EXPERIMENTAL
CELL
RESEARCH
195,
509-515
(1991)
Stimulation of Chondrogenesis in Limb Bud Mesoderm Cells by Recombinant Human Bone Morphogenetic Protein 2B (BMP-2B) and Modulation by Transforming G rowth Factor ,8, and & PING CHEN,* JILL L. CARRINGTON,~ R. G. HAMMONDS,$ AND A. H. REDDISH’ *Bone Cell Biology Section, National Institute of Dental Research, National Institutes of Health, Bethesda, Maryland 20892; tDepartment of Anatomy and Cell Biology, Uniformed Services University of the Health Sciences, F. Edward Hkbert School of Medicine, Bethesda, Maryland 20814; and SGenentech, Inc., South San Francisco, California 94080
repair [2-41. Recently the genes for several bone morphogenetic proteins (BMP) were cloned and expressed [ 5-7 1. Recombinant BMP-2A, BMP-2B (also referred to as BMP-4), and osteogenin (BMP-3) induce endochondral bone differentiation in uiuo [6-71. However, the mechanism of action of these inductive proteins on endochondral bone formation is unclear and needs to be further studied using appropriate responding cells. BMP-2B (BMP-4) is one of a family of cartilage and bone-inductive proteins derived from bone matrix. These are, in turn, part of a larger family of developmentally important molecules related to TGF-/3 [5]. In postnatal and adult animals, such bone morphogenetic proteins may be involved in bone repair. It is likely that they may be involved in cartilage and bone differentiation during limb morphogenesis. In this study, we have used stage 24-25 chick limb bud mesoderm cells in culture to examine whether BMP-2B regulates chondrogenesis and its modulation by transforming growth factor p, and &.
Bone morphogenetic protein 2B (BMP-2B) also called BMP-4 is one of a family of cartilage and bone-inductive proteins derived from bone matrix and belongs to the transforming growth factor ,f3(TGF-/3) superfamily. These bone-inductive proteins isolated from adult bone may be involved in bone repair. However, they may also play a role in cartilage and bone formation during embryonic development. To test whether BMP-2B influences cartilage formation by embryonic cells, recombinant human BMP-2B was applied to cultured limb bud mesoderm plated at three different densities. BMP-2B stimulated cartilage formation as assessed by Alcian blue staining and incorporation of radioactive sulfate into sulfated proteoglycans. Cells cultured at all three densities in the presence of 10 rig/ml BMP-2B formed a nearly continuous sheet of cartilage with abundant extracellular matrix and type II collagen. In addition, when cells were cultured in 0.5% serum in the presence of 10 nglml of BMP-2B for 5 days there was an increase in alkaline phosphatase as detected by histochemical and biochemical methods. Transforming growth factor p isoforms (TGF-8, and TGF-8,) inhibited sulfate incorporation into proteoglycans in a dose-dependent manner. This inhibition by TGF/3 was overcome by recombinant BMP-2B. This study demonstrates that recombinant BMP-2B stimulates cartilage formation by chick limb bud mesoderm in vitro and is further modulated by a 1991 Academic Press, Inc. TGF-/3 isoforms.
MATERIALS
AND METHODS
Cell culture. Fertile chicken eggs (Truslow Farms, Inc. Chestertown, MD) were incubated and embryos were removed from the eggs for dissection at Hamburger and Hamilton stages 24-25 [8]. Wing buds were collected in sterile Tyrode’s salt solution containing gentamicin. Then they were placed in 1% ethylenediaminetetra-acetic acid in double-strength calcium-magnesium-free Tyrode’s solution at 37°C for 30 min to loosen the ectoderm. Ectoderms were removed from the mesoderm in cold 10% fetal bovine serum in Ham’s F-12 medium and discarded. The mesoderm cells were dissociated in trypsin according to the procedure of Ahrens et al. [9]. Cells were suspended at three different densities (low density, 0.4 x lo7 cells/ml; medium density, 0.8 X lo7 cells/ml; high density, 2 X lo7 cells/ml) in Ham’s F-12 medium with 5% fetal bovine serum, L-glutamine, and gentamicin. A 20.~1 drop of cell suspension was placed in the center of each of the eight center wells of a 24.well plate. After 1 h incubation at 37”C, 0.5 ml of medium containing 0.5% fetal bovine serum was added. Growth and differentiation factors were also added at this time. The next day, an additional 0.5 ml medium and test factors were added, and each day thereafter 0.5 ml medium was removed and exchanged for fresh medium. BMP-2B and T G F 0, and & were also renewed daily except as specified for individual experiments.
INTRODUCTION Recent investigations have shown that polypeptide growth factors may play an important role in the development and growth of cartilage and bone [ 11. The developmental sequence of matrix-induced new bone formation is morphologically similar to that seen in embryonic endochondral bone formation and adult fracture 1 To whom reprint requests should be addressed at National Institutes of Health, Building 30, Room 211, Bethesda, M D 20892. 509
0014.4827/91$3.00
Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
510
CHEN
BMP-2B was obtained as deBMP-PB and TGFO. Recombinant scribed [7] and it was active in uiuo [7]. The protein was quantitated by amino acid analysis and dissolved in 5 m M HCl with 2% bovine serum albumin and stored at 4°C. TGF 8, and&were purchased from R & D Systems (Minneapolis, MN). Alkaline phosphatase activity was Alkaline phosphatase actiuity. detected both histochemically and biochemically [lo]. For quantitation, 0.5 ml of buffer containing 0.15 M NaCl, 3 mMNaHCO,, pH 9.3, was added to the cultures. The cultures were then scraped from the dish into tubes and sonicated. Aliquots of cell extracts were incubated with p-nitrophenyl phosphate for 60 min at 37°C and the formation of p-nitrophenol was determined on a spectrophotometer at 400 nm. A unit is defined as units/mg protein. For histochemical staining, cultures were rinsed and incubated with a mixture of 0.05% naphthol AS-MX phosphate, 1% N,N-dimethylformamide, 0.01 M MgCl,, and 0.1% fast blue BB salt in 0.1 M Tris buffer, pH 8.5, at 37°C for 15 min [ll]. After color development, cultures were washedin PBS and fixedin 10% formaldehyde. Cultures were washed with phosphate-buffA&an blue staining. ered saline (PBS) and fixed in 10% formaldehyde in PBS overnight. After a rinse in 3% glacial acetic acid, pH 1, cultures were stained overnight in 0.5% Alcian blue in 3% glacial acetic acid (pH 1). Cultures were destained in 3% glacial acetic acid and stored in PBS. Cell proliferation. Cell proliferation was assessed by incorporation of [3H]thymidine into acid precipitable material. Medium was removed from the cells and replaced with Ham’s F-12 without thymidine but containing 0.2% fetal bovine serum. [3H]Thymidine was added at 5 &i/ml and cells were incubated for 2 h at 37°C. Incorporation of radioactivity was determined according to the method of Danielpour et al. [ 121. Synthesis of sulfated proteoglycans. The synthesis of sulfated proteoglycans was assessed by adding [?S]sulfate to each culture at 30 pCi/ml for 6 h. After the labeling period, the medium was removed and saved and the cells were dissolved in 6 M guanidine-HCl, pH 7.4. The fraction of radioactive sulfate incorporated into macromolecules in the cell layer and in the medium was determined as previously described [lo]. Results were normalized to the quantity of DNA and the statistical significance was analyzed using t test. DNA was quantitated using the method of Quantitation of DNA. Labarca and Paigen [13]. The same lot of DNA standard and bisbenzimide were used throughout the experiments to minimize variations. Detection of type II collagen. Cultures were washed in PBS, fixed in buffered 3% formalin for 15 min, and washed again in PBS. Endogenous peroxidase activity was inactivated in 0.5% H,O, in methanol for 1 h and nonspecific binding was blocked with 5% bovine serum albumin in 0.01 MT&buffered saline, pH 7.4. Affinity-purified goat antibody to type II collagen (Southern Biotechnology, Inc., Birmingham, AL) was applied to the cultures at 1:lO dilution (controls had diluted normal goat serum) for 1 h at room temperature. The secondary antibody was biotinylated rabbit anti-goat antibody. ABCperoxidase (Vector Labs, Inc., Burlingame, CA) was applied to detect antibody binding according to manufacturer’s directions and 3,3’diaminobenzidine was used as the peroxidase substrate. Separation of chondrocytes and nonchondrocytes from mixed cultures. Limb bud mesoderm was grown for 8 days in culture at 2 X lo7 cells in 600 ~1, then subjected to three sequential collagenase digestions (620 units/ml, Worthington, CLS2) according to the method of Lennon et al. [14]. After each digestion, single cells were separated from cell clumps using a 20-Km pore size Nitex filter. In this way, cells were separated into predominantly nonchondrogenic, mixed, and predominantly chondrogenic cell populations [ 141. These cells were replated at 8 X lo6 cells/ml in 20 pl drops in separate wells of a 24.well plate. After 1 h, 0.5 ml medium containing 0.5% fetal
ET AL bovine serum was added along with BMP-2B. as described above.
Medium
was changed
RESULTS Cultures of Limb Bud Mesoderm In the presence of 0.5% serum, the mesoderm cells formed typical cartilage nodules surrounded by a layer of fibroblast-like cells on Day 2. At this time, cartilage matrix was detected by staining with Alcian blue. The synthesis of cartilage matrix as indicated by incorporation of [35S]sulfate into proteoglycans increased after 1 day in culture at all three cell densities (Fig. 1). Proteoglycan synthesis was greatest in the highest density culture. Type II collagen was detected immunochemically in the nodules of cartilage distributed throughout the culture, but not in the internodular regions (Figs. 2a and 2~). Effects of BMP-2B In high density Synthesis of sulfated proteoglycans. cultures the response to BMP-2B at 0.1 rig/ml was undetectable. However, at 1 rig/ml of BMP-2B there was an almost twofold increase in [35S]sulfate incorporation, and more than a 40-fold increase in incorporation of [35S]sulfate into proteoglycans at the dose of 10 rig/ml of BMP-2B (data not shown). A dose of 10 rig/ml was chosen to examine the time course of response to BMP-2B at all three different cell densities (Fig. 1). After 2 days of application, the cells at medium and high densities showed a significant increase in sulfate incorporation over controls (P < 0.005). After 3 days of exposure, cells at all three densities showed significantly increased sulfate incorporation (P < 0.005), 1.5-fold (low density), 2.7-fold (medium density), and 19-fold (high density) compared to respective controls. After 4 days, the highest effect of BMP-2B was in the high density cultures (32-times that of control). Type II collagen. Cells cultured at high density without BMP-2B showed distinct cartilage nodules interspersed with noncartilage cells (Fig. 2a). However, in the presence of 10 rig/ml BMP-BB, these cells formed a nearly continuous sheet of cartilage (Fig. 2b). The uniformity of cartilage formation was reflected in the distribution of type II collagen. In cultures exposed to BMP-BB, the antibody to type II collagen bound over the entire cell layer rather than to just nodules (Fig. 2d). The incorporation of thymidine by DNA synthesis. cells was measured at all three cell densities after 1,2,3, and 4 days of culture (Table 1). During the first 2 days, exposure to 10 rig/ml BMP-2B depressed DNA synthesis by cells at all three densities (P < 0.05). However, after 3 and 4 days of exposure, the cells at high density
BMP-2B
0
AND
Window
*Or
140 0 x
r b
105
Day
Combined Factor-P
Day by FIG. 1. Time course of synthesis of sulfated proteoglycans chick limb bud mesoderm with and without added BMP-2B. (a) High density (2 X 10’ cells/ml): (b) medium density (8 X 10’ cells/ml): (c) low density (4 X lo6 cells/ml). Data represent the mean -t standard deviation for four cultures at each time point and treatment. (O), control cultures; CO), cultures treated with 10 rig/ml BMP-2B. The scale for the increase in sulfate incorporation over 4 days with BMP2B treatment greatly Qurpasses the increase in control cultures. Note that the scale for % incorporation is different at the three different densities. Synthesis was greatest in the high density cultures.
had increased d 0.05).
DNA
synthesis
compared
of Exposure
In order to determine when BMP-2B acts during the differentiation of these mesodermal cells, we applied it for two separate time periods to cells cultured at high density. In one experiment, BMP-2B was added only during the first 2 days of culture and removed. The cells were cultured in control medium for an additional 3 days. In this experiment there was a greater than 12fold increase in incorporat,ion of [35S]sulfate into proteoglycans at 10 rig/ml BMP-2B (control, 5.7 k 0.4 cpm X lO”/pg DNA; BMP-BB, 71 + 3.2 cpm X lO”lpg DNA). In a second experiment, cells were cultured in control medium during the first 3 days and 10 rig/ml BMP-2B was added only during the last 2 days. The cells increased synthesis of sulfated proteoglycans to 32-times the control value when exposed only during the last 2 days of culture (control, 4.9 _t 0.4 cpm X 103/pg DNA; BMP-BB, 156.5 t 6.8 cpm X 103/pg DNA). These results demonstrate that at later stages there is a marked increase in the response of mesodermal cells to BMP-2B.
Day
*
511
CHONDROGENESIS
to controls
(P
Alkaline phosphatase. In 0.5% serum, alkaline phosphatase activity was not detectable in control cultures, or cultures exposed to BMP-2B at 0.1 and 1 rig/ml for 5 days (data not shown). However, when cells were exposed to 10 rig/ml of BMP-2B for 5 days, alkaline phosphatase activity was detected both histochemically and biochemically. Alkaline phosphatase activity was maximal in high density cultures (low density, 0.09 + 0.01 units/mg protein; medium density, 0.12 + 0.03 units/mg protein; high density, 0.27 + 0.02 units/mg protein).
Effects of BMP-2B and Transforming (TGF-/3, or TGF-&)
Growth
Previous studies have indicated that TGF-@ may modulate cartilage or bone differentiation [15, 26, 29, 301. We studied the effects of combining transforming growth factor-p (TGF-P, or TGF-&) and BMP-2B on sulfated matrix production by chicken limb bud mesoderm. In the presence of 0.5% serum, TGF-@, and TGF8, alone depressed incorporation of sulfate into macromolecules after 5 days (Figs. 3 and 4). The addition of TGF-0, and BMP-2B together at 1 rig/ml resulted in increased sulfate incorporation 6.6-times that of controls, this was more than BMP-2B alone. However, addition of TGF-P, and BMP-2B together at 10 rig/ml resulted in a l7-fold increase of sulfate incorporation compared to the control cultures; this was less than BMP-2B alone at this higher dose. After the addition of TGF-& and BMP-2B together at 1 rig/ml, there was an increase in sulfated proteoglycan synthesis to g-times that of controls, again, higher than BMP-2B alone at this dose. With a 10 rig/ml dose of both TGF-0, and BMP-BB, sulfate incorporation increased 44-fold compared to controls (Figs. 3 and 4). Effects of BMP-2B Bud Mesoderm
on Subpopulations Cells
of Cultured
Limb
Stage 24-25 chick limb bud mesoderm contains several populations of cells and includes chondrogenic and myogenic lineages [ 161. To examine the effects of BMP2B on chondrogenic and nonchondrogenic cells, we separated cells into three populations. The first population
512
CHEN
ET AL.
FIG. 2. Influence of BMP-PB on chondrogenesis. Alcian blue staining (a and b) and distribution of type II collagen (c and d) in cultured chick limb bud mesoderm. Mesoderm cells were cultured at high density (2 X IO7 cells/ml) without BMP-2B (a and c) or with 10 rig/ml BMP-2B (b and d) for 5 days. Both the Alcian blue and anti-type II collagen antibody staining show nodules of cartilage surrounded by a network of fibroblast-like cells in controls, In the presence of BMP-2B both staining methods revealed a nearly continuous sheet of cartilage.
(fraction 1) was predominantly nonchondrogenic cells as indicated by morphology, Alcian blue staining, and synthesis of sulfated proteoglycans (Table 2); the second population (fraction 2) was a mixed population; and the third population (fraction 3) was predominantly chondrogenic cells which stained intensely with Alcian blue and synthesized large amounts of sulfated proteoglycans (Table 2). BMP-ZB was applied at 10 rig/ml for
5 days to these three populations of cells cultured at medium density. Synthesis of sulfated proteoglycans was increased nearly g-fold by addition of BMP-2B to the first population of predominantly nonchondrogenic cells. The increase was 11-fold in the second mixed population of cells, and a similar increase was observed in the third, predominantly chondrogenic, population of cells (Table 2). These results demonstrate that sulfate
BMP-2B
AND
TABLE Effects
of BMP
on Incorporation
of [3H]Thymidine
Cell Density (cells/ml)
1
by Chick
Limb
Bud
Day [“Hlthymidine
Low (4 x 106) Medium
Control BMP-2Bb Control BMP-2B Control BMP-2B
(8 x 106)
High (2 x 107)
513
CHONDROGENESIS
1
2
1150 +- 210” 540 +- 190 3110 +- 450 1440 I 190 9140 k 820 3960 +- 880
1150 -+ 80 53Ok 110 4220-+ 310 1670 f 130 12600 f 670 9260+- 1270
Cells
at High,
Medium
incorporation
and Low
Density
(CPM/Culture) 3 2490 *
4
590
198Ok 85Ok
285Ok 622Ok
1220 580 5900 * 880 16930 t 1760 25830 k 910
388Ok
360 210 230
4190 t 710 965Ok 1620 18720 + 2100
’ Data are presented as means + standard deviation for four cultures. * BMP-2B was added daily when the medium was changed at a dose of 10 rig/ml.
incorporation of both nonchondrogenic genie cells is stimulated by BMP-2B.
and chondro-
DISCUSSION
BMP-2B increased cartilage formation by chick limb bud mesoderm. Under the influence of BMP-2B at 10 rig/ml, the difference was discernible morphologically, with Alcian blue staining, and by increased radio-sulfate incorporation after 2 days of incubation when compared with control cultures. The dose eliciting a response in culture was similar to but somewhat lower than that eliciting a response in postnatal animals in uiuo. With
an in viuo osteogenesis assay of recombinant BMP-BB, a dose of 30 ng induced cartilage and 100 ng induced cartilage and significantly increased alkaline phosphatase activity [7]. In the present experiment, cells cultured at all three densities in the presence of 10 rig/ml BMP-2B formed a nearly continuous sheet of cartilage with sulfated proteoglycans and type II collagen. This was in contrast to the control cultures, which showed distinct cartilage nodules interspersed with fibroblastic cells. BMP-2B initially inhibited cell proliferation. However at later times, in cells plated at high density, proliferation was increased by BMP-2B. In medium and low density cultures, BMP-2B increased the synthesis of cartilage matrix despite decreased cell division. However, in high density cultures, the later increase in cell
2 0 TGF-Pl BMP-2B
-
- .I 1 10 -
I -
10 -
.l ,l
1 1 10 1 10 10
Dosehg/ml)
FIG. 3. Combined effects of BMP-2B and TGF-0, on proteoglycan synthesis by cultured chick limb bud mesoderm at high density. Doses of BMP-2B and TGF-@, were applied for 5 days. Data represent the mean f standard deviation of increases and decreases relative to control. Data are shown for eight cultures of controls, eight cultures at 1 rig/ml TGF-/3,, and four cultures for all other treatments. Standard deviations are depicted by error bars where large enough to appear on this scale.
TGF-P2
-
-
BMP-2B
-
110
-
1
10
1
1
10
-
-
1
10
10
Dose (g/ml)
FIG. 4. Combined effects of BMP-2B and TGF-0, on proteoglycan synthesis by cultured chick limb bud mesoderm at high density. Doses of BMP-2B and TGF-& were applied for 5 days. Data represent the mean i standard deviation of increases and decreases relative to control. Data represent four cultures for all treatments. Standard deviations are depicted by error bars where large enough to appear on this scale.
514
CHEN
TABLE
2
Effects of BMP-2B on Sulfate Proteoglycan Synthesis by Subpopulations of Chick Limb Bud Mesoderm Separated by Collagenase Digestion
Treatment Control BMP-2B Fold Increase
Cell population
?3 incorporation
Fraction
Fraction
1
,570 f 20” 4920 i 160 8.6
2
1410 5 340 15,520 t 670 11
CPM/pg Fraction
DNA 3
9580 +- 610 110,240 f 11,140 11.5
’ Data are presented as means + standard deviation of four cultures from fractions obtained after selective collagenase digestion as described under Materials and Methods.
division may have contributed to the overall increase in proteoglycan synthesis. The incorporation of [35S]sulfate into matrix macromolecules by chick limb bud mesoderm is maximal at 8 days [17]. There is an approximately three- to fourfold increase in sulfate incorporation between Days 5 and 8, followed by decreased incorporation from Day 10 through Day 16 [17]. Results presented here showed that the [35S]sulfate incorporation into proteoglycans increased about 40-fold after treatment with BMP-2B at 10 rig/ml for 5 days. Therefore, it is unlikely that the control cultures would ever reach the level of the synthesis of proteoglycan seen in BMP-2B treated cultures. This was consistent with morphological observations showing some cartilage nodules, but also a large number of noncartilage cells in control cultures, while in BMP2B treated cultures, the cells all appeared to be cartilage cells. Stage 23 to 24 chick limb bud mesoderm cells have the potential to differentiate into several tissue phenotypes, including cartilage, muscle, and bone [ 181. Seeding of such undifferentiated cells in culture at high initial plating density has been shown to promote chondrogenesis [19, 201. Therefore, cultures of embryonic chick limb endochondral bone precursors are a particularly useful way to study cells at early stages of differentiation. BMP-2B was able to stimulate cartilage formation regardless of whether it was present early or late in the culture period. However, the increase in cartilage formation when BMP-2B was added only during Days 3 and 4 was equivalent to that when BMP-2B was present throughout the culture period. Therefore, it is clear that BMP-2B acted on these cells not only early in the culture period, before the cells were overtly differentiating, but even more strongly during the later culture period, when the cells were already overtly differentiating. In addition, when chondrogenic and nonchondrogenic cells were separated from the cultures and replated, BMP-2B increased the incorporation of sulfate by both the originally nonchondrogenic and the chondrogenic
ET AI,.
cells. It appears, then, that BMP-2B can increase cartilage formation by acting at multiple stages during cartilage morphogenesis. A number of reports indicate that alkaline phosphatase activity increases prior to mineralization of matrix and this enzyme may play an obligatory role in the mineralization process [al]. Banks et al. [22] demonstrated that alkaline phosphatase activity aids in production of hydroxyapatite crystals on synthetically crosslinked collagen bathed in @-glycerophosphate solutions. It has also been suggested that alkaline phosphatase may act as a pyrophosphatase to remove the pyrophosphates which are known inhibitors of mineralization. Therefore, alkaline phosphatase activity is often used as a distinctive marker of osteogenic expression [3,23,24]. It is also expressed in resting and hypertrophic chondrocytes [ 111. In the present study, in 0.5% serum, alkaline phosphatase activity was below the level of detection in control cultures. However, when cultures were exposed to BMP-2B for 5 days at 10 rig/ml, alkaline phosphatase was detected both histochemically and biochemically. The stimulation of alkaline phosphatase activity may be an important function of BMP-2B in the maturation of chondroblasts and osteoblasts. Many well-known growth factors and cytokines have been implicated in bone growth and repair [l, 25, 261. TGF-0, is synthesized and secreted by bone cell cultures [27] and is a multifunctional autocrine regulator of bone formation [15]. In this study, we tested the effects of combined BMP-2B and TGF-0 isoforms (0, and 8,) on embryonic limb bud mesoderm. TGF-fi, or TGF-& alone, in 0.5% serum, inhibited sulfate incorporation into proteoglycans in a dose-dependent manner. BMP2B was able to overcome the inhibition of TGF-0, at 1 or at 10 rig/ml. At 1 rig/ml, the combination of TGF-0, and BMP-2B actually increased the synthesis of cartilage matrix to 3.5 times that of BMP-2B alone. However, at 10 rig/ml, the combination of TGF-P, and BMP-2B led to a level of sulfated proteoglycan synthesis, that, while still above control levels, was lower than with 10 rig/ml BMP-2B alone. Therefore, there appears to be a dosedependent synergistic action of BMP-2B and TGF-p,. While TGF-& alone inhibited sulfated proteoglycan synthesis, when TGF-6, was combined with BMP-2B either at lower dose or at higher dose, it acted synergistically with the BMP-2B. Therefore, in this experiment, at high doses, the potencies of these two forms of TGFB appear to differ when applied to limb bud mesoderm in uitro, particularly in combination with BMP-2B. This is consistent with other experiments testing effects of TGF-P, and TGF-& on cultures of limb bud mesoderm from stage 23-24 chicken embryos [as]. Although TGFp, and TGF-0, are homologous to BMP-2A and BMP-3 (about 34%), they do not induce new cartilage or bone formation when tested in an in vim osteogenesis assay in nonskeletal sites [29]. However, they do increase the
BMP-2B
AND
CHONDROGENESIS
synthesis of cartilage specific macromolecules by fibroblasts in vitro as well as modulate the expression of other extracellular matrix components by a variety of other cell types [30, 311. Recently we have tested native osteogenin (BMP-3) in this experimental system (manuscript in preparation). Osteogenin increased cartilage formation by chick limb bud mesoderm: the synthesis of sulfated proteoglycans in high density cultures increased about IO-fold under the influence of osteogenin at 10 rig/ml. Interestingly, after application of osteogenin to these cells for 5 days no alkaline phosphatase activity was detected in the presence of 0.5% serum. Therefore, the actions of each member of the BMP family of proteins may differ somewhat. Recombinant human BMP-2B induces endochondral bone differentiation in viva [7]. We have now shown that it also stimulates cartilage formation by chick limb bud mesoderm in uitro. The fact that embryonic cells respond to osteoinductive factors such as BMP-2B and osteogenin indicates these factors may play a role in cartilage development during limb morphogenesis. We thank Dr. Noreen S. Cunningham and Dr. Vishwas Paralkar for their constructive criticism ofthis manuscript. The work was supported by the USIJHS Protocol No. RO70AN. The opinions or assertions contained herein are the private ones of the authors and are not to he construed as official or reflecting the views of the DOD or the USIJHS. The experiments reported herein were conducted according to the principles set forth in the “Guide for Care and Use of Laboratory Animals” Institute of Laboratory Animal Resources, National Research Council, DHEW. Puh. No. (NIH) 78-23.
REFERENCES 1.
2. 3. 4. 5.
6.
Hauschka, P. V., Mavrakos, A. E., Zafrati, M. D., Doleman, S. E., and Klagshrun, M. (1986) J. Hiol. Chem. 261, 12,66512,674. Urist, M. R. (19fi5) Science 150, 893-899. Reddi, A. H., and Huggins. C. B. (1972) Proc. Nat/. Acad. Sci. IJSA 69, 1601~~1605. Firschein, H. E., and Urist, M. R. (1972) Clin. Orthop. R&t. Rex 84, 263-275. Wozney, J. M., Rosen, V., Celeste, A. J., Mitsock, L. M., Whitters, M. J., Kriz, R. W., Hewick, R. M., and Wang, E. A. (1988) Science 242, 1528-1534. Wang, E. A., Rosen, V., D’Alessandro, J. S., Bal:luy, M., Cordes, P., Harada, T., Israel, D. I., Hewick, R. M., Kerns, K. M., LaPan,
Received December 17. 1990 Revised March 19. 199 I
515
P., Luxenherg, D. I’., McQuaid, D., Moutsatsos, 1. K., Nave, J., and Wozney, J. M. (1990) Proc. N&l. Acad. Sci. IJSA 87, 22202224. Hammonds, R. G., Schwall, R., Dudley, A., Lai, C., Berkemeier, L., Cunningham, N., Reddi, A. H., Wood, W. I., and Mason, A. J. (1991) Mol. Endocrinol. 4, 149-155. 8. Hamburger, V., and Hamilton, H. L. (1951) J. Morphol. 88,4992. 9. Ahrens, P. B., Solursh, M., and Reiter, R. S. (1977) neu. Biol. 60, 69-82. J. L., and Reddi, A. H. (1990) Exp. Cell Res. 186, 10. Carrington, 368-373. and its Applica11. Burstone, M. S. (1962) Enzyme Histochemistry tion in the Study of Neoplasms, pp. 160-280, Academic Press, New York. 12. Danielpour, D., Dart, L. L., Flanders, K. C., Roberts, A. B., and Sporn, M. B. (1989) J. Cell Physiol. 138, 79-86. 13. Labarca, C., and Paigen, K. (1980) Anal. Biochem. 102, 344352. 14. Lennon, D. I’., Osdoby, P., Carrino, D. A., Vertel, B. M., and Caplan, A. I. (1983) J. Craniofacial Gen. DPU. Riol. 3, 235-251. 15. Centrella, M., McCarthy, T. L., and Canalis, E. (1987) J. Riol. Chem. 262, 2869-2874. 16. Zanetti, N. C., and Solursh. M. (1986) I>eu. Hiol. 113, 110-118. 17. Hascall, V. C., Oegema, 7’. R., Brown, M., and Caplan, A. I. (1976) J. Riol. Chem. 251, 3511-3519. 18. Zwilling, E. (1968) Deu. Biol. Suppl. 2, 184-204. 19. Caplan, A. I. (1970) Exp. Cell Res. 62, 341-355. 20. Umansky, R. (1966) Deu. Biol. 13,31-56. and Physiology of Bone” 21. Bourne, G. (1974) “The Biochemistry (Bourne, G. H., Ed.), 2nd ed., Vol. III, pp. 79-120. Academic Press, New York. 22. Banks, S., Wakajima, S., Shapiro, L., and Tilcuitz, 0. (1977) Science 198, 1164-1166. 23. Robison, R. (1923) Biochem. J. 17, 286-293. 24. Wlodarski, K. H., and Reddi, A. H. (1986) Calcif. Tissue Znt. 39, 382-385. 25. Simpson, E. (1984) Trends Biochem. Sci. 9.527-530. 26. Triffitt, J. T. (1987) Acta Orthop. &and. 58, 673-684. 27. Robey, P. G., Young, M. F., Flanders, K. G., Roche, N. S., Kondaiah, P., Reddi, A. H., Termine, J. D., Sporn, M. B., and Roberts, A. B. (1987) J. Cell Riol. 105, 457-463. 28. Kulyk, W. M., Rodgers, B. J., Greer, K., and Kosher, R. A. (1989) I)eu. Riol. 135, 424-430. N., and Reddi, A. H. (1987) 29. Sampath, T. K., Muthukumaran, Proc. Natl. Acad. Sci. IJSA 84, 7109-7113. 30. Seyedin, S. M., Thompson, A. Y., Bentz, H., Rosen, D. M., McPherson, J. M., Conti, A., Siegel, N. R., Galluppi, G. R., and Piez, K. A. (1986) J. Riol. C,hem. 261, 5693-5695. 31. Sporn, M. B., Roberts, A. B., Wakefield, L. M., and Assoian, R. K. (1986) Science 233, 532-534.
CELL
RESEARCH
195,
509-515
(1991)
Stimulation of Chondrogenesis in Limb Bud Mesoderm Cells by Recombinant Human Bone Morphogenetic Protein 2B (BMP-2B) and Modulation by Transforming G rowth Factor ,8, and & PING CHEN,* JILL L. CARRINGTON,~ R. G. HAMMONDS,$ AND A. H. REDDISH’ *Bone Cell Biology Section, National Institute of Dental Research, National Institutes of Health, Bethesda, Maryland 20892; tDepartment of Anatomy and Cell Biology, Uniformed Services University of the Health Sciences, F. Edward Hkbert School of Medicine, Bethesda, Maryland 20814; and SGenentech, Inc., South San Francisco, California 94080
repair [2-41. Recently the genes for several bone morphogenetic proteins (BMP) were cloned and expressed [ 5-7 1. Recombinant BMP-2A, BMP-2B (also referred to as BMP-4), and osteogenin (BMP-3) induce endochondral bone differentiation in uiuo [6-71. However, the mechanism of action of these inductive proteins on endochondral bone formation is unclear and needs to be further studied using appropriate responding cells. BMP-2B (BMP-4) is one of a family of cartilage and bone-inductive proteins derived from bone matrix. These are, in turn, part of a larger family of developmentally important molecules related to TGF-/3 [5]. In postnatal and adult animals, such bone morphogenetic proteins may be involved in bone repair. It is likely that they may be involved in cartilage and bone differentiation during limb morphogenesis. In this study, we have used stage 24-25 chick limb bud mesoderm cells in culture to examine whether BMP-2B regulates chondrogenesis and its modulation by transforming growth factor p, and &.
Bone morphogenetic protein 2B (BMP-2B) also called BMP-4 is one of a family of cartilage and bone-inductive proteins derived from bone matrix and belongs to the transforming growth factor ,f3(TGF-/3) superfamily. These bone-inductive proteins isolated from adult bone may be involved in bone repair. However, they may also play a role in cartilage and bone formation during embryonic development. To test whether BMP-2B influences cartilage formation by embryonic cells, recombinant human BMP-2B was applied to cultured limb bud mesoderm plated at three different densities. BMP-2B stimulated cartilage formation as assessed by Alcian blue staining and incorporation of radioactive sulfate into sulfated proteoglycans. Cells cultured at all three densities in the presence of 10 rig/ml BMP-2B formed a nearly continuous sheet of cartilage with abundant extracellular matrix and type II collagen. In addition, when cells were cultured in 0.5% serum in the presence of 10 nglml of BMP-2B for 5 days there was an increase in alkaline phosphatase as detected by histochemical and biochemical methods. Transforming growth factor p isoforms (TGF-8, and TGF-8,) inhibited sulfate incorporation into proteoglycans in a dose-dependent manner. This inhibition by TGF/3 was overcome by recombinant BMP-2B. This study demonstrates that recombinant BMP-2B stimulates cartilage formation by chick limb bud mesoderm in vitro and is further modulated by a 1991 Academic Press, Inc. TGF-/3 isoforms.
MATERIALS
AND METHODS
Cell culture. Fertile chicken eggs (Truslow Farms, Inc. Chestertown, MD) were incubated and embryos were removed from the eggs for dissection at Hamburger and Hamilton stages 24-25 [8]. Wing buds were collected in sterile Tyrode’s salt solution containing gentamicin. Then they were placed in 1% ethylenediaminetetra-acetic acid in double-strength calcium-magnesium-free Tyrode’s solution at 37°C for 30 min to loosen the ectoderm. Ectoderms were removed from the mesoderm in cold 10% fetal bovine serum in Ham’s F-12 medium and discarded. The mesoderm cells were dissociated in trypsin according to the procedure of Ahrens et al. [9]. Cells were suspended at three different densities (low density, 0.4 x lo7 cells/ml; medium density, 0.8 X lo7 cells/ml; high density, 2 X lo7 cells/ml) in Ham’s F-12 medium with 5% fetal bovine serum, L-glutamine, and gentamicin. A 20.~1 drop of cell suspension was placed in the center of each of the eight center wells of a 24.well plate. After 1 h incubation at 37”C, 0.5 ml of medium containing 0.5% fetal bovine serum was added. Growth and differentiation factors were also added at this time. The next day, an additional 0.5 ml medium and test factors were added, and each day thereafter 0.5 ml medium was removed and exchanged for fresh medium. BMP-2B and T G F 0, and & were also renewed daily except as specified for individual experiments.
INTRODUCTION Recent investigations have shown that polypeptide growth factors may play an important role in the development and growth of cartilage and bone [ 11. The developmental sequence of matrix-induced new bone formation is morphologically similar to that seen in embryonic endochondral bone formation and adult fracture 1 To whom reprint requests should be addressed at National Institutes of Health, Building 30, Room 211, Bethesda, M D 20892. 509
0014.4827/91$3.00
Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
510
CHEN
BMP-2B was obtained as deBMP-PB and TGFO. Recombinant scribed [7] and it was active in uiuo [7]. The protein was quantitated by amino acid analysis and dissolved in 5 m M HCl with 2% bovine serum albumin and stored at 4°C. TGF 8, and&were purchased from R & D Systems (Minneapolis, MN). Alkaline phosphatase activity was Alkaline phosphatase actiuity. detected both histochemically and biochemically [lo]. For quantitation, 0.5 ml of buffer containing 0.15 M NaCl, 3 mMNaHCO,, pH 9.3, was added to the cultures. The cultures were then scraped from the dish into tubes and sonicated. Aliquots of cell extracts were incubated with p-nitrophenyl phosphate for 60 min at 37°C and the formation of p-nitrophenol was determined on a spectrophotometer at 400 nm. A unit is defined as units/mg protein. For histochemical staining, cultures were rinsed and incubated with a mixture of 0.05% naphthol AS-MX phosphate, 1% N,N-dimethylformamide, 0.01 M MgCl,, and 0.1% fast blue BB salt in 0.1 M Tris buffer, pH 8.5, at 37°C for 15 min [ll]. After color development, cultures were washedin PBS and fixedin 10% formaldehyde. Cultures were washed with phosphate-buffA&an blue staining. ered saline (PBS) and fixed in 10% formaldehyde in PBS overnight. After a rinse in 3% glacial acetic acid, pH 1, cultures were stained overnight in 0.5% Alcian blue in 3% glacial acetic acid (pH 1). Cultures were destained in 3% glacial acetic acid and stored in PBS. Cell proliferation. Cell proliferation was assessed by incorporation of [3H]thymidine into acid precipitable material. Medium was removed from the cells and replaced with Ham’s F-12 without thymidine but containing 0.2% fetal bovine serum. [3H]Thymidine was added at 5 &i/ml and cells were incubated for 2 h at 37°C. Incorporation of radioactivity was determined according to the method of Danielpour et al. [ 121. Synthesis of sulfated proteoglycans. The synthesis of sulfated proteoglycans was assessed by adding [?S]sulfate to each culture at 30 pCi/ml for 6 h. After the labeling period, the medium was removed and saved and the cells were dissolved in 6 M guanidine-HCl, pH 7.4. The fraction of radioactive sulfate incorporated into macromolecules in the cell layer and in the medium was determined as previously described [lo]. Results were normalized to the quantity of DNA and the statistical significance was analyzed using t test. DNA was quantitated using the method of Quantitation of DNA. Labarca and Paigen [13]. The same lot of DNA standard and bisbenzimide were used throughout the experiments to minimize variations. Detection of type II collagen. Cultures were washed in PBS, fixed in buffered 3% formalin for 15 min, and washed again in PBS. Endogenous peroxidase activity was inactivated in 0.5% H,O, in methanol for 1 h and nonspecific binding was blocked with 5% bovine serum albumin in 0.01 MT&buffered saline, pH 7.4. Affinity-purified goat antibody to type II collagen (Southern Biotechnology, Inc., Birmingham, AL) was applied to the cultures at 1:lO dilution (controls had diluted normal goat serum) for 1 h at room temperature. The secondary antibody was biotinylated rabbit anti-goat antibody. ABCperoxidase (Vector Labs, Inc., Burlingame, CA) was applied to detect antibody binding according to manufacturer’s directions and 3,3’diaminobenzidine was used as the peroxidase substrate. Separation of chondrocytes and nonchondrocytes from mixed cultures. Limb bud mesoderm was grown for 8 days in culture at 2 X lo7 cells in 600 ~1, then subjected to three sequential collagenase digestions (620 units/ml, Worthington, CLS2) according to the method of Lennon et al. [14]. After each digestion, single cells were separated from cell clumps using a 20-Km pore size Nitex filter. In this way, cells were separated into predominantly nonchondrogenic, mixed, and predominantly chondrogenic cell populations [ 141. These cells were replated at 8 X lo6 cells/ml in 20 pl drops in separate wells of a 24.well plate. After 1 h, 0.5 ml medium containing 0.5% fetal
ET AL bovine serum was added along with BMP-2B. as described above.
Medium
was changed
RESULTS Cultures of Limb Bud Mesoderm In the presence of 0.5% serum, the mesoderm cells formed typical cartilage nodules surrounded by a layer of fibroblast-like cells on Day 2. At this time, cartilage matrix was detected by staining with Alcian blue. The synthesis of cartilage matrix as indicated by incorporation of [35S]sulfate into proteoglycans increased after 1 day in culture at all three cell densities (Fig. 1). Proteoglycan synthesis was greatest in the highest density culture. Type II collagen was detected immunochemically in the nodules of cartilage distributed throughout the culture, but not in the internodular regions (Figs. 2a and 2~). Effects of BMP-2B In high density Synthesis of sulfated proteoglycans. cultures the response to BMP-2B at 0.1 rig/ml was undetectable. However, at 1 rig/ml of BMP-2B there was an almost twofold increase in [35S]sulfate incorporation, and more than a 40-fold increase in incorporation of [35S]sulfate into proteoglycans at the dose of 10 rig/ml of BMP-2B (data not shown). A dose of 10 rig/ml was chosen to examine the time course of response to BMP-2B at all three different cell densities (Fig. 1). After 2 days of application, the cells at medium and high densities showed a significant increase in sulfate incorporation over controls (P < 0.005). After 3 days of exposure, cells at all three densities showed significantly increased sulfate incorporation (P < 0.005), 1.5-fold (low density), 2.7-fold (medium density), and 19-fold (high density) compared to respective controls. After 4 days, the highest effect of BMP-2B was in the high density cultures (32-times that of control). Type II collagen. Cells cultured at high density without BMP-2B showed distinct cartilage nodules interspersed with noncartilage cells (Fig. 2a). However, in the presence of 10 rig/ml BMP-BB, these cells formed a nearly continuous sheet of cartilage (Fig. 2b). The uniformity of cartilage formation was reflected in the distribution of type II collagen. In cultures exposed to BMP-BB, the antibody to type II collagen bound over the entire cell layer rather than to just nodules (Fig. 2d). The incorporation of thymidine by DNA synthesis. cells was measured at all three cell densities after 1,2,3, and 4 days of culture (Table 1). During the first 2 days, exposure to 10 rig/ml BMP-2B depressed DNA synthesis by cells at all three densities (P < 0.05). However, after 3 and 4 days of exposure, the cells at high density
BMP-2B
0
AND
Window
*Or
140 0 x
r b
105
Day
Combined Factor-P
Day by FIG. 1. Time course of synthesis of sulfated proteoglycans chick limb bud mesoderm with and without added BMP-2B. (a) High density (2 X 10’ cells/ml): (b) medium density (8 X 10’ cells/ml): (c) low density (4 X lo6 cells/ml). Data represent the mean -t standard deviation for four cultures at each time point and treatment. (O), control cultures; CO), cultures treated with 10 rig/ml BMP-2B. The scale for the increase in sulfate incorporation over 4 days with BMP2B treatment greatly Qurpasses the increase in control cultures. Note that the scale for % incorporation is different at the three different densities. Synthesis was greatest in the high density cultures.
had increased d 0.05).
DNA
synthesis
compared
of Exposure
In order to determine when BMP-2B acts during the differentiation of these mesodermal cells, we applied it for two separate time periods to cells cultured at high density. In one experiment, BMP-2B was added only during the first 2 days of culture and removed. The cells were cultured in control medium for an additional 3 days. In this experiment there was a greater than 12fold increase in incorporat,ion of [35S]sulfate into proteoglycans at 10 rig/ml BMP-2B (control, 5.7 k 0.4 cpm X lO”/pg DNA; BMP-BB, 71 + 3.2 cpm X lO”lpg DNA). In a second experiment, cells were cultured in control medium during the first 3 days and 10 rig/ml BMP-2B was added only during the last 2 days. The cells increased synthesis of sulfated proteoglycans to 32-times the control value when exposed only during the last 2 days of culture (control, 4.9 _t 0.4 cpm X 103/pg DNA; BMP-BB, 156.5 t 6.8 cpm X 103/pg DNA). These results demonstrate that at later stages there is a marked increase in the response of mesodermal cells to BMP-2B.
Day
*
511
CHONDROGENESIS
to controls
(P
Alkaline phosphatase. In 0.5% serum, alkaline phosphatase activity was not detectable in control cultures, or cultures exposed to BMP-2B at 0.1 and 1 rig/ml for 5 days (data not shown). However, when cells were exposed to 10 rig/ml of BMP-2B for 5 days, alkaline phosphatase activity was detected both histochemically and biochemically. Alkaline phosphatase activity was maximal in high density cultures (low density, 0.09 + 0.01 units/mg protein; medium density, 0.12 + 0.03 units/mg protein; high density, 0.27 + 0.02 units/mg protein).
Effects of BMP-2B and Transforming (TGF-/3, or TGF-&)
Growth
Previous studies have indicated that TGF-@ may modulate cartilage or bone differentiation [15, 26, 29, 301. We studied the effects of combining transforming growth factor-p (TGF-P, or TGF-&) and BMP-2B on sulfated matrix production by chicken limb bud mesoderm. In the presence of 0.5% serum, TGF-@, and TGF8, alone depressed incorporation of sulfate into macromolecules after 5 days (Figs. 3 and 4). The addition of TGF-0, and BMP-2B together at 1 rig/ml resulted in increased sulfate incorporation 6.6-times that of controls, this was more than BMP-2B alone. However, addition of TGF-P, and BMP-2B together at 10 rig/ml resulted in a l7-fold increase of sulfate incorporation compared to the control cultures; this was less than BMP-2B alone at this higher dose. After the addition of TGF-& and BMP-2B together at 1 rig/ml, there was an increase in sulfated proteoglycan synthesis to g-times that of controls, again, higher than BMP-2B alone at this dose. With a 10 rig/ml dose of both TGF-0, and BMP-BB, sulfate incorporation increased 44-fold compared to controls (Figs. 3 and 4). Effects of BMP-2B Bud Mesoderm
on Subpopulations Cells
of Cultured
Limb
Stage 24-25 chick limb bud mesoderm contains several populations of cells and includes chondrogenic and myogenic lineages [ 161. To examine the effects of BMP2B on chondrogenic and nonchondrogenic cells, we separated cells into three populations. The first population
512
CHEN
ET AL.
FIG. 2. Influence of BMP-PB on chondrogenesis. Alcian blue staining (a and b) and distribution of type II collagen (c and d) in cultured chick limb bud mesoderm. Mesoderm cells were cultured at high density (2 X IO7 cells/ml) without BMP-2B (a and c) or with 10 rig/ml BMP-2B (b and d) for 5 days. Both the Alcian blue and anti-type II collagen antibody staining show nodules of cartilage surrounded by a network of fibroblast-like cells in controls, In the presence of BMP-2B both staining methods revealed a nearly continuous sheet of cartilage.
(fraction 1) was predominantly nonchondrogenic cells as indicated by morphology, Alcian blue staining, and synthesis of sulfated proteoglycans (Table 2); the second population (fraction 2) was a mixed population; and the third population (fraction 3) was predominantly chondrogenic cells which stained intensely with Alcian blue and synthesized large amounts of sulfated proteoglycans (Table 2). BMP-ZB was applied at 10 rig/ml for
5 days to these three populations of cells cultured at medium density. Synthesis of sulfated proteoglycans was increased nearly g-fold by addition of BMP-2B to the first population of predominantly nonchondrogenic cells. The increase was 11-fold in the second mixed population of cells, and a similar increase was observed in the third, predominantly chondrogenic, population of cells (Table 2). These results demonstrate that sulfate
BMP-2B
AND
TABLE Effects
of BMP
on Incorporation
of [3H]Thymidine
Cell Density (cells/ml)
1
by Chick
Limb
Bud
Day [“Hlthymidine
Low (4 x 106) Medium
Control BMP-2Bb Control BMP-2B Control BMP-2B
(8 x 106)
High (2 x 107)
513
CHONDROGENESIS
1
2
1150 +- 210” 540 +- 190 3110 +- 450 1440 I 190 9140 k 820 3960 +- 880
1150 -+ 80 53Ok 110 4220-+ 310 1670 f 130 12600 f 670 9260+- 1270
Cells
at High,
Medium
incorporation
and Low
Density
(CPM/Culture) 3 2490 *
4
590
198Ok 85Ok
285Ok 622Ok
1220 580 5900 * 880 16930 t 1760 25830 k 910
388Ok
360 210 230
4190 t 710 965Ok 1620 18720 + 2100
’ Data are presented as means + standard deviation for four cultures. * BMP-2B was added daily when the medium was changed at a dose of 10 rig/ml.
incorporation of both nonchondrogenic genie cells is stimulated by BMP-2B.
and chondro-
DISCUSSION
BMP-2B increased cartilage formation by chick limb bud mesoderm. Under the influence of BMP-2B at 10 rig/ml, the difference was discernible morphologically, with Alcian blue staining, and by increased radio-sulfate incorporation after 2 days of incubation when compared with control cultures. The dose eliciting a response in culture was similar to but somewhat lower than that eliciting a response in postnatal animals in uiuo. With
an in viuo osteogenesis assay of recombinant BMP-BB, a dose of 30 ng induced cartilage and 100 ng induced cartilage and significantly increased alkaline phosphatase activity [7]. In the present experiment, cells cultured at all three densities in the presence of 10 rig/ml BMP-2B formed a nearly continuous sheet of cartilage with sulfated proteoglycans and type II collagen. This was in contrast to the control cultures, which showed distinct cartilage nodules interspersed with fibroblastic cells. BMP-2B initially inhibited cell proliferation. However at later times, in cells plated at high density, proliferation was increased by BMP-2B. In medium and low density cultures, BMP-2B increased the synthesis of cartilage matrix despite decreased cell division. However, in high density cultures, the later increase in cell
2 0 TGF-Pl BMP-2B
-
- .I 1 10 -
I -
10 -
.l ,l
1 1 10 1 10 10
Dosehg/ml)
FIG. 3. Combined effects of BMP-2B and TGF-0, on proteoglycan synthesis by cultured chick limb bud mesoderm at high density. Doses of BMP-2B and TGF-@, were applied for 5 days. Data represent the mean f standard deviation of increases and decreases relative to control. Data are shown for eight cultures of controls, eight cultures at 1 rig/ml TGF-/3,, and four cultures for all other treatments. Standard deviations are depicted by error bars where large enough to appear on this scale.
TGF-P2
-
-
BMP-2B
-
110
-
1
10
1
1
10
-
-
1
10
10
Dose (g/ml)
FIG. 4. Combined effects of BMP-2B and TGF-0, on proteoglycan synthesis by cultured chick limb bud mesoderm at high density. Doses of BMP-2B and TGF-& were applied for 5 days. Data represent the mean i standard deviation of increases and decreases relative to control. Data represent four cultures for all treatments. Standard deviations are depicted by error bars where large enough to appear on this scale.
514
CHEN
TABLE
2
Effects of BMP-2B on Sulfate Proteoglycan Synthesis by Subpopulations of Chick Limb Bud Mesoderm Separated by Collagenase Digestion
Treatment Control BMP-2B Fold Increase
Cell population
?3 incorporation
Fraction
Fraction
1
,570 f 20” 4920 i 160 8.6
2
1410 5 340 15,520 t 670 11
CPM/pg Fraction
DNA 3
9580 +- 610 110,240 f 11,140 11.5
’ Data are presented as means + standard deviation of four cultures from fractions obtained after selective collagenase digestion as described under Materials and Methods.
division may have contributed to the overall increase in proteoglycan synthesis. The incorporation of [35S]sulfate into matrix macromolecules by chick limb bud mesoderm is maximal at 8 days [17]. There is an approximately three- to fourfold increase in sulfate incorporation between Days 5 and 8, followed by decreased incorporation from Day 10 through Day 16 [17]. Results presented here showed that the [35S]sulfate incorporation into proteoglycans increased about 40-fold after treatment with BMP-2B at 10 rig/ml for 5 days. Therefore, it is unlikely that the control cultures would ever reach the level of the synthesis of proteoglycan seen in BMP-2B treated cultures. This was consistent with morphological observations showing some cartilage nodules, but also a large number of noncartilage cells in control cultures, while in BMP2B treated cultures, the cells all appeared to be cartilage cells. Stage 23 to 24 chick limb bud mesoderm cells have the potential to differentiate into several tissue phenotypes, including cartilage, muscle, and bone [ 181. Seeding of such undifferentiated cells in culture at high initial plating density has been shown to promote chondrogenesis [19, 201. Therefore, cultures of embryonic chick limb endochondral bone precursors are a particularly useful way to study cells at early stages of differentiation. BMP-2B was able to stimulate cartilage formation regardless of whether it was present early or late in the culture period. However, the increase in cartilage formation when BMP-2B was added only during Days 3 and 4 was equivalent to that when BMP-2B was present throughout the culture period. Therefore, it is clear that BMP-2B acted on these cells not only early in the culture period, before the cells were overtly differentiating, but even more strongly during the later culture period, when the cells were already overtly differentiating. In addition, when chondrogenic and nonchondrogenic cells were separated from the cultures and replated, BMP-2B increased the incorporation of sulfate by both the originally nonchondrogenic and the chondrogenic
ET AI,.
cells. It appears, then, that BMP-2B can increase cartilage formation by acting at multiple stages during cartilage morphogenesis. A number of reports indicate that alkaline phosphatase activity increases prior to mineralization of matrix and this enzyme may play an obligatory role in the mineralization process [al]. Banks et al. [22] demonstrated that alkaline phosphatase activity aids in production of hydroxyapatite crystals on synthetically crosslinked collagen bathed in @-glycerophosphate solutions. It has also been suggested that alkaline phosphatase may act as a pyrophosphatase to remove the pyrophosphates which are known inhibitors of mineralization. Therefore, alkaline phosphatase activity is often used as a distinctive marker of osteogenic expression [3,23,24]. It is also expressed in resting and hypertrophic chondrocytes [ 111. In the present study, in 0.5% serum, alkaline phosphatase activity was below the level of detection in control cultures. However, when cultures were exposed to BMP-2B for 5 days at 10 rig/ml, alkaline phosphatase was detected both histochemically and biochemically. The stimulation of alkaline phosphatase activity may be an important function of BMP-2B in the maturation of chondroblasts and osteoblasts. Many well-known growth factors and cytokines have been implicated in bone growth and repair [l, 25, 261. TGF-0, is synthesized and secreted by bone cell cultures [27] and is a multifunctional autocrine regulator of bone formation [15]. In this study, we tested the effects of combined BMP-2B and TGF-0 isoforms (0, and 8,) on embryonic limb bud mesoderm. TGF-fi, or TGF-& alone, in 0.5% serum, inhibited sulfate incorporation into proteoglycans in a dose-dependent manner. BMP2B was able to overcome the inhibition of TGF-0, at 1 or at 10 rig/ml. At 1 rig/ml, the combination of TGF-0, and BMP-2B actually increased the synthesis of cartilage matrix to 3.5 times that of BMP-2B alone. However, at 10 rig/ml, the combination of TGF-P, and BMP-2B led to a level of sulfated proteoglycan synthesis, that, while still above control levels, was lower than with 10 rig/ml BMP-2B alone. Therefore, there appears to be a dosedependent synergistic action of BMP-2B and TGF-p,. While TGF-& alone inhibited sulfated proteoglycan synthesis, when TGF-6, was combined with BMP-2B either at lower dose or at higher dose, it acted synergistically with the BMP-2B. Therefore, in this experiment, at high doses, the potencies of these two forms of TGFB appear to differ when applied to limb bud mesoderm in uitro, particularly in combination with BMP-2B. This is consistent with other experiments testing effects of TGF-P, and TGF-& on cultures of limb bud mesoderm from stage 23-24 chicken embryos [as]. Although TGFp, and TGF-0, are homologous to BMP-2A and BMP-3 (about 34%), they do not induce new cartilage or bone formation when tested in an in vim osteogenesis assay in nonskeletal sites [29]. However, they do increase the
BMP-2B
AND
CHONDROGENESIS
synthesis of cartilage specific macromolecules by fibroblasts in vitro as well as modulate the expression of other extracellular matrix components by a variety of other cell types [30, 311. Recently we have tested native osteogenin (BMP-3) in this experimental system (manuscript in preparation). Osteogenin increased cartilage formation by chick limb bud mesoderm: the synthesis of sulfated proteoglycans in high density cultures increased about IO-fold under the influence of osteogenin at 10 rig/ml. Interestingly, after application of osteogenin to these cells for 5 days no alkaline phosphatase activity was detected in the presence of 0.5% serum. Therefore, the actions of each member of the BMP family of proteins may differ somewhat. Recombinant human BMP-2B induces endochondral bone differentiation in viva [7]. We have now shown that it also stimulates cartilage formation by chick limb bud mesoderm in uitro. The fact that embryonic cells respond to osteoinductive factors such as BMP-2B and osteogenin indicates these factors may play a role in cartilage development during limb morphogenesis. We thank Dr. Noreen S. Cunningham and Dr. Vishwas Paralkar for their constructive criticism ofthis manuscript. The work was supported by the USIJHS Protocol No. RO70AN. The opinions or assertions contained herein are the private ones of the authors and are not to he construed as official or reflecting the views of the DOD or the USIJHS. The experiments reported herein were conducted according to the principles set forth in the “Guide for Care and Use of Laboratory Animals” Institute of Laboratory Animal Resources, National Research Council, DHEW. Puh. No. (NIH) 78-23.
REFERENCES 1.
2. 3. 4. 5.
6.
Hauschka, P. V., Mavrakos, A. E., Zafrati, M. D., Doleman, S. E., and Klagshrun, M. (1986) J. Hiol. Chem. 261, 12,66512,674. Urist, M. R. (19fi5) Science 150, 893-899. Reddi, A. H., and Huggins. C. B. (1972) Proc. Nat/. Acad. Sci. IJSA 69, 1601~~1605. Firschein, H. E., and Urist, M. R. (1972) Clin. Orthop. R&t. Rex 84, 263-275. Wozney, J. M., Rosen, V., Celeste, A. J., Mitsock, L. M., Whitters, M. J., Kriz, R. W., Hewick, R. M., and Wang, E. A. (1988) Science 242, 1528-1534. Wang, E. A., Rosen, V., D’Alessandro, J. S., Bal:luy, M., Cordes, P., Harada, T., Israel, D. I., Hewick, R. M., Kerns, K. M., LaPan,
Received December 17. 1990 Revised March 19. 199 I
515
P., Luxenherg, D. I’., McQuaid, D., Moutsatsos, 1. K., Nave, J., and Wozney, J. M. (1990) Proc. N&l. Acad. Sci. IJSA 87, 22202224. Hammonds, R. G., Schwall, R., Dudley, A., Lai, C., Berkemeier, L., Cunningham, N., Reddi, A. H., Wood, W. I., and Mason, A. J. (1991) Mol. Endocrinol. 4, 149-155. 8. Hamburger, V., and Hamilton, H. L. (1951) J. Morphol. 88,4992. 9. Ahrens, P. B., Solursh, M., and Reiter, R. S. (1977) neu. Biol. 60, 69-82. J. L., and Reddi, A. H. (1990) Exp. Cell Res. 186, 10. Carrington, 368-373. and its Applica11. Burstone, M. S. (1962) Enzyme Histochemistry tion in the Study of Neoplasms, pp. 160-280, Academic Press, New York. 12. Danielpour, D., Dart, L. L., Flanders, K. C., Roberts, A. B., and Sporn, M. B. (1989) J. Cell Physiol. 138, 79-86. 13. Labarca, C., and Paigen, K. (1980) Anal. Biochem. 102, 344352. 14. Lennon, D. I’., Osdoby, P., Carrino, D. A., Vertel, B. M., and Caplan, A. I. (1983) J. Craniofacial Gen. DPU. Riol. 3, 235-251. 15. Centrella, M., McCarthy, T. L., and Canalis, E. (1987) J. Riol. Chem. 262, 2869-2874. 16. Zanetti, N. C., and Solursh. M. (1986) I>eu. Hiol. 113, 110-118. 17. Hascall, V. C., Oegema, 7’. R., Brown, M., and Caplan, A. I. (1976) J. Riol. Chem. 251, 3511-3519. 18. Zwilling, E. (1968) Deu. Biol. Suppl. 2, 184-204. 19. Caplan, A. I. (1970) Exp. Cell Res. 62, 341-355. 20. Umansky, R. (1966) Deu. Biol. 13,31-56. and Physiology of Bone” 21. Bourne, G. (1974) “The Biochemistry (Bourne, G. H., Ed.), 2nd ed., Vol. III, pp. 79-120. Academic Press, New York. 22. Banks, S., Wakajima, S., Shapiro, L., and Tilcuitz, 0. (1977) Science 198, 1164-1166. 23. Robison, R. (1923) Biochem. J. 17, 286-293. 24. Wlodarski, K. H., and Reddi, A. H. (1986) Calcif. Tissue Znt. 39, 382-385. 25. Simpson, E. (1984) Trends Biochem. Sci. 9.527-530. 26. Triffitt, J. T. (1987) Acta Orthop. &and. 58, 673-684. 27. Robey, P. G., Young, M. F., Flanders, K. G., Roche, N. S., Kondaiah, P., Reddi, A. H., Termine, J. D., Sporn, M. B., and Roberts, A. B. (1987) J. Cell Riol. 105, 457-463. 28. Kulyk, W. M., Rodgers, B. J., Greer, K., and Kosher, R. A. (1989) I)eu. Riol. 135, 424-430. N., and Reddi, A. H. (1987) 29. Sampath, T. K., Muthukumaran, Proc. Natl. Acad. Sci. IJSA 84, 7109-7113. 30. Seyedin, S. M., Thompson, A. Y., Bentz, H., Rosen, D. M., McPherson, J. M., Conti, A., Siegel, N. R., Galluppi, G. R., and Piez, K. A. (1986) J. Riol. C,hem. 261, 5693-5695. 31. Sporn, M. B., Roberts, A. B., Wakefield, L. M., and Assoian, R. K. (1986) Science 233, 532-534.
