Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

In most cellular systems, G protein-coupled receptors (GPCRs) desensitize due to prolonged agonist stimulation. An important mechanism of GPCR desensitization is receptor phosphorylation of the agonist-bound receptor by specific G protein-coupled receptor kinases, and the binding of the inhibitory protein, -arrestin to the phosphorylated receptor, which in turn blocks the interaction of receptor with G proteins. The binding of agonists to GPCRs often also initiates the internalization of receptors into the cell interior (Bogatkewitsch et al., 1996). The prevailing view is that -arrestin, bound to the receptor, binds with high affinity to clathrin, and immobilizes the receptors in clathrin-coated pits, which subsequently bud from the plasma membrane (Zhang et al., 1996, Goodman et al., 1996). The budding of the clathrin-coated pits is controlled by the monomeric GTPase dynamin (Zhang et al., 1996). Other plasma membrane receptors, which do not couple to G proteins such as transferrin receptors and low-density lipoprotein receptors, can also internalize in clathrin-coated vesicles (Van der Bliek et al., 1993). In addition to clathrin-mediated sequestration, GPCRs can sequester via alternative sequestration pathways (Pals-Rylaarsdam et al., 1997, de Weerd and Leeb-Lundberg, 1997, Vögler et al., 1998).

Recently, the monomeric GTPase RhoA has been reported to regulate clathrin-mediated sequestration of transferrin receptors. Overexpression of constitutively active RhoA, but not of wild-type RhoA, significantly inhibited the sequestration of transferrin receptors in HeLa cells, whereas RhoGDI, the GDP dissociation inhibitor of Rho, as well as Clostridium botulinum C3 transferase, which inactivates RhoA by ADP-ribosylation, enhanced transferrin receptor endocytosis in permeabilized A431 cells (Lamaze et al., 1996). Because muscarinic acetylcholine receptors (mAChRs) and other GPCRs rapidly stimulate the translocation of RhoA from the cytosol to the plasma membrane and thereby allow activation of RhoA by GDP/GTP exchange (Fleming et al., 1996, Keller et al., 1997), we investigated whether sequestration of GPCRs is regulated by RhoA. Recently, we and others have demonstrated that m1, m3, and m4 mAChRs, transfected in HEK-293 cells, sequester via dynamin-dependent clathrin-coated vesicles, and recycle back to the plasma membrane (Vögler et al., 1998, Tolbert and Lameh, 1996). In contrast, endocytosis of mAChRs of the m2 subtype in HEK-293 cells is fully dynamin-independent and almost completely irreversible, indicating that m2 mAChRs sequester by nonclathrin-coated vesicles (Pals-Rylaarsdam et al., 1997, Vögler et al., 1998). We therefore used m1 and m2 mAChRs expressed in HEK-293 cells as model systems of GPCRs possessing different sequestration pathways of GPCRs to analyze the role of RhoA in GPCR sequestration.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Materials. [³H]Quinuclidinyl benzilate ([³H]QNB, specific activity 43 Ci/mmol) and N-[³H]methylscopolamine ([³H]NMS, specific activity 84 Ci/mmol) were purchased from New England Nuclear (Boston, MA). Anti-RhoA monoclonal antibody (26C4) was obtained from Santa Cruz Biotechnology (Santa Cruz, CA).

Plasmid Construction. All recombinant DNA procedures were carried out following standard protocols. DNA encoding murine m1 mAChR (Shapiro et al., 1988) and porcine m2 mAChR (Peralta et al., 1987) were subcloned into pCD-PS expression vector. DNA encoding human RhoA, which was subcloned into pRK5, and myc-tagged C. botulinum C3 transferase supplied in pEF were kind gifts from Dr. Alan Hall.

Cell Culture and Transfection. HEK-293 tsA201 cells were grown in Dulbecco's modified Eagle's (DME)/F-12 medium supplemented with 10% fetal calf serum, penicillin G (100 U/ml), and streptomycin (100 µg/ml) in an atmosphere of 5% CO₂. Cells on 150-mm plates were transiently transfected with either 12.5 or 25 µg of pCD-PS containing m1 or m2 mAChR DNA, together with 50 µg of pRK5 containing RhoA wild-type DNA, myc-tagged C. botulinum C3 transferase DNA, or 50 µg of pRK5 (Van Koppen and Nathanson, 1990). Transfection efficiency was 10 to 20% as measured by -galactosidase assays on HEK-293 tsA201 cells cotransfected with 50 µg of pSV (Promega, Madison, WI) encoding -galactosidase and 25 µg of mAChR DNA in pCD-PS per 150-mm tissue culture plate.

Immunoblot Analysis of RhoA Expression A total of 48 to 60 h after DNA transfection, cells on 150-mm plates were washed twice with phosphate-buffered saline (PBS; 150 mM NaCl, 6.5 mM Na₂HPO₄, 2.7 mM KCl, pH 7.4) and lysed by the addition of 1.0 ml of boiling lysis buffer [1% sodium dodecyl sulfate (SDS), 10 mM Tris-HCl, pH 7.4]. Lysate was transferred to a microcentrifuge tube and boiled for 5 min. After five passages through a 25-gauge needle, samples were centrifuged for 5 min to remove insoluble material and diluted to an equal amount of protein as measured by the bicinchoninic acid method (Pierce, Rockford, IL) with lysis buffer. A total of 100 µl of electrophoresis sample buffer (250 mM Tris-HCl, pH 6.8, 4% SDS, 10% glycerol, 0.006% bromphenol blue, 2% 2-mercaptoethanol) was added to 100 µl of the diluted samples and boiled for another 5 min. After SDS-polyacrylamide gel electrophoresis on 15% polyacrylamide gels, protein was blotted onto nitrocellulose. Nitrocellulose was then blocked with 150 mM NaCl, 10 mM Tris-HCl, pH 7.5, and 5% bovine serum albumin (fraction V; Sigma Chemical Co., St. Louis, MO). After washing three times for 5 min in 150 mM NaCl, 10 mM Tris-HCl, pH 7.5, 0.1% Tween 20, the blot was incubated with anti-RhoA monoclonal antibody (0.2 µg/ml) in blocking buffer for 1 h. After three washes for 5 min in wash buffer, the blot was incubated with 0.2 µg/ml horseradish peroxidase-conjugated goat anti-mouse antibody (Dianova, Hamburg, Germany) at room temperature. After 1 h, the blot was washed again, and immunoreactivity was visualized by enhanced chemiluminescence (Amersham Buchler, Braunschweig, Germany).

Muscarinic Receptor Sequestration and Binding Assays. Briefly, 24 h after transfection, cells from 150-mm plates were replated on 24-well plates and allow to reattach and grow for another 24 h. The cells were then incubated with or without carbachol (10^-7, 10^-5, or 10^-3 M) for 0 to 60 min in 25 mM HEPES-buffered DME/F-12 medium (pH 7.4). Cells were washed with ice-cold 25 mM HEPES-buffered saline, pH 7.4, containing 113 mM NaCl, 6 mM glucose, 3 mM CaCl₂, 3 mM KCl, 2 mM MgSO₄, and 1.0 mM NaH₂PO₄, and for each manipulation, cells on four wells were incubated with 500 µl of 1 nM [³H]NMS in 500 µl of HEPES-buffered saline at 4°C to measure total binding. Two additional wells received also 3 µM atropine to measure nonspecific binding. After a 4-h incubation, cells were washed with ice-cold HEPES-buffered saline, solubilized in 1% Triton X-100, scraped, and transferred to scintillation vials, which received 3.5 ml of scintillation fluid before radioactivity counting. Sequestration is expressed as (1  quotient of cell surface receptors of carbachol-treated and untreated cells) x 100%. For determination of the equilibrium dissociation constant K_d of [³H]NMS, intact cells were incubated at 4°C for 12 h instead of 4 h (to attain binding equilibrium at low radioligand concentrations) with 10 to 1000 pM. [³H]NMS in the absence and presence of 3 µM atropine. For monitoring dissociation of [³H]NMS from m1 mAChRs, intact transfected cells on 24-well plates were first incubated with 1 nM [³H]NMS in the presence and absence of 3 µM atropine for 4 h at 4°C, followed by a brief rinse with ice-cold HEPES-buffered saline and adding 500 µl ice-cold HEPES-buffered saline containing 3 µM atropine to prevent reassociation of [³H]NMS. Total mAChR number was determined by binding of the membrane-permeable muscarinic antagonist [³H]QNB to crude cell homogenates at receptor-saturating concentrations of 600 pM (Van Koppen and Nathanson, 1990). Untransfected HEK-293 tsA201 cells do not express detectable levels of mAChRs as measured by [³H]QNB binding to total cell homogenates.

Phalloidin Staining of Actin Cytoskeleton. Twenty-four hours after transfection with green fluorescent protein (GFP)-encoding pEGFP-C1 (Clontech, Palo Alto, CA) together with RhoA pRK5 or control pRK5 (50 µg/150-mm plate each), HEK-293 tsA201 cells were replated and grown overnight on poly-L-lysine-coated 18 × 18-mm glass coverslips. The cells were incubated in 25 mM HEPES-buffered DME/F12 medium (pH 7.4) containing 5 µg/ml cytochalasin B (Sigma) or vehicle (dimethyl sulfoxide, final concentration of 0.2%) for 10 min at 37°C. Cells were rinsed twice with PBS and fixed with 3% paraformaldehyde in PBS for 15 min at room temperature. Then cells were washed twice with PBS for 5 min each and permeabilized in 0.05% Triton X-100 in PBS for 2 min. After washing three times for 5 min with PBS, cells were incubated for 15 min in 10 µg/ml tetramethylrhodamine isothiocyanate-conjugated phalloidin (Sigma) in PBS. After three washes with PBS, stained cells were mounted with Moviol (Calbiochem, San Diego, CA) and viewed by fluorescence microscopy using a Zeiss Axiovert microscope. Transfected cells were identified by green fluorescence of expressed GFP, and actin by tetramethylrhodamine isothiocyanate-phalloidin fluorescence using standard wavelength settings.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Inhibition of m1 and m2 mAChR Sequestration by RhoA. To test a regulatory role of RhoA in GPCR sequestration, we transiently coexpressed wild-type RhoA with mAChRs in HEK-293 tsA201 cells. Transfection with RhoA DNA (50 µg/150-mm plate) resulted in approximately 20- to 40 -fold overexpression of RhoA as measured by densitometry and corrected for a transfection efficiency of 10 to 20% in HEK-293 tsA201 cells (Vögler et al., 1998) (Fig. 1A). Sequestration of m1 mAChRs in HEK-293 tsA201 cells was inhibited dose dependently by transfecting cells with increasing amounts of RhoA DNA (Fig. 1B). Transfection with 5.0 µg of pRK5 RhoA DNA per 150-mm tissue culture plate reduced m1 mAChR sequestration in response to incubation with 1 mM carbachol for 60 min from 53 ± 3 to 36 ± 4%, and transfection with 50 and 150 µg pRK5 RhoA DNA diminished m1 mAChR sequestration to 21 ± 4 and 21 ± 6%, respectively. A time course of m1 mAChR internalization in control and RhoA-overexpressing cells is shown in Fig. 2 (left panel). Inhibition of m1 mAChR sequestration by RhoA was apparent at low concentrations of carbachol as well. For example, m1 mAChR sequestration induced by incubation with 10^-5 M carbachol for 60 min was reduced from 10 ± 5% in control cells to 2 ± 3% in RhoA-overexpressing cells (data not shown). To investigate whether RhoA also interferes with clathrin-independent GPCR sequestration, the effect on m2 mAChR sequestration was determined. As shown in Fig. 2 (right panel), transfection of HEK-293 cells with RhoA DNA reduced m2 mAChR sequestration in response to treatment with 1 mM carbachol for 60 min from 79 ± 3% to 29 ± 3%. Similar to m1 mAChR sequestration, RhoA overexpression reduced m2 mAChR sequestration at low carbachol concentrations as well. The m2 mAChR sequestration induced by incubation with 10^-5 M carbachol for 60 min was reduced from 65 ± 2% in control cells to 17 ± 3% in RhoA overexpressing cells (data not shown).

View larger version (18K): [in this window] [in a new window]   Fig. 1.   Overexpression of RhoA wild-type in HEK-293 tsA201 cells. Influence of RhoA expression on m1 mAChR sequestration is shown. A, detection of RhoA in total cell lysates of HEK-293 tsA201 cells transiently transfected with 50 µg of pRK5 (pRK5) or pRK5 RhoA (RhoA). Equal amounts of cell lysates (50 µg protein/lane) were resolved on polyacrylamide gels, transferred to nitrocellulose, and immunoblotted using anti-RhoA monoclonal antibody. B, HEK-293 tsA201 cells were transfected with 25 µg of m1 mAChR DNA in pCD-PS along with 150 µg of pRK5 or 0.5 to 150 µg of pRK5 RhoA per 150-mm tissue culture plate. Total DNA per plate was kept constant at 175 µg. Cells were treated with or without 1 mM carbachol for 60 min at 37°C, and cell surface receptor number was determined by specific [3H]NMS binding with a receptor-saturating concentration of 1 nM [3H]NMS. [3H]NMS binding to untreated cells (i.e., not preincubated with carbachol) transfected with 150 µg pRK5, 0.5, 5.0, 50, and 150 µg of pRK5 RhoA was 334 ± 61, 193 ± 47, 154 ± 54, 35 ± 21, and 8 ± 5 fmol/mg protein, respectively. The data represent the mean ± S.E.M. of three independent experiments.

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Time course of m1 and m2 mAChR sequestration in HEK-293 tsA201 cells. The effect of overexpressing RhoA wild-type is shown. HEK-293 tsA201 cells on 150-mm plates were transfected with 50 µg of pRK5 RhoA (RhoA) or pRK5 (pRK5) DNA along with 12.5 (m1) or 25 µg (m2) mAChR DNA. Forty-eight hours later, cells were incubated in the presence of 1 mM carbachol for 10, 30, or 60 min at 37°C. Thereafter, sequestration was assessed by [3H]NMS binding to intact cells at 4°C. Data are the mean ± S.E.M. from six sets of experiments each. Specific [3H]NMS binding to cells transfected with m1 mAChR DNA with and without pRK5 RhoA was 110 ± 13 and 219 ± 32 fmol/mg protein, respectively. Specific [3H]NMS binding to m2 mAChR-expressing cells transfected with and without pRK5 RhoA was 25 ± 4 and 76 ± 29 fmol/mg protein, respectively.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Inhibition of m1 and m2 mAChR sequestration by RhoA. Effect of C3 transferase overexpression is shown. HEK-293 tsA201 cells were transfected with 12.5 µg of m1 or 25 µg of m2 mAChR in pCD-PS along with pRK5 (pRK5), 50 µg of pRK5 RhoA DNA (RhoA), 50 µg of pEF C3 transferase (C3), or 50 µg of pRK5 RhoA + 50 µg of pEF C3 transferase (RhoA + C3) per 150-mm plate. Total DNA per plate was kept constant at 125 µg. Cells were treated with or without 1 mM carbachol for 60 min at 37°C. The data represent the mean ± S.E.M. of three independent experiments each. Specific [3H]NMS binding to m1 mAChR-expressing cells was 81 ± 12 (pRK5), 43 ± 10 (RhoA), 38 ± 9 (C3), and 108 ± 46 (RhoA + C3) fmol/mg protein, respectively. Specific [3H]NMS binding to m2 mAChR-expressing cells was 37 ± 10 (pRK5), 32 ± 8 (RhoA), 26 ± 9 (C3), and 58 ± 17 (RhoA + C3) fmol/mg protein, respectively.

View larger version (139K): [in this window] [in a new window]   Fig. 4.   Effect of cytochalasin B on actin cytoskeleton in control and RhoA-overexpressing HEK-293 tsA201 cells. HEK-293 tsA201 cells transiently transfected with GFP encoding pEGFP-C1, along with control pRK5 (A and B) or pRK5 RhoA (C and D), were incubated with either vehicle (0.2% dimethyl sulfoxide) (A and C) or 5 µg/ml cytochalasin B (B and D) for 10 min at 37°C. Actin was stained with 10 µg/ml tetramethylrhodamine isothiocyanate- (TRITC) conjugated phalloidin and viewed by fluorescence microscopy. Transfected cells that were identified by expressing GFP are marked by an arrow. Scale bar, 20 µM.

Inhibition of m1 and m2 mAChR Cell Surface Targetting by RhoA Overexpression. Rho GTPases have also been implicated in the intracellular movement of organelles (Murphy et al., 1996). We therefore investigated whether RhoA affects trafficking of receptors to the plasma membrane. As shown in Fig. 5, RhoA overexpression decreased the expression of m1 and m2 mAChRs on the cell surface by 60% and 31%, respectively, as determined by [³H]NMS binding to intact cells and [³H]QNB binding to cell homogenates in parallel (P < .01, t test). Whereas in control cells, 82 ± 13% of m1 and 80 ± 6% of m2 receptors were present at the cell surface, in RhoA-overexpressing cells, only 33 ± 5% of m1 and 55 ± 6% of m2 receptors were on the cell surface, respectively. In contrast, RhoA overexpression did hardly reduce total m1 and m2 receptor number (see legend of Fig. 5). Control experiments demonstrated that RhoA overexpression did not change 1) the equilibrium dissociation constant of [³H]NMS [K_d values of 126 ± 14 and 164 ± 30 PM in cells transfected with pRK5 RhoA and control pRK5, respectively (mean ± S.D., n = 2 independent transfection experiments)] or 2) the dissociation rate constant of [³H]NMS (k values of 0.03 ± 0.01 and 0.04 ± 0.01 min¹, respectively (mean ± S.D., n = 2 independent experiments). Thus, the reduced number of cell surface receptors in RhoA-overexpressing cells is not caused by reduced binding affinity of the receptor, or increased dissociation of [³H]NMS from the receptor during washing at the end of the [³H]NMS binding assay. Cotransfection with C3 transferase DNA blocked the effect of RhoA on m1 and m2 mAChR cell surface expression completely. Overexpression of C3 transferase alone did not change the subcellular distribution of m1 and m2 mAChRs in HEK-293 cells.

View larger version (15K): [in this window] [in a new window]   Fig. 5.   Inhibition of cell surface expression of m1 and m2 mAChRs by RhoA. HEK-293 tsA201 cells were transfected with either 25 µg m1 or m2 mAChR DNA in pCD-PS together with 100 µg of pRK5 (pRK5), 50 µg of pRK5 RhoA + 50 µg of pRK5 (RhoA), 50 µg of pEF C3 transferase + 50 µg of pRK5 (C3), or 50 µg of pRK5 RhoA + 50 µg of pEF C3 transferase (RhoA + C3) per 150-mm tissue culture plate. C3 transferase overexpression did not influence RhoA overexpression as verified by Western blotting. Cell surface receptor number was determined by specific [3H]NMS binding to intact cells on six-well plates. Total receptor number was measured by [3H]QNB binding to cell homogenates from six-well plates in parallel (m1 + pRK5: 883 ± 197; m1 + RhoA: 685 ± 256; m1 + C3: 411 ± 125; m1 + RhoA + C3: 596 ± 281; m2 + pRK5: 425 ± 34; m2 + RhoA: 340 ± 36; m2 + C3: 224 ± 18, and m2 + RhoA + C3: 240 ± 43 fmol/mg protein). The data represent the mean ± S.E.M. of 6 (m1) or 12 (m2) independent transfection experiments, except for "m2 + C3" and "m2 + RhoA + C3," which are the mean ± S.E.M. of 5 experiments. Cell surface expression of m1 and m2 mAChRs was statistically larger in pRK5-transfected cells than in pRK5 RhoA-transfected cells (P < .01, two-tailed t test). Total expression of m1 and m2 mAChRs in pRK5-transfected cells (pRK5) was not statistically different from that in pRK5 RhoA-transfected (RhoA) (P = .55 and = .10, respectively, two-tailed t test).

View larger version (15K): [in this window] [in a new window]   Fig. 6.   Sequestration of [3H]NMS occupied m1 mAChRs in HEK-293 tsA201 cells and the effect of RhoA overexpression are shown. Receptor-mediated uptake of [3H]NMS was determined by incubation of the cells, transiently expressing m1 mAChRs with or without RhoA, with 1 nM [3H]NMS (±3 µM atropine) at 37°C for 0 to 120 min. Thereafter, cells were washed with 50 mM glycine buffer, pH 3.0, containing 150 mM NaCl to remove plasma membrane-bound [3H]NMS. Uptake of [3H]NMS is expressed as a percentage of total specific [3H]NMS binding to plasma membrane receptors in control and RhoA-overexpressing cells (260 ± 22 and 61 ± 9 fmol/mg protein, respectively). The data represent the mean ± S.E.M. of four sets of experiments.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

Evidence implicating RhoA in the regulation of clathrin-mediated sequestration pathway has recently been presented by Schmid and coworkers (Lamaze et al., 1996). Constitutively active RhoA impedes sequestration of transferrin receptors, whereas inhibitors of RhoA activation (RhoGDI and C3 transferase) stimulate transferrin receptor internalization. In the present study, we demonstrated that overexpression of wild-type RhoA strongly inhibited agonist-induced sequestration of both m1 and m2 mAChRs, and that this effect was prevented by coexpression of C3 transferase. However, overexpression of C3 transferase alone had no influence on m1 and m2 mAChR sequestration. We therefore conclude that RhoA at physiological concentrations is not a regulator of mAChR sequestration, but when overexpressed, significantly inhibits clathrin-dependent and -independent mAChR sequestration.

In addition to inhibition of mAChR sequestration, we noted that RhoA overexpression also leads to intracellular accumulation of m1 mAChRs and m2 mAChRs, without significantly affecting total receptor expression, as determined by [³H]QNB binding to cell homogenates. As the intracellularly accumulated mAChRs retained their capacity to bind [³H]QNB, we conclude that they are probably correctly folded in intracellular vesicles. As the intracellular concentration of receptors can be considered to be an equilibrium between constitutive receptor internalization and transport of intracellular receptors to the plasma membrane, the intracellular accumulation of mAChRs could be the result of either enhanced constitutive receptor internalization or decreased transport of intracellular receptors to the plasma membrane. We observed that constitutive internalization of [³H]NMS-occupied m1 mAChRs was significantly inhibited by RhoA overexpression, suggesting that RhoA overexpression most likely inhibits transport of mAChRs to the plasma membrane rather than increases constitutive receptor internalization.

The mechanism by which RhoA inhibits agonist-induced mAChR sequestration in HEK-293 cells has yet to be identified. The decreased agonist-induced internalization in RhoA-overexpressing cells could be due to a diminished number of receptors at the plasma membrane. This explanation, however, seems very unlikely. The percentage of internalized m1 or m2 receptors as a result of agonist exposure is larger in HEK-293 tsA201 cells expressing low numbers than high numbers of cell surface receptors per cell (our unpublished observations; Pals-Rylaarsdam and Hosey, 1997). A more plausible (and unifying) mechanism is that RhoA inhibits mAChR internalization primarily by inhibiting receptor recycling to the plasma membrane, resulting in a loss of cell surface receptors and saturation of intracellular receptor pools, which in turn inhibits the flow of internalized receptors from the plasma membrane. However, we cannot exclude that RhoA overexpression inhibits receptor internalization as well as transport to the plasma membrane independently from each other.

The molecular mechanisms by which RhoA interferes with receptor trafficking remain to be elucidated. It is possible that overexpressed RhoA competitively inhibits one or more, as yet unknown, monomeric GTP-binding proteins, which regulate particular mAChR transport pathways. It is also possible that overexpressed RhoA acts via downstream effector enzymes, which change the phospholipid composition of the plasma membrane or vesicles, thereby altering the motility of receptors. RhoA is able to stimulate various lipid kinases, including phosphatidylinositol 4-phosphate 5-kinase, an enzyme that is essential for the production of phosphatidylinositol-4,5-bisphosphate, and phosphatidylinositol 3-kinase, producing phosphatidylinositol-3,4-bisphosphate, and phosphatidylinositol-3,4,5-trisphosphate (Machesky and Hall, 1996). By changing the phospholipid composition, RhoA may also indirectly regulate the activity of key proteins like dynamin, clathrin adaptor AP2 subunits and other proteins (such as those required for dynamin-independent GPCR sequestration), which are involved in receptor trafficking including sequestration and which activity is regulated by negatively charged phospholipids (Lin and Gilman, 1996, Gaidarov et al., 1996). Several lines of evidence suggest that phosphoinositides in specific intracellular locations can signal the recruitment or activation of proteins essential for vesicular transport (De Camilli et al., 1996). It is therefore also possible that overexpression of RhoA may have induced missorting of proteins necessary for vesicular trafficking by changing the phospholipid composition of the intracellular vesicles. Another mechanism of RhoA action may be the activation of the tyrosine kinases, p125 focal adhesion kinase and paxillin (Seckl et al., 1995), or the stimulation of the serine/threonine kinases, protein kinase N (Amano et al., 1996) and Rho-kinase (Amano et al., 1997). The availability of dominant-negative and constitutively active forms of these lipid and protein kinases will be helpful to identify the RhoA targets that interfere with mAChR trafficking in HEK-293 cells.

Of particular interest was the observation that [³H]NMS occupied m1 mAChRs sequester into the cell interior of HEK-293 cells. This suggests that not only agonist-activated but also antagonist-occupied mAChRs internalize into the cell interior. However, sequestration of the antagonist-occupied receptors was much slower than that of agonist-occupied receptors. For example, 41 and 50% of agonist-activated m1 mAChRs are endocytosed within 30 and 60 min, respectively, whereas about 14 and 22% of antagonist-occupied receptors sequester within the same time periods. These findings provide additional evidence for the recent notion that antagonist-occupied GPCRs, such as the cholecystokinin receptors (Roettger et al., 1997) and angiotensin II type 1 receptors (Conchon et al., 1994) can internalize. It is possible that m1 mAChR sequestration is stimulated by the binding of NMS in a similar way as by agonists. However, this is not likely because 1) NMS is an inverse antagonist which stabilizes the inactive state of m1 mAChRs, preventing it from interacting with G proteins (Jakubík et al., 1995) and 2) NMS does not induce receptor down-regulation, but on the contrary, may up-regulate receptor number during hours of antagonist incubation (Fukamauchi et al., 1993). The sequestration kinetics of NMS-occupied m1 mAChRs is remarkably similar to the kinetics of constitutive delivery of mAChRs to the surface of various cell types (Koenig and Edwardson, 1994, 1996). We therefore hold the view that [³H]NMS occupied m1 mAChRs sequester by a constitutive internalization process. In the absence of antagonist, this pathway is in equilibrium with constitutive delivery of intracellular mAChRs to the cell surface (i.e., as a result of de novo receptor synthesis in the endoplasmic reticulum or delivery from other internal stores), keeping receptor number at the plasma membrane at a steady-state level. After binding of NMS, receptor internalization is presumably slightly inhibited as prolonged incubation with NMS can result in receptor upregulation.

In summary, our study has demonstrated that RhoA is not a regulator of mAChR sequestration in HEK293 cells. However, when overexpressed, RhoA inhibits both clathrin-dependent and -independent mAChR sequestration, presumably by inhibiting recycling of mAChRs to the plasma membrane. The identification of the RhoA targets involved will provide new information on the mechanisms that regulate GPCR trafficking in HEK-293 cells.