A Mechanism Regulating G Protein-coupled Receptor Signaling That Requires Cycles of Protein Palmitoylation and Depalmitoylation* ♦

Background: The functions of palmitate turnover in signal transduction are poorly understood. Results: Inhibiting palmitate turnover on R7BP redistributed R7BP-R7 RGS complexes from the plasma membrane to endomembranes, dissociated them from GIRK channels, and delayed Gi/o deactivation and channel closure. Conclusion: Palmitate turnover on R7BP promotes GIRK channel deactivation. Significance: Inhibiting palmitate turnover on R7BP could enhance GIRK activity in neurological disorders. Reversible attachment and removal of palmitate or other long-chain fatty acids on proteins has been hypothesized, like phosphorylation, to control diverse biological processes. Indeed, palmitate turnover regulates Ras trafficking and signaling. Beyond this example, however, the functions of palmitate turnover on specific proteins remain poorly understood. Here, we show that a mechanism regulating G protein-coupled receptor signaling in neuronal cells requires palmitate turnover. We used hexadecyl fluorophosphonate or palmostatin B to inhibit enzymes in the serine hydrolase family that depalmitoylate proteins, and we studied R7 regulator of G protein signaling (RGS)-binding protein (R7BP), a palmitoylated allosteric modulator of R7 RGS proteins that accelerate deactivation of Gi/o class G proteins. Depalmitoylation inhibition caused R7BP to redistribute from the plasma membrane to endomembrane compartments, dissociated R7BP-bound R7 RGS complexes from Gi/o-gated G protein-regulated inwardly rectifying K+ (GIRK) channels and delayed GIRK channel closure. In contrast, targeting R7BP to the plasma membrane with a polybasic domain and an irreversibly attached lipid instead of palmitate rendered GIRK channel closure insensitive to depalmitoylation inhibitors. Palmitate turnover therefore is required for localizing R7BP to the plasma membrane and facilitating Gi/o deactivation by R7 RGS proteins on GIRK channels. Our findings broaden the scope of biological processes regulated by palmitate turnover on specific target proteins. Inhibiting R7BP depalmitoylation may provide a means of enhancing GIRK activity in neurological disorders.

Palmitoylation (S-palmitoylation or S-acylation) is a posttranslational modification of proteins in which palmitate or other long-chain fatty acids are attached via thioester linkage to cysteine residues (1,2). Hundreds of proteins in eukaryotic cells, including monomeric and heterotrimeric G proteins, G protein-coupled receptors, ion channels, regulatory enzymes, and scaffold proteins are palmitoylated. Palmitoylation has diverse functions, including anchoring hydrophilic proteins to membranes, sorting proteins into membrane lipid microdomains, and regulating intracellular protein trafficking, proteinprotein interactions, and degradation. Indeed, palmitoylation has important roles in several diseases, including cancer, Huntington disease, and neuronal ceroid lipofuscinosis (3)(4)(5)(6).
Despite such evidence, whether palmitate turnover provides a regulatory switch that controls protein function remains a central question. This concept is best supported by studies of palmitoylated Ras isoforms. Inhibiting depalmitoylation with palmostatin B, a small molecule designed to inhibit acylprotein thioesterase 1 (APT1), 3 redistributes H-or N-Ras from the plasma membrane to endomembrane compartments and blunts growth factor-evoked activation of Ras on the Golgi apparatus (7). These and other findings have indicated that H/N-Ras is depalmitoylated globally by acylprotein thioesterases, repalmitoylated by endomembrane-localized palmitoyltransferases, and then delivered for anterograde transport to the plasma membrane (21). A variant of this model in which palmitoylation occurs both on endomembranes and the plasma membrane has been suggested by studies of the dynamically palmitoylated postsynaptic scaffold protein PSD-95 (18). Proteomic analysis using alkynyl palmitate analogs and pulse-chase analysis has confirmed these findings, identifying a subset of enzymatically regulated palmitoylated proteins in mouse T-cells, including Ras family GTPases, subunits of heterotrimeric G proteins (including G s ␣ and G 13 ␣), membrane-associated guanylate kinases, leucine-rich repeat and PDZ domain (LAP) proteins, and other cancer-related scaffolding proteins (22).
Whereas dynamic palmitoylation occurs on select proteins, many important questions remain because the functional consequences of depalmitoylation are nearly completely unknown. How widely is palmitate turnover, as distinguished from palmitoylation per se, required for function or regulation of specific, palmitoylated proteins? What mechanisms determine whether palmitate turnover occurs rapidly or slowly on palmitoylated proteins? Does regulated palmitate turnover serve as a switch to control protein function?
To elucidate functions of palmitate turnover in neuronal G protein-coupled receptor signaling, we have studied the regulator of G-protein signaling 7 (R7 RGS) family (RGS6, -7, -9 -1, -9 -2, and -11) and its control by R7 RGS-binding protein (R7BP), a hydrophilic, palmitoylated SNARE-like protein (23). R7 RGS proteins form obligate heterodimers with G␤5 and function as GTPase-activating proteins that accelerate deactivation of G i/o class G␣ subunits (24,25). When activated allosterically upon association with R7BP, R7 RGS-G␤5 complexes assemble with G protein-regulated inwardly rectifying K ϩ (GIRK) channels to facilitate G i/o deactivation and consequent channel closure (26). Palmitoylation is required for localizing R7BP to the plasma membrane, whereas palmitoylation blockade results in transport of R7BP into the nucleus (27). At steady state, palmitate turnover occurs rapidly on R7BP, which can be negatively regulated by G i/o signaling (28). Accordingly, palmitate turnover on R7BP may provide a mechanism to regulate GIRK activity evoked by G i/o -coupled receptors. Here, we have tested this hypothesis by analyzing the consequences of inhibiting the rate-limiting depalmitoylation step of palmitate turnover on R7BP trafficking and function. Our studies identify a novel function for palmitate turnover in neuronal cell signaling.

EXPERIMENTAL PROCEDURES
Vertebrate Animals-All experimental protocols involving vertebrate animals were approved by the Animal Studies Committee at Washington University.
Cell Culture and Transfection-Neuro2A cells endogenously expressing RGS7 and G␤5 but not R7BP, Neuro2A cells stably transfected with FLAG-R7BP, and HEK293T cells were cultured as described previously (28,29). Cells were transfected with Lipofectamine 2000 (Invitrogen) according to the supplier's recommendations. Cells were treated with DMSO, palmostatin B (Palm B), hexadecyl fluorophosphonate (HDFP), and/or cycloheximide (Sigma), in Opti-MEM (Invitrogen) at 37°C unless otherwise noted. Neuro2A cells stably transfected with the APT1 knockdown plasmid pLKO.1-APT1 were selected and maintained under puromycin selection. Hippocampal neurons were isolated from wild type and R7BP Ϫ/Ϫ mice as described previously (28). Neurons were cultured for 14 days (days in vitro 14) before drug treatment and electrophysiological recording.
Fluorescence Confocal Microscopy-Neuro2A cells seeded on poly-D-lysine (10 g/ml)-coated coverslips were transfected with indicated plasmids. At 24 h post-transfection, cells were treated 4 -6 h as indicated and fixed with paraformaldehyde (4%) for 10 min at room temperature. Cells were mounted in VECTASHIELD mounting medium (Vector Laboratories), and fluorescence images at different wavelengths were captured sequentially on an Olympus FV500 microscope using Fluoroview software. Images were adjusted for brightness and contrast, assembled as montage, and analyzed using ImageJ software.
Metabolic Labeling, Immunoprecipitation, and On-Bead Click Chemistry-Neuro2A cells stably expressing FLAG-R7BP or transiently expressing GFP-N-Ras and GFP-SNAP25 were labeled with 25 M 17-ODYA for 2 and 6 h for GFP-SNAP25 in DMEM (Invitrogen) supplemented with 5% dialyzed FBS, 0.1 mM sodium pyruvate, and 0.1 mM nonessential amino acids. Cells were washed with phosphate-buffered saline (PBS) prior to being chased with media containing 200 M palmitate in the presence of either vehicle (DMSO), Palm B, or HDFP for the indicated periods of time. Cells were then washed and suspended in lysis buffer (PBS supplemented with 2.5 M PMSF, 1ϫ EDTA-free Complete Protease Inhibitors, and 1% Triton X-100). Cleared lysates were immunoprecipitated overnight at 4°C with mouse anti-FLAG beads or rabbit anti-GFP beads. After three washes with lysis buffer, beads were suspended in PBS. Click chemistry reaction protocols were adopted from previous publications (30,31). Immunoprecipitated samples were reacted with click chemistry reagents (20 M TAMRAazide, 1 mM tris(2-carboxyethyl)phosphine hydrochloride, 100 M tris-(benzyltriazolylmethyl)amine, and 1 mM CuSO 4 ) for 1 h at room temperature with periodic mixing. Reactions were stopped by addition of reducing SDS-PAGE sample buffer and boiling for 5 min at 100°C. Samples were separated by SDS-PAGE and analyzed by in-gel fluorescence analysis (Typhoon 9400 laser scanner, GE Healthcare) and Western blotting. Bands of interest were quantified using ImageJ (TAMRA fluorescence) or ImageLab (Western blotting) software under conditions where detection was demonstrated by control experiments to be in the linear range.
Bioluminescence Resonance Energy Transfer (BRET)-Measurement of BRET in intact cells between GIRK2c-Rluc8 and split Venus-tagged R7-RGS-G␤5 complexes was performed as described previously (26). As indicated, cells also were co-transfected with a plasmid expressing FLAG-tagged R7BP. The results are expressed as means Ϯ S.E. Statistical comparisons between groups were done using Student's t test.
Activity-based Labeling of APT1 and APT2-Neuro2A cells were transfected with plasmids expressing GFP, GFP-APT1, or GFP-APT2, lysed by sonication, treated 30 min with or without reversible APT1-and APT2-selective inhibitors (compounds 21 and 1, 10 M each(33)), and probed with an activity-dependent fluorescent probe (PEGylated rhodamine-labeled fluorphosphonate) for 10 min at room temperature (33). Activitydependent labeling of GFP-APT1 or -APT2 resolved by SDS-PAGE was quantified by fluorescence scanning normalized to the level of expressed protein determined by quantitative immunoblotting (LI-COR).

RESULTS
Inhibition of Palmitate Turnover on R7BP-Because our prior studies showed that R7BP undergoes palmitate turnover (28), we investigated the functions of this process by using Palm B or HDFP to inhibit APT1 and other enzymes in the serine hydrolase family that mediate protein depalmitoylation (9,22,34). Because HDFP and Palm B are globally acting irreversible inhibitors of depalmitoylation, and Palm B can covalently modify proteins in addition to serine hydrolases (9), the experiments described below employed extensive controls to determine the specificity of effects observed.
First, we used established methods (20,31) to measure the depalmitoylation rate of FLAG-tagged R7BP expressed stably at physiological levels with endogenous RGS7-G␤5 complexes in neuroblastoma (Neuro2A) cells (29). Control, Palm B-, or HDFP-treated cells were pulse-labeled with the palmitate analog 17-ODYA, chased with conventional palmitate, and lysed at various time points. After immunoprecipitation, 17-ODYA-labeled proteins were conjugated with an azide-linked fluorescent dye (TAMRA) by using click chemistry. TAMRA-labeled FLAG-R7BP was quantified relative to total FLAG-R7BP. TAMRA labeling of FLAG-R7BP during the chase indicated the rate and extent of depalmitoylation in cells.
The results indicated that FLAG-R7BP is depalmitoylated by mechanisms sensitive to Palm B or HDFP (Fig. 1). In control cells, FLAG-R7BP was depalmitoylated ϳ70% within 1 h (Fig.  1A), consistent with prior [ 3 H]palmitate pulse-chase labeling studies (28). Palm B or HDFP inhibited depalmitoylation of FLAG-R7BP in a dose-dependent manner (Fig. 1A). Pulsechase analysis indicated that HDFP (10 M) slowed the depalmitolyation rate of FLAG-R7BP ϳ5-fold (Fig. 1B). Likewise, depalmitoylation of transiently expressed GFP-tagged N-Ras was markedly inhibited by HDFP (Fig. 1C). In contrast, transiently expressed GFP-SNAP25 did not undergo significant depalmitoylation within 2 h with or without HDFP, consistent with prior evidence indicating that palmitate turnover on this protein occurs slowly (19,35). These results indicated that Palm B and HDFP inhibit depalmitoylation of proteins that undergo rapid palmitate turnover.
Palmitate Turnover Is Required for Localizing R7BP to the Plasma Membrane-Because models based on studies of H/N-Ras indicate that palmitate turnover is required for localizing palmitoylated hydrophilic proteins to the plasma membrane (21), we analyzed the effects of Palm B and HDFP on the subcellular localization of R7BP. Neuro2A cells transfected with GFP-R7BP were treated with cycloheximide to block new protein synthesis, incubated 2 h with vehicle, Palm B, or HDFP, and then fixed and analyzed by confocal fluorescence microscopy. In control cells, GFP-R7BP localized primarily to the plasma membrane and to a lesser extent on endomembranes (Fig. 2), similar to endogenous R7BP in neurons (29). Following Palm B or HDFP treatment in the presence of cycloheximide, pre-existing GFP-R7BP was depleted from the plasma membrane and accumulated on endomembrane compartments overlapping with Golgi (Fig. 2, A and B) and endosome (Fig. 2C) markers, indicating that redistribution had occurred. Redistribution of pre-existing GFP-R7BP following Palm B or HDFP administration was confirmed by time-lapse fluorescence confocal microscopy (Fig. 2D). Palm B and HDFP had similar effects on FLAG-R7BP stably expressed in Neuro2A cells at physiological levels (Fig. 2E). Palm B and HDFP increased the extent that RFP-R7BP and GFP-N-Ras co-localized on endomembrane compartments (Fig. 2F), indicating that inhibition of palmitate turnover affects the trafficking of these proteins similarly. In contrast, under the same conditions Palm B or HDFP did not affect the localization of GFP-SNAP-25 (Fig. 2G) or a form of GFP-R7BP in which its C-terminal polybasic region was preserved but its palmitoylation sites (CCLVSS) were replaced with the irreversible lipid modification (geranylgeranyl) motif (CLIL) of RhoA (GFP-R7BP-GG; Fig. 2H). Furthermore, HDFP and Palm B did not impair cell viability or organelle (Golgi, endosome) distribution or identity. Therefore, Palm B and HDFP had the specific effect of causing proteins that undergo palmitate turnover to redistribute from the plasma membrane to endomembranes.
The time course of R7BP redistribution (t1 ⁄ 2 ϳ60 min) elicited by Palm B or HDFP was similar to that shown previously for N-Ras in Palm B-treated cells (7) but much slower than the intrinsic rate that palmitoylated proteins translocate to endomembranes (seconds to minutes (36)). Presumably, redistribution kinetics of R7BP or Ras reflect the relatively slow rate that

Palm B or HDFP enters cells and inactivates serine hydrolases sufficiently to inhibit depalmitoylation globally.
Because prior studies showed that constitutively active G o ␣ inhibits R7BP depalmitoylation in Neuro2A cells (28), we determined whether expression of a constitutively active G o ␣ mutant that does not bind RGS proteins (G o ␣*; G o ␣-Q205L/ G184S) affected R7BP localization. Indeed, G o ␣* caused GFP-R7BP to accumulate on endomembrane compartments (Fig.  2I), indicating that R7BP depalmitoylation and intracellular trafficking can be regulated by activated G o ␣ by a mechanism independent of RGS-G␣ interaction.
Palmitate Turnover on R7BP Regulates Trafficking and Function of R7 RGS-G␤5 Complexes-Because R7BP associates with R7 RGS⅐G␤5 complexes (27), we next determined whether palmitate turnover regulates the trafficking of R7 RGS-G␤5-R7BP heterotrimers. Split Venus-tagged forms of the R7 RGS protein RGS9-2 and G␤5 were co-expressed transiently with RFP-R7BP in Neuro2A cells. In control cells these proteins colocalized extensively on the plasma membrane, whereas in HDFP-treated cells they co-localized on endomembrane compartments (Fig. 2J). Because RGS9-2⅐G␤5 complexes expressed transiently without R7BP localized diffusely to the cytoplasm and nucleoplasm (27), the effects of Palm B and HDFP indicated that inhibiting palmitate turnover on R7BP redistributes RGS9-2-G␤5-R7BP heterotrimers from the plasma membrane to endomembranes.
To address whether palmitate turnover is required for the function of R7 RGS⅐G␤5⅐R7BP complexes, we determined whether HDFP affects the ability of R7BP to allosterically modulate R7 RGS⅐G␤5 complexes. This allosteric mechanism facilitates association of R7 RGS⅐G␤5 complexes with GIRK channels, as indicated by BRET experiments using cells co-expressing luciferase-tagged GIRK2c as donor and split Venustagged forms of R7 RGS proteins and G␤5 as acceptors (26). Consistent with prior studies, FLAG-R7BP augmented BRET between GIRK2c and RGS9-2-G␤5 heterodimers ϳ5-fold relative to controls lacking R7BP (Fig. 3A). HDFP significantly reduced BRET in R7BP-expressing cells (Fig. 3A) but did not affect BRET between GIRK2c and RGS9-2-G␤5 heterodimers in cells lacking R7BP.
Next, we determined whether HDFP affects R7BP-dependent regulation of GIRK channel activity by R7 RGS⅐G␤5 complexes in cultured hippocampal pyramidal neurons. GIRK channels are opened by G␤␥ subunits and closed by channelbound G i/o ␣ subunits that have hydrolyzed GTP to GDP and reformed G␣␤␥ heterotrimers (37,38). By accelerating G i/o ␣-  FEBRUARY 28, 2014 • VOLUME 289 • NUMBER 9 mediated GTP hydrolysis, R7 RGS-G␤5 complexes recruited to GIRK channels by R7BP facilitate channel closure (26). Accordingly, we treated hippocampal neurons from wild type and R7BP Ϫ/Ϫ mice with vehicle or HDFP and recorded whole-cell GIRK currents in response to application and subsequent washout of baclofen, a GABA B receptor agonist. Ablating R7BP or treating wild type neurons with HDFP impaired GIRK current deactivation kinetics (offset) similarly (Fig. 3, B and C). In R7BP Ϫ/Ϫ neurons, however, HDFP had a nonsignificant effect on GIRK current deactivation (Fig. 3, B and C). Because these results indicated that HDFP requires the presence of R7BP to affect GIRK current deactivation, they suggested that palmitate turnover on R7BP is essential for R7 RGS⅐G␤5 complexes to regulate GIRK channel activity.

Palmitate Turnover Regulates GPCR-GIRK Signaling
Whereas R7BP is required for HDFP to affect GIRK channel deactivation, we used a further approach to establish whether HDFP exerts its effects specifically by inhibiting depalmitoylation of R7BP rather than other proteins in this system such as G i/o ␣ subunits that also undergo palmitate turnover (39,40). We determined whether R7BP bearing a polybasic region and geranylgeranylation motif (R7BP-GG) is functional and resistant to HDFP as assessed by the kinetics of GIRK channel deactivation. Neuro2A cells endogenously expressing RGS7⅐G␤5 complexes were transiently transfected with GABA B receptors, GIRK2c, and either GFP as a control or GFP-tagged forms of wild type R7BP or R7BP-GG. Analysis of whole-cell GIRK currents upon application and subsequent washout of baclofen indicated that, in the absence of HDFP, palmitoylated and geranylgeranylated forms of GFP-R7BP were functionally equivalent as indicated by GIRK deactivation kinetics relative to GFP controls (Fig. 4). However, although GIRK channel deactivation kinetics in cells expressing wild type GFP-R7BP was sensitive to HDFP, they were resistant to HDFP in cells expressing GFP-R7BP-GG (Fig. 4). These and the preceding results therefore provided three independent lines of evidence that HDFP affects GIRK channel deactivation kinetics specifically by inhibiting palmitate turnover on R7BP.
Serine Hydrolases Other Than or in Addition to APT1 and APT2 Depalmitoylate R7BP-APT1 is the only serine hydrolase shown to date that depalmitoylates proteins oriented toward the cell cytoplasm (10), suggesting that this enzyme and/or its close relative APT2 potentially depalmitoylate R7BP. We tested this hypothesis by knocking down APT1 and APT2 and/or inhibiting them with a potent, reversible inhibitor specific for each enzyme (compounds 21 and 1, respectively (33)). Efficient knockdown or inhibition of APT1 and -2 was achieved but failed to affect R7BP localization (Fig. 5, A-D). Moreover, combining knockdown of APT1 and -2 with inhibitors of both enzymes did not impair R7BP depalmitoylation (Fig. 5E). These results indicated that serine hydrolases other than or in addition to APT1 and APT2 in neuronal cells depalmitoylate R7BP.

DISCUSSION
Palmitate turnover is well established as a widespread biochemical process because hundreds of proteins, even in single cell types (19,31,(41)(42)(43)(44), are palmitoylated, and many proteins are dynamically palmitoylated or change palmitoylation status in response to cell activation or disease (11,19,22). However, whether palmitate turnover is essential for protein function or regulation remains poorly understood. As discussed below, our studies of R7BP establish a novel, essential function for palmitate turnover for regulating neuronal G protein-coupled receptor signaling and indicate that depalmitoylation in neuronal cells involves serine hydrolases other than or in addition to APT1.
We have found that palmitate turnover on R7BP is required for timely deactivation of G i/o -gated GIRK channels by R7 RGS⅐G␤5 complexes in neuronal cells. Palmitate turnover is required to accumulate R7BP on the plasma membrane, similar to H/N-Ras (7). Blunting palmitate turnover redistributes R7BP to endomembranes, thereby removing R7BPbound R7-RGS⅐G␤5 complexes from GIRK channels at the plasma membrane. Because R7 RGS⅐G␤5 complexes are GTPase-activating proteins for G i/o ␣ subunits docked on GIRK channels, this redistribution process slows the rate of G i/o ␣ deactivation and consequent channel closure. Like H/N-Ras, stably palmitoylated R7BP may translocate by diffusion from the plasma membrane to endomembranes (21). Conversely, R7BP undergoing palmitate turnover could be utilized as a substrate for endomembrane-localized palmitoyltransferases that facilitate its subsequent anterograde transport and steady state localization at the plasma membrane. Cycles of depalmitoylation and repalmitoylation of R7BP also may occur locally at the plasma membrane, similar to PSD-95 (18). Indeed, both R7BP and PSD-95 are palmitoylated by DHHC2 (18,28), which localizes to endomembranes and the plasma membrane.
GIRK channel regulation by dynamic palmitoylation of R7BP may have important physiological functions. By hyperpolarizing neurons, GIRK channels provide a principal mechanism for inhibitory modulation of synaptic transmission by scores of G i/o -coupled receptors and have been implicated in neurological disorders, including chronic pain, epilepsy, Parkinson disease, and Down syndrome (45). By impairing G i/o deactivation, genetic ablation of R7BP augments GIRK activity and enhances the antinociceptive effects of morphine (26), which are evoked by G i/o -coupled -opioid receptors and GIRK channels (46). By impairing R7BP localization and function at the plasma membrane, depalmitoylation inhibitors therefore may provide a means of enhancing morphine action in chronic pain or augmenting GIRK activity in neurological disorders.
We have found that R7BP palmitoylation and trafficking are regulated strikingly by G i/o signaling. Interruption of G i/o sig-FIGURE 5. Effect of APT1 and APT2 knockdown and inhibition on R7BP localization and depalmitoylation. A, APT1 and APT2 knockdown. Knockdown efficiency was determined by measuring the level of GFP-APT1 or -APT2 protein relative to FLAG-R7BP in Neuro2A cells transiently transfected with an APT1 shRNA-expressing plasmid or APT2 RNAi as compared with a control shRNA-expressing plasmid and control RNAi. B, effect of APT1 and APT2 knockdown on localization of GFP-R7BP in Neuro2A cells stained for the Golgi marker GS28. C, inhibition of GFP-APT1 and -APT2 activity. Lysates of Neuro2A cells expressing GFP, GFP-APT1, or GFP-APT2 were treated 2 h with vehicle or specific reversible inhibitors of APT1 or APT2 (compounds 21 and 1, respectively; 10 M), labeled with a fluorescent probe (PEGylated rhodamine-labeled fluorphosphonate) that covalently modifies only active enzyme, resolved by SDS-PAGE, and quantified by fluorescence scanning. Enzyme inhibition is indicated by reduction in fluorescent labeling. D, effect of APT1 and APT2 inhibition on GFP-R7BP localization. Neuro2A cells transiently expressing GFP-R7BP were treated 2 h with vehicle or APT1-and APT2-specific inhibitors (compounds 21 and 1; 10 M each), fixed, and stained for GS28. E, effect of simultaneous knockdown and inhibition of APT1 and APT2 on R7BP depalmitoylation. Neuro2A cells transiently expressing FLAG-R7BP and control RNAi or APT1-and -2-specific RNAi were pulse-labeled 2 h with 17-ODYA in the presence of vehicle or APT1 and APT2 inhibitors (compounds 21 and 1, 10 M each) and chased 1 h with unlabeled palmitate in the continued presence of vehicle or inhibitors. 17-ODYA incorporated into immunoprecipitated FLAG-R7BP before and after the chase was derivatized with TAMRA-azide for fluorescence detection and quantification. naling in neurons leads to accumulation of fully depalmitoylated R7BP in the nucleus (28), whereas sustained G i/o signaling inhibits R7BP depalmitoylation and promotes endomembrane localization. Regulation of palmitate turnover on R7BP therefore may determine whether R7 RGS⅐G␤5 complexes control G i/o signaling at the plasma membrane or intracellular compartments or potentially transduce signals between compartments.
The serine hydrolases that depalmitoylate R7BP remain to be identified. Because knockdown and inhibition of the established acylprotein thioesterase APT1 and its relative APT2 did not affect R7BP localization or depalmitoylation in neuronal cells, these enzymes may be uninvolved. Alternatively, APT1 and/or APT2 could function redundantly with other serine hydrolases to depalmitoylate R7BP. Once the relevant acylprotein thioesterases have been identified, the mechanisms that regulate palmitate turnover on R7BP can be investigated.
In conclusion, approaches analogous to those used by us and others (7,47) should enable investigators to define the functions of palmitate turnover on specific proteins in a variety of biological processes. Such knowledge promises to reveal cellular and disease processes that potentially could be modulated by inhibitors of palmitate attachment or removal.