two hormones, insulin and the rat-derived, somatomedin- like polypeptide, multiplication-stimulating activity (MSA),' stimulate proteoglycan biosynthesis in rat ...
Vol. 256, No.4, Issue of February 25. pp. 2053-2058, 19.91 Printedin U.S.A.
Characterization of Proteoglycans Synthesizedby Rat Chondrosarcoma Chondrocytes Treatedwith Multiplication-stimulatingActivity and Insulin* (Received for publication, July 21, 1980)
Richard L. Stevens+and VincentC. Hascall From the Laboratory . of . Biochemistry, National Instituteof Dental Research, National Institutesof Health, Bethesda, Maryland 20205
The structures of proteoglycans synthesized andse- MSA was found to be time-dependent, but was less timecretedbySwarmratchondrosarcomachondrocytes dependent than the response after insulin treatment. Based cultured in the presence of insulin, multiplication-stimon radioisotope incorporation into proteoglycan, insulin and ulating activity(MSA), and fetal calf serum were inves- MSA may affect the chondrosarcoma chondrocyte differently. tigated. Proteoglycans produced in all cultures were This investigation, then, was undertaken to compare the found as link protein stabilized aggregates. Monomericstructures of proteoglycans synthesized in response to the proteoglycans from cultures treated with MSAor in- hormones with those synthesized in medium alone or medium sulin were slightly larger in hydrodynamic size than with serum. those synthesized by chondrocytes maintained in unsupplemented medium.This increase reflected a 25 to EXPERIMENTALPROCEDURES 30% increase in thelengthofthecovalentlybound Materials-Sepharose CL-2B, Sephadex G-200, and Sephadex Gchondroitin sulfate side chains. No free glycosamino- 25,PD-10 columns were obtained from Pharmacia; chondroitinase glycans were detected in any cultures, although the ABC, 2-acetamido-2-deoxy-3-0-(~-~-gluco-4-enepyranosyluronic at acid)D-galactose (ADiOS) and 2-acetamido-2-deoxy-3-0-(~-~-glucopresence of a lower molecular weight proteoglycan 4-enepyranosyluronic acid)-6-O-sdfo-~-galactose (ADi6S) were oba significant concentration was observed in the medium MSA and insulin tained from Miles Laboratories. Other materials are described in the fraction of serum-maintained cultures. treatments did not result in the synthesisof a proteo- preceding paper (1). Isolation a n d Culturing of Cells and Labeling of Newly Syntheglycan containing other types of glycosaminoglycans Proteoglycan-Chondrocytes derived from the transplantable of sized nor did they alter the degree or position of sulfation Swarm rat chondrosarcoma were maintained in culture as previously the chondroitin sulfate glycosaminoglycans. The small described (1) in the presence of serum-free Dulbecco’smodified changes in proteoglycan structure do not account for Eagle’s medium supplemented with glucose (hereafter referred to as the larger differences in thelevels of incorporation of Medium A). Treated cultures were grown in Medium A alone or radioactivity indicating that the primary effect of MSA Medium A supplemented with either 15%fetal calf serum, MSA (0.05 and insulin on proteoglycan synthesis in this in vitro to 2 pg/ml), or insulin (1 ng to 0.5 pg/ml). Cells were labeled for 3 h system is an increased rate of synthesis and secretion with 20 to 50 pCi of [%]sulfate, [3H]serine,or [3H]glucosamine/mlof medium as previously described (1). The structures of the newly of proteoglycan.
The preceding paper presents data which demonstrate that two hormones, insulin and the rat-derived, somatomedinlike polypeptide, multiplication-stimulating activity (MSA),’ stimulate proteoglycan biosynthesis in rat chondrosarcoma chondrocytes (1). When insulin at 10 n g / d was present in serum-free medium for 3 days, incorporation of [35S]sulfate into proteoglycans was stimulated 2- to %fold over that obtained in cultures maintained in medium alone, and the level of proteoglycan synthesis was as high as that obtained for cultures maintained in medium supplemented with 15%fetal calf serum. MSA also stimulated proteoglycan synthesis, but generally not to the same level, and the molar concentration required for half-maximal stimulation was 50to 100 times that for insulin. The stimulation of proteoglycan biosynthesis by
’This work was supported by the Arthritis Foundation, Atlanta, Georgia. The costs of publication of this articlewere defrayed in part by the payment of page charges. This article must thereforebe hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. $Present address, Department of Medicine, Harvard Medical School and Robert B. Brigham Hospital, Boston, MA 02115. The abbreviations used are: MSA, multiplication-stimulating activity; aAI-Dl, proteoglycan monomer isolated from the ratchondroSarcoma as described in Ref. 16.
synthesized proteoglycans localized in the two major radiolabeled fractions, namely the medium and the 4 M guanidine HCl extract of the cell layer, were investigated. Aggregation of Proteoglycans Locatedin the Medium FractionsChondrocyte cultures, which were maintained for 3 days in Medium A alone or Medium A supplemented with MSA, ins.ulin, or fetal calf serum, were labeled with [35S]sulfatefor 3 h, the medium removed and frozen a t -4OOC until analyzed. Aliquots of the medium fractions (125 pl) were mixed with 100 pl of 1.2 M sodium acetate, pH 6.8, and with 25 pl of a rat chondrosarcoma monomer proteoglycan (aAI-Dl) solution (10 mg/ml). After 2 h at 4”C, samples were applied to an analytical Sepharose CL-2B column (0.6 X 120 cm), eluting under the associative conditions of 0.5 M sodium acetate, 0.1 M sodium sulfate, pH 6.8. Each 0.5-ml fraction was mixed with 1 ml of H 2 0and 12.5 ml of Aquasol, and analyzed for radioactivity. The elution position of aggregated chondrosarcoma proteoglycan (aAl) and free [3sS]sulfate were used to indicate the position of the excluded ( Vo)and total ( V t ) volumes, respectively. Relative Hydrodynamic Sizes of Proteoglycans-The hydrodynamic sizes of the newly synthesized radiolabeled proteoglycans were investigated by Sepharose CL-2B chromatography by eluting the column under the dissociative conditions of 4 M guanidine HC1, which prevents proteoglycan aggregation to hyaluronic acid (2). Aliquots (125 p1) of medium fractions from treated cultures were mixed with 8 M 25 pl of 10 m g / d of proteoglycan monomer and 125plof guanidine HCI. Aliquots of 4 M guanidine HC1 extracts were analyzed directly after the addition of the unlabeled proteoglycan monomer. Samples were applied to a Sepharose CL-2B column (0.6 X 120 cm) and elutedwith 4 M guanidine HCl, 0.1 M Tris-HC1, and 0.1 M NaS04, pH 7.0. Fractions of 0.5 ml were mixed with an equal volume of 70%
Proteoglycans from MSA Insulin-treated and
ethanol and, after the addition of 12.5 ml of Aquasol, analyzed for radioactivity. Molecular Weight Determination of Glycosaminoglycuns-0Glycosidically linked glycosaminoglycans were released from the proteoglycans by p-elimination (3,4). Carrier proteoglycan monomer was added to medium and cell extract fractions and samples were dialyzed overnight a t 4'C against Hz0 and lyophilized. Samples were then dissolved in 0.5 ml of 0.05 M NaOH, 1.0 M NaBH4, and heated a t 45OC for 48 h, afterwhich time the samples were cooledand neutralized by the addition of sufficient quantities of 2 M acetic acid. Aliquots (250 4) of each sample were applied to an analytical Sephadex G-200 column (0.6 X 120 cm) and eluted with 0.5 M sodium acetate. The radioactivity was determined for each fraction. The column's Voand Vt were determined using proteoglycan monomer and [35S]sulfate, respectively. The average molecular weights, M,, of released glycosaminoglycans were estimated by the gel tiltration method of Wasteson (5) using the partition coefficient (Kav)of the eluted radioactivity profile. Equilibrium Density Gradient Centrifugation-Aliquots of medium fractions were chromatographed on Sephadex G-25,PD-10 columns in the presence of 4 M guanidine HCl and protease inhibitors ( 6 ) to remove unincorporated isotope (7). The PD-10-excluded volume fractions of medium samples, which contained newly synthesized macromolecules,were adjusted to a density of 1.5g/ml by the addition of solid cesium chloride (CsCl), and equilibrium density gradient centrifugation was carried out under the dissociative conditions of 4 M guanidine HC1 for 48 h at 15°C and 34,OOO rpm in a Beckman SW 50.1 rotor(8). Aliquots of the 4 M guanidine HC1 extracts were centrifuged directly without the PD-10 chromatography step. Gradients were divided into four approximately equal fractions by means of a tube slicer. The radioactivity in each fraction was measured and the bottom (Dl) fractions were dialyzedat 4"C, initially against 0.005 M sodium acetate to remove the majority of CsCl and guanidine HC1, then against 2 M sodium acetate to convert the proteoglycans into their sodium salt, and finally exhaustively against water. Chondroitinase Digestcon-Samples of lyophilized D l proteoglycans from cultures labeled with either [3H]glucosamineor [35S]sulfate were dissolved in 150 pl of 0.1 M sodium acetate, 0.1 M Tris-HC1, pH 8.0. Carrier proteoglycan monomer (25 pl of 10 mg/ml solution) was added, and samples were incubated for 5 h at 37°C with 0.5 unit of chondroitinase ABC as described by Saito et al. (9). NaF (0.001 M) was included in the hydrolysis buffer to inhibit anysulfatase contaminant present in the enzyme preparation. Digests were applied to cellulose thin layer plates and approximately 20 pg of carrier ADi6S and ADiOS added. After a 24-h desalting step using n-butanokethanol: water (78:48:24 v/v), chromatograms were developed with n-butanol: acetic acid2 M NH,OH (2:3:1)? Carrier disaccharides were detected under ultraviolet light, and the unsaturated disaccharide spots were scraped from the plate into scintillation vials. Radioactivity in each sample was then eluted from the cellulose by the addition of 1 ml of 0.1 M HC1 followed by heating for 17 h at 40°C. Aquasol (12.5 m l ) was added and radioactivity determined. The recovery of radioactivity from the thin layer plates was consistently found to be close to 100%. RESULTS
Aggregation of Newly Synthesized Proteoglycans-Aliquots of the medium fractions from a series of cultures maintained in Medium A alone or supplemented with either MSA, insulin, or fetal calf serum were chromatographed in the presence of an excess concentration of proteoglycan monomer carrier on Sepharose CL-2B in associative solvent conditions (Fig. 1). Approximately 90% of the newly synthesized radiolabeled proteoglycans from each culture were localizedin the column's excluded volume. This indicates that the large majority of [35S]sulfate-labeled proteoglycans in the medium fraction for each treated culture were present as link proteinstabilized aggregates (7, 10). Aliquots of the 4 M guanidine HC1 extracts from the different cultures were dialyzedagainst 0.5 M sodium acetate (associative conditions) and after the addition of proteoglycan monomer, they were eluted on the
* This modification of the original procedure for separating disaccharide products in chondroitinase digests was developed by Dr. Roger Mason, Charing Cross Hospital Medical School, London, England.
FIG. 1 (left). Sepharose CL-ZB chromatography of proteoglycans located in the medium. An aliquot of the medium from cultures maintained in Medium A alone ( A )or in Medium A supplemented with MSA ( E ) , insulin ( C ) , or fetal calf serum (D) was analyzed for proteoglycan aggregation. Columns were run under associative conditions. VO and V, mark the column's void and total volumes, respectfully. Free unincorporated radioactivity elutes at V,. FIG. 2 (right). Sepharose CGZB chromatography of proteoglycans located in the medium. An aliquot of the medium fraction from cultures maintained in either Medium A alone ( A )or in Medium A supplemented with MSA ( E ) ,insulin (C), or fetal calf serum (D) was analyzed for relative hydrodynamic size. Columns were run under dissociative conditions. Free unincorporated radioactivity elutes at
same column. The elution profiles of the labeled macromolecules from the cell layer were very similar to those obtained for the medium fractions with about 90% of the proteoglycans recovered in the excludedvolume (data not shown). The stimulation of proteoglycan synthesis by MSA and insulin, therefore, did not result in the production of a class of proteoglycans incapable of formingaggregates. The presence of aggregated proteoglycan under these different conditions of culture also indicates that the cells weresynthesizingsufficient amounts of hyaluronic acid and link protein to accomodate the amount of proteoglycan monomer synthesized. Hydrodynamic Sizes of Newly Synthesized Proteoglycans-Ahquots of the medium fractions were made 4 M in guanidine HC1 (dissociative conditions) and eluted on Sepharose CL-2B in the presence of 4 M guanidine HC1 (Fig. 2). The profiles for all samples except those from the culture maintained with fetal calf serum showed a single, broad peak characteristic of monomer proteoglycans synthesized by chondrocytes. After removal of unincorporated radioactivity by a PD-10 chromatography step prior to Sepharose CL-2B chromatography, no free glycosaminoglycan chains were found in any culture (data not shown). A second population of smaller proteoglycans, representing about 30%of the total35S-labeled macromolecules found in the medium, was observed in the chromatograph of the medium sample from the culture maintained in 15% fetal calf serum. This low molecular weight proteoglycan was not observed in the cell extract fraction of the serum-treated culture and therefore represented about 10%of the total 35S-labeled macromoleculesin this culture. It
Proteoglycans from MSA Insulin-treated and Chondrocytes is likely that this population of proteoglycan was synthesized by the fibroblastic-like cell population which was observedto increase with time in culture when cells were maintained in serum (1). This low molecular size proteoglycan was not prominent by Sepharose CL-2B chromatography when eluted under associative conditions (Fig. Id), suggesting that this species was also able to interact either specifically or nonspecifcally with other macromolecules present in the medium to elute earlier in the column (11). The radiolabeled proteoglycan monomers synthesized in cultures maintained with either MSA or insulin were similar in hydrodynamic size, but were larger than those synthesized in culture with Medium A alone (Fig. 2). When cultures were maintained in Medium A with increasing concentrations of either MSA or insulin, the average sizes of the monomer proteoglycans showed small, but progressive increases. The elution profiles of newly synthesized monomers in medium fractions of day 4 cultures maintained with 0.05 to 2.0 pg/ml of MSA, showthis trend (Fig. 3). Nearly identical results were obtained with a setof cultures maintained in Medium A with from 0 to 10 n g / d of insulin (data notshown). Thus, although MSA and insulin affected proteoglycan synthesis differently (l),the final biosynthetic proteoglycan product after stimulation by these two hormones appeared to be the same in terms of the average molecular size and ability of the proteoglycans to aggregate. Proteoglycans associated with the cell layer-matrix, when extracted using 4 M guanidine HCl, were also found to be of large hydrodynamic size and similar to those isolated from the medium (Fig. 4).Again, the proteoglycans produced by cultures stimulated with MSA or insulin (Fig. 4, B and C) were slightly larger than the proteoglycans produced by cultures maintained in unsupplemented Medium A (Fig. 4A). The presence of a significant proportion of proteoglycans with smaller size as was observed in the medium fraction of the serum-maintained culture was not detected in the extract fractions from any of the cultures. Molecular Weight of Newly Synthesized Glycosaninoglycans-The 35S-labeled glycosaminoglycans, cleavedfrom the
intact proteoglycans by the mild alkaline-borohydride treatment, eluted in the included column volume onSephadex G200. Based on the K,, and the molecular weight standard curves published by Wasteson ( 5 ) ,the molecular weightof the released glycosaminoglycans fromproteoglycans produced by cultures maintained in Medium A alone (Fig. 5A), was approximately 12,500. However, the average molecular weight of the glycosaminoglycans obtained in cultures maintained in higher concentrations of MSA or insulin increased to about 15,500 (Fig. 5, B and C). Cultures maintained in lower concentrations of either of the two hormones, yielded newly synthesized proteoglycans with intermediate molecular sizes(Fig. 3), and correspondingly gave free glycosaminoglycan chains intermediate in sizebetween that observed for MediumA alone and Medium A with high concentrations of hormones. Thus, the increase in hydrodynamic size of newly synthesized monomers in response to hormone treatment correlated with an increase in size of the chondroitin sulfate chains in the molecules. The 25 to 30% net increase in the average molecular weight of the glycosaminoglycans suggests that hormone treatment of the chondrosarcoma chondrocyte influenced, at least indirectly, glycosaminoglycan chain elongation and termination mechanisms. The 25 to 30%net increase in average glycosaminoglycan chain length, however, cannot byitself account for the 200 to 500% increase in incorporation of [35S]sulfateinto macromolecules which occurs when the cultures are maintained in MSA- or insulin-supplemented Medium A (1). Chondroitinase Digestion-Aliquots (250pl) of medium
3.0 2.0 1 .o
3.0 2.0 1 .o
FIG. 3. The relative size of medium proteoglycans produced by chondrocytes maintained in increasing concentration of
MSA. Cultures were exposed to no MSA (-), 0.05 p g MSA/ml (W), 0.1 p g MSA/ml (H or) to, 2.0 .ua - MSA/ml (A-A). Columns wererununder dissociative conditions. Free unincorporated [35S]sulfateis located at V,.
FIG. 4. Sepharose CL-PB chromatography of proteoglycans located in the cell extract fractions. An aliquot of the 4 M guanidine HC1 extract of the cell lavers from cultures maintained in Medium A alone ( A )or in Medium'A supplemented with MSA ( B ) , insulin (C), or fetal calf serum ( D ) was analyzed for relative hydrodynamic size. Columns were under run dissociative conditions.
Proteoglycans from MSA and Insulin-treated Chondrocytes
and cell extract fractions from cultures maintained in Medium A alone or Medium A supplemented with either serum, MSA, or insulin were analyzed by PD-10 gel filtration chromatography before density gradient centrifugation for susceptibility of the 35S-labeled molecules to digestion by chondroitinase ABC. In all cultures, 98 to 99%of the macromolecular radioactivity shifted from the column's excluded volume to the included volume followingincubation with this enzyme, indicating that the [3SS]sulfate was incorporated almost exclusively into chondroitin sulfate. Thus, MSA and insulin stimulation of the chondrosarcoma chondrocytes did not result in the production of an additional type of proteoglycan contain-
ing other types of glycosaminoglycan,such as heparan sulfate, heparin, or keratan sulfate. Proteoglycans labeled with [3H]glucosaminewhichwere isolated and purified froma set of cultures as described under "Experimental Procedures,''were also digested with chondroitinase ABC, and the released disaccharides were analyzed by thin layer chromatography (Table I). In all samples, from both the medium and extract fractions, approximately 95% of thetotal 3H-label wasdigested to disaccharides. Cultures stimulated with MSA and insulin revealed similar distributions of [3H]glucosamine in the different disaccharides for both the medium and cell extract fractions. A change of k5W in the distribution of radioactivity was consideredto be within the errorof this &say system. The undigested 3H-radioactivity remaining localized at theorigin on the thin layer chromatography plate is consistent with the presence of oligosaccharides on the core protein as described by De Luca et al. (12) and by Lohmander et al. (13) and with the presence of residual disaccharides on some of the chondroitin sulfate linkage oligosaccharides (14, 15). The majority of the 3H-radioactivity,
2.0 1 .o
TABLEI Disaccharide analysis of chondroitinase ABC-digested D l proteoglycan This experiment was done three times, twice with [%]sulfate, and once with [3H]GlcNH2-labeledproteoglycan. In all three experiments, 72 to 87% of the total radioactivity was localized at ADi4S after enzyme treatment, no matter what the above culture conditions were.
2 2.0 v) Ln F7
Per cent distributionof I3H]GlcNH;. Fraction Culture Ori- ADi-diS ADi-GS ADi-4S ADiOS" gin
3.5 0 0 9.3 Medium A Medium 86.0 0 0 13.8 Medium A Cell extract 4.3 81.9 3.9 0 1.6 9.2 15%FCSb Medium 84.5 5.8 2.7 7.6 7.9 15% FCS Cell extract 76.7 4.6 0.8 1.4 11.5 Insulin Medium 80.1 0 3.2 13.0 4.0 77.7 Cell extract Insulin 0 0.4 9.8 3.1 86.7 Medium MSA 8.1 9.4 4.5 2.9 MSA Cell extract 72.4 a Although some of the radioactivity detected in this disaccharide spot may originate from hyaluronic acid, this radioactivity is primarily due to chondroitin, since hyaluronic acid would be removed by the equilibrium density gradient centrifugation step, and the modified disaccharide separation technique discussed in the text effectively separates hyaluronic acid disaccharides from chondroitin disaccharides. 'FCS, fetal calf serum.
FIG. 5. Sephadex G-200 chromatography of liberated %-labeled glycosaminoglycans. Glycosaminoglycans from proteoglycans located in medium were liberated by treatment with alkalineborohydride as described in text. Cultures were maintained in Medium A alone ( A )or in Medium A supplemented with MSA ( B ) or insulin (C) .
TABLE I1 Distribution of radioactivity in CsCk density gradients Medium A fraction" Gradient* fraction
D4 D3 3.6 D2 63.9 D l
32.5 0 8.7 0 0 100
37.8 5.6 9.0 48.5 47.7
3.7 0 3.6 0 7.6 0 85.0100
Medium A + insulin Medium
26.8 7.6 11.0 54.6
+ serum ['HISerine
Cell extract fraction" Gradient fraction
Medium A + Medium MSA
Medium A [sS]Sulfate
D4 D3 D2 5.5 D8.9 l
87.7 0 0 01.4 100
83.1 4.4 3.6
0 0 6.1
88.6 3.31.8 1.2
+ insulin [3H]Serme
1.6 8.4 Bottom 93.9 % of total radioactivity in each fraction. Essentially all [35S]sulfate-labeledmacromolecules should localize at thebottom of the gradient in Fractions D l and D2. The finding of 91 to 100%of the total 35S-radioactivityin these two fractions indicates the reproducibility of the separation conditions.
Proteoglycans from MSAInsulin-treated and 70 to 808,migrated with the ADi4S standard, indicating that the newly synthesized chondroitin sulfate chains from all cultures were primarily 4-sulfated. Unsulfated disaccharides, representing about 10% of the total disaccharides, were detected in proteoglycans isolated from all cultures; however, no large increase was found in the cultures maintained in Medium A alone. The much lower incorporation of 35S-labelin the latter culture was therefore not the result of a deficiency in sulfation of the newly synthesized molecules. Disulfated disaccharides were not observed at significant levels in MSAor insulin-treated cultures, indicating that the increased incorporation of 36S-radioactivityin these cultures over MediumA cultures was not the result of synthesis of oversulfated chondroitin sulfate. MSA and insulin, therefore, did not appear to affect the degree or position of sulfation of newly synthesized chondroitin sulfate. Distribution of Radioactivity in CesiumChZoride Gradients-Cultures maintained in Medium A alone or Medium A supplemented with insulin, MSA, or serum were labeled with [35S]sulfateand [3H]serine and the radioactivity measured in equilibrium density-gradient fractions of varying density. As expected, the majority of 35S-labeled macromolecules from all cultures and from both the medium and cell extract fractions were recovered in the fraction of highest buoyant density, the bottom D l fraction (Table 11).This was consistent with other data which indicated that [35S]sulfate was being incorporated specifically into proteoglycans. However, when [3H]serine was used as the radioisotope precursor, less than 10% of the incorporated radioactivity of the principal [3H]serine-labeled fraction, namely the cell extract (1) was recovered in the D l fraction, and this proportion was not altered by hormone treatment. Thus,the observed increase in [3H]serine incorporation by chondrocytes maintained in Medium A with insulin or MSA compared to cultures maintained in Medium A alone (1)appeared to be the result of an overall increase in synthesis of total protein rather thana stimulation of just proteoglycan protein core synthesis. DISCUSSION
19). In MSA- and insulin-treated cultures both glycosaminoglycan chain size and net synthesis of proteoglycan monomer increase significantly. The increases in [3H] serine incorporation into proteoglycans after MSA and insulin treatment, suggests that protein core synthesis is stimulated, accounting for much of the net increase in proteoglycan production. The presence of larger chondroitin sulfate chains after hormone treatment, in this instance, may result from an increase in the intracellular concentrations of the precursor sugar nucleotides involved in glycosaminoglycansynthesis. The radiolabeled proteoglycans produced in insulin- or MSA-stimulated cultures were susceptible to digestion by chondroitinase ABC, indicating that no new type of proteoglycan containing heparan sulfate or heparin glycosaminoglycans was preferentially synthesized in response to the hormones. Preliminary termination of keratan sulfate chains with sialic acid or thelack of ability of the chondrosarcoma chondrocytes to elongate keratan sulfate chains has been noted for these particular chondrocytes when maintained in culture in 15% fetal calf serum (13). The hormone treatments did not stimulate synthesis of keratan sulfate on the core protein. Disaccharide analyses of the chondroitinase digests revealed that sulfation of glycosaminoglycans was not appreciably different in the different culture conditions. The general similarity of the biosynthetic product regardless of the growth conditions employed suggests that MSA and insulin stimulation of the Swarm rat chondrosarcoma chondrocyte results primarily in an increase in the rate of synthesis and secretion of the amountof proteoglycan. Hajek and Solursh ( Z O ) , using chick embryo chondrocytes reported that proteoglycan synthesis may be morepreferentially stimulated than collagen synthesis in response to insulin treatment. However, the fact that insulin- and MSA-stimulated incorporation of radioisotope into total protein indicates that the increase in extracellular proteoglycan accumulation (1) probably reflects a general increase in protein synthesis as well as core protein synthesis. Acknowledgments-We express our gratitudeto Drs. Peter Nissley
Studies presented in the preceding paper demonstrated that and James Kimura for stimulating discussions and for providing insulin and the somatomedin-like polypeptide, MSA, material support for these studies. The technical help of Ms. Thelma Prather was also greatly appreciated. stimulate incorporation of radioactive precursors into macromolecules and the accumulation of an extracellular matrix in REFERENCES cultures of chondrocytes derived from the Swarm rat chon1. Stevens, R. L., Nissley, S. P., Kimura, J. H., Rechler, M. M., drosarcoma. The increase in incorporation of radioactivity Caplan,A. I., and Hascall, V. C. (1981) J. Biol. Chem. 256, was not the result of a mitogenic response by the cells to the 2045-2052 two hormones at the concentrations investigated. Data pre2. Hardingham, T. E., and Muir, H. (1972) Biochzm. Biophys. Acta sented in this report indicate that in MSA- and insulin-stim279,401-405 ulated, as well as in unstimulated Medium A cultures, [35S]- 3. Anderson, B., Hoffman, P., and Meyer, K. (1965) J. Biol. Chem. 240, 156-167 sulfate is indeed being incorporated into proteoglycans which 4. Carbon, D. M. (1968) J.Biol. Chem. 243, 616-626 are similar to those produced in vivo by this tumor (6, 16) 5. Wasteson, A. (1971) J. Chromtogr. 59, 87-97 since they are able to form aggregates. No low molecular 6. Oegema, T. R., Jr., Hascall, V. C., and Dziewiatkowski, D. D. weight fibroblast-like proteoglycan, typical of that produced (1975) J.Bwl. Chem. 250,6151-6159 in small amounts and secreted into the medium of serum7. Kimura, J. H., Hardingham, T. E., Hascall, V. C., and Solursh, M. treated cultures, was observed in insulin- and MSA-stimulated (1979) J. Biol. Chem. 254,2600-2609 8. Hascall, V. C., andSajdera, S. W. (1969) J. Biol. Chem. 244, cultures. Insulin and MSA treatment, on the other hand, 2384-2396 resulted in an increase in the sue of the newly synthesized 9. Saito, H., Yamagata, T.,and Suzuki, S. (1968)J.Biol. Chem. 243. proteoglycan monomers, which wasprimarily, if not entirely, 1536-1542 due to a 25 to 30% increase in the molecular weight of the 10. Kimura, J. H., Hardingham, T. E., and Hascall, V. C . (1980) J. individual chondroitin sulfate chains. Previous studies have B i d . Chem. 255, 7134-7143 shown that theaverage chondroitin sulfate chain lengths vary 11. Coster, L., Carlstedt, I., and Malmstrom, A. (1979) Biochem. J. 183,669-681 inversely with the amount of newly synthesized chondroitin sulfate. Chondroitin sulfate chain size decreases when chon- 12. De Luca, S., Lohmander, L. S., Nilsson, B., and Hascall, V. C., and Caplan, A. I. (1980) J . B i d Chem. 255,6077-6083 drocytes are stimulated tomake more net chondroitin sulfate 13. Lohmander, L.S., De Luca, S., Nilsson, B., Hascall, V. C., Caputo, by the addition of P-xylosides to cultures (17, 18), and chain C. B., Kimura, J. H., and Heineglrd, D. (1980) J . Biol. Chem. length increases significantly as net protein core synthesis 255,6084-6091 decreases after cycloheximide treatment of chondrocytes (18, 14. Hascall, V. C.. Riolo, R. L., Hayward, J., and Reynolds, C. C.
Proteoglycans from MSA Insulin-treated and
(1972) J.Biol. Chem. 247,4521-4528 15. Roden, L., and Smith, R.(1966) J.Biol. Chern. 241, 5949-5954 16. Faltz, L. L., Reddi, A. H., Hascall, G. K., Martin, D., Pita, J. C., and Hascall, V. C . (1979) J.Biol. Chem. 264, 1375-1380 17. Lohmander, L. S., Madsen, K., and Hinek, A. (1979) Arch. Biochern. Biophys. 192,148-157
18. Kato, Y., Kimata, K., Ito, K., Karasawa, K., and Suzuki, S. (1978) J. Biol. Chern. 253,2784-2789 19. Kimura, J. H., Hascall, V. C., and Caputo, C . B. (1980) Fed. Proc. 39, 1638 20. Hajek, A. S., and Solursh, M.(1975) Gen. Comp. Endocrinol. 25, 432-446