size and failed to show any appreciable amount of hyaluronic-acid-binding capacity (Vasan & Lash, 1979). It has been recognized, that hyaluronic-acid-binding by proteoglycan monomers is necessary for the large aggregate formation. The decrease in hydrodynamic size, and the elution close to the total volume are typical features of the proteoglycans obtained from vitamin A-treated culture. This could be due to a decrease in the length of core protein. The decrease in the length of core protein could be either at the hyaluronic-acid-binding region or chondroitin-sulphate-rich region. Dissociated proteoglycan molecules (A1–D1) obtained from vitamin A-treated chondrocytes showed a decrease in the hyaluronic-acid-binding capacity (Fig. 5). This could be due to the enzymatic cleavage of the protein region or due to the failure of the synthesis of the hyaluronic-acid-binding region. Thus the vitamin A-treated culture produces proteoglycans which are not only incapable of forming large aggregates (as the non-aggregating proteoglycans described by Muir and coworkers), but chromatographically resemble the non-chondrogenic type of proteoglycans synthesized by transformed chondrocytes, muscle cells and tendon fibroblasts (unpublished observations). Work is in progress to gain more knowledge on the chemical nature of these smaller proteoglycan molecules obtained from different tissues and chondrocytes grown under various conditions. The result of the present investigation shows, that Vitamin A, in addition to causing a definite increase in the degradation of proteoglycans, also inhibited the differentiation of chondrocytes mostly by altering the synthetic mechanism.

I am grateful to Dr Stanley Shapiro and Dr Peddrick Weis for the critical evaluation of the manuscript and Dr Paul Goetinck for experimental suggestions. I like to thank Mr Markus Meyenhofer for assistance in photography. This investigation was supported by a grant from National Institute of Health Biomedical Research Support 5 S07 RR05393 to the College of Medicine and Dentistry of New Jersey, New Jersey Medical School.

Fell
,
H. B.
&
Dingle
,
J. T.
(
1963
).
Studies on the mode of action of excess of vitamin A6. Lysosomal protease and the degradation of cartilage matrix
.
Biochem. J
.
87
,
403
408
.
Fell
,
H. B.
&
Mellanby
,
E.
(
1952
).
The effect of hypervitaminosis A on embryonic limbbones cultivated in vitro
.
J. Physiol
.
116
,
320
349
.
Goodman
,
D. W.
,
Smith
,
J. E.
,
Hembry
,
R. M.
&
Dingle
,
J. T.
(
1974
).
Comparison of the effect of vitamin A and its analogs upon rabbit ear cartilage in organ culture and upon the growth of vitamin A deficient rats
.
J. Lipid Res
.
15
,
406
414
.
Hardingham
,
T. E.
&
Muir
,
H.
(
1972
).
The specific interaction of hyaluronic acid with cartilage proteoglycans
.
Biochem. Biophys. Acta
.
279
,
401
405
.
Hascall
,
V. C.
&
Heinegard
,
D.
(
1974
).
Aggregation of cartilage proteoglycans. I. The role of hyaluronic acid
.
J. biol. Chem
.
249
,
4232
4241
.
Hassell
,
J. R.
,
Pennypacker
,
J. P.
,
Yamada
,
K. M.
&
Pratt
,
R. M.
(
1978
).
Changes in cell surface proteins during normal and vitamin A inhibited chondrogenesis in vitro
.
Ann. N. Y. Acad. Sci
.
312
,
406
409
.
Jetten
,
A. M.
,
Jetten
,
M. E.
&
Sherman
,
M. I.
(
1979
).
Stimulation of differentiation of several Murine embryonal carcinoma cell lines by retinoic acid
.
Expl Cell Res
.
124
,
381391
.
Kimata
,
K.
,
Okayama
,
M.
,
Oqhira
,
A.
&
Suzuki
,
S.
(
1974
).
Heterogeneity of proteo-chondroitin sulfates produced by chondrocytes at different stages of cytodifferentiation
.
J. biol. Chem
.
249
,
1641
1653
.
Kochhar
,
D. M.
,
Dingle
,
J. T.
&
Lucy
,
J. A.
(
1968
).
The effect of vitamin A (retinol) on cell growth and incorporation of labeled glucosamine and proline by mouse fibroblasts in culture
.
Expl Cell Res
.
52
,
591
601
.
Lash
,
J. W.
,
Ovadia
,
M.
&
Vasan
,
N. S.
(
1978
).
Microheterogeneities, non-equivalence, and embryonic induction
.
Medical Biol
.
56
,
333
338
.
Lewis
,
C. A.
,
Pratt
,
R. M.
,
Pennypacker
,
J. P.
&
Hassell
,
J. R.
(
1978
).
Inhibition of limb chondrogenesis in vitro by vitamin A: Alterations in cell surface characteristics
.
Devl Biol
.
64
,
31
47
.
Lucy
,
J. A.
,
Dingle
,
J. T.
&
Fell
,
H. B.
(
1961
).
Studies on the mode of action of vitamin A. 2. A possible role of intracellular proteases in the degradation of cartilage matrix
.
Biochem. J
.
79
,
500
508
.
Marzullo
,
G.
&
Lash
,
J. Xv
. (
1970
).
Control of phenotypic expression in cultured chondrocytes: Investigations on the mechanism
.
Devl Biol
.
22
,
638
654
.
Oegema
,
R. T.
Jr.
,
Hascall
,
V. C.
&
Dziewiatkowski
,
D. D.
(
1975
).
Isolation and characterization of proteoglycans from the swarm rat chondrosarcoma
.
J. biol. Chem
.
250
,
6151
6159
.
Pennypacker
,
J. P.
,
Lewis
,
C. A.
&
Hassell
,
J. R.
(
1978
).
Altered proteoglycan metabolism in mouse limb mesenchyme cell cultures treated with vitamin A
.
Arch. Biochem. Biophys
.
186
,
351
358
.
Saito
,
H.
,
Yamagata
,
T.
&
Suzuki
,
S.
(
1968
).
Enzymatic methods for the determination of small quantities of isomeric chondroitin sulfate
.
J. biol. Chem
.
243
,
1536
1542
.
Shapiro
,
S. S.
&
Poon
,
J. P.
(
1976
).
Effect of retinoic acid chondrocyte glycosaminoglycans biosynthesis
.
Arch. Biochem. Biophys
.
174
,
74
81
.
Solursh
,
M.
&
Meier
,
S.
(
1973
).
The selective inhibition of mucopolysaccharide synthesis by vitamin A treatment of cultured chick embryo chondrocytes
.
Calcif. Tiss. Res
.
13
,
131
142
.
Vasan
,
N. S.
&
Lash
,
J. W.
(
1975
).
Chondrocytes metabolism as affected by vitamin A
.
Calcif. Tiss. Res
.
19
,
99
107
.
Vasan
,
N. S.
&
Lash
,
J. W.
(
1978
).
Proteoglycan heterogeneity in embryonic chick articular and epiphyseal cartilages
.
Conn. Tiss. Res
.
6
,
191
199
.
Vasan
,
N. S.
&
Lash
,
J. W.
(
1979
).
Monomeric and aggregate proteoglycan in the chondrogenic differentiation of embryonic chick limb buds
.
J. Embryol. exp. Morph
.
49
,
47
59
.
Yoshimura
,
M.
,
Lash
,
J. W.
,
Vasan
,
N. S.
&
Kaji
,
A.
(
1980
).
The effect of oncogenic virus on the in vitro chondrogenic differentiation of embryonic chick somites (manuscript submitted to Devl Biol.)
.
Wolbach
,
S. B.
(
1947
).
Vitamin A deficiency and excess in relation to skeletal growth
.
J. Bone Joint Surg
.
29
,
171
192
.