- Email: [email protected]
Acceleration of Cartilage and Bone Differentiation on Collagenous Substrata
Collagen ReI. Res. Vol. 7/1987, pp. 351-370
Acceleration of Cartilage and Bone Differentiation on Collagenous Substrata GILA MAORI, KLAUS VON DER MARK2, HARI REDDI 3, DICK HEINEGARD 4 , AHNDERS FRANZEN4 and MICHAEL SILBERMANN 1 Laboratory for Musculoskeletal Research, The Rappaport Institute for Research in the Medical Sciences, Faculty of Medicine, Technion, Haifa, Israel, 2 Laboratory for Connective Tissue Research, The Max Planck Institute for Biochemistry, Martinsried, FRG, 3 Bone Cell Biology Section, National Institute of Dental Research, N.I.H., Bethesda, MD, USA and 4 Department of Physiological Chemistry, University of Lund, Lund, Sweden. 1
Abstract Chondroprogenitor cells of newborn murine mandibular condyles were cultured on top of collagen sponges for up to 18 days. After 24 h in culture, new chondroblasts developed which subsequently matured showing signs of hypertrophy, while the extracellular matrix revealed positive reactivity for type II collagen, cartilage proteoglycans and mineralization. Light and electron microscopy examinations showed signs of new osteoid formation, a feature that was preceeded by positive immunohistochemical reaction for type I collagen, fibronectin and bone specific sialoprotein. A close temporal and spatial association was noted between the development of mature, mineralized cartilage and new osteoid. The differentiation of new cartilage and bone cells was linked to an increased activity of DNA synthesis and cellular proliferation. The de novo bone formation was accompanied by increasing rates of alkaline phosphatase activity and uptake of [45 Ca] features that were found to be tightly correlated to each other. The collagen substrata appeared also to facilitate the migration of cells, their replication and their subsequent differentiation to their respective cellular lineage. Hence, collagen sponges in vitro appear to serve as a promising substrata for culture systems involved with the growth and differentiation of mineralizing tissues such as cartilage and bone. Key words: bone, cartilage, collagen, differentiation, progenitor cells.
Introduction Local and systemic factors regulate the growth, differentiation and maturation of skeletal tissues. The vertebrate skeletal system has the potential for considerable regeneration and remodelling even after growth ceases, as exemplified by the multi-step sequential process of fracture healing (Reddi, 1985). A thorough analysis of the local
352
G. Maor et al.
and systemic factors that influence cartilage and bone differentiation requires the use of an appropriate in vitro experimental model. The extracellular matrix of mineralizing tissues such as cartilage and bone may well be of major importance in defining the local microenvironment for cell growth and differentiation. The principal component of extracellular matrix in cartilage and bone is collagen, which functions as a substratum for the progenitor cells. Most normal animal cells must attach to a surface such as collagen in order to spread and proliferate - a property termed "anchorage dependence" by Stoker et al. (1968). The cell surface glycoprotein fibronectin mediates the attachment of cells to collagen (Kleinman et aI., 1981). Further, the adhesion to matrix may play an important role in determining cell responsivity to other regulatory factors such as matrical proteoglycans (Reddi, 1985). From the foregoing, it is evident that collagen could very possibly be a repository for factors that modify various phases of cartilage and bone development: chemotaxis, proliferation and differentiation. Hence, the purpose of the present study was to elucidate whether collagen may aid in the promotion of cartilage and bone differentiation in a tissue culture system using skeletal progenitor cells. Materials and Methods Collagen substrata
This study utilized collagen sponges (U257-16DS, Helitrex Inc., Princeton, NJ) that were made of type I collagen purified from bovine achilles tendon. The above material comprised of long fibrils that were depleted of telopeptides but were left with endogenous and exogenous crosslinks. The density of the sponges was about 0.09, while over 90% of the sponge volume was open space with the fibrillar strands of collagen organized into a lattice network. As the collagen sponges are supplied in a dehydrated state they swell when immersed in tissue culture media. Swelling normally occurred as a 2- to 3-fold increase in thickness with minor alterations in the other dimensions. Also, during the production process collagen sponges acquire different textures on opposite sides. One surface tends to be fairly smooth and regular whereas the other one becomes more patterned. The two sides differ somewhat in their microstructure: rough and smooth, yet the typical primary and secondary banding patterns of native collagen are preserved. Culture system
ICR mice, 1- to 2-days old, were anesthetized with ether. Using a surgical microscope, the mandibular condyles were aseptically dissected away from the mandibles, cleaned of all soft tissue and of the underlying bone and placed in freshly prepared bovine pancreas 0.1 % trypsin type III, (Sigma, St. Louis, MO) solution (dissolved in BGJb medium; Gibco, Grand Island, NY) for 10 min at 37°C. Tissues were then transferred into a fresh medium containing 10% fetal calf serum, where the perichondrial envelope surrounding the cartilage was gently separated from the cartilaginous core. The separated strands of tissue, consisting almost exclusively of cartilage progenitor cells (Weiss et aI., 1986) were transferred onto 16 mm-wide collagen sponges cemented to stainless steel grids (Michigan Dynamics, Garden City, MI) that were placed in plastic disposable culture dishes (20 mm; Falcon, Irvine, CAl. The culture medium, pH 7.2 - 7.4, consisted of BGJb medium (Fitton-Jackson modifica-
Collagen and Bone Differentiation
353
tion) supplemented with 10% fetal calf serum, ascorbic acid (300 ~g/ml) penicillin (100 U/mi), and streptomycin (100 ~g/ml). Cultures were carried out for as long as 18 days in a humidified incubator at 3rC in an atmosphere of 5% CO r 95% air. The medium was changed every 48 h, and specimens were obtained after 1, 2, 3, 4, 6, 9, 14 and 18 days in culture.
Light and electron microscopy Tissues that were designated for structural studies were initially washed in Hank's balanced salt solution and subsequently fixed in 3% glutaraldehyde in 0.1 M cacodylate buffer, pH 7.2, for 90 min postfixed in 1 % OS04 for 60 min, dehydrated and embedded in Epon 812. Thick sections (1 ~m) were stained with 1 % toluidine blue or with Pearson's silver gelatin stain. Adjacent thin sections (600-700A) were cut on LKB Ultratone III, mounted on No. 300 copper grids and stained for 10 min in saturated aqueous uranyl acetate followed by staining for 10 min in 0.2% lead citrate before examination with a Jeol 100B electron microscope, operating at 60 kV.
Indirect immunofluorescence studies Explants what were designated for immunofluorescence studies were not fixed but were washed and immediately frozen in liquid nitrogen (-186°C) and cut in a cryostat (-25°C), and the sections (8 ~m thick) were mounted on microscope slides. All the frozen sections were initially treated with 0.3 M ethylenediaminetetracetic acid (EDTA) pH 7.5, for 30 min at room temperature, followed by 30 min treatment with hyalase (ovine testes hyaluronidase, Serva # 25118), 1 mg/ml in phosphate-buffered saline (PBS), pH 7.2, at room temperature. The following antibodies were used: rabbit anti-rat type I collagen; guinea pig anti-chick type II collagen; rabbit anti-rat cartilage proteoglycans; rabbit anti-chick cartilage anchorin ClI; and rabbit anti-calf bone specific sialoprotein. Following incubation with the above antibodies diluted 1:5 for 30 min at room temperature in a moist chamber, the sections were rinsed X 3 with PBS and were subsequently reacted with fluorescein isothiocyanate (FITC)-conjugated goat anti-rabbit gamma globulin (1 :20) or with rabbit anti-guinea pig gamma globulin (Behringwerke, Marburg, FRG) (1:20) for 30 min at room temperature. Sections incubated with non-immune serum or with FITC-conjugated rabbit anti-guinea pig gamma globulin alone served as controls.
Histochemistry For the demonstration of alkaline phosphatase activity in explants the azo-coupling method was used whereby frozen sections were incubated in media comprising of 0.2 mM of naphthol AS-TR phosphate (Sigma), 100 mM Tris buffer and 1.6 mM of fast red violet LB, pH 8.4, for 20 min at room temperature. As controls, sections were incubated in media from which the substrate had been omitted; in media containing 5 mM of levamisole or in media that were preheated to 60°C for 20 min. For the localization of active sites of mineralization cultures were supplemented with 10 mg/ml of oxytetracycline (Sigma) for the final 24 h of incubation. The specimens were then fixed with glutaraldehyde alone, embedded in Epon 812 and were examined in a fluorescent microscope.
354
G. Maor et al.
[3 Hj- Thymidine autoradiography
Cultures were supplemented with 5 !-ICi/ml of eH]-methyl thymidine (4.8 Ci/mmol, Nuclear Res. Center, Israel), for the last 24 h of the incubation period. The explants were then washed with Hank's buffer solution containing 1 mM of thymidine, fixed in glutaraldehyde and embedded in Epon. Section 2 !-1m thick were mounted on glass slides, coated with nuclear track emulsion (NTB-2, Kodak, Rochester, NY) and placed in light-tight black boxes. After 3 weeks of exposure at 4°C, autoradiographs were developed in D-170 developer (Kodak) at 18°C, fixed and lightly stained with toluidine blue. rHj-thymidine incorporation into DNA
Explants were transferred into a cold solution of 10% trichloroacetic acid (TCA) with 1 mM thymidine and were homogenized at 4°C with a polytron homogenizer (Type PT 10/35 Kinematica). Following further washing with 10% TCA the precipitates were hydrolyzed at 90°C for 20 min in 1.5 ml10% TCA and aliquots were used to determine TCA insoluble counts. The remaining 1 ml of the hydrolyzate was used to determine the DNA content of the specimens using a modification of Burton's method (Burton, 1968). Sigma's calf thymus DNA was used as standard. Biochemical assay for alkaline phosphatase
Explants were homogenized in ice-cold 0.15 M NaCl containing 3 mM NaHCO~, pH 7.4, centrifuged and the supernatant was assayed for alkaline phosphatase, pH 9.3, with p-nitrophenyl phosphate (0.01 M, Sigma 104) as substrate (Halpern et ai., 1972). Protein was determined in the enzyme extracts (Lowrey et ai., 1951). 45Ca uptake
45CaClz (17.49 mCi/mgj Amersham, U.K.) was added at a dose of 2 !-ICi/ml to the culture medium 18 h prior to the termination of the cultures. Explants were homogenized in buffer as described above, and after the separation of the supernatant for enzyme assays the sediments were washed with 0.1 M CaCl 2 in 0.05 M Tris-HCI, pH 7.4, at 25°C for 30 min in order to remove the exchangeable [45 Ca], and were then hydrolyzed in 0.5 M HCI for at least 2 h at 25°C. Aliquots were used to determine [45 Ca] counts. Means and standard erors were calculated from all numerical findings and the comparison of the means were determined by student's t-test. Correlation tests utilized the least square regression line and standard error of estimate.
Results In newborn mice the cartilage of the mandibular condyle exhibits four distinct zones of cell populations: articular, chondroprogenitor, chondroblasts and hypertrophic chondrocytes (Weiss et aI., 1986). As indcated, our culture specimens were comprised of groups of progenitor cells that revealed strong basophilic properties following staining with toluidine blue (Fig. 1). The isolated tissue reacted positively for type I collagen
Collagen and Bone Differentiation
355
Fig. 1. The appearance of an explant immediately following the trypsinization procedure and its separation from the condylar core. Note the homogeneous population of basophilic cells that comprise condylar progenitor zone. Toluidine blue, X 240.
Fig. 2. Frozen section through an explant immediately following its separation. The tissue was reacted with antibodies against type I collagen. Note the strong reactivity for this type of collagen. X 384.
356
G. Maor et al.
Fig. 3. Section through an explant after 24 h in culture. The tissue attained a form of a conglomerate in which the innermost portion reveals the initial structural signs of cartilagelike tissue. The latter is represented by increasing amounts of extracellular matrix separating between the chondroblast-like cells (C). Toluidine blue, x 240.
Fig. 4. Frozen section through an explant after 24 h in culture and treated with antibodies against cartilage proteoglycans. Note the strong reactivity at the periphery of the newly formed chondroblasts as well as in the extracellular matrix. X 384.
Collagen and Bone Differentiation
357
Fig. 5. Portion of explant that has been cultured for 48 h on top of collagen sponge. Note the distinct nodules of cartilage that is underlied by a thick layer of fibroblast-like cells. The latter attach the newly formed cartilage onto the collagen sponge. Toluidine blue, X 480.
Fig. 6. A section through the newly formed cartilage shown in Figure 5 (48 h in culture). The cell shows an intracellular accumulation of glycogen (G), strands of rough endoplasmic reticulum (arrows) and mitochondria (m). The matrix exhibits typical collagen fibrils along with electron dense granules matrical proteoglycans. x 16,000.
358
G. Maar et al.
Fig. 7. Autoradiograph of an explant after 48 h in culture and labelled in vitro with [3H]_ thymidine. Note the marked labelling, especially along the peripheral sites of the newly formed nodules of cartilage (C). Tissue lightly stained with toluidine blue, x 240.
Fig. 8. Portion of an explant that has been cultured for 3 days. The cartilaginous nodule is already deeply embedded with the collagen sponge. Note the gradation of cartilage cell maturation from younger chondroblasts at the periphery of the nodule (C) to mature hypertrophic chondrocytes (H) within the inner part of the nodule. In addition, a dense population of mesenchyme-like cells (M) is seen adjacent to the well developed cartilage. Toluidine blue, x 240.
Collagen and Bone Differentiation
359
Fig. 9. General view of cartilage after 3 days in culture following reaction with antibodies against cartilage proteoglycans. It can be seen that each individual cell is highly reactive for this antigen. Frozen section, x 154.
Fig. 10. A similar section to that shown in Figure 9 but was reacted with antibodies against cartilage anchorin cn. A positive reactivity is noted throughout. x 154.
360
G. Maar et al.
Fig. 11. A similar section to those shown in Figures 8 and 9 but was reacted with antibodies against type I collagen. Only the peripheral portions of the explants reveal positive reactivity, whereas the newly formed cartilage (C) is negative. X 96.
Fig. 12. Section through explant that was cultured for 8 days and received a single dose of oxytetracycline for 24 h. The fluorescent activity (indicative of matrix mineralization) appears in the inner portion of the tissue. x 240.
Collagen and Bone Differentiation
361
Fig. 13. Frozen section obtained after 3 days in culture and reacted with antibodies against bone specific sialoprotein. Note the intense intracellular reactivity in the newly differentiated cells. x 154.
Fig. 14. Frozen section through the strands of collagen sponge adjacent to the explant after 6 days in culture. The section was reacted with antibodies against bone specific sialoprotein and shows an intense reactivity along the strands of the collagen sponge. All control sections were found negative. x 384.
362
G. Maor et al.
(Fig. 2) but not for type II collagen or for cartilage proteoglycans (not shown). After 24 h in culture, the explants appeared as condensed conglomerates of basophilic cells with a relatively small amount of intercellular matrix between them (Fig. 3). By that time the explant already revealed strong reactivity for cartilage proteoglycans (Fig. 4). By 48 h distinct nodules of cartilage were evident that were attached to the collagen sponge via a dense layer of fibroblast-like cells (Fig. 5). The newly formed young chondroblasts contained the typical lakes of glycogen as well as a distinct system of rough endoplasmic reticulum and a small number of rounded mitochondrial profiles that were randomly dispersed throughout the cytoplasm (Fig. 6). Fine collagen fibrils and proteoglycan granules were always encountered in the extracellular matrix (Fig. 6). By the same time, many cells, especially along the cartilage periphery, were labelled with eH]thymidine as shown with autoradiography (Fig. 7). After 3 days in culture, the cartilage nodules showed large numbers of mature, hypertrophic chondrocytes that, for the most part, were located within the inner portions of the cultures (Fig. 8) and showed strong reactivity for cartilage proteoglycans (Fig. 9) and anchorin CII (Fig. 10). The periphery of this culture reacted positively for type I collagen (Fig. 11). Moreover, these cartilage nodules were found to undergo an active phase of mineralization as visualized via vital staining with oxytetracycline (Fig. 12). Of great interest was the finding that by 3 days almost all of the cells within the explants also reacted positively for bone specific sialoprotein (Fig. 13). With further progression of the culture period (6 days) the fibrilar strands of collagen within the sponges exhibited an intense reactivity for both bone sialoprotein and fibronectin along their surfaces (Fig. 14). Concomitantly more cells appeared to have penetrated into the sponges inner mass (Fig. 15) and were undergoing an active phase of cell proliferation (Fig. 16). By 9 days new loci of differentiation developed adjacent to the existing foci of mature mineralized cartilage. The latter foci were inhabited by cells that differed structurally from the foregoing cartilage cells and were embedded in a well developed extracellular matrix (Fig. 17). The latter matrix which reacted positively for bone specific sialoprotein (not shown) was also found to be occupied by newly formed collagen fibers that reacted positively for type I collagen, but not for type II collagen and appeared to be tightly attached to the sponges' original collagen strands (Fig. 18). in addition, the matrix revealed multiple matrix vesicles in between collagen fibers (Fig. 19) a feature that is characteristic of embryonic woven bone. Also, the cultures revealed the appearance of multinucleated giant cells (Fig. 20) as well as phagocyte-like cells that appeared to be involved in the degradation of the sponges' collagen (Fig. 21). The cellular population in the newly formed nodule of osteoid reacted positively for alkaline phosphatase (Fig. 22). All the control tests for this enzyme revealed negative results. Biochemical determinations for alkaline phosphatase activity indicated a relative low rate of activity during the first 4 days in culture with a substantial increase thereafter (Fig. 23). A very similar pattern was noted with regard to the uptake of 4S [ Ca] (Fig. 23). A very high degree of correlational relationship was found between alkaline phosphatase activity and uptake of [4S Ca] (r = 0.96). Measuring the rate of [3H]-thymidine incorporation into DNA indicated the existence of two distinct peaks: (1) on day 2 and (2) on day 6 (Fig. 24). The first peak coincided with the differentiative phase of cartilage whereas the second peak occurred at about the same time interval of new bone formation.
Collagen and Bone Differentiation
363
-I'
'1 - ~ \ . ·:..~'.I1. ./~~ ....,10<,...,,""""'" Fig. 15. Electron micrograph showing an osteoblast (OB) in close contact with the collagen sponge (CS). At the lower portion of the photomicrograph newly formed collagen fibers (F) can be seen. x 10,000,
Fig, 16, Autoradiograph [3Hl-thymidine-labelled representing a deeper zone within the collagen sponge - farther away from the original explant, Note the immense labelling of cells that have emigrated out of the explant and migrated along the sponge's fibrils (arrows), x 480, 25
Collagen 715
364
G. Maar et al.
Fig. 17. Section through a newly formed osteoid in an explant that was cultured for 9 days. The cells resemble bone cells to a great extent while the matrix (M) appears rich in fibrillar material. Pearson's silver stain, X 288.
Fig. 18. A higher magnification of a portion of a cell and its extracellular matrix in an area shown in Figure 17. Note the large number of collagen fibers that are closely attached to the cell as well as to the sponge's fibrillar material (5). The newly formed collagen fibers (COL) were found to react positively for type I collagen and fibronectin but not for type II collagen. X 9800.
Collagen and Bone Differentiation
365
Fig. 19. A section through the extracellular matrix of the newly formed osteoid. Note the multiple matrix vesicles (arrows) in between the collagen fibers, a characteristic feature of embryonic, woven bone. These vesicular structures serve as the initial locus of mineralization in this tissue. X 9800.
Fig. 20. The appearance of a giant multinuclear cell located at a peripheral site of the mineralized cartilage. This cell resembles a chondroclast and could be involved in the resorption of mineralized matrix. Toluidine blue, x 384.
366
G. Maor et al.
Fig. 21. Electron micrograph of cell within the collagen sponge that appears to be involved in degrading the sponge's fibrils (SF). The cell reveals large number of lysosome-like organelles (arrows) that presumably take part in the digestion of the phagocytized material. x 9800.
Fig. 22. Frozen section through the newly formed osteoid that was reacted for alkaline phosphatase. Pronounced reactivity is noted within most of the cells. x 240.
367
Collagen and Bone Differentiation 4 5 CA UPTAKE AND ALKALINE PHOSPHo ACTIVITY 10
60
9
~
CI.
'0"
~
Q.
•c 1\1
~
8
0 0
.
,..
n
a 3
7
......
6
Z
5
01
4
0
::s.
..'"
......
'C
l> C c: Z a l> DI '-"
.
.. 0
.. Jt: It
0
2
:::J
0
(")
CD
3
c
~ (,II
0
2
4
6
8
10
0
Days in Culture
Fig. 23. Kinetics of [4S Ca] uptake and alkaline phosphatase activity in explants cultured on collagen sponges. Both parameters reveal a marked and steady increase following 4 days in culture. Each mean was based upon 6 determinations . • - . = [4S Ca] uptake; 0 - 0 = alkaline phosphatase activity.
3 H THYMIDINE INCORPORATION INTO DNA
... .0 .
80
0
Col
70
c~
-0(
QlZ
cO
60
"001 .- :::l
E,
>OE
.c . I-CI. Col
50
40
30
0
'0
Days in Culture
Fig. 24. Kinetics of [3H]-thymidine incorporation in explants cultured on collagen sponges. Prominent peaks are noticed on day 2 and 6 at which time the differentiation of cartilage and bone cells took place respectively. Each mean was based upon 6 determinations.
368
G. Maor et a!.
Discussion The differentiation of cartilage and bone cells culminates in the formation of a collagenous extracellular matrix with non-collagenous components that ultimately undergoes mineralization. In this regard mineralized tissues present a formidable challenge to understand the molecular and cellular basis of mineralization. Analysis of this problem is in part simplified by the use of appropriate in vitro models of cartilage and bone differentiation, maturation and mineralization. Although cartilage and bone cells have been isolated from embryonic organs (Silbermann and Maor, 1984), the process of progenitor cells differentiation in vitro and the subsequent formation of specific matrices needs further work-up. One possible reason for the current situation may relate to the lack of optimal environment conditions that are conductive to the differentiation of mineralizing tissues. During the course of systematic investigations on bone differentiation in organ and tissue cultures of the mandibular condyle (Silbermann and Reddi, 1985; Weiss et a!., 1986; Silbermann et a!., 1987), it became apparent that collagen substrata might serve as an important promoter to bone cell differentiation and mineralization. In the present study we were able to demonstrate that tissue culture of newborn condylar progenitor cells not only maintain the capacity to express their chondrogenic phenotype, but also promote bone cell differentiation when cultured on collagen sponges. In a very short period of time, the collagenous microenvironment succeeded in elaborating the cascade of events involving progenitor cell proliferation, cartilage cell differentiation, cartilage matrix mineralization and bone cell differentiation. The present findings indicate that if condylar progenitor cells are cultured on top of collagen substrata they increase the ability to synthesize DNA and proliferate. Also, it became evident that during the first 24 hours in culture, almost all of the progenitor cells in the explant attained chondroblastic phenotype showing positive reactivity for cartilage proteoglycans. The differentiation of cartilage cells and the elaboration of cartilage specific extracellular matrix capable of undergoing mineralization was succeeded by osteoid formation already by 6 days in culture. It became very clear that the development of new osteoid was intimately associated with the differentiation of cartilaginous tissue, a fact that had been alluded to by Scott-Savage and Hall (1980). The foregoing findings indicate that the collagenous scaffold facilitated the differentiation of cartilage and bone progenitor cells and promoted the laying down of mineralized extracellular matrix in vitro. In a way the latter microenvironment appears very reminiscent of the normal ecology of cartilage and bone cells. As already mentioned in the Introduction, the growth and differentiation of cells is anchorage dependent and it is the extracellular collagenous framework that provides the anchorage in vivo. It seems, therefore, reasonable to assume that the collagen sponge provides the necessary milieu of cell-cell and cell-matrix interactions essential for the function of mineralizing cells. Moreover, the collagen sponge may concentrate cartilage and bone essential macromolecules such as cartilage proteoglycans, anchorin CII, bone specific sialoprotein (Franzen and Heinegard, 1985) and osteopontin (Oldberg et aI., 1986), in such a way that they may attain critical concentrations to initiate tissue specific morphogenesis and mineralization. It is also likely that local growth factors such as bone-derived growth factor (BDGF), cartilage-derived factor, an ll-kDa somatomedin-C-like substance, cationic cartilage-derived growth factor (CDGF), bone morphogenetic protein, an 18.5-kDa protein with bone-indictive activity, transforming growth factor beta, skeletal growth factor and insulin-like growth factor I (Hauschka et aI., 1986) that
Collagen and Bone Differentiation
369
might be released by the condylar progenitor cells and may be trapped by ionic and/or hydrophobic interactions in the collagen sporlge. Cells vary in their attachments to collagen matrices. Many cells will adhere just as readily to polystyrene as to collagen. However, some types of cells such as condylar progenitor cells exhibit a definite requirement for collagen (unpublished data), preferentially binding to the trihelical portion or interacting with the end chains. In addition, cell attachment may involve fibronectin or other factors, which may have been supplied by the cultured cells. It is difficult to predict the preferential sponge type for a particular cell line or tissue type. In our hand the U257-16DS sponge was found appropriate for cartilage and bone cells adherance, growth, differentiation, maturation and matrix mineralization. The collagen sponges were found to undergo degradation with time. The rate of degradation depends among others on pH, ionic strength and the amount of collagenase and other proteases elaborated by the cells. Sponges composed of more highly crosslinked collagen appear to survive for longer periods of time, potenially extending into weeks or months. The condylar progenitor cells did not appear to exert any appreciable force on the collagen matrix, as we did not face collapse of sponges. In conclusion, the present in vitro tissue culture system may have far-reaching implications for answering some of the following questions: (1) Which growth factors playa role in the growth and development of the skeleton? What regulates their synthesis, and how are are their effects mediated? (2) Does loss of progenitor cells function contribute to the development of skeletal anomalies? If so, why does the loss occur? How can the loss of function be corrected? (3) What regulates the synthesis of noncollagenous proteins in cartilage and bone matrix? And what is the latters' function? (4) How is the action of hormones such as glucocorticoids, sex hormones, parathyroid hormone and 1,25-dihydroxyvitamin D3 mediated on cartilage and bone forming cells? (5) What is the physiologic role, if any, of lymphokines and cytokines on skeletal tissues differentiation? How are their effects mediated? (6) Is the above in vitro system suitable for the growth of human progenitor cells from both normal subjects and patients? Hence, the ability of collagen substrata to accelerate cartilage and bone differentiation in vitro demonstrated in this study is encouraging as it seems feasible that additional factors such as fibronectin, laminin and bone growth factors could be incorporated in such collagen sponges. Such interactions may fasten the differentiation processes via increasing the affinity of the growing cells onto their substrata. Acknowledgements This study was supported in part by the Fund for the Promotion of Research at the Technion, Grant 180-650, and by the European Molecular Biology Organization, Grant 4448. References Burton, K.: Methods in Enzymology 12: 163-166, 1968. Franzen, A. and Heinegard, D.: Isolation and characterization of two sialoproteins present only in bone calcified matrix. Biochem. J. 232: 715-724, 1985. Halpern, E. P., Rosoff, S. ]. and Weiner, S.: Use of p-nitrophenyl phosphate dicyclohexylammonium salt as substrate for alkaline phosphatase determination on the SMA 12/60. Clin. Chern. 18: 593-594, 1972.
370
G. Maor et al.
Hauschke, P. V., Mavrakos, A. E., Iafrati, M. D., Doleman, S. E. and Klagsburn, M.: Growth factors in bone matrix. J. Bioi. Chern. 261: 12665-12674, 1986. Kleinman, H. K., Klebe, R.]. and Martin, G. R.: Role of Collagenous matrices in adhesion and growth of cells. J. Cell. BioI. 88: 473-485, 1981. Lowry, O. H., Rosenbrough, N. J., Farr, A. L. and Randall, R. J.: Protein measurement with the folin phenol reagent. J. BioI. Chern. 143: 265-275, 1951. Mollenhauer, ]., Bee,]. A., Lizarbe, M. A. and von der Mark, K.: Role of anchorin CIl, a 31,000-mol-wt membrane protein in the interaction of chondrocytes with type II collagen. J. Cell. BioI. 98: 1572-1578, 1984. Oldberg, A. and Heinegard, D.: Cloning and sequence analysis of rat bone sialoprotein (osteopontin) eDNA reveals an Arg-Gly-Asp cell binding sequence. Proc. Natl. Acad. £Ci. USA 83: 8819-8823, 1986. Reddi, A. H.: Regulation of bone differentiation by local and systemic factors. In: Bone and Mineral Research, Vol. 3, ed. by Peck, W. A., Elsevier, Amsterdam, 1985, pp. 27-48. Scott-Savage, P. and Hall, B. K.: Differentiative ability of the tibial periosteum from the embryonic chick. Acta Anat. 106: 129-140, 1980. Silbermann, M. and Reddi, A. H.: Collagen sponge promotes in vitro bone formation and mineralization. In: Current Advances in Skeletogenesis, ed. by Ornoy, A., Harell, A. and Sela, ]., Elsevier, Amsterdam, 1985, pp. 7-13. Silbermann, M. and Maor, G.: Organ and tissue culture of cartilage and bone. In: Methods of Calcified Tissue Preparation, ed. by Dickson, G. R., Elsevier, Amsterdam, 1984, pp. 467-530. Silbermann, M., Reddi, A. H., Hand, A. R., Leapman, R., von der Mark, K. and Franzen, A.: Cells with osteoblast characteristics arise from cartilage progenitor cells in organ culture of mandibular condyles. J. Craniofacial Genet. and Dev. Bioi. 7: 59-79, 1987. Stoker, M. G. P., O'Neill, c., Berryman, S. and Waxman, V.: Anchorage and growth regulation in normal and virus-transformed cells. Int. J. Cancer 3: 687-693, 1968. Weiss, A., von der Mark, K. and Silber mann, M.: A tissue culture supporting cartilage cell differentiation, extracellular mineralization and subsequent bone formation using mouse condylar progenitor cells. Cell Diff. 19: 103-113, 1986. Dr. Michael Silbermann, Technion School of Medicine, P.O. Box 9649, Haifa, 31096, Israel.
Acceleration of Cartilage and Bone Differentiation on Collagenous Substrata GILA MAORI, KLAUS VON DER MARK2, HARI REDDI 3, DICK HEINEGARD 4 , AHNDERS FRANZEN4 and MICHAEL SILBERMANN 1 Laboratory for Musculoskeletal Research, The Rappaport Institute for Research in the Medical Sciences, Faculty of Medicine, Technion, Haifa, Israel, 2 Laboratory for Connective Tissue Research, The Max Planck Institute for Biochemistry, Martinsried, FRG, 3 Bone Cell Biology Section, National Institute of Dental Research, N.I.H., Bethesda, MD, USA and 4 Department of Physiological Chemistry, University of Lund, Lund, Sweden. 1
Abstract Chondroprogenitor cells of newborn murine mandibular condyles were cultured on top of collagen sponges for up to 18 days. After 24 h in culture, new chondroblasts developed which subsequently matured showing signs of hypertrophy, while the extracellular matrix revealed positive reactivity for type II collagen, cartilage proteoglycans and mineralization. Light and electron microscopy examinations showed signs of new osteoid formation, a feature that was preceeded by positive immunohistochemical reaction for type I collagen, fibronectin and bone specific sialoprotein. A close temporal and spatial association was noted between the development of mature, mineralized cartilage and new osteoid. The differentiation of new cartilage and bone cells was linked to an increased activity of DNA synthesis and cellular proliferation. The de novo bone formation was accompanied by increasing rates of alkaline phosphatase activity and uptake of [45 Ca] features that were found to be tightly correlated to each other. The collagen substrata appeared also to facilitate the migration of cells, their replication and their subsequent differentiation to their respective cellular lineage. Hence, collagen sponges in vitro appear to serve as a promising substrata for culture systems involved with the growth and differentiation of mineralizing tissues such as cartilage and bone. Key words: bone, cartilage, collagen, differentiation, progenitor cells.
Introduction Local and systemic factors regulate the growth, differentiation and maturation of skeletal tissues. The vertebrate skeletal system has the potential for considerable regeneration and remodelling even after growth ceases, as exemplified by the multi-step sequential process of fracture healing (Reddi, 1985). A thorough analysis of the local
352
G. Maor et al.
and systemic factors that influence cartilage and bone differentiation requires the use of an appropriate in vitro experimental model. The extracellular matrix of mineralizing tissues such as cartilage and bone may well be of major importance in defining the local microenvironment for cell growth and differentiation. The principal component of extracellular matrix in cartilage and bone is collagen, which functions as a substratum for the progenitor cells. Most normal animal cells must attach to a surface such as collagen in order to spread and proliferate - a property termed "anchorage dependence" by Stoker et al. (1968). The cell surface glycoprotein fibronectin mediates the attachment of cells to collagen (Kleinman et aI., 1981). Further, the adhesion to matrix may play an important role in determining cell responsivity to other regulatory factors such as matrical proteoglycans (Reddi, 1985). From the foregoing, it is evident that collagen could very possibly be a repository for factors that modify various phases of cartilage and bone development: chemotaxis, proliferation and differentiation. Hence, the purpose of the present study was to elucidate whether collagen may aid in the promotion of cartilage and bone differentiation in a tissue culture system using skeletal progenitor cells. Materials and Methods Collagen substrata
This study utilized collagen sponges (U257-16DS, Helitrex Inc., Princeton, NJ) that were made of type I collagen purified from bovine achilles tendon. The above material comprised of long fibrils that were depleted of telopeptides but were left with endogenous and exogenous crosslinks. The density of the sponges was about 0.09, while over 90% of the sponge volume was open space with the fibrillar strands of collagen organized into a lattice network. As the collagen sponges are supplied in a dehydrated state they swell when immersed in tissue culture media. Swelling normally occurred as a 2- to 3-fold increase in thickness with minor alterations in the other dimensions. Also, during the production process collagen sponges acquire different textures on opposite sides. One surface tends to be fairly smooth and regular whereas the other one becomes more patterned. The two sides differ somewhat in their microstructure: rough and smooth, yet the typical primary and secondary banding patterns of native collagen are preserved. Culture system
ICR mice, 1- to 2-days old, were anesthetized with ether. Using a surgical microscope, the mandibular condyles were aseptically dissected away from the mandibles, cleaned of all soft tissue and of the underlying bone and placed in freshly prepared bovine pancreas 0.1 % trypsin type III, (Sigma, St. Louis, MO) solution (dissolved in BGJb medium; Gibco, Grand Island, NY) for 10 min at 37°C. Tissues were then transferred into a fresh medium containing 10% fetal calf serum, where the perichondrial envelope surrounding the cartilage was gently separated from the cartilaginous core. The separated strands of tissue, consisting almost exclusively of cartilage progenitor cells (Weiss et aI., 1986) were transferred onto 16 mm-wide collagen sponges cemented to stainless steel grids (Michigan Dynamics, Garden City, MI) that were placed in plastic disposable culture dishes (20 mm; Falcon, Irvine, CAl. The culture medium, pH 7.2 - 7.4, consisted of BGJb medium (Fitton-Jackson modifica-
Collagen and Bone Differentiation
353
tion) supplemented with 10% fetal calf serum, ascorbic acid (300 ~g/ml) penicillin (100 U/mi), and streptomycin (100 ~g/ml). Cultures were carried out for as long as 18 days in a humidified incubator at 3rC in an atmosphere of 5% CO r 95% air. The medium was changed every 48 h, and specimens were obtained after 1, 2, 3, 4, 6, 9, 14 and 18 days in culture.
Light and electron microscopy Tissues that were designated for structural studies were initially washed in Hank's balanced salt solution and subsequently fixed in 3% glutaraldehyde in 0.1 M cacodylate buffer, pH 7.2, for 90 min postfixed in 1 % OS04 for 60 min, dehydrated and embedded in Epon 812. Thick sections (1 ~m) were stained with 1 % toluidine blue or with Pearson's silver gelatin stain. Adjacent thin sections (600-700A) were cut on LKB Ultratone III, mounted on No. 300 copper grids and stained for 10 min in saturated aqueous uranyl acetate followed by staining for 10 min in 0.2% lead citrate before examination with a Jeol 100B electron microscope, operating at 60 kV.
Indirect immunofluorescence studies Explants what were designated for immunofluorescence studies were not fixed but were washed and immediately frozen in liquid nitrogen (-186°C) and cut in a cryostat (-25°C), and the sections (8 ~m thick) were mounted on microscope slides. All the frozen sections were initially treated with 0.3 M ethylenediaminetetracetic acid (EDTA) pH 7.5, for 30 min at room temperature, followed by 30 min treatment with hyalase (ovine testes hyaluronidase, Serva # 25118), 1 mg/ml in phosphate-buffered saline (PBS), pH 7.2, at room temperature. The following antibodies were used: rabbit anti-rat type I collagen; guinea pig anti-chick type II collagen; rabbit anti-rat cartilage proteoglycans; rabbit anti-chick cartilage anchorin ClI; and rabbit anti-calf bone specific sialoprotein. Following incubation with the above antibodies diluted 1:5 for 30 min at room temperature in a moist chamber, the sections were rinsed X 3 with PBS and were subsequently reacted with fluorescein isothiocyanate (FITC)-conjugated goat anti-rabbit gamma globulin (1 :20) or with rabbit anti-guinea pig gamma globulin (Behringwerke, Marburg, FRG) (1:20) for 30 min at room temperature. Sections incubated with non-immune serum or with FITC-conjugated rabbit anti-guinea pig gamma globulin alone served as controls.
Histochemistry For the demonstration of alkaline phosphatase activity in explants the azo-coupling method was used whereby frozen sections were incubated in media comprising of 0.2 mM of naphthol AS-TR phosphate (Sigma), 100 mM Tris buffer and 1.6 mM of fast red violet LB, pH 8.4, for 20 min at room temperature. As controls, sections were incubated in media from which the substrate had been omitted; in media containing 5 mM of levamisole or in media that were preheated to 60°C for 20 min. For the localization of active sites of mineralization cultures were supplemented with 10 mg/ml of oxytetracycline (Sigma) for the final 24 h of incubation. The specimens were then fixed with glutaraldehyde alone, embedded in Epon 812 and were examined in a fluorescent microscope.
354
G. Maor et al.
[3 Hj- Thymidine autoradiography
Cultures were supplemented with 5 !-ICi/ml of eH]-methyl thymidine (4.8 Ci/mmol, Nuclear Res. Center, Israel), for the last 24 h of the incubation period. The explants were then washed with Hank's buffer solution containing 1 mM of thymidine, fixed in glutaraldehyde and embedded in Epon. Section 2 !-1m thick were mounted on glass slides, coated with nuclear track emulsion (NTB-2, Kodak, Rochester, NY) and placed in light-tight black boxes. After 3 weeks of exposure at 4°C, autoradiographs were developed in D-170 developer (Kodak) at 18°C, fixed and lightly stained with toluidine blue. rHj-thymidine incorporation into DNA
Explants were transferred into a cold solution of 10% trichloroacetic acid (TCA) with 1 mM thymidine and were homogenized at 4°C with a polytron homogenizer (Type PT 10/35 Kinematica). Following further washing with 10% TCA the precipitates were hydrolyzed at 90°C for 20 min in 1.5 ml10% TCA and aliquots were used to determine TCA insoluble counts. The remaining 1 ml of the hydrolyzate was used to determine the DNA content of the specimens using a modification of Burton's method (Burton, 1968). Sigma's calf thymus DNA was used as standard. Biochemical assay for alkaline phosphatase
Explants were homogenized in ice-cold 0.15 M NaCl containing 3 mM NaHCO~, pH 7.4, centrifuged and the supernatant was assayed for alkaline phosphatase, pH 9.3, with p-nitrophenyl phosphate (0.01 M, Sigma 104) as substrate (Halpern et ai., 1972). Protein was determined in the enzyme extracts (Lowrey et ai., 1951). 45Ca uptake
45CaClz (17.49 mCi/mgj Amersham, U.K.) was added at a dose of 2 !-ICi/ml to the culture medium 18 h prior to the termination of the cultures. Explants were homogenized in buffer as described above, and after the separation of the supernatant for enzyme assays the sediments were washed with 0.1 M CaCl 2 in 0.05 M Tris-HCI, pH 7.4, at 25°C for 30 min in order to remove the exchangeable [45 Ca], and were then hydrolyzed in 0.5 M HCI for at least 2 h at 25°C. Aliquots were used to determine [45 Ca] counts. Means and standard erors were calculated from all numerical findings and the comparison of the means were determined by student's t-test. Correlation tests utilized the least square regression line and standard error of estimate.
Results In newborn mice the cartilage of the mandibular condyle exhibits four distinct zones of cell populations: articular, chondroprogenitor, chondroblasts and hypertrophic chondrocytes (Weiss et aI., 1986). As indcated, our culture specimens were comprised of groups of progenitor cells that revealed strong basophilic properties following staining with toluidine blue (Fig. 1). The isolated tissue reacted positively for type I collagen
Collagen and Bone Differentiation
355
Fig. 1. The appearance of an explant immediately following the trypsinization procedure and its separation from the condylar core. Note the homogeneous population of basophilic cells that comprise condylar progenitor zone. Toluidine blue, X 240.
Fig. 2. Frozen section through an explant immediately following its separation. The tissue was reacted with antibodies against type I collagen. Note the strong reactivity for this type of collagen. X 384.
356
G. Maor et al.
Fig. 3. Section through an explant after 24 h in culture. The tissue attained a form of a conglomerate in which the innermost portion reveals the initial structural signs of cartilagelike tissue. The latter is represented by increasing amounts of extracellular matrix separating between the chondroblast-like cells (C). Toluidine blue, x 240.
Fig. 4. Frozen section through an explant after 24 h in culture and treated with antibodies against cartilage proteoglycans. Note the strong reactivity at the periphery of the newly formed chondroblasts as well as in the extracellular matrix. X 384.
Collagen and Bone Differentiation
357
Fig. 5. Portion of explant that has been cultured for 48 h on top of collagen sponge. Note the distinct nodules of cartilage that is underlied by a thick layer of fibroblast-like cells. The latter attach the newly formed cartilage onto the collagen sponge. Toluidine blue, X 480.
Fig. 6. A section through the newly formed cartilage shown in Figure 5 (48 h in culture). The cell shows an intracellular accumulation of glycogen (G), strands of rough endoplasmic reticulum (arrows) and mitochondria (m). The matrix exhibits typical collagen fibrils along with electron dense granules matrical proteoglycans. x 16,000.
358
G. Maar et al.
Fig. 7. Autoradiograph of an explant after 48 h in culture and labelled in vitro with [3H]_ thymidine. Note the marked labelling, especially along the peripheral sites of the newly formed nodules of cartilage (C). Tissue lightly stained with toluidine blue, x 240.
Fig. 8. Portion of an explant that has been cultured for 3 days. The cartilaginous nodule is already deeply embedded with the collagen sponge. Note the gradation of cartilage cell maturation from younger chondroblasts at the periphery of the nodule (C) to mature hypertrophic chondrocytes (H) within the inner part of the nodule. In addition, a dense population of mesenchyme-like cells (M) is seen adjacent to the well developed cartilage. Toluidine blue, x 240.
Collagen and Bone Differentiation
359
Fig. 9. General view of cartilage after 3 days in culture following reaction with antibodies against cartilage proteoglycans. It can be seen that each individual cell is highly reactive for this antigen. Frozen section, x 154.
Fig. 10. A similar section to that shown in Figure 9 but was reacted with antibodies against cartilage anchorin cn. A positive reactivity is noted throughout. x 154.
360
G. Maar et al.
Fig. 11. A similar section to those shown in Figures 8 and 9 but was reacted with antibodies against type I collagen. Only the peripheral portions of the explants reveal positive reactivity, whereas the newly formed cartilage (C) is negative. X 96.
Fig. 12. Section through explant that was cultured for 8 days and received a single dose of oxytetracycline for 24 h. The fluorescent activity (indicative of matrix mineralization) appears in the inner portion of the tissue. x 240.
Collagen and Bone Differentiation
361
Fig. 13. Frozen section obtained after 3 days in culture and reacted with antibodies against bone specific sialoprotein. Note the intense intracellular reactivity in the newly differentiated cells. x 154.
Fig. 14. Frozen section through the strands of collagen sponge adjacent to the explant after 6 days in culture. The section was reacted with antibodies against bone specific sialoprotein and shows an intense reactivity along the strands of the collagen sponge. All control sections were found negative. x 384.
362
G. Maor et al.
(Fig. 2) but not for type II collagen or for cartilage proteoglycans (not shown). After 24 h in culture, the explants appeared as condensed conglomerates of basophilic cells with a relatively small amount of intercellular matrix between them (Fig. 3). By that time the explant already revealed strong reactivity for cartilage proteoglycans (Fig. 4). By 48 h distinct nodules of cartilage were evident that were attached to the collagen sponge via a dense layer of fibroblast-like cells (Fig. 5). The newly formed young chondroblasts contained the typical lakes of glycogen as well as a distinct system of rough endoplasmic reticulum and a small number of rounded mitochondrial profiles that were randomly dispersed throughout the cytoplasm (Fig. 6). Fine collagen fibrils and proteoglycan granules were always encountered in the extracellular matrix (Fig. 6). By the same time, many cells, especially along the cartilage periphery, were labelled with eH]thymidine as shown with autoradiography (Fig. 7). After 3 days in culture, the cartilage nodules showed large numbers of mature, hypertrophic chondrocytes that, for the most part, were located within the inner portions of the cultures (Fig. 8) and showed strong reactivity for cartilage proteoglycans (Fig. 9) and anchorin CII (Fig. 10). The periphery of this culture reacted positively for type I collagen (Fig. 11). Moreover, these cartilage nodules were found to undergo an active phase of mineralization as visualized via vital staining with oxytetracycline (Fig. 12). Of great interest was the finding that by 3 days almost all of the cells within the explants also reacted positively for bone specific sialoprotein (Fig. 13). With further progression of the culture period (6 days) the fibrilar strands of collagen within the sponges exhibited an intense reactivity for both bone sialoprotein and fibronectin along their surfaces (Fig. 14). Concomitantly more cells appeared to have penetrated into the sponges inner mass (Fig. 15) and were undergoing an active phase of cell proliferation (Fig. 16). By 9 days new loci of differentiation developed adjacent to the existing foci of mature mineralized cartilage. The latter foci were inhabited by cells that differed structurally from the foregoing cartilage cells and were embedded in a well developed extracellular matrix (Fig. 17). The latter matrix which reacted positively for bone specific sialoprotein (not shown) was also found to be occupied by newly formed collagen fibers that reacted positively for type I collagen, but not for type II collagen and appeared to be tightly attached to the sponges' original collagen strands (Fig. 18). in addition, the matrix revealed multiple matrix vesicles in between collagen fibers (Fig. 19) a feature that is characteristic of embryonic woven bone. Also, the cultures revealed the appearance of multinucleated giant cells (Fig. 20) as well as phagocyte-like cells that appeared to be involved in the degradation of the sponges' collagen (Fig. 21). The cellular population in the newly formed nodule of osteoid reacted positively for alkaline phosphatase (Fig. 22). All the control tests for this enzyme revealed negative results. Biochemical determinations for alkaline phosphatase activity indicated a relative low rate of activity during the first 4 days in culture with a substantial increase thereafter (Fig. 23). A very similar pattern was noted with regard to the uptake of 4S [ Ca] (Fig. 23). A very high degree of correlational relationship was found between alkaline phosphatase activity and uptake of [4S Ca] (r = 0.96). Measuring the rate of [3H]-thymidine incorporation into DNA indicated the existence of two distinct peaks: (1) on day 2 and (2) on day 6 (Fig. 24). The first peak coincided with the differentiative phase of cartilage whereas the second peak occurred at about the same time interval of new bone formation.
Collagen and Bone Differentiation
363
-I'
'1 - ~ \ . ·:..~'.I1. ./~~ ....,10<,...,,""""'" Fig. 15. Electron micrograph showing an osteoblast (OB) in close contact with the collagen sponge (CS). At the lower portion of the photomicrograph newly formed collagen fibers (F) can be seen. x 10,000,
Fig, 16, Autoradiograph [3Hl-thymidine-labelled representing a deeper zone within the collagen sponge - farther away from the original explant, Note the immense labelling of cells that have emigrated out of the explant and migrated along the sponge's fibrils (arrows), x 480, 25
Collagen 715
364
G. Maar et al.
Fig. 17. Section through a newly formed osteoid in an explant that was cultured for 9 days. The cells resemble bone cells to a great extent while the matrix (M) appears rich in fibrillar material. Pearson's silver stain, X 288.
Fig. 18. A higher magnification of a portion of a cell and its extracellular matrix in an area shown in Figure 17. Note the large number of collagen fibers that are closely attached to the cell as well as to the sponge's fibrillar material (5). The newly formed collagen fibers (COL) were found to react positively for type I collagen and fibronectin but not for type II collagen. X 9800.
Collagen and Bone Differentiation
365
Fig. 19. A section through the extracellular matrix of the newly formed osteoid. Note the multiple matrix vesicles (arrows) in between the collagen fibers, a characteristic feature of embryonic, woven bone. These vesicular structures serve as the initial locus of mineralization in this tissue. X 9800.
Fig. 20. The appearance of a giant multinuclear cell located at a peripheral site of the mineralized cartilage. This cell resembles a chondroclast and could be involved in the resorption of mineralized matrix. Toluidine blue, x 384.
366
G. Maor et al.
Fig. 21. Electron micrograph of cell within the collagen sponge that appears to be involved in degrading the sponge's fibrils (SF). The cell reveals large number of lysosome-like organelles (arrows) that presumably take part in the digestion of the phagocytized material. x 9800.
Fig. 22. Frozen section through the newly formed osteoid that was reacted for alkaline phosphatase. Pronounced reactivity is noted within most of the cells. x 240.
367
Collagen and Bone Differentiation 4 5 CA UPTAKE AND ALKALINE PHOSPHo ACTIVITY 10
60
9
~
CI.
'0"
~
Q.
•c 1\1
~
8
0 0
.
,..
n
a 3
7
......
6
Z
5
01
4
0
::s.
..'"
......
'C
l> C c: Z a l> DI '-"
.
.. 0
.. Jt: It
0
2
:::J
0
(")
CD
3
c
~ (,II
0
2
4
6
8
10
0
Days in Culture
Fig. 23. Kinetics of [4S Ca] uptake and alkaline phosphatase activity in explants cultured on collagen sponges. Both parameters reveal a marked and steady increase following 4 days in culture. Each mean was based upon 6 determinations . • - . = [4S Ca] uptake; 0 - 0 = alkaline phosphatase activity.
3 H THYMIDINE INCORPORATION INTO DNA
... .0 .
80
0
Col
70
c~
-0(
QlZ
cO
60
"001 .- :::l
E,
>OE
.c . I-CI. Col
50
40
30
0
'0
Days in Culture
Fig. 24. Kinetics of [3H]-thymidine incorporation in explants cultured on collagen sponges. Prominent peaks are noticed on day 2 and 6 at which time the differentiation of cartilage and bone cells took place respectively. Each mean was based upon 6 determinations.
368
G. Maor et a!.
Discussion The differentiation of cartilage and bone cells culminates in the formation of a collagenous extracellular matrix with non-collagenous components that ultimately undergoes mineralization. In this regard mineralized tissues present a formidable challenge to understand the molecular and cellular basis of mineralization. Analysis of this problem is in part simplified by the use of appropriate in vitro models of cartilage and bone differentiation, maturation and mineralization. Although cartilage and bone cells have been isolated from embryonic organs (Silbermann and Maor, 1984), the process of progenitor cells differentiation in vitro and the subsequent formation of specific matrices needs further work-up. One possible reason for the current situation may relate to the lack of optimal environment conditions that are conductive to the differentiation of mineralizing tissues. During the course of systematic investigations on bone differentiation in organ and tissue cultures of the mandibular condyle (Silbermann and Reddi, 1985; Weiss et a!., 1986; Silbermann et a!., 1987), it became apparent that collagen substrata might serve as an important promoter to bone cell differentiation and mineralization. In the present study we were able to demonstrate that tissue culture of newborn condylar progenitor cells not only maintain the capacity to express their chondrogenic phenotype, but also promote bone cell differentiation when cultured on collagen sponges. In a very short period of time, the collagenous microenvironment succeeded in elaborating the cascade of events involving progenitor cell proliferation, cartilage cell differentiation, cartilage matrix mineralization and bone cell differentiation. The present findings indicate that if condylar progenitor cells are cultured on top of collagen substrata they increase the ability to synthesize DNA and proliferate. Also, it became evident that during the first 24 hours in culture, almost all of the progenitor cells in the explant attained chondroblastic phenotype showing positive reactivity for cartilage proteoglycans. The differentiation of cartilage cells and the elaboration of cartilage specific extracellular matrix capable of undergoing mineralization was succeeded by osteoid formation already by 6 days in culture. It became very clear that the development of new osteoid was intimately associated with the differentiation of cartilaginous tissue, a fact that had been alluded to by Scott-Savage and Hall (1980). The foregoing findings indicate that the collagenous scaffold facilitated the differentiation of cartilage and bone progenitor cells and promoted the laying down of mineralized extracellular matrix in vitro. In a way the latter microenvironment appears very reminiscent of the normal ecology of cartilage and bone cells. As already mentioned in the Introduction, the growth and differentiation of cells is anchorage dependent and it is the extracellular collagenous framework that provides the anchorage in vivo. It seems, therefore, reasonable to assume that the collagen sponge provides the necessary milieu of cell-cell and cell-matrix interactions essential for the function of mineralizing cells. Moreover, the collagen sponge may concentrate cartilage and bone essential macromolecules such as cartilage proteoglycans, anchorin CII, bone specific sialoprotein (Franzen and Heinegard, 1985) and osteopontin (Oldberg et aI., 1986), in such a way that they may attain critical concentrations to initiate tissue specific morphogenesis and mineralization. It is also likely that local growth factors such as bone-derived growth factor (BDGF), cartilage-derived factor, an ll-kDa somatomedin-C-like substance, cationic cartilage-derived growth factor (CDGF), bone morphogenetic protein, an 18.5-kDa protein with bone-indictive activity, transforming growth factor beta, skeletal growth factor and insulin-like growth factor I (Hauschka et aI., 1986) that
Collagen and Bone Differentiation
369
might be released by the condylar progenitor cells and may be trapped by ionic and/or hydrophobic interactions in the collagen sporlge. Cells vary in their attachments to collagen matrices. Many cells will adhere just as readily to polystyrene as to collagen. However, some types of cells such as condylar progenitor cells exhibit a definite requirement for collagen (unpublished data), preferentially binding to the trihelical portion or interacting with the end chains. In addition, cell attachment may involve fibronectin or other factors, which may have been supplied by the cultured cells. It is difficult to predict the preferential sponge type for a particular cell line or tissue type. In our hand the U257-16DS sponge was found appropriate for cartilage and bone cells adherance, growth, differentiation, maturation and matrix mineralization. The collagen sponges were found to undergo degradation with time. The rate of degradation depends among others on pH, ionic strength and the amount of collagenase and other proteases elaborated by the cells. Sponges composed of more highly crosslinked collagen appear to survive for longer periods of time, potenially extending into weeks or months. The condylar progenitor cells did not appear to exert any appreciable force on the collagen matrix, as we did not face collapse of sponges. In conclusion, the present in vitro tissue culture system may have far-reaching implications for answering some of the following questions: (1) Which growth factors playa role in the growth and development of the skeleton? What regulates their synthesis, and how are are their effects mediated? (2) Does loss of progenitor cells function contribute to the development of skeletal anomalies? If so, why does the loss occur? How can the loss of function be corrected? (3) What regulates the synthesis of noncollagenous proteins in cartilage and bone matrix? And what is the latters' function? (4) How is the action of hormones such as glucocorticoids, sex hormones, parathyroid hormone and 1,25-dihydroxyvitamin D3 mediated on cartilage and bone forming cells? (5) What is the physiologic role, if any, of lymphokines and cytokines on skeletal tissues differentiation? How are their effects mediated? (6) Is the above in vitro system suitable for the growth of human progenitor cells from both normal subjects and patients? Hence, the ability of collagen substrata to accelerate cartilage and bone differentiation in vitro demonstrated in this study is encouraging as it seems feasible that additional factors such as fibronectin, laminin and bone growth factors could be incorporated in such collagen sponges. Such interactions may fasten the differentiation processes via increasing the affinity of the growing cells onto their substrata. Acknowledgements This study was supported in part by the Fund for the Promotion of Research at the Technion, Grant 180-650, and by the European Molecular Biology Organization, Grant 4448. References Burton, K.: Methods in Enzymology 12: 163-166, 1968. Franzen, A. and Heinegard, D.: Isolation and characterization of two sialoproteins present only in bone calcified matrix. Biochem. J. 232: 715-724, 1985. Halpern, E. P., Rosoff, S. ]. and Weiner, S.: Use of p-nitrophenyl phosphate dicyclohexylammonium salt as substrate for alkaline phosphatase determination on the SMA 12/60. Clin. Chern. 18: 593-594, 1972.
370
G. Maor et al.
Hauschke, P. V., Mavrakos, A. E., Iafrati, M. D., Doleman, S. E. and Klagsburn, M.: Growth factors in bone matrix. J. Bioi. Chern. 261: 12665-12674, 1986. Kleinman, H. K., Klebe, R.]. and Martin, G. R.: Role of Collagenous matrices in adhesion and growth of cells. J. Cell. BioI. 88: 473-485, 1981. Lowry, O. H., Rosenbrough, N. J., Farr, A. L. and Randall, R. J.: Protein measurement with the folin phenol reagent. J. BioI. Chern. 143: 265-275, 1951. Mollenhauer, ]., Bee,]. A., Lizarbe, M. A. and von der Mark, K.: Role of anchorin CIl, a 31,000-mol-wt membrane protein in the interaction of chondrocytes with type II collagen. J. Cell. BioI. 98: 1572-1578, 1984. Oldberg, A. and Heinegard, D.: Cloning and sequence analysis of rat bone sialoprotein (osteopontin) eDNA reveals an Arg-Gly-Asp cell binding sequence. Proc. Natl. Acad. £Ci. USA 83: 8819-8823, 1986. Reddi, A. H.: Regulation of bone differentiation by local and systemic factors. In: Bone and Mineral Research, Vol. 3, ed. by Peck, W. A., Elsevier, Amsterdam, 1985, pp. 27-48. Scott-Savage, P. and Hall, B. K.: Differentiative ability of the tibial periosteum from the embryonic chick. Acta Anat. 106: 129-140, 1980. Silbermann, M. and Reddi, A. H.: Collagen sponge promotes in vitro bone formation and mineralization. In: Current Advances in Skeletogenesis, ed. by Ornoy, A., Harell, A. and Sela, ]., Elsevier, Amsterdam, 1985, pp. 7-13. Silbermann, M. and Maor, G.: Organ and tissue culture of cartilage and bone. In: Methods of Calcified Tissue Preparation, ed. by Dickson, G. R., Elsevier, Amsterdam, 1984, pp. 467-530. Silbermann, M., Reddi, A. H., Hand, A. R., Leapman, R., von der Mark, K. and Franzen, A.: Cells with osteoblast characteristics arise from cartilage progenitor cells in organ culture of mandibular condyles. J. Craniofacial Genet. and Dev. Bioi. 7: 59-79, 1987. Stoker, M. G. P., O'Neill, c., Berryman, S. and Waxman, V.: Anchorage and growth regulation in normal and virus-transformed cells. Int. J. Cancer 3: 687-693, 1968. Weiss, A., von der Mark, K. and Silber mann, M.: A tissue culture supporting cartilage cell differentiation, extracellular mineralization and subsequent bone formation using mouse condylar progenitor cells. Cell Diff. 19: 103-113, 1986. Dr. Michael Silbermann, Technion School of Medicine, P.O. Box 9649, Haifa, 31096, Israel.