PTHR1 (type 1 parathyroid hormone receptors) mediate the effects of PTH (parathyroid hormone) on bone remodelling and plasma Ca2+ homoeostasis. PTH, via PTHR1, can stimulate both AC (adenylate cyclase) and increases in [Ca2+]i (cytosolic free Ca2+ concentration), although the relationship between the two responses differs between cell types. In the present paper, we review briefly the mechanisms that influence coupling of PTHR1 to different intracellular signalling proteins, including the G-proteins that stimulate AC or PLC (phospholipase C). Stimulus intensity, the ability of different PTH analogues to stabilize different receptor conformations (‘stimulus trafficking’), and association of PTHR1 with scaffold proteins, notably NHERF1 and NHERF2 (Na+/H+ exchanger regulatory factor 1 and 2), contribute to defining the interactions between signalling proteins and PTHR1. In addition, cAMP itself can, via Epac (exchange protein directly activated by cAMP), PKA (protein kinase A) or by binding directly to IP3Rs [Ins(1,4,5)P3 receptors] regulate [Ca2+]i. Epac leads to activation of PLCϵ, PKA can phosphorylate and thereby increase the sensitivity of IP3Rs and L-type Ca2+ channels, and cAMP delivered at high concentrations to IP3R2 from AC6 increases the sensitivity of IP3Rs to InsP3. The diversity of these links between PTH and [Ca2+]i highlights the versatility of PTHR1. This versatility allows PTHR1 to evoke different responses when stimulated by each of its physiological ligands, PTH and PTH-related peptide, and it provides scope for development of ligands that selectively harness the anabolic effects of PTH for more effective treatment of osteoporosis.
Regulation of plasma Ca2+ concentration and bone remodelling by PTH (parathyroid hormone)
PTH plays a central role in regulating turnover of bone and in plasma Ca2+ (and phosphate) homoeostasis. PTH is released from chief cells in the parathyroid glands when the plasma Ca2+ concentration decreases and then, via its direct (kidney and bone) or indirect (intestines) actions, it contributes to restoring the plasma Ca2+ concentration . These responses are mediated by PTHR1 (type 1 PTH receptors), GPCRs (G-protein-coupled receptors) that bind PTH and PTHrP (PTH-related peptide) with similar affinity . The response of bone to PTH depends on interactions between osteoblasts, the cells that deposit bone, and osteoclasts, the cells that mediate bone resorption, because only osteoblasts express PTHR1. Exploitation of PTHR1 as a target for treatment of osteoporosis seeks selectively to harness the anabolic effects of PTH while minimizing the osteoclast-mediated resorption and associated hypercalcaemia . At present, this relies on pulsatile delivery of PTH, but a clearer understanding of the underlying mechanisms may lead to less intrusive and more effective therapy. Although PTH is best known for its role in plasma Ca2+ homoeostasis and bone remodelling, consonant with the high levels of expression of PTHR1 in bone and kidney, PTHR1 is also expressed in many diverse tissues, including vascular smooth muscle, hepatocytes, cardiac myocytes and lymphocytes .
Circulating ‘PTH’ includes the full-length peptide, PTH-(1–84), perhaps some short-lived N-terminal fragments [although not PTH-(1–34), see below], and large amounts of long-lived C-terminal fragments generated before and after secretion. Many of the C-terminal fragments are biologically active, often opposing the effects of PTH-(1–84), but their receptors have not been identified . The biological actions of PTH at PTHR1 are replicated by its N-terminal sequence, PTH-(1–34), which is conserved in all vertebrates [1,4] (Figure 1a). PTHR1, like other class II GPCRs, has a long cytosolic N-terminus . The C-terminal α-helical end of PTH-(1–34) binds first to this region (the ‘N-domain’) and thereby, via a flexible linking loop, positions the α-helical N-terminal of PTH-(1–34) to bind more slowly to the juxtamembrane region of the receptor (‘J-domain’) [3,5] (Figure 1b). These interactions with the J-domain lead to receptor activation.
PTH binding to PTHR1
Versatile signalling mechanisms of PTHR1
PTHR1 shares with other class II GPCRs an ability to activate several intracellular signalling pathways, including those involving cAMP and Ca2+. The importance of cAMP in mediating the effects of PTH is evident from patients with pseudohypoparathyroidism, where a maternal mutation of Gs (the only allele expressed in kidney tubules) causes renal PTH-resistance. The immediate consequences of activating PTHR1 differ between cell types, and even between studies of the same cell lines. In some cells, PTH activates AC (adenylate cyclase) without activating PLC (phospholipase C) [6,7], in others, it activates PLC and not AC (see references in ), and in further cells still, notably osteoblasts and kidney tubules, it activates both . Even within the same proximal tubule cell, PTH activates PLC from the apical surface and AC from the basolateral surface [10,11]. Heterologously expressed PTHR1 can also mediate stimulation of both PLC and AC , confirming that a single receptor can elicit both effects. Several factors contribute to whether PTHR1 stimulates AC and/or PLC.
First, expression levels of PTHR1 and of the G-proteins with which it can interact influence the fidelity of G-protein coupling. All levels of expression of PTHR1 allow activation of AC via Gs , while high levels allow additional coupling to PLC via Gq or Gi/Go . Overexpression of Gq can also allow PTHR1 that would otherwise stimulate only AC to activate PLC [6,15,16]. Intense stimulation  or high levels of expression of the signalling proteins can therefore allow PTHR1 to stimulate PLC  (Figure 2).
Signalling to AC and PLC from PTHR1
Secondly, different analogues of PTH cause PTHR1 to adopt different conformations and thereby to interact selectively with different downstream signalling partners, for example, G-proteins and the GRKs (G-protein receptor kinases) that phosphorylate the C-terminal tail of PTHR1 allowing β-arrestins to bind. This phenomenon, where a single receptor responds differently to different ligands, is described as ‘stimulus trafficking’  (Figure 2). It is particularly relevant to PTHR1 because its endogenous ligands, PTH and PTHrP, evoke different physiological responses. Most significant for the treatment of osteoporosis is the observation that both increase bone density, but PTHrP causes less bone resorption and hypercalcaemia than does PTH [3,19].
For PTH, residues within its N- and C-termini were suggested to contribute selectively to stimulation of AC and PLC respectively. Although that now seems too simple an interpretation , different PTH analogues can stimulate AC without activating PLC [e.g. Trp1PTHrP-(1–36) or Gly1,Arg19PTH-(1–28)] [17,21], and point mutations within the first , second  or third  intracellular loop of PTHR1 selectively affect coupling to AC or PLC (Figure 1b). The versatility of PTHR1 signalling is increased further by its coupling to β-arrestin in a receptor conformation described as R0 (to distinguish it from the conformation that binds to G-proteins) . Ligands, via different interactions with the J-domain of PTHR1, differ in the extent to which they stabilize this R0 conformation: PTH, for example, is better than PTHrP [5,19]. The association of PTHR0 with β-arrestin has at least two important consequences: stimulation of the MAPK (mitogen-activated protein kinase) cascade  and trafficking of PTHR1 to early endosomes . β-Arrestin is commonly associated with desensitization of GPCRs (including PTHR1 ), but some PTHR1 ligands that stabilize R0 [e.g. PTH-(1–34)] cause more long-lasting activation of AC than those that do not [e.g. PTHrP-(1–36)] [5,19,25]. Exciting insight into this apparent paradox is provided by recent work showing that AC and Gs are internalized with the PTH–PTHR1–β-arrestin complex. Within early endosomes, functional interactions between PTHR1, Gs and AC then allow very sustained cAMP production within the cell [5,25]. Receptors for other peptide hormones behave similarly . Analogues of PTH differ in their abilities to activate Gs or promote association of PTHR1 with β-arrestin: Trp1PTHrP-(1–36) activates Gs without recruiting β-arrestin (and so stimulates AC only at the plasma membrane and fails to activate the MAPK cascade), whereas D-Trp12,Tyr34-PTH-(7–34) promotes recruitment of β-arrestin but not activation of Gs (and so stimulates MAPK, but not AC) . The latter analogue selectively enhances osteoblast activity without the accompanying recruitment of osteoclasts and ensuing bone resorption . This suggests that it may be an attractive candidate for treatment of osteoporosis with less risk of the hypercalcaemia associated with current daily injections of PTH. These results highlight both the versatility provided by stimulus trafficking and the potential utility of ‘biased ligands’ as a means of diminishing adverse clinical effects.
A third influence on the coupling of PTHR1 to AC and PLC is provided by scaffold proteins, notably the oligomeric NHERF (Na+/H+ exchanger regulatory factor) proteins. Their association with PTHR1 influences the specificity of coupling to G-proteins and β-arrestins  (Figure 2). In addition, NHERFs regulate subcellular targeting of PTHR1, its desensitization and internalization, and inhibition of phosphate transport in proximal tubules by PTH [7,28].
NHERF1 and NHERF2 are cytosolic proteins each with a MERM (merlin/ezrin/radixin/moesin) domain and two PDZ domains, through which they interact with various proteins, including several GPCRs . The MERM domain, via its interactions with the actin cytoskeleton, allows targeting of PTHR1 to the apical membrane of polarized cells  and inhibition of renal phosphate uptake by PTH . Most relevant to the present discussion are the effects of NHERF on the interactions of PTHR1 with G-proteins and β-arrestins (Figure 2). Association with NHERF1 allows PTHR1 to stimulate Gq and thereby PLC without compromising its ability to stimulate Gs and AC [9,31]. This effect probably accounts for the ability of apical PTHR1 associated with NHERF1 in proximal tubules to activate PLC, whereas basolateral PTHR1 activates only AC . NHERF2 promotes coupling of PTHR1 to Gq and Gi, while compromising its interaction with Gs [8,9]. NHERF2 can thereby switch PTHR1 from activating AC to activating PLC . Different patterns of expression of NHERF among cells that express PTHR1 undoubtedly contribute to the diversity of signalling pathways activated by PTH . However, the effects of NHERFs may themselves vary between cells: in rat osteosarcoma cells, for example, NHERF1 allows PTHR1 to associate with the cytoskeleton, but it does not facilitate coupling to PLC . NHERF1 also competes with β-arrestin2 for binding to phosphorylated PTHR1, thereby slowing desensitization . This interaction would presumably also uncouple PTHR1 from the MAPK pathway and prevent sustained cAMP signalling from endosomes. Finally, calpain (a Ca2+-activated protease) also associates with the C-terminal tail of PTHR1  and in response to PTH and an increase in [Ca2+]i (cytosolic free Ca2+ concentration), it cleaves the receptor releasing the NHERF-binding sequence. This uncoupling from NHERF may, at least in part, account for the increased stimulation of AC by calpain-cleaved PTHR1. In summary, tissue-specific associations of PTHR1 with NHERFs, and their acute regulation, are a further and important determinant of the downstream signalling partners recruited by activated PTHR1.
Regulation of Ca2+ signals by cAMP
In most settings, activation of PTHR1 causes activation of Gs and thereby stimulation of AC and formation of cAMP. Additional links between PTH and cytosolic Ca2+ signals are provided by this cAMP. The most common target of cAMP is PKA (protein kinase A), which phosphorylates and so regulates various Ca2+ channels. Within a signalling complex formed by NHERF1, PTHR1 causes activation of PKA and thereby phosphorylation and activation of L-type Ca2+ channels, leading to enhanced Ca2+ entry across the plasma membrane . All three subtypes of mammalian IP3R [Ins(1,4,5)P3 receptor] (IP3R1–IP3R3) are probably phosphorylated by PKA, and at least for IP3R1 and IP3R2, this causes a modest increase in their sensitivity to Ins(1,4,5)P3 [36–38]. Although PKA can potentiate signals evoked by IP3R, our results with heterologously expressed PTHR1 suggest that it does not contribute to the responses evoked by PTH [39–41] (see below). Another widely expressed target of cAMP is Epac (exchange protein directly activated by cAMP) . For other receptors linked to stimulation of AC, Epac1 provides a link with Ca2+ signalling. Binding of cAMP to Epac1 allows it to catalyse GTP binding to a small G-protein, Rap2B, and it then activates PLCϵ leading to formation of Ins(1,4,5)P3 and so release of Ca2+ from intracellular stores . Our results suggest that this pathway is not required for PTH to regulate signalling via IP3R (see below).
Direct regulation of IP3Rs by cAMP at AC–IP3R junctions
In HEK (human embryonic kidney)-293 cells stably expressing modest levels of human PTHR1 (~105 receptors/cell), PTH stimulates AC, but not PLC, and neither does it directly increase [Ca2+]i . However, PTH massively potentiates the Ca2+ signals evoked by either receptors that stimulate PLC, such as the endogenous M3 muscarinic receptors of HEK-293 cells, or by flash-photolysis of caged Ins(1,4,5)P3. In intact cells, these effects of PTH are entirely mimicked by high concentrations of the membrane-permeant analogue of cAMP, 8-Br-cAMP (8-bromo-cAMP); and in permeabilized cells, they are mimicked by cAMP [38,41]. Indeed, the activity of all three subtypes of mammalian IP3Rs is potentiated by cAMP . Considerable evidence suggests that the ability of PTH to potentiate Ins(1,4,5)P3-evoked Ca2+ release is entirely mediated by cAMP . However, the effect does not require activation of PKA and neither is it mimicked by activation of PKA [38,41]. Furthermore, the effects of 8-Br-cAMP in intact cells are not mimicked by a similarly membrane-permeant analogue of cAMP that selectively activates Epac . These results indicate that high concentrations of cAMP, at least 100-fold higher than required to activate PKA, can sensitize IP3R to Ins(1,4,5)P3 without involvement of the classical targets of cAMP, PKA and Epac. We suggest that a novel low-affinity cAMP-binding site on either the IP3R itself or a tightly associated protein allows cAMP to directly increase the efficacy of Ins(1,4,5)P3.
Three initially perplexing observations suggest that the delivery of cAMP to IP3R in intact cells requires a very specific association of IP3R with AC. First, despite evidence that cAMP entirely mediates the effects of PTH, measurements of total cAMP levels in cells suggest that global concentrations of cAMP are unlikely to be sufficient to sensitize IP3R . Secondly, all three IP3R subtypes are sensitized by Ins(1,4,5)P3, yet our use of siRNA (small interfering RNA) selectively to reduce expression of IP3R1 and IP3R2 (the major subtypes in HEK-293 cells) suggests that potentiation of Ins(1,4,5)P3-evoked Ca2+ release by PTH is selectively attenuated by loss of IP3R2. Furthermore, although AC6 accounts for only approximately 5% of the AC expressed in HEK-293 cells, it appears also to be selectively required for PTH to potentiate Ca2+ signals . These selective requirements for AC6 and IP3R2 are supported by immunoprecipitation analyses which reveal a specific association of the two proteins . Thirdly, there is no clear relationship within cells between cAMP formation and potentiation of Ca2+ signals. Different stimuli (PTH, forskolin, or isoprenaline to activate endogenous β-adrenoceptors) evoke comparable potentiation of Ca2+ signals, but with hugely different effects on cAMP. Even substantial inhibition of AC (>80%) has no effect on the ability of any concentration of PTH, maximal or submaximal, to potentiate Ins(1,4,5)P3-evoked Ca2+ signals. These and related observations  led us to propose that an intimate association between AC6 and IP3R2, an ‘AC–IP3R junction’, allowed cAMP to be delivered directly and at very high local concentration to an associated IP3R, maximally sensitizing it to Ins(1,4,5)P3 (Figure 3). With this scheme, each junction operates as a binary on–off switch, and with a considerable safety margin because the AC delivers more cAMP to each IP3R than is required to cause its maximal sensitization. The concentration-dependent effects of PTH, we suggest, then arise not from graded changes in the activity within each junction (they are either on or off), but from graded recruitment of individual junctions: the higher concentrations of PTH allow more junctions to be activated. These results reveal yet another link between PTH and [Ca2+]i that requires co-stimulation of a receptor linked to activation of PLC. PTH, by stimulating delivery of cAMP within the AC6–IP3R2 junction, then potentiates the responses to Ins(1,4,5)P3.
Signalling from PTH to Ca2+ signals via cAMP
PTHR1, as the receptor that mediates the most important biological actions of PTH and as a proven target for treatment of osteoporosis, commands attention, but it also provides a striking illustration of the versatility of the signalling mechanisms emanating from a single GPCR. Stimulus intensity, stimulus trafficking and association with scaffold proteins all contribute to defining the interactions of PTHR1 with intracellular signalling proteins (G-proteins or β-arrestins) (Figure 2). Furthermore, the cAMP that is usually produced when PTHR1 is activated contributes to generation of intracellular Ca2+ signals via each of the conventional targets of cAMP, Epac (stimulation of PLCϵ) and PKA (phosphorylation of Ca2+ channels), and by interaction with a novel site on the IP3R (Figure 3).
Signalling 2011: a Biochemical Society Centenary Celebration: A Biochemical Society Focused Meeting held at the University of Edinburgh, U.K., 8–10 June 2011. Organized and Edited by Nicholas Brindle (Leicester, U.K.), Simon Cook (The Babraham Institute, U.K.), Jeff McIlhinney (Oxford, U.K.), Simon Morley (University of Sussex, U.K.), Sandip Patel (University College London, U.K.), Susan Pyne (University of Strathclyde, U.K.), Colin Taylor (Cambridge, U.K.), Alan Wallace (AstraZeneca, U.K.) and Stephen Yarwood (Glasgow, U.K.).
cytosolic free Ca2+ concentration
exchange protein directly activated by cAMP
G-protein receptor kinase
human embryonic kidney
mitogen-activated protein kinase
Na+/H+ exchanger regulatory factor
protein kinase A
type 1 PTH receptor
Our work is supported by the Wellcome Trust [grant number 085295] and the Biotechnology and Biological Sciences Research Council [grant number BB/H009736].