DEVELOPMENTAL DYNAMICS 1 9 3 3 0 0 4 1 3 (1992)

Effect of Systemic Calcium Deficiency on the Expression of Transforming Growth Factor-p in Chick Embryonic Calvaria TOMOTARO SAT0 AND ROCKY S. TUAN Department of Orthopaedic Surgery, Thomas Jefferson University, Philadelphia, Pennsylvania 19107

ABSTRACT The developmental process of intramembranous ossification involves bone formation directly from mesenchymal differentiation without a cartilage intermediate. We have previously observed that systemic calcium deficiency in the developing chick embryo, produced by long-term shell-less culture, results in the appearance of chondrocyte-like cells in the calvarium, a parietal bone which normally develops via intramembranous ossification. This investigation aims to analyze the mechanism underlying this calcium deficiency-related, aberrant appearance of cartilage phenotype in the chick embryonic calvarium. In view of the reported involvement of transforming growth factor p (TGF-P)in osteogenesis and chondrogenesis, we have examined and compared here the expression of TGF-p in the chick embryonic calvaria of normal (in ovo development, NL), shell-less (SL), and calcium-supplemented SL (SL+ Ca) embryos. TGF-f3 expression was analyzed at the mRNA level by blot and in situ cDNA hybridization, and at the protein level by immunohistochemistry and immunoblotting. The results presented here indicate that: 1)TGF-p is expressed in the chick embryonic calvarium by both periosteal cells and osteocytes, as revealed by in situ hybridization and immunohistochemistry; 2) TGF-P expression is significantly increased in SL calvarium compared to NL calvarium, at both protein and mRNA levels; 3) the number of TGF-f3expressing cells increases in the SL calvarium, particularly along the central, subcambial core region of the bone; and 4) exogenous calcium repletion to the SL embryo affects the expression of TGF-P such that the pattern approaches that in the NL embryo. Taken together, these results indicate that altered TGF-f3expression accompanies the aberrant appearance of cartilage phenotype caused by systemic calcium deficiency. We postulate that normal cellular differentiation along the osteogenic pathway during embryonic intramembranous ossification is crucially dependent on regulated TGF-0 expression. o 1992 Wiley-Liss, Inc. Key words: Shell-less chick embryo culture, In situ hybridization, Immunohistochemistry, TGF-f3, Extracellular ma0

1992 WILEY-LISS, INC.

trix, Mineralization, Bone development, Intramembranous ossification, Chondrogenesis

INTRODUCTION Calvaria, flat bones of the skull, characteristically develop directly from cranial ectomesenchyme by a process known as intramembranous ossification, which does not involve a cartilage intermediate (Gilbert, 1988). During this process, mesenchymal cells undergo proliferation and condensation and differentiate into osteoblasts, followed by mineralization (Jackson and Randall, 1956; Bernard and Pease, 1969). The major extracellular matrix (ECM) component of the calvarial bone is type I collagen (Bornstein and Sage, 1980; Jacenko and Tuan, 1986). We have recently shown that severe systemic calcium deficiency produced by longterm shell-less (SL) culture of the chick embryo (Tuan, 1980;Ono and Tuan, 1986), where serum calcium level decreases by half, induces the appearance of a cartilage phenotype in the calvaria. This is indicated by tissue histology and by the expression of a cartilage-specific ECM component, collagen type I1 (Tuan and Lynch, 1983, Jacenko and Tuan, 1986). Immunohistochemistry revealed that collagen type I1 expression is localized to the central core of the SL calvaria (Jacenko and Tuan, 1986). Subsequently, collagen type I1 mRNA was also detected by in situ hybridization to be localized to similar, undermineralized area of the SL calvaria (McDonald and Tuan, 1989). Interestingly, cells positive for collagen type I1 mRNA could also be detected in the calvaria of normal chick embryo, whereas collagen type I1 protein was less detectable (McDonald and

Received September 4, 1991; accepted February 27, 1992. Address reprint requests to Rocky S. Tuan, Ph.D., Department of Orthopaedic Surgery, Thomas Jefferson University, 1015 Walnut Street, Philadelphia, PA 19107-5092. Abbreviations: TGF-p, transforming growth factor type p; ECM, extracellular matrix; NL, normal chick embryos developing in ova; SL, chick embryos developing in long-term shell-less culture; SL + Ca, SL embryos supplemented with exogenous calcium; SSC, 0.15 M NaCU0.015M sodium citrate, pH 7.0; SDS, sodium dodecyl sulfate; PBS, 10 mM phosphate-buffered saline, pH 7.2; TBS, Tris buffered saline, 10 mM Tris and 0.9% NaCl; BMP, bone morphogenetic protein; EDTA, ethylenediaminetetraaceticacid.

TGF-P IN CHICK EMBRYONIC CALVARIA

Tuan, 1989).This aberrant appearance of chondrocytelike cells in the SL calvaria suggests that perhaps a cartilage intermediate may indeed be involved, albeit transiently, in intramembranous ossification, and that perhaps systemic calcium deficiency has effectively arrested this process. Transforming growth factor p (TGF-P)and members of its superfamily are potent growth regulatory factors, which have been postulated to also regulate cell proliferation and differentiation during embryonic development (Massague, 1987; Sporn et al., 1987; Heine et al., 1987; Sporn and Roberts, 1990; Sandberg et al., 1988). In particular, the BMPs (bone morphogenetic proteins) derived from bone matrix, which are members of the TGF-p superfamily, have been shown to be chondro-and osteo-inductive factors in vivo and in vitro (Wang et al., 1990).TGF-P has also been shown to promote cartilage and bone growth when introduced subperiosteally (Noda and Camilliere, 1989; Joyce et al., 1990). The target of such TGF-p action may be mesenchymal cells, chondrocytes, preosteoblast, osteoblasts, and/or osteocytes (Centrella, 1986; Sporn and Roberts, 1990; Noda and Camilliere, 1989; Joyce et al., 1990; Massague, 1990). It has been shown that TGF-P acts through regulation of ECM synthesis (Rizzino, 1988; Massague, 1990; Sporn and Roberts, 1990), e.g., increasing the expression of fibronectin (Ignotz and Massague, 1986; Dean et al., 1988) and collagen (Seyedin et al., 1987; Ingotz and Massague, 1986; Roberts et al., 1986; Varga et al., 1987). In regard to embryonic skeletogenesis, there is increasing speculation that TGF-p or members of its superfamily are likely to be intimately involved in the expression of cartilage and bone phenotype, in particular during limb chondrogenesis (Heine et al., 1987; Pelton et al., 1989; Hayamizu et al., 1991; Carrington and Reddi, 1988; Carrington et al., 1991). In this study, we aimed to test the postulate that TGF-P is functionally involved in the aberrant chondrogenesis of the SL embryonic calvaria which accompanies systemic calcium deficiency. Specifically, we postulate that the effect could result from changes in the temporal, spatial, or quantitative aspects of TGF-p expression in the calvaria. Our postulate is partly built upon the following findings. First, TGF-P is produced by the calvaria in situ (Sandberg et al., 1988; Heine et al., 1987) and as cultured cells (Centrella and Canalis, 19851,and is found abundantly in the mineralized bone matrix (Seyedin et al., 1987). Second, TGF-p is active in calvarial bone growth, since direct injection of TGFp l and TGF-P2 into the subperiosteal space induced growth (Noda and Camilliere, 1989;Joyce et al., 1990). Third, calvarial cells are stimulated by TGF-P to increase synthesis of collagen (Centrella et al., 1987) and osteocalcin (Noda, 1989). Finally, chondrogenesis by limb mesenchyme is significantly affected by TGF-P in vivo (Hayamizu et al., 1991) and stimulated in vitro (Rosen et al., 1988, Seyedin et al., 1988; Kulyk et al., 1989; Carrington et al., 1990). The in vitro action of TGF-P has been shown to involve the enhancement of

301

cellular condensation and fibronectin expression (Leonard et al., 1991). Taken together, these observations suggest that ossification, both intramembranous and endochondral, is influenced and perhaps regulated by TGF-P. To test this postulate, the calvaria of three types of chick embryos are analyzed here with respect to TGF-p expression, including normal (NL), SL, and SL partially supplemented with exogenous calcium (SL+ Ca). TGF-p expression is analyzed a t the protein level immunochemically, and at the mRNA level based on cDNA hybridization. Both quantitative (immunoblotting, immunosorbent assay, and Northerdslot blots) and spatial (immunohistochemistry and in situ hybridization) analyses are carried out. The results reported here support the hypothesis that altered TGF-p expression accompanies and may be responsible for the aberrant chondrogenesis in the calvaria of the calcium deficient, SL chick embryo.

RESULTS Histology and Matrix Composition of Chick Embryonic Calvaria Systemic calcium deficiency of the chick embryo, produced by long-term SL culture, resulted in calvarial bones which displayed obvious signs of undermineralization. Comparison of the histology of NL and SL calvaria sectioned as described in Figure l and stained with hematoxylin-eosin (Fig. 2A,C) showed that the SL calvaria were generally stained deep pink-purple, consistent with a more acidic extracellular bone matrix (Kahn and Simmons, 1977). On the other hand, NL calvarial sections showed prominent, hematoxylinstained basophilic regions resembling mature bony trabecular structures. The undermineralization was clearly evident upon staining with alizarin red (Fig. 2B,D), whereupon SL sections displayed only patchy areas of calcification. These observations confirmed our previous findings (Jacenko and Tuan, 1986; McDonald and Tuan, 1989). With respect to the composition of the bone matrix, we (Jacenko and Tuan, 1987) have previously observed that collagen type I expression was not significantly affected by calcium deficiency and was detected in abundance in both NL and SL calvaria. Interestingly, we observed here that osteocalcin expression was significantly reduced in the SL calvarium (Fig. 3). Since osteocalcin is a specific component of the mature, calcified bone matrix (see reviews by Cole and Hanley, 1991; Stein et al., 19901, this finding is thus consistent with the undermineralized status of the SL calvaria. TGF-f3 Expression in Chick Embryonic Calvaria To assess the effect of systemic calcium deficiency on TGF-p expression in the chick embryonic calvaria, TGF-P was assayed at both the mRNA and protein levels. TGF-p mRNA. The expression of TGF-p mRNA in day-14 NL and SL embryonic calvaria was character-

302

SAT0 AND TUAN

A

B SUPERIOR

t

Fig. 1. Diagrammatic illustration of the orientation of chick embryonic calvarial sections. A: Lateral view of chicken head indicating the calvarium (darkened area). Dotted line with arrowhead represents the plane along which the calvarium was sectioned serially. B: Cross-sectional view of the calvarium after sectioning along the plane indicated in A. The calvarium is arbitrarily divided into the superior, orbital, and temporal zones. (Figure adapted from Jacenko and Tuan, 1986; McDonald and Tuan, 1989.)

ized by means of both blot hybridization and in situ hybridization. Blot hybridization. Polyadenylated RNAs isolated from NL and SL embryonic calvaria were analyzed by Northern blotting and probed using radiolabelled human TGF-Pl cDNA. As shown in Figure 4A, a 3.0 kb hybridization band was seen in both NL and SL calvaria. The size of the band corresponded to that of chick TGF-P3 reported by Jakowlew et al. (1988b).Even prolonged exposure (14 days, data not shown) failed to reveal any additional bands with sizes corresponding to those reported for other chick TGF-P subtypes (for example, TGF-P1, 2.5 kb; TGF-P2, 8, 4.3, and 3,9 kb; TGF-P4, 1.7 kb; Jakowlew et al, 1991). As shown in Figure 4B, the identity of the TGF-P3 band was further confirmed by probing with chick TGF-P3 cDNA (Jakowlew et al., 1991), which also hybridized to the same 3.0 kb band. It should be noted that the 3.0 kb band also cross-hybridized to chick TGF-P2 cDNA (not shown). These results taken together indicate that, at the mRNA level, the calvaria of both NL and SL embryos express TGF-P, primarily in the form of TGF-P3. To quantify the relative level of TGF-P expression, RNAs isolated from NL and SL calvaria were slotblotted, and hybridized separately with TGF-P cDNA and p-actin cDNA probes. The apparent higher level of TGF-P expression in the SL calvaria shown in Figure 4C was confirmed by quantitative comparisons of the densitometric signals, with endogenous p-actin signal as an internal standard. As shown in Figure 5, this analysis showed that the level of TGF-f3 mRNA in SL calvaria was a t least 30% higher than that in NL calvaria.

I n situ hybridization. Paraffin sections of NL and SL calvaria were processed for in situ hybridization using biotin-labelled TGF-f3cDNA and alkaline phosphatase conjugated streptavidin as a secondary reagent as described in Experimental Procedures. As shown in Figure 6, positive hybridization signals were seen mainly in the centrally located osteocytes and the periosteum of both inner and outer surfaces of the bone. In general, the more calcified matrix of the NL calvaria tended to have higher background staining of the more calcified matrix (Fig. 6A). [Note: As reported previously (McDonald and Tuan, 19891, longer decalcification time in EDTA during pre-hybridization, although useful in reducing the background, frequently resulted in premature loosening of the sections from the slide during the procedure.] A more prominent periosteal staining was also seen in the NL calvaria. These staining patterns clearly demonstrated that the chick embryonic calvaria contain TGF-f3 expressing cells. To quantify the abundance of TGF-P expressing cells in the NL and SL calvaria, we examined a large number of sections and undertook a systematic counting of positive cells in random 40x microscopic fields from different parts of the calvaria, including the superior, orbital, and temporal regions (see Fig. 1). The data are presented in Table 1and are also expressed as percentages of the total number of cells found in the microscopic field (Fig. 7). These results revealed several interesting features: 1)in both NL and SL calvaria, the largest number of positive cells was found in the orbital region, particularly in the less calcified bone surface; 2) however, the percentage of positive cells in each type of calvaria remained rather constant in the three zones,

TGF-p IN CHICK EMBRYONIC CALVARIA

303

Fig. 2. Histology of NL and SL calvaria. Paraffin sections of the orbital region of NL (A,B) and SL (C,D) calvaria were stained with hematoxylin-eosin (A, C) or alizarin red (B, D) and viewed by bright-field photography. Bone matrix was stained pinkish, and osteocytes embedded in lacunae were prominently seen on the superficial and internal areas (A,B). Note distinct undermineralization in SL calvaria compared to NL calvaria as revealed by poor alizarin red staining (B vs. D). Bar = 44 pm.

i.e., the total number of cells and the number of TGF-f3 positive cells changed correspondingly; and 3) most importantly, in all regions of the SL calvaria, the percentage of TGF-P positive cells was considerably higher than the corresponding value in NL calvaria. Taken together with the blot hybridization data, these observations strongly suggest that systemic calcium deficiency, as a result of SL culture, significantly enhances TGF-P expression in the calvarium, primarily by increasing the number of TGF-P expressing cells. TGF-p protein. The expression of TGF-P in day-14 NL and SL embryonic calvaria was next analyzed at

the protein level by means of immunohistochemistry and immunochemistry [blotting and enzyme-linked immunosorbent assay (ELISA)]. Immunohistochemistry. The calvarial sections were immunostained using antibodies prepared against porcine TGF-P1 and -P2 (Fig. 8). Similar to in situ hybridization, immunohistochemistry revealed TGF-P positive cells localized in the central core (osteocytes) and in the periosteum of both NL (Fig. 8A) and SL (Fig. 8B) calvaria. In the periosteum, TGF-P staining often appeared to be associated with the extracellular matrix (Fig. 8B). Little background staining was seen (Fig.

304

SAT0 AND TUAN

Fig. 3. lmrnunohistochemical localization of osteocalcin in NL and SL calvaria. The procedure was as described in Experimental Procedures. A,B,C: NL; D,E,F: SL calvaria. The sections were viewed by phase-contrast optics (A,D), Nornarski differential interference contrast optics (B,E),

and epifluoresence (C,F). Note the generally more calcified matrix (arrows) in the NL section, which corresponded to an overall higher level of osteocalcin staining. Osteocalcin staining was also clearly detected in cells of both NL and SL calvaria (C,F). Bar = 44 p,m,

8D), and no extensive decalcification of the sections was required. Quantitative aspects concerning the distribution of TGF-P immunopositive cells were analyzed as described above for in situ hybridization and the results are presented in Table 1 and Figure 9. Again, compared to NL calvaria, SL calvaria appeared to house a higher percentage of TGF-P positive cells, particularly in the temporal and orbital regions (Fig. 9). There was

no apparent difference between the data obtained by immunohistochemistry and those from in situ hybridization (see Table 1).The only exception was that the superior region did not display a statistically significant increase in TGF-P immunopositive cells in the SL calvaria (Fig. 9). Zmmunochemistry. Protein extracts from NL and SL calvaria were blotted onto nitrocellulose and immunoreacted with TGF-P antibodies. As shown in Figure

TGF-P IN CHICK EMBRYONIC CALVARIA

A size

NL

kb

5.0

3.0

1.7

305

B SL

kb

-

5.0

-

3.0

-

1.7

NL

SL

-

-

C NL SL 2.5

5

1

IClSl Fig. 4. Analysis of TGF-p mRNA by Northern and slot blot hybridization. A,B: Northern blots. Poly-A+ RNAs (5 pg) isolated from NL and SL calvaria were fractionated on a formaldehyde gel, blotted onto Gene Screen, and probed with 32P-labelledTGF-p cDNA probes (see Experimental Procedures). A: Human TGF-p1 probe; B: chicken TGF-p3 probe.

Note the presence of a prominent band of 3 kb size in both NL and SL RNA samples, with the latter being more intense. C: Slot blot. Total RNAs (5, 2.5, and 1 pg) of NL and SL calvaria were slot-blotted onto Gene Screen and probed with human TGF-p1 cDNA. The results showed higher amount of signal in the SL sample.

10A, on a protein basis, SL calvaria contained a significantly higher amount of TGF-P (ranging from 40 to 130% increase). This was also confirmed by means of ELISA, which showed a twofold to threefold increase in TGF-P level in the SL calvarial extract (Fig. 10B). These results on TGF-P protein level are therefore in general agreement with those on TGF-P mRNA, confirming that TGF-P expression was indeed elevated in the SL calvaria. Effect of Calcium Supplementation to SL Embryos on TGF-f3Expression in Calvaria The data presented above clearly demonstrated that systemic calcium deficiency was accompanied by an apparent increase in TGF-(3 expression in the calvaria of the developing chick embryo, suggesting a likely causal relationship. To investigate whether calcium deficiency was principally responsible for the enhanced TGF-P expression, SL embryos were supplemented

I-

5

2.5

1

Total RNA load (pg) Fig. 5. Quantitative comparison of TGF-p mRNA levels in NL and SL chick embryonic calvaria by slot blot hybridization. The slot blot shown in Figure 3 was scanned by transmittance densitometry. The data showed that the SL calvaria contained at least 30% more TGF-p mRNA than the NL calvaria.

306

SAT0 AND TUAN

with calcium as described in Experimental Procedures. Such supplementation resulted in serum calcium levels intermediate between those of the NL and SL embryos (Tuan and Nguyen, 1987; Gawande and Tuan, 19901, and restored calvarial collagen phenotype of the SL embryo to that of the NL embryo (Jacenko and Tuan, 1987). In this study, we examined the effect of calcium supplementation on TGF-P expression by means of both in situ hybridization and immunohistochemistry, and the results are presented in Figures 6-9 and in Table 1. In general, no gross histological differences could be detected with respect to the distribution of the TGF-P positive cells as detected by in situ hybridization or immunohistochemistry, i.e., positive cells were found in the periosteum and the central core of the bone (Figs. 6C,8C). Interestingly, quantitation of positive cells clearly revealed that the percentage of TGF-P positive cells in all regions of the SL + Ca calvaria was intermediate between the values found in NL and SL calvaria (Figs. 7,9, Table 1). This was not the result of major redistribution of the cells within the calvaria, since in terms of total cell number, the relationship of orbital > (temporal or superior) still held true for all specimens. The finding that TGF-P expressing cells were specifically affected is therefore consistent with the notion that systemic calcium deficiency was a principal cause of enhanced TGF-P expression in the SL calvaria.

Fig. 6. Localization of TGF-P mRNA in chick embryonic calvaria by in situ hybridization. Calvarial sections from the orbital region were probed with biotin-labelled human TGF-Pl cDNA (A-C) as described in Experimental Procedures. A,D: NL; B: SL; C: SL+ Ca calvaria. The sections were examined with Nomarski differential interference optics. Note positive hybridization signals localized to osteocytes (arrows) in the central core of the bone, and to the periosteal area. Control (D) was done without labelled probe. Bar = 44 pm.

DISCUSSION Our previous observations (Jacenko and Tuan, 1986; McDonald and Tuan, 1989) have shown that experimentally induced systemic calcium deficiency in the developing chick embryo results in the aberrant appearance of cartilage-like histology and chondrocytelike cells in the calvarium, a parietal head bone which normally develops via the intramembranous ossification pathway, without a cartilage intermediate (Jackson and Randall, 1956; Bernard and Pease, 1969). Results from the present study clearly demonstrate that a higher level of TGF-P expression, most likely a result of an increase in the number of TGF-P producing cells, also takes place in the calcium-deficient calvarium of the SL embryo. Within the day-14 calvarium, the orbital and temporal regions show the largest increase in the number of cells positive for TGF-P mRNA and protein during calcium deficiency. That the systemic calcium deficiency is causally responsible for the enhanced TGF-P expression in the calvaria of SL embryos is indicated by the partial restoration towards the normal pattern of TGF-P expression when the SL chick embryos are supplemented with exogenous calcium. The TGF-Ps are a class of growth factors with wideranging functions and biological activities and almost ubiquitous tissue and cellular distribution (Massague, 1987,1990; Sporn and Roberts, 1990). The involvement of TGF-P in the process of chondrogenesis and osteogenesis, and in the regulation of cellular functions in

307

TGF-P IN CHICK EMBRYONIC CALVARIA

TABLE 1. Localization and Quantitation of TGF-S Expressing Cells in Chick Embryonic Calvaria ~~~~

C e l l num ber in tissue zonesa

TGF-P detection

Embryo

In s i t u h yb ri d i za ti o n

NL

SL SL+Ca

NL

Immunohistochemistry

SL SL+Ca

Superior Stained Total

7?1 622 7? 1 8+2 8 k 1 8+1

11+2 8*2 1023 13'3 1222 11'2

Orbital Stained Tot al

30'5 27'4 25+3 36+3 26c4 2422

47+6 3255 32t4 52217 32+5 3144

Temporal Stained Total

1Ok2 6+2 722 5+1 6 2 2 7 k 1

1523 7+3 1023 8+3 8?2 1021

Sum nb

Stained

To t a l

34 33 27 13 12 8

47+6 39t5 39+4 4925 40+5 3952

73k7 4727 5226 7326 5228 52+4

"The cells were counted as cells present in a giv en 40 X objective field. Values represent m ean t SD. bn = N u m b e r of sections examined.

-

3

Q)

.-.-5

80

ul 0

n

a z a

60

E

-?

LL

40

superior

orbital

temporal

Location in Calvarial Section Fig. 7. Quantitation of the abundance of TGF-P expressing cells in chick embryonic calvaria based on in situ hybridization. In random microscopic (40 x ) fields, calvarial cells showing positive hybridization were counted, and calculated as a percentage of the total number of cells. Three arbitrary zones, superior, orbital, and temporal (see Fig. I ) , were examined (see Table 1 for details of data). The data showed that SL calvaria had a significantly higher percentage of stained cells than the NL calvaria, whereas SL Ca calvaria were intermediate between NL and SL calvaria. All data were analyzed by Student's t-test and P values are indicated.

+

these processes has been strongly implicated in a large number of in vitro and in vivo studies (see Introduction). In particular, it has been suggested that in the embryonic limb TGF-P, or a member of its family, may be the natural promoter of mesenchymal differentiation along the chondrogenic pathway (Kulyk et al., 1989; Leonard et al., 1991). Our present finding that aberrant chondrogenesis in the calvarium, which results from systemic calcium deficiency of the developing chick embryo, is accompanied by elevated TGF-P expression is thus consistent with this postulate. On the other hand, it should be noted that the effect of TGF-P on chondrogenesis is likely to be a function of the developmental stage and commitment state of the precartilage cells, as shown by TGF-P implantation in the chick embryonic limb in vivo (Hayamizu et al., 1991). In this study, we have observed that elevated TGF-(3

expression in the SL calvaria most likely results from an increase in the number of expressing cells, although the in situ hybridization and immunohistochemical techniques used here do not conclusively rule out a concomitant higher level of gene expression in individual cells. Similarly, we (Jacenko and Tuan, 1986; McDonald and Tuan, 1989) have previously observed that the number of cells expressing collagen type I1 is also increased in the SL calvarium, particularly in the central, subcambial core of the orbital and temporal regions, which is less calcified than the superior region of the bone at day 14 of development (McDonald and Tuan, 1989). It is thus tempting to draw a correlation between these findings, viz. undermineralization of the bone matrix leads to increased TGF-P expression, which in turn is responsible for enhanced chondrogenic differentiation. Interestingly, the undermineralization of the matrix of the SL calvaria is also accompanied by a substantial decrease in matrix osteocalcin (Fig. 3). Since osteocalcin is a bone-specific protein associated with osteoblast maturation and calcification (Stein et al., 19901, the SL calvarium thus represents a less mature bone than the NL counterpart, consistent with its potential to express the chondrogenic phenotype. That the embryonic calvarium expresses TGF-P has been shown by a number of investigators, e.g., TGF-P1 in human (Sandberg et al., 1988) and rodents (Heine et al., 1987). In the chick embryo, Jakowlew et al. (1991) have also shown a wide distribution of TGF-p isotype expression in various tissues, including bone and cartilage. The interspecies protein sequence homology between the TGF-P isotypes is substantial, as high as 64-100% (Derynck et al., 1985; Sporn et al., 1987; Jakowlew et al., 1988a,b,c, 1990; Kondaiah et al., 1990). Nevertheless, the spatial and temporal profiles of expression of each TGF-P isotype appear to be highly specific during embryonic development (Heine et al., 1987; Lehnert and Akhurst, 1988; Miller et al., 1989). Thus, although the administration or treatment with one TGF-P type, most often TGF-P1 or TGF-P2 because of their availability, is found to elicit a biological response in a given tissue in situ or in vitro, the natural effector may very well be another isotype. From our findings reported here and those of Jakowlew et al.

308

SAT0 AND TUAN

* * -#-.

superior

orbital

temporal

Location in Calvarial Section Fig. 9. Quantitation of the abundance of TGF-P expressing cells in chick embryonic calvaria based on immunohistochemical staining. In random microscopic (40 x ) fields, calvarial cells showing positive immunostaining were counted and calculated as a percentage of total cells. Three arbitrary zones, superior, orbital, and temporal (see Fig. l ) , were examined (see Table 1 for details). The data showed that SL calvaria had a significantly higher percentage of stained cells than NL calvaria, whereas SL+ Ca calvaria were intermediate between NL and SL calvaria. All data were analyzed by Student's t-test and P values are indicated.

(19911,TGF-p3 may be the biologically active molecule in the developing chick embryonic calvarial bone, although specific immunoreactivity against TGF-P3 antibodies is needed for confirmation. Interestingly, it has been recently reported that the relative level of TGF-p type expression in the chick embryonic long bone is p3 > p4 > p2 > p l (O'Donnell et al., 1991). The fact that TGF-p is normally expressed in the chick embryonic calvarium suggests that it is important for formation of the tissue during development. Osteoblasts are stimulated to proliferate in vitro when treated with TGF-p (Robey et al., 1987; Centrella et al., 1986, 19871, whereas bone resorption is suppressed (Pfeilschifter et al., 1988). It may be speculated that during normal development, the growth of the calvarium is indeed regulated by controlled expression of TGF-p. Interestingly, in vivo administration of TGF-P accelerates wound healing (Mustoe et al., 19871, suggesting that TGF-P may be involved in the wound healing and repair response in general (Massague, 1990).It is thus possible that long-term systemic calcium deficiency, resulting in a substantially undermineralized

Fig. 8 . lmmunohistochemical localization of TGF-p protein in chick embryonic calvaria. The procedure was as described in Experimental Procedures, using antibodies directed against human TGF-p1,2and the tissue sections were made from the same calvaria as in the in situ hybridization study (see Fig. 6). The sections were viewed by Nomarski differential interference contrast optics. A,D: NL; B: SL; C : SL+ Ca calvaria. Note positive staining (arrows) in osteocytes as well as the periosteum, the latter showing significant matrix staining. D: Control using rabbit serum instead of primary antibodies. Bar = 44 km.

309

TGF-p IN CHICK EMBRYONIC CALVARIA 0.6 r

A

B

E C

Q)

0

C

NL

a L

0

rA

2 c Q)

z

SL 20

10

5

clg

Sample dilution (log 2) Fig. 10. lmmunoquantitation of TGF-P in chick embryonic calvaria. A: Immunoblotting. Extracts of NL and SL calvaria were serially diluted (20, 10, and 5 pg) and blotted onto nitrocellulose and TGF-p detected immunohistochemically as described in Experimental Procedures. The apparent, higher signal in the SL calvarial extract was confirmed by reflectance

densitometry (40-130% increase). B: ELISA. Calvarial extracts were serially diluted and processed by ELISA as described in Experimental Procedures. The ,,A, values f SD from four separate assays were net values over background (i.e., no extract).

matrix, is perceived as an adverse condition by the calvarium, and the tissue responds in a “damage-repair” mode. It is also possible that the undermineralized matrix is less able to sequester the extracellular TGF-p, which is then permitted to act upon the bone cells to stimulate further TGF-P expression (Van ObberghenSchilling et al., 1988). In addition, in view of the interactive relationship between TGF-p and the extracellular matrix (Massague, 1990; Sporn and Roberts, 19901, it is also possible that the calvarial cells of the SL embryo respond t o the altered matrix of the calvaria, mediated via cell surface moieties or receptors (Massague, 1990), by expressing TGF-p. The central issue in our investigation is the observation that long-term calcium deficiency results in the emergence of “chondrocyte-like”cells in the otherwise totally osteogenic calvarium (Jacenko and Tuan, 1986; McDonald and Tuan, 1989; Tuan, 1991). Since seemingly parallel increases in the number of cells expressing TGF-p and collagen type I1 are observed, it may be speculated that the additional cells expressing TGF-P in fact correspond to those eventually expressing type I1 collagen, i.e., chondroprogenitor cells. A similar observation has been made in the embryonic limb bud, where the central condensing core of mesenchymal cells indeed express higher levels of TGF-(3 and later form the cartilage anlage (Kulyk et al., 1989). The mechanisms by which TGF-p influences cellular proliferation and differentiation are currently not completely understood, and are being actively studied by a large number of investigators (see review by Mas-

sague, 1990; Sporn and Roberts, 1990). For example, the exact identity of TGF-P receptors and their mechanisms of action are still open issues. Finally, the results obtained here point to another issue concerning the developmental program of intramembranous ossification, which normally does not involve a cartilage intermediate. However, we (Jacenko and Tuan, 1986; McDonald and Tuan, 1989; Wong and %an, 1990) and others (Aubin et al., 1982; Villanueva et al., 1989) have clearly observed the presence of cells with multiple differentiation potentials in the calvarium, suggesting that intramembranous ossification may in fact involve a chondrogenic phase, albeit highly abbreviated. A recent report by Lengele et al. (1990) presented radiographic and histological evidence for the presence of chondroid tissue in the chick embryonic calvaria during early development, up to incubation day 14. These authors concluded that chondroid tissue should be considered a transient autonomous phase during development of the cranial bones. For these reasons, the mesenchymal cells of the intramembranous bone are likely to resemble their counterparts in endochondral bones, at least with respect to their chondrogenic potential. Interestingly, our in situ localization studies reveal that the additional TGF-P positive cells, which arise as a result of calcium deficiency in the SL calvaria, are mostly localized to the central and subperiosteal/subcambial areas of the temporal and orbital regions (Figs. 6 and 7). These areas correspond to the same location where we (Jacenko and Tuan, 1986; McDonald and Tuan, 1989) have previ-

310

SAT0 AND TUAN

ously observed the presence of type I1 collagen producing, "chondrocyte-like" cells, and where Lengele e t al. (1990) recently detected chondroid tissue. It should be noted that these areas normally express less TGF-p than the periosteum (Figs. 6,8), thus suggesting that controlled TGF-P expression may be imporant in regulating the differentiative fate of these cells along either the chondrogenesis or osteogenesis pathway. Overexpression of TGF-p in situ, caused by long-term systemic calcium deficiency as a result of SL culture of the chick embryo, thus permits or influences these pluripotent cells to undergo chondrogenesis.

EXPERIMENTAL PROCEDURES Chick Embryos and Calvarial Tissues Fertilized White Leghorn chicken eggs were purchased from Truslow Farms (Chestertown, MD) and incubated in a humidified egg incubator a t 37.5"C with regular rotation. Some embryos were incubated in ovo for 3 days and transferred to ex ovo (shell-less) culture for additional 11 days (total 14 days). The protocol of shell-less culture has been described previously (Dunn and Boone, 1976; Tuan, 1980). Calcium was supplemented to cultured SL embryos as described previously (Jacenko and Tuan, 1987) by daily addition of a CaCO, slurry (10% w/v in water), applied as several spots, onto the chorioallantoic membrane using the following dosage: 10 mg (day 111, 20 mg (day 12), 30 mg (day 13), and 40 mg (day 14). In some instances, calcium was supplemented in the form of sterilized eggshell pieces as described earlier (Tuan, 1983; Tuan and Nguyen, 1987). After incubation day 14, calvaria were removed from NL, SL, and SL + Ca embryos, washed immediately i n ice-cold PBS (0.01 M phosphate-buffered saline, pH 7.2), blotted with filter paper to remove associated connective tissues, and used immediately for the following experiments. Blot Analysis of TGF-P mRNA RNA isolation. RNA was isolated from tissues by the single-step guanidinium thiocyanate method (Chomczynski and Sacchi, 1987; Tuan et al., 1991). Pieces of calvaria were frozen in liquid N2 and homogenized in a 4 M guanidinium thiocyanate buffer, followed by extraction with phenol/chloroform/isoamyl alcohol. RNA was isolated by repeated precipitation with isopropanol, washed in 70% ethanol, and its concentration estimated based on Azso. In some experiments, mRNA was further isolated by oligo-dT affinity chromatography (Aviv and Leder, 1977). TGF-P cDNA probes. The cDNA probes used were: 1) human TGF-P1 cDNA, 2 kb EcoRI fragment excised from the plasmid phTGF-p2 (Kasid et al., 1988), kindly provided by Dr. G.I. Bell (NIH);2) chicken TGFP3-specific cDNA, EcoRI fragment of plasmid pTGFP-ChX17 (Jakowlew et al., 1988b), kindly provided by Dr. Sonia B. Jakowlew (NIH). Northern blot analysis. Northern blotting was per-

formed with 32P-labeled cDNA probe prepared using a random priming procedure (Feinberg and Vogelstein, 1983) (1601Y kit, Amersham Corp., Chicago, IL). Polyadenylated RNA (5 pg) was denatured in formamide and formaldehyde, separated by electrophoresis on 1.0% agarose gel calibrated with size marker (0.24-9.5 kb RNA Ladder., BRL, Gaithersburg, MD), transferred to nylon membrane (Gene Screen Plus, New England Nuclear, Boston, MA) in 10 x SSC (1 x SSC = 0.15 M NaCU0.015 M sodium citrate, pH 7.0). The blots were denatured and then neutralized. Prehybridization of the blot was carried out for 1h r at 42°C in 50% formamide buffer containing l % sodium dodecyl sulfate (SDS). Hybridization was carried out for 16 h r at 42°C in the same buffer used in prehybridization, to which ,'P-labeled cDNA insert (lo6 c p d m l ) was added with denatured salmon sperm DNA. The blots were washed twice in 2 x SSC a t room temperature for 5 min, twice in 2 x SSC-1% SDS at 65°C for 30 min, and then twice in 0.1 x SSC for 30 min at room temperature. After washing, the blots were exposed to X-ray film (XOMAT AR, Eastman Kodak, Rochester, NY) for 12 hr to 14 days at -80°C. In general, for confirmation, Northern blotting was performed using both human and chicken TGF-p cDNA probes. Replicate hybridization was also carried out using a p-actin cDNA probe (kindly provided by Dr. D. Cleveland, Johns Hopkins University) to test for mRNA integrity. Slot blot hybridization analysis. Slot blot analysis was performed to quantify TGF-@mRNA. Samples of total RNA (5, 2.5, and 1 pg) were slot-blotted (VacuSlot-VS, VacuSystems, American Bionetics, Inc., Hayward, CA) onto nylon membrane (Gene Screen Plus). The blot was processed for hybridization as described above, and autoradiographic signals were quantified by scanning densitometry (GS 300 Transmittance/Reflectance Scanning Densitometer, Hoefer Scientific Instruments, San Francisco, CA) and analyzed with the Hoefer software. Both human and chick TGF-p probes were used for confirmation of results. Normalization of the hybridization signals with respect to total mRNA content was carried out by replicate hybridization of the blot with p-actin cDNA probe.

Localization of TGF-P mRNA by In Situ Hybridization The method used here was identical to that described previously (McDonald and Tuan, 1989; Oshima et al., 1989). Briefly, TGF-p cDNA probe (total plasmid) was biotin-labelled by nick-translation in the presence of biotin-dUTP (16-mer, Enzo Biochem., New York, NY). Calvarial tissues were fixed in modified Carnoy's solution at -2o"C, dehydrated, embedded in paraplast, and sectioned at 8pm thickness. The sections were digested with hyaluronidase and Proteinase K, denatured with formamide, and hybridized for 48 h r a t 37°C with the biotinylated TGF-p probe. After washing to remove ex-

TGF-p IN CHICK EMBRYONIC CALVARIA

311

cess unhybridized probe, the sections were incubated Immunochemical Analysis of TGF-9 in with streptavidin conjugated with alkaline phos- Calvarial Extract phatase (Detek I-alk, Enzo), followed by chromogenic Chicken calvaria were pulverized in liquid Nz, and histochemistry using bromochloroindoyl phosphate extracted with 0.5 M EDTA; the extracts were dialyzed and nitroblue tetrazolium. Controls included omission against 10 mM NH4HC03 and lyophilized. [Note: It of probe or the use of irrelevant DNA as probe. was found that 0.5 M EDTA efficiently extracted TGFp from day-14 embryonic calvaria, in comparison to other bone extraction protocols which contained addiImmunohistochemistry tional guanidine (e.g., Carrington et al., 1988),presumTGF-9. Immunohistochemistry was performed using ably because these bones were in a relatively early rabbit polyclonal antibodies (IgG) for TGF-p1, p2, and state of mineralization.] Protein concentration in the pl,2 from porcine platelet (R&D Systems Inc., Minne- extracts was determined using the BCA assay apolis, MN). For control, the TGF-P antibodies were (PIERCE, Rockford, IL). For immunoblotting, extracts substituted with normal rabbit serum. Tissue sections with known amounts of protein were dissolved in TBS used for immunohistochemistry were the same as those (Tris buffered saline; 10 mM Tris, pH 7.4, 0.9% NaCU used for in situ hybridization. The immunohistochem- and blotted in serial dilution (20, 10, and 5 pg) onto ical staining procedure was essentially as described re- nitrocellulose membrane. Immunodetection of TGF-P cently (Ono and Tuan, 1991), processed by the Strepta- was carried out using anti-TGF-p antibodies (R&D vidin-biotin system of ZYMED Laboratories Inc. Systems) at 1:1,000 dilution, followed by the peroxi(South San Francisco, CA). Briefly, after deparaffiniza- dase immunoblot system kit (Immunoblot SP kit for tion and hydration, calvarial sections were washed Rabbit Antibody, ZYMED Laboratories) using the with PBS for 10 min, and demineralized overnight by manufacturer’s protocols. The immunohistochemical incubation in 0.1 N acetic acid at 4°C. After peroxidase signals were quantified by means of reflectance densiblock with 3% hydrogen peroxidase in methanol for 10 tometry using the Hoefer Scanning Densitometer as min, serum blocking with nonimmune rabbit serum for described above. For ELISA, extracts of NL and SL 10 min, and washing with PBS, the sections were in- calvaria were serially diluted in Na,CO,, pH 9.6, from cubated with anti-TGF-p antibody (1:lOO dilution) in a an initial concentration of 1 mg/ml, coated onto mimoist chamber for 30 min at 37°C. After rinsing, the crotitre wells, washed with PBS, and incubated with sections were incubated with biotinylated anti-rabbit anti-TGF-p antibodies (1:1,000 dilution). After further antibodies (1:40 dilution) for 10 min at room tempera- washing with PBS, incubation was continued with alture, rinsed, and incubated with streptavidin-peroxi- kaline phosphatase conjugated anti-rabbit IgG antidase conjugate. Finally, substrate, chromogen solution, bodies (Sigma). Enzyme reaction was carried out using and hydrogen peroxide were added for 5 min a t 37°C. the ALP-10 kit (Sigma) according to the manufacturOsteocalcin. Rabbit-derived antibodies to osteocal- er’s protocol, and A,,, values were determined. cin were kindly provided by Dr. Jane Lian (University of Massachusetts). Calvarial sections were prepared for ACKNOWLEDGMENTS immunohistochemistry as described above for TGF-P The authors thank Kenneth J. Shepley for his expert (150 dilution for primary antibodies), except that imassistance in in situ hybridization and immunohistomunostaining was visualized using FITC-conjugated chemistry, Dr. Sonia B. Jakowlew for providing anti-rabbit IgG antibodies (Sigma Chemical Co., St. chicken TGF-p cDNA, and Dr. G.I. Bell for human Louis, MO) at 1:lOO dilution. TGF-P1 cDNA. This work was supported in part by research grants from NIH (HD158221,March of Dimes Birth Defects Foundation (1-11461, and U S . DepartMicrography and Cell Counting ment of Agriculture (90-37200-52651,and the OrthoTissue sections hybridized with TGF-p probes or paedic Research and Education Foundation. stained with TGF-p antibody were observed and photographed on Kodak Ektar 125 film using an Olympus REFERENCES BH2 microscope equipped with Nomarski differential Aubin, J., Heersche, N., Merrilees, M., and Sodek, J . (1982) Isolation interference contrast optics. In addition, the number of of bone cell clones with differences in growth, hormone responses, and extracellular matrix production. J. Cell Biol., 92:452-461. stained cells and total number of cells in various regions of the calvaria, i.e., superior, orbital, and tempo- Aviv, H., and Leder, P. (1972) Purification of biologically active globin messenger RNA by chromatography on oligothymidylic acid cellural (see Fig. l),as examined in random 40 x magnifilose. Roc. Natl. Acad. Sci. U.S.A., 69:1408-1412. cation microscopic fields, were counted. The ratios of Bernard, G., and Pease, D. (1969) An electron microscopic study of the number of stained cells to the total number of cells initial intramembranous osteogenesis. Am. J. Anat. 125:271-290. in the calvaria of different embryos were compared sta- Bornstein, P., and Sage, H. (1980) Structurally distinct collagen types. Annu. Rev. Biochem. 49:957-1003. tistically by Student’s t-test. J.L., and Reddi, A.H. (1988) Temporal changes in the For osteocalcin immunof luorescence, the sections Carrington, response of chick limb bud mesodermal cells to transforming were viewed by epifluorescence illumination and phogrowth factor p-type 1. Exp. Cell Res. 186:368-373. tographed using Kodak T-Max film. Carrington, J.L., Roberts, A.B., Flanders, K.C., Roche, N.S., and

312

SAT0 AND TUAN

Reddi, A.H. (1988) Accumulation, localization, and compartmentation of transforming growth factor p during endochondral bone development. J . Cell Biol. 107:1969-1975. Carrington, J.L., Chen, P., Yanagishita, M., and Reddi, A.H. (1991) Osteogenin (bone morphogenetic protein-3) stimulates cartilage formation by chick limb bud cells in vitro. Dev. Biol. 146:406415. Centrella, M., and Canalis, E. (1985) Transforming and nontransforming growth factors are present in medium conditioned by fetal rat calvaria. Proc. Natl. Acad. Sci. U.S.A. 82:7335-7339. Centrella, M., Massague, J., and Canalis, E. (1986) Human platelet derived transforming growth factor-p stimulates parameters of bone growth in fetal rat calvariae. Endocrinology 119:2306-2312. Centrella, M., McCarthy, T.L., and Canalis, E. (1987) Transforming growth factor f3 is a bifunctional regulator of replication and collagen synthesis in osteoblast-enriched cell cultures from fetal rat bone. J . Biol. 262:2869-2874. Chomczynski, P., and Sacchi, N. (1987) Single step method of RNA isolation by acid guanidium thiocyanate-phenol-chroloformextraction. Anal. Biochem. 162:156-159. Cole, D.E., and Hanley, D.A. (1991) Osteocalcin. In “Bone. Vol. 11. Bone Matrix and Bone Specific Products,” Hall, B.K. (ed.) Boca Raton: CRC Press, pp. 239-294. Dean, D.C., Newby, R.F., and Bourgeosis, S. (1988) Regulation of fibronectin biosynthesis by dexamethasone, transforming growth factor-p, and CAMP in human cell lines. J. Cell Biol. 106:21592170. Derynck, R., Jarrett, J.A., Chen, E.Y., Eaton, D.H., Bell, J.R., Assoian, R.K., Roberts, A.B., Sporn, M.B., and Goeddel, D.V. (1985) Human transforming growth factor-p complementary DNA sequence and expression in normal and transformed cells. Nature 316~701-705. Dunn, B., and Boone, E. (1976) Growth of the chick embryo in vitro. Poult. Sci. 551067-1071. Feinberg, A.P., and Vogelstein, B.A. (1983) A technique for radiolabelling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem. 132:6-13. Gawande, S.R., and Tuan, R.S. (1990) Action of 1,25(0H), vitamin D, on calcium metabolism and skeletal development of the developing chick embryo. Trans. Orthop. Res. Soc. 15:401. Gilbert, S.F. (1988) Developmental Biology, 2nd ed. Sunderland: Sinauer Associates, Chapt. 6, pp. 196-242. Hayamizu, T.F., Sessions, S.K., Wanek, N., and Bryant, S.V. (1991) Effects of localized application of transforming growth factor p l on developing chick limbs. Dev. Biol. 145164-173. Heine, U.I., Munoz, E.F., Flanders, K.C., Ellingworth, L.R., Lam, H.Y .P., Thompson, N.L., Roberts, A.B., and Sporn, M.B. (1987) Role of transforming growth factor-p in the development of the mouse embryo. J . Cell Biol. 105:2861-2876. Ignotz, R.A., and Massague, J . (1986) Transforming growth factor-@ stimulates the expression of fibronectin and collagen and their incorporation into the extracellular matrix. J. Biol. Chem. 261:43374345. Jacenko, O., and Tuan, R.S. (1986) Calcium deficiency induces expression of cartilage-like phenotype in chick embryonic calvaria. Dev. Biol. 133:221-234. Jacenko, O., and Tuan, R.S. (1987) Changes in extracellular matrix during experimentally induced calcium deficiency in chick embryos. Prog. Clin. Biol. Res. 217B:401-404. Jackson, S.F., and Randall, J . (1956) Fibrogenesis and the formation of matrix in developing bone. In: “Ciba Foundation Symposium of Bone Structure and Metabolism,” Wolstenholm, G., and OConnor, H. (eds). Boston: Little, Brown, pp. 47-62. Jakowlew, S.B., Dillard, P.J., Sporn, M.B., and Roberts, A.B. (1988a) Nucleotide sequence of chicken transforming growth factor beta 1 (TGF-B1). Nucleic Acids Res. 168730. Jakowlew, S.B., Dillard, P.J., Kondaiah, P., Sporn, M.B., and Roberts, A.B. (1988b) Complementary deoxyribonucleic acid cloning of a novel transforming growth factor-p messenger ribonucleic acid from chick embryo chondrocytes. Mol. Endocrinol. 2:747-755. Jakowlew. S.B.. Dillard. P.J.. h o r n . M.B.. and Roberts. A.B. (1988~) Complementary deoxyribonucleic acid cloning of a messenger riboI

,

I

I

_

I

nucleic acid encoding transforming growth factor p4 from chicken embryo chondrocytes. Mol. Endocrinol. 2:1186-1195. Jakowlew, S.B., Dillard, P.J., Sporn, M.B., and Roberts, A.B. (1990) Complementary deoxyribonucleic acid cloning of a n mRNA encoding transforming growth factor-p2 from chicken embryo chondrocytes. Growth Factors 2:123-133. Jakowlew, S.B., Dillard, P.J., Winokur, T.S., Flanders, K.C., Sporn, M.B., and Roberts, A.B. (1991) Expression of transforming growth factor-ps 1-4 in chicken embryo chondrocytes and myocytes. Dev. Biol. 143:135-148. Joyce, M.E., Roberts, A.B., Sporn, M.B., and Bolander, M.E. (1990) Transforming growth factor-p and the initiation of chondrogenesis and osteogenesis in the rat femur. J . Cell Biol. 110:2195-2207. Kahn, A.J., and Simmons, D.J. (1977) Chondrocyte-to-osteocyte transformation in grafts of perichondrium-free epiphyseal cartilage. Clin. Orthop. 129:299-304. Kasid, A., Bell, G.I., and Director, E.P. (1988) Effects of transforming growth factor-p on human lymphokine-activated killer cell precursors-autocrine inhibition of cellular proliferation and differentiation to immune killer cells. J. Immunol. 141:690-698. Kondaiah, P., Sands, M.J., Smith, J.M., Fields, A,, Roberts, A.B., Sporn, M.B., and Melton, D.A. (1990) Identification of a novel transforming growth factor-p (TGF-p5) mRNA in Xenopus Zueuis. J. Biol. Chem. 2651089-1093. Kulyk, W.M., Rodgers, B.J., Greer, K., and Kosher, R.A. (1989) Promotion of embryonic chick cartilage differentiation by transforming growth factor-p. Dev. Biol. 135:424-430. Lehnert, S.A., and Akhurst, R.J. (1988) Embryonic expression pattern of TGF-p type 1 RNA suggests both paracrine and autocrine mechanisms of action. Development 104:263-273. Lengele, B., Dhem, A., and Schowing, J . (1990) Early development of the primitive cranial vault in the chick embryo. J . Craniofac. Genet. Dev. Biol. 10:103-112. Leonard, C.M., Fuld, H.M., Frenz, D.A., Downie, S.A., Massague, J., and Newman, S.A. (1991) Role of transforming growth factor-p in chondrogenic pattern formation in the embryonic l i m b Stimulation of mesenchymal condensation and fibronectin gene expression by exogeneous TGF-p and evidence for endogenous TGF-@like activity. Dev. Biol. 145:99-109. Massague, J. (1987) The TGF-p family of growth and differentiation factors. Cell 49:437-438. Massague, J. (1990) The transforming growth factor-p family. Annu. Rev. Cell Biol. 6597-641. McDonald, S.A., and Tuan, R.S. (1989) Expression of collagen type transcripts in chick embryonic bone detected by in situ cDNAmRNA hybridization. Dev. Biol. 133:221-234. Miller, D.A., Lee, A,, Matsui, Y., Chen, E.Y., Moses, H.L., and Derynck, R. (1989) Complementary DNA cloning of the murine transforming growth factor-P3 (TGF-p3) precursor and the comparative expression of TGF-p3 and TGF-@lmessenger RNA in murine embryos and adult tissues. Mol. Endocrinol. 3:1926-1934. M u s h , T.A., Pierce, G.F., Thomason, A., Gramates, P., Sporn, M.B., and Deuel, T.F. (1987) Transforming growth factor beta induces accelerated healing of incisional wounds in rats. Science 237:13331336. Noda, M. (1989) Transcriptional regulation of osteocalcin production by transforming growth factor-@in rat osteoblast-like cells. Endocrinology 124:612-617. Noda, M., and Camilliere, J.J. (1989) In vivo stimulation of bone formation by transforming growth factor-p. Endocrinology 124: 2991-2994. ODonnell, L.R., Majmudar, G., and Bonadio, J. (1991) Temporal and spatial distribution of mRNA for transforming growth factor+ types 1, 2, 3 and 4 in embryonic chick bone. Calcif. Tissue Int. 48:449. Ono, T., and Tuan, R.S. (1986) Effect of experimentally induced calcium deficiency on development, metabolism, and liver morphogenesis of the chick embryo. J . Embryol. Exp. Morphol. 92:207-222. Ono, T., and Tuan, R.S. (1991) Vitamin D and chick embryonic yolk calcium mobilization: Identification and regulation of expression of vitamin D-dependent CaZ -binding protein, calbindin-D,,,, in the yolk sac. Dev. Biol. 144:167-176. +

TGF-p IN CHICK EMBRYONIC CALVARIA Oshima, O., Leboy, P.S., McDonald, S.A., Tuan, R.S., and Shapiro, I.M. (1989) Developmental expression of genes in chick growth cartilage detected by in situ hybridization. Calcif. Tissue Int. 45:182192. Pelton, R.W., Nomura, S., Moses, H.L., and Hogan, B.L.M. (1989) Expression of transforming growth factor p-2 RNA during murine embryogenesis. Development 106:759-768. Pfeilschifter, J., Seyedin, S.M., and Mundy, G.R. (1988) Transforming growth factor beta inhibits bone resorption in fetal rat long bone cultures. J . Clin. Invest. 82:680-685. Rigby, P.W., Dreckmann, M., Rhodes, C., and Berg, P. (1977) Labeling deoxy-ribonucleic acid to high specific activity in vitro by nick translation with DNA polymerase. I. J . Mol. Biol. 113:237-251. Rizzino, A. (1988) Transforming growth factor-p: Multiple effects of cell differentiation and extracellular matrices. Dev. Biol. 130:411422. Roberts, A.B., Sporn, M.B., Assoian, R.K., Smith, J.M., Roche, N.S., Wakefield, L.M., Heine, U.I., Liotta, L.A., Falanga, V., Kehrle, J.H., and Fauch, A.S. (1986) Transforming growth factor type p: Rapid induction of fibrosis and angiogenesis in vivo and stimulation of collagen formation in vitro. Proc. Natl. Acad. Sci. U.S.A. 83: 4167-4171. Robey, P.G., Young, M.F., Flanders, K.C., Roche, N.S., Kondaiar, P, Reddi, H., Termine, J.D., Sporn, M.B., and Roberts, A.B. (1987) Osteoblasts synthesize and respond to tranforming growth factortype p (TGF-p) in vitro. J. Cell Biol. 105:457-463. Rosen, D.M., Stempien, S.A., Thompson, A.Y., and Seyedin, S.M. (1988) Transforming growth factor-beta modulates the expression of osteoblasts and chondroblasts phenotype in vitro. J . Cell Physiol. 134:337-345. Sandberg, M., Autio-Harmainen, H., and Vuorio, E. (1988) Localization of the expression of type I, 111, and IV collagen, TGF-p1 and c-fos genes in developing human calvarial bones. Dev. Biol. 130: 324-334. Seyedin, S.M., Segarini, P.R., Rosen, D.M., Thompson, A.Y., Bentz, H., and Graycar, J . (1987) Cartilage-inducing factor P is a unique protein structurally and functionally related to transforming growth factor-p. J . Biol. Chem. 262:1946-1949. Seyedin, S.M., Rosen, D.M., and Segarini, P.R. (1988) Modulation of chondroblast phenotype by transforming growth factor-p. Pathol. Immunopathol. Res. 7:38-42. Sporn, M.B., Roberts, A.B., Wakefield, L.M., and de Crombrugghe, B. (1987) Some recent advances in the chemistry and biology of transforming growth factor-beta. J. Cell Biol. 105:1039-1045.

313

Sporn, M.B., and Roberts, A.B. (1990)TGF-P: problems and prospects. Cell Regulation 12375-882. Stein, G., Lian, J., and Owen, T. (1990) Relationship of cell growth to the regulation of tissue-specific gene expression during osteoblast differentiation. FASEB J . 4:3111-3123. Tuan, R.S. (1980) Calcium transport and related functions in the chorioallantoic membrane of cultured shell-less chick embryos. Dev. Biol. 74:196-204. Tuan, R.S. (1983) Supplemented eggshell restores calcium transport in chorioallantoic membrane of cultured shell-less chick embryos. J. Embryol. Exp. Morphol. 74:119-131. Tuan, R.S., and Lynch, M.H. (1983) Effect of experimentally induced calcium deficiency on the developmental expression of collagen types in chicken embryonic skeleton. Dev. Biol. 100:374-386. Tuan, R.S., and Nguyen, H.Q. (1987) Cardiovascular changes in calcium-difficient chick embryos. J. Exp. Med. 1651418-1423. Tuan, R.S., Moore, C.J., Brittingham, J.W., Kinvin, J.J., Akins, R.E., and Wong, M. (1991) In vitro study of placental trophoblast calcium uptake using JEG-3 human cholocarcinoma cells. J. Cell Sci. 98: 333-342. Tuan, R.S. (1991) Ionic regulation of chondrogenesis. In: “Molecular Biology of Cartilage,” Hall, B., and Newman, S. (eds). Boca Raton: CRC Press, pp. 153-178. Van Obberghen-Schilling, E., Roche, N.S., Flanders, K.C., Sporn, M.B., and Baker, C.C. (1988) Transforming growth factor p l positively regulates its own expression in normal and transformed cells. J . Biol. Chem. 263:7741-7746. Varga, J., Rosenbloom, J., and Jimenez, S. (1987) Transforming growth factor-p (TGF-p) causes a persistent increase in steady-state amounts of type I and type I11 collagen and fibronectin mRNAs in normal human dermal fibroblasts. Biochem. J. 247:597-604. Villanueva, J.E., Nishimoto, S.K., and Nimni, M.E. (1989) Cells isolated from fetal rat calvaria by isopycnic separation express different collagen phenotypes. Matrix 9:40-48. Wang, E.A., Rosen, V., DAlessandro, J.S., Bauduy, M., Cordes, P., Harada, T., Israel, D.I., Hewick, R.M., Kerns, K.M., LaPan, P., Luxenberg, D.P., McQuaid, D., Moutsatsos, I.K., Nove, J., and Wozney, J.M. (1990) Recombinant human bone morphogenetic protein induces bone formation. Proc. Natl. Acad. Sci. U.S.A. 87:22202224. Wong, M., and Tuan, R.S. (1990) Modulation of chondrogenic activity of chick embryonic calvarial cells in vitro. Teratology 41:600.