[Frontiers in Bioscience 3, d113-124, January 15, 1998]

	[Frontiers in Bioscience 3, d113-124, January 15, 1998] Reprints PubMed CAVEAT LECTOR
	CONTROL OF TGF-BETA RECEPTOR EXPRESSION IN BONE Michael Centrella, Changhua Ji, Thomas L. McCarthy Department of Surgery, Plastic Surgery Section, Yale University School of Medicine, 333 Cedar Street, PO Box 208041, New Haven, CT 06520-8041, Received 12/1/97 Accepted 12/5/97 8. REGULATION OF TGF-BETA RECEPTORS ON BONE CELLS Variations in TGF-beta receptor levels appear to parallel the changes in TGF-beta activity that occur as bone cell differentiation proceeds. Notably, the relative amounts of betaglycan, measured by both radio-iodinated TGF-beta binding assay and by mRNA analysis, are significantly lower on differentiated osteoblasts (11,21,68); this predicts that betaglycan expression is reduced as osteoblasts mature. The proportions of TGF-betaRI and TGF-betaRII also vary, and do so independently from each other, during this process. Specifically, in fetal rat bone cells, the relative amount of TGF-beta binding to TGF-betaRII also decreases when osteoblast-like activity increases and betaglycan levels fall, while the proportion of TGF-beta binding to TGF-betaRI increases significantly. These patterns too are consistent with changes in steady state mRNA levels for TGF-betaRI and TGF-betaRII (21). Some studies in the pre-osteoblastic neonatal murine MC3T3-E1 cell line suggest large decreases in all three TGF-beta receptors and in TGF-beta function during the processes of collagen matrix deposition and mineralization (76,77) in vitro. Changes like this do not appear to occur in primary osteoblast-enriched cell cultures from fetal rat bone (45, and other unpublished studies). These differences may relate to the phenotype of MC3T3-E1 cells that endogenously express very little TGF-betaRI (68). Thus, in the absence of specific osteogenic inducers, changes in the proportion of TGF-betaRI may be difficult to note when its levels are initially so low. In contrast, and perhaps more consistent with results in primary bone cell cultures, immunohistochemical staining during mouse organogenesis showed a more ubiquitous staining pattern for TGF-betaRI than for TGF-betaRII. In these studies, staining for TGF-betaRII was preferentially found in regions of undifferentiated cells that thereafter differentiated into bone (78), much like the changes that occur in fetal rat cell cultures (21). Furthermore, in human osteosarcoma derived MG-63 cell cultures, differentiation induced by 1,25 dihydroxyvitamin D3, suppresses the levels of TGF-betaRII in parallel with a decrease in certain aspects of TGF-beta activity (79). Agents that increase TGF-beta binding to betaglycan tend to reduce the stimulatory effects of TGF-beta on osteoblast function. Small but significant increases in binding occur in response to parathyroid hormone, and larger and more sustained effects occur with glucocorticoid treatment (11,19,80). However, although betaglycan levels can vary dramatically on bone cells with differentiation (described above), with these osteotropic hormones, or with growth factor treatment (see below), lack of promoter information has severely limited our ability to understand how these events occur at the molecular level. Glucocorticoid decreases TGF-beta binding to a small extent to TGF-betaRII, an effect that increases with longer exposure to hormone (19). Nonetheless, initial studies with TGF-betaRII promoter constructs (63) show no consistent changes in promoter activity when bone cells are treated with glucocorticoid , or with BMP-2, TGF-beta, PGE2, or retinoic acid (unpublished data). These results and earlier studies showing large changes in betaglycan in response to these agents (19,21,73, and unpublished data) support that TGF-beta binding to TGF-betaRII is in part regulated indirectly by the proportion of betaglycan present on the cell surface (21,41,81). Little else is currently known about the molecular mechanisms that mediate changes in betaglycan or TGF-betaRII expression in bone cells or in any tissue. In contrast to small or negligible changes in TGF-betaRII, the proportion TGF-beta binding to TGF-betaRI is rapidly and significantly affected by positive and negative regulators of osteoblast activity (19,21). New studies, described in more detail below, predict how changes in TGF-betaRI expression may occur by variations in the level of a more recently appreciated, osteoblast-restricted nuclear factor, CBFa1. As described earlier, other TGF-beta supergene family members that regulate skeletal cells include the BMPs. These agents were first defined by a functional assay where they initiated cartilage formation that was then replaced by bone at ectopic sites in vivo in the rat (20,22). Some BMPs are more effective on osteoblasts, and others are more effective on less differentiated bone cells. Certain effects induced by several BMP family members are distinct from those induced by TGF-betas, and in some instances differ even from each other. BMPs enhance the synthesis of cartilage proteoglycan, alkaline phosphatase, osteocalcin, and PTH receptors in cultures of uncommitted stromal cells, chondrocytes, and/or osteoblast-like cells (21,45,81-92). Mutations in the mouse short ear gene (that is genetically related to human BMP-5) correlate with abnormal growth, skeletal formation, and fracture repair in these animals (93). BMPs act through specific BMP receptors and do not directly compete for binding at high affinity TGF-beta receptors. Nevertheless, BMP-2 and BMP-4 potently alter TGF-beta binding on osteoblasts. In direct contrast to the effects of glucocorticoid, these BMPs rapidly decrease TGF-beta binding to TGF-betaRII and betaglycan, and increase its binding to TGF-betaRI. In so doing, BMP treatment alters the effects of TGF-beta on bone cell replication, matrix protein synthesis, and alkaline phosphatase activity in patterns that are consistent with progressive increases in osteoblast differentiation (21). Other studies recently showed that BMP-2 can oppose the inhibitory effect of glucocorticoid on TGF-beta binding and activity. Surprisingly, BMP-2 also suppresses the stimulatory effect of glucocorticoid on a gene promoter containing a positive glucocorticoid response element (94). Some osteoblast regulators may therefore alter TGF-beta induced osteoblast activity by re-distributing TGF-beta binding among its various receptors. Changes in TGF-beta binding may in part depend on new TGF-beta receptor synthesis (19,20,73). While these findings suggest that effects by several regulators of bone cell activity converge at the molecular level on TGF-beta receptor expression, detailed information requires the new molecular tools that have become available only recently.

[Frontiers in Bioscience 3, d113-124, January 15, 1998]
Reprints
PubMed
CAVEAT LECTOR

CONTROL OF TGF-BETA RECEPTOR EXPRESSION IN BONE

Michael Centrella, Changhua Ji, Thomas L. McCarthy

Department of Surgery, Plastic Surgery Section, Yale University School of Medicine, 333 Cedar Street, PO Box 208041, New Haven, CT 06520-8041,

Received 12/1/97 Accepted 12/5/97

8. REGULATION OF TGF-BETA RECEPTORS ON BONE CELLS

Variations in TGF-beta receptor levels appear to parallel the changes in TGF-beta activity that occur as bone cell differentiation proceeds. Notably, the relative amounts of betaglycan, measured by both radio-iodinated TGF-beta binding assay and by mRNA analysis, are significantly lower on differentiated osteoblasts (11,21,68); this predicts that betaglycan expression is reduced as osteoblasts mature. The proportions of TGF-betaRI and TGF-betaRII also vary, and do so independently from each other, during this process. Specifically, in fetal rat bone cells, the relative amount of TGF-beta binding to TGF-betaRII also decreases when osteoblast-like activity increases and betaglycan levels fall, while the proportion of TGF-beta binding to TGF-betaRI increases significantly. These patterns too are consistent with changes in steady state mRNA levels for TGF-betaRI and TGF-betaRII (21).

Some studies in the pre-osteoblastic neonatal murine MC3T3-E1 cell line suggest large decreases in all three TGF-beta receptors and in TGF-beta function during the processes of collagen matrix deposition and mineralization (76,77) in vitro. Changes like this do not appear to occur in primary osteoblast-enriched cell cultures from fetal rat bone (45, and other unpublished studies). These differences may relate to the phenotype of MC3T3-E1 cells that endogenously express very little TGF-betaRI (68). Thus, in the absence of specific osteogenic inducers, changes in the proportion of TGF-betaRI may be difficult to note when its levels are initially so low. In contrast, and perhaps more consistent with results in primary bone cell cultures, immunohistochemical staining during mouse organogenesis showed a more ubiquitous staining pattern for TGF-betaRI than for TGF-betaRII. In these studies, staining for TGF-betaRII was preferentially found in regions of undifferentiated cells that thereafter differentiated into bone (78), much like the changes that occur in fetal rat cell cultures (21). Furthermore, in human osteosarcoma derived MG-63 cell cultures, differentiation induced by 1,25 dihydroxyvitamin D3, suppresses the levels of TGF-betaRII in parallel with a decrease in certain aspects of TGF-beta activity (79).

Agents that increase TGF-beta binding to betaglycan tend to reduce the stimulatory effects of TGF-beta on osteoblast function. Small but significant increases in binding occur in response to parathyroid hormone, and larger and more sustained effects occur with glucocorticoid treatment (11,19,80). However, although betaglycan levels can vary dramatically on bone cells with differentiation (described above), with these osteotropic hormones, or with growth factor treatment (see below), lack of promoter information has severely limited our ability to understand how these events occur at the molecular level. Glucocorticoid decreases TGF-beta binding to a small extent to TGF-betaRII, an effect that increases with longer exposure to hormone (19). Nonetheless, initial studies with TGF-betaRII promoter constructs (63) show no consistent changes in promoter activity when bone cells are treated with glucocorticoid , or with BMP-2, TGF-beta, PGE2, or retinoic acid (unpublished data). These results and earlier studies showing large changes in betaglycan in response to these agents (19,21,73, and unpublished data) support that TGF-beta binding to TGF-betaRII is in part regulated indirectly by the proportion of betaglycan present on the cell surface (21,41,81). Little else is currently known about the molecular mechanisms that mediate changes in betaglycan or TGF-betaRII expression in bone cells or in any tissue.

In contrast to small or negligible changes in TGF-betaRII, the proportion TGF-beta binding to TGF-betaRI is rapidly and significantly affected by positive and negative regulators of osteoblast activity (19,21). New studies, described in more detail below, predict how changes in TGF-betaRI expression may occur by variations in the level of a more recently appreciated, osteoblast-restricted nuclear factor, CBFa1.

As described earlier, other TGF-beta supergene family members that regulate skeletal cells include the BMPs. These agents were first defined by a functional assay where they initiated cartilage formation that was then replaced by bone at ectopic sites in vivo in the rat (20,22). Some BMPs are more effective on osteoblasts, and others are more effective on less differentiated bone cells. Certain effects induced by several BMP family members are distinct from those induced by TGF-betas, and in some instances differ even from each other. BMPs enhance the synthesis of cartilage proteoglycan, alkaline phosphatase, osteocalcin, and PTH receptors in cultures of uncommitted stromal cells, chondrocytes, and/or osteoblast-like cells (21,45,81-92). Mutations in the mouse short ear gene (that is genetically related to human BMP-5) correlate with abnormal growth, skeletal formation, and fracture repair in these animals (93). BMPs act through specific BMP receptors and do not directly compete for binding at high affinity TGF-beta receptors. Nevertheless, BMP-2 and BMP-4 potently alter TGF-beta binding on osteoblasts. In direct contrast to the effects of glucocorticoid, these BMPs rapidly decrease TGF-beta binding to TGF-betaRII and betaglycan, and increase its binding to TGF-betaRI. In so doing, BMP treatment alters the effects of TGF-beta on bone cell replication, matrix protein synthesis, and alkaline phosphatase activity in patterns that are consistent with progressive increases in osteoblast differentiation (21). Other studies recently showed that BMP-2 can oppose the inhibitory effect of glucocorticoid on TGF-beta binding and activity. Surprisingly, BMP-2 also suppresses the stimulatory effect of glucocorticoid on a gene promoter containing a positive glucocorticoid response element (94).

Some osteoblast regulators may therefore alter TGF-beta induced osteoblast activity by re-distributing TGF-beta binding among its various receptors. Changes in TGF-beta binding may in part depend on new TGF-beta receptor synthesis (19,20,73). While these findings suggest that effects by several regulators of bone cell activity converge at the molecular level on TGF-beta receptor expression, detailed information requires the new molecular tools that have become available only recently.