| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Massachusetts General Hospital Boston, Massachusetts 02114
 |
Introduction |
|---|
 Top Introduction References |
|---|
PTH has been known for decades as a crucial regulator of blood calcium. By causing bone resorption, renal tubular calcium reabsorption, and activation of vitamin D, PTH raises blood calcium levels. Calcium then feeds back on the parathyroid cell through the recently characterized parathyroid calcium sensor and decreases the synthesis and secretion of PTH (1). This feedback loop strongly suggests that PTH’s teleological purpose is primarily the defense of blood calcium. In this paradigm, PTH doesn’t care about bone—it just uses bone to maintain blood calcium. It is no surprise, then, that in states of PTH excess, bone is weakened by excess resorption of matrix and mineral.
From this perspective, the observation that intermittent administration of PTH to rodents and humans leads to increases in bone mass (2) makes no sense at all. The resolution of this paradox—that PTH can both increase and decrease bone mass—will require greater understanding of the complexity of regulation of bone-forming cells (osteoblasts) and bone resorbing cells (osteoclasts). One part of this resolution may follow from the observation that osteoblast precursors and osteoblasts secrete PTH-related protein (PTHrP). Unlike PTH, PTHrP is predominantly a paracrine factor; PTHrP in bone may have roles well removed from simply the maintenance of calcium homeostasis (3).
Despite the distinct physiologic missions of PTH and PTHrP, many of the actions of PTH and PTHrP are mediated by a common PTH/PTHrP receptor, which interacts with the similar amino-terminal regions of each ligand (4). This receptor is found on osteoblasts and not on osteoclasts. Studies of PTH/PTHrP receptor knock-out mice have shown that this receptor functions physiologically as both a PTH receptor and a PTHrP receptor. Bones from these mice do not release 45Ca in response to PTH (Lanske, unpublished observations), nor do their chondrocytes differentiate normally in response to PTHrP (5). Activation of the PTH/PTHrP receptor stimulates both the Gs-adenylate cyclase and Gq family-phospholipase C pathways. This complexity of second messenger signals has led many groups to try to define the functions of each pathway in mediating the diverse actions of PTH and PTHrP. For example, the stimulation of collagenase 3 synthesis in osteoblasts depends virtually solely on the cAMP pathway (6), whereas the inhibition of renal proximal tubular transport of phosphate requires actions of both pathways (7).
The actions of PTH (and presumably PTHrP) on osteoblasts are hard to characterize because specific actions depend dramatically on the degree of differentiation of the osteoblast and on the nature of signals from adjacent cells and matrix. As noted earlier, in intact bone, intermittent administration of PTH leads to net bone formation. DNA synthesis is induced in osteoblast precursors and collagen synthesis is increased by intermittent PTH administration in organ culture (8). Despite these anabolic actions of PTH seen in organ culture and in vivo, PTH also has direct actions to inhibit the formation of bone nodules by osteoprogenitor cells in culture (9) and to decrease collagen production by cultured osteoblasts (10). A series of papers, including the work of Yu and Chandrasekhar in this issue of Endocrinology, explore the effect of PTH on the synthesis of osteocalcin, a protein synthesized only by the most mature osteoblasts. Perhaps not surprisingly, these papers differ widely in their conclusions; these differences most likely reflect the variable effects of PTH on less and more differentiated osteoblasts. But before discussing the results of Yu and Chandrasekhar (11), it would be useful first to consider osteocalcin as another familiar friend with unexpected behavior.
Osteocalcin has received attention for years because it is the most
prevalent noncollagenous protein in bone (12). It is a small 5.8-kDa
protein synthesized almost exclusively by mature osteoblasts and
odontoblasts. Like several clotting factors, it contains several
-carboxyglutamic acid residues that bind calcium. Clinical
investigators have found that osteocalcin levels in blood are a useful
index of bone formation (13). The recent study of mice missing the
osteocalcin gene (14) has provided the first glimpses of the
physiologic role of osteocalcin. These mice mineralize and resorb bone
normally but have osteoblasts that lay down more bone than normal.
Consequently, the bones of these mice contain more matrix and mineral
and are stronger than the bones of normal littermates. This result
leads to the hypothesis that one action of osteocalcin is to slow down
the anabolic activities of osteoblasts. From this perspective, the
synthesis of osteocalcin only by the most mature osteoblasts is a
homeostatic activity designed to dampen the activity of these
cells.
The osteocalcin gene in mice is actually two virtually identical, clustered genes, followed by a related gene. The promoter regions of the mouse and closely related rat osteocalcin genes contain multiple binding sites for transcription factors, including OSF 2 (15), a member of a highly conserved family (15, 16); OSF 2 is likely to influence the transcription of many genes in osteoblasts.
The study of Yu and Chandrasekhar (11) analyzes the effects of PTH on the osteocalcin gene’s promoter in an osteoblast-like human cell line, SaOS-2. Others have shown that SaOS-2 cells synthesize osteocalcin, though this synthesis has not always been found to be stimulated by amino-terminal fragments of PTH (17). Yu and Chandrasekhar do not report whether PTH stimulates the transcription of the endogenous PTH gene in the cells and under the conditions that they use. They do show that a luciferase reporter gene driven by a 1905-bp fragment of the rat osteocalcin promoter is stimulated by PTH detectably within 1 h, impressively after 4–8 h, and much less after 24 h. The mechanism of the transient nature of this stimulation is not explored but may reflect multiple direct and indirect effects of PTH to affect the levels and activities of many different transcription factors, as well as the changing sensitivity of the PTH/PTHrP receptor signaling system. The 1905-bp promoter responds to PTH more dramatically than do shorter promoter fragments. This result suggests that either this upstream region participates directly in PTH signaling or that it may, instead, synergize with downstream factors that mediate PTH signaling.
Yu and Chandrasekhar use a number of approaches to show that the
cAMP-protein kinase A pathway participates in the mediation of the PTH
signal. Forskolin, an activator of adenylate cyclase, mimics the effect
of PTH, and IBMX, a phosphodiesterase inhibitor that prolongs the
half-life of cAMP, dramatically potentiates the action of PTH. Further,
analogs of PTH that activate adenylate cyclase stimulate the
osteocalcin promoter, whereas analogs that fail to activate adenylate
cyclase do not. Finally, a relatively selective protein kinase A
inhibitor, H89, partly blocks the action of PTH, as does an
oligonucleotide complementary to the sequence encoding
RI
, one of four cAMP-binding regulatory subunits of cAMP
(18). In future studies, it would be of interest to see whether the
phospholipase C pathway plays an important role in mediating the PTH
signal, as well. Action through the phospholipase C pathway could
explain the synergy that Yu and Chandrasekhar note between the actions
of forskolin and PTH and might also explain the only partial effects of
H89 and the oligonucleotide (though other explanations for these
findings are also possible).
The authors also present results suggesting that the phospholipase C pathway does not participate in the osteocalcin promoter’s response to PTH. PTH analogs thought to activate phospholipase C but not adenylate cyclase fail to activate the osteocalcin promoter, and an analog thought to activate adenylate cyclase but not phospholipase C does activate the promoter. The effects of these analogs vary from system to system, however. The authors will need to demonstrate that these fragments behave as expected in their SaOS cells to allow more precise interpretation of these experiments.
PTH, then, can activate the promoter region of a gene associated with the phenotype of well differentiated osteoblasts. Is this part of PTH’s (or PTHrP’s) anabolic program? Or does the activation of the osteocalcin promoter represent another example of PTH turning off osteoblast function by activating a powerful inhibitor of osteoblasts? Of course, these are not mutually exclusive possibilities. It will take some time to make teleological sense out of the actions of PTH and osteocalcin. The rewards are likely to be new tools for treating bone disease and understanding the normal physiology of bone.
Received May 30, 1997.
|
References |
|---|
|
TopIntroductionReferences |
|---|
/AML-1-related factor increases osteocalcin promoter activity
through its binding to an osteoblast-specific cis-acting
element. J Biol Chem 270:30973–30979
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
