Homologous regulation of the α2C-adrenoceptor subtype in human hepatocarcinoma, HepG2

Abstract

1. Previous studies of the regulation of the α2C-adrenoceptor in OK and in transfected cells have led to discrepant conclusions. In the present work, we examined the homologous regulation of the human α2C-adrenoceptor in the hepatocarcinoma cell-line, HepG2; a model which expresses this subtype spontaneously.

2. Short-period treatment of the cells with UK14304 provoked neither a diminution of the potency of the α2-agonist to inhibit forskolin-induced cyclic AMP-accumulation nor a change in the degree of receptor coupling to G-proteins.

3. Long-period exposure to UK14304 resulted in a large reduction of [³H]MK912 binding sites (55% decrease). The action of UK14304 was dose-dependent (EC₅₀=190±45 nM), rapid (t_1/2 =4.2 h) and reversible. Receptor down-regulation was also observed with clonidine or (−)adrenaline (38 and 36% decrease, respectively) and was blocked by the addition of α2-antagonists.

4. Conversely to that observed with α2-agonists, treatment of the cells with RX821002 or yohimbine alone, but not with phentolamine, promoted a significant increase of the receptor expression.

5. The observed alterations of receptor density are not the reflection of changes at the α2C4 mRNA level. Estimation of the receptor protein turnover and measurement of its half-life demonstrated that down-regulation by α2-agonists and up-regulation by α2-antagonists, with inverse-agonist efficacy, are respectively the consequence of increased and decreased rate of receptor degradation.

6. In conclusion, our data show that α2C-adrenoceptor does not undergo desensitization but is down-regulated in HepG2. The lack of desensitization agrees with previous results obtained in cells transfected with the α2C4 gene, but not with observations made in OK cells. Inversely, down-regulation fits with results obtained in OK but not in transfected cells. The reasons for these discrepancies are discussed. Our results also demonstrated that certain α2-antagonists behave as inverse agonist on the HepG2 model and thus provide for the first time evidence of inverse efficacy of antagonists on a cellular model expressing physiological level of a wild-type α2-adrenoceptor.

Introduction

Many observations made both in vivo and in vitro have demonstrated that sustained stimulation of G-protein-coupled receptors (GPCRs) generally results in a subsequent attenuation of their triggered biological responses. The magnitude and the determinants of this adaptive process appear to vary with the type of GPCR considered; but according to our knowledge of the most well-defined models, it is now clear that the emergence of refractoriness is the consequence of the combination of a set of timely-operated phenomena including desensitization, internalization and down-regulation of the receptor (for review see Böhm et al., 1997).

Receptor desensitization occurs within seconds to minutes following exposure to the agonist. It corresponds to a functional uncoupling of the receptor from its G-proteins and it is primarily due to receptor phosphorylation by protein kinases belonging either to the second-messenger protein kinase family (heterologous desensitization) or to G-protein-coupled receptor kinases (homologous desensitization). Receptor internalization, on the other hand, occurs within minutes to hours following agonist exposure and may contribute to further attenuation of the cellular receptivity by translocating surface receptors to an intracellular compartment where physical interactions with the agonist are no longer possible. The precise mechanisms involved in this process are not fully elucidated yet, but it is considered that the predominant pathway is via the clathrin-coated pits. Finally, down-regulation is characterized by a net decrease in receptor number and is only observed after prolonged exposure to agonist for hours to days. The molecular mechanisms responsible for this phenomenon largely depend on the type of GPCR considered and the diminution of receptor number reflects enhanced degradation and/or altered synthesis of the receptor polypeptide.

The effects of catecholamines on cyclic AMP production are mediated through interactions with α2- and β-adrenoceptors which are negatively and positively coupled to adenylyl cyclase, respectively. While most of our present understanding of the determinants of GPCR regulation derives from extensive studies of the β-adrenoceptor system (Collins et al., 1991) far less is known about α2-adrenoceptors. The α2-adrenoceptors are currently classified into three pharmacological subtypes, namely: α2A, α2B and α2C (Bylund et al., 1994). The genes encoding these subtypes have been cloned in different species and were respectively termed α2C10, α2C2 and α2C4 in human (Kobilka et al., 1987; Regan et al., 1988; Lomasney et al., 1990). According to the results obtained on transfected cells, a major divergence between α2-adrenoceptor subtypes lies in their ability to undergo desensitization and down-regulation. Studies carried out on cells transfected with the human genes indicated that exposure to adrenaline resulted in a rapid attenuation of α2-agonist-induced inhibition of adenylyl cyclase in cells expressing the α2A- or the α2B-subtype, while no desensitization was observed with the α2C (Eason & Liggett, 1992; Kurose & Lefkowitz, 1994). Moreover, long term exposure to the catecholamine provoked a down-regulation of the human α2A- or α2B-subtype, but not of the α2C (Eason & Liggett, 1992). On the other hand, experiments performed on the OK cell-line (a model which spontaneously expresses an α2C-adrenoceptor pharmacologically identical to human, but from opossum origin) demonstrated that treatment with noradrenaline results in a rapid decrease in the potency of α2-agonists to inhibit PTH-induced cyclic AMP production (Jones et al., 1990). Furthermore, long-term exposure to the neurotransmitter induces a significant reduction in receptor number (Shreve et al., 1991; Pleus et al., 1993) indicating that both desensitization and down-regulation of α2C-adrenoceptor subtype occurs in these cells.

In contrast to α2A-adrenoceptor, which largely benefited of the existence of cell-lines such as HT29 (Paris et al., 1987) or HEL (McKernan et al., 1988), the study of the regulation of the human α2C-adrenoceptor has been so far restricted to transfected cells because of the lack of model constitutively expressing this subtype. The recent discovery of the human hepatocarcinoma HepG2 as exhibiting α2C-adrenoceptors negatively coupled to adenylyl cyclase (Schaak et al., 1997a) allows now to investigate the regulation of this subtype in a cell-line natively expressing this receptor subtype.

The present study carried out on HepG2 shows that α2C-adrenoceptor does not undergo rapid desensitization but is largely down-regulated in response to long-term exposure to α2-agonists. Conversely cell treatment with certain α2-antagonists resulted in an increase of receptor density. According to estimation of receptor protein turnover and to measurement of its half-life, the changes induced by agonists and antagonists are respectively due to increased and decreased rate of the receptor degradation. These observations bring new insights onto the mechanisms of homologous regulation of α2C-adrenoceptors.

Methods

Cell culture and treatments

The human hepatocarcinoma HepG2 was cultured in Dulbecco's Modified Eagle's Medium (DMEM) containing 25 mM glucose, 100 μg ml⁻¹ streptomycin, 100 IU ml⁻¹ penicillin, 2 mM glutamine and 10% foetal calf serum. Unless otherwise specified, all treatments were performed on confluent attached cells and in the absence of FCS. At zero time of the treatment, the hormone or drug to be tested was added to the culture from frozen stock solution. At the indicated time, the medium was removed and the cell-layers rinsed twice with phosphate buffered saline (PBS). According to their intended use, cells were either frozen at −80°C until analysis (binding experiments, RNA extraction) or isolated in PBS containing 0.5 mM ethylenediaminetetraacetic acid (EDTA) (measurement of intracellular cyclic AMP).

Generation and synthesis of α2C4 and β-actin riboprobes

The probe for detection of α2C4 mRNAs was obtained by subcloning a 370 bp fragment (SmaI-MaeIII) corresponding to nucleotides 1014–1382 of the α2C4 coding region into pBluescriptII KS+ (pKS+, Stratagene, La Jolla, CA, U.S.A). The specificity of this probe has been previously assessed (Valet et al., 1993). The β-actin probe was obtained by PCR and TA-cloning into the EcoRV site of pKS+, the amplified fragment (236 bp) corresponds to nucleotides 415–650 of the cDNA (exon 3). For synthesis of the radiolabelled probes, the two plasmids were cut with the appropriated restriction enzyme and antisense RNAs were synthesized in the presence of [³²P]UTP using the T3 RNA polymerase (Promega).

RNA preparation and RNase protection assays (RPA)

Cellular RNAs were isolated using the guanidium isothiocyanate/phenol-chloroform extraction method (Chomczynski et al., 1987). The integrity of the preparations was assessed by agarose gel electrophoresis and the RNA concentrations were measured by u.v. spectrophotometry. RPA were performed as previously described (Schaak et al., 1997a). Lyophilised RNAs (100 μg) were taken in 30 μl of hybridization buffer (80% deionized formamide, 0.4 M NaCl, 1 mM EDTA, 40 mM Pipes, pH 6.7) containing an excess of ³²P-labelled riboprobe. The samples were heated to 95°C for 5 min and then immediately placed at 55°C for 14 h. Non-hybridized probe was eliminated by the addition of 0.3 ml of TEN buffer (10 mM Tris-HCl, 5 mM EDTA, 300 mM NaCl, pH 7.5) containing RNases A (40 μg ml⁻¹) and T₁ (2 μg ml⁻¹). After 2 h at 37°C, digestion was stopped by the addition of 5 μl of proteinase K (10 mg ml⁻¹) and samples were further incubated for 15 min at 37°C. Carrier tRNA (10 μg) in 0.3 ml of solution D (4 M guanidium thiocyanate, 25 mM sodium citrate, 0.1 M 2-mercaptoethanol and 0.5% sarkosyl, pH 7.0) were then added and protected hybrids precipitated with isopropyl alcohol. After washing with 70% ethanol, RNA pellets were dissolved in 10 μl of sample buffer (97% deionized formamide, 0.1% SDS, 10 mM Tris-HCl, pH 7.0) and loaded onto a 5% acrylamide/7 M urea gel. The amounts of protected radiolabelled probe were quantified using a Phosphor-Imager (Molecular Dynamics, Sunnyvale, CA, U.S.A.).

Receptor quantification

The number of α2-adrenoceptors was measured on crude membrane preparations using the selective α2-adrenergic antagonist [³H]MK912 (2s, 12bs-1′,3′ dimethyspiro (1,3,4,5′,6,6′,7,12b-octahydro-2H-benzo[b]furo [2,3-a] quinazoline)-2,4′-pyrimidin-2′one) (Pettibone et al., 1989). Frozen cells were homogenized in 25 ml of TE buffer (50 mM Tris-HCl, 5 mM EDTA, pH 7.5) and centrifuged at 39,000×g for 10 min. The particulate fraction was washed in TE buffer and the final crude membrane pellet was taken up in the appropriate volume of TM buffer (50 mM Tris-HCl, 0.5 mM MgCl₂, pH 7.5). The protein concentration was determined using the Coomassie blue method (Bradford, 1976). Total binding was measured by incubating 100 μl of cell membrane preparation with the radioligand in a total volume of 400 μl of TM buffer. After a 45 min period of incubation at 25°C, bound radioactivity was separated from free by filtration through GF/C Whatman filters using a Millipore Manifold Sampling unit. Filters were rapidly washed with ice-cold TM buffer and membrane-bound radioactivity was determined by liquid spectrometry. Specific binding was defined as the difference between total and non-specific binding measured in the presence of 10⁻⁵ M phentolamine. For saturation studies, the final concentrations of [³H]MK912 ranged from 0.04–3 nM.

Determination of intracellular cyclic AMP content

Cells were detached in PBS containing 0.5 mM EDTA and collected by gentle centrifugation (900×g, 5 min, 4°C). The pellet was suspended in DMEM buffered with 10 mM HEPES (pH 7.4). Aliquots of the cell suspension were incubated in a 200 μl final volume of HEPES-buffered DMEM containing 0.2 mM 3-isobutyl-1-methylxanthine and the indicated concentration of the drug to be tested. After 15 min at 37°C, the reaction was stopped by adding 1.8 ml of methanol/formic acid (95/5, v/v). The cell lysate was centrifuged (3000×g, 10 min, 4°C) and aliquots of the supernatant were evaporated. The dry samples were taken in acetate buffer containing 0.1% NaN₃ and their cyclic AMP content was determined by radio-immunological assay (Steiner et al., 1972).

Drugs and chemicals

[³H]MK912 (79 Ci mmol⁻¹) was from New England Nuclear (Boston, MA, U.S.A.) and [α-³²P]UTP from ICN (Costa Mesa, CA, U.S.A.). Phentolamine and clonidine were donated by Ciba-Geigy (Basel, Switzerland) and Boehringer-Ingelheim (Ingelheim, Germany). UK14304 (5-bromo-6-(2-imidazolin-2-ylamino)-quinoxaline) and RX821002 (2-(2-methoxy-1,4-benzodioxan-2-yl)-2-imidazoline) were gifts from Pfizer (Sandwich, U.K.) and Reckitt and Colman Laboratories (Kingston-upon-Hull, U.K.). RNase A, RNase T₁ and proteinase K were purchased from Boehringer Mannheim (Meylan, France). Yohimbine, (−)adrenaline, benextramine, forskolin, concanavalin A, cycloheximide and all other chemicals were from Sigma (St. Louis, MI, U.S.A.). Foetal calf serum was purchased from Gibco-BRL (Cergy Pontoise, France). Radioimmunoassay kits for cyclic AMP determination were from Immunotech (Luminy, France).

Analysis of data

Binding data were analysed using the EBDA-LIGAND computer programs (McPherson, 1985). All data are expressed as means±s.e.mean and statistical analysis was performed by Student's t-test. Statistical differences for P less than 0.05 and 0.001 were indicated with asterisks (^* and ^***, respectively).

The kinetics of receptor recovery after benextramine treatment were fitted according to a model which validity was already verified for other receptors (Neve & Molinoff, 1986). This model assumes that within each group (untreated and treated cells) (i) the rate of receptor synthesis remains constant during the period of receptor recovery and (ii) the rate of receptor degradation is proportional to its concentration. The equation which fits this model is [R_t]=V/k (1−e^−Kt), where R_t is the receptor density at discrete time t, V its rate of synthesis and k its disappearance rate constant. When t tends to ∞, then [R_t] approaches V/k and represents receptor density at steady state [R_ss]. Therefore the equation can be written [R_t]=[R_ss] (1−e^−Kt). Its logarithmic transformation (In [R_ss] / [R_ss−R_t]=kt) gives straight line and allows to calculate k.

Results

Effect of short-term agonist-exposure on α2C-adrenoceptor responsiveness

HepG2 cells were demonstrated to express α2C-adrenoceptors negatively coupled to adenylyl cyclase very recently (Schaak et al., 1997a) and no information is available yet about the mechanisms regulating this receptor subtype in this model. Previous studies of α2C-adrenoceptor desensitization in OK cells (Jones et al., 1990; Shreve et al., 1991) and in transfected cells having led to divergent conclusions (Eason & Liggett, 1992; Kurose & Lefkowitz, 1994), we thought of interest to examine first whether the functional efficacy of HepG2 receptor underwent desensitization with prior exposure to α2-agonist. To do so, HepG2 cell-layers were incubated with 10 μM UK14304 (full agonist). After a 30 min period of treatment, cells were extensively washed in PBS, rapidly isolated in cold and assayed for α2-adrenoceptor mediated inhibition of forskolin-induced cyclic AMP accumulation. Results from this experiment, which are depicted in Figure 1a, indicated that neither basal nor forskolin-stimulated cyclic AMP levels were affected by the pretreatment. Maximal inhibition of cyclic AMP accumulation by UK14304 reached similar extend (56±4% in control and 51±5% in pretreated cells). Dose-response curves were superimposable and no shift in the potency of UK14304 was observed. Computer-assisted analysis of the data indicated that EC₅₀ values for UK14304 to inhibit cyclic AMP accumulation were not significantly different (28±4 and 53±15 nM in control and pretreated-cells, respectively). Strictly similar results were obtained after a 15 min period of pretreatment with UK14304 (not shown), indicating that the α2C-adrenoceptor did not undergo desensitization in HepG2 cells.

Figure 1.

Effect of short-term exposure to UK14304 on the functional efficacy and degree of coupling of the α2C-adrenoceptor. HepG2 cell layers were pre-incubated without or with 10 μM UK14304 for 30 min and were extensively washed in cold PBS. (a) Control (○) and agonist-treated cells (•) were detached in cold PBS containing 0.5 mM EDTA. After mild centrifugation at 4°C, cells were suspended and incubated 15 min at 37°C in 200 μl of HEPES-buffer DMEM containing either vehicle (basal) or 10 μM forskolin or 10 μM forskolin plus indicated concentrations of UK14304. Concentration of cyclic AMP was measured as described in the ‘Methods' section. Results are expressed as picomoles of cyclic AMP mg−1 of cellular protein and are means±s.e.mean (n=6). (b) Cell-layers from control (□) or agonist-treated (▪) dishes were rapidly rinsed and frozen. Membranes were then prepared as described in the Methods section and incubated in the presence of 0.8 nM [³H]MK912 and increasing concentrations of UK14304. The amount of specifically bound radioligand was determined using 10⁻⁵ M phentolamine to estimate non-specific binding. The amounts of bound [³H]MK912 are expressed as per cent of control and the presented data are the mean±S.D. from three independent experiments. Analysis of the inhibition curves according to a two-site model indicated that the per cent of receptor under high-affinity state for agonist was 53±6% and 58±8% in control and UK14304-treated cells respectively. K_iH and K_iL values for UK14304 were respectively 0.41±0.17 nM and 43±8 nM in control cells and 1.70±0.25 nM and 161±67 nM and agonist-treated cells (mean±s.e.mean, n=3).

This conclusion was further confirmed by the study of the inhibition of the binding of the radiolabelled antagonist [³H]MK912 by UK14304. The inhibition curves obtained on washed membranes prepared from control and from agonist-treated cells are depicted in Figure 1b. Analysis of the data according to a two-site inhibition model allowed to calculate that the percentage of high-affinity agonist binding sites was 53±6% in control cells and 58±8% in UK14304-pretreated cells (mean±s.e.mean, n=3). Exposure to the agonist provoked a slight rightward shift of the inhibition curve resulting in an small increase (4 fold) of both K_iH and K_iL values. However, the ratio of K_i values of UK14304 for low- and high-affinity state (K_iL : K_iH) was unchanged (106±55 in control versus 80±29 in treated-cells, mean±s.e.mean, n=3) indicating that the degree of receptor coupling to G_i-proteins was unaffected by the exposure to the agonist.

Effect of long-term exposure to α2-adrenergic compounds on α2-adrenoceptor density

As determined by saturation binding isotherms with [³H]MK912, the density of α2-adrenoceptors in HepG2 cells cultured in the presence of 10% FCS was 55.7±3.7 femtomoles mg⁻¹ of membrane protein (n=17). In contrast to that previously observed for α2A-adrenoceptor subtype in HT29 cells (Devedjian et al., 1991), a 48 h period of serum deprivation did not provoke an increase in receptor number. All further treatments were however carried out in serum-deprived medium in order to avoid any interference of FCS with the tested compounds. Under such culture conditions, a 48 h exposure to UK14304 (10 μM) induced a large reduction (55%) of receptor number (Figure 2a). No alteration of receptor affinity was observed, the dissociation constant for [³H]MK912 being unchanged by the treatments (0.08±0.03 nM in control versus 0.09±0.02 nM in treated cells). As shown in the Figure 2b, exposure to 10 μM (−)adrenaline or to 10 μM clonidine resulted also in a significant decrease of receptor expression (38 and 36% decrease, respectively). The effect of 10 μM UK14304 was prevented by the prior addition of 100 μM RX821002 (α2-antagonist), thus demonstrating that receptor occupancy by the agonist is necessary for receptor down-regulation to occur. A blockade of UK14304-induced down-regulation was also observed with 100 μM yohimbine or 100 μM phentolamine (not shown). It is remarkable that treatment with solely RX821002 or yohimbine, but not phentolamine, resulted in a significant increase in receptor number, indicating that certain α2-antagonists have the potency to up-regulate the receptor and behave as inverse agonists in this regard.

Figure 2.

Effect of long-term exposure to α2-adrenergic compounds on α2C-adrenoceptor density. (a) HepG2 cells were maintained for 48 h in serum-free DMEM containing either 10 μM UK14304 (•) or not (○). Membranes were prepared as described in the Methods section and incubated in the presence of various concentrations of [³H]MK912. The amount of specifically bound radioligand was determined using 10⁻⁵ M phentolamine to estimate non-specific binding. The data from a typical experiment are displayed as a Scatchard plot. Each point is the mean of determination in duplicate. (b) HepG2 cells were exposed for 48 h to various α2-adrenergic compounds. The tested drugs were 10 μM UK14304, 100 μM RX821002, 10 μM UK14304 plus 100 μM RX821002 (UK+RX), 10 μM (−)adrenaline, 10 μM clonidine, 100 μM phentolamine or 100 μM yohimbine. Membranes were prepared and assayed for their capacity to bind [³H]MK912. The maximum number of binding sites expressed as femtomoles mg⁻¹ of proteins was calculated by analysis of the radioligand saturation isotherms according to a one-component model. Reported values are the means±s.e.mean from four determinations (^*** and ^* indicate values significantly different from control at P<0.001 and P<0.05, respectively).

The dose-dependence and the time-course of the receptor down-regulation are depicted in Figure 3a and b. The decrease in receptor expression provoked by UK14304 was dose dependent (EC₅₀=190±45 nM) and reached its maximum (55% diminution) at 5 μM concentration of agonist. On the other hand, the decline of α2-adrenoceptor number induced by UK14304 was rather rapid and was completed within 12 h following exposure. The process was consistent with a mono-exponential model with a half-life of disappearance of 4.2 h. Finally, when HepG2 cells exposed to UK14304 for 12 h were washed and then further incubated in drug-free medium, α2-adrenoceptors were observed to recover up to control levels within 24 h following agonist removal.

Figure 3.

Dose-dependence, kinetics and reversibility of α2C-adrenoceptor down-regulation induced by UK14304. (a) HepG2 cells were exposed for 48 h to final concentrations of UK14304 ranging from 1 nM to 10 μM. Membranes were prepared and assayed for their capacity to bind [³H]MK912 as described in the legend of Figure 2. Each data point represents the mean±s.e.mean from three experiments. Analysis of the dose-response indicated that half-maximal down-regulation of the receptor was achieved at 190±45 nM UK14304. (b) HepG2 cells were incubated with 10 μM UK14304. Cells were harvested at the indicated times and membrane preparations assayed for their capacity to bind [³H]MK912 as described in the legend of Figure 2. After a 12 h period of exposure to UK14304, cells were extensively washed and then replaced in agonist-free medium (○). The number of [³H]MK912 binding sites is expressed as per cent of control (untreated-cells). Data represent the mean of two experiments.

Measurement of α2C4 mRNA levels

Previous studies of the molecular mechanisms accounting for β2-adrenoceptor down-regulation have established that the diminution in receptor number induced by β-agonists is the reflection of decreased levels of receptor mRNA. This change results from an accelerated rate of mRNA degradation (Böhm et al., 1997) which is the consequence of the binding of a 35 kD protein to AUUUA pentamers located in the 3′UTR of the β2-adrenoceptor transcript (Port et al., 1992; Tholanikunnel et al., 1995). Such a sequence being present in the 3′UTR of the α2C4 mRNA (Schaak et al., 1997b), we thus wondered whether such a mechanism also occurred in HepG2 cells. HepG2 cells were incubated or not in the presence of the different α2-adrenergic drugs and the amounts of α2C4 mRNA were measured by RPA using a ³²P-labelled α2C4 antisense RNA. β-actin was used as an internal standard to normalize the quantification. A typical autoradiogram is presented in Figure 4. Analysis of the data obtained from cells treated for different periods of time (6, 12, 24 and 48 h) showed that neither α2-agonists (UK14304, (−)adrenaline, clonidine) nor the α2-antagonist (RX821002) affected the steady state levels of the α2C4 mRNAs.

Figure 4.

Effect of α2-adrenergic compounds on α2C4 mRNA level. Cellular RNAs prepared from HepG2 cells maintained for 12 h under different culture conditions were hybridized with labelled antisense riboprobe for α2C4 or β-actin. The samples were digested with a mixture of RNases and the resistant hybrids were separated by electrophoresis. The treatments tested were the following: none (control), 10 μM UK14304, 10 μM (−)adrenaline, 10 μM clonidine or 100 μM RX821002. A representative autoradiogram is shown. Analysis of four different experiments by direct quantification of the radioactive bands using a Phosphor-Imager indicated that none of the tested drugs provoked a significant change in the amount of α2C4 mRNA.

UK14304 increases and RX821002 decreases the rate of α2C-adrenoceptor degradation

The decrease in receptor density induced by UK14304 being rather rapid and not ascribable to a change in the amount of α2C4 mRNA, we hypothesized that the effect of the agonist results from a modification of the receptor turnover. The half-life of the receptor protein and the rate of its degradation were thus examined to investigate in this way. As previously done for the α2A-adrenoceptor in HT29 cells (Paris et al., 1987), the half-life of the α2C-adrenoceptor was first determined by following the recovery of the binding capacity of the HepG2 cells after blockade of pre-existing receptors with the irreversible ligand, benextramine. This alkylating agent having never been used on α2C-adrenoceptors, preliminary experiments were carried out to assess its apparent affinity for this receptor subtype and to verify the irreversibility of the blockade. The apparent affinity of benextramine for the α2C-adrenoceptor was estimated by determining its potency to inhibit [³H]MK912 binding to HepG2 cell membrane preparations. The inhibition curves obtained with benextramine and with the selective α2-antagonist RX821002 are shown in Figure 5a.

Figure 5.

Irreversible effect of benextramine on α2C-adrenoceptors. (a) Membranes prepared from HepG2 cells were incubated in the presence of 0.5 nM [³H]MK912 and increasing concentrations of benextramine (□) or RX821002 (▪). According to Hill coefficient values not significantly different from unity, inhibition curves were fitted to a one-site model. The calculated EC₅₀ values were 49±5 nM for benextramine and 17±1 nM for RX821002. (b) HepG2 cells were incubated for 20 min at 37°C in the presence of either 2 μM benextramine (Benex), 20 μM RX821002 (RX) or 2 μM benextramine plus 20 μM RX821002 (RX+Benex). Cells were extensively washed, membranes were prepared and α2-adrenoceptors were quantified by analysis of [³H]MK912 saturation isotherms.

Analysis of the data according to a one-site inhibition model indicated that the value of EC₅₀ for benextramine (49±5 nM) was close to that obtained for the specific α2-antagonist, RX821002 (17±1 nM), demonstrating this compound had a rather good affinity for the α2C-adrenoceptor. A benextramine concentration of 2 μM was used in all further experiments on intact cells. As shown in Figure 5b, the treatment of HepG2 cells with such a dose of benextramine resulted in a total abolition of the capacity of cell membrane to bind [³H]MK912. As expected, this effect was not observed with the reversible antagonist RX821002. It was moreover prevented by the prior addition of an excess of RX821002, indicating that the irreversible ligand docks in the receptor binding pocket. Given these preliminary data, a series of culture dishes were treated with benextramine 20 min at 37°C. Cells were then extensively washed in drug-free DMEM and allowed to recover their receptors over a period up to 3 days. The measurement of the reappearance of [³H]MK912 binding sites demonstrated that receptor recovery was a mono-exponential process (Figure 6a). The restoration was rapid (90% of the steady state level of the receptor being recovered within 24 h) and was exclusively due to receptor neosynthesis since it was totally abolished by the addition of cycloheximide (50 μg ml⁻¹) into the culture medium (not shown). The fitting of the kinetic of receptor reappearance according to the model described in the Methods section (Figure 6b), allowed to determine the rate constant of the receptor degradation (k_cont=0.063 h−1). From this value, it can be calculated that the rate constant for receptor synthesis was 3.65±0.5 fmol mg⁻¹ prot h⁻¹ and that the receptor half-life was 11 h, under basal conditions. Results from the parallel study of the receptor turnover on UK14304-treated cells yielded a significantly higher value for the rate constant of the receptor degradation (k_uk=0.112 h−1). According to this value, the rate constant for receptor synthesis was unchanged by the treatment (3.36±0.6 fmol mg⁻¹ prot h⁻¹). The receptor half-life was by contrast reduced to 6.2 h, thus suggesting that the decrease in receptor density by UK14304 was the consequence of an accelerated receptor degradation. Such a conclusion was further confirmed by the direct estimation of the kinetics of receptor degradation in the presence of an inhibitor of protein synthesis (Figure 7). Indeed, the measurement of the decay of [³H]MK912 binding sites in the presence of cycloheximide (50 μg ml⁻¹) yielded values for receptor half-life of 13.4 h and 6.9 h in control and UK14304-treated cells, respectively. Inversely to UK14304, RX821002 increased receptor stability (t_1/2=20.2 h) demonstrating that the up-regulation provoked by antagonists was primarily due to a decrease in the rate of receptor degradation. No significant modification of receptor half-life was observed in cells treated with phentolamine (not shown).

Figure 6.

Kinetics of the α2C-adrenoceptor reappearance after irreversible blockade with benextramine. (a) A series of culture dishes was incubated for 20 min at 37°C in the presence of 2 μM benextramine. Cells layers were then extensively washed and finally replaced either in DMEM (control, ○) or in DMEM containing 10 μM UK14304 (UK14304, •). Cell dishes were collected at the indicated times, membranes were prepared and assayed for their capacity to bind [³H]MK912 as described in the legend of Figure 2. Data represent the means±s.e.mean from three experiments. (b) Semi-logarithmic plot of the kinetics of α2-adrenoceptor recovery. The kinetics of receptor recovery in control (○) or in UK14304-treated cells (•) were plotted according to the equation ln [R_ss] / [R_ss−R_t]=kt, in which R_ss is the number of receptors at steady state and R_t the number of receptors at discrete time. The slopes of the plots which represent the rate constant for receptor degradation (k) were determined by linear regression analysis.

Figure 7.

Estimation of the rate of receptor degradation in the presence of cycloheximide. HepG2 cells were pre-incubated for 12 h in the presence of 10 μM UK14304 (•), 100 μM RX821002 (▪) or not (○). Cycloheximide (50 μg ml⁻¹) was then added into the culture medium and cells were harvested at the indicated times. Membranes were prepared and assayed for their capacity to bind [³H]MK912 as described in the legend of Figure 2. Binding site number is expressed as per cent of the corresponding control (untreated with cycloheximide). Each data point represents the mean±s.e.mean from three experiments.

Studies carried out on cells transfected with cDNA encoding rat or mouse α2C-adrenoceptors have demonstrated that this subtype has both membranous and intracellular localization. The subcellular distribution of α2C-adrenoceptor was not examined in HepG2, but we questioned whether receptor down-regulation requires sequestration into an internal compartment to occur. To address this, HepG2 were treated for 6 h with 10 μM UK14304 under conditions where sequestration was blocked. As shown in the Table 1, blockade of sequestration by 0.25 mg ml⁻¹ concanavalin A or 0.4 M sucrose completely prevented receptor down-regulation.

Table 1.

 Effects of concanavalin A and sucrose on receptor down-regulation

Discussion

The molecular mechanisms underlying the regulation of the expression of α2-adrenoceptors of the α2C-subtype are still unclear. While experiments on transfected cells showed that human α2C-adrenoceptor is refractory to short-term desensitization (Eason & Liggett, 1992; Kurose & Lefkowitz, 1994) and long-term down-regulation (Eason & Liggett, 1992), other studies, carried out with the opossum kidney-derived cell-line OK, suggested that both processes occurred (Jones et al., 1990; Shreve et al., 1991). The objective of the present work was therefore to examine the regulation of the α2C-adrenoceptor in HepG2, a human cell-line which spontaneously expresses this receptor subtype.

The experiments, carried out on this model, show that short-term exposure to α2-agonist results neither in an attenuation of receptor functionality nor in an alteration of its degree of coupling to G_i-proteins, thus demonstrating that α2C-adrenoceptor does not undergo rapid desensitization in HepG2. In this respect, our results agree with what previously found in COS or CHO cells transfected with the α2C4 gene, but do not fit with observations made on OK cells. According to previous studies on purified recombinant α2-adrenoceptors or on transfected cells (Kurose & Lefkowitz, 1994; Jewell-Motz & Liggett, 1996), the failure of the human α2C-adrenoceptor to undergo desensitization is due to the fact that this receptor subtype is not a substrate for G protein-coupled receptor kinases. The reasons for the discrepancy between the results obtained in HepG2 and OK are not fully elucidated. The most likely explanation is that OK cells express a receptor which amino acid sequence exhibits large divergence with the human α2C-adrenoceptor (Blaxall et al., 1994). Major differences between the two polypeptides are found in the third intracellular loop, a region which appears important in receptor phosphorylation and desensitization. In regard to this latter point, it is noteworthy that the third intracellular loop of the OK receptor contains an EESSTS motif (amino acids 306–312) which resembles the EESSSS sequence that is phosphorylated by β-adrenergic receptor kinase (βARK) in the human α2A-adrenoceptor (amino acids 293–299) (Eason et al., 1995). Such a motif is absent from all other α2C-adrenoceptors cloned so far from other species. The possibility therefore exists that OK α2C-adrenoceptor represents an exception in that it undergoes desensitization because it is a substrate for βARK. This hypothesis would merit examination.

In contrast to what found in transfected cells where human α2C-adrenoceptor was refractory to down-regulation (Eason & Liggett, 1992), our current results demonstrate that long-lasting exposure of HepG2 cells to α2-agonists induces a marked decrease in receptor number. HepG2 being from human origin, differences in the primary structure of the receptors cannot be suspected here. CHO cells used in the above-cited study expressed approximately 20 fold higher levels of α2C-adrenoceptors than HepG2, furthermore a large proportion of these receptors was found in an intracellular compartment, a situation that might not fully reflect the physiological reality. Alternatively, the extend of down-regulation could depend of the cell-type. Study of other human cell-lines expressing α2C-subtype endogenously, such as SK-N-MC or Y79 (Schaak et al., 1997a; Gleason & Hieble, 1992), should bring a beginning of answer to this latter point. Conversely to that observed for desensitization, the characteristics of receptor down-regulation in HepG2 are fairly similar to those in OK (Pleus et al., 1993). The maximal extend of receptor decrease after 24 h exposure to 10 μM UK14304 in HepG2 cells (55%) is very similar to that provoked by 30 μM noradrenaline in OK cells (58%). The mechanisms accounting for this phenomenon were not investigated in OK cells. According to the present study, down-regulation of α2C-adrenoceptor was not due to a reduction of α2C4 mRNA levels nor to a diminution of receptor synthesis, but to an approximately 2 fold increase in its rate of degradation. Such a process was recently demonstrated to also account for down-regulation of human α2A- and rat α2B-adrenoceptor in HT29 and transfected-CHO cells (Heck & Bylund, 1997). Our work did not investigate the mechanisms leading to accelerated degradation of α2C-adrenoceptor. However, increased degradation is not the consequence of change in cyclic AMP level because α2-agonists have no significant effect on basal cyclic AMP concentration in HepG2. Moreover, down-regulation is abolished when HepG2 are exposed to concanavalin A or when 0.4 M sucrose is added to the culture medium indicating that internalization is a prerequisite to α2C-adrenoceptor degradation. The receptor being not ‘phosphorylable', it is likely that the internalization process is independent of β-arrestin. As for β2-adrenoceptor (Zhang et al., 1996), further studies, including transfection of HepG2 cells with dominant negative forms of this protein and of dynamin, should afford more information on the exact route of α2C-adrenoceptor processing.

Interestingly, exposure of HepG2 cells to RX821002 or yohimbine provoked a significant rise in the receptor expression which is primarily the consequence of an attenuated rate of receptor degradation. Our experiments being carried out in serum-free conditions and serum deprivation having no consequence on receptor density, there is no concern these effects might reflect blockade of the action of catecholamines or catecholamine metabolites present in FCS. Such a possibility is also eliminated by the observation that phentolamine, which is able to block the receptor down-regulation induced by UK14304, does not affect by itself receptor number and thus behaves as a neutral antagonist on this system. Up-regulation by α2-antagonist has never been reported for any of the α2-adrenoceptor subtypes; however antagonist-induced augmentation of receptor expression was already observed for other GPCRs including α1- and β-adrenoceptors. Increase in receptor expression following exposure to antagonists was for instance observed in NG108-15 cells and in Rat1 fibroblasts permanently transfected with the wild-type human β2-adrenoceptor (Lee et al., 1997) and with a constitutively active mutant (CAM) of the hamster α1B-adrenoceptor (MacEwan & Miligan, 1996). Because solely β-antagonists with inverse agonist property are efficient on the wild-type β2-adrenoceptor whereas all α1-antagonists are efficient on the CAM α1B-adrenoceptor, it is generally believed that up-regulation of GPCR by antagonists is the special feature of inverse-agonists acting on receptors displaying agonist-independent coupling to G-protein (Milligan & Bond, 1997). Whether antagonist-induced up-regulation of HepG2 α2C-adrenoceptor was dependent of its coupling to G_i-proteins was not assayed in the present work, but future studies should address this issue.

Altogether the data provided in this manuscript show that human α2C-adrenoceptor is refractory to rapid desensitization but is down-regulated in response to long-term exposure to α2-agonists. They furthermore demonstrated that certain α2-antagonists behave as inverse agonist on the HepG2 model. To our knowledge this report is the first evidence of inverse efficacy of antagonists on a cell expressing physiological level of a wild-type α2-adrenoceptor.

Abbreviations

βARK

β-adrenergic receptor kinase

DMEM

Dulbecco's Modified Eagle's Medium

EDTA

ethylenediaminetetraacetic acid

GPCR

G-protein-coupled receptor

MK912

2s, 12bs-1′,3′dimethyspiro (1,3,4,5′,6,6′,7,12b-octahydro-2H-benzo[b]furo [2,3-a]quinazoline)-2,4′-pyrimidin-2′one

PBS

phosphate buffered saline

RX821002

2-(2-methoxy-1,4-benzodioxan-2-yl)-2-imidazoline

UK14304

5-bromo-6-(2-imidazolin-2-ylamino)-quinoxaline