Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

Serotonin (5-HT)_2A and 5-HT_2C receptors are closely related members of the seven transmembrane spanning (7-TMS) superfamily of receptors, which are best known for their capacity to signal via G proteins. In the central nervous system, these receptors are widely distributed and participate in regulating a large variety of physiological functions and behaviors, including appetite control, sleep, hallucinogenesis, schizophrenia, neuroendocrine secretions, and thermoregulation (Zifa and Fillion, 1992; Boess and Martin, 1994; Hoyer et al., 1994). Overall amino acid sequence identity between the 5-HT_2A and 5-HT_2C receptors is ~50% and is ~80% within the transmembrane spanning domains. Given such close sequence homology, it is not surprising that the pharmacological characteristics of these receptors are also quite similar with relatively few selective ligands available. Moreover, both of these receptors couple to the same cellular signal transduction pathways [phospholipase C (PLC) and phospholipase A₂] (Felder et al., 1990; Berg et al., 1996). In spite of these many similarities, important differences in cellular signaling between these two receptor systems are beginning to emerge (Berg et al., 1994, 1996; Roth et al., 1998; Grotewiel and Sanders-Bush, 1999).

Continuous or repeated activation of 7-TMS receptors by agonists is often associated with reduced cellular responsiveness (desensitization). Rapid (within minutes) desensitization of many 7-TMS receptors is mediated by receptor phosphorylation, which can be elicited by a variety of kinases, including receptor-specific protein kinases [e.g., G protein receptor kinases (GRKs) and/or "effector" kinases such as protein kinase A, PKC, and calcium/calmodulin (CaM)-dependent protein kinase. GRK-mediated receptor phosphorylation is enhanced by agonist occupancy of the receptor and leads to the binding of arrestin, which uncouples the receptor from G proteins and initiates receptor internalization processes. For some 7-TMS receptors, internalization (sequestration) participates in response desensitization as well as being involved in the subsequent recovery (resensitization) of receptor-mediated responses (Krupnick and Benovic, 1998). Receptor phosphorylation by effector kinases generally does not require agonist binding and does not promote arrestin binding, rather the phosphorylation alone seems to inhibit coupling of the receptor to G proteins (Lefkowitz et al., 1990; Bunemann et al., 1999). Consequently, agonist-mediated, rapid desensitization of 7-TMS receptor systems occurs through a variety of mechanisms, which may differ depending upon the receptor system (Freedman and Lefkowitz, 1996; Bunemann et al., 1999). Given that there are differences in signal transduction between 5-HT_2A and 5-HT_2C receptors, there may be differences in desensitization between these receptors as well.

Although there have been several investigations of desensitization of 5-HT_2A receptor system responsiveness (Ivins and Molinoff, 1991; Kagaya et al., 1993; Roth et al., 1995; Vouret-Craviari et al., 1995; Berry et al., 1996), there are relatively few studies with the 5-HT_2C receptor system (Boddeke et al., 1993; Akiyoshi et al., 1995). Furthermore, assessment of the similarities and differences in characteristics and mechanisms of desensitization between these two receptor systems is difficult given the considerable methodological differences used. Such differences include not only time of agonist exposure (minutes to hours to days) but also differences in the cells in which the receptors are expressed. As pointed out by Roth et al. (1998), the actions of agonists on 5-HT₂ receptor systems appear to depend upon the cellular milieu in which the receptors are expressed, which may differ in the type and/or quantity of signaling molecules present. An example of this complexity is the different roles of protein kinase C (PKC) in agonist-induced loss of responsiveness of 5-HT_2A and 5-HT_2C receptor systems expressed in different cell types (Roth et al., 1986; Kagaya et al., 1990; Boddeke et al., 1993; Akiyoshi et al., 1995; Roth et al., 1995; Vouret-Craviari et al., 1995). In only one study have the characteristics of desensitization been compared with 5-HT_2A and 5-HT_2C receptors expressed in the same cell background; however, mechanisms of desensitization were not examined (Briddon et al., 1998).

In this study we evaluated rapid, agonist-induced desensitization of both 5-HT_2A and 5-HT_2C receptor-mediated PLC activation with the receptors expressed in the same cell background and at equivalent receptor densities. We found that maximal desensitization of both receptor systems elicited by 5-HT occurred within a similar time frame, was dependent upon the degree of receptor occupancy, and recovered within minutes of agonist removal. Although rapid desensitization of both receptor systems appeared to be mediated by phosphorylation events, both PKC and CaM kinase II play roles in agonist-mediated desensitization of the 5-HT_2A, but not the 5-HT_2C, receptor system. In contrast, GRKs play a role in the loss of responsiveness of the 5-HT_2C, but not the 5-HT_2A, receptor system.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials. The following materials were purchased from commercial sources: myo-[³H]inositol, [³H]ketanserin (PerkinElmer Life Science Products, Boston, MA); [³H]mesulergine (Amersham Pharmacia Biotech, Chicago, IL); ketanserin, phorbol-12-myristate-13-acetate (PMA), staurosporine, okadaic acid, pertussis toxin, 5-HT HCl, KN-62, and W-7 (Sigma/RBI, Natick, MA); and bisindolylmaleimide, chelerythrine chloride, and hygromycin (Calbiochem, La Jolla, CA). Fetal bovine serum was from Gemini Bioproducts Inc. (Calabasas, CA). All other tissue culture reagents were purchased from Invitrogen (Carlsbad, CA). All other drugs and chemicals (reagent grade) were purchased from Sigma Chemical (St. Louis, MO). SB 206553 was kindly provided by Dr. Tom Blackburn (SmithKline Beecham, Harlow, UK).

Cell Culture. Chinese hamster ovary (CHO) cell lines that express stably human 5-HT_2A (CHO-5-HT_2A) or 5-HT_2C (CHO-5-HT_2C) receptors at ~100 to 300 fmol/mg of protein were used for these studies. Cells were maintained in -MEM supplemented with 5% FBS and 300 µg/ml hygromycin. Unless otherwise specified, for all experiments, cells were seeded into 15-cm or 24-well tissue culture plates at a density of 4 × 10⁴ cells/cm². After a 24-h plating period, cells were washed with Hanks' balanced salt solution and placed into Dulbecco's-MEM/F-12 (1:1) with 5 µg/ml insulin, 5 µg/ml transferrin, 30 nM selenium, 20 nM progesterone, and 100 µM putrescine (serum-free media). Cells were grown in serum-free media 24 h prior to experimentation.

Inositol Phosphate (IP) Accumulation Measurements. Cells were labeled with 1 µCi/ml myo-[³H]inositol in serum-free medium for 24 h. Total [³H]inositol phosphate accumulation (IP₁, IP₂, IP₃, collectively referred to as IP) in the presence of 20 mM LiCl in response to agonist stimulation for 10 min at 37°C was determined as previously described (Berg et al., 1994). To elicit desensitization, cells were exposed to agonist (in the absence of LiCl) for various periods of time (0-120 min pretreatment time). After this agonist exposure, responsiveness of the receptor system was assessed by measuring IP accumulation in response to application of a maximal concentration of 5-HT (100 µM for 5-HT_2A or 10 µM for 5-HT_2C receptors) for 10 min in the presence of 20 mM LiCl.

Cell Permeabilization. Cells were made permeable by electroporation with a BTX model ECM8300 electroporator (Genetronics Inc., San Diego, CA), which delivers square-wave pulses. For these experiments, cells were grown in T175 flasks in the presence of 5% FBS, harvested using Dulbecco's phosphate-buffered saline containing 5 mM EGTA, and washed once with standard iso-osmotic pulsing buffer (250 mM sucrose, 1 mM MgCl₂, and 10 mM potassium phosphate buffer, pH 7.4). Cells were resuspended in pulsing buffer at a density of 3.25 × 10⁶ cells/ml and 800 µl of cell suspension was placed in 4-mm gap cuvette. Cells were electropulsed twice for a duration of 500 ms at a field strength of 1500 V/cm. Under these conditions, more than 90% of the cells were permeable as determined by trypan blue dye exclusion. After a 15-min recovery period in pulsing buffer at room temperature, 80 to 90% of the cells were again able to exclude trypan blue. After electroporation, cells were plated into 12-well dishes at a density of 4 × 10⁴ cells/cm² in the presence of 5% FBS and the antagonists ketanserin (5-HT_2A) or SB206553 (5-HT_2C). Experiments were done 16 h later to allow cells to adhere and to complete myo-[³H]inositol labeling prior to measurement of receptor-mediated IP accumulation.

Data Analysis. For desensitization experiments, IP accumulation data were expressed as a percentage according to the following equation:

(1)

where control dpm = IP accumulation in response to 5-HT (100, µM 5-HT_2A; 10 µM, 5-HT_2C) in the absence of pretreatment; control basal dpm = IP accumulation in the absence of 5-HT and LiCl; experimental dpm = IP accumulation in response to 5-HT after various 5-HT pretreatment paradigms; and experimental basal dpm = IP accumulation in the absence of 5-HT and LiCl after various 5-HT pretreatment paradigms.

(2)

(3)

(4)

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

Characteristics of 5-HT_2A and 5-HT_2C Receptor-Mediated IP Accumulation

Incubation of CHO-5-HT_2A or CHO-5-HT_2C cells with 5-HT (10 min, 37°C) increased total IP accumulation by 557% above basal ± 203% and 520% above basal ± 87%, respectively. EC₅₀ values for 5-HT were 324 nM (pEC₅₀ = 6.49 ± 0.08) and 51 nM (pEC₅₀ = 7.29 ± 0.10) at 5-HT_2A and 5-HT_2C receptors, respectively (mean ± S.E.M., n = 3-4). Treatment of cells with the alkylating agent phenoxybenzamine (PBZ) (5-HT_2A: 300 nM PBZ, 15 min; 5-HT_2C: 2 µM PBZ, 30 min) decreased the maximal response to 5-HT by approximately 50% (239% above basal ± 134% for 5-HT_2A and 278% above basal ± 28% for 5-HT_2C) with no change in EC₅₀ (427 nM, pEC₅₀ = 6.37 ± 0.10 and 69 nM, pEC₅₀ = 7.16 ± 0.13, for 5-HT_2A and 5-HT_2C receptors, respectively; Fig. 1). These data demonstrate the absence of receptor reserve for 5-HT in both the CHO-5-HT_2A and CHO-5-HT_2C cell lines, and consequently the EC₅₀ values for 5-HT were used as approximations of the K_A (Kenakin, 1997) for occupancy calculations (eq. 3).

View larger version (19K): [in this window] [in a new window]   Fig. 1.   Characteristics of 5-HT2A and 5-HT2C receptor-mediated IP accumulation. CHO-5-HT2A cells were treated for 15 min with 300 nM PBZ or dimethyl sulfoxide (0.01%; vehicle) (A) and CHO-5-HT2C cells were treated with 2 µM PBZ or dimethyl sulfoxide (0.02%; vehicle) for 30 min (B) before measurement of arachidonic acid release and IP accumulation in response to various concentrations of 5-HT for 10 min. Data are expressed as the mean (±S.E.M.) maximal response from vehicle-treated cells in paired experiments and represent data from three or four experiments measured in triplicate. Individual concentration curves from each experiment were fit to eq. 1. Mean Emax and EC50 values are provided in the text. For CHO-5-HT2A cells, basal IP accumulation was 440 ± 8 and 478 ± 24 dpm in the presence of vehicle or PBZ, respectively. For CHO-5-HT2C cells, basal IP accumulation was 569 ± 138 and 532 ± 99 dpm in the presence of vehicle or PBZ, respectively.

Characteristics of 5-HT_2A and 5-HT_2C Receptor System Desensitization

Pretreatment with 5-HT produced rapid loss of responsiveness for both the 5-HT_2A and 5-HT_2C receptor systems (Fig. 2). At high receptor occupancy (>90%) during pretreatment, the k_des was the same (~0.4 min¹) for both receptor systems. However, although the k_des for the 5-HT_2C receptor system was directly correlated with fractional occupancy, the k_des for the 5-HT_2A system was not. Although the maximal loss of responsiveness was the same for both receptor systems (~65%), there was a difference in the level of receptor occupancy by 5-HT required to produce maximal desensitization (Table 1). The 5-HT_2C receptor system was more sensitive to the desensitizing stimulus, with maximal desensitization occurring at ~50% receptor occupancy. In contrast, maximal desensitization of the 5-HT_2A receptor system required at least 75% occupancy. Another interesting difference between the two receptor systems was the time-dependent "resensitization" of response that occurred after 20 min of the continued presence of agonist for only the 5-HT_2A receptor system when desensitization was elicited using a concentration of 5-HT (100 µM) to produce full receptor occupancy (>99%; Fig. 2). Because of this return of responsiveness, the maximal desensitization of the 5-HT_2A receptor system after 30 min of agonist exposure at high receptor occupancy (>99%) was actually less than that produced by some lower receptor occupancies (75 and 90%). Resensitization of the 5-HT_2C receptor system did not occur at similar occupancy levels (Fig. 2) nor did it occur when 100 µM 5-HT (99.99% occupancy) was used as the desensitizing stimulus, suggesting that the effect in the CHO-5-HT_2A cells was not due to nonspecific actions of a high concentration of 5-HT.

View larger version (29K): [in this window] [in a new window]   Fig. 2.   Time course of desensitization of 5-HT2A and 5-HT2C receptor-mediated IP accumulation in response to pretreatment with 5-HT. CHO-5-HT2A (A) and CHO-5-HT2C (B) cells were pretreated with the indicated concentrations of 5-HT from 0 to 120 min. After pretreatment, IP accumulation in response to a maximal concentration of 5-HT (5-HT2A, 100 µM; 5-HT2C, 10 µM) applied for 10 min was measured. Data are normalized to the percentage of the control 5-HT response for each experiment and represent the mean ± S.E.M. of three to five experiments. Data were fit to eq. 4 to determine maximal loss of response (Emax) and the rate constant for desensitization (kdes), which are presented in Table 1. 5-HT-stimulated IP accumulation averaged 222% above basal levels ± 24% (n = 10) and 439% above basal levels ± 29% (n = 18) for CHO-5-HT2A and CHO-5-HT2C cells, respectively.

As shown in Fig. 2, pretreatment with maximal concentrations of 5-HT produced only partial desensitization for both 5-HT_2A and 5-HT_2C receptor systems. The response to agonist that remained after maximal desensitization (50 and 40% for 5-HT_2A and 5-HT_2C receptor systems, respectively) was blocked completely in the presence of the antagonists ketanserin (10 µM, CHO-5-HT_2A) or mianserin (10 µM, CHO-5-HT_2C) (data not shown).

Characteristics of 5-HT_2A and 5-HT_2C Receptor System Recovery from Desensitization

Desensitization elicited by pretreatment with 5-HT at 50 and 100% receptor occupancy for 10 min began to reverse within 15 min after agonist removal for both receptor systems (Fig. 3, A and B). For 10-min desensitizing stimuli at 50% receptor occupancy, full recovery of the IP response for both receptor systems was observed at 15 min of agonist washout. However, after desensitization with 10 min pretreatment at full occupancy, the IP response recovered only partially. Interestingly, the recovery of the 5-HT_2A receptor system from the desensitized state elicited by full occupancy for 10 min appeared to decay over time. In other words, the 5-HT_2A receptor system appeared to redesensitize with time after washout of 5-HT such that at 30 to 60 min of recovery, the responsiveness of the system was the same as if there was no recovery period. In contrast, the partial recovery of the 5-HT_2C receptor system after a 10-min desensitizing stimulus at full receptor occupancy was stable for up to 1 h after agonist removal. When the pretreatment period with 5-HT at full receptor occupancy was extended to 60 min, the recovery of both receptor systems was abolished (Fig. 3, C and D).

View larger version (31K): [in this window] [in a new window]   Fig. 3.   Recovery of receptor-mediated IP accumulation after agonist washout. A and B, after pretreatment (10 min) with EC50 or maximal concentrations of 5-HT (~50 and 100% receptor occupancy), CHO-5-HT2A cells (A) or CHO-5-HT2C cells (B) were washed to remove agonist and further incubated in the absence of 5-HT for 0 to 60 min. IP accumulation was subsequently measured in response to a maximal concentration of 5-HT for 10 min. Data were analyzed using analysis of variance followed by the Newman-Keuls post hoc test. *p < 0.05 compared with nondesensitized control; p < 0.05 compared with desensitized, no-recovery point. C and D, CHO-5-HT2A (C) or CHO-5-HT2C (D) cells were pretreated with maximal concentrations of 5-HT for 10 or 60 min, washed, and IP accumulation in response to maximal 5-HT was measured immediately (0 min of recovery) or after 15 min of recovery. Data are normalized to the control (no pretreatment) response and represent the mean ± S.E.M. of three to five experiments. *p < 0.05 compared with desensitized control.

Lack of Heterologous Desensitization

Activation of the endogenous P_2Y purinergic receptor in CHO cells with a maximal concentration of ATP (100 µM, 10 min) increased IP accumulation by 72% above basal ± 2% in CHO-5-HT_2A cells and 153% above basal ± 4% in CHO-5-HT_2C cells (n = 3). Pretreatment with ATP had no effect on 5-HT-mediated IP accumulation in either cell line although responsiveness to ATP was completely abolished (homologous desensitization). In CHO-5-HT_2A cells, 5-HT-mediated IP accumulation was 360% above basal ± 14% in vehicle-treated cells versus 358% above basal ± 50% in cells pretreated with ATP (mean ± S.E.M., n = 4). In CHO-5-HT_2C cells, 5-HT-mediated IP accumulation was 509% above basal ± 2% in vehicle-treated cells versus 496% above basal ± 21% in cells pretreated with ATP (mean ± S.E.M., n = 3). Similarly, pretreatment of cells with maximal concentrations of 5-HT had no effect on ATP-mediated IP accumulation in either CHO-5-HT_2A cells or CHO-5-HT_2C cells. These data suggest that agonist-stimulated heterologous regulation does not occur between the purinergic receptor system and either the 5-HT_2A or 5-HT_2C receptor system in CHO cells.

Mechanisms of 5-HT_2A and 5-HT_2C Receptor System Desensitization

Inhibition of PKC. Phosphorylation is a common mechanism of desensitization for many receptor systems (Krupnick and Benovic, 1998; Bunemann et al., 1999) and in some cell systems phosphorylation by PKC has been implicated as a mediator of 5-HT_2A (Roth et al., 1986; Berry et al., 1996) and 5-HT_2C (Boddeke et al., 1993) receptor system desensitization. Acute treatment with the PKC activator PMA (15 min, 1 µM) significantly reduced 5-HT_2A, but not 5-HT_2C, receptor-mediated IP accumulation. In CHO-5-HT_2A cells, IP accumulation in response to the 5-HT_2A/2C partial agonist DOI (1 µM, 10 min) was 110% above basal ± 15% and 70% above basal ± 5% in the absence and presence of PMA (mean ± S.E.M., n = 3; p < 0.05), whereas in CHO-5-HT_2C cells IP accumulation in response to DOI was 250 ± 50 and 300 ± 50% in the absence and presence of PMA, respectively (mean ± S.E.M., n = 3; p > 0.05). Consistent with these data, prior treatment with the protein kinase inhibitors staurosporine or bisindolylmaleimide (1 µM, 5 min) enhanced by ~2-fold the 5-HT_2A receptor-mediated IP accumulation in response to DOI (120% above basal ± 5% versus 300% above basal ± 75%, vehicle and staurosporine, respectively; mean ± S.E.M., n = 3; p < 0.05) and to 5-HT (see legends of Figs. 4 and 5). In contrast to 5-HT_2A receptor-mediated IP accumulation, staurosporine treatment had no effect on 5-HT_2C receptor-mediated IP accumulation (see legend of Fig. 4).

View larger version (22K): [in this window] [in a new window]   Fig. 4.   Effect of inhibition of PKC on desensitization of 5-HT2A and 5-HT2C receptor-mediated IP accumulation. CHO-5-HT2A (A) or CHO-5-HT2C (B) cells were treated with the protein kinase inhibitor staurosporine (1 µM) or vehicle (0.01% dimethyl sulfoxide) for 5 min before pretreatment with a maximal concentration of 5-HT for 0 to 30 min. After pretreatment, IP accumulation in response to a maximal concentration of 5-HT (5-HT2A, 100 µM; 5-HT2C, 10 µM) for 10 min was measured. Data are presented as the percentage of control (no pretreatment) and represent the mean ± S.E.M. of four experiments. For CHO-5-HT2A cells, basal IP accumulation was 798 ± 99 and 739 ± 83 dpm for vehicle and staurosporine-treated cells, respectively. 5-HT-stimulated IP accumulation was 327% above basal ± 39% and 590% above basal ± 132% in the presence of vehicle and staurosporine, respectively. For CHO-5-HT2C cells, basal IP accumulation was 652 ± 29 and 650 ± 111 dpm for vehicle and staurosporine-treated cells, respectively. 5-HT-stimulated IP accumulation was 613% above basal ± 41% and 718% above basal ± 59% in the presence of vehicle and staurosporine, respectively. *p < 0.05 compared with vehicle.

View larger version (10K): [in this window] [in a new window]   Fig. 5.   Effect of PKC inhibitors and PKC down-regulation on resensitization of 5-HT2C receptor-mediated IP accumulation. CHO-5-HT2A cells were treated for 24 h with the phorbol ester PMA (1 µM) or vehicle (0.01% dimethyl sulfoxide) or with the selective PKC inhibitors bisindolylmaleimide (Bis, 1 µM) or chelerythrine chloride (Chel, 1 µM) for 5 min, before pretreatment with a maximal concentration of 5-HT (100 µM) for 10 min (maximal desensitization) or 30 min (resensitization). After pretreatment, IP accumulation in response to a maximal concentration of 5-HT (100 µM) for 10 min was measured. Data are presented as the percentage of resensitization (calculated from the percentage of decrease of the maximal desensitization) and represent the mean ± S.E.M. of four experiments. *p < 0.05 compared with vehicle.

Inhibition of Calmodulin and CaM Kinase II. Calcium/calmodulin-activated kinases have also been reported to participate in receptor system desensitization phenomena (Zamani and Bristow, 1996). As shown in Fig. 6, 5-HT-mediated desensitization in CHO-5-HT_2A cells was significantly reduced (by 50%) in the presence of either the calmodulin antagonist W-7 or the CaM kinase II inhibitor KN-62, indicating that a phosphorylation step by this calmodulin-dependent kinase is involved in 5-HT_2A receptor system desensitization. Furthermore, analysis of time course experiments, done with pretreatment with maximal concentrations of 5-HT, also showed a reduction in the maximal desensitization (by 40%), but no effect on the k_des (0.40 ± 0.07 versus 0.46 ± 0.16, n = 3) in the presence of KN-62 (1 µM). In contrast to effects on the 5-HT_2A receptor system, blockade of calmodulin or inhibition of CaM kinase II had no effect on receptor-mediated desensitization of IP accumulation in CHO-5-HT_2C cells (Fig. 6).

View larger version (14K): [in this window] [in a new window]   Fig. 6.   Effect of calmodulin antagonist W-7 and CaM kinase II inhibitor KN-62 desensitization of 5-HT2A and 5-HT2C receptor-mediated IP accumulation. CHO-5-HT2A (A) and CHO-5-HT2C (B) cells were incubated with vehicle (H2O or 0.01% dimethyl sulfoxide), W-7 (100 µM), or KN-62 (10 µM) for 5 min before pretreatment with a maximal concentration of 5-HT for 10 min. After pretreatment, IP accumulated in response to 5-HT (5-HT2A, 100 µM; 5-HT2C, 10 µM) for 10 min was measured. Data are presented as the percentage of control (vehicle, W-7 or KN-62 in the absence of 5-HT pretreatment) and represent the mean ± S.E.M. of four experiments. Because the presence of dimethyl sulfoxide or H2O had no effect on basal or 5-HT-stimulated IP levels or 5-HT-mediated desensitization in either cell line, these data were pooled for each experiment and are represented by a single column. For CHO-5-HT2A cells (A), basal IP accumulation was 714 ± 53 dpm and 5-HT-stimulated IP accumulation was 182% above basal ± 13%. The presence of KN-62 had no effect on basal or 5-HT stimulated IP levels, however, W-7 increased basal levels by 26% (901 ± 69 dpm) and consequently decreased 5-HT-stimulated IP levels (to 74% above W-7 basal ± 8%). Similarly, for CHO-5-HT2C cells (B), basal IP accumulation was 651 ± 75 dpm and 5-HT-stimulated IP accumulation was 154% above basal ± 11%. The presence of KN-62 had no effect on basal or 5-HT-stimulated IP levels; however, basal levels in the presence of W-7 increased by 40% (937 ± 112 dpm) and thus decreased 5-HT-stimulated IP levels (to 86% above W-7 basal ± 7%). *p < 0.05 compared with vehicle.

Inhibition of GRK. Figure 7 shows the effect of stable expression of a dominant-negative mutant of GRK2 (dnGRK2_K220R) on 5-HT-mediated IP accumulation in DNG-5-HT_2A and DNG-5-HT_2C cells. 5-HT-stimulated IP accumulation, normalized for receptor expression, was enhanced approximately 2-fold in DNG-5-HT_2C cells compared with that in CHO-5-HT_2C cells. In contrast, expression of dnGRK2_K220R did not alter the response to 5-HT_2A receptor activation (Fig. 7).

View larger version (13K): [in this window] [in a new window]   Fig. 7.   Effect of dnGRK2 on 5-HT2A and 5-HT2C receptor-mediated IP accumulation. IP accumulation in response to a maximal concentration of 5-HT (5-HT2A, 100 µM; 5-HT2C, 10 µM) for 10 min was measured in cells expressing the 5-HT2A or 5-HT2C receptor with (DNG-5-HT2A and DNG-5-HT2C) or without (CHO-5-HT2A and CHO-5-HT2C) coexpression of the dominant-negative mutant of GRK2 (dnGRK2K220R). Data are presented as the total dpm of IP accumulated per femtomole per milligram of protein of receptor and represent the mean ± S.E.M. of four experiments. Basal IP accumulation was 588 ± 83, 562 ± 63, 702 ± 24, and 751 ± 51 dpm for CHO-5-HT2A, DNG-5-HT2A, CHO-5-HT2C, and DNG-5-HT2C cells, respectively. 5-HT-stimulated IP accumulation was 1683 ± 155, 977 ± 60, 2263 ± 237, and 7297 ± 681 dpm for CHO-5-HT2A, DNG-5-HT2A, CHO-5-HT2C, and DNG-5-HT2C cells, respectively. *p < 0.05 compared with response in the absence of dnGRK2.

View larger version (20K): [in this window] [in a new window]   Fig. 8.   Effect of heparin on 5-HT2A and 5-HT2C receptor-mediated IP accumulation. CHO-5-HT2A and CHO-5-HT2C cells were electropermeabilized in the presence of 10 µM heparin or vehicle (pulsing buffer; control). 16 h later, cells were pretreated for 10 min with a maximal concentration of 5-HT (5-HT2A, 100 µM; 5-HT2C, 10 µM) or vehicle (0 pretreatment), and IP accumulation was measured in response to stimulation with a maximal concentration of 5-HT for 10 min. For stimulation experiments, data are presented as percentage above basal. For pretreatment-induced desensitization experiments, data are presented as percentage of the 5-HT stimulation response in the absence of pretreatment. Data represent the mean ± S.E.M. of six experiments. Basal IP accumulation was 441 ± 21 and 457 ± 16 dpm for vehicle and heparin-treated CHO-5-HT2A cells, respectively. Basal IP accumulation was 556 ± 21 and 482 ± 59 dpm for vehicle and heparin-treated CHO-5-HT2C cells, respectively. *p < 0.05 compared with vehicle.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

Although the 5-HT_2A and 5-HT_2C receptor systems are often considered to be indistinguishable, evidence suggesting fundamental differences between them is accumulating (Berg et al., 1994, 1996; Roth et al., 1998; Grotewiel and Sanders-Bush, 1999). Here, we studied characteristics and mechanisms of rapid desensitization of both receptor systems expressed at equivalent levels in the same cell background. Not surprisingly, we found similarities between the two receptor systems in the magnitude of maximal desensitization and in the rate constants for desensitization (k_des) at maximal receptor occupancy. Furthermore, both receptor systems recovered rapidly and to a similar extent after removal of the desensitizing stimulus. Finally, phosphorylation events appear to be involved in the desensitization of both receptor systems.

Given the high degree of sequence homology between the 5-HT_2A and 5-HT_2C receptors, similarities in signaling mechanisms are not surprising. However, there were major differences in the characteristics and mechanisms of rapid desensitization. Perhaps the first sign of a difference was in the occupancy-desensitization relationship. The 5-HT_2C receptor system was more sensitive to desensitization than was the 5-HT_2A receptor system. Similar results were reported for 5-HT_2A and 5-HT_2C receptors in SH-SY5Y neuroblastoma cells (Briddon et al., 1998). Furthermore, in our study the rate constants for pretreatment-induced desensitization were agonist concentration-dependent for the 5-HT_2C receptor system, but independent of 5-HT concentration for the 5-HT_2A receptor system, suggesting the possibility that mechanisms for desensitization differ between the two receptor systems.

PKC has been implicated in mechanisms of desensitization of several receptor systems (Francesconi and Duvoisin, 2000). Our results suggest that PKC is involved in rapid desensitization of the 5-HT_2A, but not the 5-HT_2C, receptor system. Neither the protein kinase inhibitor staurosporine nor PKC down-regulation altered the characteristics of desensitization elicited by 5-HT pretreatment in CHO-5-HT_2A or CHO-5-HT_2C cells, as we reported previously (Berg et al., 1998). However, 5-HT_2A, but not 5-HT_2C, receptor-mediated IP accumulation was enhanced in the presence of PKC inhibitors staurosporine and bisindolylmaleimide and by PKC down-regulation. This enhancement of 5-HT-stimulated IP accumulation in CHO-5-HT_2A cells suggests that negative feedback actions of PKC occur rapidly, within the 10-min responsiveness test interval in our paradigm. That the loss of responsiveness elicited by pretreatment with 5-HT was not altered by blockade of PKC indicates that there are additional mechanisms of desensitization for the 5-HT_2A receptor system.

Although phosphorylation of 5-HT_2A receptors has yet to be shown, the mechanism by which PKC reduces responsiveness of the 5-HT_2A receptor system could involve phosphorylation of the receptor itself. The human 5-HT_2A receptor has eight potential phosphorylation sites for PKC in regions of the receptor (four in third intracellular loop and four in the C-terminal tail) that are important for receptor-G protein coupling. Alternatively, PKC could alter the function of the signaling molecules coupled to the receptor. PKC activation decreases G11 stimulation of PLC (Filtz et al., 1999) and phosphorylation of Gi2 by PKC reduces receptor-mediated inhibition of adenylyl cyclase activity (Palaparti and Anand-Srivastava, 1998). Furthermore, PKC-mediated phosphorylation of PLC2 reduces its capacity to be activated by G protein (Cunningham et al., 1999). However, unless the signaling molecules activated by the 5-HT_2A receptor differ from those activated by the 5-HT_2C receptor in CHO cells, one would expect that a mechanism of PKC that targeted transducer or effector molecules associated with the 5-HT_2A receptor should also influence the responsiveness of the 5-HT_2C receptor system.

It would appear that the role played by PKC in desensitization of 5-HT_2A and 5-HT_2C receptor systems is dependent upon the cellular system studied and the response measured, as suggested by Roth et al. (1998). For example, Roth et al. (1995) reported that agonist-induced desensitization of 5-HT_2A receptor-mediated PLC activity in NIH 3T3 cells was mediated by PKC only between 2 and 6 h of agonist exposure. A similar time frame of PKC-mediated desensitization was found for 5-HT_2A agonist mobilization of [Ca²⁺]_i in C6 glioma cells (Kagaya et al., 1993). In contrast, Kagaya et al. (1990) describe an effect of PKC within minutes for desensitization of 5-HT_2A agonist-elicited [Ca²⁺]_i release in human platelets. Activation of PKC may also play a role in desensitization of 5-HT_2C-receptor-mediated outward calcium currents in A9 cells, although 5-HT_2C agonist-mediated desensitization required 24-h agonist exposure. However, PKC inhibitors and activators were without effect on desensitization of the calcium response to 5-HT_2C receptor activation in CHO cells (Akiyoshi et al., 1995), as we found here for 5-HT_2C-mediated IP accumulation. Such differences between cell systems underscore the need to study differences between receptor systems in the same cell background. It is likely that the differences in the actions of PKC between the 5-HT_2A and 5-HT_2C receptor systems in this study were due to differences in the receptors themselves because they were expressed in the same cell background.

Interestingly, various inhibitors of PKC as well as PKC down-regulation blocked the time-dependent resensitization of the 5-HT_2A receptor system in response to high receptor occupancy (>99%) by agonist. This suggests the involvement of PKC in the mechanism of resensitization of the 5-HT_2A receptor system. Recently, it has been shown that 5-HT_2A receptors can be rapidly internalized by an endosomal pathway (Berry et al., 1996). After internalization, many receptors can rapidly recycle back to the cell surface in the continued presence of agonist (Morrison et al., 1996). Furthermore, Shih and Malbon (1996) have shown that PKC is required for recycling of 2 adrenergic receptors to the plasma membrane after agonist-promoted sequestration. Although rapid recycling of 5-HT_2A receptors to the cell surface has yet to be reported, it is possible that the resensitization process that occurs in CHO cells for the 5-HT_2A receptor system at high receptor occupancy is due to PKC-dependent stimulation of receptor recycling back to the cell surface.

The CaM kinase II inhibitor KN-62 and the calmodulin antagonist W-7 reduced 5-HT_2A, but not 5-HT_2C, receptor-mediated desensitization in response to pretreatment with 5-HT. Kagaya et al. (1993) reported that W7 blocked 5-HT-mediated desensitization of 5-HT_2A receptor-mediated increases in [Ca²⁺]_i in C6 glioma cells, also suggesting a role for calmodulin. CaM kinase II can directly phosphorylate receptors, leading to decreased responsiveness (Mestek et al., 1995; Koch et al., 1997). Although direct phosphorylation of 5-HT_2A receptors has yet to be shown, the 5-HT_2A receptor has one site in the second and two sites in the third intracellular loop that are potential targets for CaM kinase II phosphorylation.

Even though the effector kinases PKC and CaM kinase II appear to play roles in desensitization of the 5-HT_2A receptor system, activation of the endogenous purinergic receptor system in CHO cells did not alter the responsiveness of the 5-HT_2A receptor system. This was somewhat surprising because stimulation of the CHO purinergic receptor activates PLC (Berg et al., 1996, 1999) and thus presumably activates PKC and CaM kinase II. Furthermore, agonist activation of either the 5-HT_2A or 5-HT_2C receptor did not desensitize the purinergic receptor system. The lack of agonist-induced heterologous desensitization between the 5-HT_2A/2C and purinergic receptor systems in CHO cells may indicate that the signal transduction cascades coupled to these receptors are compartmentalized. We have previously reported that constitutive, agonist-independent 5-HT_2C receptor activity in CHO cells reduces responsiveness of the purinergic receptor-mediated IP accumulation (Berg et al., 1999). Perhaps the long-term signaling mechanisms elicited by constitutive receptor activity (which can lead to heterologous desensitization) may differ from those caused by short-term agonist activation.

A role for GRK in desensitization of the 5-HT_2C, but not 5-HT_2A, receptor system is suggested because blockade of GRK by stable expression of dnGRK2_K220R or treatment with heparin enhanced 5-HT-stimulated IP accumulation during the 10-min responsiveness test in CHO-5-HT_2C, but not CHO-5-HT_2A, cells. This suggests that the inhibitory effect of GRK occurs very rapidly, within the 10-min test interval. Heparin treatment, however, did not block desensitization elicited by pretreatment with 5-HT, suggesting that there are additional mechanisms involved in desensitization of 5-HT_2C receptor-mediated IP accumulation.

GRK2 contains a regulator of G protein signaling (RGS) homology domain in its N-terminal region (Siderovski et al., 1996) and can function as a GTPase-activating protein for Gq G proteins (Carman et al., 1999; Sallese et al., 2000; Usui et al., 2000). By increasing the rate of GTP hydrolysis by the G protein  subunit, GTPase-activating proteins decrease signaling. Interestingly, Sallese et al. (2000) have reported that expression of dnGRK2_K220R reduces 5-HT_2C receptor-mediated PLC activation in human embryonic kidney 293 cells. However, we found in CHO cells that expression of dnGRK2_K220R enhanced the responsiveness of the 5-HT_2C receptor system and had no effect on IP accumulation in CHO-5-HT_2A cells. This is contrary to what one would expect if the GRK kinase-deficient mutant could act as an RGS protein on G proteins expressed in CHO cells. Although the CHO cells used here express G11, we do not detect the presence of Gq by using selective antibodies (K. A. Berg, J. D. Cropper, and W. P. Clarke, unpublished observations). Perhaps the RGS binding domain of dnGRK2_K220R may distinguish between different members of the Gq/11 family, and the differences between our findings and those of Sallese et al. (2000) may be due to differences in G protein  subtype expression between the different cell types.

In conclusion, although there are some similarities in the characteristics of agonist-induced desensitization between the 5-HT_2A and 5-HT_2C receptor systems, there are major differences in desensitization mechanisms. The 5-HT_2A receptor system is less sensitive to agonist-induced desensitization than is the 5-HT_2C receptor system. In addition, the 5-HT_2A, but not the 5-HT_2C, receptor system undergoes a PKC-dependent resensitization at high receptor occupancy. Both PKC and CaM kinase II play roles in agonist-mediated desensitization of the 5-HT_2A receptor system, whereas neither effector kinase is involved with 5-HT_2C desensitization. In contrast, GRKs play a partial role in the loss of responsiveness of the 5-HT_2C, but not the 5-HT_2A, receptor system. These differences in desensitization further underscore the differences in signaling between these two highly homologous receptor systems. Perhaps it may be possible to exploit these differences to develop drugs to selectively alter 5-HT_2A or 5-HT_2C receptor signaling.