Abstract
Under standard culture conditions, chondrogenic expression by stage-21 embryonic chick limb bud mesenchyme is dependent upon high cell plating densities. Alternatively, when cultured in suspension aggregating limb bud cells differentiate exclusively as cartilage. We have previously demonstrated that the aggregation of prechondrogenic limb bud cells is specifically mediated by a Ca2+-dependent mechanism. In the present paper, we examine the involvement of calcium cations in chondrogenic expression in vitro. During cartilage differentiation, we demonstrate that limb bud cells elevate their intracellular Ca2+ levels to achieve a conserved plateau level. This increase in intracellular Ca2+ levels does not occur in sparse cell cultures, which also fail to demonstrate cartilage differentiation. Although elevation of extracellular Ca2+ concentration effects precocious chondrogenesis, ultimately this is substantially lower than in control cultures. In contrast, elevation of intracellular Ca2+ levels by the addition of 0·1 μM-A23187 readily stimulates precocious and extensive cartilage differentiation. 0· μM-A23187 initially elevates intracellular Ca2+ levels to that required for cartilage differentiation but this then continues to increase concomitant with a reduction in cartilage nodule size. 10 μM-retinoic acid completely inhibits chondrogenesis in vitro and elevates intracellular Ca2+ to particularly high levels. Our data indicate the central role of controlled intracellular Ca2+ levels to normal chondrogenic expression. Deviation from this level by cells that either fail to achieve or that exceed it inhibits subsequent cartilage development, and can cause a loss of phenotypic expression by differentiated cartilage.
Introduction
Within the developing avian limb bud, areas of prospective skeletal development are first identified by both an increase in relative cell density (Fell, 1925; Thorogood & Hinchliffe, 1975) and an enhanced immunolocalization of collagen type I and fibronectin (Dessau, von der Mark, von der Mark & Fischer, 1980). This pattern persists until stage 25 (Hamburger & Hamilton, 1951) and is followed by the de novo synthesis and accumulation of a cartilage-specific extracellular matrix (von der Mark & Conrad, 1979).
Cell culture analysis indicates that avian limb bud chondrogenesis occurs by a series of distinct steps. The first clonable cartilage cells can be isolated from stage-25 embryonic chick limb buds (Solursh & Reiter, 1975). Between stages 21 and 24, limb bud chondrogenesis in vitro is dependent upon high cell density (micromass) culture (Solursh, Ahrens & Reiter, 1978). Alternatively, when cultured in suspension the aggregating limb bud cell subpopulation differentiates exclusively as cartilage (Levitt & Dorfman, 1972; von der Mark & von der Mark, 1977). Such observations suggest that the selective association amongst cells with the same developmental fate promotes their chondrogenic potential. Consistent with this, Lewis, Pratt, Pennypacker & Hassell (1978) have demonstrated that retinoic acid not only abolishes limb bud chondrogenesis in micromass culture but also elicits an alteration in the pattern of cell surface proteins. However, direct interaction with the extracellular matrix is also important since isolated stage-22 to -23 limb bud cells are capable of undergoing chondrogenesis when cultured in collagen gels (Solursh, Linsenmayer & Jensen, 1982).
Prior to stage 20, limb bud mesenchyme cultured at high density only undergoes chondrogenesis following treatment with dibutyryl cAMP (Ahrens, Solursh & Reiter, 1977; Solursh, Reiter, Ahrens & Vertel, 1981). Thus, although extensive intercellular association is required for the penultimate stages of chondrogenesis, additional factors are necessary to initiate cartilage development. The mechanism of dibutyryl cAMP action remains to be elucidated although its stimulatory effect on older limb bud mesenchyme is mimicked by first culturing the cells in suspension for a brief period (Solursh et al. 1978). Since suspension culture necessarily specifies a rounded cell morphology, dibutyryl cAMP could mediate its effect directly on cell shape via the cytoskeleton.
We have previously demonstrated that the aggregation of limb bud cells that precedes their differentiation into cartilage is specifically governed by a Ca2+-dependent adhesion mechanism (Bee & von der Mark, 1987). Since intracellular levels of cAMP are intimately associated with those of calcium, in the present paper we have examined the relationship between this cation and cartilage differentiation in vitro.
Materials and methods
Preparation of cells
All embryos used in this study were from White Leghorn eggs incubated in a forced-draught incubator at a temperature of 38 ± 0·5°C. Embryos were staged according to the developmental table of Hamburger & Hamilton (1951). Limb buds were dissected from stage-21 chick embryos in calcium- and magnesium-free Tyrode’s solution, pH7-4. They were dissociated in the same solution containing 0·1% trypsin (Sigma Chem. Co., Poole, England) and 0·1 % collagenase (Sigma Chem. Co.) by constant agitation for 30 min at 37 °C. The resulting cell suspension was diluted with an equal volume of Ham’s F12 nutrient medium supplemented with 10 % fetal calf serum and antibiotics (F12S10; all components from Flow Laboratories, Rick-mansworth, England), and filtered through Nitex to remove cell aggregates. Dissociated single cells were collected by centrifugation at 460g for 10 min, washed twice with F12S10 and finally resuspended at 2×107 cells ml−1 in F12S10.
Preparation of primary cultures
Micromass cultures of limb bud cells were established according to Solursh et al. (1978). Briefly, 10 μl samples containing approximately 2×105 cells were applied to 35 mm plastic tissue culture dishes. Cells were allowed to attach for 1 h at 37 °C in a humidified atmosphere containing 5 % CO2. After this period, 2 ml F12S10 was added to each of the dishes containing control cultures. Each 35 mm dish contained five micromass cultures. To assess its effect on limb bud chondrogenesis, cultures were maintained in F12S10 supplemented with CaCl2 to a final concentration of 5 mM. Similarly, retinoic acid (all trans; Sigma Chem. Co.) was dissolved in 95 % ethanol (1·5 mg ml−1): 200 μl of this solution was added to 100 ml F12S10. The calcium ionophore A23187 (Sigma Chem. Co.) was dissolved in 95% ethanol and then diluted with F12S10. We first examined the effect of A23187 concentration on limb bud cells and differentiated cartilage behaviour. The majority of the experiments reported here were performed at a final concentration of 0T/ZM-A23187. In addition, to examine the effect of either elevated extracellular Ca2+ or A23187 on established limb bud cultures, in certain experiments cultures were maintained for the first three days in standard F12S10. After this period, they were fed medium supplemented with either component for the remaining culture period. For sparse cultures, 105 cells were plated in 3 ml medium in 60 mm tissue culture dishes. Sparse cultures were fed fresh F12S10 5 h after initial plating. All cultures were fed daily with fresh F12S10.
Identification of cartilage differentiation
Cultures were washed three times in phosphate-buffered saline, pH 7·4 and fixed in Karnovsky’s fixative for Ih at 4°C. Following fixation, cultures were washed with distilled water and stained for 30min with 10% alcian blue, pHl. Stained cultures were differentiated and dehydrated through 95 % ethanol, absolute ethanol and mounted in glycerol. Stained cultures were photographed on a Wild dissecting microscope using a red filter.
Measurement of intracellular Ca2+
Limb bud cultures were incubated with F12S10 supplemented with 10 μCi ml−1 [45Ca]CaCl2 (Amersham International, Amersham, England) for 18 h. After the labelling period, cultures were washed three times with phosphate-buffered saline supplemented with calcium and magnesium and then twice with ice-cold 0·9% NaCl: 5 mw-CaCl2: lOmM-Hepes, pH 7-4. Cells were scraped off the dish in lml NaCl:CaCl2:Hepes and centrifuged in a Beckman microfuge for 30 s. The pellet was resuspended in NaCl: CaCl2:Hepes and centrifuged twice more before resuspending in 200 μl 5mM-CaCl2. Cells were lysed in 5MM-CaCl2 for 1·5 h at 4°C and then centrifuged at 10000g for 20min (Bareis, Hirata, Schiffmann & Axelrod, 1982). Pelleted material was discarded and aliquots of the supernatant counted for radioactivity in 10 ml liquid scintillation fluid (Amersham).
Equivalent unlabelled cultures were detached and dissociated with 0·1% trypsin: 0·1% collagenase in calcium- and magnesium-free Tyrode’s solution, pH7·4. Dispersed cells were collected by centrifugation and resuspended in Tyrode’s solution. Total cell number per culture was determined by counting cells with a haemocytometer. Each experiment was performed in triplicate. Within each experiment, both intracellular 45Ca levels and total cell number were determined from three parallel dishes each containing five micromass cultures. Mean 45Ca levels were expressed with respect to mean cell number, each determined from a total of nine culture dishes (45 micromass cultures). Variance of 45Ca levels per unit cell number was calculated according to the Taylor series approximation to the variance of a ratio (Kendall & Stuart, 1977). Variance between parallel experiments was not significant.
Results
Cartilage differentiation in control cultures
Micromass cultures of stage-21 chick limb bud mesenchyme were fixed and stained with alcian blue at daily intervals commencing on day 1 (Fig. 1A). Although background staining is observed on day 2 (Fig. IB), the first distinct nodules of cartilage are detected on day 3 (Fig. 1C). Extensive chondrogenesis is demonstrated on day 4 (Fig. ID) and this becomes more prominent through days 5 and 6 (Fig. 1E,F, respectively).
Cartilage differentiation in control micromass cultures. Cultures established from stage-21 embryonic chick limb buds were fixed and stained with alcian blue at daily intervals (A–F). Slight staining of the culture on day 2 (B) becomes more prominent on day 3 (C). Numerous cartilage nodules are present on day 4 (D) and become more extensive on days 5 (E) and 6 (F).
Cartilage differentiation in control micromass cultures. Cultures established from stage-21 embryonic chick limb buds were fixed and stained with alcian blue at daily intervals (A–F). Slight staining of the culture on day 2 (B) becomes more prominent on day 3 (C). Numerous cartilage nodules are present on day 4 (D) and become more extensive on days 5 (E) and 6 (F).
Cultures were labelled with [45Ca]CaCl2 for 18 h and total uptake into the free intracellular pool determined for each of the days of in vitro development. Average total cell numbers were obtained from parallel, unlabelled cultures and expressed with respect to the mean of intracellular 45Ca2+ levels. This value was further normalized to 106 cells. Fig. 2 demonstrates that during micromass culture under control conditions 45Ca2+ levels are initially low, increase slightly on day 2 and then decrease on day 3. Intracellular 45Ca2+ levels increase linearly concomitant with the development and progression of cartilage differentiation. By day 5, this increase starts to level off and day 6 and 7 cultures show variation about a plateau level of 45Ca2+ of approximately 9500 cts min−110−6 cells. This plateau level is further demonstrated by day 8 and 9 micromass cultures (data not shown) indicating that by culture day 6 cells have achieved the stable levels of intracellular Ca2+ characteristic of chondrogenic expression.
Intracellular 45Ca2+ levels in control micromass cultures. Total 45Ca2+ uptake was corrected for cell number and expressed per 106 cells. A slight elevation on day 2, during condensation, is followed by a decrease in intracellular Ca2+ levels. However, as cells develop into distinct cartilage nodules levels are linearly elevated and then stabilize levels of 45Ca2+ at approximately 104cts min−110−6 cells. Error bars indicate +S.D.
Intracellular 45Ca2+ levels in control micromass cultures. Total 45Ca2+ uptake was corrected for cell number and expressed per 106 cells. A slight elevation on day 2, during condensation, is followed by a decrease in intracellular Ca2+ levels. However, as cells develop into distinct cartilage nodules levels are linearly elevated and then stabilize levels of 45Ca2+ at approximately 104cts min−110−6 cells. Error bars indicate +S.D.
When cultured at low densities under standard conditions, stage-21 limb bud cells fail to differentiate into cartilage. Analysis of 45Ca2+ uptake into the intracellular pool demonstrates that intracellular calcium is initially very low (Fig. 3). Over the succeeding days this level increases gradually and begins to plateau on culture day 6 (Fig. 3). At all stages, the level of intracellular 45Ca2+ is significantly lower in sparsely plated cultures compared with corresponding micromass cultures which readily demonstrate cartilage development (Figs 1, 2). While micromass cultures increase their intracellular 45Ca2+ levels from approximately 5×103cts min−110−6 cells on day 1 to over 9×103cts min−110−6 cells on day 7, sparsely plated cells demonstrate an increase from 103cts min−1 to less than 4×103cts min−110−6 cells over the same period.
Intracellular 45Ca2+ levels in limb bud cells cultured at low density. When cultured at low density, stage-21 limb bud cells fail to differentiate into cartilage. Under these conditions, intracellular 45Ca2+ levels are initially very low and are not elevated to those achieved by cells cultured at high density and differentiating into cartilage. Over the culture period, intracellular 45Ca2+ increases gradually to achieve a stable level of 45Ca2+ of 3750 cts min−110−6 cells - significantly less than that demonstrated by micromass cultures. Scale bars indicate +S.D.
Intracellular 45Ca2+ levels in limb bud cells cultured at low density. When cultured at low density, stage-21 limb bud cells fail to differentiate into cartilage. Under these conditions, intracellular 45Ca2+ levels are initially very low and are not elevated to those achieved by cells cultured at high density and differentiating into cartilage. Over the culture period, intracellular 45Ca2+ increases gradually to achieve a stable level of 45Ca2+ of 3750 cts min−110−6 cells - significantly less than that demonstrated by micromass cultures. Scale bars indicate +S.D.
Elevated extracellular calcium
Micromass cultures maintained in F12S10 supplemented with calcium to a final concentration of 5 mM were fixed and stained with alcian blue at daily intervals commencing on day 1. Compared with control cultures (Fig. 1), cultures maintained in high extracellular Ca2+ demonstrated a few prominent nodules of cartilage on day 2 (Fig. 4B). This precocious development of cartilage was more pronounced on day 3 (Fig. 4C). It should be noted that under such conditions, cartilage develops in the centre of the culture and as dispersed nodules at the periphery. These two regions are separated by an area relatively sparse in cartilage nodule development. This pattern continues to predominate through culture days 4, 5, and 6 (Fig. 4D,E,F, respectively). In addition, the ultimate amount of cartilage development demonstrated on day 6 (Fig. 4F) is considerably less than in control cultures.
The effect of elevated extracellular Ca2+ levels on cartilage development. In the presence of 5mM-Ca2+, the first distinct cartilage nodules are demonstrated on day 2 (B). These become more distinct on day 3 (C) as a central cluster and a peripheral ring. This pattern of cartilage is most distinct on days 5 (E) and 6 (F), demonstrating a banded region with a low incidence of cartilage development.
The effect of elevated extracellular Ca2+ levels on cartilage development. In the presence of 5mM-Ca2+, the first distinct cartilage nodules are demonstrated on day 2 (B). These become more distinct on day 3 (C) as a central cluster and a peripheral ring. This pattern of cartilage is most distinct on days 5 (E) and 6 (F), demonstrating a banded region with a low incidence of cartilage development.
Elevated intracellular calcium
Cultures of stage-21 embryonic chick limb bud mesenchyme were maintained in F12S10 supplemented with 1, 0·1 or 0·01 μM-A23187 for the first 48 h. After this period they were fed control medium and fixed and stained on day 6. 1 μM-A23187 caused a marked inhibition of cartilage development (Fig. 5A) with only a few discrete alcian blue-positive nodules developing at the periphery of the culture. In contrast, following treatment with either 0·1μM- (Fig. 5B) or 0·01 μM- (Fig. 5C) A23187, extensive cartilage development occurred. Similarly, the addition of 0·1 μM- A23187 on day 3 to cultures previously maintained in control medium permits normal chondrogenic expression during the remaining culture period (data not presented).
The effect of A23187 concentration on cartilage development. Micromass cultures were maintained in 1 μM (A), 0·1 μM (B), or 0·01 μM (C) A23187 for the first 48 h and fixed 4 days later on culture day 6. Such treatment with 1 μM-A23187 almost completely abolished cartilage differentiation (A), while either 0·1 μM (B) or 0 01 μM (C) did not exert a detectable effect on normal culture development.
The effect of A23187 concentration on cartilage development. Micromass cultures were maintained in 1 μM (A), 0·1 μM (B), or 0·01 μM (C) A23187 for the first 48 h and fixed 4 days later on culture day 6. Such treatment with 1 μM-A23187 almost completely abolished cartilage differentiation (A), while either 0·1 μM (B) or 0 01 μM (C) did not exert a detectable effect on normal culture development.
When maintained in 0·1 μM-A23187 throughout the culture period (Fig. 6) the ionophore stimulates extensive precocious cartilage differentiation. Although faint, extensive alcian blue staining of the cultures is demonstrated on day 1 (Fig. 6A) together with a number of small dispersed nodules. By day 2, numerous highly stained cartilage nodules are present throughout the culture (Fig. 6B) and the nodules become more extensive on day 3 (Fig. 6C). However, this initial stimulation of chondrogenesis is followed by a progressive reduction in nodule size and number through days 4, 5 and 6 (Fig. 6D,E,F, respectively).
The effect of continuous exposure to 01 fZM-A23187 on cartilage differentiation. When added on day 0, 0·1 μM-A23187 stimulates alcian-blue staining of day 1 cultures (A). By day 2 (B) numerous densely stained nodules are demonstrated. These are more prominent on day 3 (C), a pattern which persists on day 4 (D), and is followed by a progressive reduction in the size of the cartilage nodules through days 5 (E) and 6 (F).
The effect of continuous exposure to 01 fZM-A23187 on cartilage differentiation. When added on day 0, 0·1 μM-A23187 stimulates alcian-blue staining of day 1 cultures (A). By day 2 (B) numerous densely stained nodules are demonstrated. These are more prominent on day 3 (C), a pattern which persists on day 4 (D), and is followed by a progressive reduction in the size of the cartilage nodules through days 5 (E) and 6 (F).
The effect of 0·1 μM-A23187 on limb bud mesenchyme intracellular 45Ca2+ levels was analysed with respect to culture development (Fig. 7). Day 1 cultures demonstrate levels of 45Ca2+ of approximately 9750 cts min−1 10−6 cells. This decreases by approximately 40 % on day 2, achieving a level which is maintained through day 3. After this period, intracellular 45Ca2+ levels increase dramatically to 20×103cts min−1 on day 7 and continue to rise after this (data not shown). Thus, the precocious development of cartilage nodules in vitro correlates with a precise initial elevation of intracellular Ca2+. The progressive reduction of alcian blue staining is reflected by an extensive and continuous elevation of intracellular calcium levels.
The effect of continuous exposure to 0·1 μM-A23187 on intracellular 45Ca2+ levels. In the presence of A23187, intracellular [45Ca2+] is elevated to approximately 104cts min−110−6 cells on day 1. A decline in [45Ca2+] until day 3 is followed by a sharp, linear elevation to achieve 2×104cts min−110−6 cells on day 7. Error bars indicate +S.D.
The effect of continuous exposure to 0·1 μM-A23187 on intracellular 45Ca2+ levels. In the presence of A23187, intracellular [45Ca2+] is elevated to approximately 104cts min−110−6 cells on day 1. A decline in [45Ca2+] until day 3 is followed by a sharp, linear elevation to achieve 2×104cts min−110−6 cells on day 7. Error bars indicate +S.D.
The effect of retinoic acid on intracellular calcium levels
10 μM-retinoic acid completely inhibited limb bud chondrogenesis in micromass culture. During the first two days of micromass culture development, 10 μM-retinoic acid has little detectable effect on intracellular Ca2+ levels (Fig. 8). In contrast to control or A23187-treared cultures, 45Ca2+ levels increase gradually but sequentially. However, by days 3 and 4, intracellular 45Ca2+ has increased to approximately 20×103cts min−1 and 43×103cts min−1 per 106 cells, respectively, and increase still further to achieve 75×103ctsmin-1 on day 6. This value decreases by day 7 although intracellular 45Ca2+ levels remain high at between 50×103 and 80×103cts min−1 per 106 cells at later stages (data not shown).
The effect of retinoic acid on intracellular 45Ca2+ levels. 45Ca2+ levels remain low during the first two days of culture and then increase fourfold on day 3 to 2×104cts min−110−6 cells. This initial elevation of [45Ca2+] is followed by a dramatic increase to 8×104cts min−110−6 cells on day 6. Although levels decline to 5×104 cts min−110−6 cells on day 7, they remain abnormally high. Error bars indicate +S.D.
The effect of retinoic acid on intracellular 45Ca2+ levels. 45Ca2+ levels remain low during the first two days of culture and then increase fourfold on day 3 to 2×104cts min−110−6 cells. This initial elevation of [45Ca2+] is followed by a dramatic increase to 8×104cts min−110−6 cells on day 6. Although levels decline to 5×104 cts min−110−6 cells on day 7, they remain abnormally high. Error bars indicate +S.D.
Discussion
The differentiation of limb bud mesenchyme into cartilage, either in vivo or in micromass culture, is preceded by a spontaneous elevation of cell density. We have previously shown that the aggregation of limb bud cells in suspension prior to chondrogenesis is specifically governed by a Ca2+-dependent adhesion mechanism (Bee & von der Mark, 1987). The ability of limb bud mesenchyme in vitro to first undergo an elevation of intercellular association and then differentiate into cartilage is a direct reflection of the stage of embryonic development: until stage 21, although cells are capable of establishing prechondro-genic nodules they are only competent to differentiate into cartilage in the presence of dibutyryl cAMP and theophylline (Solursh et al. 1981). Between developmental stages 21 and 23, chondrogenesis is dependent upon high cell densities or culture in collagen gels (Ahrens et al. 1977; Solursh et al. 1982). At these later stages, chondrogenesis is also stimulated by dibutyryl cAMP and theophylline (Solursh et al. 1978). In the present paper, we have focused upon the role of Ca2+ in chondrogenic expression.
Compared with control cultures, elevation of extracellular Ca2+ to a final concentration of 5 HIM stimulates stage-21 embryonic chick limb bud cells to differentiate into cartilage. By day 2, discrete cartilage nodules are present in the cultures. Although they are not extensive, the number of nodules increases by day 3. Similar observations have recently been reported by San Antonio & Tuan (1986) who demonstrated an increase in both alcian blue and collagen type II staining during the first 60 h of culture in the presence of 5 mM-calcium. The data presented herein also show that elevated extracellular [Ca2+] stimulates chondrogenic potential. However, when cultures are maintained in high concentrations of exogenous Ca2+ cartilage development is ultimately reduced in a characteristic pattern. The addition of high [Ca2+] to cultures maintained for the first three days in control medium, during which nodule formation and prechondrocyte condensation are established, exerts an identical effect on subsequent culture development. Thus, extracellular Ca2+ appears to be required to initiate the process of cartilage development and this process is partially stimulated by elevating the levels of this cation. However, sustained high levels of extracellular Ca2+ are detrimental to the later stages of cartilage expression.
In contrast to extracellular Ca2+, manipulation of intracellular Ca2+ levels exerts profound effects on cartilage differentiation in vitro. Although 1 μM-A23187 almost completely inhibits chondrogenic expression, exposure to either 0·1 or 0·01 μM-A23187 during the initial 48 h of culture is not detrimental to cartilage differentiation or cartilage expression. The continual presence of 0·1 μM-A23187 not only stimulates initial cartilage development but ultimately leads to a progressive regression of cartilage nodules. Consequently, these data suggest that precocious elevation of prechondrocyte intracellular [Ca2+] stimulates differentiation but that continued elevation of intracellular [Ca2+] is detrimental to the chondrocyte. This effect is apparently delayed since the addition of 0·1 μM-A23187 from culture day 3 onwards does not effect loss of cartilage during the remaining culture period.
Analysis of intracellular 45Ca2+ in control cultures shows levels to be relatively low during the initial stages of culture development. However, commencing on day 3 intracellular [45Ca2+] rises to achieve a plateau level concomitant with extensive cartilage development. This change in intracellular Ca2+ level is most distinct by day 5 indicating a shift from relatively low prechondrogenic levels to a significantly higher chondrogenic value. Consistent with this, the mean values for group (i) (culture days 1–3) and group (ii) (culture days 5–70) are significantly different from one another. However, within each group there is no significant difference in the mean. These data suggest that intracellular Ca2+ levels are low in the prechondrocyte but that they increase concomitant with ultimate differentiation. A parallel study of limb bud cells cultured at low densities, which do not permit cartilage differentiation, reveals that cells fail to elevate intracellular 45Ca2+ levels. However, the possibility of selection against potentially chondrogenic cells cannot be ruled out.
In the presence of 0·1 μM-A23187 intracellular [45Ca2+] is spontaneously elevated concomitant with precocious cartilage development. Intracellular 45Ca2+ levels then decline as chondrogenesis continues and again begin to increase on day 3. With the progressive reduction of alcian blue-detectable cartilage is a simultaneous elevation of intracellular Ca2+ to high levels. Consequently, intracellular 45Ca2+ levels in the presence or absence of A23187 correlates directly with the development of cartilage. They demonstrate that a controlled elevation of intracellular [Ca2+] from the initial level parallels the expression of cartilage and that these levels normally remain constant in differentiated cartilage. Support for this comes from equivalent studies in which we have found that differentiated cartilage cells isolated from the chick sternum have a stable level of 45Ca2+ of approximately 8000 cts min−110−6 cells (Broadley & Bee, manuscript in preparation). Since continuous exposure to A23187 has a negative effect on cartilage expression and elevates intracellular [45Ca2+] to well above 10x103ctsmin-110−6 cells we suggest that elevation of intracellular [Ca2+] above a threshold level causes loss of chondrogenic expression.
Chick limb bud chondrogenesis in vitro is completely inhibited by retinoic acid (Lewis et al. 1978).
Since this is opposite to the effect of 0·1 μM-A23187 reported here, we assumed that retinoic acid would inhibit elevation of intracellular Ca2+ levels. Analysis of intracellular Ca2+ levels in the presence of retinoic acid demonstrates that initially low levels of 45Ca2+ rise dramatically to achieve and maintain an abnormally high value. Thus, in contrast to our expectation, retinoic acid stimulates Ca2+ uptake into the intracellular environment. Since the levels are well above an estimated threshold level, the inhibition of chondrogenesis by retinoic acid probably reflects not only its ability to elevate intracellular [Ca2+] but also the detrimental effect of high intracellular [Ca2+] on chondrogenic expression. Although retinoic acid effects loss of the differentiated cartilage phenotype in vitro (Fell & Dingle, 1963; Shapiro & Poon, 1976; Solursh & Meier, 1973; Vasan & Lash, 1975) it is not yet clear whether the inhibitory effect of retinoic acid is directed at cartilage differentiation per se or at the initiation of expression by committed chondrocytes. Nevertheless, it exerts dramatic effects on the pattern of limb development in vivo (Tickle, Alberts, Wol-pert & Lee, 1982; Summerbell, 1983).
The effects of A23187 and retinoic acid on cartilage differentiation and intracellular 45Ca2+ levels do not appear to be due to selective stimulation or inhibition of cell proliferation. In all experiments, the cell number per culture was remarkably similar to control. Although elevated extracellular [Ca2+] initially and transiently promotes cartilage differentiation, we conclude that ultimate chondrogenesis is directly related to changes in intracellular Ca2+ levels. Conditions that promote chondrogenesis, notably dense versus sparse culture, correlate directly with intracellular 45Ca2+ levels. Similarly, inhibition or stimulation of chondrogenesis also appears to be mediated through modification of the intracellular Ca2+ pool which in turn is influenced by cell density. Further analysis is required to determine whether the changes in intracellular [Ca2+] we demonstrate herein are regulated by cell–cell contact. Since intracellular Ca2+ and cAMP levels are intimately associated and elevated cAMP is capable of stimulating chondrogenesis, the mechanism by which Ca2+ enters the cell and influence developmental behaviour requires investigation.
ACKNOWLEDGEMENTS
This work was supported by grants from the Nuffield Foundation and the Medical Research Council to JAB.