Sequestration of type VI collagen in the pericellular microenvironment of adult chondrocytes cultured in agarose

Osteoarthritis and Cartilage (1996) 4, 275-285 © 1996 Osteoarthritis Research Society

1063-4584/96/040275 + 11 $12.00/0

OSTEOARTHRITIS and CARTILAGE questration of type VI c o l l a g e n in the pericellular nvi onment of adult c h o n d r o c y t e s c u l t u r e d in a g a r o s e BY JIANG CHANG AND C. ANTHONY POOLE

~

t

of A•natomy, School of Medicine, University of Auckland, Private Bag 92019, Auckland, New Zealand

Summary ~:~epresents the chondrocyte and its pericellular microenvironment and plays an important role in the ~ost~•Oa~thritis. Type VI collagen is preferentially localized in the pericellular microenvironment of adult ~ i l a g ~ ~nd increases during osteoarthritis. In this study, we characterized the pericellular sequestration ~:~|tageh: in long-term chondrocyte-agarose cultures, and assessed the action of interleukin-1 on type VI :~~~~na~ n~ ::id ~ ; ~ '::! :: assembly. ~'~~i~°emical and biochemical analysis showed that cultured chondrocytes initiate type VI collagen ~ratiort immediately upon plating and continue pericellular matrix sequestration in a time dependent manner. C ~ l m~roscopy confirmed the cell surface localization and pericellular accumulation of type VI collagen, while i ~ analysis identified a 'cargo-net like' organization of type VI collagen around each chondrocyte. Quantitative ~ i s revealed a primary phase of rapid cell division and low levels of type VI collagen sequestration, followed by •g~ndar~• phase of relative growth stability and high levels of type VI collagen deposition. Interleukin-1 treated c t t | ~ s showed increased sequestration and retention of type VI collagen in an expanded microenvironment sur-~l~rlding the chondrocytes. The~data suggests a role for type VI collagen in the differentiation of the pericellular microenvironment in vitro. Th~ ~¢reased type VI collagen sequestration promoted by interleukin-1 was consistent with previous studies on 0steo~thritic cartilage, and implies a functional role for type VI collagen in the chondron remodeling associated with ~ t ~ g e degradation. K~ywords: Type VI collagen, Pericellular microenvironment, Chondrocytes-agarose culture, lnterleukin-1.

Introduction

c o n n e c t i v e tissues, and is expressed in t h e form of thin, beaded filaments, s t r u c t u r a l l y distinct from o t h e r fibrillar collagens [12, 13]. P u b l i s h e d studies i ndi cat e t h a t the triple-helical domain of type VI collagen cont ai ns several Arg-Gly-Asp sequences involved in cell binding [14], while i n t e r a c t i o n s with cell-surface receptors, such as the in te g rin s [15] and t he NG-2 r e c e p t o r [16], have been reported. In addition, the globular domains of type VI collagen c o n t a i n a num ber of r e p e a t i n g amino •acid motifs homologous with the fibrillar collagen-binding A domain of von Willebrand f a c t o r [17] suggesting a funct i onal i n t e r a c t i o n b e t w e e n type VI "•collagen and the h e t e r o t y p i c collagen fibers of th e e x t r a c e l l u l a r m at ri x [18]. Type VI collagen interacts also with a range of matrix m acrom o le c u le s, including type II collagen, decorin and fibromodulin [19], h y a l u r o n a n [20], and fibronectin [21 ]. U l t r a s t r u c t u r a l studies on the type VI collagen m at ri x produced by cul t ured fibroblasts h a v e d e m o n s t r a t e d both cell--matrix and matrix- m a t r i x interactions, while light, confocal and e l e c t r o n microscopic analyses of isolated c a n i n e c h o n d r o n s have confirmed t h a t type VI collagen i n t e r a c t s

Tttt~i:i:~ONDRON has previously been identified as a dis r ~ miCrOanatomical unit of adul t a r t i c u l a r cartiI~g~: [1,•2]. Each c h o n d r o n consists of a •b h o r ~ y t e s u r r o u n d e d by a pericellular microen~:i~onme~t••that contains collagen types II, VI, IX [h~iii:~| andi:i:::XI [5], the p r o t e o g l y c a n s a g g r e c a n and de~brin [6~7,8], and glycoproteins, such as fib~ectin:ii:i::[9] and link protein [10]. Of these p e ~ t t u l a r i : : : c o m p o n e n t s , type VI collagen has eme~d a~::::a discrete m ol e c ul a r m a r k e r of ch~on m~croanatomy, i n t e r a c t i n g with b ot h the c e ll- $ ~f ace and the h e t e r o g e n o u s matrix componen~::of th e pericellular m i c r o e n v i r o n m e n t [11]. ~:: VI collagen is widely distributed in •

~

C

.............

"

S~hrai~q~:::i::3 January 1996; accepted 6 March 1996. :This::~e~arch was supported by grants from the Auckland M~icgti::i:i::~Research Foundation (J.C.), the Health Research ~i~ncii::i:~Jt~::New Zealand (C.A.P.), the Arthritis Foundation of N~ve Zealand and the New Zealand Lottery Grants Board. Addres~ correspondence to Dr C Anthony Poole, HRC Senior R$~earch •Fellow, Department of Anatomy, School of Medicine, U!~iVersity of Auckland, Private Bag 92019, Auckland, New

Z~tland.

275 1063-4584/96/040275 + 11 $12.00/0

276

Chang and Poole: Type VI collagen in chondrocyte cultures

directly with the chondrocyte membrane, and extends o u t w a r d from the cell surface to define the structural heterogeneity of the chondron [3, 11]. It is likely, therefore, that type VI collagen performs a bridging function in articular cartilage, not only between the chondrocyte and its pericellular matrix, b u t also b e t w e e n the spatially integrated matrix macromolecules of the pericellular microenvironment. Interleukin-1 (IL-1) is a critical mediator of cartilage catabolism during osteoarthritis [22], and has consistently been shown to down-regulate proteoglycan and collagen production in isolated chondrocyte and cartilage explant cultures [22, 23]. Simultaneously, IL-1 increases the degradation of these components by up-regulating the production of matrix metalloproteinases such as stromelysin and collagenase, but not the tissue inhibitors of the metalloproteinases (TIMPs) [24, 25, 26]. Interestingly, type VI collagen appears resistant to metalloproteinase attack [26, 27], and higher levels of type VI collagen with a broad pericellular distribution have been reported in osteoarthritic cartilage [28, 29, 30]. Because there are currently no reports on the pericellular assembly of type VI collagen by chondrocytes, we have used confocat immunohistochemistry and specific analytical techniques to characterize the progressive pericellular sequestration of type VI collagen during long-term agarose gel culture of adult articular chondrocytes. The modulation of type VI collagen assembly by IL-1 was also examined, and shown to promote significant increases in pericellular type VI collagen sequestration, which parallels the microanatomical changes in type VI collagen which have been reported for osteoarthritic cartilage.

Materials and methods MATERIALS

Hams F-12 culture medium, fetal bovine serum (FBS) and bovine serum albumin (BSA) were obtained from GIBCO, Life Technologies, New Zealand. Sea plaque agarose was obtained from FMC Bioproducts U.S.A. Pronase, and human recombinant IL-I~ in the form of a sterile solution, were obtained from Boehringer Mannheim, New Zealand, while collagenase and testicular hyaluronidase were obtained from Sigma Chemical Co., U.S.A. Cell Tracker Green (5-chloromethylfluorescein diacetate) and ethidium homodimer-1 were obtained from Molecular Probes, OR, U.S.A. The ECL western blotting detection kit was purchased from Amersham , U.K. Anti-type VI

collagen antibody was kindly donated by Dr Shirley Ayad, School of Biological Science, University of Manchester, U.K. I S O L A T I O N OF C H O N D R O C Y T E S A N D A G A R O S E G E L CULTURE

Fresh cartilage samples were collected through the full depth of the tibial plateaus and femoral condyles of mature, healthy crossbred dogs (2-4 years; 20-25 kg) euthanized by the Auckland City P o u n d under veterinary supervision. Cartilage samples were pooled and chondrocytes isolated as previously described [31]. Briefly, diced cartilage was treated with pronase (0.8% w/v, at 37°C for 90 min) in Ham's F-12 medium containing 5% FBS, followed by bacterial collagenase digestion (0.4% w/v, at 37°C for 19h) in Ham's F-12 medium containing 5% FBS. Once digested, cells were centrifuged and washed in medium, and total cell number determined by hemocytometer counting. Cells were suspended in a solution of 1% (w/v) agarose in Ham's F-12 medium with 10% FBS at a density of 1 x 106 celts/35 mm dish and maintained in culture at 37°C for up to 12 weeks. To assess the effects of IL-1 on type VI collagen sequestration, cultures in the early stage of incubation (3-17 days) were used because type VI sequestration was shown to be low and reasonably constant during this period. IL-I~ was added directly to the culture medium at concentrations of 0.5 and 5.0 ng/ml, and the media replaced daily. Triplicate cultures from each treatment regime were collected at selected time points (3, 7, 14 days). One culture was immediately treated with 25#M/ml cell tracker green at 37°C for 2 h to stain viable cells, washed and then fixed in 4% paraformaldehyde for 30 min [32]. After washing, chondrocyte-agarose plugs were cored from the centre of the culture dish as previously described [7], and stored in phosphate-buffered saline (PBS) containing 1% BSA and 0.05% azide at 4°C for subsequent immunohistochemistry. The remaining two cultures were analyzed biochemically for type VI collagen. IMMUNOHISTOCHEMISTRY

The immunohistochemical techniques for staining type VI collagen in c h o n d r o n - a g a r o s e plugs have previously been published [7, 11]. In this study, chondrocyte-agarose plugs were digested with testicular hyaluronidase (2mg/ml in 0.1 Tris saline, pH 5.5, at room temperature for 60 min) to remove proteoglycans that otherwise mask antigenic sites on the type VI collagen molecule.

Osteoarthritis and Cartilage Vol. 4 No. 4 Antl-type ~ p e VI collagen antibody was prepared at~&~•~h~on of 1.~500 in 0.1 M Tris saline with 1% ~ ~ e plugs labeled for 24-48 h at 4°C with ~ s agitation. Treatment with normal ~ : was used as a control. A biotinylated ~ ~irabbit antibody and streptavidin~ ~ i R e d were used for detection. Labeled i ~ ~:~qui!ibrated and mounted in Vectai ~ ~ r L~ib0ratories, CA, U.S.A.), and the i ~ i ~ ~he coverslip sealed with clear nail

277

fixed cultures at each specified time point, and stained for 1 h with 4 ZM ethidium homodimer in Dulbecco's PBS. Three areas from each plug were randomly selected and photographed at a c o n s t a n t magnification using a Nikon Diphot inverted microscope fitted with an epi-illuminated fluorescein isothiocyanate (FITC) filter set. Negative images were projected onto a digitizing pad, and the total number of cells within three defined regions of the field was computed using in-house morphometric analysis software to determine cell • number per unit area.

C O N F O C A L MICROSCOPY

~ VI collagen distribution in chondro~ s e •plugs was examined using a Leica ~~S•4d confocal laser scanning microscope ~ • : : • a k r y p t o n - a r g o n laser (Leica, Heidel~ , : G : e ~ m a n y ) . Single optical sections approxi~]y 0i5 # m thick were collected at intervals of 1-p,m through the z-axis of each cell, and the z~eries data sets stored on magneto-optical disks. These data sets were processed in one of two ways using the Leica ScanWare software, which operates the confocal microscope. Single optical sections were selected from the mid-region of the cell and used for image assessment and densitomettic analysis. Alternatively, z-series projections were• •constructed which represent computer averaged assemblies of selected optical sections in the data set, and result in a perfectly focused image through the depth of the specimen. Images for publication were cropped, calibrated and labeled in Adobe Photoshop 2.5.1 and transferred to P a g e M a k e r 4.0.1 for assembly of the final plate which was printed on a Techtronix Phaser 440 dye sublimation printer. IMAGE A N A L Y S I S

Image analysis was used to establish a semiqua~ntitative profile of type VI collagen labeling i n t ~ i t y . To assess the surface distribution of type VI collagen, a z-series projection of half the images in a data set was prepared, and transferred to NIH Image shareware for processing using a surface plot module of the program. Additional intensity profile densitometry was performed on single optical sections from the mid-region of the cell to clarify •the circumferential distribution of type VI collagen in the vicinity of the chondrocyte. CELL D E N S I T Y A N A L Y S I S

To assess cell density, three c h o n d r o c y t e agarose plugs were sampled from the center of

BIOCHEMICAL ANALYSIS

Cell-associated type VI collagen was extracted from the agarose gel using 4M guanidinium chloride, 0.05M Tris, pH7.4, 2 4 h at 4°C [33]. Because this study was focused on the pericellular sequestration of type VI collagen during the development of the chondrocyte microenvironment, we did not examine the type VI collagen content of the culture medium. However, immunohistochemical data indicates a strong retention of type VI collagen in the pericellular microenvironment, suggesting the media fraction was likely to be low. Extracted type VI collagen was analyzed using an immunodot-binding assay [33] modified for use with an ECL detection system (Amersham, U.K.). Serial dilutions of the cell extracts and the corresponding protein standard were prepared in Tris-buffered saline, pH 7.2, (TBS) and transferred in volumes of 100 ~1 onto nitrocellulose membranes using a BIO-DOT microfiltration a p p a r a t u s (BioRad, U.S.A.). Nonspecific binding sites on the membrane were blocked by incubation with 1% BSA in TBS for 2 h at room temperature. The membranes were then treated for 2 h at room temperature with anti-type VI collagen antibody diluted 1.:100 in TBS containing 1% BSA. After washing with 0.5% Tween 20 in TBS (TBS-T), the membranes were treated with the ECL Western blotting detection kit as described in the manufacturers instructions (Amersham, U.K.). The membranes were incubated with donkey anti-rabbit Ig horseradish peroxidase-linked whole antibody for 1 h, rinsed with TBS-T once for 15 min and twice for 5 min each, and treated with equal volumes of detection solutions I and II for 1 min. The membranes were wrapped in plastic and exposed to X-ray film for times ranging from 30 s to 10 rain to achieve an optimized signal strength. Standard curves were prepared using type VI collagen extracted from calf intervertebral disc after enzyme digestion with chondroitin ABC lyase and streptomyces hyaluronidase [34]. Protein

278

Chang and Poole: Type VI collagen in chondrocyte cultures

standards ranged from 20-1000 ng/dot, and the s t a n d a r d curve showed a reasonably linear dose-dependent profile. Results IMMUNOHISTOCHEMISTRY

Immunolabeling and confocal microscopy revealed type VI collagen sequestered a r o u n d each chondrocyte within 24h of plating [Fig. l(a)]. Initially, a stippled staining p a t t e r n was observed on the surface of most cells in the mixed culture [Fig. l(a)]. However, over time, a small proportion of the chondrocytes consistently failed to sequester type VI collagen in the pericellular microenvironment, suggesting a unique subpopulation of chondrocytes which express low levels of type VI collagen. After 1 week in culture, the stippled sequestration became more uniform, single optical sections revealing a thin, discontinuous layer of type VI collagen surrounding the chondrocyte [Fig. l(b)]. By 6 weeks of culture, intensely stained bands of type VI collagen appeared woven around most cells, the limited staining between these bands

creating the impression of a 'cargo-net-like' enclosure a r o u n d the cells [Fig. 1(c)]. At the end of the culture period (12 weeks) considerable variability was observed in the s t r u c t u r a l organization of type VI collagen around individual cells. The most consistent p a t t e r n to emerge was a layered pattern, concentric rings of type VI collagen forming around m a n y of the chondrocytes [Fig. l(d)]. Type IV Collagen staining was consistently high in a broad band immediately adjacent to the chondrocyte with successive layers showing progressively weaker staining intensities and increased disorganization.

IMAGE

ANALYSIS

Analyses of surface plots from z-series projections obtained after 1 day culture are shown in Fig. 2(a), (b), and illustrate peaks of intense punctate staining on the cell surface with deep intervening trough s of minimal intensity. However, after 6 weeks culture, a marked decrease in the width and depth of the troughs was clearly evident [Fig. 2(c), (d)], while the peaks of staining intensity did not increase appreciably but tended

FIG. 1. Digital confocal images of chondrocytes cultured in agarose gel for different tiine periods and labeled with anti-type VI collagen antibody. (a) Z-series projection of a single chondrocyte after I day in culture. A stippled pattern of type VI collagen staining was clearly evident on the cell surface. Bar = 10 #m. (c) A single optical section from a: chondrocyte cultured for 1 week illustrating the thin, discontinuous layer of type VI collagen sequestered on the cell surface. Bar--10 #m. (b) Z-series projection showing a single chondrocyte after 6 weeks in culture. Intensely-stained bands of type VI collagen create the impression of a 'cargo-net-like' enclosure woven around the cell. Bar = 10 #m. (d) A single optical section through a chondrocyte after 12 weeks in culture showing concentric layers of type VI collagen with different labeling intensities. The pericellular layer adjacent to the cell surface consistently showed the strongest staining. Bar = 10 #m.

O s t e o a r t h r i t i s a n d C a r t i l a g e Vol. 4 N o . 4

279

d

FI~. 2. Image analysis of the surface distribution of type VI collagen sequestered by chondrocytes in agarose culture. (a) Z-series projection of" a single chondrocyte after 1 day in culture. The square box represents the area analyzed. Bar= t0 tan. (b) Surface plot analysis of (a) revealed peaks of intense punctate staining on the cell surface with deep intervening troughs of minimal intensity. (c) Z-series projection of a single chondrocyte after 6 weeks in culture showing the area analyzed. Bar = 10 tun. (d) Surface plot analysis of C showing that peaks of staining intensity tended to coalesce and form ridges of type VI collagen labeling over the cell surface. to coalesce to form ridges of intense labeling over the cell surface [Fig. 2(d)]. By the end of the cult.ure period a uniform intensity was achieved around the cell with the remaining troughs presenNng as broad and shallow depressions (r e sul~s :ii::i::~0t. shown). Dens~tometric analysis of the intensity profiles obtaine~.. ~rom single optical sections revealed a thin banit Ofdense staining on the cell surface after I. week's e~lture, while two or. three concentric peaks of.pt:0gressively weaker density were evident after 12 .weeks' culture (results not shown).

CELL DENSITY

Changes in chondrocyte density during the culture period are shown in Fig. 3. During the first 2--3 weeks of the culture, chondrocyte density doubled, indicating a primary phase of rapid cell proliferation. However, over the remaining 9 10

weeks of culture chondrocyte density stabilized, a 10 20% increase in cell density indicating a secondary phase characterized by a marked reduction in the rate of chondrocyte proliferation. Total cell numbers increased 2.6 times during the 12-week culture period.

BIOCHEMICAL

ANALYSIS

]'he immunodot binding analysis of cell associated type VI collagen is presented in F i g . . 4 . Clearly, adult ehondroeytes sequestered type VI collagen in a time dependent manner, characterized also by two distinct phases of activity. In the first month of culture, the rate of type VI collagen deposition was low, while the later 2 months were characterized by a steady increase in the rate of type VI collagen sequestration. By the end of the culture period (12 weeks), the pericellular deposition of type VI collagen increased fivefold

7O

I

IL-1 TREATMENT

6O

~ 50 ©

~ 40 rj 30

20

~ 0

I 20

,

I 40

,

L 60

~

I 80

~

I 100

Time (days)

FIG. 3. Chondrocyte density in agarose gel measured as cell number/unit area. Cell density increased rapidly during the first 2-3 weeks of culture and stabilized over the remaining 9-10 weeks of culture. Total cell number increased 2.6 times during the culture period. compared with day 1 cultures. On a per-cell basis, the type VI collagen deposition appeared to decrease during the first m o n t h of culture, while the net increase in deposition was 2.7-fold by the end of the culture period. The reason for this observation, as identified by cell tracker dye and immunohistochemistry, was t h a t one subpopulation of chondrocytes consistently sequestered low levels of type VI collagen, but were actively proliferating active during the first phase of culture. 3000

Discussion

2000

C~

i000 t

0

Immunohistochemical staining of IL-1 treated cultures is shown in Fig. 5(a) (b). Cell clusters typical of control cultures at 17 days were surrounded by a stippled type VI c o l l a g e n sequestration p a t t e r n [Fig. 5(a)]. While s i m i l a r to the p a t t e r n seen in Fig. l(a)-(b), the cells in these clusters frequently shared a narrow pericellular margin with neighboring chondrocytes. In contrast, substantial deposits of type VI collagen were observed a r o u n d the cells in clusters of IL-1 treated cultures [Fig. 5(b)]. In particular, each cell was associated with a distinct pericellular matrix rich in type VI collagen, the overall integrity of the cluster being m a i n t a i n e d by the integrated architecture of overlapping microenvironments [Fig. 5(b)]. The cell density of IL-I~ treated cultures was not significantly different from control cultures (data not shown), indicating t h a t IL-la did not stimulate the rate of chondrocyte proliferation over the 2-week incubation period. In contrast, the immunodot binding analysis showed t h a t IL-I~ treatment at concentrations of 0.5 and 5.0 ng/ml, both produced a marked increase in type VI collagen s e q u e s t r a t i o n (Fig. 6). Although little change was evident after 3 days exposure to IL-I~, a significant increase in type VI collagen deposition was observed at both 7 and 14 days of IL-la treatment (Fig. 6). No significant differences were identified between cultures treat,ed with 0.5 or 5.0 ng/ml IL-la.

20

40 60 80 100 Time (days) FIG. 4. Immunodot binding analysis of cell-associated type VI collagen. Two phases of type VI collagen deposition could be identified, a primary phase in the first 4 weeks characterized by low levels of type VI collagen deposition, followed by a secondary phase characterized by a steady increase in type VI collagen sequestration. Type VI collagen levels had increased fivefold at 12 weeks compared with day 1 culture.

Confocal immunohistochemistry in combination with specific biochemical analysis has shown t h a t articular chondrocytes cultured in agarose gel initiate type VI collagen deposition upon plating and continue pericellular matrix sequestration in a time dependent manner. In particular, confocal microscopy and image analysis revealed a thin, stippled layer of type VI collagen which forms on the cell surface during the first week of culture. After 6 weeks in culture, a more compact, 'cargo-net-like', structure progressively encapsulated the chondrocyte and ultimately formed a series of concentric layers which defined the collagenous pericellular matrix by 12 weeks of culture. Similar pericellular staining patterns have been reported for the proteoglycans and fibrillar collagens produced by chondrocytes in agarose [25], alginate [25, 35] a n d high-density cultures [36], but none of these studies have focused specifically on type VI collagen, which we

FI~. 5. Digital confocal images of chondrocytes cultured in agarose gel for 17 days and labeled with anti-type VI collagen antibody. (a) A single optical section through a typical chondrocyte cluster in culture without IL-I~ treatment. (Jells at the periphery of the cluster formed a strong stippled pattern of type VI collagen sequestration, while cells in the center showed a weak pericel]ular reaction. Bar = 10 ]lm. (b) A single optical section through a chondrocyte cIuster after 14 days exposure to IL-la (5.0 ng/ml). A nmrked increase in the width and intensity of type VI collagen sequestration was evident around cells in the cluster. Bar = 10 pm. consider an a c c u r a t e m a r k e r of c h o n d r o n morphology in adult tissue [3, 11]. Colocalization studies are now in progress using a range of matrix antibodies. These studies suggest t h a t the fibrillar collagens are more closely associated with the type VI collagen m i c r o e n v i r o n m e n t than the glyc osaminoglycans and proteoglycans, which have been found to have a significantly broader

t

•a

.

*

i~

.

looo

0

?

.

.

~

c,~: "~

~ 7 14 Time (days) Fia. 6. The immunodot binding analysis of cell associated type V1 collagen in cultures treated with IL-I~ at concentrations of 0.5 (~) and 5.0 (~) ng/ml. Comparison with untreated controls (m) reveals significant increases in type VI collagen deposition at both 7 and 14 days of IL-I:~ treatment (*P < 0.05). 3

pericellular distribution (Chang and Poole, in preparation). However, it should be emphasized t h a t a h e t e r o g e n e o u s population of a r t i c u l a r chondrocytes was used in these studies, and a distinct subpopulation of cells was identified which showed minimal type VI collagen sequestration. The possibility t h a t this subpopulation of cells may derive from the superficial layer, previously shown to yield c h o n d r o c y t e s with a limited capacity to form a pericellular matrix in vitro [31], is c u r r e n t l y being examined in more homogeneous cultures of superficial and deep layer chondrocytes. Specific biochemical analysis confirmed the immediate sequestration and progressive accumulation of type VI collagen in c h o n d r o c y t e agarose culture. However, the c o r r e l a t i o n between the cell-density data and analysis of cell associated type VI collagen suggests two distinct phases in the pericellular sequestration of type VI collagen. The primary phase, which occurred over the first 2-4 weeks of culture, was characterized by rapid cell proliferation (as measured by cell density) and a comparatively low rate of type VI collagen sequestration. In contrast, the secondary phase of activity was characterized by a marked r e d u c t i o n in the rate of c h o n d r o c y t e proliferation and an equally marked increase in type VI collagen sequestration. Thus, after 12 weeks of culture, cell density increased 2.6-fold, while pericellular type V[ collagen increased to five times the level of day

282

Chang and Poole: Type VI collagen in chondrocyte cultures

1 controls. Further studies are in progress which will attempt to correlate the rate of type VI collagen synthesis with the rate of pericellular sequestration. Previous data from cultured fibroblasts have also shown that type VI collagen was initially expressed during the later part of the proliferative phase, and that a large proportion of the threefold increase in newly synthesized type VI collagen was deposited into the cell-associated matrix [33]. A similar profile of type VI collagen mRNA expression has recently been reported in fibroblasts during wound healing [37], while our own data suggests that the initial proliferative phase was followed by a period of active type VI collagen sequestration to produce a complete pericellular matrix enclosing the chondrocyte. These results are consistent with the view that connective tissue cells such as fibroblasts and chondrocytes first initiate proliferation which is then followed by high rates of type VI collagen synthesis when the normal interactions between the cells and their extracellular matrix are destroyed or disrupted by mechanical or enzymatic perturbation. The mechanisms by which synthesized type VI collagen is sequestered and retained at the cell surface were not examined in this study. However, published reports suggest that a range of cell-surface receptors, including the integrins [14, 38], the hyaluronan receptor C D 4 4 [20,39], and the developmental receptor NG2 [16], could potentially mediate the interaction between the chondrocyte and type VI collagen [11]. A specific cell-surface localization of type VI collagen was identified in this study, and in conjunction with previous studies on isolated chondrons, suggests that type VI collagen could mediate both cell-matrix and matrix-matrix interactions critical for the maintenance of the pericellular superstructure and the macromolecular heterogeneity of the pericellular microenvironment in adult articular cartilage [3,11]. We speculate that type VI collagen, synthesized and secreted in a pre-assembled tetrameric form [40], is retained at the cell surface by integral membrane receptors and that the sequestration and lateral assembly of type VI collagen may lead to the progressive differentiation of a pericellular microenvironment around each chondrocyte. We argue that this adult chondrocyte-agarose gel culture system represents an appropriate model with which to examine the role of type VI collagen in the formation, maintenance and remodeling of the pericellular microenvironment typical of mature and osteoarthritic chondrocytes. IL-I~, an important mediator of osteoarthritis,

has been characterized in the synovial fluid of osteoarthritic joints [41, 42], and is known to be synthesized by both synoviocytes and chondrocytes in vivo and in vitro [43,44]. In the experiments reported here, IL-la stimulated the pericellular sequestration of type VI collagen, presumably by up-regulation of type VI collagen synthesis. Confocal microscopy confirmed the increased accumulation of type VI collagen in IL-la treated cultures, with some cells forming clusters similar to those previously seen in osteoarthritic cartilage labeled for type VI collagen (Poole, personal observation). It is now well established that IL-1 promotes cartilage catabolism by stimulating the production of prostaglandin E2 and neutral metalloproteinases including collagenase, stromelysin and plasminogen activator [24,45,46]. Simultaneously, IL-1 down-regulates proteoglycan production [25, 47] and modulates collagen synthesis [48-50]. In human chondrocyte cultures for example, IL-1 down-regulates the expression of type II and IX collagens, but increases the synthesis of type I and III collagens, signaling a marked change in the expression of the chondrocyte phenotype [22]. However, in contrast to the fibrillar collagens, studies on the effect of IL-1 .on type VI collagen synthesis are limited. Bathon and co-workers [51] recently demonstrated an IL-1 induced suppression of type VI collagen mRNA expression in cultured human synovial fibroblast-like cells. This observation is at variance with the reported increase in type VI collagen in the synovium of arthritic joints [52], and stands in marked contrast to the IL-1 induced increase in pericellular type VI collagen sequestration reported in this study. Thus, it is entirely possible that IL-1 regulation of type VI collagen metabolism may be in part cell specific. Irrespective of these arguments, the significant increase in pericellular type VI collagen sequestration induced by I1-1 is consistent with studies reporting elevated levels of type VI collagen in osteoarthritic cartilage [28-30], and clearly highlights an important functional role for type VI collagen during the pericellular remodeling associated with osteoarthritic degeneration [53]. One possible explanation for these results is that the IL-1 induced suppression of normal proteoglycan and fibrillar collagen synthesis is coupled with a stimulation of type VI collagen production to ensure continued cell-matrix integrity during osteoarthritic remodeling. Alternatively, type VI collagen may preferentially accumulate around the cell because it is endowed with a remarkable degree of resistance to the metalloproteinases induced by IL-1 [26, 27]. Type VI collagen however,

Osteoarthritis and Cartilage Vol. 4 No. 4 is catabolized by t he serine proteases [27]. IL-1 stimulates th e a c t i v i t y of tissue plasminogen a c t i v a t o r in c a r t i l a g e e x p l a n t cultures, while s i m u l t a n e o u s l y i n c r e a s i n g the mRNA expression of the serine p r o t e i n a s e inhibitor, pl as mi nogen a c t i v a t o r inhibitor-1 (PAL1) [54]. In fibroblasts, IL-1 stimulates a second pl as m i nogen a c t i v a t o r inhibitor, PAI-2 [55], while the p r o t e a s e nexin, which inhibits serine proteinases, is up- r egul at ed by IL-1 in b o th c h o n d r o c y t e and synovial fibroblast cultures [49]. It is possible, therefore, t h a t type VI collagen's r es is t ance to m e t a l l o p r o t e i n a s e attack, coupled with the p r o t e c t i v e effect of IL-1 i nduced s e r i n e p r o t e i n a s e inhibitors, provides an appropriate e n v i r o n m e n t for t he i ncr e a s ed p e r i c e l l u l a r a c c u m u l a t i o n o f type VI collagen. In summary, c o n f o c a l microscopy and image analysis h av e d e m o n s t r a t e d the spatial and temporal d if f er ent i at i on of type v i collagen in the pericellular m i c r o e n v i r o n m e n t of adult a r t i c u l a r c h o n d r o c y t e s c u l t u r e d long-term in agarose gel. C o n c u r r e n t biochemical analysis confirmed a fivefold in cr eas e in p e r i c e l l u l a r type VI collagen a c c u m u l a t i o n over the c u l t u r e period, and an IL-1 induced increase in the per i c e l l ul a r s e q u e s t r a t i o n of type VI collagen. This model is now being used to examine th e role of fibrillar collagens, glycoproteins, p r o t e o g l y c a n s and glycosaminoglycans during the d ev el opm ent and di f f er e nt i a t i on of the pericellular microenvironment, relating the c h a r a c t e r i s t i c di s t r i but i on of type VI collagen to t h a t of s t r u c t u r a l m a c r om ol e c ul e s previously identified in the c h o n d r o n s of n o r m a l and o s t e o a r t h r i t i c cartilage.

Acknowledgment We wish to t h a n k Mr M a r c Roos for t e c h n i c a l assistance during aspects of this study.

References 1. Poole CA, Flint MH, Beaumont BW. Chondrons in cartilage: ultrastructural analysis of the pericellular microenvironment in adult human articular cartilage. J Orthop Res 1987;5:509-22. 2. Poole CA. Structure and function of articular cartilage matrices. In Woessner JF, Howell DS, Eds. Joint Cartilage Degradation: Basic and clinical aspects. New York, Marcel Dekker 1993:1-35. 3. Poole CA, Ayad S, Schofield JR. Chondrons from articular cartilage (I). Immunolocalisation of type VI collagen in the pericellular capsule of isolated canine chondrons. J Cell Sci 1988;90:635-45. 4. Poole CA, Wotton S, Duance VC. Localisation of type IX collagen in chondrons isolated from porcine articular cartilage and rat chondrosarcoma. Histochem J 1988;20:567-74.

283

5. Smith GN, Hasty KA, Brandt KD. Type XI collagen is associated with the chondrocyte surface in suspension culture. Matrix 1989;9:186-92. 6. Poole CA, Honda T, Skinner SJM, Schofield JR. Chondrons from articular cartilage (II). Analysis of the glycosaminoglycans in the cellular microenvironment of isolated canine chondrons. Connective Tiss Res 1990;24:319-30. 7. Poole CA, Glant T, Schofield JR. Chondrons from articular cartilage (IV). Immunolocalisation of proteoglycan epitopes in isolated canine tibial chondrons. J Histochem Cytochem 1991;39:117587. 8. Poole CA, Gilbert RT, Ayad S, Plass AHK. Immunolocalisation of type VI collagen, decorin and fibromodulin in articular cartilage and isolated chondrons. Trans Orthop Res Soc 1993; 18:644. 9. Poole CA. Chondrons extracted from articular cartilage: methods and applications. In Maroudas A, Kuettner K, Eds. Methods in cartilage research. Academic Press, Ltd, London. 1990:78~83. 10. Ratcliffe A, Fryer PR, Hardingham TE. The distribution of aggregating proteoglycan in articular cartilage: comparison of quantitative immunoelectron microscopy with radioimmunoassay and biochemical analysis. J Histochem Cytochem 1984;32:193. 11, Poole CA, Ayad S, Gilbert RT. Chondrons from articular cartilage (V). Immunohistochemical evaluation of type VI collagen organisation in isolated chondrons by light, confocal and electron microscopy. J Cell Sci 1992;103:1101-10. 12. Burgeson RE. New collagens, new concepts. Annu Rev Cell Biol 1988;4:551-77. 13. Van der Rest M, Aubert-Foncher E, Dublet B, Eichenberger D, Font B, Gotdschmidt D. Structure and function of the fibril-associated collagens. Biochem Soc Trans 1991;19:820-4. 14. Aumailley M, Mann K, v o n d e r Mark H, Timpl R. Cell attachment properties of collagen type VI and A r g ~ l y - A s p dependent binding to its u2(VI) and a3(VI) chains. Exp Cell Res 1989;181:753-61. 15. Aumailley M, Specks U, Timpt R. Cell adhesion to type VI collagen. Biochem Soc Trans 1991;19:843-7. 16. Stallcup WB, Dahlin K, Healy P. Interaction of the NG2 chondroitin sulfate proteoglycan with type VI collagen. J Cell Biol 1990;111:3177-88. 17. Koller E, Winterhalter KH, Trueb B. The globular domains of type VI collagen are related to the collagen-binding domains of cartilage matrix protein and yon Willebrand factor. EMBO J 1989;8:1073-7. 18. Bruckner P, van der Rest M. Structure and function of cartilage collagens. Microscopy Research and Technique 1994;28:378-84. 19. Bidanset DJ, Guidry C, Rosenberg LC, Choi HU, Timpl R, Hook M. Binding of the proteoglycan decorin to collagen type VI. J Biol Chem 1992;267:5250~. 20. Kielty CM, Whitaker SP, Grant ME, Shuttleworth CA. Type VI microfibrils: evidence for a structural association with hyaluronan. J Cell Biol 1992;118:979-90. 21. McDevitt CA, Marcelino J. Intact type VI collagen binds to fibronectin. Trans Orthop Res Soc 1993;18:72.

284

Chang and Poole: Type VI collagen in chondrocyte cultures

22. Goldring MB. Degradation of articular cartilage in culture, regulatory factors. In Woessner JF, Howell DS, Eds. Joint Cartilage Degradation: Basic and Clinical Aspects. New York, Marcel Dekker, 1993:281-345. 23. Tyler JA, Bolls S, Dingle JT, Middleton JFS. Mediators of matrix catabolism. In: Kuettner KE, Schleyerbach R, Peyron JC, Hascall VC, Eds. Articular Cartilage and Osteoarthritis. New York, Raven Press, 1992:251-64. 24. Martel-Pelletier J, Zafarullah M, Kodama S, Pelletier J-M. I n vitro effects of interleukin 1 on the synthesis of metalloproteinases, TIMP, plasminogen activators and inhibitors in h u m a n articular cartilage. J Rheum 1991;18 $27:80-4. 25. Aydelotte MB, Raiss RX, Caterson B, Kuettner KE. Influence of interleukin-1 on the morphology and proteoglycan metabolism of cultured bovine articular chondrocytes. Connective Tis~ Res 1992;28: 143-59. 26. Okada Y, Shinmei M, Tanaka O, Naka K, Kimura A, Nakanishi I, et al. Localization of metalloproteinase 3 (stromelysin) in osteoarthritic cartilage and synovium. Lab Invest 1992;66:680-90. 27. Kielty CM, Lees M, Shuttleworth A, Woolley D. Catabolism of intact type VI collagen microfibrils: susceptibility to degradation by serine proteinases. Biochem Bioph Res Comm 1993;191:123(~6. 28. Ronzi6re M-C, Ricard-Blum S, Tiollier J, H a r t m a n n DJ, Garrone R, Herbage D. Comparative analysis of collagens solubilized from human foetal, normal and osteoarthritic adult articular cartilage, with emphasis on type VI collagen. Biochim Biophys Acta 1990;1038:222-30. 29. McDevitt CA, Pahl JA, Ayad S, Miller RR, Uratsuji M, Andrish JT. Experimental osteoarthritic articular cartilage is enriched in guanidine-soluble Type VI collagen. Biochem Biophys Res Comm 1988;157:250-5. 30. Ayad S, Brierley VH, Marriott A, Grant ME. Characterization of the collagens in normal and osteoarthritic human articular cartilage. Trans Orthop Res Soc 1991;16: 249. 31. Aydelotte MB, Kuettner KE. Differences between sub-populations of cultured bovine articular chondrocytes. I Morphology and cartilage matrix production. Connective Tiss Res 1988;18:205-22. 32. Poole CA, Brookes NH, Gilbert RT, Beaumont BW, Crowther A, Scott L, et al. Detection of viable and non-viable cells in connective tissue explants using the fixable fluoroprobes 5-chloromethylfluorescein diacetate and ethidium homodimer-1. Conn Tiss Res 1996;33:233-41. 33. Hatamochi A, Aumailley M, Mauch C, Chu M-L, Timpl R, Krieg T. Regulation of type VI collagen expression in fibroblasts. Effects of cell density, cell-matrix interactions, and chemical transformation. J Biol Chem 1989;264:3494-9. 34. Wu JJ, Eyer DR, Slayter HS. Type VI collagen of the intervertebral disc. Biochem J 1987;248:373-81. 35. Mok SS, Masuda K, H~iuselmann HJ, Aydelotte MB, Thonar EJMA. Aggrecan synthesized by mature bovine chondrocytes suspended in alginate. J Biol Chem 1994;269:33021-7. 36. Ruggiero F, Petit B, Ronziere MC, Farjanel J, H a r t m a n n DJ, Herbage D. Composition and organization of the collagen network produced by

fetal chondrocytes cultured at high density. J Histochem Cytochem 1993;41:867-75. 37. Oono T, Specks U, Eckes B, Majewski S, Hunzelmann N, Timpl R, et al. Expression of type Vi collagen mRNA during wound healing. J Invest Dermatol 1993;100:329-34. 38. Salter DM, Hughes DE, Simpson R, Gardner DL. Integrin expression by human articular chondrocytes. Br J Rheumatot 1992;31:231-4. 39. McDevitt CA, Marcelino J, Tucker L. Interaction of type VI collagen with hyaluronan. FEBS Lett 1991;294:167-70. 40. Timpl R, Engel J. Type VI collagen. In Mayne R, Burgeson RE, Eds. Structure and Function of Collagen Types. Orlando, FL, Academic Press. 1987:105-43. 41. Wood DD, Ihrie EJ, Dinarello CA, et al. Isolation of an interleukin-l-like factor from joint effusions. Arthritis Rheum 1983;26:975~83. 42. Morris EA, McDonald BS, Webb AC, et al. The identification of interleukin-1 in equine osteoarthritic joint effusions. Am J Vet Res 1990;51:59 64. 43. Towte CA, Trice M, Ollivierre F, et al. Regulation of cartilage remodelling by IL-I: evidence for autocrine synthesis of IL-1 by chondrocytes. J Rheum 1987;14:11-13. 44. Wood DD, Ihrie EJ, Hamerman D. Release of interleukin-1 from human synovial tissue in vitro. Arthritis Rheum 1985;28:853-62. 45. McGuire-Goldring MB, Meats JE, Wood DD, et al. I n vitro activation of human chondrocytes and synoviocytes by a human interleukin-l-like factor. Arthritis Rheum 1984;27:654~2. 46. Gowen M, Wood DD, Ihrie EJ, et al. Stimulation by human interleukin-1 of cartilage breakdown and production of collagenase and proteoglycanase by human chondrocytes but not by human osteoblasts in vitro. Biochim Biophys Acta 1984;797:186-93. 47. Benton HP, Tyler JA. Inhibition of cartilage proteoglycin synthesis by interleukin 1. Biochem Biophys Res Commun 1988;154:421-8. 48. Goldring MB, Birkhead J, Sandell LJ, Kimura T, Krane SM. Interleukin-1 suppresses expression of cartilage-specific types II and IX collagens and increases types I and III collagens in human chondrocytes. J Clin Invest 1988;82:2026-37. 49. McGuire-Goldring MB, Krane SM. Release of protease nexin by synovial cells and chondrocytes is stimulated by mononuclear cell factor. Arthritis Rheum 1984;174:251-7. 50. Beresford JN, Gallagher JA, Gowen M, et al. The effects of monocyte-conditioned medium and interleukin-1 on the synthesis of collagenous and non-collagenous proteins by mouse bone and human bone cells in vitro. Biochim Biophys Acta 1984;801:58-65. 51. Bathon JM, Hwang JJ, Shin LH, Precht PA, Towns MC, Horton WE. Type VI collagen-specific messenger RNA is expressed constitutively by cultured human synovial fibroblasts and is suppressed by interleukin-1. Arthritis Rheum 1994; 37:1350-6. 52. Wolf J, Carsons SE. Distribution of type VI collagen expression in synovial tissue and cultured synoviocytes. Ann Rheum Dis 1991;50:494-6. 53. Poole CA, Matsuoka A, Schofield JR. Chondrons

O s t e o a r t h r i t i s a n d Cartilage Vol. 4 No. 4

from articular cartilage (III). Morphologic changes in the cellular microenvironment of chondrons isolated from osteoarthritic human and canine cartilages. Arthritis Rheum 1991;34:22-35. 54. Treadwell BV, Pavia M, Towle CA, Cooley VJ, Mankin HJ. Cartilage synthesizes the serine protease inhibitor PAI-I: support for the involve-

285

ment of serine proteases in cartilage remodelling. J Orthop Res 1991;9:309-16. 55. Michel JB, Quertermous T. Modulation of mRNA levels for urinary- and tissue-type plasminogen activator and plasminogen activator inhibitors 1 and 2 in human fibroblasts by interleukin 1. J Immunol 1989;143:890-5.