EXPERIMENTAL

CELL

RESEARCH

192,

‘L66-“?0

(1991)

Cell Hypertrophy and Type X Collagen Synthesis in Cultured Articular Chondrocytes MAURIZIO

PACIFICI,*”

ELEANOR B. GOLDEN,*

Departments of ‘AnatomJl-Histology University of Pennsyluania,

SHERRILL

and tBiochemistry, School Philadelphia, Pennsylvania

Articular cartilage is a permanent tissue whose cells do not normally take part in the endochondral ossification process. To determine whether articular chondrocytes possess the potential to express traits associated with this process such as cell hypertrophy and type X collagen, chondrocytes were isolated from adult chicken tibia1 articular cartilage and maintained in long-term suspension cultures. As a positive control in these experiments, we used parallel cultures of chondrocytes from the caudal portion of chick embryo sternum. Both articular and sternal chondrocytes readily proliferated and progressively increased in size with time in culture. Many had undergone hypertrophy by 4-5 weeks. Analysis of medium-released collagenous proteins revealed that both articular and sternal chondrocytes initiated type X collagen synthesis between 3 and 4 weeks of culture; synthesis of this macromolecule increased with further growth. Immunofluorescence analysis of B-week-old cultures showed that about 15% of articular chondrocytes and 30% of sternal chondrocytes produced type X collagen; strikingly, there appeared to be no obvious relationship between type X collagen production and cell size. The results of this study show that articular chondrocytes from adult chicken tibia possess the ability to express traits associated with endochondral ossification when exposed to a permissive environment. They suggest also that the process of cell hypertrophy and initiation of type X collagen synthesis are independently regulated both in articular and sternal chondrocytes. e issi Academic PRSS, I~C.

INTRODUCTION The behavior and developmental fate of articular chondrocytes differ markedly from those of epiphyseal chondrocytes. Articular chondrocytes are present throughout postnatal life and maintain largely unchanged their morphological and biosynthetic features [l]. In contrast, growth plate chondrocytes persist only through puberty and undergo profound phenotypic ’ To whomreprint requestsshouldbeaddressed. 0014.4827/91 $3.00 Copyright 0 1991 by Academic Press, All rights of reproduction in any form

266 Inc. reserved.

L. ADAMS,* AND IRVING M. SHAPIRO? of Dental

Medicine,

19104-6003

changes as they participate in endochondral ossification [2]. During this process, small resting and proliferating chondrocytes mature into large hypertrophic cells; synthesis of a unique collagen (type Xl is activated [3-51; and the extracellular matrix surrounding hypertrophic chondrocytes is mineralized [6]. The mature hypertrophic chondrocytes are eventually replaced by bone cells. The mechanisms that regulate these strikingly diverse developmental programs in articular and growth plate chondrocytes are unknown. Several groups of investigators have attempted to determine whether the phenotypic changes occurring during the endochondral process are genetically predetermined or are under microenvironmental control [7, 81. Chondrocytes isolated from the caudal region of chick embryo sternum were utilized in these studies. This region contains immature chondrocytes, remains cartilaginous during early postnatal life, and undergoes endochondral ossification only in the adult animal [9]. When grown in suspension culture or within collagen gels, these cells were found to proliferate, increase in average size, and initiate type X collagen synthesis by 3 to 4 weeks of culture. Thus removal from the in uiuo environment and growth under permissive culture conditions clearly induce precocious cell hypertrophy and type X collagen synthesis in chondrocytes from the caudal region of the sternum. These observations have led to the conclusion that chondrocytes possessthe intrinsic potential to express traits associated with endochondral ossification and that the active expression of these traits is controlled by microenvironmental cues [7, 8, lo]. Because articular cartilage is present throughout postnatal life and does not undergo endochondral ossification, its cells need not possessthis intrinsic potential displayed by sternal chondrocytes and other chondrocytes destined to be replaced by bone cells. To address this issue, chondrocytes were isolated from tibia1 articular cartilage of &week-old chickens and maintained under permissive culture conditions for several weeks. We found that articular chondrocytes are very similar to caudal sternal chondrocytes grown in parallel cultures. Both cell types underwent hypertrophy and produced type X collagen with time in culture.

TYPE

X COLLAGEN

SYNTHESIS

BY

ARTICIJLAR

267

CHONDROCYTES

F 1

amounts of incorporated radioactivity were electrophoresed on 6% SDS-polyacrylamide gels under reducing conditions 1161. Gels were processed for fluorography and exposed to Kodak XAR-5 films at -70°C. Type X collagen was identified by its intense labeling with proline, its sensitivity to bacterial collagenase as described [15], its relative electrophoretic mobility [3], and cyanogen bromide peptide mapping [5]. Moreover, we carried out partial amino acid sequencing of gel-purified type X collagen; the sequence obtained

1 C

Ser-Asp-Gly-Tyr-PheeSerCrlu-Arg-

FIG. 1. Micrograph ofa longitudinal section through tibia1 articular cartilage. Proximal tibia1 articular cartilage from R-week-old chickens was processed for histological analysis and stained with hematoxylin-eosin. The central (C) portion of articular cartilage that lies below the fibrocartilage (F) was used as a source of articular chondrocytes. Bar, 300 ym.

MATERIALS

AND

METHODS

Cell cultures. Proximal tibia1 articular cartilage was removed from the epiphysis of R-week-old chickens under sterile conditions. The central portion was surgically separated from the underlying cells and the overlying layer of fibrocartilage [ll] (see Fig. 1). The tissue was finely minced with surgical blades and incubated overnight at 37°C in high glucose Dulbecco’s modified Eagle’s medium containing 2 mM L-glutamine, 0.25% trypsin, 0.1 % collagenase, 50 IT/ml each of penicillin and streptomycin, 1.25 pg/ml fungizone, and 10 pg/ml gentamitin. The released cells were separated from tissue fragments by filtration through 20 pm Nitex, recovered by centrifugation, suspended in complete medium (see below), and plated in primary culture at a density of 3 x lo6 cells/lOO-mm tissue culture dish (Falcon) in 10 ml of medium/dish. Pure populations of floating chondrocytes were harvested from the medium of Day 6-7 primary cultures by centrifugation. These cells were maintained in suspension culture thereafter, using loo-mm bacteriologic dishes plated at an initial density of 445 x 106. Cultures were fed twice a week with complete medium and were subcultured weekly at the same initial density. The complete medium consisted of high glucose Dulbecco’s medium containing 10% fetal calf serum (Hyclone), 2 mM L-glutamine, and 50 U/ml each of penicillin and streptomycin [12]. The caudal one-third portion of sternal cartilage was dissected from l&day-old chick embryos. The minced tissue was incubated for 3 h at 37°C in balanced salt solution containing 0.25% trypsin and 0.1% collagenase. The released cells were then maintained under culture conditions identical to those described above and served as a positive control in this study. T>jpe X collagen s,ynthesis. Cultures were labeled for 16 h with 10 pCi/ml of [5-aH]proline (Amersham; 35 Ci/mmol) in complete medium containing 10 pg/ml ascorbic acid and 100 pg/ml fi-aminopropionitrile fumarate. Medium and cells were separated after labeling. Cell layer samples were not analyzed further, because they contained low levels of type X collagen [13; our unpublished observations]. Collagenous proteins were precipitated from the medium by addition of ammonium sulfate to 30% of saturation [14, 151. Precipitated material was recovered by centrifugation at 13,OOOg for 15 min. Pellets were washed with 30% ammonium sulfate and then 70% ethanol and were solubilized in gel sample buffer [16]. Aliquots containing similar

Tyr-GlnX-Gln-SerrSer-Ile.

is in complete accordance with the sequence predicted by cDNA analysis [17]. Type II collagen and its intermediate forms were identified as described [14]. Identification of type IX collagen subunits was based on electrophoretic comigration with purified type IX collagen subunits prepared as described [IS]. An&v-urn preparation and immunocytochemistry. A detailed description of the preparation and characterization of type X collagen antiserum used in the present study will be provided elsewhere (Pacifici et al., in preparation). Briefly, type X collagen was isolated from the medium of chick sternum chondrocyte cultures as described [13]. Collagenous proteins recovered from the culture medium after two precipitations with 30% ammonium sulfate were salt-fractionated at acid pH. The material precipitated by 2 M NaCl in 0.5 M acetic acid (consisting primarily of type X collagen) was separated on a preparative 6% SDS -polyacrylamide gel under reducing conditions. The prominent type X collagen band, visualized by incubating the gel in 0.25 A4 KC1 at 4°C for 1 h, was cut out of the gel and electroeluted. Approximately 200 pg of the electroeluted protein was mixed with 1 ml of complete Freund’s adjuvant and injected intradermally into a rabbit. A booster injection with 100 Kg of type X collagen in incomplete adjuvant was performed after 3 weeks, and blood was collected 1 week later. Specificity of the antiserum was established in several ways. In immunofhiorescence assays, the antiserum stained the matrix surrounding hypertrophic chondrocytes but not that of proliferating chondrocytes or surrounding noncartilaginous tissues in chicken tibia1 growth cartilage. In immunoprecipitation assays [12], the antiserum reacted with type X collagen synthesized by cultured hypertrophic chondrocytes but did not react with types II, IX, and XI collagen (see Fig. 3, lane 10) or with type I collagen synthesized by cultured dermal fibroblasts. In ELISA, the antiserum reacted with type X collagen but failed to react with types I. II, IX, and XI collagens. Immunocytochemistry was carried out as described previously [12]. Floating articular and sternal chondrocytes from 5-week-old cultures were treated with 0.25% trypsin for 20 min to remove the pericellular matrix and allowed to adhere to 4.5.mm tissue culture plates for 24 h in the presence of 4 U/ml of testicular hyaluronidase (Calbiochem) 112, 19, 201. This combination of enzymatic treatments results in the adhesion of the entire cell population. Cultures were then fixed with 70% ethanol for 5 min and exposed for 1 h to a 1:200 dilution of type X collagen antiserum in phosphate-buffered saline containing 10% normal goat serum to block nonspecific antibody binding. After rinsing, bound antibodies were localized by incubation with a 1:125 dilution of rhodamine-conjugated goat anti-rabbit IgGs (Cappell) for 45 min. Samples were viewed under epifluorescence using a Zeiss photomicroscope. Percentage of positive cells was determined by examining a total of about 200 cells in random fields. RESULTS

Chondrocyte

Cultures

and HypertrophJ

In initial experiments we examined the behavior of chicken tibia articular chondrocytes in the culture con-

268

ET

AL.

majority of chondrocytes remained relatively small, but a few chondrocytes had reached a diameter of 15-20 pm (Figs. 2B and 2E). By 4-5 weeks of culture, many chondrocytes had grown in size and some of them exceeded 25 pm in diameter (Figs. 2C and 2F). Interestingly, the average size of articular chondrocytes in these cultures was appreciably smaller than that of sternal chondrocytes (cf. Figs. 2C and 2F). Thus, in terms of behavior in culture, articular and sternal chondrocytes are very similar. Initially both cell types attached to the substrate and after a few days became detached. In addition, the cells exhibited similarities in morphological appearance and proliferative potential. The data also show that both types of cells gave rise to hypertrophic cells with time in culture. However, there were fewer hypertrophic cells in the articular chondrocyte cultures. Type X Collagen Synthesis To study type X collagen synthesis, chondrocytes in l-, 2-, 3-, 4-, and 5-week-old suspension cultures were labeled for 16 h with [3H]proline and collagenous proFIG. 2. Phase micrographs of live chondrocytes in culture. (A teins secreted into the culture medium were analyzed by and D) Articular and sternal chondrocytes, respectively, during the SDS-PAGE and fluorography. Both l- and 2-week-old first 2 days of culture. Note that both cultures contained round cells articular and sternal chondrocytes synthesized large which were of small average size. Some of the elongated cells present could be contaminating fibroblastic cells. (B and E) 2-week-old susamounts of type II collagen and lesser amounts of type pension cultures of articular and sternal chondrocytes, respectively. IX collagen, but no detectable type X collagen (Fig. 3, Note that some cells have undergone cell hypertrophy. (C and F) lanes 1, 2, 6, and 7). Between 3 and 4 weeks of culture, S-week-old suspension cultures of articular and sternal chondrocytes, both types of cells activated the synthesis of type X colrespectively. Most cells have increased in size. However, the size inlagen (Fig. 3, lanes 3, 4, 8, and 9). In articular chondrocrease is heterogeneous, and the average size of articular chondrocytes is smaller than that of sternal chondrocytes. Bar, 60 pm. cyte cultures type X collagen synthesis became prominent by 5 weeks (Fig. 3, lane 5). To determine the proportion of cells producing type X collagen in 4- and 5-week-old cultures, we performed ditions routinely used in our laboratories [12, 19, 201. single cell analysis using immunofluorescence. ChonAfter plating in primary cultures, the cells attached to drocytes were allowed to adhere to tissue culture dishes the tissue culture dishes for the first 2-3 days. Most of the cells were round and had an average diameter of for 24 h prior to processing for immunofluorescence. lo-12 pm (Fig. 2A). By Day 6-7 of culture, the cells had Interestingly, some of both the articular and the sternal increased substantially in number and most of them had chondrocytes remained round after attachment to the detached from the substrate, forming a homogeneous substrate while the remainder displayed a very flat polypopulation of floating cells (not shown). Their average gonal shape (Figs. 4A and 4C). The basis for this differsize had not changed substantially. This behavior of ar- ence in morphology is not clear at present. About 15% of titular chondrocytes was quite similar to that of caudal articular chondrocytes and 30% of sternal chondrocytes sternal chondrocytes grown in parallel cultures. These exhibited positive intracellular staining with the type X cells also attached to the substrate were small in size collagen antiserum (Figs. 4B and4D, respectively), probduring the initial 2-3 days of culture (Fig. 2D). By Day ably representing type X collagen accumulated in intra6-7, they had formed a pure population of floating cells. cellular organelles. Staining was associated with both flat and round cells. Strikingly, there was no apparent Floating articular and sternal chondrocytes harvested from the media of Day 6-7 primary cultures were relationship between positive staining and cell size or maintained thereafter in suspension culture using bac- diameter. teriologic dishes. Both types of chondrocytes proliferated extensively and were subcultured at weekly interDISCUSSION vals. Microscopic inspection revealed that the cells grew in size with time in culture. This process, however, apThe results of the study show for the first time that peared to be asynchronous. In 2-week-old cultures, the articular chondrocytes isolated from chicken tibia1 car-

TYPE Artvxlar chondrocytes

I

X COLLAGEN

SYNTHESIS

Sternal chondrocytes II

I

kd

205

-

pro al (II) pc al(ll) pN al(ll) al (II) -al(lX) ;

116 97

1

2

3456789

10

FIG. 3. Fluorograms of SDSpolyacrylamide gels to analyze collagenous proteins. Articular chondrocytes grown in suspension for 1, 2, 3, 4, and 5 weeks (lanes 1 to 5, respectively) and sternal chondrocytes grown for 1, 2, 3, and 4 weeks (lanes 6 to 9, respectively) were labeled with [“Hlproline for 16 h. Collagenous proteins secreted into the medium were analyzed by SDS-PAGE and fluorography. Note that both articular and sternal chondrocytes activated type X collagen synthesis between 3 and 4 weeks of culture. Lane 10, a sample similar to that in lane 9 immunoprecipitated with the type X collagen antiserum. Note that the antiserum specifically immunoprecipitates type X collagen but not the other collagen types. Pro-al(B), PCtul(II), pN-al(B), and tul(II1, precursor and mature forms of type II collagen; nl(IX) and tu3(IX), two of the three subunits of type IX collagen. The third subunit, ru2(IX), cannot be resolved with this electrophoretic system because it comigrates with the mature ol(II) polypeptide; al(X), the single subunit of type X collagen.

tilage undergo cell hypertrophy and activate type X collagen synthesis with time in suspension culture. Thus, articular chondrocytes, like sternal chondrocytes, appear to have the intrinsic ability to express these and other traits [21, 221 associated with endochondral ossification in response to favorable microenvironmental cues or conditions. These findings raise the possibility that the cartilage microenvironment in uiuo does not permit expression of traits associated with endochondral ossification because it may contain factors inhibiting chondrocyte maturation. Alternatively, the tissue may not contain factors that are actually needed to promote chondrocyte maturation and endochondral ossification. Our observations and conclusions correlate well with those in a recent study by Iwamoto et al. [23]. These authors isolated chondrocytes from the permanent region of rabbit rib cartilage and maintained them under permissive culture conditions. They found that the levels of alkaline phosphatase and 1,25-dihydroxychole-

BY

ARTICULAR

CHONDROCYTES

269

calciferol receptor increased 40- and loo-fold with time in culture compared to in uiuo levels. Moreover, the expression of these proteins correlated with a marked accumulation of calcium. Thus, permanent rabbit rib chondrocytes can be induced to express calcification-related traits under permissive culture conditions even though they fail to do so under normal in viva conditions. Not all of the articular and sternal chondrocytes had activated type X collagen synthesis by 5 weeks of culture. This was not surprising as Solursh et al. [7] had shown that many cells in sternal chondrocyte cultures do not exhibit evidence of type X collagen production. There are two explanations of this phenomenon. First, a longer period of culture may be necessary to allow maturation of the entire chondrocyte population. This possibility correlates well with our observations that cell maturation in culture is an asynchronous process. An alternative possibility is that the articular and sternal cell populations may be heterogeneous; they may contain cells that are able to develop into mature, type X collagen-synthesizing cells and cells that are unable to do so in the permissive culture conditions used. The observation that chondrocytes within the same culture express somewhat different phenotypes supports this view. For example, two subpopulations of mammalian chondrocytes can be distinguished using a monoclonal antibody to keratan sulfate, one population expressing this cartilage-characteristic macromolecule and one not express-

FIG. 4. Phase (A and C) and immunofluorescence (B and D) micrographs of articular (A and B) and sternal (C and D) chondrocytes. Cells from 5-week-old suspension cultures were first allowed to attach to tissue culture plates and then processed for immunofluorescence with the type X collagen antiserum. The percentage of positive cells in both articular and sternal cultures varied from experiment to experiment; on average, 15% of articular chondrocytes and 30% of sternal chondrocytes stained positively with the antiserum. Note that there is no obvious relationship between positive staining and cell size or diameter in both articular (B) and sternal (D) cultures. Bar, 60 pm.

270

PACIFIC1

ing it [24]. In addition, differences in cell morphology and metabolism have been observed in different areas of articular cartilage [ 11. Our immunofluorescence data show that type X collagen production is not correlated with cell size and diameter (Fig. 4); thus, based on cell size or diameter, no predictions could be made as to which cells would exhibit positive or negative staining with type X collagen antibodies. This may appear to be in sharp contradiction with the widely accepted notion that type X collagen synthesis is the unique property of hypertrophic cells. Our data, however, are in full agreement with those in a recent study using in situ hybridization [25]. The authors analyzed the expression of several genes encoding extracellular matrix components, including type X collagen, in chicken tibia1 growth cartilage. They found that, whereas type X collagen transcripts are absent in resting immature cells, they are first detectable in late proliferating, oval-shaped, prehypertrophic cells. We have now found by immunohistochemistry that these same cells actually synthesize type X collagen, confirming that type X collagen production is first activated in late proliferating, prehypertrophic cells and continues in hypertrophic cells (Pacifici et al., in preparation). We conclude from these studies and the data in Fig. 4 that cell hypertrophy and type X collagen gene expression are not coordinately regulated and that cell hypertrophy is not necessarily accompanied by activation of type X collagen production. This work was supported by Grants AR39705, AR34411, and DE08239 from the National Institutes of Health. We thank Dr. W. Abrams for his help with the amino acid sequencing.

REFERENCES 1. 2.

Treadwell, B. V., and Mankin, H. J. (1986) Clin. Res. 213,50-61. Bloom, W., and Fawcett, D. W. (1968) A Textbook ninth ed., pp. 212-222, Saunders, Philadelphia.

Received Revised

July 10, 1990 version received

August

22, 1990

ET 3.

Schmid, T. M., 12,444-12,450.

4.

Capasso, O., F. S., Tajana, 197-206.

Cell hypertrophy and type X collagen synthesis in cultured articular chondrocytes.

Articular cartilage is a permanent tissue whose cells do not normally take part in the endochondral ossification process. To determine whether articul...
2MB Sizes 0 Downloads 0 Views