Inhibition of Chemoattractant N-Formyl Peptide Receptor Trafficking by Active Arrestins T. Alexander Key 1,2,3 , Charlotte M. Vines 1,3 , Brant M. Wagener 1,3 , Vsevolod V. Gurevich 4 , Larry A. Sklar 2,3 and Eric R. Prossnitz 1,3, * 1 Department of Cell Biology and Physiology, 2 Department of Pathology, and 3 Cancer Research and Treatment Center, University of New Mexico Health Sciences Center, Albuquerque, NM 87131, USA 4 Department of Pharmacology, Vanderbilt University Medical Center, Nashville, TN 37232, USA *Corresponding author: Eric R. Prossnitz, [email protected]Recent studies have highlighted the emergence of a class of G protein-coupled receptors that are internalized in an arrestin-independent manner. In addition to demonstrating that the N-formyl peptide receptor belongs in this family, we have recently shown that recycling of the receptor requires the presence of arrestins. To further elucidate mechanisms of arrestin-dependent regulation of G protein- coupled receptor processing, we examined the effects of altering the receptor–arrestin complex on ternary complex formation and cellular trafficking of the N-formyl peptide receptor by studying two active arrestin-2 mutants (trun- cated arrestin-2 [1–382], and arrestin-2 I386A, V387A, F388A). Complexes between the N-formyl peptide receptor and active arrestins exhibited higher affinity in vitro than the complex between the N-formyl peptide receptor and wild-type arrestin and furthermore were observed in vivo by colocalization studies using confocal microscopy. To assess the effects of these altered interactions on receptor trafficking, we demonstrated that active, but not wild-type, arrestin expression retards N-formyl peptide receptor inter- nalization. Furthermore, expression of arrestin-2 I386A/ V387A/F388A but not arrestin-2 [1–382] inhibited recycling of the N-formyl peptide receptor, reflecting an expanded role for arrestins in G protein-coupled receptor processing and trafficking. Whereas the extent of N-formyl peptide receptor phosphorylation had no effect on the inhibition of internalization, N-formyl peptide receptor recycling was restored when the receptor was only partially phosphory- lated. These results indicate not only that a functional interaction between receptor and arrestin is required for recycling of certain G protein-coupled receptors, such as the N-formyl peptide receptor, but that the pattern of receptor phosphorylation further regulates this process. Key words: arrestin, desensitization, endocytosis, formyl peptide receptor, G protein-coupled receptor, internalization, phosphorylation, recycling Received 15 June 2004, revised and accepted for publication 14 October 2004 The modulation of G protein-coupled receptor (GPCR) activity, including that of the b 2 -adrenergic (b 2 AR), angio- tensin II type 1 A, adenosine A 2 , D2 dopamine, CCR-5, and vasopressin V2 receptors, is effected through the con- certed actions of kinases, arrestins, and endocytic proteins (reviewed in (1–3)). Agonist activation induces the rapid phosphorylation of serines and threonines in the carboxyl- terminus and/or intracellular loops of GPCRs by a G protein-coupled receptor kinase (GRK). Arrestins then physically prevent the signaling of activated, phospho- rylated receptors by sterically precluding heterotrimeric G protein binding. For many GPCRs, arrestin binding also results in the simultaneous recruitment of endocytic pro- teins to the plasma membrane followed by translocation of the receptor–arrestin complex to clathrin-coated pits. As receptors spatially segregate from both agonists and G proteins upon internalization, they are further desensitized. Internalized GPCRs can re-sensitize after ligand and arrestin dissociation, phosphatase activity, and efferent trafficking (4). Arrestins therefore likely play a broad range of roles in the processing of GPCRs, with effects on signaling, internalization, trafficking and recycling. In recent years, however, GPCRs have been described that internalize in an arrestin-independent manner. These include the m2-muscarinic (5,6), protease-activated-1 (7), gonado- tropin-releasing hormone (8), and N-formyl peptide recep- tors (FPR (9–12)). Following agonist binding, these receptors are phosphorylated by a GRK (13,14). However, unlike class- ical GPCRs, phosphorylation in some cases may be suffi- cient for partial desensitization (15,16). Thus, arrestin binding may not be absolutely required to quench the ago- nist-stimulated signaling of such receptors. Furthermore, studies have revealed the presence of a novel endocytic pathway in the internalization of some of these receptors, since dominant-negative clathrin, dynamin and arrestin (or a subset thereof) do not affect receptor kinetics of endocyto- sis or recycling (5,9). Despite these dissimilarities, as with classical GPCRs, arrestins do in fact translocate to the mem- brane following agonist stimulation, bind to these GPCRs and can traffic with internalized receptors (15). Furthermore, in vitro studies with liganded, phosphorylated receptors in this class suggest complex interactions of the two proteins (15–17). Nevertheless, a clear-cut role for arrestins in the biology of certain receptors remains uncertain (18). The formyl peptide receptor (FPR) is a chemoattractant GPCR that couples to a pertussis toxin-sensitive G protein in leukocytes, where it mediates superoxide formation, degranulation, and chemotaxis (19). The FPR may be Traffic 2005; 6: 87–99 Copyright # Blackwell Munksgaard 2005 Blackwell Munksgaard doi: 10.1111/j.1600-0854.2004.00248.x 87
13
Embed
Inhibition of Chemoattractant N-Formyl Peptide Receptor Trafficking by Active Arrestins
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Inhibition of Chemoattractant N-Formyl PeptideReceptor Trafficking by Active Arrestins
T. Alexander Key1,2,3, Charlotte M. Vines1,3,Brant M. Wagener1,3, Vsevolod V. Gurevich4,Larry A. Sklar2,3 and Eric R. Prossnitz1,3,*1Department of Cell Biology and Physiology,2Department of Pathology, and3Cancer Research and Treatment Center, University ofNew Mexico Health Sciences Center, Albuquerque, NM87131, USA4Department of Pharmacology, Vanderbilt UniversityMedical Center, Nashville, TN 37232, USA*Corresponding author: Eric R. Prossnitz,[email protected]
Recent studies have highlighted the emergence of a class ofG protein-coupled receptors that are internalized in anarrestin-independent manner. In addition to demonstratingthat the N-formyl peptide receptor belongs in this family,we have recently shown that recycling of the receptorrequires the presence of arrestins. To further elucidatemechanisms of arrestin-dependent regulation of G protein-coupled receptor processing, we examined the effects ofaltering the receptor–arrestin complex on ternary complexformation and cellular trafficking of the N-formyl peptidereceptor by studying two active arrestin-2 mutants (trun-cated arrestin-2 [1–382], and arrestin-2 I386A, V387A,F388A). Complexes between the N-formyl peptide receptorand active arrestins exhibited higher affinity in vitro thanthe complex between the N-formyl peptide receptor andwild-type arrestin and furthermore were observed in vivoby colocalization studies using confocal microscopy. Toassess the effects of these altered interactions on receptortrafficking, we demonstrated that active, but not wild-type,arrestin expression retards N-formyl peptide receptor inter-nalization. Furthermore, expression of arrestin-2 I386A/V387A/F388A but not arrestin-2 [1–382] inhibited recyclingof the N-formyl peptide receptor, reflecting an expandedrole for arrestins in G protein-coupled receptor processingand trafficking. Whereas the extent of N-formyl peptidereceptor phosphorylation had no effect on the inhibitionof internalization, N-formyl peptide receptor recycling wasrestored when the receptor was only partially phosphory-lated. These results indicate not only that a functionalinteraction between receptor and arrestin is requiredfor recycling of certain G protein-coupled receptors, suchas the N-formyl peptide receptor, but that the pattern ofreceptor phosphorylation further regulates this process.
aCompiled from this work and refs. (15–17).bRefers to the ability of the indicated FPR state to form a complex with G protein in vitro and initiate cell signaling in vivo.c— indicates an affinity >> 30 mM and not measurable; þ indicates an affinity >10 mM and weakly measurable; þþ indicates an affinity
between 1 and 10mM; þþþ indicates an affinity < 1mM. Where accurate values are known, they are shown in parentheses.dA qualitative determination inferred from both the direction and magnitude of the shift in the ligand dissociation curve of the indicated
complex. See also footnote a.eNot determinable due to the lack of observable complex formation. By default the receptors exist in a low affinity state for ligand.
Active Arrestins Inhibit FPR Trafficking
Traffic 2005; 6: 87–99 91
requires receptor phosphorylation yet is not mediated by
arrestins (12), in contrast to the b2AR.
Effects of active arrestins on FPR recycling
Given the prior results, one might hypothesize that the
apparent inhibition of internalization could be the result of
an enhanced recycling rate such that active arrestin
expression would lead to more rapid trafficking of internal-
ized receptors back to the cell surface, as has recently
been described for the b2AR (27). Additional support for
this hypothesis lies in the fact that the kinetics of arrestin
binding to the carboxyl-terminus of many GPCRs dictates
their kinetics of recycling (34). However, belying this
hypothesis, high affinity complexes are less likely to
dissociate than low affinity ones. In that sense, ‘active’
arrestins could stay associated with the FPR for longer
time periods, since they are of higher affinity, and there-
fore could retard, not enhance, recycling.
To assess directly the effects of active arrestin expression
on FPR recycling, we used the same flow cytometric
system as employed for internalization assays. However,
after agonist stimulation at 37 �C and washing on ice, cells
were rewarmed to 37 �C in the absence of ligand for vary-
ing times in order to allow trafficking from endosomes
back to the cell surface. As shown in Figure 7, overexpres-
sion of arrestin [1–382]–GFP had no observable effect on
the rate of wild-type FPR recycling in comparison with
nontransfected control cells. Surprisingly, however,
endosomal arrestin dissociation in addition to receptor
dephosphorylation. Since the principal difference between
the DA and DB receptors in terms of ternary complex
formation is agonist affinity, it is possible that differences
in complex stability (i.e. ligand affinity) account for the
differences in re-sensitization kinetics. That is, since the
DA receptor binds to arrestin-2–3 A and forms a high ago-
nist affinity complex, such assemblies are on average less
likely to dissociate in an acidic endosomal environment
than low agonist affinity assemblies composed of DB
receptors and are, concomitantly, less likely to recycle. It
should also be noted, however, that the somewhat lower
affinity of the arrestin-2–3 A mutant for the DB receptor
compared with the WT and DA FPR, could allow the for-
mer receptor but not the latter receptors to recycle.
A further explanation may lie in the loss of key regulatory
sites as a result of carboxy-terminal truncation of the
arrestin. The distal tail, which is intact in the 3 A, but not
the truncated arrestin-2 mutant, has been demonstrated to
contain several domains through which arrestins interact
with protein factors, as well as critical sites of regulation.
These protein factors include the b2-adaptin subunit of the
adaptor protein AP-2 (38), which has been shown to func-
tion in classical GPCR internalization (39), as well as
N-ethylmaleimide-sensitive factor, an ATPase which
binds to the distal tip of arrestin and functions in receptor
transport and sorting (40). Furthermore, it has been shown
that dephosphorylation of serine 412 in the tail of arrestin
is necessary for classical receptor–arrestin complexes to
engage the endocytic machinery (41). The truncated pro-
tein also lacks this important phosphorylation site. And
finally, there may be unknown conformational differences,
including sites of ubiquitination (42), between the two
proteins that would account for the differences in recy-
cling. The presence or lack of a critical domain may affect
not only the kinetics of trafficking, but also the endosomal
compartmentalization of internalized assemblies. Our
results suggest that the inhibition of recycling may in fact
result from a defect in the ability of the FPR to traffic
properly from early endosomes to recycling endosomes
in the presence of the arrestin-2–3 A mutant.
In summary, our recent results have demonstrated that
the presence of arrestin is required for FPR recycling (12).
The results presented here extend this observation to
reveal how the activation state of arrestin is a critical
determinant in regulating receptor recycling. Thus, an
arrestin mutant that is perhaps incapable of returning to
its basal inactive state (i.e. dissociating from the receptor)
is capable of blocking intracellular trafficking and thus pre-
vent the recycling of the FPR. Our results suggest that the
carboxy-terminus of arrestin as well as the ability of the
receptor to attain a conformation upon arrestin binding
displaying high affinity for the ligand are required for this
inhibition. These studies further suggest that dissociation
of the GPCR–arrestin complex is a critical determinant in
receptor recycling. It will be of interest to examine
whether disruptions in the timing and quality of arrestin
associations will similarly disrupt the endocytic profile of
other GPCRs that internalize in an arrestin-independent
manner. Lastly, our results underscore the fundamental
differences in the mechanisms involved in the internaliza-
tion and subsequent trafficking of the FPR as compared to
receptors such as classic b2AR.
Materials and Methods
ReagentsfNle-Leu-Phe-Nle-Tyr-Lys-Alexa546 and -Alexa633 were synthesized by
incubating Alexa Fluor-546 and -633 carboxylic acid succinimidyl esters
(Molecular Probes, Eugene, OR) in anhydrous DMSO (containing 100 mM
triethylamine as catalyst) with equimolar fNle-Leu-Phe-Nle-Tyr-Lys (Sigma,
St. Louis, MO) at room temperature overnight (17). GTPgS (guanosine 50-3-
O-(thio)triphosphate) was from Sigma. Frozen stocks were thawed imme-
diately prior to use. Monomeric RFP (mRFP1) was a generous gift from
Dr. Roger Tsien (43).
Cell stimulation and membrane preparationAs previously described, U937 cells stably expressing the FPR were cul-
tured in RPMI at 37 �C with 5% CO2 and passaged every 3–4 days by
reseeding at 2�105 cells/mL. For membrane harvest, cells were expanded
in sealed Pyrex spinner flasks, stimulated for 8 min at 37 �C with 10 mM
fMLF, which results in near maximal receptor phosphorylation, and trans-
ferred to ice (16).
Stimulated cells were collected by centrifugation, resuspended in ice-cold
cavitation buffer (10 mM HEPES, 100 mM KCl, 30 mM NaCl, 3.5 mM MgCl2,
600 mg/mL ATP, pH 7.3), and bombed by nitrogen cavitation for 20 min at
500 psi. Following centrifugation, the membrane fraction was resuspended
in HEPES sucrose buffer (200 mM sucrose, 25 mM HEPES, pH 7.0) with
protease inhibitor and phosphatase inhibitor cocktails (Calbiochem,
San Diego, CA) and flash frozen. Aliquots were stored until use at � 80 �C.
Detergent solubilizationThawed membranes were diluted in an intracellular binding buffer (BB:
30 mM HEPES, 100 mM KCl, 20 mM NaCl, 1 mM EGTA, 0.1% w/v BSA,
Active Arrestins Inhibit FPR Trafficking
Traffic 2005; 6: 87–99 97
0.5mM MgCl2), isolated by centrifugation, and resuspended in BB contain-
ing protease inhibitor cocktail set I, phosphatase inhibitor cocktail, and 1%
n-dodecyl b-D-maltoside (Calbiochem/EMD Biosciences, San Diego, CA).
Following solubilization for 90min at 4 �C, the soluble fraction was collected
by centrifugation for immediate experimentation.
Receptor reconstitutionSolubilized FPR was incubated with either bovine brain Gi/Go heterotrimer,
purified arrestins, or BB, as well as with fluorescent agonist, N-formyl-met-
leu-phe-lys-fluorescein 5-isothiocyanate (10 nM). The liganded samples
were gently mixed at 4 �C for up to 120min. Blocked samples received a
large excess of unlabeled peptide, N-formyl-met-leu-phe-phe (fMLFF), prior
to fluorescent ligand addition.
Spectrofluorometric analysisFluorescence was measured with an SLM 8000 spectrofluorometer (Spec-
tronics, Urbana IL) in time acquisition, photon-counting mode. Excitation
was fixed at 490 nm with a 490 nm band pass filter and the excitation
monochromater. Emission was monitored using a 520 nm band pass inter-
ference filter and a 500-nm-long pass filter. Antibody and nucleotide add-
itions were made through a microinjection port above the sample holder.
Following reconstitution at 4 �C, samples were diluted with room tem-
perature BB with 0.1% n-dodecyl b-D-maltoside and inhibitors, transferred
to glass cuvettes, and placed into the spectrofluorometer with gentle stirring.
For the first 10s, equilibrium levels were obtained. At 10s, excess antifluor-
escein antibodies, as previously described, were added to the sample. The
antibodies rapidly quench the fluorescence of the unbound, fluorescein-con-
jugated ligand. For some samples, excess GTPgS (guanosine 50-3-O-(thio)tri-
phosphate) was added at 40 s in order to disrupt G protein-receptor coupling.
ElectroporationsThe arrestin-2–GFP construct in pEGFPN1 was a generous gift from
Dr. Jeffrey Benovic (Thomas Jefferson University). Mutant arrestin
constructs were generated through PCR amplification using arrestin-specific
primers and verified in their entirety by dideoxy sequencing. For
transfections, DNA (25mg) was added to 8� 106 FPR-expressing U937
cells in serum-free RPMI. Cells were electroporated at 200V, 2000O, with
a 50ms pulse time on a BTX (Holliston, MA) electroporator and transferred to
T25 flasks with serum-containing RPMI. After 24–48h, cells were collected
by centrifugation, and resuspended in serum-free RPMI for assay.
Confocal microscopyTransfected cells were incubated with 10nM fNle-Leu-Phe-Nle-Tyr-Lys-
Alexa546 or 633 for the indicated time at either 0 �C or 37 �C. Stimulated
samples were immediately fixed with ice-cold, 2% paraformaldehyde for
30min. Cells werewashed, resuspended in Vectashield (Vector Laboratories,
Burlingame, CA), and mounted onto glass slides. Fluorescence images were
acquired on a Zeiss LSM510 confocal microscope (Carl Zeiss Inc., Thornwood,
NY, USA) to localize both the FPR (red) and arrestin (green) signals.
Internalization and recyclingTransfected cells in serum-free RPMI were stimulated with 10 mMfMet-Leu-Phe (FPR) or buffer for up to 30min at 37 �C. Samples were
repetitively washed in ice-cold PBS and resuspended in PBS with either
10 nM fNle-Leu-Phe-Nle-Tyr-Lys-Alexa633. Samples were kept on ice until
analysis. Fluorescence measurements were acquired on a Beckman Dick-
inson FACS Calibur. Live cells were initially gated by forward scatter vs.
side scatter plots. GFP-expressing cells were then gated by FL1 histogram
in comparison to untransfected control cells. Finally, the receptor-asso-
ciated fluorescence was resolved FL4 histogram analysis. The mean chan-
nel fluorescence for 50 000 events per sample was collected and analyzed.
For recycling assays, prior to fluorescent ligand/antibody addition, cells
were warmed for up to 30min at 37 �C to allow for receptor recycling,
quenched on ice, and collected by centrifugation. Acquisition was other-
wise identical to that for internalization assays.
Data analysisData were analyzed using FCSQuery (44) and Prism software (Graphpad,
San Diego, CA). In general, data were plotted as normalized fluorescent
intensity for receptor reconstitution or mean channel fluorescence for
internalization and recycling as a function of time. Non-linear regression
analyses were performed to generate dose–response curves and normal-
ized to Bmax values. For repetitive spectrofluorometric reconstitution
assays, sample data were pooled prior to analysis.