Biosynthetic Expression of Type X Collagen in Embryonic Chick ...

VOl. 261. No,, 6, Issue of February 25, pp. 2897-2904 1986 Printed in C.S.A.

THEJOURNAL OF BIOLOGICAL CHEMISTRY Q 1986 by The American Society of Biological Chemists, Inc.

Biosynthetic Expression of Type X Collagen in Embryonic Chick Sternum Cartilage duringDevelopment* (Received for publication, July 3, 1985)

Anthony M. ReginatoS, James W. Lash& and Sergio A. JimenezSll From the Departments of $Medicine and §Anatomy of the University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania 191 04

To investigate the temporal and topographic changes in the expression of various collagens during theprocess of endochondral bone formation, qualitative and quantitative analysis of the collagens synthesized by organ cultures from the separated presumptive calcification and permanent cartilaginous regions of embryonic chick sternum at various stages of development was performed. Special emphasis was placed on the study of Type X collagen, a recently described species that may play a role in tissue calcification. We found that Type X collagen is biosynthesized exclusively by cartilage from the zone of presumptive calcification and that its biosynthetic expression is acquired at stage 43 (day 17) of sternal development. Quantitative analysis indicated that Type X was the biosynthetic product which showedthe most dramatic changes increasing markedly with increased sternal X collagen could be detected at age. While no Type stage 40, it represented about 12%of the total collagen synthesized at stage 43, further increasing to 45% at stage 46 of sternal development. Theincrease in Type X collagen in the presumptive calcification region was accompanied by a relative decrease in the proportion of la, 2a,3a, d(II),and Type IX collagens. In contrast, the permanent hyaline cartilage did not displaydetectable synthesis of Type X collagen at any sternal age. The strict topographic distribution and the temporal expression of Type X collagen biosynthesis coincident with the development of sternal calcification, confirm the notion that this collagen may play an important role in the extracellular matrixremodeling associated with the initiation and progression of tissue calcification.

Normal development of the skeletal system in vertebrates requires precise temporal and spatial coordination of growth, and reorganization of the various structural elements, particularly cartilage and bone. These changes involve simultaneous de nouo synthesis and deposition of matrix and removal of the pre-existing matrix. During long bone development, the matrix which is initially comprised of hyaline cartilage undergoes appositional growth to form a dumbbell-shaped mass surrounded by perichondrium with a central portion destined to become the bone diaphysis and the two distal portions, or

* This work was supported in partby National Institutes of Health Grant AM-15091. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solelyto indicate this fact. 1To whom correspondence and requests for reprints should be addressed at: Rm 570 Maloney Bldg., Hospital of the University of Pennsylvania, 3600 Spruce St., Philadelphia, PA 19104.

epiphyses, which will form the future articular regions. The cartilage matrix participates inendochondral bone formation by becoming calcified. During this process, the chondrocytes become hypertrophic and subsequently degenerate leaving the interconnecting spaces for osteogenesis (1). In addition to endochondral ossification, intramembraneous ossification is involved in the skeletal development of mammals and birds. This process involves bone formation within fibrocartilage membranes of condensed primitive mesenchymal tissue. Mesenchymal cells differentiate intoosteoblasts which begin synthesis and secretion of osteoid matrix at the centers of ossification. The deposited osteoid rapidly becomes mineralized. It has been well documented that the hyaline cartilage matrix involved in endochondral bone formation exhibits a high degree of molecular heterogeneity particularly regarding its collagenous components (2). Although it is generally accepted that skeletal development involves primarily changes in Type I1 the principal cartilage collagen, and Type I the main bone collagen (3-5), the role of the more recently identified minor cartilage collagens designated as la, 2a, 3a (6, 7), Type IX (8-13), and Type X (14-22) has not been fully examined. In thepresent paper, we have studied the expression of the various collagens in embryonic chick cartilage during the developmental process leading to its calcification. We found that organ cultures of embryonic chick sternal cartilage exhibit profound spatial andtemporal changes in theexpression of Type X collagen biosynthesis during differentiation. The results demonstrated that Type X collagen biosynthesis is exclusively confined to cartilage from the presumptive calcification region and that its expression is acquired only at stage 43 of sternal development. A marked acceleration of Type X biosynthesis occurred at stage 46, corresponding with the onset of ossification (23, 24). EXPERIMENTALPROCEDURES

Biosynthetic Studies-Sterna from White Leghorn chick embryos corresponding to developmental stages 40 (14 day), 43 (17 day), and 46 (20 day) as described by Hamburger and Hamilton (24)were removed and carefully dissected from surrounding perichondrium. The caudal two-thirds comprising the permanent hyaline and the cephalic one-third comprising the presumptive calcification cartilages, respectively, were separated with a scalpel. Microscopic examination confirmed that the cephalic portion obtained with this procedure contained almost entirely hypertrophic chondrocytes. The tissues were incubated in Eagle’s minimal essential medium containing ascorbic acid (100 hg/ml), 0-aminopropionitrile (2.5 mM),and 1% streptomycin. Cultures were labeled for 48 h with 1 pCi/ml [“CC] proline (275 mCi/mmol, Amersham Corp.). After incubation, the media were removedand frozen after a solution of protease inhibitors was added to give a final concentration of 50 mM disodium EDTA, 0.2 m M phenylmethylsulfonyl fluoride, 5.0 mM N-ethylmaleimide, and 1.0 mM p-aminobenzamidine/HCl. Collagen Solubilization-After labeling, the minced cartilage was

2897

2898

Type X Collagen Biosynthesis in Chick Embryo Development

homogenized with a Polytron homogenizer for 3 min ina buffer containing 1.0 M NaCl, 50 mM Tris-HC1, pH 7.4, at 4 "C and the concentrations of protease inhibitors listed above. The homogenized tissues were extracted with the same buffer for 72 h at 4 "C. The solubilized material was removed by centrifugation and insoluble material was sequentially extracted for 48 h in a manner previously described by Jimenez and Bashey (25). Briefly, this involved sequential extraction with first 0.5 M acetic acid, then with 0.5 M acetic acid containing 1mg/ml pepsin (Sigma), followed by two extractions with a solution containing 0.15 M NaCI, 50 mM Tris-HC1, 25 mM dithiothreitol (DTT'), pH 7.4, at 4 "C and finally a repeated extraction with the 0.5 M acetic acid/l mg/ml pepsin solution. This method resulted in complete solubilization of the tissues. Following centrifugation, the supernatants were removed and after extensive dialysis against 0.15 M NaCI, 50 mM Tris-HC1, pH 7.4, buffer, aliquots from each of the extracts were taken for determination of their [14C] hydroxyproline content by the method of Juva and Prockop (26). Proteolytic Digestion-Lyophilized samples used for collagenase digestion were resuspended in a buffer containing 0.15 M NaCI, 5.0 mM CaCl,, 50 mM Tris-HC1, 10 mM N-ethylmaleimide, 0.2 mM phenylmethylsulfonyl fluoride, pH 7.4, a t 37 "C. Purified bacterial collagenase (Form 111, Advance Biofactures) was added (150 units/ ml) and the samples were digested for 6 h at 37 "C. The reaction was terminated by addition of EDTA to a final concentration of 10 mM and cooling a t 4 "C. For digestion with pepsin, lyophilized samples were suspended in 0.5 M acetic acid containing 300 pg/ml pepsin and incubated for 6 h a t 15 "C. The reaction was terminated by addition of 0.1 ml of 6 M NaOH and the digested samples were then dialyzed at 4 "C against 0.1 M ammonium bicarbonate and lyophilized. Gel Electrophoresis-The labeled proteins were examined by SDSpolyacrylamide slab gel electrophoresis on 4.0-10% acrylamide exponential gradient gels as described previously (21). Lyophilized samples were dissolved in sample buffer (20 mM Tris borate, 2.0 M urea, 2.0% SDS, 1.0% glycerol, pH 8.5) and heated for 1 h at 60 "C. Reduced samples were treated with 1.0% (v/v) 2-mercaptoethanol. Samples were electrophoresed for 3.5 h at 250 V. After electrophoresis, the gelswere processed for fluorography with EN3HANCE (New England Nuclear) and exposed to X-Omat AR film (Eastman Kodak). Two-dimensional cyanogen bromide peptide mapping was performed by a modification of the method of Barsh and Byers (27) as previously described (21). Quantitative Analysis of the Relative Proportion of Various Collagen Types Synthesized in Culture by the Developing Sternum-In order to examine the relative proportions of the various collagen types synthesized during the organ cultures, fluorographs from neutral salt and DTT extracts from the permanent hyaline cartilage and presumptive calcification regions at the three stages of development were scanned in a lineardrive densitometer at 540 nm and theareas under each collagen peak were quantified employing a planimeter. The densitometric analysis was carried outin slightly under-exposed fluorographs to assure linearitybetween the calculated areas and the absolute amount of radioactivity in each band. The acetic acid and pepsin extracts were not analyzed because they represented less than 10% of the total collagen extracted. All fluorographs examined were from samples electrophoresed after limited pepsin digestion under reducing conditions to allow cleavage to noncollagenous domains and dissociation of the disulfide-bonded species of Type IX collagen. To compare the rates of Type I1 and Type X collagen production per cell between chondrocytes from the permanent cartilaginous and presumptive calcification regions, parallel experiments were performed to determine the number of chondrocytes present in each of the separate regions at stages 40 and 46 of sternal development as described by Jimenez et a2. (28). At stage 40 of sternal development, the number of chondrocytes/sternum from the permanent hyaline cartilage and thepresumptive calcification region was similar (65.8 X IO4 and 63.4 X lo4, respectively). At stage 46, however, the number of cells obtained from the permanent hyaline cartilage was substantially greater than that obtained from the presumptive calcification region (150.7 X lo4 and 76.5 X lo4 cells/sternum, respectively). Histological Preparations-Sterna were removed from chick embryo at stages 40, 43, and 46 of development and were prepared for differential histochemical staining of cartilage and bone with alizarin red, alcian blue, and silver nitrate stains asdescribed (29, 30). The abbreviations used are: DTT, dithiothreitol; SDS, sodium dodecyl sulfate.

RESULTS

Protein and Collagen Biosynthesis by Permanent Hyaline and Presumptive CalcificationCartilage at Three Stages of Development-Determination of total [14C]proline incorporation and [14C]hydroxyprolinesynthesis by cartilage from the separate sternalregions at the three stages of development showed that thepresumptive calcification region exhibited 2fold greater [14C]proline incorporation and about %fold greater [14C]hydroxyprolinesynthesis at the threesternal ages. Similarly, the degree of proline hydroxylation in the newly synthesized proteins was higher in the presumptive calcification cartilage at the three stages of development (44.5, 46.7, and 45.7% versus 36.9, 40.3, and 40.6%, respectively). Extraction of Newly Synthesized Collagens-Table I shows the patterns of solubilization of the newly synthesized collagens from the separated regions of sternal cartilage at the three stages of development. The amount of collagen present in each of the sequential extractions was combined into three different populations corresponding to readily extractable (NaCl and acetic acid extracts), pepsin-'extractable, and pepsin-resistant (DTT extracts) molecules. We found that treatment of the organ cultures with @-aminopropionitrileduring labeling, greatly facilitated the extraction since greater than 70% of the newly synthesized collagen from all specimens was extracted under nonproteolytic conditions with 1.0 M NaCl/ 50 mM Tris, pH 7.4, buffer followedby0.5 M acetic acid. Subsequent proteolytic digestion with 1mg/ml pepsin at 4 "C solubilized less than 6% of newly synthesized collagen from both presumptive cartilaginous and calcification regions at the three different sternal ages examined. Treatment of the pepsin-insoluble residues under reducing conditions with 0.15 M NaC1/50 mM DTT, pH7.4, buffer followed byan additional pepsin digestion at 4 "C completely solubilized the remainder of the labeled collagen from all specimens. Comparison of the patternsof collagen solubilization from the permanent hyaline cartilage and presumptive calcification regions showed differences with increasing sternal age. The proportion of newly synthesized collagen present in the readily extractable pool from the hyaline cartilage region exhibited a moderate increase from 79.3 to 83.4% during development from stage 43 to 46, whereas it decreased from 81.1 to 70.6% in the presumptive calcification cartilage (Table I). The pool of pepsin-extractable molecules showed a decrease in both zones with increasing sternal age. The pool of pepsin-resistant molecules, however, showed notable differences between the two zones. While the proportion of these molecules remained constant during development of the hyaline cartilage, it increased from 13.5% to 27.6% in the cartilage from the presumptive calcification zone (Table I). SDS-Gel Electrophoresis-The labeled proteins present in the neutral salt and DTT extracts from the permanent hyaline cartilage and presumptive calcification regions of chick sternum at different stages of development were examined by SDS-gel electrophoresis. The very low amounts of collagen extracted with acetic acid and pepsin were not sufficient to permit further study of these fractions. Identification of collagenous bands in the gels was performed by comparison of their relative migration to standard collagen chains and by their susceptibility to digestion with purified bacterial collagenase. Electrophoretic and fluorographic examination of the 1.0 M NaCl extracts from the two separate regions at stage 40 of sternal development before and after proteolytic digestion is shown in Fig. 1. The pattern of labeled proteins present in the samples from both regions under reducing conditions (lanes 2 and 6) shows a prominent band corresponding to


2899

TABLE I Extractability of newly synthesized collagen from the cartilaginous region and presumptive calcification regions of stages 40, 43, and 46 embryonic chick sternum Embryonic chick sterna from stages 40,43, and 46 were separated into permanent hyaline (caudal) and presumptive (cephalic) cartilaginous regions, minced, and labeled with [“Clproline as described under “Experimental Procedures.” After labeling the tissue was homogenized in 1.0 M NaCl, 50 mM Tris-HCI, pH 7.4, buffer and sequentially extracted as described previously (25). Aliquots of the extracts were hydrolyzed after extensive dialysis and their [‘4C]hydroxyprolinecontent assayed by the method of Juva and Prockop (26). Presumptive calcification cartilage

Permanent hyaline cartilage Extractants

1.0 M NaCl

Embryonic stages

40 43 46 40 43 46 40 43 46

DTT + pepsin after DTT

2

3

4

% of total

5

6

617.8 1624.3 1530.6 47.3 83.0 28.9 113.7 276.2 275.9

7

8

2

FIG. 1. SDS-polyacrylamide gel electrophoresis fluorograph of 1 .O M NaCl extracts from permanent hyaline cartilage and presumptive calcification cartilage from stage 40 sternum. Samples of intact or pepsin-digested 1.0 M NaCl extracts were denatured and electrophoresed under reducing and nonreducing conditions in the presence or absence of 1%(v/v) P-mercaptoethanol and 1% SDS asdescribed under “ExperimentalProcedures” and were electrophoresed in 4-10% polyacrylamide exponential gels as described previously (21, 22). Lane l , permanent hyaline cartilage extract without reduction before pepsin digestion; lane 2, as lane 1 after reduction; lune 3, pepsin-digested extract without reduction; lane 4, as lane 3 after reduction; lane 5, presumptive calcification cartilage extract without reduction before pepsin digestion; lane 6, as lane 5 after reduction; lane 7, pepsin-digested extract without reduction; lane 8, as lane 7 after reduction.

al(I1) and other minor bands migrating above it. Three of these bands correspond to thepre-pepsin forms of la, 2a, and 3a and the remaining bands appear to correspond to intact and partially processed Type I1 procollagen molecules. As expected, after pepsin digestion these molecules migrated more rapidly due to removal of their noncollagenous extensions. As described previously, lap and 201, migrated above al(II),, and 3a, co-migrated with it. The electrophoretic patterns obtained underreducing and nonreducing conditions from intact and pepsin-treated extracts from both regions were essentially identical (Fig. 1, lanes 1-4 and 5-8). Bands with electrophoretic migration corresponding to Type IX and Type X collagens were not detected in either region at stage

[“C]Hydroxyproline

% of total

dprn X IO”

dpm X

+ 0.5 M acetic acid

Pepsin (1 mg/ml)

1

[“C]Hydroxyproline

79.3 81.9 83.4 6.1 4.2 1.6 14.6 13.9 15.0

1583.2 2438.7 1996.7 94.7 122.6 54.4 262.6 385.9 762.5

81.6 82.3 70.6 4.9 4.2 1.8 13.5 13.1 27.6

40 of development even when overloaded samples were examined. The electrophoretic analysis of the labeled collagens present in the 1.0 M NaCl extracts at stage 43 showed marked differences between the two regions (Fig. 2 A , lane 1 and Fig. 2B, lane I). In the extracts from the presumptive calcification zone a prominent band migrating with an apparent M, = 60,000 was observed and represented about 16% of the total radioactivity migrating into the gel. This band was absent in the region of permanent hyaline cartilage. The M , 60,000 band did not change its electrophoretic mobility under reducing and nonreducing conditions (Fig. 2B, lanes 1 and 2). After pepsin digestion the M , = 60,000 band migrated into a 45,000 M, position indicating proteolytic removal of noncollagenous domains (Fig. 2B, lane 3). The migration of the 45,000 M, was unchanged under reducing and nonreducing conditions (Fig. 2B, lune 4). The electrophoretic mobility of this band is consistent with the behavior of Type X collagen before and after pepsin treatment (14,15,18,28). The collagenous nature of the labeled proteins migrating in this position was confirmed by their disappearance after collagenase digestion (not shown). Similar analysis of the permanent hyaline cartilage region (Fig. 2 A , lanes 1-4) failed to show any radioactive bands corresponding to Type X collagen. Additional bands migrating below al(I1) chains in the pepsin-digested samples from both regions displayed electrophoretic mobilities similar to those of Type IX collagen after pepsin digestion and disulfide bond reduction as previously described (8, 13,19). The labeled molecules present in the 1.0 M NaCl extracts from stage 46 sterna showed a similar pattern to those from sterna at stage 43 of development (Fig. 3B). The M, 60,000 band corresponding to Type X collagen was found exclusively in the presumptive calcification region and represented approximately 38% of the totalradioactivity migrating into the gel (Fig. 3A, lane 1 and Fig. 3B, lane 1). Pepsin digestion also resulted in the removal of noncollagenous proteins yielding the expected pepsin-cleaved Type X (Mr= 45,000) band (Fig. 3B, Lanes 3 and 4). The migration pattern under reducing and nonreducing conditions did not change for either form of Type X collagen (Fig. 3B, lunes 3-4). The dramatic increase in the proportion of Type X collagen in the pepsin-digested extracts from the presumptive calcification region during development from stage 40 to 46 was accompanied by a concomitant decrease in the proportion of Type IX collagen

2900

Type X Collagen Biosynthesis in Chick Embryo Development 2

3

B I

4

FIG. 2. SDS-polyacrylamide gel electrophoresis fluorogaph of 1.0 M NaCl extracts from permanent hyaline cartilage ( A ) and presumptive calcification cartilage ( B ) from stage 43 sternum. Samples of intact or pepsin-digested 1.0 M NaCl extracts were denatured and electrophoresed under reducing and nonreducing conditions as described in the legend to Fig. 1. Lane 1, electrophoresis without reduction of intactextract before pepsin digestion; lane 2, as lane 1 after reduction; lane 3, electrophoresis of pepsin-digested extract without reduction; lane 4, as lane 3 after reduction.

54K

45K

-

+

‘84K 69K 60K 45K 35K

*

2 FIG. 3. SDS-polyacrylamide gel electrophoresis fluorogaph of 1.O M NaCl extracts from permanent hyaline cartilage ( A ) and presumptive calcification cartilage ( B ) from stage 46 sternum. Samples of intact or pepsin-digested 1.0 M NaCl extracts were denatured and electrophoresed under reducing and nonreducing conditions as described in the legend to Fig. 1. Lane I , electrophoresis without reduction of intactextract before pepsin digestion; lane 2, as lane I after reduction; lane 3, electrophoresis of pepsin-digested extract without reduction; lane 4, as lane 3 after reduction.

2

3

4

---

3

2

3

4

I

69K-c

60 Ki

polypeptides from 19 to 5% and areduction in the proportion of la and 2a collagen from 22 to 11%. Electrophoretic analysis of the DTT extracts is shown in Fig. 4. The electrophoretograms of these extracts disclosed dramatic qualitative and quantitative changes in the expression of Type X collagen in cartilage from the presumptive calcification cartilage with increasing developmental age. Whereas no radioactivity migrating in the position of pepsincleaved Type X collagen (MI = 45,000)could be detected in samples from stage 40, even when overloaded gels were examined, large amounts of radioactivity were found migrating in this region at stages 43 and 46 of development. Type X comprised about 11% of the total radioactivity at stage 43 sternum. This proportion increased to 45% at stage 46 of development. Bands corresponding to Type X collagen were not detected at all in the extracts from the permanent hyaline cartilage at anyof the threestages. The figure also shows that the extracts from both zones at thethree developmental stages exhibited several bands migrating below al(I1) chains. The apparent molecular weight of these bands estimatedby com-

45K-c

parison to the migration of Type I1 collagen CNBr peptides was calculated to be in decreasing order: 80,000, 69,000, 54,000,and 49,000 Mr. In most experiments the 69,000 M, band was the most prominent. These bands are similar to those observed after pepsin digestion and subsequent reduction of Type IX collagen polypeptides described by Gibson et al. (19)and Yasui et al. (13). CNBr Peptide Pattern Analysis-In order to further characterize the collagens present in the pepsin-digested 1.0 M NaCl extracts of the presumptive calcification region, twodimensional CNBr mapping of extracts from both regions of stage 46 sterna was performed. Comparison of the peptide patterns obtained from the permanent hyaline and presumptive calcification cartilage extracts showed the presence of the major CNBr peptides of Type I1 collagen. The four prominent peptides observed (arrowheads,Fig. 5) correspond in decreasing order to CB 10.5 (31,000 Mr), CB 11 (25.000 M,), CB 8 (13,000Mr), and CB 9.7 (10,soO M,).The major CNBr peptides of Type X collagen exhibited a M. below 13,000 and were found only in the presumptive calcification region (smoll

Type X Collagen Biosynthesis in Chick Embryo Development I

2

3

4

5

6

2901

arrows). These results are similar to the CNBr peptide patterns of Type X collagen from chick embryo tibiotarsus (14, 15)or from long term culture of chick chondrocytes grown in three-dimensional gels (18). lap Quantitative Analysis of the Relative Proportion of Collagen ZapSynthesized by the Developing Embryonic Sterna-To compare the relative proportion of the various collagens synthesized by the separate sternal regions during development, 8 4 K -. fluorographs of the 1.0 M NaCl and DTT extracts were 6 9 K -. examined by densitometric analysis and the areas under each collagen chain were calculated by planimetry. The results 45K -. shown in Table I1 indicate that the sternum exhibits both spatialand temporal expression of collagenous molecules during embryonic development. The permanent hyaline cartilage region displayed a constant relative proportion of 3a, + ul(II), biosynthesis at the three different sternal stages whereas the proportion of la, + 2a, decreased from 29.5% at stage 40 to 21% at stage 43 and 23% at stage 46. On the other hand, the proportion of Type IX collagen exhibited an inFIG.4. SDS-polyacrylamide gel electrophoresis fluoro- crease from 9% at stage 40, to 18% at stage 43, and to 19% at graph of DTT extracts from permanent hyaline cartilage and stage 46 of development. As expected, the permanent hyaline presumptive calcification cartilage at three stages of development. Samples of DTT extracts were electrophoresed under re- cartilage region did not display any detectable biosynthesis of ducing conditions in 4-10% polyacrylamide exponential gels as de- Type X collagen. In contrast, Type X collagen comprised scribed in the legend to Fig. 1. Lane l , stage 40 sternum permanent approximately 12% of the newly synthesized collagen from hyaline cartilage; lane 2, presumptive calcification region; lane 3, the presumptive calcification region at stage 43 of developstage 43 sternum permanent hyaline cartilage; lane 4, presumptive ment, increasing to 45.6% at stage 46. No expression of Type calcification region; lane 5, stage 46 sternum permanent hyaline X collagen biosynthesis could be detected in sternum from cartilage; lane 6, presumptive calcification region. stage 40 embryos. The increase in Type X collagenwas accompanied by a decrease in the relative proportion of all the other collagens. Type IX collagen declined from 11.9% at stage 40 to 5.2% at stage 46; while the la, and 2a, declined from 32% to 13% and 3a, + al(II), decreased from 56 to 37%. A comparison of the rates of protein synthesis and collagen biosynthesis corrected for the absolute number of cells obtained from each of the separate regions demonstrated that chondrocytes from the presumptive calcification region were substantially more active than those from the zone of permanent hyaline cartilage at both sternal stages (Table 111). A t stage 40, the presumptive calcification chondrocytes incorporated two times as much [“Clproline and synthesized 2.5 greater amounts of [ “C] hydroxyproline when calculated on a per cell basis. Similarly, at stage 46 of development, cells from the presumptive calcification region incorporated 2.2 times as much [“Clproline and synthesized about 3 times more [“C] hydroxyproline than cells from the permanent hyaline cartilage region. When the rates of synthesis of Type I1 and Type X collagens synthesized per cell by the permanent hyaline and presumptive calcification chondrocytes were examined, it was found that Type X collagen represented the major biosynthetic product of presumptive calcification chondrocytes at stage 46, since these cells synthesized essentially equal amounts of Type X and Type I1 collagens. Furthermore, a comparison of the amount of Type X collagen produced by stage 46 presumptive calcification chondrocytes to the amount of Type I1 collagen produced by permanent hyaline cartilage chondrocytes at the same stage demonstrated that the presumptive calcification cartilage chondrocytes synthesized FIG.5. Fluorograph of two-dimensional CNBr peptide map- 250% more Type X collagen than the amount of Type I1 ping of permanent hyaline cartilageand presumptive calcifi- collagen synthesized by the permanent hyaline cartilage choncation cartilage collagen chains. Two-dimensional CNBr-peptide drocytes. pping was performed as previously described (21,22).First dimension Morphologic Changes during theDevelopment of Embryonic samples were pepsin-digested reduced 1.0 M NaCl extracts from the Sterna-To correlate the changes in collagen biosynthetic permanent hyaline cartilage and presumptive calcification cartilages expression with gross morphologic changes occumng in the of stage 46 embryonic sterna. Direction of electrophoresis is indicated by large arrows. The peptides of Type X collagen are shown by small developing chick sterna, whole sterna from embryos at stages 40, 43, and 46 of development were differentially stained for arrows. The main CNBr peptides from al(I1) chain are shown by proteoglycans with alcian blue (Fig. 6A) and forcalcified arrowheads.

-.


2902

TABLErr Relative proportion of various collagen chains synthesized by the permanent hyaline and presumptive calcification regions of embryonic chicksterna at various stages of development The relative proportion of the various collagen chains was calculated from electrophoretic analysis of pepsindigested 1.0 M NaCl extracts and DTTextracts from the permanent hyaline cartilage and presumptive calcification regions as described under "Experimental Procedures." After fluorography the gels were scanned at 540 nm in a linear drive densitometer and theareas under the peaks corresponding to each of the collagen chains were calculated with a planimeter. The values shown are theaverages of values obtained with two separate preparationsof extracts. The acetic acid and pepsin extracts were not included in this analysis because of their low radioactivity. Proportion of each collagen chain

Embryonic stage

Sternal region

la

+ zap

3ap + al(II),

Type IX,

Type X, ~

Permanent hyaline 8.9 cartilage 18.5

61.6 61.4 58.6 55.9 51.7

Presumptive calcification 11.3 region

40 43 46 40 43 46

29.5 20.7 23.1 32.7 27.9 5.2 12.9

TABLE I11 Comparison of the rates of f4C]proline incorporation, P4C] hydroxyproline synthesis, and production of the various collagens on a per cell basis betweenpermanent hyaline and presumptive calcification regionsof embryonic sternum a t stages 40 and 46 of development The total amounts of [14C]prolineincorporated and ['4C]hydroxyproline synthesized by organ cultures from the separate sternal regions of the two sternal ages during a 48-h labeling period were corrected for the number of cells obtained by enzymatic dissociation of sternum as described under "Experimental Procedures." Each value represents an average of two separate experiments andis expressed as dpm X 10-3/i03 cells/48 h. Permanent hyaline cartilage

Presumptive calcification region

Stage 40 Stage 46 Stage 40 Stage 46

dpm X lO"/l6 cells/48 h

Total I 4 C 445.5 199.5 216.7 Total ['4C]hydroxyproline242.6 204.0 81.0 78.9 ['4C]Hydroxyproline in" 26.1 lap 2aP collagen 49.5 Type I1 collagen' 3.3 Type IX collagen 0 Type X collagen

+

dpm X 10-3/103 cells/48 h

523.1 34.1 39.7 7.2 0

82.4 110.6 11.0 0

42.2 94.1 7.8 98.7

The amount of [14C]hydroxyprolinein each collagen was calculated from densitometric analysis of fluorographs from the pepsintreated neutral salt and the DTTextracts with two separate preparations of extracts and was corrected by the number of cells obtained after enzymatic dissociation of chondrocytes in parallel experiments. In selected cases direct determination of ['4C]hy~oxyprolineafter excision of the corresponding band from the gel and subsequent hydrolysis was performed. Includes 301 collagen chains.

tissue with silver nitrate and alizarin red stains (Fig. 6B). Sterna from stages 40 and 43 displayed uniform staining with alcian blue throughout the tissue indxating that the entire matrix was cartilaginous. However, at stage 46 two circular areas which did not stain with alcian blue corresponding to proteolgycan-depleted matrix appeared in the presumptive calcification region. Silver nitrate staining for CaZf deposits (Fig. 6B) showed marked localization of large Caz+ deposits (arrows)within the two circular areas. The Ca2+deposits were restricted t o the cephalic portion of the proteoglycan-depleted circular areas and were observed only at stage 46 of development. These findings indicate the occurrence of proteoglycan depletion concomitantly with the development of calcification.

~~

19.4 8.3 36.7

12.1 45.6

DISCUSSION

A significant finding of this study is the demonstration of a spatial andtemporal expression of Type X collagen biosynthesis by the presumptive calcification region in thedeveloping embryonic chick sternum. The Type X collagen synthesized by the embryonic chick sterna organ cultures was characterized by electrophoretic analysis, proteolytic digestion, and CNBr-peptide mapping and itwas shown to be similar to Type X collagen from chick embryo tibiotarsus (14, 15) and from cultures of chick chondrocytes grown in three-dimensional gels (18,ZO). In agreement with these studies we found that tissue extracts obtained without protease digestion contain intactType X collagen whichmigrate electrophoretically with an apparent M, of 60,000. These chains do not contain disulfide bonds since they do not change electrophoretic mobility after reduction. Limited proteolytic digestion of the extracts results in cleavage of non-helical regions of these chains yielding 45,000 M , protease-resistant domains. In contrast with previous studies demonstrating a remarkable resistance of Type X collagen to extraction from the tissues (28), we found that the majority of Type X collagen could be extracted with a NaCl buffer at neutral pH when the organ cultures were labeled in thepresence of P-aminopropionitrile. Thus, it appears that the majority of Type X collagen molecules in the matrix are stabilized by P-aminopropionitrilesensitive bonds. A small proportion of Type X collagen remained in the tissue in the form of a pepsin-resistant population of molecules which become solubilized only following disulfide-bond reduction. Comparison of [14C]prolineincorporation and ['4C]hydroxyproline synthesis between the two regions showed that the presumptive calcification cartilage was substantially more active than the permanent hyaline region at the threestages analyzed since it displayed %fold greater [14C]prolineincorporationand about 2.5-3-fold greater ['4C]hydrox~roline biosynthesis. Remarkable differences were found when the biosynthetic products of the permanent hyaline cartilage and presumptive calcification regions at various developmental stages were compared. Although both regions synthesized Type 11, l a , Za, 3a, and Type IX collagen chains, only cartilage from the presumptive calcification region displayed synthesis of Type X collagen. Quantitative analysis of the proportion of the various collagens synthesized by both regions exhibited temporal changes as well. For example, in the presumptive calcification region at stage 40 of sternal development no detectable Type X collagen was found. At stage 43, however, Type X represented approximately 12% of the newly synthesized

Type X Collagen Biosynthesis in Chick Embryo Development FIG. 6. A, gross morphology of whole sternum at stages 40, 43, and 46 of embryonic development. Whole sterna from embryos at stages 40, 43, 46 were removed and cleaned and after fixation, they were stained with alcian blue (29). 1 , stage 40; 2, stage 43; 3, stage 46. B, gross morphology of stage 46 sternum stained with alcian blue/alizarin red, and silver nitrate stains. Stage 46 sternum was stainedas described with alcian blue, alizarin red, and silver nitrate as described (29,30). 1, silver nitrate stain; 2, alcian bluelalzarin red stain.

2903

A

B

f I

2

collagen and increased to 45% at stage 46 of sternal development. The increase in Type X was accompanied by a decrease in the relative proportion of la, 2 a , 3a, al(II), and Type IX collagens. In contrast, the permanenthyaline cartilage exhibited an increase in the proportion of Type IX collagen and a slight decrease in the proportion of l a , 2a, 3 a , and al(I1). The mechanisms responsible for the striking changes in the relative proportions of the various collagen subpopulations with sternal development observed in the presumptive calcification region are not clear. The decrease in levels of la, 2 a , and 3a collagens, which are referred to as thecell-associated collagens (31, 3 2 ) , may be related to the loss of membrane integrity and release of lysosomal content from chondrocytes undergoing calcification. The decrease of Type IX collagen in the presumptive calcification region with sternal agemay represent the transformation of the permanent hyaline cartilage matrix to one of calcification. The dramatic increase in Type X collagen production may also reflect changes occurring on the extracellular matrix upon the onset of calcification. The results of quantitative analysis at stage 46 of development demonstrated that Type X collagen represents a major biosynthetic product of these cells and when calculated on a per cell basis it was found that a chondrocyte from the presumptive calcification region produced 250% greater amounts of Type X collagen than the amount of Type I1 collagenproduced by a chondrocyte from the permanent hyaline cartilage region. Whether the initiation of Type X collagen biosynthesis at stage 43 of development demonstrated here reflects changes in chondrocyte gene expression or is the result of expansion of a unique subpopulation of chondrocytes selectively committed to Type X collagen production, is not known. Nevertheless, the strict topographical and temporal distribution of Type X collagen synthesis coincident with initiation of sternal calcification indicates that cells within the presumptive calcification region are intimately involved in the process of endochondral bone formation. In recent studies, Capasso et al. (33, 34) and Schmid and Linsenmayer (17) showed biosynthetic and immunohistochemical acquisition of Type X collagen in the developing tibiotarsus. Capasso et al. (34) suggested that the Type X production may be the sequela of blood vessel ingrowth, in which a serum factor might be responsible for initiating its synthesis. The studies reported here, however, indicate that the earliest biochemically detectable synthesis of Type X collagen occurred at stage 43, before the appearance of any form of vascularization in the tissue, which is usually observed at the later stage of development (stage 46). It should be noted, however, that maximal Type X collagen biosynthesis was coincident with morphological observations of proteoglycan depletion and Ca2+deposition at themost cephalad region

3

I

2

of the presumptive calcification cartilage. Thus, it could be postulated that Type X collagen could provide a permissive matrix for calcification, but the process of calcification being under a different set of regulatory controls. Elucidation of this possibility, however, will require in vitro studies on the calcification of Type X collagen and the demonstration of mineral deposits characteristic of calcification after tissue implantation of purified Type X collagen. These studies are currently being performed. Acknowledgments-The expert assistance of E. Lobb in preparation of the manuscript is gratefully acknowledged. REFERENCES 1. Ham, A. W., and Cormack, D. H. (1979) in Histology (Ham, A. W. and Cormack, D.H., eds) 8th Ed, pp. 377-422, J. B. Lippincott Co., Philadelphia 2. Bornstein, P., and Traub, W. (1979) in The Proteins (Neurath, H., and Hill, R. L., eds) Vol. 4, pp. 411-632, Academic Press, New York 3. von der Mark, H., von der Mark, K., and Gay, S. (1976) Deu. Biol. 48,237-249 4. von der Mark, K., von der Mark, H., and Gay, S. (1976) Deu. Biol. 53. 153-170 5. von der Mark, K., and von der Mark, H. (1977) J. Bone Jt. Surg. Br. Vol. 59,458-464 6. Burgeson, R. E., and Hollister, D. W. (1979) Biochem. Biophys. Res. Commun. 87,1124-1131 7. Burgeson, R. E., Hebda, P. A., Morris, N. P., and Hollister, D. W. (1982) J. Biol. Chem. 2 5 7 , 7852-7856 8. Reese, C.A., Wiedemann, H., Kuhn, K., and Mayne, R. (1982) Biochemistry 21,826-830 9. Richard-Blum, S., Hartmann, D. J., Herbage, D., Payen-Meyran, C., and Ville, G. (1982) FEBS Lett. 146,343-347 10. Duance, V. C., Wotton, S. F.,Voyle, C. A., and Bailey, A. J. (1984) Biochem. J . 221,885-889 11. von der Mark, K., von Menxel, M., and Wiedemann, H. (1984) Eur. J. Biochem. 138,629-633 12. van der Rest, M., Mayne, R., Ninomiya, Y., Seidah, N. E., Chretein, M., and Olsen, B. R. (1985) J. Biol. Chem. 260,220225 13. Yasui, N., Benya, P. D., and Nimni, M. E. (1984) J . Biol. Chem. 259,14175-14179 14. Schmid, T. M., and Conrad, H. E. (1982) J. Biol.Chem. 2 5 7 , 12444-12450 15. Schmid, T. M., and Conrad, H. E. (1982) J. Biol.Chem. 2 5 7 , 12451-12457 16. Schmid, T. M., and Linsenmayer, T. F. (1983) J. Bwl.Chem. 258,9504-9509 17. Schmid, T. M., and Linsenmayer, T. F. (1985) Deu. Biol. 1 0 7 , 373-381 18. Gibson, G. J., Schor, S. L., and Grant, M. E. (1982) J. Cell Biol. 93,767-774 19. Gibson, G. J., Kielty, C. M., Garner, C., Schor, S. L., and Grant, M. E. (1983) Biochem. J . 211,417-426 20. Gibson, G. J., Beaumont, B. W., and Flint, M. H. (1984) J. Cell Biol. 99,208-216 21. Remington, M. C., Bashey, R. I., Brighton, C. T., and Jimenez, S. A. (1983) Collagen Relat. Res. 3 , 271-278

2904


22. Remington, M. C., Bashey, R. I., Brighton, C. T., and Jimenez, S. A. (1984) Biochem. J. 224,227-233 23. Shinz, H. R., and Zangerl, R. (1937) Morphologisches Jahrbuch Gegenbaurs 80, 620-628 24. Hamburger, V., and Hamilton, H. L. (1951) J. MorphoL 8,49-92 25. Jimenez, S. A., and Bashey, R. I. (1978) Biochem. J. 173, 337340 26. Juva, K., and Prockop, D. J. (1966) Anal. Biochem. 15, 77-83 27. Barsh, G., and Byers, P. (1978) Prnc. Natl. Acad. Sei. U. S. A. 78,5124-5146 28. Jimenez, S. A., Yankowski, R., and Reginato, A. M. (1986) Biochem. J., in press

29. Dingerkus, G., and Uhler, L. D. (1977) Stain Technol. 52, 229232 30. Conn, H. J., Darrow, A., and Emmel, V. M. (1960) in Stain Procedures. Biological Stain Commission 2nd Ed, pp. 151-152, Williams & Wilkins Co., Baltimore 31. Reese, C. A., and Mayne, R. (1981) Biochemistry 20,5443-5448 32. Liotta, L. A., Kalebic, T., Reese, C. A., and Mayne, R. (1982) Biochem. Biophys. Res. Commun. 104,500-506 33. Capasso, O., Quarto, N., Descalzi-Cancedda, R., and Cancedda, R. (1984) EMBO J. 3,823-827 34. Capasso, O., Tajana, G., and Cancedda, R. (1984) Mol. Cell. Biol. 4,1163-1168