709

DIFFERENCES IN THE COLLAGEN TYPES SYNTHESIZED BY LAPINE ARTICULAR CHONDROCYTES IN SPINNER AND MONOLAYER CULTURE DAVID P. NORBY, CHARLES J. M A L E M U D , and LEON S O K O L O F F In five different secondary monolayer cultures of rabbit articular chondrocytes, 72-89% of the collagen synthesized was Type I, as determined by a chain separation and CNBr-cleavage peptide analysis. When sister cells were transferred to spinner bottles after primary monolayer culture growth, 88% of the collagen formed in four separate experiments was Type 11. A reversion to Type I collagen synthesis occurred when the spinnercultured cells were returned to monolayer flasks. The change in the species of collagen depended on the suspension condition and not on the low CaZ+content (0.33 mM) of the spinner medium. These findings parallel the switch to phenotypic glycosaminoglycan synthesis that also occurs when monolayer-cultured cells are transferred to spinner bottles. Whether this phenomenon arises through a genetic control mechanism rather than environmental selection for particular cohorts of cells has not yet been determined.

From the Department of Pathology, State University of New York at Stony Brook, Stony Brook, New York. Supported by grant AM-17258-01 from the National Institute o f Arthritis. Metabolism. and Digestive Diseases, and by the New York Chapter of The Arthritis Foundation. David P. Norby. Ph.D.: Instructor in Pathology; Charles J. Malemud. Ph.D.: Instructor in Pathology; Leon Sokoloff, M.D.: Professor of Pathology. Address reprint requests to David P. Norby, Ph.D., Department of Pathology, Health Sciences Center, SUNY at Stony Brook, Stony Brook, New York 11794. Submitted for publication May 18, 1976 accepted August 12, 1976. Arthritis and Rheumatism, Vol. 20, No. 2 (March 1977)

In monolayer culture, lapine articular chondrocytes have been shown to display two major differences from normal cartilage in the matrix components produced: a) an apparent shift of collagen synthesis from tissue-specific Type I1 to Type I ( I ) ; and b) large changes in the amount and profile of glycosaminoglycan (GAG) species (2). In a recent study, however, it was shown that transfer of monolayered-cultured chondrocytes to suspension conditions resulted in a dramatic return of chondroid expression of GAGS (2). The following study demonstrates an analogous phenomenon with respect to the types of collagen formed in cultures from a common cell inoculum: I n monolayers, the collagen is primarily Type I, whereas in spinner bottles, the bulk is Type I I .

MATERIALS AND METHODS Seven separate cultures of lapine articular chondrocytes were studied. Media and Reagents. These materials were acquired from the following commercial suppliers. Ham’s Nutrient F12, minimal Eagle’s (MEM), minimal Eagle’s for spinner culture (SMEM), Dulbecco-Vogt (DMEM), and a modification of the latter for spinner conditions ( S D M E M ) (see below) were from Grand Island Biological Co. Fetal Calf Serum (FCS) came from Associated Biomedic Sytems. CM-cellulose (CMC, Whatman CM32) and DEAE-cellulose (DEAE. Whatman D E 52) were acquired from H. Reeve Angel. BioGel P-2 was from BioRad Laboratories, and CNBr from Eastman Organic Chemicals. SH-2-Glycine(9.4 Ci/mM ) and Liquifluor came from the New England Nuclear Corp. Pepsin (PM), collagenase (CLS), testicular hyaluronidase (HSE), and trypsin (TRL) were purchased from Worthington Biochemical

710

NORBY ET AL

Corp. L-ascorbic acid (sodium salt) was from Sigma Chemical Co, and /3-aminopropionitrile (BAPN) from Aldrich Chemical co. Preparation of Tissue. Cartilage was taken from the shoulder, hip, and knee joints of 2-3-month-old NZW albino rabbits. The specimens were treated for 3 minutes at room temperature with 15 ml of 0.05% hyaluronidase in Gey's solution to eliminate possible traces of adherent synovial or blood cells. They were then washed twice with Gey's solution. Chondrocytes were dissociated from the cartilage with trypsin and collagenase as described previously (3). I n one instance skin from the ventral thorax was treated in the same way. Techniques of Culture. Chondrocytes ( lo5)were grown in 75 cm2 Falcon flasks in primary culture, with Ham's F-12 medium supplemented with 10% FCS and a 0.1% penicillinstreptomycin mixture (3,4). After 7 days, when the cells became confluent, they were subcultured under two different conditions. One portion was grown as a monolayer-lo5 cells/ml medium in Falcon plastic flasks containing DMEM supplemented as above. The other group of trypsinized cells was transferred to spinner bottles (Bellco) at the same initial inoculum size as above and in the same medium modified by deleting Ca2+ and adding 0.1% Pluronic F-68 (Wyandotte Chemical Co.). The cultures were fed with fresh medium on the third and fifth days. Most of the analyses were carried out on these secondary cultures. In preliminary experiments additions of 0,40, or 100 pg/ml of Na ascorbate were made at each feeding. In the absence of vitamin C, not enough label was incorporated to obtain chromatograms. At the 100 pg/ml level, cytotoxic effects were observed. For this reason the data presented here were recorded at the 40 pg/ml level. Labeling was initiated at the fifth day feeding by

changing the medium to MEM or SMEM containing sH-2glycine (0.5 pCi/ml). After incubating for 20 hours, the medium and the cells were harvested and analyzed separately. These data have been pooled in the tables, but differences in the partition of the two fractions are noted where relevant in the following discussions. Preparation of Carrier Collagen. Minced young rabbit dermis was sequentially extracted with 0.2 M NaCI, 0.05 M Tris, pH 7.5; 1.O M NaCI, 0.05 M Tris, pH 7.5; and 4% acetic acid (HAc). Collagen in the insoluble residue was rendered soluble by using pepsin in 4% HAc. After inactivation of the pepsin at pH 8 (5), the collagen was purified by precipitation procedures (6) and by DEAE-cellulose chromatography (7). N o attempt was made to separate Type I collagen from the Type 111 collagen that was probably present (8,9) (Figure I ) . Preparation of Samples for Chromatography. Immediately after the cultures were harvested, 20 mg of carrier collagen i n 10 ml of 4% HAc were added to each sample (cells or medium). Glacial HAc was then added to make the total sample 4% HAc. The samples were then either processed directly or frozen for processing at a later date. To ensure that all collagen was present in the soluble monomeric form and that any procollagen was converted to collagen, limited proteolysis of nonhelical regions was carried out with pepsin (5). Inactivation of pepsin was carried out in 0.2 M NaCI, 0.05 M Tris, pH 8.0. The sample was applied to a DEAE-cellulose column equilibrated with the same buffer to remove G A G S that might interfere with C M C chromatography (7). Collagen i n the unadsorbed fraction of the eluant was precipitated by addition of dry NaCI to 5%(w/v), followed by glacial HAc to 4%(v/v). The precipitate was redissolved in 0.04 M {Na') acetate, 1.0 M urea, pH 4.8, and dialyzed against the same buffer.

10

08

06

g

O4

: N

02

$!

3

00 12

8 3

08

T-

06

34

'3 2

00 0

50

100

150

200

255

300

EFFLUENT VOLUME (ml.) Fig 1. C M C elurion parterns of spinner and nionolayer culture media labeled with 3H-glycine and chromatographed wirh rabbit skin carrier collagen. n l ( l l 1 ) in carrier is injerred from siudies of orhers with pepsin-solubilized collagen in human skin (8.9).

71 1

CHONDROCYTE COLLAGEN PRODUCTION

MONOLAYER a1 PEAK CNBr PEPTIDES

-

c

..m

5w

300

*

L

78

% N

96

T

I

,

.

-T-

: .

JS

-

9z

-

-

--

I

n

f

I

b .

800

,--..___ ,...._.,' . --7

100

P200

1

300

7

-

r

~

400

400

1

5M

600

7cQ

800

900

EFFLUENT VOLUME (ml)

Fig 2. CMC' elurion parterns 0fC"Brpeptides deriuedfrom a I peak (above) and 0 2 peak (below)from monolayers chromatographed nirh cvrresponding carrier rabbit skin a chain peptides. Presumed identities of peaks were assigned by comparison ofpublished elution patterns 01 CNBr digesrs oj Tjlpe I collagen from other species (40-46). Nomenclarure is that of Fietzek and Kuhn (391.

Alpha Chain Separation. C M C chromatography was carried out at 40°C with a 2 X 8 cm column (10). The total gradient volume was 400 ml. Five-milliliter fractions were ) ~ (Tables 1 and 2 ) collected. The proportion of ( ~ 1 molecules mas calculated from the excess of a 1 over an N 1 : a 2 ratio of 2. CNBr Peptide Analysis. To facilitate recognition of marker peptides in carrier collagen, an additional 50 mg of corresponding N chains were added t o lyophilized desalted peaks from C M C chromatography. CNBr treatment was performed as described by Miller et a l ( 1 I ) , except that the samples were dried by flash evaporation at the end of the digestion. C M C chromatography was carried out (12) by using 5 mM ( N a - ) citrate, pH 3.6, as the starting buffer and 5 mM {Na') citrate, pH 3.6, + 0.18 M NaCl as the limiting buffer. The total gradient volume was 1 liter. Five-milliliter fractions were collected. CNBr cleavage peptides of collagen. prepared from rabbit rib cartilage (free of perichondrium) by the methods described above, were also prepared as reference material for t u I ( I I ) radioactivity peaks. Measurement of Radioactivity. A 0.7-1111 aliquot of each fraction was mixed with 10 ml of a solution of 2% (v/v) Liquifluor in a I : I mixture of toluene and 2-methoxyethanol. Samples were counted for radioactivity and corrected for background. Additional Experiments. a ) T o test for reversibility of chondroid expression. cells that had been suspended in a spinner bottle in SDMEM for 5 days were returned untrypsinized to a stationary flask containing DMEM. b) T o determine whether the chondroid expression observed in spinner bottles was the result of the suspension condition rather than the low Ca" content of S D M E M , chondrocytes were grown as a monolayer in stationary flasks containing the same spinner medium. c ) The effect of deletion of FCS from the culture medium during the labeling period was studied to compare results with other studies published previously. d) The effect of BAPN (100 &mI) present only during the labeling period was studied for similar reasons. e) The species of collagen synthesized by a culture of rabbit skin fibrocytes was studied as a control of the collagen chain characterization.

RESULTS Monolayer Culture. Approximately 80% of the radioactivity appeared in the cell pellet and 20% in the medium. These values have been pooled in the tables and following presentation. A distinct a2 peak was found in each of five separate monolayer cultures of articular chondrocytes (Figure 1 ). I n chromatograms from some of the cultures, the a 1 : a 2 ratio was nearly 2. These al peaks were pooled and digested with CNBr. The C M C elution pattern revealed that the preponderant al chain was al(1) (Figure 2 ) . However the a1 :a2 ratio was variable and gave an average value in secondary culture of 3.1 0.32 (Table I ) . The small amount of (Y 1 chains in excess of the 2 : I : : aI : c u 2 of Type I collagen was not characterized in these other cultures. Therefore it cannot be stated whether this excess material was Type 11 collagen or (a1 (13). Spinner Culture. In contrast to the monolayer cultures, the bulk (80%)of the radioactivity appeared in the medium rather than the cell fraction. No radioactivity peak was detectable in the a2 region in any of the 10 spinner bottles derived from cultures of 4 separate rabbits (Figure 1 ). Counts above background appearing in the a2 region were used to calculate a 1 : a 2 ratios. These ranged from 14 to 30 (Table 1 ) . CNBr peptide radioactivity peaks did not cochromatograph with al(1) peaks from rabbit skin carrier (Figure 3). The configuration of the peaks conformed closely to that of Type I1 collagen derived from rabbit rib cartilage (Figure 3). The slight shift in the elution position occurs commonly when, as in the present instance, the separations were not carried out simul-

+

712

NORBY ET AL

Table 1. Collagen Species Synrhesized by Chondrocytes in Monolayer and Spinner Culture (Medium -+ Cell Pellet)

Monolayer

3.3 f 0.60 (292) 3.1 f 0.32 (13.5) 6.9

lo 2" 3"$

69.4

30.6

73.2.t

26.8

38.0

62.0

12.9

87. IQ

(1.1)

2"

Spinner

22.2 f 3.74 (7.4)

* Mean

f standard error. The first figure in parentheses is the total number of flasks analyzed; the second is the number of separate cultures from which the preceding was derived. t Proven Type I by CNBr peptide analysis (Figure 2). $ Returned to monolayer conditions following spinner culture. 5 Proven Type I I by CNBr peptide analysis (Figure 3).

taneously, and minor technical parameters (e.g., jacket temperature or limiting buffer concentration) may vary. Reversibility of Chondroid Expression. When spinner-cultured cells were returned to stationary flasks, a monolayer outgrowth began within 2 days. Within 5 days the cells became confluent, although several heaped-up clusters of cells persisted. The a 1 : a 2 ratio of 6.9: 1 indicated a substantial reversion to the monolayer pattern observed originally (Table 1 ). Effect of Low Caz+ Concentration. The Ca2+concentrations of spinner and monolayer media supplemented with 10% FCS. measured with an atomic ab-

lo

08 06

sorption spectrometer (Perkin-Elmer, Model 403), were 0.33 and 2.45 mM respectively. The a l : a 2 ratios in corresponding monolayer cultures were comparable to each other (Table 2). Thus the low Ca2+content of the spinner medium did not account for the change in collagen species occurring in spinner culture. The chondrocytes grew as well in the low as in the high Ca2+containing media. Miscellaneous Findings. BAPN had no effect on the amount or type of collagen synthesized, except that there was a shift of radioactivity from the cell-associated fraction to the medium in the monolayer cultures. Deprivation of serum during the labeling period reduced the incorporation of 3H-glycine 30% below control values in the monolayer culture. This reduction was associated with a distinct shriveling and some fragmentation of the cells, There was, however, no change in the a 1 : a 2 ratio (Table 2). However, in the spinner culture deprived of serum, the reduction in radioactivity was accompanied by a decrease in the a1 : a 2 ratio to 5.1 (Table 2). The cul : a 2 ratio of the fibroblast cultures was 2.3: 1.

DISCUSSION Although reservations about the differentiated state of cultured mammalian chondrocytes have been expressed ( l4), the present experiments establish that under certain conditions of culture, Type I1 collagen is indeed synthesized by lapine articular chondrocytes.

7 SPINNER-a1PEAK-CNBr PEPTIDES -

P

0

I00

200

XI0

6M)

EFFUEM a U M E (rnl.)

Fig 3. CMC elution parrern of CNBr peptides derivedfroni a1 peak of spinner medium chromatographed with a1 chain peptides of rabbit skin collagen (above). The CNBr peptide pattern of the a1 peak of rabbir cartilage collagen is below (see text/. Identities of peaks were inferred by comparison with published elution patterns of CNBr digests of cartilage collagen from other mammalian species (47-491.

713

CHONDROCYTE COLLAGEN PRODUCTION

Table 2. ESject of Serum, Calcium Concenrration. and BAPN on Collagen Species Synthesized by Cultured Chondrocyles (Medium + Cell Pellet) Experiment No.

I 2

3

Culture Monolayer Monolayer Monolayer Monolayer Spinner Spinner Monolayer M onolayer Spinner Spinner

Serum

+ + +

-

+ + + + +

-

Ca(mM)

BAPN

2.45 0.33 2.45 2.13 0.33 0.01 2.45 2.45 0.33 0.33

(Yl:a2* 2.9 3.2 2.5 2.4 20.2 5.8 2.4 2.4 16.3 14.6

(aI)2LY2(%)

bI)J(%)

.

23 28 14 13 86 55 II 11 83 81

(2.7,3.1) (2.9,3.5)

(3.8.6.4.7.2) (2. I , 3.0.2. I ) (2.3.2.5) (13.6, 19.0) (14.1, 15.1)

77 72 86 87 14 45 89 89 17 19

* Mean. The figures in parentheses are the data of the individual flasks analyzed.

This phenomenon parallels the switch to phenotypic GAG synthesis under the same spinner conditions (2). It is, however, more directly related to a genetic control mechanism than the latter, in which the proportions rather than the types of molecules synthesized are altered. The nature of the relationship between the two phenotypic expressions is not known, but it is noted that in monolayer cultures there is a higher proportion of hyaluronate than in the spinner cultures (2). Linsenmayer (15) has reported that in the early stages of chondrogenesis of chick embryo limb buds, high concentrations of hyaluronate are associated with Type I rather than Type I 1 collagen synthesis. Three reports indicate that hyaluronate interferes with chondrogenic expression of cultured embryonic chondrocytes (16-18). The failure of prolonged monolayer cultures of mammalian and embryonic chick chondrocytes to produce phenotypic GAG and collagen has been noted by several investigators (1,2,13,19,20). Three general explanations have been proposed: a) the cells are overgrown by extraneous fibroblast-like cells (21,22); b ) the chondrocytes undergo dedifferentiation to fibroblast-like cells (23); and c) the cells undergo a process of senescence characterized by a loss of replicating activity and the formation of an unusual trimeric form of collagen, (I)}, (24). The first of these possibilities cannot be eliminated because articular cartilage is always heterogeneous. Fibrocartilaginous regions constitute a variable proportion of articular cartilage in all mammalian joints, particularly at the margins (22). They form the predominant constituent of the articular cartilage in avian species (25,26). Type I collagen is the predominant species of the articular semilunar fibrocartilage in the pig (27). It is thus not surprising that two species of

collagen have been identified chemically in avian articular cartilage (28). Even after careful dissection of rabbit articular cartilage, approximately 2% of the tissue contains some atypical matrix staining with fast green and reduced amounts of safranin 0 (22). These regions are transitional to fibrocartilage. The possibility that even this small cohort of atypical hyaline chondrocytes might become a sizeable part of the ultimate cell population is shown by the following oversimplified estimation. It neglects other important biologic considerations, such as possible cell type interactions (21), and is based only on the known population doubling times of monolayercultured fibroblasts and chondrocytes of rabbits (29) using the following formula.

where N, and N, are the numbers of fibroblasts and chondrocytes, t is the time past the start of culture, No is the number of cells at start of culture, and tdf and tdc are the doubling times of fibroblasts and chondrocytes (10 and 16 hours respectively). In 5 days of culture the fibroblastic cells would constitute 36% of the total. The possibility of fibroblastic dedifferentiation is unlikely unless qualified. Both the molecular species of collagen observed here and the GAG profile reported previously (2) are quite different from that of cultured dermal fibroblasts. There are greater similarities to fibrocartilage, which also has a Type I rather than a Type I 1 collagen (27) as well as some hyaluronate (30) and a sizeable dermatan sulfate (31) content. The suggestion that the phenomenon is akin to senescence is unlikely for two principal reasons: a) Unlike the cells in the studies of Mayne et al(24), there was no decay in the proliferative capability of the chon-

NORBY ET AL

7 14

drocytes. Indeed these cells grew aggressively, and the same phenomenon was observed in primary as well as in secondary and tertiary cultures. b) The same cells, when transferred to spinner bottles, again expressed their phenotype through Type I 1 collagen synthesis. In recent papers preliminary evidence was presented that neonatal or fetal chondrocytes, grown in primary culture, synthesize predominantly Type I1 collagen (32). The cells formed clusters that did not grow out in monolayer form. Several other investigators have similarly noted that in chick embryo chondrocyte cultures, the cells remain rounded and aggregated during the early days of culture and resemble cartilage. This stage is followed by a spreading out and ameboid wandering of the cells, which is accompanied by a loss of sulfated GAG synthesis (14,20,23). In early cultures (24,33) Type I1 collagen is synthesized, but Type I collagen is predominant in the later monolayer growth (24). By immunofluorescent means. Miiller et a1 (19) found that the early rounded cells stained for Type I1 collagen, but the later growing monolayer cultures were more readily stained by antiserum to Type I collagen. Thus the chick embryo and neonatal mammalian cells, when grown as clones or in high density cultures before much spreading out and monolayer proliferation occur, express both tissue-characteristic collagen and GAGS. Following extensive proliferation as monolayers, the change to the “fibrocartilaginous” pattern emerges. This interpretation is not ruled out by the possibility that the small amount of synthesized in monolayer culture may be ofthe al(1) rather than al(I1) species. A proportion of cells may have become “senescent.” If the trimer this finding would be conshould instead be {al(ll))3, sistent with the occurrence of a proportion of Type I 1 collagen in another sort of fibrocartilage: anulus fibrosus (34). The basis for the reemergence of the chondroid expression when such monolayer cultures are transferred to spinner bottles remains unsettled. Two possibilities entertained previously are possible: either there is a selection under suspension conditions for a cohort of hyaline chondrocytes, or there is an environmental regulation of the synthetic activity of a uniform population of cells. The latter possibility is supported by observations of a switch in the type of collagen produced in cultures derived from a single chick embryo chondroblast (24). Speculation about the cellular mechanisms in the latter event must include alterations in the cell membrane attending the attachment to the surface of the culture chamber. The fact that the low Ca concentration was n o t

the basis for the chondrogenic expression in the spinner bottles is well established by the findings. Changes in the ionic environment of cultured chondrocytes have been found previously to affect GAG formation and degradation (35,36), but not the synthesis of collagen (37). Aside from the theoretical mechanisms that may be involved, the present findings add weight to the use of cell culture methods in the study of phenotypic products of mammalian articular chondrocytes.

ADDENDUM As this manuscript is submitted for publication, Cheung and coworkers (38) report the “derepression” of nonphenotypic collagen synthesis in monolayer cultures of lapine articular chondrocytes. The type of a1 chain synthesized was not characterized by CNBr cleavage peptide analysis. The level of ascorbate employed at the time of labeling (100 kg/ml) was cytotoxic in the present authors’ experiments.

ACKNOWLEDGMENT The authors are indebted t o Dr. Edward J . Miller for critical review of the manuscript.

REFERENCES I . Layman DL, Sokoloff L, Miller EJ: Collagen synthesis by articular chondrocytes in monolayer culture. Exp Cell Res 73: 107- I 12, 1972 2. Srivastava VML, Malemud CJ, Sokoloff L: Chondroid expression by lapine articular chondrocytes in spinner culture following monolayer growth. Connect Tissue Res 2: 127- 136. 1974 3. Green WT Jr: Behavior of articular chondrocytes in cell culture. Clin Orthop 75248-260, 1971 4. Sokoloff L. Malemud CJ, Green WT Jr: Sulfate incorporation by articular chondrocytes in monolayer culture. Arthritis Rheum 13:l 18-124. 1970 5. Miller EJ: Structural studies on cartilage collagen employing limited cleavage and solubilization with pepsin. Biochemistry I1:4903-4909, 1972 6. Deshmukh K, Nimni ME: Soluble collagen with high aldehyde content extracted from insoluble collagen with rnercaptoethylamine. Biochim Biophys Acta 154:258-260, 1968

7. Miller EJ: Isolation and characterization of the cyanogen bromide peptides from the a1 (11) chain of chick cartilage collagen. Biochemistry 10:3030-3035, 1971 8. Chung E, Miller EJ: Collagen polymorphism: characterization of molecules with the chain composition [a1 in human tissues. Science 183:1200-1201, 1974

CHONDROCYTE COLLAGEN PRODUCTION

9. Chung E, Keele EM, Miller EJ: Isolation and characterization of the cyanogen bromide peptides from the a1 ( I l l ) chain of human collagen. Biochemistry 13:34593464, 1974 10. Miller EJ, Woodall DL, Vail MS: Biosynthesis of cartilage collagen: use of pulse labeling to order the cyanogen bromide peptides in the a l (11) chain. J Biol Chem 248: 1 666-167 I , 1973 1 I . Miller EJ, Epstein EH, Piez KA: Identification of three genetically distinct collagens by cyanogen bromide cleavage of insoluble human skin and cartilage collagen. Biochem Biophys Res Commun 42:1024-1029, 1971 12. Miller EJ: Isolation and characterization of a collagen from chick cartilage containing three identical a chains. Biochemistry 10:1652-l659, 1971 13. Mayne R, Vail MS, Miller EJ: Analysis of changes in collagen biosynthesis that occur when chick chondrocytes are grown i n 5-bromo-2'-deoxyuridine. Proc Natl Acad Sci USA 72:4511-4515, 1975 14. Levitt D, Dorfman A: Concepts and mechanisms of cartilage differentiation. Curr Top Dev Biol 8:103-149, 1974 15. Linsenmayer TF: Temporal and spatial transitions in collagen types during embryonic chick limb development. 11. Comparison of the embryonic cartilage collagen molecule with that from adult cartilage. Dev Biol40:372-377, 1974 16. Wiebkin O W , Muir H: The inhibition of sulphate incorporation in isolated adult chondrocytes by hyaluronic acid. FEBS Lett 37:42-46, 1973 17. Toole BP. Jackson G, Gross J: Hyaluronate in morphogenesis. Inhibition of chondrogenesis in vitro. Proc Natl Acad Sci USA 69:1384-1386, 1972 18. Solursh M , Vaerewyck SA, Reiter RS: Depression by hyaluronic acid of glycosaminoglycan synthesis by cultured chick embryo chondrocytes. Dev Biol 41:233-244, 1974 19. Miiller P, Lernmen C, G a y S, et al.: Biosynthesis of collagen by chondrocytes in oifro, Extracellular Matrix Influences on Gene Expression. Edited by H C Slavkin, R C Greulich. New York, Academic Press, 1975, pp 293-302 20. Shulman HJ, Meyer K: Protein-polysaccharide of chicken cartilage and chondrocyte cell cultures. Biochem J I20:689-697. 1970 21. Bryan JC: Studies on clonal cartilage strains. 1. Effect of contaminant non-cartilage cells. Exp Cell Res 52:3 19-326, I968 22. Hough AJ, Sokoloff L: Tissue sampling as a potential source of error in experimental studies of cartilage. Connect Tissue Res 3:27-3 I, 1975 23. Schiltz J R , Mayne R, Holtzer H: The synthesis of collagen and glycosaminoglycans by dedifferentiated chondroblasts in culture. Differentiation 1:97-108, 1973 24. Mayne R , Vail MS, Mayne PM, et al: Changes in types of collagen synthesized as clones of chick chondrocytes grow and eventually lose division capacity. Proc Natl Acad Sci USA 73:1674-1678, 1976

715

25. Van der Stricht 0: Recherches sur le cartilage articulaire des oiseaux. (Researches on the articular cartilage of birds.) Arch Biol (Paris) 101-41, 1890 26. Lubosch W: Bau und Enstehung der Wirbeltiergelenke. Eine Morphologiskhe und Histogenetische Untersuchung. (Structure and Origin of Vertebrate Joints. A Morphological and Histogenetic Investigation.) Jena, Gustav Fischer, 1910 27. Eyre D, Muir H: The distribution of different molecular species of collagens in fibrous, elastic and hyaline cartilages of the pig. Biochem J 151:595-602, 1975 28. Seyer JM, Brickley D M , Glimcher MJ: The identification of two types of collagen in the articular cartilage of postnatal chickens. Calcif Tissue Res 17:43-55, 1974 29. Malemud CJ, Sokoloff L: Some biological characteristics of a pituitary growth factor ( C G F ) for cultured articular chondrocytes. J Cell Physiol 84: I7 1-1 80, 1974 30. Solheim K : The glycosaminoglycans of human semilunar cartilage. J Olso City Hosp 15:127-132, 1965 31. Habuchi H , Yamagata T, Iwata H, et al: The occurrence of a wide variety of dermatan sulfate-chondroitin sulfate copolymers in fibrous cartilage. J Biol Chem 248:60196028, 1973 32. Schindler FH, Ose MA, Solursh M: Synthesis of cartilage collagen by rabbit and human chondrocytes in primary cell culture. In Vitro 12:44-47, 1976 33. Handley CJ, Bateman JF, Oakes BW, et al: Characterization of the collagen synthesized by cultured cartilage cells. Biochim Biophys Acta 386:444-450, 1975 34. Eyre DR, Muir H: Collagen polymorphism: two molecular species in pig intervertebral disc. FEBS Lett 42: 192- 196, 1974 35. Lash JW, Rosene K , Minor RR, et al: Environmental enhancement of in vitro chondrogenesis. 111. The influence of external potassium ions and chondrogenic differentiation. Dev Biol 35:370-375, 1973 36. Shulrnan HJ, Opler A: The stimulatory effect of calcium on the synthesis of cartilage proteoglycan. Biochem Biophys Res Commun 59:914-919, 1974 37. Daniel JC, Kosher RA, Hamos JE, et al: Influence of external potassium on the synthesis and deposition of matrix components by chondrocytes in vitro. J Cell Biol 63:843-854, 1974 38. Cheung HS, Harvey W, Benya PD, et al: New collagen markers of "derepression" synthesized by rabbit articular chondrocytes in culture. Biochem Biophys Res Commun 68: 1371-1 378, 1976 39. Fietzek PP, Kuhn K : The primary structure of collagen, International Review of Connective Tissue Research. Vol. 7. Edited by DA Hall, DS Jackson. New York. Academic Press, 1976, pp 1-60 40. Butler WT, Piez KA: Isolation and characterization of the cyanogen bromide peptides from the a 1 chain of rat skin collagen. Biochemistry 6:3771-3780, 1967 41. Fietzek PP, Piez KA: Isolation and characterization of the

716

42.

43.

44.

45.

cyanogen bromide peptides from the a 2 chain of rat skin collagen. Biochemistry 8:2 129-21 33, I969 Rauterberg J, Kuhn K: Acid soluble calf skin collagen: characterization of the peptides obtained by cyanogen bromide cleavage of its a1 chain. Eur J Biochem 191398-407, 1971 Click EM, Bornstein P: Isolation and characterization of the cyanogen bromide peptides from the a I and a 2 chains of human skin collagen. Biochemistry 9:4699-4706, 1970 Clark CC, Bornstein P: Cyanogen bromide cleavage of guinea pig skin collagen. Isolation and characterization of peptides from the a l and a 2 chains. Biochemistry ll:1468-1474, 1972 Heinrich W, Lange PM, Stirtz T, et al: Isolation and characterization of the large cyanogen bromide peptides

NORBY ET AL

from the a I and a 2 chains of pig skin collagen. FEBS Lett 16:63-67. 1971 46. Becker U, Timpl R: Cyanogen bromide peptides of the rabbit collagen a 1 chain. FEBS Lett 27:85-88, 1972 47. Miller EJ, Lunde LG: Isolation and characterization of the cyanogen bromide peptides from the al (11) chain of bovine and human cartilage collagen. Biochemistry 1213153-3159, 1973 48. Eyre DR, Muir H : Characterization of the major CNBrderived peptides of porcine Type I 1 collagen. Connect Tissue Res 3:165-170, 1975 49. Smith BD, Martin GR, Miller EJ, et al: Nature of the collagen synthesized by a transplanted chondrosarcoma. Arch Biochem Biophys 166:181-186, 1975