PRIORITY PAPER A pool of Y2 neuropeptide Y receptors activated by modifiers of membrane sulfhydryl or cholesterol balance Steven L. Parker 1 , Michael S. Parker 2 , Justin K. Kane 1 and Magnus M. Berglund 3 1 Department of Pharmacology, University of Tennessee College of Medicine, Memphis, TN, USA; 2 Department of Microbiology and Molecular Cell Sciences, University of Memphis, TN, USA; 3 Unit of Pharmacology, Department of Neuroscience, University of Uppsala, Sweden The cloned guinea-pig Y2 neuropeptide Y (NPY) receptors expressed in Chinese hamster ovary (CHO) cells, as well as the Y2 receptors natively expressed in rat forebrain, are distributed in two populations. A smaller population that is readily accessed by agonist peptides on the surface of intact cells constitutes less than 30% of Y2 receptors detected in particulates after cell homogenization. A much larger frac- tion of cell surface Y2 sites can be activated by sulfhydryl modifiers. A fast and large activation of these masked or cryptic sites could be obtained with membrane-permeating, vicinal cysteine-bridging arsenical phenylarsine oxide. A lower activation is effected by N-ethylmaleimide, an alkyla- tor that slowly penetrates lipid bilayers. The restricted-access alkylator, 2-[(trimethylammonium)ethyl]methanethiosulfo- nate, was not effective in unmasking these sites. Some of the hidden cell surface Y2 sites could be activated by polyene filipin III through complexing of membrane cholesterol. The results are consistent with the presence of a large Y2 reserve in a compartment that can be accessed by alteration of sulfhydryl balance or fluidity of the cell membrane, and by treatments that affect the anchoring and aggregation of membrane proteins. Keywords: receptor sequestration; receptor reserve; receptor signaling; receptor masking. Synaptic discharge of many neurotransmitters produces concentrations of these receptor agonists that saturate the respective binding sites, with a potential for prolonged and excessive signaling. With receptors characterized by high binding affinities, which represent a large fraction of rhodopsin-related neurotransmitter receptors, it may not be possible to adequately constrain the signaling by dissociation of the agonist. For neuropeptide receptors, a paracrine regulation via secretion of specific peptidases would meet large difficulties in both the selectivity and the economy of action. Scavenging by cell membrane- resident ectoproteinases by way of in situ encounters with extra- cellular agonists may not satisfy the clearance needs created by agonist discharge. A much more selective (and potentially quicker) regulation could be provided by sequestration or internalization of the receptor–ligand complex and further intramembrane or intracellular processing (reviewed in [1,2]). This could be accomplished by recycling sequestration (e.g. the m1 muscarinic receptor [3]), by recycling internalization (e.g. the m2 muscarinic receptor [4] or the neuropeptide Y (NPY) Y1 receptor [5]), and by lysosome-linked disposing internalization (e.g. the endothelin-B receptor [6]), all possibly enacted in relation to the prevailing levels of the respective agonists and the extent of preservation of the respective receptor molecules. Among neuropeptide transmitters, large levels of NPY are present in many areas of the forebrain [7], enabling an important regulation of feeding [8]. The forebrain NPY receptors include all principal Y receptor types [9], with Y1 and Y2 receptors detected at largest levels [10]. The slowly internalizing Y5 receptors [11] could represent a substantial component of sustained feeding regulation by NPY. However, both the Y5 and the feeding-coregulative [8] Y1 receptors (which could be strongly driven to internalize even by picomolar concentrations of NPY [5,12]) might be overwhelmed by large NPY release, in view of high nanomolar levels of the peptide in rodent [7] and even in human forebrain locations [13]. The discharge overloads could be handled through participation of another NPY receptor, the Y2 receptor. The Y2 receptor is strongly expressed especially in hypothalamic areas [10], and exists in two affinity states, one of which shows a very high binding affinity and is linked to a large degree of receptor aggregation [14]. The Y2 receptor is also distinguished by a low rate of internalization compared to the Y1 receptor when expressed in CHO cells [12]. A large portion of the Y2 complement is not detected on membranes of intact cells, but becomes accessible to agonist peptides upon cell homogenization, or upon treatment with a membrane-penetrating crosslinker of Correspondence to S. L. Parker, Department of Pharmacology, University of Tennessee College of Medicine, Memphis, TN 38163, USA. Tel.: + 1 901 850 7617, E-mail: StevenLeonardParker@msn.com Abbreviations: NPY, neuropeptide Y; hNPY, human/rat NPY; hPYY(3–36), human peptide YY(3–36); PYY, peptide YY; NEM, N-ethylmaleimide; MTSET, 2-[(trimethylammonium)ethyl]methane- thiosulfonate bromide; PAO, phenylarsine oxide. (Received 22 February 2002, revised 18 March 2002, accepted 22 March 2002) Eur. J. Biochem. 269, 2315–2322 (2002) Ó FEBS 2002 doi:10.1046/j.1432-1033.2002.02903.x vicinal cysteines, phenylarsine oxide (PAO) [12]. This study presents evidence for activation of these sites by agents that affect the membrane sulfhydryl balance or cholesterol, and also the rate of internalization of the Y2 receptor. MATERIALS AND METHODS Chemicals The Y peptides hPYY(3–36) and hNPY were obtained from the American Peptide Company (San Diego, CA, USA). Filipin III, N-ethylmaleimide (NEM) and phenylarsine oxide were purchased from Sigma (St Louis, MO, USA). Cholesteryl hemisuccinate was obtained from Calbiochem (La Jolla, CA, USA). Filipin III and PAO were dissolved in dimethylsulfoxide and stored in aliquots at )80 °C. Filipin complex (Sigma; a mixture of three isomers of filipin) was only about 25% as active as filipin III, and hence was not used. Cholesteryl hemisuccinate was prepared as a water emulsionandalsostoredat)80 °C. Restricted-access methanethiosulfonate MTSET (2-[(trimethylammo- nium)ethyl]methanethiosulfonate bromide) was obtained from Toronto Research Chemicals (North York, Ontario, Canada) Solutions of this agent and of NEM were made within 15 min before use. Labeled peptides All iodinations of Y peptides were performed as described previously [15]. The radioactive Y peptides were 75–90% monoiodinated and had specific activities in the range of 1500–1800 CiÆmmol )1 (70–80% theoretical), as deduced by comparison in saturation assays with HPLC-purified monoiodinated 125 I-labeled Y peptides hNPY and hPYY(3–36) supplied by PerkinElmer/NEN, Cambridge, MA, USA (specific activity 2170 CiÆmmol )1 ). Cell cultures and labeling All cell types were cultured in F12/D-MEM medium (Gibco, Long Island, NY, USA) at 250 lgÆmL )1 of geneticin and 2m M GlutaMax1 (Gibco). The guinea-pig Y1 receptor (gpY1-CHO; [16]); and the guinea-pig Y2 receptor (gpY2- CHO; [17]) were expressed in Chinese hamster ovary (CHO) cells. The cell lines used in this study had stable particulate receptor density, at the level of 10–20 fmol per 100 000 cells, over up to 40 passages in F12/D-MEM at 250 lgÆmL )1 of geneticin (Gibco). At full confluence, which was reached within 48 h with seeds of 25 000 cells per cm 2 at 37 °C in 95% O 2 /5% CO 2 , the cell count was 200 000–235 000 per cm 2 . After four washes with OptiMemÒ medium (Gibco, Long Island, NY, USA) to remove any external proteinase activity, the experimental incubations were car- ried in 48-well (0.8 cm 2 Æwell )1 ) plates, in a volume of 0.25 mL, using OptiMemÒ medium. The labeling of Y peptides was carried out at 50 p M of 125 I-labeled peptide tyrosine, using 300 n M nonlabeled peptides for nonsaturat- ing (or nonspecific) binding correction. In all experiments, less than 15% of the labeled peptides were degraded over the incubation period, as verified by Bio-Gel P-4 chromatogra- phy [15]. The incubations were terminated by the removal of the medium by suction, two washes with cold Opti-Mem buffer, and extraction for 12 min at 0–4 °C with ice-cold 0.2 M CH 3 COOH/0.5 M NaCl (pH 2.6), which was found to quantitatively dissociate the cell-surface attached Y peptides, without significant extraction of internalized peptides [12,18]. The binding of Y peptides to particulates from gpY2-CHO cell or rat forebrain tissue particulates was, on the other hand, more than 90% extracted by cold acid saline, as expected from the known importance of the arginine residues of NPY in Y2 (and Y1) receptor binding [19]. Rat forebrain tissue (a pool of hypothalamic and piriform cortex slices) was diced by scissors into fragments  1-mm in diameter, and then dispersed in 0.14 M NaCl/0.01 M Na phosphate/0.001 M EDTA (NaCl/P i /EDTA buffer; final pH 7.4) at about 50 mg fresh tissue per ml, using five slow passages through a 12-gauge needle. The suspension was brought to 5 FALGPA units per mL (5 lmol of furyl- acroyl-Leu-Gly-Pro-Ala hydrolyzed per min at pH 7.5 and 25 °C) of Sigma collagenase and incubated for 60 min at 23–24 °C with five slow passages through 16-gauge and 18- gauge needles repeated at 10-min intervals. The suspension was then filtered through a 100-mesh sieve, and the filtrate sedimented for 5 min at 100 g to recover the cells. The cells (>95% excluding Trypan blue stain) were resuspended in Opti-Mem buffer, and  200 000 cells per well were imme- diately plated for experimental incubations. The density of Y2 receptors did not significantly change in the above dissociation procedure and the additional incubation of up to 2 h at 37 °C. At the end of the incubation period, the cells were brought to 0–4 °C, harvested by sedimentation at 100 g for 5 min and surface-washed with cold Opti-Mem buffer. The pellets were then dispersed in ice-cold 0.2 M CH 3 COOH/0.5 M NaCl, and after 12 min at 0–4 °C sedimented for 5 min at 4000 g to separate the extracted (originally cell-surface attached) and the residual (internal- ized) radioactive peptide. Receptor characterization The homogenization or NPY receptor assay buffer con- tained 8% sucrose, 0.2% proteinase-free BSA (Sigma), 0.025% bacitracin, 1 m M diisopropylfluorophosphate (Sigma), 4 m M CaCl 2 ,2m M MgCl 2 ,20m M hepes.NaOH (pH 7.4) and 50 l M ATP. Particulates from gpY2-CHO cells or from dispersed rat forebrain cells were isolated by homogenization in the cold NPY receptor assay buffer, applying 12 complete strokes of a Teflon pestle (clearance 0.10 mm) in a Potter-Elvehjem homogenizer at about 800 r.p.m., followed by the removal of debris for 5 min at 100 g, and the sedimentation of particulates for 15 min at 16 000 g (all at 0–4 °C). The pelleted particulates were aliquoted and kept at )80 °C. Particulate receptors were assayed as described previously [20]. The particle concen- tration was 100–125 lgÆmL )1 , the assay volume was 0.2 mL, and the incubation time was 90 min at 23–24 °C, with the appropriate competitors or inhibitors. The assay was terminated by centrifugation for 15 min at 12 000 g at 4 °C, the supernatants were discarded, and the pellets surface-washed by cold assay buffer prior to counting in a gamma-scintillation counter. The binding properties of the cell-surface receptors were characterized on monolayer cultures in Opti-Mem medium. Iodinated Y peptides were input at 50 p M , and competed by up to nine concentrations of homologous or isologous peptides in the range of 2316 S. L. Parker et al. (Eur. J. Biochem. 269) Ó FEBS 2002 3 · 10 )11 to 1 · 10 )6 M . Polyethyleneglycol precipitation of particulates was carried out as described previously [15]. Data evaluation Binding parameter calculations were carried out in the LIGAND program [21]. Multiple comparisons following a positive ANOVA were done in Tukey’s t-test [22]. RESULTS Activation of cell surface Y2 sites by phenylarsine oxide PAO inhibited the binding of Y2-selective agonist hPYY (3–36) to particulates from gpY2-expressing cells or rat forebrain only at concentrations above 1 m M (Table 1). Pretreatment with the arsenical at 100 l M decreased the affinity of the Y2 binding to CHO cell or forebrain particulates by not more than 40% (Table 2). However, PAO consistently increased the Y2 agonist binding to CHO cell monolayers by fourfold to fivefold (Fig. 1A), as previously shown [12]. This activation occurred with little change in Y2 site affinity (Fig. 1A). The binding of the Y2 agonist to dispersed rat forebrain cells was also increased by PAO up to fourfold (Fig. 1B), without a significant change in affinity (Fig. 1B). The activation by PAO saturated with increasing concentration of the arsenical between 10 and 30 l M with either the Y2-CHO cells (Fig. 1A), or with rat forebrain cells (Fig. 1B). Pretreatment of gpY1-CHO cells at 24 or 37 °C with 30 l M PAO did not produce significant change in subsequent surface labeling by 125 I-labeled hNPY or 125 I-labeled (Leu31,Pro34)hPYY (data not shown). The activation of Y2 sites on intact cells by PAO did not involve more than half of the receptors detectable in particulates after cell homogenization (Fig. 1A). Homoge- nization of gpY2-CHO cells alone resulted in a large activation of Y2 sites, to about twice the level measured after treatment of cell monolayers by PAO, and exceeding the monolayer control binding by a factor of at least four in the absence of PAO treatment. Pretreatment or cotreatment of either the cell monolayers or the isolated particulates by PAO (at 30 l M ) did not produce a consistent increase in particulate Y2 binding relative to control preparations (Fig. 1A). Likewise, the Y2 binding to particulates isolated from homogenates of dispersed rat forebrain cells was increased more than sixfold relative to whole cells, but not Table 1. Inhibition of Y2 receptor binding by the agents used. Particu- lates from gpY2-CHO cells or rat piriform cortex were labeled over 90 min at 23–24 °C by 50 p M 125 I-labeled hPYY(3–36) in the presence of 10–12 different concentrations of the respective agents, between 0.01 and 10 m M . The results are averages of three separate experiments, shown ± SEM. The assay conditions are specified in the Methods section. The nonspecific binding was defined at 300 n M of nonlabeled hPYY(3–36). Percent of the total specific binding displaced at 10 m M of an agent is shown in parenthesis after the corresponding K I value. K I ,m M Agent guinea pig Y2-CHO rat forebrain Dithiothreitol 3.37 ± 0.66 (46%) 3.35 ± 0.44 (34%) Phenylarsine oxide 2.24 ± 0.75 (78%) 1.51 ± 0.34 (94%) N-Ethylmaleimide 0.146 ± 0.014 (75%) 0.708 ± 0.104 (73%) MTSET 0.264 ± 0.027 (72%) 1.86 ± 0.14 (65%) Filipin III > 0.1 > 0.1 Table 2. Effect of pretreatment with various sulfhydryl-active agents on theaffinityofY2binding. The K d values are in p M hPYY(3–36), ± SEM. The particulates were preincubated for 30 min at 24 °C in the assay buffer in presence of the indicated molarities of SH-active agents, then sedimented to remove the agents, surface-washed and assayed (see Materials and methods). Data represent triplicate competition assays employing 8–10 concentration points in the range of 10–30 000 p M nonlabeled Y2 agonist, and 50 p M 125 I-labeled hPYY(3–36). Defining the nonspecific binding at 100 n M nonlabeled hPYY(3–36), more than 90% displacement of 125 I-labeled hPYY(3–36) was observed at 30 n M unlabeled hPYY(3–36) after any treatment. Note that the Y2 binding inbothCHOandratforebraincellscouldberesolvedintotwosig- nificant components with K d values of 5–15 and 300–700 p M , respectively [14]. K d ,p M Agent and molarity used gpY2-CHO cells Rat forebrain cells Control 422 ± 40 316 ± 43 Dithiothreitol (1 m M ) 541 ± 70 Phenylarsine oxide (100 l M ) 627 ± 70 334 ± 52 N-Ethylmaleimide (30 l M ) 601 ± 72 MTSET (100 l M ) 637 ± 46 Fig. 1. Effects of phenylarsine oxide (PAO) on the binding of Y2 agonist 125 I-labeled hPYY (3–36) to gpY2-CHO monolayers or particulates, and to dispersed rat forebrain cells. The labeled Y2 agonist was input at 50 p M in all cases. For assay details see Methods. All data, shown ± SEM, are averages of three experiments. Asterisks indicate differ- ences significant vs. the control binding in the Tukey t-test at the level of 95% (*) or 99% (**) confidence. (A) The surface binding to monolayers or particulates from gpY2-CHO cells as related to pre- treatment (30 min at 37 °C) of monolayers with PAO in the range of 3–100 l M , and pretreatment or cotreatment of particulates with 30 l M of the arsenical. Competition by nonlabeled hPYY(3–36) showed K d of 442 ± 37 p M for control surface binding, and of 638 ± 74 p M for surface binding after pretreatment by 30 l M PAO (n ¼ 3). (B) The surface binding to dispersed rat forebrain cells after pretreatment (30 min at 37 °C) with 3–100 l M PAO,andalsothebindingtopar- ticulates from cells pretreated for 30 min at 37 °C with 30 l M PAO. Competition by nonlabeled hPYY(3–36) showed K d of 407 ± 38 p M for control surface binding, and of 451 ± 24 p M for surface binding after pretreatment by 30 l M PAO (n ¼ 3). Ó FEBS 2002 Masked Y2 NPY receptors (Eur. J. Biochem. 269) 2317 further augmented significantly by a pretreatment with 30 l M PAO (Fig. 1B). Theeffectof30l M PAO could be largely suppressed by 100 l M sulfhydryl protector dithiothreitol, and was com- pletely prevented by 1 m M dithiothreitol (Fig. 2). Dithio- threitol reduced the binding of hPYY(3–36) to a fraction (<50%) of particulate Y2 sites at a K I value of about 3.5 m M with either the gpY2 or the rat forebrain Y2 receptor (Table 1). However, dithiothreitol did not affect the binding to cell surface sites at up to 1 m M (Fig. 2). Pretreatment with dithiothreitol at up to 1 m M did not affect gpY2 internalization relative to controls, and also neutralized any decrease in receptor- linked Y2 ligand internalization by PAO (Fig. 2). Effects of the alkylators NEM and MTSET Effects of alkylators on availability of Y2 sites were examined only with gpY2 receptor expressed in CHO cells. NEM, an alkylator slowly penetrating the lipid bilayer [23], also significantly activated the monolayer Y2 sites. The increase appeared to saturate between 10 and 100 l M (Fig. 3), and was followed by a significant decrease at 300 l M , reflecting inactivation of a fraction of the Y2 sites at high concentrations of the alkylator [14]. The restricted- access alkylator MTSET [24] did not produce a significant stimulation of the gpY2 surface binding at 10–100 l M ,and was inhibitory at 300 l M (Fig. 3). Both alkylators essen- tially prevented internalization of hPYY(3–36) at 300 l M , and their activity at that concentration was fully neutralized by 1 m M dithiothreitol (Fig. 3); a highly selective blockade of internalization by NEM was also found for the gpY1- CHO receptor (data not shown). Activation of Y2 sites by filipin III Possible effects of the cholesterol-complexing polyene filipin III on the availability of cell-surface Y2 receptors were also assessed in gpY2-CHO cells. A substantial increase in surface Y2 binding could be shown, dependent on concen- tration of the antibiotic. The increase reached about twice the control level at 3 l M was somewhat reduced at 10 l M , and then dropped to almost the control levels at 30 l M (Fig. 4). In the same set of experiments, the increase in surface Y2 binding at 10 l M PAO was, as routinely observed, close to five times the control value (Fig. 4). Stimulation of the binding by filipin was largely prevented by equimolar cholesterol, and was essentially cancelled at a cholesterol concentration three times in excess to that of filipin (Fig. 4). The unmasking of the Y2 sites by 10 l M PAO was not altered by equimolar cholesterol (Fig. 4). Cholesterol alone at 10 l M slightly increased the surface binding of the Y2 agonist (Fig. 4). Activation of the Y2 sites by nonionic detergents or emulsifiers could not be satisfactorily studied with cell monolayers, due to loss of cell attachment. With either gpY2-CHO or rat forebrain particulates, pretreatment with polyoxyethylene sorbitan emulsifiers Tween 40 (monopalm- itate) or Tween 80 (monooleate) at up to 10 m M produced less than 10% activation (data not shown). Dynamics of activation of the surface Y2 sites by phenylarsine oxide indicates little accumulation due to receptor externalization The dynamics of appearance of additional surface sites at 30 l M PAO indicated a fast activation, as the increase in the labeling by agonist peptides relative to control values Fig. 2. Unmasking of the binding sites for Y2 selective ligand 125 I-labeled hPYY (3–36) on gpY2-CHO cells by phenylarsine oxide is prevented by dithiothreitol. The cells were pretreated for 30 min at 37 °C with 30 l M PAO with or without dithiothreitol (100 l M )1m M ), or with 1 m M dithiothreitol alone, washed and labeled with 125 I-labeled hPYY(3–36) for 30 min at 37 °C,followedbyextractionofsurface- bound ligand with acid saline at 0–4 °C (see Materials and methods). The data are averages of three experiments. In this and further graphs, asterisks indicate differences significant vs. the control cell surface binding at the level of 95% (*) or 99% (**) confidence in Tukey t-tests, while ampersands (&, &&) show the corresponding differences for the internalized binding. DTT, dithiothreitol. Fig. 3. Effects of two alkylating agents on surface binding and inter- nalization of 125 I-labeled hPYY (3–36) in gpY2-CHO cells. The labeling at 50 p M of the Y2 agonist was carried out for 40 min at 37 °C,inthe presence of the indicated concentrations of NEM, MTSET or DTT. The separation of surface and internalized tracer was done as in Figs 1 and2.Inthesamesetofexperiments(n ¼ 3), the surface binding of the Y2 agonist in the presence of 10 l M PAO was 101 ± 2.8 fmolÆmg )1 cell protein. Significance in Tukey t-testsisindicatedin Fig. 2. DTT, dithiothreitol. 2318 S. L. Parker et al. (Eur. J. Biochem. 269) Ó FEBS 2002 saturated within 20 min at 24 °C, and in less than 10 min at 37 °C (Fig. 5A). More than 50% of the total increase in surface gpY2-CHO sites by PAO was observed within 3 min of labeling at any temperature (Fig. 5A). A similar fast activation was observed for Y2 receptors of rat forebrain cells (not shown). This was contrasted by a gradual, steady accumulation of surface Y1 sites in gpY1- CHO cells at 3 l M filipin or 10 l M PAO, related to inhibition of Y1 receptor internalization by these agents (Fig. 5B). With the gpY1-CHO receptor (which is inten- sively internalized and recycled, as different from the gpY2 receptor expressed in the same cell type [12]), there was a gradual relative increase (over eightfold) in labeled surface sites between 3 and 60 min of incubation at 24 °C in the presence of either filipin III or PAO (Fig. 5B). DISCUSSION Several rhodopsin-family receptors were described as partly cryptic, hidden, masked, or compartmentalized, for example the pituitary LHRH [25], the protease-activated sevenfold receptors such as the thrombin receptor [26], the a2- adrenergic receptor reserve (60–70% of all sites [27]), and the 5-HT1B receptor reserve (up to 90% of the total receptor [28]). The muscarinic acetylcholine receptor, the classic example of an aggregated rhodopsin-like receptor, is usually examined in relation to its sequestration by agonists. However, its massive aggregation by agrins, which also involves cell-surface dystroglucans [29], could induce a degree of constitutive, agonist-unrelated sequestration. Among the Y receptors, the hypothalamic Y2 receptors induced by estrogen show fast density changes by pro- gesterone [30], possibly related to receptor masking. In this study, competition by Y2 agonist after treatment by PAO indicated an essentially normal affinity for the unmasked Y2 sites, which should derive from a constitutively sequestered membrane compartment. This work finds a considerable activation of cell mem- brane Y2 sites by a vicinal dithiol-bridging agent, and a lower activation by the alkylating agent NEM, or by the choles- terol-complexing agent filipin III. With PAO, the cell-surface activation reaches  50% of the receptor numbers found in particulates derived by mechanical disruption of the cells. In cell cultures, many of the surface receptors may not be readily available for interaction with larger peptidic ligands due to cell–cell interactions or interactions with the substra- tum. It is therefore reasonable to assume that the arsenical would expose a majority of the sites that can be labeled in the absence of cell disruption by 34–36-residue peptides used for the Y2 labeling. Lack of Y2 activation by the restricted membrane-access alkylator MTSET [24] indicates a large role for a cell membrane compartment in the masking of Y2 sites. From our previous work [14], a substantial part of Y2 receptors in this compartment should be aggregated and anchored to cytoskeletal proteins, some of which contain PAO-sensitive vicinal dithiol and even trithiol motifs. Activation of masked populations of surface receptor sites by PAO was shown previously for macroglobulin, Fig. 4. Unmasking of surface Y2 sites in gpY2-CHO cells by filipin III and effect of cholesterol. The cell monolayers were labeled by 125 I-labeled hNPY at 50 p M for 40 min at 37 °C at 1, 3, 10 or 30 l M filipin III (FIII) without or with 10 l M cholesteryl hemisuccinate (Chol), and the tracer attached to surface receptors was extracted by acid saline at 0–4 °C (see Materials and methods). Unmasking by 10 l M PAO resulted in surface binding of 90 ± 1.47 fmolÆmg )1 pro- tein without Chol, and of 89.3 ± 0.9 fmolÆmg )1 protein with 10 l M Chol (n ¼ 3). Significance vs. the control binding was evaluated as in Fig. 1. Fig. 5. Comparative dynamics of labeling of Y2 sites in gpY2-CHO cells and of Y1 sites in gpY1-CHO cells in the presence of phenylarsine oxide or filipin III. The ligands (see below) were used at 50 p M for the indi- cated periods of time. After removal of the unbound tracer and washing, the cells were extracted with cold acid saline to separate the cell surface-bound and internalized ligand (see Materials and meth- ods). The results are expressed as percent of control labeling. The data are averages of three experiments. (A) Surface labeling (ext) and internalization (int) of 125 I-labeled hPYY(3–36) in gpY2-CHO cells at 10 l M PAO. The control labeling at 60 min was 20.3 ± 0.6 and 6.9 ± 0.23 (37 °C), and 24 ± 0.83 and 3.3 ± 0.15 fmolÆmg )1 total cell protein (24 °C) for the cell-surface and internalized fraction, respectively. For any time point, the elevation of Y2 binding in the presence of PAO was highly significant vs. the respective control binding in Tukey t-testing. (B) Surface labeling (ext) and internaliza- tion (int) of 125 I-labeled hNPY in gpY1-CHO cells at 24 °C in the presence of 10 l M PAO or 3 l M filipin III. After 60 min at 24 °C,the control labeling was 7.5 ± 0.23 and 37.7 ± 0.3 fmolÆmg )1 total cell protein for the cell-surface and internalized fraction, respectively. Note that the internalized fraction of the Y1 binding with either inhibitor was less than 5% of control values. Ó FEBS 2002 Masked Y2 NPY receptors (Eur. J. Biochem. 269) 2319 transferrin and mannose-tipped glycoprotein receptors [31], some of which are C-lectins possessing internal vicinal dithiols [32], and could be activated by shedding or disengaging the membrane neighbors through bridging by PAO. This study finds that the surface Y2 binding to either CHO cells or rat forebrain cells is strongly activated by PAO below 100 l M . The Y2 receptor that contains no vicinal dithiols [17] is inhibited by PAO only above 1 m M , i.e. at molarities many orders of magnitude in excess of those producing the maximum unmasking of cell-surface Y2 sites. The activation of the Y2 receptor by PAO could mainly result from alterations in arrangement of neighbors posses- sing vicinal dithiols. The communication between the extracellular matrix (ECM) and the actin cytoskeleton (especially the a-actinin component) indicates a direct involvement of basal lamina molecules such as fibronectin, collagen IV, or laminin in cytoskeletal handling of the plasma membrane nicotinic receptor [33]. This could also apply to the Y2 receptor. The shear-stress signal transduction proposed by Kano et al. [34] could relate to the known aggregation of Y2 receptors [14], and to an involvement of basement membrane constituents. The shear can result from PAO-imposed bridging of vicinal cysteines in ECM molecules, as well as of tandem dithiols in a-actinins, adaptins and selectins. A role for single intramembrane thiols could also be assumed, based on the lack of Y2 activation by MTSET, and limited activation by NEM, an alkylator that would access both the exposed and the membrane-hidden thiols, but in a different time frame [23]. Examination of the activity of the alkylating agents, however, is complicated by their partial inactivation of Y2 binding at molarities below 1 m M ([14]; this work). A thiol/disulfide redox system could regulate cell surface Y2 availability mainly through neighboring or interacting membrane proteins. De-anchoring by PAO should also be considered in the context of thiol-disulfide redox equilibria. Molecules sensi- tive to trivalent arsenicals could be sought especially among the strongly expressed adhesion and interaction factors such as integrins (e.g. a1-integrin, a laminin and collagen receptor), cadherins, and glypicans. Adhesion system ligands possessing multiple vicinal dithiols such as melusin [35], and multifunctional receptor/ligand proteins, e.g. laminins, could also be modified by PAO [36], and this might change the membrane protein arrangement in the vicinity of Y2 sites. Anchoring of selectins (e.g. L -type; [37]), as well as of other adhesion molecules [38] can be reduced by dithiol bridging, and this type of change might alter the availability of Y2 sites for agonist binding. However, PAO can also act on intramembrane and intracellular sites, as it penetrates the cell membrane with ease, and does accumulate in the cytosol [39]. Among cell membrane molecules that can be affected by the arsenical to alter the state of aggregation of the Y2 receptor, certain types of phosphatases could be of importance [40]. This, however, may not be connected directly to the Y2 receptor, which is poorly internalized in response to agonist binding ([12]; this work). Most of the activation of the Y2 sites appears to be related to unmasking of occluded surface receptors, and is also accomplished by shearing involved in cell homogenization. Alkylators and PAO might also increase cell permeability, as observed in the case of occludin proteolysis [41]. Dethiolation of protein disulfides via thioreductase could also be sensitive to PAO [42], and the loss of disulfide dynamics could be partly responsible for the observed Y2 accumulation. Stretch receptors and mechanoreceptors could also participate in regulation of Y2 sites. Tight junction-permeability could be increased by tyrosine phosphorylation [43], to loosen the structure of these aggregates, and PAO might act to improve the phosphorylation by inhibiting tyrosine protein phospha- tases [44]. Decrease in affinity of cell-surface Y2 sites after PAO is small, and could mainly reflect disaggregation of the sites. The change should be related to modifications of the membrane environment of the receptor, as there are no vicinal cysteines in the molecule of the rodent Y2 [17]. Unlike the metabotropic glutamate receptors or the adren- ergic receptors, the Y receptors, except the Y5 [45], do not contain vicinal cysteines that could impart a particular sensitivity to PAO in ligand binding. Thus, the particulate Y2 has a PAO K I above 1 m M , the Y1 of  400 l M , while the Y5 shows a K I of only about 10 l M [46] (implicating the seventh transmembrane segment in Y5 ligand binding). Various proteasome subunits possess vicinal cysteines [47], which can be modified by PAO to alter either the proteolytic function, the anchoring, or the assembly of proteasome complexes. The fast activation of Y2 binding by PAO would indicate cell membrane rather than intracellular targets. However, proteasomes are known to be associated with cell membrane as well as with intracellular membrane systems. Intramembrane targets of PAO could also include G-protein b-andc-subunits that contain vicinal cysteines, as well as some rab and ras G-proteins. The stimulation of Y2 binding by filipin should not result from accumulation of surface sites (as we have shown in parallel experiments for the rapidly internalizing Y1 recep- tor expressed in CHO cells), but rather from a direct unmasking due to complexing of membrane cholesterol, as cholesterol was able to abolish the effect of the polyene. This may involve glycosylphosphatidylinositol anchors, known to undergo a constitutive cholesterol-dependent sequestra- tion into early endosomes [48]. The Y2 receptor activation by alteration of sulfhydryl or cholesterol balance was found in this study for cells with quite different plasma membrane systems. The CHO cells are of epithelial derivation. The forebrain cells expressing the Y2 receptor might, in addition to neuronal cells [49], also include glia, in an analogy to kidney epithelia [50]. The Y2 activation indicates involvement of a general mechanism that can operate across cell types to ensure rapid engage- ment of a sequestered receptor pool in response to stimuli that might range from micromechanical to oxidoreductive. This segregation is not present with rapidly recycling Y1 receptors, but could be shared by other receptors that show low rates of ligand-induced internalization [12]. ACKNOWLEDGEMENTS This research was partly supported by NIH grants 13703 and 12844, and by State of Tennessee Educational Funds. REFERENCES 1. Mukherjee, S., Ghosh, R.N. & Maxfield, F.R. (1997) Endocytosis. Physiol. Rev. 77, 759–803. 2320 S. L. Parker et al. (Eur. J. Biochem. 269) Ó FEBS 2002 2. Ferguson, S.S. (2001) Evolving concepts in G protein-coupled receptor endocytosis: the role in receptor desensitization and signaling. Pharmacol. Rev. 53, 1–24. 3. Wei, H.B., Yamamura, H.I. & Roeske, W.R. (1994) Down- regulation and desensitization of the muscarinic M1 and M2 receptors in transfected fibroblast B82 cells. Eur. J. Pharmacol. 268, 381–391. 4. Tsuga, H., Kameyama, K., Haga, T., Honma, T., Lameh, J. & Sadee, W. (1998) Internalization and down-regulation of human muscarinic acetylcholine receptor m2 subtypes. Role of third intracellular m2 loop and G protein- coupled receptor kinase 2. J. Biol. Chem. 273, 5323–5330. 5. Parker, S.L., Parker, M.S., Kane, J.K. & Li, M.D. (2001) Inter- nalization of neuropeptide Y (NPY) Y1 and Y2 receptors in dis- persed rat forebrain cells. Society for Neuroscience 2001 Annual Meeting, San Diego, CA, Abstract 461.7. 6. Oksche,A.,Boese,G.,Horstmeyer,A.,Furkert,J.,Beyermann, M.,Bienert,M.&Rosenthal,W.(2000)Lateendosomal/lyso- somal targeting and lack of recycling of the ligand- occupied endothelin B receptor. Mol. Pharmacol. 57, 1104–1113. 7. Allen, Y.S., Adrian, T.E., Allen, J.M., Tatemoto, K., Crow, T.J., Bloom, S.R. & Polak, J.M. (1983) Neuropeptide Y distribution in the rat brain. Science 221, 877–879. 8. Kalra, S.P., Dube, M.G., Pu, S., Xu, B., Horvath, T.L. & Kalra, P.S. (1999) Interacting appetite-regulating pathways in the hypothalamic regulation of body weight. Endocrinol. Rev. 20, 68–100. 9. Parker, R.M. & Herzog, H. (1999) Regional distribution of Y-receptor subtype mRNAs in rat brain. Eur. J. Neurosci. 11, 1431–1448. 10. Aicher, S.A., Springston, M., Berger, S.B., Reis, D.J. & Wahles- tedt, C. (1991) Receptor-selective analogs demonstrate NPY/PYY receptor heterogeneity in rat brain. Neurosci. Lett. 130, 32–36. 11. Parker, E.M., Balasubramaniam, A., Guzzi, M., Mullins, D.E., Salisbury, B.G., Sheriff, S., Witten, M.B. & Hwa, J.J. (2000) [D-Trp(34)] neuropeptide Y is a potent and selective neuropeptide Y Y(5) receptor against with dramatic effects on food intake. Peptides 21, 393–399. 12. Parker, S.L., Kane, J.K., Parker, M.S., Berglund, M.M., Lundell, I.A. & Li, M.D. (2001) Cloned neuropeptide Y (NPY) Y1 and pancreatic polypeptide Y4 receptors expressed in Chinese hamster ovary cells show considerable agonist-driven internalization, in contrast to the NPY Y2 receptor. Eur. J. Biochem. 268, 877–886. 13. Allen, J.M., Ferrier, I.N., Roberts, G.W., Cross, A.J., Adrian, T.E., Crow, T.J. & Bloom, S.R. (1984) Elevation of neuropeptide Y (NPY) in substantia innominata in Alzheimer’s type dementia. J. Neurol. Sci. 64, 325–331. 14. Parker, S.L. & Parker, M.S. (2000) Ligand association with the rabbit kidney and brain Y1, Y2 and Y5-like neuropeptide Y (NPY) receptors shows large subtype-related differences in sensi- tivity to chaotropic and alkylating agents. Regul. Pept. 87, 59–72. 15. Parker, S.L., Parker, M.S., Sweatman, T. & Crowley, W.R. (1998) Characterization of G-protein and phospholipase C-coupled agonist binding to the Y1 neuropeptide Y receptor in rat brain. Sensitivity to G-protein activators and inhibitors and to inhibitors of phospholipase C. J. Pharmacol. Exptl Therap. 286, 1–12. 16. Berglund, M.M., Holmberg, S.K., Eriksson, H., Gedda, K., Maffrand, J.P., Serradeil-Le Gal, C., Chhajlani, V., Grundemar, L. & Larhammar, D. (1999) The cloned guinea pig neuropeptide Y receptor Y1 conforms to other mammalian Y1 receptors. Peptides 20, 1043–1053. 17. Sharma, P., Holmberg, S.K., Eriksson, H., Beck-Sickinger, A.G., Grundemar, L. & Larhammar, D. (1998) Cloning and functional expression of the guinea pig neuropeptide Y Y2 receptor. Regul. Pept. 75–76, 23–28. 18. Haigler, H.T., Maxfield, F.R., Willingham, M.C. & Pastan, I. (1980) Dansylcadaverine inhibits internalization of 125I- epidermal growth factor in BALB 3T3 cells. J. Biol. Chem. 255, 1239–1241. 19. Beck-Sickinger, A.G., Wieland, H.A., Wittneben, H., Willim, K.D.,Rudolf,K.&Jung,G.(1994)Complete L -alanine scan of neuropeptide Y reveals ligands binding to Y1 and Y2 receptors with distinguished conformations. Eur. J. Biochem. 225, 947–958. 20. Parker, M.S., Berglund, M.M., Lundell, I. & Parker, S.L. (2001) Blockade of pancreatic polypeptide-sensitive neuropeptide Y (NPY) receptors by agonist peptides is prevented by modulators of sodium transport. Implications for receptor signaling and regu- lation. Peptides 22, 887–898. 21. Munson, P.J. & Rodbard, D. (1980) LIGAND: a versatile com- puterized approach for characterization of ligand-binding pro- teins. Anal. Biochem. 107, 220–239. 22. Zar, J.H. (1984) Biostatistical Analysis. Prentice Hall, Englewood Cliffs, NJ. 23. Zarbiv, R., Grunewald, M., Kavanaugh, M.P. & Kanner, B.I. (1998) Cysteine scanning of the surroundings of an alkali-ion binding site of the glutamate transporter GLT-1 reveals a con- formationally sensitive residue. J. Biol. Chem. 273, 14231–14237. 24. Karlin, A. & Akabas, M.H. (1998) Substituted-cysteine accessi- bility method. Methods Enzymol. 293, 123–145. 25. Leblanc, P., L’Heritier, A., Rasolonjanahary, R. & Kordon, C. (1994) Neuropeptide Y enhances LHRH binding to rat gonado- trophs in primary culture. Neuropeptides 26, 87–92. 26. Kawabata, A. & Kuroda, R. (2000) Protease-activated receptor (PAR), a novel family of G protein-coupled seven trans-mem- brane domain receptors: activation mechanisms and physiological roles. Jpn J. Pharmacol. 82, 171–174. 27. Adler, C.H., Meller, E. & Goldstein, M. (1987) Receptor reserve at the alpha-2 adrenergic receptor in the rat cerebral cortex. J. Pharmacol. Exp. Ther. 240, 508–515. 28. Adham, N., Ellerbrock, B., Hartig, P., Weinshank, R.L. & Branchek, T. (1993) Receptor reserve masks partial agonist activity of drugs in a cloned rat 5-hydroxytryptamine1B receptor expression system. Mol. Pharmacol. 43, 427–433. 29. Montanaro, F., Gee, S.H., Jacobson, C., Lindenbaum, M.H., Froehner, S.C. & Carbonetto, S. (1998) Laminin and alpha dystroglycan mediate acetylcholine receptor aggregation via a MuSK-independent pathway. J. Neurosci. 18, 1250–1260. 30. Parker, S.L., Carroll, B.L., Kalra, S.P., St-Pierre, S., Fournier, A. & Crowley, W.R. (1996) Neuropeptide Y Y2 receptors in hypothalamic neuroendocrine areas are up-regulated by estradiol and decreased by progesterone cotreatment in the ovariectomized rat. Endocrinology 137, 2896–2900. 31. Kaplan, J., Ward, D.M. & Wiley, H.S. (1985) Phenylarsine oxide- induced increase in alveolar macrophage surface receptors: evi- dence for fusion of internal receptor pools with the cell surface. J. Cell Biol. 101, 121–129. 32. McKay, P.F., Imami, N., Johns, M., Taylor-Fishwick, D.A., Sedibane, L.M., Totty, N.F., Hsuan, J.J., Palmer, D.B., George, A.J., Foxwell, B.M. & Ritter, M.A. (1998) The gp200-MR6 molecule which is functionally associated with the IL-4 receptor modulates B cell phenotype and is a novel member of the human macrophage mannose receptor family. Eur. J. Immunol. 28, 4071– 4083. 33. Dmytrenko, G.M. & Bloch, R.J. (1993) Evidence for transmem- brane anchoring of extracellular matrix at acetylcholine receptor clusters. Exp. Cell Res. 206, 323–334. 34. Kano, Y., Katoh, K. & Fujiwara, K. (2000) Lateral zone of cell- cell adhesion as the major fluid shear stress-related signal trans- duction site. Circ. Res. 86, 425–433. 35. Brancaccio,M.,Guazzone,S.,Menini,N.,Sibona,E.,Hirsch,E., De Andrea, M., Rocchi, M., Altruda, F., Tarone, G. & Silengo, L. (1999) Melusin is a new muscle-specific interactor for beta (1) integrin cytoplasmic domain. J. Biol. Chem. 274, 29282–29288. Ó FEBS 2002 Masked Y2 NPY receptors (Eur. J. Biochem. 269) 2321 36. Bernier,S.M.,Utani,A.,Sugiyama,S.,Doi,T.,Polistina,C.& Yamada, Y. (1995) Cloning and expression of laminin alpha 2 chain (M-chain) in the mouse. Matrix Biol. 14, 447–455. 37. Bennett, T.A., Edwards, B.S., Sklar, L.A. & Rogelj, S. (2000) Sulfhydryl regulation of L -selectin shedding: phenylarsine oxide promotes activation-independent L -selectin shedding from leuko- cytes. J. Immunol. 164, 4120–4129. 38. Arribas, J., Coodly, L., Vollmer, P., Kishimoto, T.K., Rose-John, S. & Massague, J. (1996) Diverse cell surface protein ectodomains are shed by a system sensitive to metalloprotease inhibitors. J. Biol. Chem. 271, 11376–11382. 39. Frost, S.C. & Schwalbe, M.S. (1990) Uptake and binding of radiolabelled phenylarsine oxide in 3T3-L1 adipocytes. Biochem. J. 269, 589–595. 40. Dai, Z. & Peng, H.B. (1998) A role of tyrosine phosphatase in acetylcholine receptor cluster dispersal and formation. J. Cell Biol. 141, 1613–1624. 41. Wachtel, M., Frei, K., Ehler, E., Fontana, A., Winterhalter, K. & Gloor, S.M. (1999) Occludin proteolysis and increased perme- ability in endothelial cells through tyrosine phosphatase inhibition. J. Cell Sci. 112, 4347–4356. 42. Park, E.M. & Thomas, J.A. (1989) The mechanisms of reduction of protein mixed disulfides (dethiolation) in cardiac tissue. Arch. Biochem. Biophys. 274, 47–54. 43. Staddon, J.M., Herrenknecht, K., Smales, C. & Rubin, L.L. (1995) Evidence that tyrosine phosphorylation may increase tight junction permeability. J. Cell Sci. 108, 609–619. 44. Durden, D.L., Rosen, H., Michel, B.R. & Cooper, J.A. (1994) Protein tyrosine phosphatase inhibitors block myeloid signal transduction through the Fc gamma RI receptor. Exp. Cell Res. 211, 150–162. 45. Herzog, H., Darby, K., Ball, H., Hort, Y., Beck-Sickinger, A. & Shine, J. (1997) Overlapping gene structure of the human neuro- peptide Y receptor subtypes Y1 and Y5 suggests coordinate transcriptional regulation. Genomics 41, 315–319. 46. Parker,E.M.,Lundell,I.,Berglund,M.M.&Parker,S.L.(2001) Internalization of pancreatic polypeptide Y4 receptors: correlation of receptor intake and affinity. Society for Neuroscience 2001 Annual Meeting, San Diego, CA, Abstract 461.8. 47. Foss, G.S., Larsen, F., Solheim, J. & Prydz, H. (1998) Constitutive and interferon-gamma-induced expression of the human protea- some subunit multicatalytic endopeptidase complex-like 1. Bio- chim. Biophys. Acta 1402, 17–28. 48. Mayor, S., Sabharanjak, S. & Maxfield, F.R. (1998) Cholesterol- dependent retention of GPI-anchored proteins in endosomes. EMBO J. 17, 4626–4638. 49. Bleakman, D., Colmers, W.F., Fournier, A. & Miller, R.J. (1991) Neuropeptide Y inhibits Ca 2+ influx into cultured dorsal root ganglion neurones of the rat via a Y2 receptor. Br.J.Pharmacol. 103, 1781–1789. 50. Sheikh, S.P., Sheikh, M.I. & Schwartz, T.W. (1989) Y2-type receptors for peptide YY on renal proximal tubular cells in the rabbit. Am. J. Physiol. 257, F978–F984. 2322 S. L. Parker et al. (Eur. J. Biochem. 269) Ó FEBS 2002 . PRIORITY PAPER A pool of Y2 neuropeptide Y receptors activated by modifiers of membrane sulfhydryl or cholesterol balance Steven L. Parker 1 , Michael. compartment that can be accessed by alteration of sulfhydryl balance or fluidity of the cell membrane, and by treatments that a ect the anchoring and aggregation