Answers are grounded in this paper and cite their sources.
Parathyroid hormone (PTH) is known to be a major regulator of extracellular calcium homeostasis and can defend against reductions in extracellular calcium at least in part by enhancing bone resorption. Although the major glandular form of PTH is a straight-chain 84 amino acid peptide, most if not all of its bioactivity is believed to reside within the NH2-terminal 34 residues. Much additional interest in PTH has been generated in recent years by the discovery that it can exert profound anabolic effects on bone1 and that these effects may be of therapeutic value in osteoporosis.2 Although PTH-related protein (PTHrP), a peptide comprised of 139–173 amino acids, was also initially recognized for its skeletal catabolic effects as part of the syndrome associated with hypercalcemia of malignancy, this entity has also been shown to exert anabolic effects on bone similar to those of PTH.3 Furthermore, the NH2-terminal 34–36 amino acids of PTHrP appear to be sufficient for eliciting anabolic activity in bone although other actions have been ascribed to more COOH regions of the molecule.4, 5 The mechanistic basis for the anabolic effects of PTH and PTHrP remain incompletely defined. Considerable insight into this issue was gained by the observation that PTH (and PTHrP) can bind directly to cells of the osteoblastic lineage6 and can modulate downstream activities in osteoblasts.7, 8 In fact, effects on osteoclasts to cause bone resorption appear indirect and mediated by the direct action of these hormones on osteoblastic cells.9 Although the precise function of the different cell types comprising the osteoblast family remain to be defined, these observations nevertheless suggested that cells capable of inducing bone formation could be direct targets of PTH and PTHrP. PTHrP synthesized locally by the osteoblast appears to be important for normal bone development.10 Additionally, the bone-building effects of PTH and PTHrP seems to involve their capacity to locally up-regulate growth factors such as insulin-like growth factor via their actions on the osteoblast.11 Despite these observations, most of the cell and molecular physiology underlying the anabolic effects of these hormones is unknown, including the molecular basis for the observation that continuous high-dose administration of PTH causes predominantly osteoclastic resorption, whereas lower intermittent doses elicit anabolic effects.1 Further insight into the mechanism of action of these two hormones was provided by the discovery of the type I PTH/PTHrP receptor (PTHR).12 This G protein–coupled seven-transmembrane spanning receptor was shown to bind the NH2-terminal regions of both PTH and PTHrP with approximately equal affinity. These observations seemed to provide a molecular basis for the finding that synthetic NH2-terminal fragments of the two hormones exerted similar effects on skeletal and renal physiology.13-15 Abundant evidence has shown that the PTHR activates the adenylate cyclase/cAMP system in which Gsα is known to be a transducing agent, and the phospholipase C/inositol phosphate/diacyl glycerol/intracellular calcium system in which the transducing protein Gq has been implicated.12 Expression of the PTHR was observed in the classical PTH target tissues, bone and kidney, but was also found in many other tissues.16, 17 Inasmuch as osteoblastic cells exhibit the capacity to synthesize PTHrP as well as to express the PTHR,10 the efficacy of added exogenous PTH or PTHrP to activate osteoblasts will depend most likely not only on the local concentrations of these administered agonists but also on the state of occupancy and abundance of the PTHR as influenced by locally produced endogenous PTHrP and by circulating endogenous PTH. Although a type II PTH/PTHrP receptor which selectively binds PTH has been found and other receptors for PTH and PTHrP ligands have been postulated, their role in skeletal physiology is currently uncertain. Considerable effort has been expended by several groups over many years to define the regions of the NH2-terminal domain of the PTH molecule (and more recently of the PTHrP molecule) that are required to interact with the PTHR and stimulate downstream signaling. These studies examined full agonist, partial agonist, and antagonist activities of a variety of synthetic analogs of PTH (1–34), generally employing in vitro adenylate cyclase-stimulating activity as an index of biological response. Early studies demonstrated that either deletion or extension of the NH2 terminus of PTH markedly inhibited the capacity of this hormone to increase cAMP in target tissues.18, 19 Subsequently, the NH2 terminus of PTHrP was shown to manifest similar sensitivity.14, 20 This work suggested that the NH2-terminal domain of these hormones is a critical site for activation of the adenylate cyclase signaling system. Other work focused on the role of the COOH-terminal domain of PTH (1–34) in receptor binding, attempting to dissect a COOH-terminal “binding” domain in this entity from the NH2-terminal “activation” domain. Stimulated by the initial observation that poorly active NH2-terminal deleted analogs of PTH could still function as antagonists,19 this work ultimately suggested that the 14–34 portion of PTH contains the predominant binding domain.21 A number of studies have now been reported in which the structural basis of PTHrP function has also been analyzed, generally in comparison with PTH. Such studies have shown for example that although PTH and PTHrP lack significant amino acid sequence homology in the 14–34 domains, these domains bind to osteoblastic cells with similar avidity, suggesting that these regions retain similar molecular topologies.22, 23 Despite many additional studies, however, it remains premature to assign exact functional equivalence to most of the different domains and residues in these two hormones. Both PTH and PTHrP assume only a modestly ordered structural configuration in aqueous solution. Several nuclear magnetic resonance spectroscopy studies of PTH(1–34) and of PTHrP(1–34) in nonaqueous solutions (that are intended to mimic the membrane-bound receptor milieu) have disclosed some elements of secondary structure and have assigned two α-helical domains to these molecules, one less stable, extending for 7–11 amino acids in the NH2-terminal region, and another more stable, of 9–15 residues in the COOH domain.24-29 Some, but not all, studies have suggested interaction between the helices via a hydrophobic interface. In addition, the linker between the two helices is believed to be a turn or flexible hinge region. Cyclization of PTH and PTHrP analogs by a specific lactam apparently promotes α-helical conformation, and the stabilization of the α-helix in the putative COOH-terminal receptor-binding domain of PTH and PTHrP30, 31 has been correlated with enhanced bioactivity of these hormones. To date, however, determination of the three dimensional X-ray structure of PTH or PTHrP has not been feasible because of the difficulty in obtaining high-quality crystals. Extensive analysis of the PTHR has also been performed over the past few years to attempt to delineate the structural basis of its function. Such studies employing mutant and chimeric PTHR have indicated that the NH2-terminal extracellular domain of the PTHR in particular is important for ligand binding and appears to interact with the COOH region of PTH or PTHrP.32 Nevertheless, amino acid residues elsewhere in the PTHR, such as in the third extracellular loop, may also contribute to this binding.33 Residues in the membrane-spanning helices of the PTHR may be required for binding residues in the NH2-terminal domain of the ligand. Amino acids in the extracellular loops and the transmembrane regions are involved in signaling by PTH analogs and residues in several intracellular loops and the COOH-terminal intracellular tail are implicated in linkage to both Gsα and Gqα. Residues in the second intracellular loop appear to be especially critical for interaction with Gqα.34 Other studies have identified sites in the COOH tail of the PTHR that may be important for receptor internalization and phosphorylation.35 Consequently, a great deal of important information has been gleaned and continues to be generated on specific sites in the PTHR which may be necessary for hormone binding, hormone signaling, and PTHR regulation. Nevertheless, a comprehensive understanding of the detailed interaction of PTH and of PTHrP with the PTHR will have to await the determination of the three-dimensional structure of the bimolecular complex. Some of the most interesting insights into receptor function have come from experiments of nature. Thus, naturally occurring mutations in which arginine is substituted for histidine at position 223 in the second transmembrane segment of the PTHR and in which proline is substituted for tryptophan at position 410 of the sixth transmembrane domain region each led to constitutively active receptors capable of stimulating adenylate cyclase in the absence of a PTH agonist.36 Ultimately the overactive receptors cause the autosomal dominant syndrome of Jansen-type metaphyseal chondrodysplasia. This activation, occurring in the absence of ligand, almost certainly results from a critical change in the conformation of the receptor. This would therefore support the notion that the important function of the ligand is to induce or stabilize a receptor conformation that is favorable for signaling. Ultimate testing of this hypothesis must await crystallization of the PTHR and cocrystallization of a PTH or PTHrP ligand with the PTHR, followed by determination and comparison of the three-dimensional structure of the inactive and active forms. Are PTHR-linked signal transduction mechanisms other than adenylate cyclase required to elicit downstream anabolic effects of this hormone? The capacity of ligand to stimulate the phospholipase C system via the PTHR is complex and may be cell type–dependent, cell cycle–dependent, and dependent on the receptor abundance within the target cell.37 Several studies have suggested that the phospholipase C stimulating domain of the PTH(1–34) molecule resides in the COOH terminus of the molecule.38-41 Nevertheless, NH2-terminally truncated PTH forms which fail to stimulate adenylate cyclase but apparently retain the capacity to stimulate the phospholipase C pathway lack anabolic activity in vivo.42 Consequently, adenylate cyclase stimulation seems absolutely necessary for in vivo anabolic activity, although it may or may not be uniquely sufficient. Further studies are therefore required to clarify these points. In this issue of the Journal, Frolick et al.43 demonstrate that a COOH-terminally substitutedanalog of PTHrP(1–34), RS-66271, binds to the PTHR with only 1/100th of the affinity of PTH(1–34) and is less potent than PTH(1–34) in stimulating inositol phosphate production. Nevertheless, it exhibits equivalent activity to PTH(1–34) in stimulating cAMP in vitro, in enhancing bone resorption in vitro, and in producing an increase in trabecular and cortical bone mass in rats in vivo. The structural modification in this analog is the substitution of a model amphipathic α-helical peptide for amino acids 22–31 in human PTHrP(1–34), a modification that has the apparent effect of enhancing the helical structure in the 16–32 region of PTHrP(1–34) in aqueous solution.31 Stabilization of theα-helical conformation in the RS-6627 analog of PTHrP(1–34), although producing an increase in adenylate cyclase stimulation in vitro and anabolic activity in vivo, therefore does not enhance receptor binding in this study. In general, previous studies have shown a good correlation between receptor affinity of a PTH ligand and its efficacy as an agonist in stimulating adenylate cyclase. Assuming that the discordance identified by Frolick et al.43 is not due to technical limitations of the in vitro approach, and is not due to binding to an alternative (perhaps as yet unidentified) receptor, this “exception” raises the issue of the nature of the rules governing ligand–receptor interactions which permit full signal transduction irrespective of binding affinity. Currently, there are at least two models that attempt to explain the distinct contributions of ligand domains to binding affinity and to signaling efficacy.43 In the “conformational induction” model, ligands bind to the receptor and induce it to assume a conformational change that leads to either an active state characterized by enhanced G-protein coupling or to an inactive state characterized by G-protein uncoupling. Some sites on the ligand may therefore be critical for docking to the receptor, whereas others may be critical for inducing a conformational change in the receptor. In the “conformational selection” model, the receptor spontaneously fluctuates between inactive and active conformers. Agonists have a higher binding affinity for the active state and agonist binding causes an increased concentration of active receptors. Inverse agonists have a higher affinity for the inactive state and therefore enhance the concentration of inactive receptors. Although in both models the separate effects of ligand domains on affinity and efficacy can be appreciated, the requirement for a positive correlation of ligand affinity with ligand efficacy seems most compelling with the “conformational selection” model. Thus, the capacity of an agonist to bind with relatively weak affinity but still maximally induce an active state as reported in the study of Frolick et al.43 is most compatible with the “conformational induction” model. The capacity to develop small molecule mimics of PTH and/or PTHrP with anabolic effects could complement available antiresorptive therapy for osteoporosis and would provide a therapeutic approach of great value. Consequently, new insights into the mechanisms of these hormones which may contribute to this goal represent important contributions to both biology and therapeutics. In this regard, there are several important practical implications of the study of Frolick et al.43 for the development of more potent anabolic PTH or PTHrP agonists. First, the study demonstrates that it would be important to be cautious in employing in vitro receptor binding data alone when evaluating the potential in vivo potency of newly designed PTH and PTHrP analogs. The strategy for improved agonist design should be to enhance the PTHR conformation required for G-protein coupling and not necessarily to enhance agonist binding affinity. Second, this study supports what has been increasingly observed, i.e., that the capacity to stimulate adenylate cyclase activity in bone cells in vitro may provide a reasonable screen for evaluating analogs with potential in vivo anabolic activity with the caveat that adenylate cyclase may be a “necessary” but not a “sufficient” signaling pathway for new bone formation. Finally, the study reemphasizes that measurement of the direct interaction of PTH with its receptor and signaling systems in isolation in vitro is no accurate substitute for in vivo assessment of anabolic potency which incorporates complexities of pharmacodynamics, metabolism, and milieu. It will be important to continue to explore and define the nature of these complexities in order to understand and harness the bone-building effects of PTH and PTHrP.
No takes yet. Share an insight, caveat, or question.
David Goltzman (1999) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: