Abstract
Arginine vasopressin (AVP) stimulates adrenocorticotropin (ACTH) secretion from corticotroph cells of the anterior pituitary via activation of the V1b vasopressin receptor, a member of the G protein-coupled receptor (GPCR) family. Recently, we have shown that treatment of ovine anterior pituitary cells with AVP for short periods results in reduced responsiveness to subsequent stimulation with AVP. The aim of this study was to investigate mechanisms involved in this desensitization process. Among the GPCR family, rapid desensitization is commonly mediated by receptor phosphorylation, with resensitization being mediated by internalization and subsequent dephosphorylation of the receptors by protein phosphatases. Since desensitization of V1a vasopressin receptors is mediated by protein kinase C-mediated receptor phosphorylation, we investigated the involvement of this enzyme in desensitization of the ACTH response to AVP. Treatment of perifused ovine anterior pituitary cells with the specific protein kinase C (PKC) activator 1,2-dioctanoyl-sn-glycerol (300 μM) did not induce any reduction in response to a subsequent 5-min stimulation with 100 nM AVP, despite potently stimulating ACTH secretion. Likewise, the results obtained using the PKC inhibitor Ro 31-8220 were not consistent with involvement of PKC in AVP desensitization: 2 μM Ro 31-8220 did not reduce the ability of a 10 nM AVP pretreatment to induce desensitization to a subsequent stimulation with 100 nM AVP. Pharmacologic blockade of receptor internalization by treatment with 0.25 mg/ml concanavalin A significantly impaired the ability of a 15-min pretreatment with 10 nM AVP to induce desensitization, rather than affecting resensitization. Treatment with 10 nM okadaic acid, an inhibitor of protein phosphatase 1 and 2A, had no effect on either resensitization or desensitization. In contrast, inhibition of protein phosphatase 2B (PP2B) with 1 μM FK506 decreased the rate of resensitization: complete recovery from desensitization took 40 min, whereas in controls recovery was complete 20 min after termination of the pretreatment. These results indicate that desensitization of the ACTH response to AVP is not mediated by PKC-catalyzed phosphorylation, suggesting subtype-specific differences in the regulation of V1a and V1b vasopressin receptors. The data demonstrate that desensitization was dependent, at least in part, upon receptor internalization and that resensitization was dependent upon PP2B-mediated receptor dephosphorylation.
Introduction
Arginine vasopressin (AVP) is an important physiologic regulator of adrenocorticotropin (ACTH) secretion from the anterior pituitary (Aguilera 1994). The effects of AVP on corticotroph cells of the anterior pituitary are mediated by activation of the V1b (also known as V3) vasopressin receptor, a phospholipase C (PLC)-coupled member of the G protein-coupled receptor (GPCR) superfamily (De Keyzer et al. 1994). Recently, we have shown that treatment of perifused ovine anterior pituitary cells with AVP can cause desensitization of the ACTH response to a subsequent stimulation with AVP (Hassan et al. 2003). This desensitization occurred at concentrations and durations of AVP treatment that were within the ranges of AVP pulses measured in the sheep portal circulation, suggesting that the desensitization process might play an important physiologic role in the regulation of ACTH secretion. Desensitization was found to be rapid (complete within 10 min of the onset of AVP treatment) and was readily reversible (resensitization was complete 40 min after the end of AVP treatment). The aim of the study reported here was to investigate molecular mechanisms involved in both desensitization and resensitization of the ACTH response to AVP.
The characteristics of the desensitization and resensitization processes give some clues as to mechanisms that might underlie them. Among the GPCR superfamily, rapid and readily reversible desensitization is commonly a result of uncoupling of the receptor from its signaling pathway (Lohse 1993). This uncoupling is typically mediated by phosphorylation of the receptor by one or more protein kinases that, directly or indirectly, prevents the receptor from activating its signaling pathway. The effector kinases protein kinase A (PKA) and protein kinase C (PKC), the G protein-coupled receptor kinase (GRK) family and casein kinase 1α (CK1α) are all capable of phosphorylating GPCRs (Lohse 1993, Tobin et al. 1997).
A model has been proposed in which resensitization of desensitized, phosphorylated receptors is mediated by their internalization to an intracellular compartment (most likely the early endosome). The receptors are subsequently recycled back to the plasma membrane in a fully functional state. In this model, receptor phosphorylation is sufficient for desensitization, and internalization of the receptor is not required for reduced responsiveness (Ferguson et al. 1996). Experimental results consistent with this model have been reported for a variety of receptors, including the β2 adrenergic receptor (Pippig et al. 1995) and the NK1 neurokinin receptor (Garland et al. 1996). Two different types of protein phosphatase, PP2A and PP2B, have been implicated in the dephosphorylation of GPCRs (Pippig et al. 1995, Shih et al. 1999). Recent evidence suggests an alternative role for internalization for some members of the GPCR family, including the sst2B somatostatin receptor (Beaumont et al. 1998) and the secretin receptor (Mundell & Kelly 1998). For these receptors, pharmacologic blockade of internalization prevents desensitization from occurring, suggesting that they must be sequestered from the cell membrane in order to be successfully uncoupled from their signaling pathways.
Vasopressin receptors found in the liver, brain and vasculature are of the V1a subtype (Birnbaumer 2000). Although these receptors are structurally distinct from V1b receptors (~45.5% amino-acid sequence homology in the human (De Keyzer et al. 1994)), they share similar signaling properties: they have almost identical affinity for AVP (Jard et al. 1986), both are coupled to PLC (Birnbaumer 2000), and both undergo desensitization after exposure to AVP. In the case of the V1a receptor, desensitization appears to be mediated, at least in part, by PKC-mediated phosphorylation of the receptor (Caramelo et al. 1991, Gallo-Payet et al. 1991, Ancellin et al. 1997, Ancellin & Morel 1998). Given the pharmacologic similarities between the two receptor subtypes, the molecular mechanisms involved in their desensitization might be expected to be similar. Alternatively, in view of their distinct tissue distributions and the different concentrations and patterns of AVP they are exposed to, V1a and V1b receptors might be predicted to be regulated by different mechanisms of desensitization.
A second group of kinases that might be involved in desensitization of V1b receptors are the GRKs. However, while GRKs have been found to be involved in the regulation of some PLC-coupled GPCRs, changes in the intracellular environment following activation of PLC appear somewhat unfavourable to GRK activity (Tobin et al. 1997). Thus, it has been suggested that in some cases alternative mechanisms might be involved in the regulation of PLC-coupled GPCRs. Recently, CK1α was shown to phosphorylate the PLC-coupled m1 and m3 muscarinic receptors (Tobin et al. 1997, Waugh et al. 1999). This phosphorylation was agonist-dependent and was associated with a rapid desensitization of the cellular responses to stimulation.
Little information exists on the mechanisms involved in resensitization of the pituitary V1b receptor. While evidence indicates that the pituitary AVP receptor is internalized after AVP stimulation (Mogensen et al. 1988, Berrada et al. 2000), it is unclear what role this process might play in regulating the ACTH response to AVP. The role that protein phosphatases might play in resensitization is also unknown.
In this study we focused on investigating the mechanisms of desensitization and resensitization of the anterior pituitary ACTH response to AVP. Specifically, we used pharmacologic agents to investigate the roles of PKC, CK1α, PP2A, PP2B and receptor internalization in these two processes. The results obtained were compared with those reported for desensitization of the V1a receptor.
Materials and Methods
Cell preparation
Pituitary glands were collected from a local abattoir in accordance with animal ethics laws of New Zealand and the guidelines of the University of Canterbury. Dispersed ovine anterior pituitary cells were prepared as previously described (Evans et al. 1985, Le Beau & Mason 1994). Briefly, whole pituitary glands were removed from sheep shortly after slaughter and placed into cold, sterile HEPES buffer (25 mM HEPES, 137 mM NaCl, 5 mM KCl, 10 mM glucose and 0.002% phenol red, pH 7.3) containing antibiotic and antimycotic agents (100 U penicillin/ml, 100 mg streptomycin/ml and 0.25 mg amphotericin B/ml). A suspension of dispersed cells from the combined anterior lobes of seven pituitaries was obtained following a combination of enzymatic (collagenase type II, 480 U/ml HEPES buffer, Sigma) and mechanical disruption. The dispersed cells were washed by centrifugation (300 g for 5 min at 4°C) and resuspension in HEPES buffer; after the final wash, they were resuspended in culture medium (Dulbecco’s modified Eagle’s medium (DMEM); Sigma) containing 3.7 mg NaHCO3/ml, 584 mg l-glutamine/ml, nonessential amino acids, antibiotic/antimycotic agents (as above) and 25 mM HEPES at pH 7.3, and supplemented with 10% newborn calf serum (NCS) (Gibco Laboratories, Life Technologies, Grand Island, NY, USA). For perifusion experiments, 4.5 × 106 cells were plated into nonadherent plastic Petrie dishes (Labserv, Auckland, New Zealand) in 20 ml DMEM plus NCS.
Multicolumn perifusion experiments
The multicolumn perifusion system, as described previously and with modifications (Evans et al. 1985, 1996), allows for the simultaneous perifusion of up to 15 cell chambers or columns. Through the use of a solenoid switching system, various trains of peptide hormones and pharmacologic agents can be applied to the cells.
After overnight incubation, the cells from each culture plate were recovered by centrifugation, and the supernatant was replaced with 170 μl slurry of preswollen Sephadex G-25 fine (Sigma) suspended at a ratio of 1:1.5 v/v in Krebs-Ringer (KR) solution (125 mM NaCl, 4.7 mM KCl, 1.2 mM KH2PO4, 2.5 mM CaCl2, 1.2 mM MgSO4, 3.6 mM NaHCO3, 25 mM HEPES, 10 mM glucose and antibiotic/antimycotic agents as above). Slurries of Bio-Gel P2 (BioRad Laboratories, Hercules, CA, USA) (80 μl suspended 1:1.5 v:v in KR solution) and Sephadex G-25 fine (80 μl suspended 1:1.5 v:v in KR solution) were added sequentially to the columns to form a bed on which the mixture of cells and Sephadex was placed. Perifusion was commenced at a flow rate of 0.16 ml/min, with the perifusion medium (KR solution containing 0.05% alkali-treated casein and 0.005% ascorbate (Sigma); KR/ATC) and the cells being maintained at 37°C by a water jacket surrounding the columns and the media reservoirs. Effluent from the columns was collected in 5- and 10-min fractions and frozen for subsequent measurement of ACTH concentration by radioimmunoassay.
During an experiment, cells were perifused with a ‘basal’ solution (that is, KR/ATC) and a ‘test’ solution containing AVP. The tubing carrying these solutions was connected to the pump tubing by a Y-junction. At any given time, one of these lines was clamped closed. Through the use of a solenoid-actuated switching system, the perifusion solution could be rapidly and precisely changed between the ‘test’ and ‘basal’ solutions. When the cells were perifused with ‘basal’ KR/ATC plus two separate ‘test’ solutions, the basal line was manually transferred between KR/ATC and the second ‘test’ solution.
A variety of different experimental protocols were used in perifusion experiments. While the protocol used depended on the aim of the particular experiment, the design of all experiments shared some common features. These are described below while the details of the specific treatment regimes used in individual experiments can be found in the Results section. Each of the 15 columns in the perifusion system was randomly assigned an experimental treatment. Typically, each of the different treatments was repeated in at least two columns. In all experiments, cells were perifused with ‘basal’ KR/ATC for at least 90 min at the beginning of the experiment, allowing them to recover from transfer into the columns. During this period, ACTH secretion dropped to a consistently low level. Experimental treatments began at the end of this run-in period.
Radioimmunoassay
ACTH radioimmunoassay was performed as previously described (Evans et al. 1985). The antiserum used (rabbit antiporcine ACTH) was a gift from Dr Richard Donald, Christchurch Hospital (Christchurch, New Zealand), and the highly purified ovine ACTH, used for 125I-radioiodination and assay standards, was a gift from Dr C H Li, Hormone Research Laboratory, University of California (San Francisco, CA, USA). Intra- and interassay coefficients of variation were 6.77% and 8.25%.
Data and statistical analysis
In perifusion experiments, the ability of a particular treatment to induce desensitization was assessed by measuring its effect on the response to a subsequent 5-min stimulation with 100 nM AVP. The magnitude of the response to each of these ‘test’ pulses was calculated as the total amount of ACTH released in the 20 min following commencement of the pulse, minus basal secretion (calculated using the mean value of the hormone concentration in the two or three fractions immediately preceding either the pulse or the pretreatment). Therefore, the term ‘response’ refers to stimulated ACTH secretion. For the purposes of analysis, the results from each perifusion column were treated as n=1. In general, for each treatment, there were two to three replicates (columns) per experiment (one-cell preparation), and there were two to three experiments. Data are reported as mean±s.e.m. Statistical significance was assessed by either Student’s t-test or ANOVA, as indicated in the text. P<0.05 was considered significant.
Results
Effect of pretreatment with 1,2-dioctanoyl-sn-glycerol (DiC8) on the ACTH response to AVP
To investigate whether PKC is involved in desensitization of the ACTH response to AVP, perifused cells were pretreated with the specific PKC activator DiC8 to determine whether it induced desensitization to a subsequent stimulation with AVP. If PKC was involved in the desensitization process, pretreatment with DiC8 would be expected to reduce the response to a subsequent stimulation with AVP. The experimental design was similar to that which we have previously used to investigate desensitization of the ACTH response to AVP (Hassan et al. 2003): perifused cells were stimulated with 5-min pulses of AVP after 100, 180 and 260 min of perifusion, resulting in three peaks of ACTH secretion (Fig. 1a). Since the mean of the responses to the first and third pulses is no different to the response to the second pulse (n=9; NS, Student’s t-test), it is assigned as control (100%). To induce desensitization, the second pulse was immediately preceded by a 15-min pretreatment with either 10 nM AVP (Fig. 1b) or DiC8 at concentrations of either 30, 100 or 300 μM (Fig. 1c). The effect of pretreatment on the ACTH response to AVP was assessed quantitatively for each perifusion column by expressing the response to the pretreated (second) pulse as a percentage of the mean of the responses to the first and third pulses (that is, control). Secretion of ACTH during pretreatment with DiC8 was concentration dependent (Fig. 1d). Treatment with 30 μM DiC8 caused a small but significant increase in ACTH secretion above basal (n=6, P<0.01, t-test). When the concentration was increased to 300 μM, stimulated ACTH secretion was not significantly different from that observed during pretreatment with 10 nM AVP (one-way ANOVA with Dunnett’s test). Despite stimulating ACTH secretion at a level equivalent to AVP, pretreatment with DiC8 did not induce desensitization. In contrast to pretreatment with 10 nM AVP (which caused a reduction in response of 48.6±1.6% (n=6, P<0.0001, t-test)), the ACTH response to AVP was increased in a concentration-dependent manner after pretreatment with DiC8 (Fig. 1e). Pretreatment with 100 μM and 300 μM DiC8 caused 9.9±3.7% (n=6, P<0.05, t-test) and 51.3±4.4% (n=6, P<0.0001, t-test) increases in response respectively.
Effect of pretreatment with the specific PKC inhibitor Ro 31-8220 on desensitization of the ACTH response to AVP
The involvement of PKC in desensitization of the ACTH response to AVP was investigated further with the potent and selective PKC inhibitor Ro 31-8220 (or, more fully, 2-{1-[3-(amidino-thio)propyl]-1H-indol-3-y}-3-(1-methylindol-3-yl)-maleimide methanesulfonate) (Davis et al. 1992). If PKC were involved in the desensitization process, treatment with Ro 31-8220 would be expected to impair the ability of AVP pretreatment to induce desensitization. In preliminary studies, we found that the effects of Ro 31-8220 were not readily reversible. Thus, it was not possible to analyze desensitization as described above, using the first and third pulses as controls in a three-pulse protocol. To induce desensitization, a 5-min 100 nM AVP pulse at 180 min was preceded by a 15-min pretreatment with 10 nM AVP. This AVP pretreatment caused a significant reduction (69.8±2.4%; n=3, P<0.05, t-test) in response compared with the response to a similar 100 nM AVP pulse in control columns (that is, not pretreated) (Fig. 2a and b). To assess the effect of PKC inhibition on desensitization of the ACTH response to AVP, cells were treated with 2 μM Ro 31-8220 from 90 min of perifusion until the completion of the experiment. During treatment with Ro 31-8220, the ACTH response to the 100 nM AVP pulse was reduced by 45.1±3.5% (n=7, P<0.0001, t-test) compared with controls that were not treated with Ro 31-8220 (Fig. 2; compare panels c and a). However, despite being able to inhibit ACTH secretion, treatment with Ro 31-8220 did not reduce the extent of desensitization observed. After pretreatment with 10 nM AVP in combination with Ro 31-8220, the response to a subsequent stimulation with 100 nM AVP was reduced by 88.8±4.5% (n=4, P<0.01, t-test) compared with the corresponding control (that is, Ro 31-8220-treated but not 10 nM AVP-pretreated) (Fig. 2; compare panels d and c). Thus, rather than reducing the extent of AVP-induced desensitization, Ro 31-8220 significantly increased the desensitization caused by AVP pretreatment (n=7, P<0.05, t-test). Results are summarized in Fig. 2e.
Effect of pretreatment with CK1–7 on desensitization of the ACTH response to AVP
The role of CK1α in the regulation of AVP-stimulated ACTH secretion was investigated with the potent and selective casein kinase 1 inhibitor CK1–7 (N-[2-aminoethyl]-5-chloroisoquinoline-8-sulfonamide) (Chijiwa et al. 1989). The involvement of CK1α in desensitization of the ACTH response to AVP was investigated by testing the effect of treatment with CK1–7 on the ability of AVP pretreatment to induce desensitization. If CK1α were involved in desensitization, the presence of CK1–7 would be expected to reduce the extent of desensitization induced by AVP pretreatment. Pretreatment with 10 nM AVP for 25 min before the second 100 nM AVP pulse in a three-pulse protocol caused a reduction in response of 49.0±3.1% (n=9, P<0.001, t-test) (Fig. 3a) compared with controls. To investigate the involvement of CK1α in desensitization, the second 100 nM AVP pulse was immediately preceded by a 30-min treatment with CK1–7 at concentrations of 10, 30 or 100 μM (that is, for 5 min before the pretreatment and then concurrently with it for 25 min) (Fig. 3b–d). Treatment with CK1–7 at all three concentrations had no effect on AVP-induced desensitization, with the reduction in response induced by AVP pretreatment not being affected by any of these concentrations of CK1–7 (Dunnett’s test). After 30-min pretreatment with CK1–7 (30 μM) alone, the response to the second AVP pulse was not significantly altered compared with the controls (n=6, NS, t-test) (Fig. 3e). Results are summarized in Fig. 3f.
Treatment with CK1–7 both alone and in combination with AVP had no effect on ACTH secretion. During pretreatment with 30 μM CK1–7 alone, ACTH secretion was not significantly increased (0.22±0.17 ng ACTH above basal during the entire 30-min CK1–7 treatment; n=3, NS, t-test). Similarly, treatment with 10 nM AVP in combination with CK1–7 had no effect on stimulated ACTH secretion during the 25-min 10 nM AVP pretreatment (10 nM AVP alone: 2.46±0.07 ng ACTH; 10 nM AVP+100 μM CK1–7: 2.45±0.09 ng ACTH; n=6, NS, t-test).
Effect of blockade of receptor internalization on the regulation of the ACTH response to AVP
The involvement of receptor internalization in the regulation of the ACTH response to AVP was investigated with the lectin concanavalin A (ConA) to block receptor internalization (Pippig et al. 1995, Trincavelli et al. 2000). The experimental protocol employed was based on that which we have previously used to investigate resensitization of the ACTH response to AVP (Hassan et al. 2003). Perifused cells were stimulated with a single, 5-min pulse of 100 nM AVP after 160 min of perifusion, resulting in a characteristic peak of ACTH secretion. To induce desensitization, this pulse was immediately preceded by 15-min pretreatment with 10 nM AVP. To observe resensitization, a recovery period of either 20 or 40 min was allowed between the end of the pretreatment and the start of the pulse. The extent of desensitization was assessed quantitatively by expressing the response to the 100 nM AVP pulse after pretreatment as a percentage of the response of control cells (which were not pretreated) to 100 nM AVP pulses.
When cells were not treated with ConA, pretreatment with 10 nM AVP for 15 min immediately prior to the 100 nM AVP pulse caused a reduction in response of 58.4±4.4% (P<0.001, one-way ANOVA with Bonferroni’s test) compared with control (Fig. 4a and b). When a recovery period was allowed between the pulse and the pretreatment, resensitization occurred (Fig. 4c and d). After a 20-min recovery period, no significant desensitization remained (NS, Bonferroni’s test); that is, resensitization was complete. When cells were treated with ConA (0.25 mg/ml) for 70 min prior to the 100 nM AVP pulse (that is, from 90 to 160 min), AVP pretreatment was still able to induce desensitization, but it was significantly reduced compared with the desensitization observed in the absence of the lectin. When the AVP pretreatment immediately preceded the 100 nM AVP pulse, the response was reduced by 22.7±3.7% (P<0.05, Bonferroni’s test) compared with the response of cells that were not pretreated with AVP, but were treated with ConA (Fig. 4e and f). This effect of ConA on the ability of AVP pretreatment to induce desensitization was statistically significant (P<0.001, Bonferroni’s test). After recovery periods of 20 and 40 min, there was no significant difference between the extent of desensitization observed in control and ConA-treated cells (Fig. 4g and h). Results are summarized in Fig. 4i. Treatment with ConA had no significant direct effect on either basal or AVP-stimulated ACTH secretion.
Effect of inhibition of protein phosphatases on resensitization of the ACTH response to AVP
The role of PP2A and PP2B in resensitization of the ACTH response to AVP was assessed by pharmacologic inhibition of these two enzymes. A design similar to that used to investigate the role of internalization was employed, except that the cells were treated with the appropriate phosphatase inhibitor for 70 min prior to the 100 nM AVP pulse rather than with ConA.
FK506 is a macrocyclic lactone derived from Streptomyces sp. that specifically inhibits PP2B (Schreiber 1991). Treatment with 1 μM FK506 did not alter the magnitude of desensitization induced by pretreatment with 10 nM AVP for 15 min (54.6±6.4% in the presence of FK506 compared with 53.7±3.5% in controls not treated with the inhibitor (NS, Bonferroni’s test)). However, treatment with FK506 decreased the rate of resensitization. In the absence of FK506, recovery from desensitization was complete after 20 min (9.2±5.1% less than control; NS, Bonferroni’s test). In contrast, when cells were treated with 1 μM FK506, a significant desensitization still remained at 20 min (38.1±3.3% less than control; P<0.001, Bonferroni’s test). Resensitization was not complete until 40 min. Representative data from individual columns in which cells were treated with 1 μM FK506 are shown in Fig. 5e–h, and summarized results are shown in Fig. 5i.
In addition to this effect on the rate of resensitization, treatment with FK506 also caused a small but reproducible increase in the ACTH response to AVP pretreatment. Treatment with FK506 at concentrations of either 50 nM or 1 μM increased the ACTH response to the 15-min 10 nM AVP pretreatment (Fig. 6). Treatment with FK506 had no significant effect on either the ACTH response to the 100 nM AVP pulse or basal ACTH secretion.
Okadaic acid (OA) potently inhibits the activity of both protein phosphatase 1 (PP1) and PP2A (Herzig & Neumann 2000). PP2A is completely inhibited by 1 nM OA whereas PP1 is unaffected by this concentration, its 50% inhibitory concentration (IC50) being 10–15 nM (Cohen et al. 1989). PP2B is relatively unaffected by OA (IC50 =3600 nM) (Bialojan & Takai 1988). Treatment with 10 nM OA for 70 min prior to the 100 nM AVP pulse had no effect on the pattern of desensitization and resensitization observed after pretreatment with 10 nM AVP for 15 min. Results are summarized in Fig. 7. There was no evidence of an effect of treatment with 10 nM OA on either basal or AVP-stimulated ACTH secretion.
Discussion
This study has shown that the mechanisms underlying desensitization of the V1b vasopressin receptor differ from those of the V1a receptor. These differences are likely to be physiologically important in the regulation of responsiveness to AVP in different tissues.
No evidence was found for the involvement of either CK1α or PKC in desensitization of the ACTH response to AVP. Despite being able potently to activate PKC, treatment with DiC8 was not capable of inducing desensitization. In fact, it markedly increased the response to subsequent AVP stimulation. Likewise, the results obtained with the PKC inhibitor Ro 31-8220 were not consistent with involvement of PKC in AVP desensitization. Despite being able to reduce ACTH secretion significantly, treatment with Ro 31-8220 did not reduce the extent of desensitization induced by pretreatment with 10 nM AVP. Taken together, these results suggest that, while PKC plays an important role in the regulation of ACTH secretion, it is not involved in desensitization.
The lack of involvement of PKC in desensitization of the ACTH response to AVP suggests AVP receptor subtype-specific differences in mechanisms of desensitization of vasopressin receptors: the V1a vasopressin receptor, which is expressed in the liver, brain and vasculature (Birnbaumer 2000), is desensitized via PKC-mediated receptor phosphorylation (Caramelo et al. 1991, Gallo-Payet et al. 1991, Ancellin et al. 1997, Ancellin & Morel 1998). Differences in the mechanisms of desensitization might play an important role in the regulation of responsiveness to AVP in different tissues. The patterns and maximal concentrations of AVP secretion into the pituitary portal circulation are quite different from those in the peripheral circulation (Caraty et al. 1990). AVP is released into the portal blood in a highly pulsatile manner and reaches relatively high concentrations, that is, up to ~6 nM during insulin-induced hypoglycaemia in sheep (Caraty et al. 1990). In contrast, changes in AVP concentration in the peripheral circulation are slow and of low magnitude; for example, after hypertonic saline infusion into ewes, there is a steady increase in plasma AVP concentration from ~2 to ~7 pM over a period of 90 min (Keller-Wood 1994). In the β-adrenergic receptor system, GRK-mediated desensitization is very rapid and requires relatively high agonist concentrations, suggesting that this process might be particularly important in environments such as the synaptic cleft, where agonist concentrations increase rapidly to high concentrations (Roth et al. 1991). Correspondingly, the PKA-mediated desensitization of the β-adrenergic receptor is slower and occurs at lower agonist concentrations, suggesting that it is better suited to the regulation of responsiveness at nonsynaptic receptors (Roth et al. 1991). Similarly, differential desensitization of the V1a and V1b receptors may allow for appropriate regulation of the responsiveness to AVP in different tissues where different concentrations and patterns of AVP exposure occur. Thus, differential desensitization of the V1a and V1b receptors provides a plausible explanation for the existence of two receptors with otherwise similar signaling properties.
The reduction in desensitization caused by inhibition of receptor endocytosis indicates that desensitization of the ACTH response to AVP is mediated, at least in part, by receptor internalization. It is of interest that, at the concentration used, ConA only partially inhibited the ability of AVP pretreatment to induce desensitization. This concentration (0.25 mg/ml) has been shown, in some instances, to inhibit completely internalization of a variety of GPCRs (Pippig et al. 1995), while in others it only partially inhibited internalization (Trincavelli et al. 2000). If receptor internalization was completely inhibited by ConA in the perifused anterior pituitary cell system used in these experiments, the remaining desensitization must have been the result of another desensitization process or processes. Alternatively, the ConA treatment used may have only partially inhibited receptor internalization and as a result only partially inhibited the development of desensitization.
While neither PKC nor CK1α appears to play a role in desensitization to AVP, the effects of selective phosphatase inhibition on resensitization strongly suggest that protein phosphorylation and dephosphorylation are involved in desensitization and resensitization of the ACTH response to AVP. While no evidence could be found for the involvement of PP2A in resensitization, the attenuation in the rate of resensitization seen during FK506 treatment indicates a role for PP2B in recovery of the ACTH response to AVP. The role of PP2B in resensitization has been most extensively investigated for the β2-adrenergic receptor, where the enzyme appears to mediate resensitization by reversing GRK2-mediated phosphorylation of the receptor (Shih et al. 1999). It appears likely that PP2B plays a similar role in the regulation of the ACTH response to AVP, mediating the dephosphorylation of V1b receptors phosphorylated during desensitization. However, because a wide variety of membrane and cytosolic proteins are phosphorylated during AVP stimulation (Liu 1994, Liu et al. 1994), the possibility that dephosphorylation of proteins other than the AVP receptor is involved in resensitization cannot be excluded.
If desensitization is mediated by V1b receptor phosphorylation, the question remains as to which intracellular protein kinase is involved in this process. Perhaps the best remaining candidates are members of the GRK family. Neill et al.(1998) have shown that GRK2, GRK3, GRK6 and β-arrestins are expressed in rat anterior pituitary cells, demonstrating that, at least in this species, pituitary cells have the intracellular machinery necessary for regulation of the V1b receptor by GRKs. Furthermore, recent evidence has shown that the V1b receptor can interact with GRKs (Berrada et al. 2000). The investigation of the involvement of GRKs in the desensitization of the ACTH response to AVP is somewhat diffcult in primary cell cultures. At present, no highly specific inhibitors of GRK activity exist, and other, more specific techniques, such as expression of dominant negative mutant GRKs (Pitcher et al. 1998), or reducing GRK expression with antisense oligo-nucleotide sequences (Shih & Malbon 1994), are either extremely diffcult or impossible to use in primary cell cultures. Although in this respect use of primary cultures is disadvantageous, it has one distinct advantage. Recent evidence has shown that the level of expression of GPCRs can affect their regulation (Spurney 1998) and that regulation of GPCRs can vary depending upon the nature of the cellular environment in which they are expressed (Ferguson 2001). In the anterior pituitary cell system used in this research, the expression of all components of the signaling pathway was under physiologic regulation. As such, the mechanisms which were found to be involved in desensitization of the ACTH response to AVP are very likely to be physiologically relevant and not artifacts of the system being investigated.
If the role of PP2B in resensitization of the ACTH response to AVP is to reverse phosphorylation of the V1b receptor (by an as yet undetermined protein kinase), the question of what role this phosphorylation might play in desensitization is raised. Receptor phosphorylation is usually, directly or indirectly, sufficient to induce desensitization; that is, internalization of the receptor is not required for its desensitization (Lohse et al. 1990). However, we have shown that desensitization of the ACTH response to AVP can, to a large extent, be inhibited by pharmacologic blockade of receptor internalization, indicating that receptor sequestration is important in this desensitization process. One possibility is that there are dual mechanisms of desensitization, with receptor phosphorylation mediating the proportion of desensitization that could not be inhibited by blockade of internalization with ConA. Another possible role for receptor phosphorylation in the regulation of the V1b receptor is as a signal that acts to promote receptor sequestration. These two hypotheses for a role of receptor phosphorylation in desensitization of the pituitary AVP receptor are not necessarily mutually exclusive. Receptor phosphorylation and subsequent binding of β-arrestins could impair, but not completely prevent, coupling of the V1b receptor to Gq/11, with subsequent phosphorylation-dependent internalization of the receptor resulting in full desensitization.
In summary, we have shown that desensitization of the ACTH response to AVP depends to a large extent, if not entirely, upon receptor internalization. As such, the pituitary V1b AVP receptor joins the small group of GPCRs for which receptor internalization is important in desensitization. Additionally, the activity of PP2B plays an important role in resensitization, most likely by dephosphorylating pituitary AVP receptors that had been phosphorylated during the desensitization process. Desensitization was not the result of PKC-mediated phosphorylation, indicating that the mechanisms involved in desensitization of the pituitary V1b receptor are different from those involved in desensitization of the V1a receptor.
Treatment with the PKC activator DiC8 stimulated ACTH secretion but did not induce desensitization. Panels a–c show representative data from perifusion columns in which cells were stimulated with 5-min pulses of 100 nM AVP after 100, 180 and 260 min of perifusion (100 nM AVP pulses are indicated by arrows). Results from a control column (that is, no pretreatment) are shown in panel a. Results from columns in which the second 100 nM AVP pulse was immediately preceded by a 15-min pretreatment with 10 nM AVP (indicated by gray bar) or 300 μM DiC8 (indicated by black bar) are shown in panels b and c respectively. DiC8-stimulated ACTH secretion during the pretreatment period was concentration dependent, with 300 μM DiC8 stimulating ACTH secretion at a level equivalent to that stimulated by 10 nM AVP (d; data are mean±s.e.m., n=6 for each treatment). Despite potently stimulating ACTH secretion, pretreatment with DiC8 did not induce desensitization to a subsequent AVP stimulation (e). Data shown are the responses, expressed as a percentage of control, to a 5-min, 100 nM AVP pulse after pretreatment for 15 min with the indicated solutions. Data are mean±s.e.m. (n=5–6 for each treatment). Asterisks indicate a statistically significant difference between the results observed and 100% (t-test; *P<0.05; ***P<0.001).
Citation: Journal of Endocrinology 184, 1; 10.1677/joe.1.05654
Treatment with the PKC inhibitor Ro 31-8220 does not prevent AVP-induced desensitization. Treatment with a 100 nM AVP pulse (indicated by arrows) resulted in a peak of ACTH secretion both in the absence (a) and presence (c) of 2 μM Ro 31-8220. Pretreatment with 10 nM AVP for 15 min (indicated by black bars) resulted in a reduction in response to a subsequent stimulation with 100 nM AVP, both in the absence (b) and presence (d) of Ro 31-8220. Representative data from four perifusion columns are shown. Data reflect the ACTH secreted in response to a 100 nM AVP pulse both with and without pretreatment with 10 nM AVP for 15 min, and in the presence (gray bars) and absence (open bars) of Ro 31-8220. Data are mean±s.e.m. (n=3–4 for each treatment). Asterisks indicate statistically signicant difference between the pretreated and control value for each pair of responses (t-test; *P<0.05; ***P<0.001).
Citation: Journal of Endocrinology 184, 1; 10.1677/joe.1.05654
Treatment with the CK1α inhibitor CK1–7 did not prevent AVP-induced desensitization of the ACTH response to a subsequent stimulation with 100 nM AVP. Data in panels a–e are from single representative perifusion columns in which the second 100 nM AVP pulse has been preceded by treatment with 10 nM AVP alone (a) and 10 nM AVP in combination with either 10 μM CK1–7 (b), 30 μM CK1–7 (c) or 100 μM CK1–7 (d), and with 30 μM CK1–7 alone (e). Black bars indicate the 10 nM AVP pretreatment, and open bars indicate the CK1–7 pretreatment, while the 100 nM AVP pulses are indicated by arrows. Summarized results are shown in panel f. Data shown are the responses, expressed as a percentage of control, to a 5-min, 100 nM AVP pulse after pretreatment with the indicated solutions. Data are mean±s.e.m. (n=6–9 for each treatment). Asterisks indicate a statistically significant difference between the results observed and 100% (t-test; ***P<0.001).
Citation: Journal of Endocrinology 184, 1; 10.1677/joe.1.05654
Effect of treatment with 0.25 mg/ml ConA on the ACTH response to AVP. Representative data from eight perifusion columns are shown in panels a–h. Cells were treated with a 5-min pulse of 100 nM AVP after 160 min of perifusion (indicated by arrows). To induce desensitization, this AVP pulse was immediately preceded by 15-min pretreatment with 10 nM AVP (indicated by black bars; b and f). When recovery periods of 20 or 40 min were allowed between the pretreatment and the pulse, resensitization occurred (c, d, g and h). The effect of pharmacologic blockade of receptor internalization on regulation of the ACTH response to AVP was investigated by treating cells with 0.25 mg/ml ConA for 70 min prior to the 100 nM AVP pulse (e–h; ConA treatment indicated by open bars). Combined results from columns in which ConA was present (closed symbols) and absent (that is, controls; open symbols) are shown in panel i. The ACTH response after pretreatment, as a percentage of control, is plotted against the duration of the recovery period. Data are mean±s.e.m. (n=5–8 for each treatment). Asterisks indicate a statistically significant difference in the extent of desensitization observed between the control and ConA-treated cells (Bonferroni’s test; ***P<0.001).
Citation: Journal of Endocrinology 184, 1; 10.1677/joe.1.05654
Treatment with 1 μM FK506 attenuates resensitization of the ACTH response to AVP. Representative data from eight perifusion columns are shown in panels a–h. The experimental protocol used was identical to that described in Fig. 3, except that, instead of ConA, cells were treated with 1 μM FK506 for 70 min immediately prior to the 100 nM AVP pulse (e–h). Combined results from columns in which FK506 was present (closed symbols) and absent (that is, controls; open symbols) are shown in panel i. The ACTH response following pretreatment, as a percentage of control, is plotted against the duration of the recovery period. Data are mean±s.e.m. (n=4–10 for each treatment). Asterisks indicate a statistically significant difference between the extent of desensitization observed between the control and FK506-treated cells (Bonferroni’s test; **P<0.01).
Citation: Journal of Endocrinology 184, 1; 10.1677/joe.1.05654
Effect of treatment with FK506 on the ACTH response to the 15-min pretreatment with 10 nM AVP. The ACTH response to the AVP pretreatment during treatment with 50 nM (gray bar) and 1 μM FK506 (black bar) is expressed as a percentage of the response to pretreatment with AVP alone in control columns (open bar). Data are from columns in which the 10 nM AVP pretreatment began 15 min after the commencement of the FK506 treatment (Fig. 4d and h). Asterisks indicate a statistically significant difference between the results observed with FK506 treatment and the control (Bonferroni’s test; **P<0.01; ***P<0.001). Data are mean±s.e.m. (n=2–4 for each treatment).
Citation: Journal of Endocrinology 184, 1; 10.1677/joe.1.05654
Effect of treatment with 10 nM okadaic acid (OA) on desensitization and resensitization of the ACTH response to AVP. The ACTH response after pretreatment with 10 nM AVP for 15 min is plotted against the duration of the recovery period. Combined results from columns in which OA was present (closed symbols) and absent (that is, controls; open symbols) are shown. Data are mean±s.e.m. (n=2–10 for each treatment).
Citation: Journal of Endocrinology 184, 1; 10.1677/joe.1.05654
The authors wish to thank Professor Richard Donald for the ACTH antiserum, and the late Dr C H Li for highly purified oACTH. Ro 31-8220 was kindly provided by Dr G Lawton of the Roche Research Centre, Welwyn Garden City, UK. FK506 was kindly provided by Dr M Tomoi of the Fujisawa Pharmaceutical Co., Ltd, Osaka, Japan. This work was supported in part by grants from the Canterbury Medical Research Foundation (Project Grant 99028) and the University of Canterbury (Grants 2314290 and U6405), and by an AMI McKessar Fellowship to Dr Hassan from the Canterbury Medical Research Foundation. The authors declare that there is no conflict of interest that would prejudice the impartiality of this scientific work.
References
Aguilera G 1994 Regulation of pituitary ACTH secretion during chronic stress. Frontiers in Neuroendocrinology 15 321–350.
Ancellin N & Morel A 1998 Homologous and heterologous acute desensitization of vasopressin V1a receptor in Xenopus oocytes. Cellular Signalling 10 217–223.
Ancellin N, Preisser L, Corman B & Morel A 1997 Role of protein kinase C and carboxyl-terminal region in acute desensitization of vasopressin V1a receptor. FEBS Letters 413 323–326.
Beaumont V, Hepworth MB, Luty JS, Kelly E & Henderson G 1998 Somatostatin receptor desensitization in NG108–15 cells. A consequence of receptor sequestration. Journal of Biological Chemistry 273 33174–33183.
Berrada K, Plesnicher CL, Luo X & Thibonnier M 2000 Dynamic interaction of human vasopressin/oxytocin receptor subtypes with G protein-coupled receptor kinases and protein kinase C after agonist stimulation. Journal of Biological Chemistry 275 27229–27237.
Bialojan C & Takai A 1988 Inhibitory effect of a marine-sponge toxin, okadaic acid, on protein phosphatases. Specificity and kinetics. Biochemical Journal 256 283–290.
Birnbaumer M 2000 Vasopressin receptors. Trends in Endocrinology and Metabolism 11 406–410.
Caramelo C, Tsai P, Okada K, Briner VA & Schrier RW 1991 Mechanisms of rapid desensitization to arginine vasopressin in vascular smooth muscle cells. American Journal of Physiology 260 F46–F52.
Caraty A, Grino M, Locatelli A, Guillaume V, Boudouresque F, Conte-Devolx B & Oliver C 1990 Insulin-induced hypoglycemia stimulates corticotropin-releasing factor and arginine vasopressin secretion into hypophysial portal blood of conscious, unrestrained rams. Journal of Clinical Investigation 85 1716–1721.
Chijiwa T, Hagiwara M & Hidaka H 1989 A newly synthesized selective casein kinase I inhibitor, N-(2-aminoethyl)-5-chloroisoquinoline-8-sulfonamide, and affinity purification of casein kinase I from bovine testis. Journal of Biological Chemistry 264 4924–4927.
Cohen P, Klumpp S & Schelling DL 1989 An improved procedure for identifying and quantitating protein phosphatases in mammalian tissues. FEBS Letters 250 596–600.
Davis PD, Elliott LH, Harris W, Hill CH, Hurst SA, Keech E, Kumar MK, Lawton G, Nixon JS & Wilkinson SE 1992 Inhibitors of protein kinase C. II. Substituted bisindolylmaleimides with improved potency and selectivity. Journal of Medical Chemistry 35 994–1001.
De Keyzer, Auzan C, Lenne F, Chérif B, Thibonnier M, Bertagna X & Clauser E 1994 Cloning and characterization of the human V3 pituitary vasopressin receptor. FEBS Letters 356 215–220.
Evans MJ, Brett JT, McIntosh RP, McIntosh JEA, Roud HK, Livesey JH & Donald RA 1985 The effect of various corticotropin-releasing factor trains on the release of adrenocorticotropin, β-endorphin and β-lipotropin from perifused ovine anterior pituitary cells. Endocrinology 117 893–899.
Evans MJ, Mulligan RS, Livesey JH & Donald RA 1996 The integrative control of adrenocorticotrophin secretion: a critical role for corticotrophin-releasing hormone. Journal of Endocrinology 148 475–483.
Ferguson SSG 2001 Evolving concepts in G protein-coupled receptor endocytosis: the role in receptor desensitization and signaling. Pharmacological Reviews 53 1–24.
Ferguson SSG, Barak LS, Zhang J & Caron MG 1996 G-protein-coupled receptor regulation: role of G-protein-coupled receptor kinases and arrestins. Canadian Journal of Physiology and Pharmacology 74 1095–1110.
Gallo-Payet N, Chouinard L, Balestre MN & Guillon G 1991 Involvement of protein kinase C in the coupling between the V1 vasopressin receptor and phospholipase C in rat glomerulosa cells: effects on aldosterone secretion. Endocrinology 129 623–634.
Garland AM, Grady EF, Lovett M, Vigna SR, Frucht MM, Krause JE & Bunnett NW 1996 Mechanisms of desensitization and resensitization of G protein-coupled neurokinin1 and neurokinin2 receptors. Molecular Pharmacology 49 438–446.
Hassan A, Chacko S & Mason D 2003 Desensitization of the adrenocorticotrophin responses to arginine vasopressin and corticotrophin-releasing hormone in ovine anterior pituitary cells. Journal of Endocrinology 178 491–501.
Herzig S & Neumann J 2000 Effects of serine/threonine protein phosphatases on ion channels in excitable membranes. Physiological Reviews 80 173–210.
Jard S, Gaillard RC, Guillon G, Marie J, Schoenenberg P, Muller AF & Sawyer WH 1986 Vasopressin antagonists allow demonstration of a novel type of vasopressin receptor in the rat adenohypophysis. Molecular Pharmacology 30 171–177.
Keller-Wood M 1994 Vasopressin response to hyperosmolality and hypotension during ovine pregnancy. American Journal of Physiology 266 R188–R193.
Le Beau A & Mason DR 1994 The effects of a chemically diverse range of calcium channel antagonists on the AVP-stimulated ACTH response in ovine corticotrophs. Cell Calcium 16 47–58.
Liu JP 1994 Studies of the mechanisms of action of corticotropin-releasing factor (CRF) and arginine vasopressin (AVP) in the ovine anterior pituitary: evidence that CRF and AVP stimulate protein phosphorylation and dephosphorylation. Molecular and Cellular Endocrinology 106 57–66.
Liu J-P, Engler D, Funder JW & Robinson PJ 1994 Arginine vasopressin (AVP) causes the reversible phosphorylation of the myristoylated alanine-rich C kinase substrate (MARCKS) protein in the ovine anterior pituitary: evidence that MARCKS phosphorylation is associated with adrenocorticotropin secretion. Molecular and Cellular Endocrinology 101 247–256.
Lohse MJ 1993 Molecular mechanisms of membrane receptor desensitization. Biochimica et Biophysica Acta 1179 171–188.
Lohse MJ, Benovic JL, Codina J, Caron MG & Lefkowitz RJ 1990 β-Arrestin: a protein that regulates β-adrenergic receptor function. Science 247 1547–1550.
Mogensen N, Saermark T & Vilhardt H 1988 Endocytosis of the vasopressin receptor by anterior pituitary cells is increased by corticotropin-releasing factor (CRF). Regulatory Peptides 20 223–231.
Mundell SJ & Kelly E 1998 The effect of inhibitors of receptor internalization on the desensitization and resensitization of three Gs-coupled receptor responses. British Journal of Pharmacology 125 1594–1600.
Neill JD, Duck LW, Musgrove LC & Sellers JC 1998 Potential regulatory roles for G protein-coupled receptor kinases and β-arrestins in gonadotropin-releasing hormone receptor signaling. Endocrinology 139 1781–1788.
Pippig S, Andexinger S & Lohse MJ 1995 Sequestration and recycling of β2-adrenergic receptors permit receptor resensitization. Molecular Pharmacology 47 666–676.
Pitcher JA, Freedman NJ & Lefkowitz RJ 1998 G protein-coupled receptor kinases. Annual Review of Biochemistry 67 653–692.
Roth NS, Campbell PT, Caron MG, Lefkowitz RJ & Lohse MJ 1991 Comparative rates of desensitization of β-adrenergic receptors by the β-adrenergic receptor kinase and the cyclic AMP-dependent protein kinase. PNAS 88 6201–6204.
Schreiber SL 1991 Chemistry and biology of the immunophilins and their immunosuppressive ligands. Science 251 283–287.
Shih M & Malbon CC 1994 Oligodeoxynucleotides antisense to mRNA encoding protein kinase A, protein kinase C, β-adrenergic receptor kinase reveal distinctive cell-type-specific roles in agonist-induced desensitization. PNAS 91 12193–12197.
Shih M, Lin F, Scott JD, Wang HY & Malbon CC 1999 Dynamic complexes of β2-adrenergic receptors with protein kinases and phosphatases and the role of gravin. Journal of Biological Chemistry 274 1588–1595.
Spurney RF 1998 Effect of receptor number on desensitization of the mouse thromboxane receptor. Biochemical Pharmacology 55 1271–1281.
Tobin AB, Totty NF, Sterlin AE & Nahorski SR 1997 Stimulus-dependent phosphorylation of G-protein-coupled receptors by casein kinase 1α. Journal of Biological Chemistry 272 20844–20849.
Trincavelli ML, Tuscano D, Cecchetti P, Falleni A, Benzi L, Klotz KN, Gremigni V, Cattabeni F, Lucacchini A & Martini C 2000 Agonist-induced internalization and recycling of the human A3 adenosine receptors: role in receptor desensitization and resensitization. Journal of Neurochemistry 75 1493–1501.
Waugh MG, Challiss RAJ, Berstein G, Nahorski SR & Tobin AB 1999 Agonist-induced desensitization and phosphorylation of m1-muscarinic receptors. Biochemical Journal 338 175–183.
