DEVELOPMENTAL

BIOLOGY

139,383-3%

(1990)

Comparison between Ectoderm-Conditioned Medium and Fibronectin in Their Effects on Chondrogenesis by Limb Bud Mesenchymal Cells’ NINA

C. ZANETTI: VIRGINIA M. DRESS, AND MICHAEL SOLURSH Department of Biology, University of Iowa, Iowa

City, Iowa 52242

Accepted February 13, 1990 Limb bud ectoderm inhibits chondrogenesis by limb bud mesenchymal cells cultured at high density or on collagen gels. This ectodermal antichondrogenic influence has been postulated to function in viva in regulating the spatial patterning of cartilage and soft connective tissue in the limb. We have developed a method for preparing ectodermconditioned medium containing antichondrogenic activity. Using a simple bioassay, we have investigated some characteristics of the ectodermal products and their effects on limb bud mesenchymal cells. Inhibition of chondrogenesis by ectoderm-conditioned medium was tested on limb bud mesenchymal cells cultured on collagen gels. The antichondrogenie influence involves enhanced cell spreading and is alleviated by agents, such as cytochalasin D, that induce cell rounding. Fihronectin resembles ectoderm-conditioned medium in its ability to inhibit chondrogenesis and promote cell spreading in collagen gel cultures of limb bud mesenchymal cells. However, Western blot analysis shows that the antichondrogenic activity of ectoderm-conditioned medium is not due to fibronectin in the medium. Peptides related to the fibronectin cell-binding domain block the antichondrogenic effect of fibronectin, but not that of ectoderm-conditioned medium. On the other hand, an antibody to an integrin, as well as heparan sulfate, alleviates the antichondrogenie effects of both fibronectin and ectoderm-conditioned medium. The antichondrogenic effect of ectoderm-conditioned medium mav be mediated bv an intearin and by a cell surface heparan sulfate proteoglycan, but it does not depend directly upon fibronectin-mediated ceil spreading. 0 1990 Academic Press, Inc. INTRODUCTION

We have previously reported that ectoderms isolated from embryonic limb buds inhibit chondrogenesis by cultured limb bud mesenchymal cells (Solursh et al., 1981, 1984). A collagen gel culture system has proven useful for investigating some of the properties of this ectodermal antichondrogenic effect. Single limb bud mesenchymal cells cultured in or on hydrated collagen gels normally undergo chondrogenesis (Solursh et al, 1982), but are inhibited from doing so in the presence of limb bud ectoderm (Solursh et aZ., 1984). Moreover, gels can be preconditioned with ectoderm, such that they retain their antichondrogenic activity even after the ectoderm has been removed. This observation has suggested that ectoderm secretes a diffusible factor capable of binding to and modifying collagen gels. The altered gel becomes a nonpermissive substrate for chondrogenesis, possibly by affecting mesenchymal cell shape (Zanetti and Solursh, 1986). We have hypothesized that similar interactions between ectoderm, extracellular matrix, and mesenchymal cell shape in viva may serve as a mechanism for determining the spatial i This research was supported by NIH Research Grant HD05505 and a faculty research stipend from Siena College, Loudonville, NY. * Present address: Department of Biology, Siena College, Loudonville, NY 12211. 383

pattern of cartilage in the developing limb (Solursh et al., 1984). In this paper we report a method for obtaining ectoderm-conditioned medium that exhibits antichondrogenie activity when tested on limb bud mesenchymal cells cultured on collagen gels. We have used this culture system to investigate the mechanism by which ectoderm may regulate limb bud chondrogenesis. Because of some similarities between ectoderm-conditioned medium and fibronectin, experiments were designed to explore the possibility that the ectodermal antichondrogenie activity resembles fibronectin in its mechanism of action. MATERIALS

Limb Bud Mesenchymal

AND METHODS

Cell Cultures

Mesenchymal cell suspensions were prepared from wing buds of stage 23-24 (Hamburger and Hamilton, 1951) chick embryos (Welp Hatchery, Bancroft, Iowa), as previously described (Ahrens et al, 1977). Wing buds were dissociated in 0.1% trypsin-collagenase and the resulting cell suspension filtered through two-ply No. 20 Nitex. Collagen gel cultures were prepared as previously described (Solursh et ok, 1984). Rat tail collagen [prepared according to the method of Elsdale and Bard (1972) or 0012-1606/90 $3.00 Copyright All rights

0 1990 by Academic Press, Inc. of reproduction in any form reserved.

DEVELOPMENTALBIOLOGY V0~~~~139,1990

384

obtained from Biomedical Technologies, Inc., Cambridge, MA] was first dialyzed against l/l0 Flz stock with antibiotics, pH 4, and then diluted with distilled water to a working concentration of approximately 2 mg/ml. For preparing gels, eight parts of the dialyzed collagen, one part of fetal calf serum, and one part 10X Flz stock were quickly mixed by vortexing and then dispensed into the appropriate culture dishes. For standard “bioassays,” 0.4 ml of the collagen mixture was dispensed into each 13-mm-diameter well of a 24well tissue culture dish. In other experiments, O.&ml aliquots of the mixture were quickly spread over the surfaces of 35mm-diameter tissue culture dishes and then allowed to gel. Mesenchymal cells were plated on the gels at a cell density of 5 X lo4 cells per 13 mm gel, or 3.6 X lo5 cells per 35 mm gel. The cells were always plated in control medium [Flz medium with 10% fetal calf serum or a serum-free, defined medium (Syftestad et al., 1984)]. After 1 hr, this medium, containing any unattached cells, was withdrawn and was replaced with test medium (0.20 ml per 13-mm-diameter well or 0.5 ml per 35-mm-diameter dish). No additional feeding was given during the remainder of the culture period. All cultures were maintained in a 37°C incubator with a humidified atmosphere of 95% air, 5% COB. After 5-7 days in culture, cells were fixed for staining with Alcian blue or by immunofluorescence techniques. Preparation

of EctodermGmditioned

Medium

Ectoderms were obtained by incubating stage 21-24 chick limb buds in 0.43% trypsin-0.25% pancreatin for 30 min at 4°C and then removing the loosened ectoderms with a fine dissecting needle (Solursh et al., 1981). The ectoderms were rinsed in horse serum diluted 1:l in Ca’+- and Mg’+-free saline G, and were then transferred to cold F12 medium until ready to be placed in culture. The ectoderms were cultured on top of a small collagen gel that was surrounded but not covered by medium (either Flz medium with 10% fetal calf serum or serum-free, defined medium). The gels were prepared by dispensing 0.3 ml of collagen mixture (described above) into the center of a 35-mm- or 60-mm-diameter tissue culture dish. This volume of collagen gelled as a mound, typically covering a circular area approximately 10 mm in diameter. The dishes were then flooded with medium, after which just enough medium was withdrawn to expose the top of the mound of gelled collagen. Fifteen to twenty-five ectoderms were then placed on top of each gel and the dishes placed into the incubator. Starting on Day 5 and on alternate days thereafter, the conditioned medium was collected and replaced with fresh medium. Control medium was ob-

tained from similar derms. Fixation

cultures

that had received no ecto-

and Histology

For quantitation of cartilage nodules, cultures were fixed in 10% buffered formalin and stained with Alcian blue at pH 1 (Lev and Spicer, 1964). Some cultures were then counter-stained with Carazzi’s hematoxylin. Cultures to be stained by immunofluorescence were fixed in a 1:l mixture of methanol:acetone and then air dried. The cultures were rehydrated in phosphate-buffered saline (PBS) and then were stained with a hybridoma supernatant containing monoclonal antibody specific to chick fibronectin (B3/D6) or to type II collagen, followed by a l/300 dilution of fluorescein-conjugated IgG fraction of goat anti-mouse IgG (Cappel, West Chester, PA). After each incubation in antibody, cultures were rinsed in PBS. After the final PBS rinse, cultures were mounted in glycerin containing paraphenylene diamine (Johnson and Araujo, 1981). Western Blots Samples of control and ectoderm-conditioned medium were concentrated with an Amicon centriconmicroconcentrator and were separated by SDS-PAGE (Laemmli, 1970) on 4-16% linear gradient gels. Proteins were transferred to nitrocellulose and immunostained (Towbin et aL, 1979) with an avian-specific monoclonal antibody to fibronectin (B3/D6). The peroxidase reaction was detected with diaminobenzadine and enhanced by including NiS04 in the reaction buffer (Scopsi and Larsson, 1986). Complete transfer of proteins to nitrocellulose was verified by silver staining the gel after transfer. Materials The monoclonal antibodies JG22 (Greve and Gottlieb, 1982) and B3/D6 (Gardner and Fambrough, 1983), an avian-specific anti-fibronectin antibody, were obtained from the Developmental Studies Hybridoma Bank (NICHD Contract NOl-HD-6-2915). The JG22 antibody was obtained in the form of ascites fluid, and B3/D6 as a hybridoma supernatant. The hybridoma supernatant containing anti-type II collagen antibodies was provided by Dr. T. Linsenmayer (Linsenmayer and Hendrix, 1980). The monoclonal antibody to the collagen binding fragment of fibronectin was obtained from Telios Pharmaceuticals (La Jolla, CA). On the basis of immunofluorescence, it binds to both human and chicken plasma fibronectin. The fibronectin peptides GRGDS, GRGES, and RGDSPASSKP were obtained from Peninsula Labs

ZANETTI, DRESS, AND SOLURSH

(Belmont, CA). Laminin and human plasma fibronectin were obtained from Bethesda Research Laboratories (Gaithersburg, MD), and the chick plasma fibronectin from Biomedical Technologies, Inc. (Cambridge, MA). Dr. Ken Yamada provided the chick cell fibronectin. Chondroitin sulfate was from Miles Laboratories (Kankakee, IL); the heparin (molecular weight of 12 kDa) and the heparitin sulfate (average molecular weight of 29-45 kDa based on gel electrophoresis) were from Upjohn Co. (Kalamazoo, MI). RESULTS

Eflect of Ectoderm-Conditioned Limb Bud Mesenchymal Cells

Medium

on

Prechondrogenic limb bud mesenchymal cells normally undergo extensive chondrogenesis when plated at low density (5 X lo4 cells/l3 mm well) on rat tail collagen gels. Cultures, grown in control medium for 5-7 days, contain a mixture of cell types, including myoblasts and fibroblast-like cells (Zanetti and Solursh, 1986), as well as numerous chondrocytes with Alcian blue staining matrix (Solursh et aL, 1982) (Fig. 1). These foci are also positive by indirect immunofluorescence with antibody directed against type II collagen (Fig. 2). Typically, control cultures produce over 50 Alcian blue staining cartilage foci per well. In contrast, if limb bud mesenchymal cells are cultured on collagen but are fed ectoderm-conditioned medium, chondrogenesis is inhibited. The antichondrogenic activity varies among different batches of ectoderm-conditioned medium, but typically inhibits the number of Alcian blue foci by

Ectoderm-Conditioned

Medium

385

60-100%. Cultures fed ectoderm-conditioned medium contain numerous fibroblastic cells (Fig. l), which do not stain for the presence of type II collagen (Fig. 2). The antichondrogenic effect of ectoderm-conditioned medium seems to have only partial tissue-specificity. Medium conditioned by other embryonic tissues such as dermis, neural tube, endoderm, and dorsal ectoderm can inhibit chondrogenesis by limb bud mesenchymal cells on collagen. Both nonridge ectoderm as well as apical ectodermal ridge produce conditioned medium with antichondrogenic activity. However, medium conditioned by either of two cultured cell lines (NRK cells [Hayman et aL, 19811 and PYS cells [Pierce and Verney, 19611) has no effect on chondrogenesis by limb bud mesenchymal cells. The lack of antichondrogenic activity in medium conditioned by these epithelial cell lines argues against the possibility that this activity might simply result from depletion of essential materials. We also have noted that dialysis of ectoderm-conditioned medium (prepared from limb bud ectoderms) does not remove its antichondrogenic activity. Furthermore, although we routinely prepare ectoderm-conditioned medium in F12 medium supplemented with 10% fetal calf serum, medium containing antichondrogenic activity can also be produced by ectoderms cultured in a serum-free-defined medium (Syftestad et al., 1984). Ectoderm-conditioned medium contains slight antichondrogenic activity (approximately 25% inhibition of chondrogenesis by limb bud mesenchymal cells on collagen) by 2 days after the ectoderms have been placed in culture (data not shown). Substantial activity (~50%

FIG. 1. Limb bud mesenchymal cells cultured on collagen gels in (a) control or (b) ectoderm-conditioned medium. The cultures were fixed ‘7 days after plating and were stained with Alcian blue to reveal cartilage foci (c). Counterstaining with hematoxylin reveals other flattened (f) fibroblast-like cells and bipolar myoblasts. Note the prevalence of flattened cells in the culture fed ectoderm-conditioned medium. Bar = 100 rm.

386

DEVELOPMENTALBIOLOGY

V0~~~~139.1990

FIG. 2. Immunofluorescence (a, b) and phase contrast (c, d) micrographs of limb bud mesenchymal cells cultured on collagen gels in control (a, c) or ectoderm-conditioned (b, d) medium and stained with anti-type II collagen antibody. Cultures fed ectoderm-conditioned medium contain mostly flattened cells that do not stain for type II collagen. Bar = 100 pm.

inhibition of chondrogenesis) appears by Day 4 of culture, and after this time remains constant in medium collected every other day. Interestingly, ectoderm-conditioned gels (Solursh et aL, 1984) acquire nearly 100% of their antichondrogenic activity by Day 2 of culture (unpublished). Probably, the secreted ectodermal product(s) bind to the gel (Solursh et ab, 1984), such that only after the gel is saturated do the product(s) accumulate in the medium. The antichondrogenic activity of ectoderm-conditioned medium appears somewhat dependent upon the cell culture conditions used in the bioassay. For example, when limb bud mesenchymal cells are cultured as

high-density “micromass” cultures on plastic rather than as low-density cultures on collagen, inhibition of chondrogenesis is detected but is much less dramatic, being manifested primarily as a decrease in size of the cartilage nodules. For this reason, and also because the low-density cultures facilitate quantitation and observation of individual cell shape, we have used collagen gel cultures as our standard bioassay system for studying the properties of ectoderm-conditioned medium. In addition to inhibiting chondrogenic differentiation, ectoderm-conditioned medium also affects the shape of limb bud mesenchymal cells. This effect is first seen within 24 hr after plating. At this time, nearly all

ZANETTI,

DRESS,

AND

SOLURSH

the cells in control medium appear spherical (Fig. 3a), whereas many of the cells in ectoderm-conditioned medium have become flattened or elongated, and appear

Ebderm-Conditioned

Medium

387

spread on the gel (Fig. 3b). This difference in cell shape becomes much more apparent by Day 7 of culture, at which time most cells in control medium display the

FIG. 3. Shape of limb bud mesenchymal cells after 24 hr in culture on collagen gels in (a) control medium, (b) ectoderm-conditioned medium, (c) ectoderm-conditioned medium containing 100 pg/ml GRGDS peptide, (d) control medium containing 50 rg/ml human plasma fibronectin, (e) control medium containing 50 pg/ml human plasma fibronectin and 100 rg/ml GRGDS peptide, (f) control medium containing 50 pg/ml human plasma fibronectin and 100 pg/ml GRGES peptide. Most cells in control medium appear rounded, whereas many cells in ectodermconditioned medium are spread. The cell spreading normally occurring in fibronectin-supplemented medium does not occur in the presence of GRGDS peptide, but does occur in the presence of the control GRGES peptide. GRGDS peptide does not prevent cell spreading in ectodermconditioned medium. Bar = 100 Wm.

388

DEVELOPMENTAL BIOLOGY

typical round or polygonal morphology of chondrocytes. In contrast, cultures receiving ectoderm-conditioned medium consist predominantly of confluent, highly flattened cells (Fig. 1). To test whether the cell flattening that occurs in ectoderm-conditioned medium is related to its antichondrogenic effect, we treated some cultures with cytochalasin D, which is known to disrupt stress fibers, induce rounding, and stimulate chondrogenesis of limb bud mesenchymal cells (Zanetti and Solursh, 1984). In the presence of this drug, enhanced chondrogenesis occurred in both control cultures and those receiving ectoderm-conditioned medium, and the two types of cultures formed similar numbers of Alcian blue-staining foci (Table 1). Effect of Fibronectin

cm Limb Bud Mesenchymal

Cells

Fibronectin is known to cause flattening and dedifferentiation of chondrocytes (Pennypacker et al., 1979; West et al, 1979,1984) and to inhibit chondrogenic differentiation by high-density cultures of proximal limb bud mesenchymal cells (Swalla and Solursh, 1934). To explore the possibility that the antichondrogenic activity in ectoderm-conditioned medium might be due to fibronectin, we examined how fibronectin might affect the shape and chondrogenic differentiation of limb bud mesenchymal cells on collagen. We found that the addition of human plasma fibronectin to the culture medium of limb bud mesenchymal cells inhibits formation of Alcian blue-staining foci in a dose-dependent manner (Fig. 4). The minimum effective dosage is 1 pg/ml, which inhibits chondrogenesis by about 30%. Fibronectin concentrations of lo-50 fig/ml cause 80-100% inhi-

TABLE 1 CYTOCHALASIN D ALLEVIATES THE ANTICHONDROGENICEFFECT OF ECTODERM-CONDITIONED MEDIUM” Culture medium Control Ectoderm-conditioned Control + 2 pg/ml cytochalasin D Ectoderm-conditioned + 2 pg/ml cytochalasin D

No. Alcian blue-staining foci per dish * 62.7 (k20) 12.7 (k9)

% Inhibition of ehondrogenesis”

[Fg

ADDED

Ipgmll

FIG. 4. Dose-response curve of effect of fibronectin on chondrogenesis by limb bud mesenchymal cells on collagen gels. Cells were plated on collagen gels in control medium (serum-free medium or Fia medium). After 1 hr, the medium was withdrawn and replaced with control medium containing varying concentrations of human plasma fibronectin (BRL). Cultures were fixed after 5-7 days and were stained with Alcian blue. The percentage of inhibition of chondrogenesis was calculated as in Table 1. Data points represent the mean of three experiments.

bition of chondrogenesis, an effect equivalent to that of ectoderm-conditioned medium. Chick fibronectin also inhibits chondrogenesis in these cultures, but is effective only at doses exceeding 10 pg/ml. In contrast, lo-100 pg/ml laminin does not inhibit the chondrogenic differentiation of limb bud mesenchymal cells in collagen gel cultures. (In a representative experiment, control cultures formed 76.3 + 9 nodules, while in the presence of 30 pg/ml laminin 90 f 14 nodules formed. These data are means of triplicate cultures and are typical of three trials.) Like ectoderm-conditioned medium, fibronectin induces mesenchymal cell spreading on collagen gels. Within 12-24 hr after plating, cells in medium containing fibronectin tended to flatten or elongate (Fig. 3d), whereas most cells in control medium remained round (Fig. 3a).

79.7

Ectoderm-Conditioned Detectable Amounts

494 (+163) 631 (k190)

VOLUME 139.1990

0

LILimb bud mesenchymal cells cultured on collagen gels in control or ectoderm-conditioned medium were fixed on Day 7 and stained with Alcian blue. * Mean and standard deviation of three dishes. c % inhibition = (No. cartilage foci in control medium - No. cartilage foci in conditioned medium)/No. cartilage foci in control medium.

Medium Does Not Contain of Fibrmectin

Because the effects of fibronectin on limb bud mesenchymal cells so closely resemble the effects of ectoderm-conditioned medium, we wanted to determine whether ectoderm-conditioned medium contains fibronectin at concentrations sufficient to account for its antichondrogenic activity. We analyzed ectoderm-conditioned medium by Western blotting, using an avianspecific monoclonal antibody to fibronectin (Fig. 5). In

ZANETTI,

DRESS,

AND SOLURSH

Ectoderm-Conditioned

Medium

389

fore, staining was more pronounced in cultures treated with ectoderm-conditioned medium. Thus, immunofluorescence staining of cultures indicates that cells in collagen gel cultures produce substantial amounts of fibronectin, which remains bound to the gel. Do Agents That Interfere with Fibrmectin-Cell Fibronectin-Substrate Interactions Interfere with Antichondrogenic Activity of Ectoderm-Conditioned Medium?

1

2

3

4

5

6

FIG. 5. Western blot of chick fibronectin (lanes l-4), ectoderm-conditioned medium (lane 5) and control medium (lane 6), using an avian-specific monoclonal antibody to fibronectin. Lanes l-4 contain 100 ~1 of fibronectin solution. Lanes 5 and 6 contain the equivalent of 0.2 ml test medium as used in the bioassay system. No detectable fibronectin appears in the control or the ectoderm-conditioned media, hence these media must contain less than 2.5 rg/ml fibronectin.

our hands, the technique allows detection of as little as 0.5 pg fibronectin (lane 3). Absence of detectable staining for fibronectin in ectoderm-conditioned medium (lane 5) indicates that the medium contains less than 2.5 pg/ml fibronectin. This concentration of fibronectin causes only 40% inhibition (maximum) of chondrogenesis by limb bud mesenchymal cells (Fig. 4), and therefore cannot account for the full antichondrogenic activity (approximately 80% inhibition) normally found in ectoderm-conditioned medium. The inability of fibronectin-like peptides to interfere with the antichondrogenie effect of ectoderm-conditioned medium (discussed below) also confirms that fibronectin is not the active factor in this medium. Limb bud mesenchymal cells cultured on collagen gels do produce fibronectin. Immunofluorescence staining with an avian-specific antifibronectin antibody showed areas of positive staining in both control cultures and in cultures containing ectoderm-conditioned medium (Fig. 6). In both types of cultures, the staining pattern reveals abundant pericellular meshworks of fibronectin; areas of the gels that are devoid of cells do not stain. Although chondrocytes stain with the antibody, fibronectin meshworks are most prevalent in areas containing numerous flattened cells, and, there-

or

Although the experiments described above indicate that fibronectin is not the active antichondrogenic factor in ectoderm-conditioned medium, it was possible that fibronectin could be indirectly involved in the antichondrogenic affect. For example, an ectodermal factor might serve to enhance fibronectin-mediated spreading. Alternatively, the antichondrogenic factor in conditioned medium might resemble fibronectin in some of its molecular characteristics. In order to explore these possibilities, we ran a series of experiments in which we added to ectoderm-conditioned medium various agents thought to be capable of perturbing interactions of fibronectin with the substrate or cell surface. Standard bioassays for antichondrogenic activity revealed that some, but not all, of these agents blocked the normal effects of ectoderm-conditioned medium (Table 2). Some of the agents tested included antibodies that were directed against part of the fibronectin molecule or against integrins (Table 2). A monoclonal antibody to the collagen binding fragment of fibronectin alleviated the antichondrogenic activity of fibronectin, but had no effect on the antichondrogenic activity of ectodermconditioned medium. Another monoclonal antibody, JG22, which is directed against the common p chains of integrins (Hynes, 1987; Ruoslahti and Pierschbacher, 1987), alleviated the antichondrogenic activity of both fibronectin and ectoderm-conditioned medium, and also prevented the cell spreading that usually occurs in these media (Fig. 7). We also tested whether synthetic peptides resembling the cell-binding domain of fibronectin (Pierschbather and Ruoslahti, 1984) could alleviate the antichondrogenic activity of ectoderm-conditioned medium (Table 2). The peptide GRGDS alleviated fibronectin’s inhibition of chondrogenesis by limb bud mesenchymal cells, presumably by competing with fibronectin for the fibronectin receptor (Yamada and Kennedy, 1984). However, this peptide had no effect on the normal antichondrogenic effect of ectoderm-conditioned medium. Similarly, a decapeptide (RGDSPASSKP) containing the RGD sequence also blocked the antichondrogenic effect of fibronectin, but failed to alleviate the anti-

390

DEVELOPMENTAL BIOLOGY

VOLIJME

1.19

1Wfl

FIG. 6. Phase contrast (c, d) or immunofluorescence micrographs (a, b) of limb bud mesenchymal cells stained with an avian-specific monoclonal antibody to fibronectin. Limb bud mesenchymal cells were cultured on collagen gels in control (a, c) or ectoderm-conditioned (b, d) medium and were fixed on Day 4. In both types of cultures fibrillar arrays of fibronectin appear associated with cells as early as 24 hr after plating (not shown). After 4 days, fibronectin is most abundant in areas of numerous flattened cells that occur in ectoderm-conditioned medium. Areas of extensive fibronectin-staining also occur in control cultures. In all cultures, areas of the gels that lack cells do not stain for the presence of fibronectin. Bar = 100 Gm.

chondrogenic effect of ectoderm-conditioned medium. It is worth noting that although GRGDS prevented the normal cell flattening induced by fibronectin (Figs. 3d and 3e), it did not prevent the cell flattening induced by ectoderm-conditioned medium (Figs. 3b and 3~). A control peptide, GRGES, had little effect on cell spreading or differentiation in either fibronectin-supplemented (Fig. 3f) or ectoderm-conditioned medium. Finally, we tested whether heparin and heparitin sulfate might interfere with the effects of fibronectin or ectoderm-conditioned medium on chondrogenesis by

limb bud mesenchymal cells. Fibronectin is known to interact with heparan-containing molecules on the cell surface (Yamada et aL, 1980; Woods et al., 1984; Singer et ah, 1987b; Hedman et aL, 1982; Klebe and Mock, 1982; Johansson and Hook, 1984), possibly by influencing actin cable formation (Woods et al., 1986). We found that both heparitin sulfate and heparin alleviated the antichondrogenic effects of both fibronectin and ectoderm-conditioned medium (Table 2). In contrast, chondroitin sulfate had no effect on the normal inhibition of chondrogenesis that occurs in these cultures.

ZANETTI, TABLE

DRESS,

AND SOLURSH

2

AGENTS THAT INTERFERE WITH FN-CELL OR I rN-SUBSTRATE INTERACTIONS: EFFECTS 01 q ANTICHONDROGENIC ACTIVITY OF FIBRONECTIN OR ECTODERM-CONDITIONED MEDIUMS Standardized % inhibition chondrogenesis’

Addition

Ectodermconditioned medium

to medium

None JG22 (15 ag/ml) Peptide GRGDS (100 ag/ml) Peptide GRGES (100 pg/ml) Peptide RGDSPASSKP (200r&ml) Heparitin sulfate (100 pg/ml) Heparin (100 pg/ml) Chondroitin sulfate (100 pg/ml) Antibody to collagen-binding fragment of fibronectin

of

Control medium +20 j6g/ml HPFN

100

100

36.3 [7]' 95.1 [5]

56.0 [6] 4.5 [3]

100 [3]

81.7 [2]

100 [l]

17.8 [l]

39 [41 45.9 [2]

0 PI 0 PI

100 [2]

99.5 [2]

86.2 [5]

28.9 [3]

a Limb bud mesenchymal cells cultured on collagen gels in control or ectoderm-conditioned medium were fixed on Day 7 and stained with Alcian blue. *Inhibition of chondrogenesis was first calculated as described in Table 1, and then was standardized to an arbitrarily designated 100% inhibition in FN-supplemented or ectoderm-conditioned medium alone. Standardized % inhibition was defined as (% inhibition in treated medium)/% inhibition in untreated medium. Standardization of data from different trials was necessary because the baseline inhibition by “standard media” (ectoderm-conditioned or FN-containing medium with no additional reagents) varied among the different trails; relatively little variation occurred within a trial. “The number in brackets indicates the number of independent trials. Each data point represents the mean of at least duplicate data points from the indicated number of trials.

DISCUSSION

Ectoclerm-Conditioned Medium Inhibits Chondrogenesis and Promotes Spreading of Limb Bud Mesenchymal Cells

In previous studies we have shown that isolated limb ectoderms have the ability to inhibit chondrogenesis by cultured limb bud mesenchymal cells (Solursh et al, 1981,1984). The ability of cultured ectoderms to “condition” hydrated collagen gels (Solursh et al, 1984; Zanetti and Solursh, 1986) suggested that ectoderm produces a factor capable of diffusing through and binding to a collagen matrix. In the present study we have shown that under appropriate culture conditions, medium conditioned by ectoderm contains a similar antichondrogenic activity. Because dialysis against fresh medium does not remove this activity from serum-free ectoderm-conditioned medium, the antichondrogenic effect cannot simply reflect depletion of some essential

Ectodm-Conditioned

Medium

391

medium components, but rather must result from secretion into the medium of some ectoderm-derived molecule. The antichondrogenic effect of ectoderm-conditioned medium is best demonstrated in collagen gel cultures of limb bud mesenchymal cells. We assume that the effect of ectoderm-conditioned medium involves the same molecular mechanisms as the demonstrated antichondrogenic effect of limb ectoderm and other epithelia on limb mesenchymal cells in micromass culture (Solursh et al, 1981) and in collagen gels (Solursh et ab, 1984). The inhibition of chondrogenesis by ectoderm-conditioned medium is invariably associated with enhanced mesenchymal cell spreading, and is blocked if cytochalasin D is included in the culture medium. These results confirm our earlier studies with ectoderm-conditioned gels, in which we showed that the antichondrogenic effect of ectoderm was mediated by cell spreading, probably involving changes in the actin cytoskeleton (Zanetti and Solursh, 1986). Recently, Gregg et al. (1989) have questioned the importance of cell shape in ectodermal effects on chondrogenesis, because they observed that nonchondrogenic mesenchymal cells near ectoderms in high-density cultures do not necessarily flatten. However, this system is unusual in that vascular endothelial cells differentiate under the ectoderm (Solursh and Reiter, 1989) making comparisons to the collagen gel culture system difficult. Fibronectin ResemblesEctoderm-Conditioned Medium in Its Antichondrogenic Efect on Limb Bud Mesenchymal Cells

Because ectoderm-conditioned medium seems to affect cell-substratum interactions, we considered whether the antichondrogenic activity in this medium might be a known extracellular matrix molecule such as fibronectin or laminin. Laminin does not inhibit chondrogenesis by limb bud mesenchymal cells on collagen. Other major basal lamina components also are probably not involved, since conditioned medium from epithelial cell lines known to secrete such components does not inhibit chondrogenesis in our bioassay. Fibronectin, on the other hand, resembles ectoderm-conditioned medium in its ability to inhibit chondrogenesis in this as well as other culture systems (Swalla and Solursh, 1984). In the collagen gel culture system, inhibition of chondrogenesis by fibronectin is always accompanied by mesenchymal cell spreading. Agents that block the antichondrogenic effect of fibronectin, such as the JG22 antibody or the synthetic peptide GRGDS, cause the cells to become round. The antibody JG22 (Greve and Gottlieb, 1982), like the CSAT antibody (Neff et aL,

392

DEVELOPMENTAL

FIG. 7. Shape of limb bud mesenchymal cells after medium containing 20 rg/ml human plasma fibronectin spreading normally occurring in ectoderm-conditioned JG22 antibody. Bar = 100 pm.

BIOLOGY

VOLUME

139,199O

24 hr of culture on collagen gels in ectoderm-conditioned medium (a, b) or in control (c, d). Cultures in (b) and (d) also received 15 pg/ml JG22 antibody. Note that the cell medium or medium supplemented with fibronectin does not occur in the presence of the

1982), recognizes a 140-kDa cell surface attachment the shape and differentiation of limb bud mesenchymal cells through interactions that involve binding of complex, the avian “integrin” (Hynes, 1987; Ruoslahti fibronectin to the avian integrin and recognition by this and Pierschbacher, 1987; Buck and Horwitz, 1987; Horwitz et ab, 1985). These antibodies have been used to integrin of the RGD sequence of the exogenous fibroperturb cellular functions that depend upon cell adhenectin. sion to fibronectin (Bronner-Fraser, 1985; Hall et aZ., 1987). Similarly, synthetic peptides containing the RGD Fibronectin Is Not the Antichondrogenic Factor in sequence can mimic or compete with the cell attachEctoderm-Conditioned Media ment activity of fibronectin, and have also been used to Because fibronectin inhibited chondrogenesis by probe the biological role of cell-fibronectin interactions limb bud mesenchymal cells in a manner similar to ec(Boucaut et ak, 1984; Yamada and Kennedy, 1984; Silmedium, it was essential that we nutzer and Barnes, 1985; Hayman et al, 1985; Lash et al., toderm-conditioned 1987). Our experiments with JG22 antibody and determine whether fibronectin was the putative antichondrogenic molecule in this medium. We have asGRGDS peptides suggest that fibronectin can influence

ZANEITI, DRESS, AND SOLURSH

sayed for the presence of fibronectin in ectoderm-conditionecl medium by Western blot analysis. This method strongly suggests that fibronectin is not present in ectoderm-conditioned medium in amounts sufficient to account for its inhibition of chondrogenesis. This interpretation is further supported by the inability of GRGDS pepticles to interfere with the antichonclrogenie activity of ectoderm-conditioned medium (below). Thus, the antichonclrogenic activity is most likely derived from some other ectodermal product. It is worth noting, however, that immunofluorescence staining of collagen gel cultures reveals that mesenchyma1 cells produce extensive meshworks of fibronectinrich extracellular matrix, whether the cells are in control or in ectoclerm-conditioned medium. This fibronectin is presumably accessible for either the cells or diffusible ectoclermal products to interact with. It was therefore important to consider the possibility that such interactions might have an indirect role in ectodermal effects on chondrogenesis by limb bud mesenchymal cells on collagen. The Antichondrogenic Effect of Ectoderm-Conditioned Medium Is Not Mediated by Fibronectin-Dependent Cell Spreading If fibronectin were involved in the antichondrogenic effect of ectoderm-conditioned medium, its most likely role would be to mediate the cell spreading that accompanies the inhibition of chonclrogenesis. Therefore, we tested whether the effects of ectoclerm-conditioned medium might be modified by agents that perturb fibronectin-cell or fibronectin-substratum interactions. Neither an antibody to the collagen-binding fragment of fibronectin nor the GRGDS peptide affected the antichonclrogenic activity of ectoderm-conditioned medium. Both agents, however, were highly effective in reducing the inhibition of chonclrogenesis that normally occurs in cultures supplemented with fibronectin. The agents blocked fibronectin-mediated spreading but not spreading promoted by ectoderm-conditioned medium. Therefore, ectoclerm-conditioned medium probably inhibits chonclrogenesis and promotes mesenchymal cell spreading by some mechanism that does not directly involve fibronectin. The Antichondrogenic Efect of Ectoderm-Conditioned Medium May Be Mediated by an Integ& Other experiments in this study suggest that the active molecule in ectoclerm-conditioned medium may share some of the properties of a fibronectin-like molecule. For example, the ability of the JG22 antibody to

Ectoderm-Canditioned

Medium

393

block both the antichondrogenic and cell flattening effects of ectoderm-conditioned medium suggests that the active molecule may bind to an integrin. Alternatively, the effects of the ectodermal product may be mediated through some other molecule that interacts with an integrin. At present we cannot distinguish between these possibilities. Many of the cell adhesive proteins known to bind integrins contain an RGD sequence as their cell recognition site (refs. above); nonetheless, many of the known integrins are not inhibited by RGD peptides. For example, Staunton et aZ. (1988) have recently reported that the ligand for a leukocyte integrin LFA-1 does not contain the RGD sequence, as also found for the human laminin receptor (Gehlsen et d., 1988). Moreover, Yamada and Kennedy (1987) have shown that even ligands that do contain the RGD sequence differ somewhat in their sensitivity to various small RGD-containing peptides. Therefore, it is reasonable to propose that a molecule involved in mediating the ectoclermal effect may be an integrin, even though its activity is not blocked by the GRGDS peptide. On the other hand, the ability of JG22 to alleviate antichonclrogenic activity of ectoclerm-conditioned medium may simply reflect its ability to cause cell rounding, similar to that caused by cytochalasin D. Possible Role of Heparan Sulfate in Eflects of Fibronectin and Ectoderm-Conditioned Medium on Limb Bud Mesenchymal Cells Another property common to fibronectin and ectoderm-conditioned medium is that heparan sulfate alleviates the antichondrogenic effect that these agents have on collagen gel cultures. Both biochemical and functional studies have provided evidence for interactions between fibronectin in the extracellular matrix and heparin-containing molecules on the cell surface (Yamada et al., 1980; Woods et cd, 1984; Singer et al., 1987b; Heclman et cd., 1982; Klebe and Mock, 1982; Johansson and Hook, 1984). Woods et al. (1986) and LeBaron et al. (1988) have shown that although the cell binding fragment of fibronectin mediates initial cell attachment, formation of focal adhesions and stress fibers requires the heparin binding domain; not all cell types, however, seem to have this requirement (Singer et ah, 1987a). Cell shape, and specifically the configuration of the actin cytoskeleton, is known to influence chonclrogenesis (Zanetti and Solursh, 1984). One possible explanation for the effects of fibronectin and heparan sulfate on limb bud mesenchymal cells is that fibronectin-induced stress fibers and/or focal adhesions cause cell flattening and inhibit chonclrogenesis; heparan sulfate, by competing with cell surface heparan

DEVELOPMENTAL BIOLOGY

394

sulfate proteoglycans for the heparan sulfate binding domain of fibronectin, may prevent these structures from forming, and therefore permit chondrogenesis in the presence of fibronectin. In this context, it is worth noting that exogenous heparan sulfate stimulates chondrogenesis by limb bud mesenchymal cells in micromass culture (San Antonio et ab, 1987). The ability of heparan sulfate to alleviate the antichondrogenic effect of ectoderm-conditioned medium suggests that the active ectodermal factor, like fibronectin, may have both cell- and heparin-binding domains, both of which are necessary to induce cell flattening and inhibit chondrogenesis. Alternatively, the antichondrogenic effect of conditioned medium may indirectly require the participation of fibronectin, which could bind to the cell via a cell surface heparan sulfate proteoglycan. Such a mechanism would be consistent with our observations that the antichondrogenic effect is not affected by agents that perturb cell interactions with the cell-binding domain of fibronectin (above). Saunders and Bernfield (1988) have described an example of a cell surface proteoglycan of mouse mammary epithelial cells that binds fibronectin in this manner, and recent studies in this laboratory have demonstrated the presence of this proteoglycan on limb bud mesenchymal cells (Solursh et czb,1990). In conclusion, we have developed a procedure for preparing conditioned medium that contains an ectoderma1 product capable of promoting cell spreading and inhibiting chondrogenesis by limb bud mesenchymal cells on collagen. This product is not fibronectin, although it resembles fibronectin in its effects on limb bud cells, and an integrin may be involved in its activity. Further attempts at identification and characterization of this molecule are merited because of its potential importance in establishing the spatial patterning of cartilage and soft connective tissue in the limb (Solursh, 1984). The methods described here for producing ectoderm-conditioned medium and the simple bioassay for its activity should facilitate such endeavors. REFERENCES AHRENS, P. B., SOLURSH, M., and REITER, R. (1977). Stage-related capacity for limb chondrogenesis in cell culture. Deu. Biol60,69-82. BOUCAIJT, J.-C., DARRIKERE, T., POOLE, T. J., AOYAMA, H., YAMADA, K. M., and THIERY, J. P. (1984). Biologically active synthetic peptides as probes of embryonic development: A compatible peptide inhibitor of fibronectin function inhibits gastrulation in amphibian embryos and neural crest cell migration in avian embryos. J. Cell Biol.

99,1822-1830.

BRONNER-FRASER, M. (1985). Alterations in neural crest migration by a monoclonal antibody that affects cell adhesion. J. CeU Biol 101, 610-617.

VOLUME 139,199O

BUCK, C. A., and HORWITZ, A. F. (1987). Cell surface receptors for ECM molecules. Annu. Rev. Cell Biol. 3,179-207. ELSDALE, T., and BARD, J. (1972). Collagen substrate for studies on cell behavior. J. Cell Biol. 54,626-637. GARDNER, J. M., and FAMBROUGH, D. M. (1983). Fibronectin expression during myogenesis. J. Cell Biol 96,474-485. GEHLSEN, K. R., DILLNER, L., ENGVALL, E., and RUOSLAHTI, E. (1988). The human laminin receptor is a member of the integrin family of cell adhesion receptors. Science 241,1228-1229. GREGG, B. C., ROWE, A., BRICKNELL, P. M., and WOLPERT, L. (1989). Ectodermal inhibition of cartilage differentiation in micromass culture of chick limb bud mesenchyme in relation of gene expression and cell shape. Development 10.5,769-778. GREVE, J. M., and GOTTLIEB, D. I. (1982). Monoclonal antibodies which alter the morphology of cultured chick myogenic cells. J. Cell Bie them

18.221-229.

HALL, D. E., NEUGEBAUER, K. M., and REICHARDT, L. F. (1987). Embryonic retinal cell response to extracellular matrix proteins: Developmental changes and effects of the cell substratum attachment antibody (CSAT). J. Cell Bid 104,623-634. HAMBURGER, V., and HAMILTON, H. L. (1951). A series of normal stages in the development of the chick embryo. J. Mwrphol. 88, 49-92.

HAYMAN, E. G., ENGVALL, E., and RUOSLAHTI, E. (1981). Concomitant loss of cell surface fibronectin and laminin from transformed rat kidney cells. J. CeU Biol. 88,352-357. HAYMAN, E. G., PIERSCHBACHER, M. D., and RUOSLAHTI, E. (1985). Detachment of cells from culture substrate by soluble fibronectin peptides. J. Cell Biol. 100.1948-1954. HEDMAN, K., JOHANSSON, S., VARHO, T., KJELLER, L., VAHERI, A., and HOOK, M. (1982). Structure of the pericellular matrix: association of heparin and chondroitin sulfates with fibronectin-procollagen fibers. CeU 28,663-671. HORWITZ, A., DUGGAN, K., GREGGS, R., DECKER, C., and BUCK, C. (1985). The cell substrate attachment antigen has properties of a receptor for laminin and fibronectin. J. CeU Biol. 101,2134-2144. HYNES, R. 0. (1987). Integrins: A family of cell-surface receptors. Cell 48,549-554. JOHANSSON, S., and HOOK, M. (1984). Substrate adhesion of rat hepatocytes: On the mechanism of attachment to fibronectin. J. Cell Biol.

98,810-817.

JOHNSON, G. D., and ARAUJO, G. M. (1981). A simple method of reducing the fading of immunofluorescence during microscopy. J. Immunol.

Methods

43,349-350.

KLEBE, R. J., and MOCK, P. J. (1982). Effect of glycosaminoglycans on fibronectin-mediated cell attachment. J. CeU Physiol 112.5-g. LAEMMLI, U. K. (1970). Cleavage of structural proteins during assembly of the head of bacteriophage T4. Nature (London) 227,680~685. LASH, J. W., LINASK, K. K., and YAMADA, K. M. (1987). Synthetic peptides that mimic the adhesive recognition signal of fibronectin: Differential effects on cell-cell and cell-substratum adhesion in embryonic chick cells. Da? BioL 123,411-420. LEBARON, R. G., ESKO, J. D., WOODS, A., JOHANSSON,S., and HOOK, M. (1988). Adhesion of glycosaminoglycan-deficient Chinese hamster ovary cell mutants to fibronectin substrata. J. Cell Biol. 106, 945-952.

LEV, R., and SPICER, S. S. (1964). Specific staining of sulfate groups with Alcian blue at low pH. J. Histochem. Cytochem 12,309. LINSENMAYER, T. F., and HENDRIX, M. J. C. (1980). Monoclonal antibodies to connective tissue macromolecules: Type II collagen. Biothem

Biophys.

Res. Commun

92,440-446.

NEFF, N. T., LOWREY, C., DECKER, C., TORAR, A., DAMSKY, C., BUCK, C., and HORWITZ, A. F. (1982). A monoclonal antibody detaches embryonic skeletal muscle from extracellular matrices. J. CeU Biol. 95, 654-666.

ZANETTI,

DRESS,

AND SOLURSH

PENNYPACKER, J. P., HASSELL, J. R., YAMADA, K. M., and PRATT, R. M. (1979). The influence of an adhesive cell surface protein on chondrogenic expression in vitro. Exp. Cell Res. 121,411-415. PIERCE, G. B. JR., and VERNEY, E. L. (1961). An in vitro and in tiwo study of differentiation in terato-carcinoma. Cancer 14,101’7-1029. PIERSCHBACHER, M. D., and RUOSLAHTI, E. (1984). The cell attachment activity of fibronectin can he duplicated by small synthetic fragments of the molecule. Nature 309,30-33. RUOSLAHTI, E., and PIERSCHBACHER, M. D. (1987). New perspectives in cell adhesion: RGD and integrins. Science 238, 491-498. SAN ANTONIO, J. D., WINSTON, B. M., and TUAN, R. S. (1987). Regulation of chondrogenesis by heparan sulfate and structurally related glycosaminoglycans. Dev. BioL 123,17-24. SCOPSI, L., and LARSSON, L. I. (1986). Increased sensitivity in peroxidase immunocytohistochemistry. Histochemistry 84,221-230. SAUNDERS, S., and BERNFIELD, M. (1988). Cell surface proteoglycan binds mouse mammary epithelial cells to fibronectin and behaves as a receptor for interstitial matrix. J. CeU BioL 106,423-430. SILNUTZER, J. E., and BARNES, D. W. (1985). Effects of fibronectin-related peptides on cell spreading. In Vitro Cell. and Dev. Biol. 21, 73-78. SINGER, I. I., KAWKA, D. W., SCOTT, S., MUMFORD, R. A., and LARK, M. W. (1987a). The fibronectin cell attachment sequence Arg-GlyAsp-Ser promotes focal contact formation during early fibroblast attachment and spreading. J. Cell BioL 104,573-584. SINGER, I. I., SCOTT, S., KAWKA, D. W., and HASSELL, J. R. (198713). Extracellular matrix fibers containing fibronectin and basement membrane heparan sulfate proteoglycan coalign with focal contacts and microfilament bundles in stationary fibroblasts. Exp. Cell Res. 173,558-571. SOLURSH, M. (1984). Ectoderm as a determinant of early tissue pattern in the limb bud. CeU lX#er. 15,17-24. SOLURSH, M., and REITER, R. S. (1989). Comparison of effects of ectoderm, FGF, and TGF-@ on cell differentiation by limb bud cell cultures. Anat. Rec. 223,109A. SOLURSH, M., JENSEN, K. L., ZANETTI, N. C., LINSENMAYER, T. F., and REITER, R. S. (1984). Extracellular matrix mediates epithelial effects on chondrogenesis in vitro. Dev. BioL 105,451-457. SOLURSH, M., LINSENMAYER, T. F., and JENSEN, K. L. (1982). Chondrogenesis from single limb mesenchyme cells. Dev. Biol. 94, 259-264.

SOLURSH, M., REITER, R. S., JENSEN, K. L., KATO, M., and BERNFIELD, M. R. (1990). Transient expression of a cell surface heparan sulfate proteoglycan (Syndecan) during limb development. Dev. BioL, in press. SOLURSH, M., SINGLEY, C. T., and REITER, R. S. (1981). The influence of

Ectcderm-Conditioned

Medium

395

epithelia on cartilage and loose connective tissue formation by limb mesenchyme cultures. Dev. BioL 86,471-482. STAUNTON, D. E., MARLIN, S. D., STRATOWA, C., DUSTIN, M. L., and SPRINGER,T. A. (1988). Primary structure of ICAM- demonstrates interaction between members of the immunoglobulin and integrin supergene families. CeU 52,925-933. SWALLA, B. J., and SOLURSH, M. (1984). Inhibition of limb chondrogenesis by fibronectin. ~erentiation 26,42-48. SYFTESTAD, G. T., TRIFFITT, J. T., URIST, M. R., and CAPLAN, A. I. (1984). An osteo-inductive bone matrix extract stimulates the in vitro conversion of mesenchyme into chondrocytes. Calcif: Tissue Int.

36,625-627.

TOWBIN, H., STAEHELIN, T., and GORDON, J. (1979). Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and some applications. Proc NatL Acad Sci USA 76,4350-4354. WEST, C. M., DE WEERD, H., DOWDY, K., and DE LA PAZ, A. (1984). A specificity for cellular fibronectin in its effect on cultured chondroblasts. Ll$erentiatim 27,67-73. WEST, C. M., LANZA, R., ROSENBLOOM,J., LOWE, M., and HOLTZER, H. (1979). Fibronectin alters the phenotypic properties of cultured chick embryo chondroblasts. CeU 17.491-501. WOODS, A., COUCHMAN, J. R., JOHANSSON, S., and HOOK, M. (1986). Adhesion and cytoskeletal organization of fibroblasts in response to fibronectin fragments. EMBO J. 5,665-670. WOODS, A., HOOK, M., KJELLEN, L., SMITH, C. G., and REES, D. A. (1984). Relationship of heparan sulfate proteoglycans to the cytoskeleton and extracellular matrix of cultured fibroblasts. J. Cell BioL 99,1743-1753.

YAMADA, K. M., and KENNEDY, D. W. (1984). Dualistic nature of adhesive protein function: Fibronectin and its biologically active peptide fragments can autoinhibit fihronectin function. J. Cell BioL 99, 29-36.

YAMADA, K. M., and KENNEDY, D. W. (1987). Peptide inhibitors of fibronectin, laminin, and other adhesion molecules: Unique and shared features. J. Cell PhysioL 130,21-28. YAMADA, K. M., KENNEDY, D. W., KIMATA, K., and PRATT, R. M. (1980). Characterization of fibronectin interactions with glycosaminoglycans and identification of active proteolytic fragments. .I. BioL Chem.

255,6055-6063.

ZANETTI, N. C., and SOLURSH, M. (1984). Induction of chondrogenesis in limb mesenchymal cultures by disruption of the actin cytoskeleton. J. Cell BioL 99, 115-123. ZANETTI, N. C., and SOLURSH, IyI. (1986). Epithelial effects on limb chondrogenesis involve extracellular matrix and cell shape. Dev. Biol. 113, 110-118.

Comparison between ectoderm-conditioned medium and fibronectin in their effects on chondrogenesis by limb bud mesenchymal cells.

Limb bud ectoderm inhibits chondrogenesis by limb bud mesenchymal cells cultured at high density or on collagen gels. This ectodermal antichondrogenic...
7MB Sizes 0 Downloads 0 Views