To the Editor:

  1. Top of page
  2. To the Editor:

In a recent study Frolick et al. reported an extensive in vivo and in vitro pharmacological comparison of human parathyroid hormone (1–34) (hPTH [1–34]) and RS-66721.(1) The latter peptide ligand is an analogue of PTH-related protein (1–34) (PTHrP[1–34]) that has residues 22–31 replaced with a model amphipathic α-helical peptide sequence. The ligand was developed as a potential candidate molecule for the treatment of human osteoporosis.(2) In the study of Frolick et al. the in vivo anabolic activity of the compound, measured in a number of assays using rats, was not significantly different from the activity of hPTH(1–34). Similarly the in vitro activity of RS-66271, measured in bone resorption, cAMP, and inositol phosphate assays, was equivalent to or only slightly less than that of hPTH(1–34). However, a large difference in binding affinity between RS-66271 and hPTH(1–34) was observed in radioligand binding assays; RS-66271 was approximately 100-fold less potent than hPTH(1–34) for inhibition of [125I]PTH(1–34) binding to cells expressing the cloned human PTH/PTHrP receptor.

The discrepancy is significant because it affects the conclusions drawn about the site of action of RS-66271.(1,3,4) It also raises the issue of the utility of radioligand binding assays in the development of antiosteoporetic ligands.(3) Frolick et al. suggest that the markedly lower binding affinity of RS-66271 could arise from different PTH/PTHrP receptor splice variants, a novel receptor, or pharmacokinetic effects.(1) In the accompanying editorial(3) Goltzman suggests that the difference in binding affinity between the two ligands could arise from a difference in the mode of interaction with the PTH/PTHrP receptor, involving the contributions of conformational selection and induction in the processes of ligand binding and receptor activation.

We have accumulated data that suggest an alternative explanation. We believe that the discrepancy arises from the different PTH/PTHrP receptor conformations presented to ligand that result from receptor-G-protein interaction, a possibility that has been raised previously in theoretical terms.(4) Our radioligand-binding data(5,6) suggest that hPTH(1–34) binds with high affinity to the PTH/PTHrP receptor coupled to G-protein (RG) and with high affinity to the uncoupled receptor (R).(5) RS-66271 binds with high affinity to RG but with a much lower affinity to R.(6) It is argued below that in whole cell binding experiments, such as the assay used by Frolick et al., the receptor population is predominantly in the uncoupled R state. This hypothesis predicts the reduced binding affinity of RS-66271 compared with hPTH(1–34) observed in whole cell assays.(1)

The effect of RG coupling on agonist-binding affinity can be evaluated using the commonly accepted ternary complex model.(7,8) In this model, agonist binding increases the affinity of receptor for G-protein and, reciprocally, RG interaction increases the affinity of the agonist for the receptor. Agonist binding to the RG state is typically identified as a high-affinity-binding state in agonist-binding assays. (Binding of agonist ligands is usually evaluated by competition versus a radiolabeled-antagonist ligand,(7) although direct binding of radiolabeled agonists is also used.) Guanine nucleotides bind to the α-subunit of the heterotrimeric G-protein. This interaction is believed to result in breakdown of the RG complex, producing a receptor population that is predominantly uncoupled from G-protein. The binding of agonist ligands to the uncoupled R can be measured by competition versus a radiolabeled antagonist in the presence of GTP, or GTP analogues such as guanosine 5′-O-(3-thiotriphosphate).

We have measured the binding of hPTH(1–34) and RS-66271 in competition binding assays using [125I][Nle8,18, Tyr34]bPTH(3–34) as an antagonist radioligand. Isolated cell membranes were used as the vehicle for receptor to enable us to modify the chemical environment of the receptor (which is considerably more difficult to control when using whole cells.) Agonist binding to the RG state was evaluated in the absence of guanine nucleotide and binding to the uncoupled R was measured in the presence of 10 μM GTPγS. For hPTH(1–34), high-affinity binding (IC50 < 10 nM) was observed both in the presence as well as the absence of GTPγS (Fig. 1), suggesting that this ligand binds with high affinity to both the uncoupled R as well as the RG state.(5) For RS-66271, a high-affinity component was observed in the absence of GTPγS (IC50 = 16 pM), representing 16% of the total inhibition of [125I][Nle8,18, Tyr34]bPTH(3–34) binding (Fig. 1).(6) The remainder of RS-66271 binding in the absence of GTPγS was of low affinity (IC50 > 100 nM, Fig. 1). In the presence of GTPγS, the high-affinity state was not detected, the entire inhibition curve being described by low-affinity interactions (Fig. 1).(6) This finding suggests that the high-affinity interaction of RS-66271 represents binding to the RG state. However, the ligand binds with low affinity to the uncoupled R, in contrast with hPTH(1–34), which binds with high affinity to this receptor state. Therefore, at the uncoupled R, RS-66271 would appear to bind with lower affinity than hPTH(1–34).

thumbnail image

Figure FIG. 1.. Inhibition of [125I][Nle8,18, Tyr34]bPTH(3–34) binding to the human PTH/PTHrP receptor by hPTH(1–34) (◯) and RS-66271 (A). Binding of the antagonist radioligand to HEK293hPTH/PTHrPr cell membranes was measured as described in refs. 5 and 6, using centrifugation to separate bound and free radioligand. Binding was measured in the absence of added GTPγS (A) and in the presence of 10 μM GTPγS (B). The figure is a composite of Fig. 5A of ref. 5 (hPTH(1–34) data) and Fig. 4B of ref. 6 (RS-66271 data). Further experimental details are provided in these references.

Download figure to PowerPoint

The receptor population of intact cells may be predominantly uncoupled from G-protein. In whole cells it has been proposed that the agonist-high-affinity RG state represents a transient intermediate in the G-protein activation cycle.(8) In this agonist-high-affinity state the guanine nucleotide binding site of the G-protein is empty, following dissociation of previously bound GDP. Intracellular concentrations of guanine nucleotides are in the millimolar range so GTP binds almost instantaneously to the empty G-protein α-subunit, leading to breakdown of the RG complex.(8) These considerations suggest that the whole cell binding assay used by Frolick et al.(1) measures the affinity of ligands for the uncoupled PTH/PTHrP receptor. The result from this whole cell assay, that RS-66271 binds with lower affinity than hPTH(1–34), is in agreement with our data for the uncoupled R in isolated cell membranes discussed above.(5,6) Indeed, the binding curves for the whole cell assay (Fig. 1 of Frolick et al.(1)) appear to be almost superimposable on the binding curves obtained with membranes in the presence of GTPγS (Fig. 1B).

An interesting question that arises from these data is the structural basis of the difference between hPTH(1–34) and RS-66271 binding. This probably does not reflect a difference between the binding affinity of hPTH(1–34) and hPTHrP (1–34) (the parent compound of RS-66271) because hPTHrP(1–34) and hPTH(1–34) bind with similar potency to the hPTH/PTHrP receptor.(5) The reduced affinity of RS-66217 at the uncoupled R could result from less favorable interactions of the receptor with the substituted residues in the 22–31 region of the ligand. Alternatively, differences in secondary structure may affect ligand-binding affinity with the uncoupled R. Structural comparisons have indicated that RS-66271 has a higher α-helical content that hPTHrP(1–34)—the helical domain extends beyond the incorporated amphipathic peptide region (22–31) up to residue 16.(4) These possibilities could be explored by manipulation of ligand and receptor structure, possibly in combination with structural modeling.

In conclusion, the isolated membrane-binding data strongly suggest that the PTH/PTHrP receptor is the site of action of RS-66271, because high-affinity binding and guanine nucleotide sensitivity were observed for this ligand. The apparent discrepancy between the binding affinities of hPTH(1–34) and RS-66271 can be accounted for by the different receptor states that arise from receptor-G protein coupling. We believe that ligand-binding assays may be useful in evaluating the structure-activity relationships of ligands directed at the PTH/PTHrP receptor, provided that the nature of the receptor states in the assay can be identified. In this regard, a radiolabeled analogue of RS-66271 may be a useful probe for the RG state of the PTH/PTHrP receptor. At the low concentrations used in binding assays, this radioligand may detectably label only the high-affinity RG state.


  1. Top of page
  2. To the Editor:
  • 1
    Frolick C, Cain RL, Sato M, Harvey AK, Chandrasekhar S, Black EC, Tashjian AH Jr, Hock J 1999 Comparison of recombinant human PTH(1-34) (LY333334) with a C-terminally substituted analogue of human PTH-related protein (1-34): In vitro activity and in vivo pharmacological effects in rats. J Bone Miner Res 14: 163172
  • 2
    Vickery BH, Avnur Z, Cheng Y, Chiou S-S, Leaffer D, Caulfield JP, Kimmel DB, Ho T, Krstenansky JL 1996 RS-66271, a C-terminally substituted analogue of human parathyroid hormone-related protein (1-34), increases trabercular and cortical bone in ovariectomized, osteopenic rats. J Bone Miner Res 11: 19431951.
  • 3
    Goltzman D 1999 Interactions of PTH and PTHrP with the PTH/PTHrP receptor and with downstream signaling pathways: Exceptions that provide the rules. J Bone Miner Res 14: 173177.
  • 4
    Pellegrini M, Bisello A, Rosenblatt M, Chorev M, Mierke DF 1997 Conformational studies of RS-66271, an analogue of parathyroid hormone-related protein with pronounced bone anabolic activity. J Med Chem 40: 30253031.
  • 5
    Hoare SRJ, De Vries G, Usdin TB 1999 Measurement of agonist and antagonist ligand-binding parameters at the human parathyroid hormone type 1 receptor: Evaluation of receptor states and modulation by guanine nucleotide. J Pharm Exp Ther 289: 13231333.
  • 6
    Hoare SRJ, Usdin TB 1999 Quantitative cell membrane-basedradioligand binding assays for parathyroid hormone receptors. J Pharm Toxicol Methods 41: 8390.
  • 7
    Deléan A, Stadel JM, Lefkowitz RJ 1980 A ternary complex model explains the agonist-specific binding properties of the adenylate cyclase-coupled β-adrenergic receptor. J Biol Chem 255: 71087117.
  • 8
    Friessmuth M, Waldhoer M, Bofill-Cardona E, Nanoff C 1999 G protein antagonists. Trends Pharmacol Sci 20: 237245.