A Nuclear Function of b-Arrestin1
in GPCR Signaling: Regulation of Histone
Acetylation and Gene Transcription
Jiuhong Kang,1,3Yufeng Shi,1,3Bin Xiang,2,3Bin Qu,1Wenjuan Su,2Min Zhu,2Min Zhang,1Guobin Bao,1
Feifei Wang,2Xiaoqing Zhang,2Rongxi Yang,1Fengjuan Fan,1Xiaoqing Chen,2Gang Pei,1and Lan Ma2,*
1Laboratory of Molecular Cell Biology, Institute of Biochemistry and Cell Biology, Shanghai Institutes for Biological Sciences,
The Graduate School, Chinese Academy of Sciences, Shanghai 200031, China
2Pharmacology Research Center, Shanghai Medical College, The Graduate School, Fudan University, Shanghai 200032, China
3These authors contributed equally to this work.
Chromatin modification is considered to
be a fundamental mechanism of regulating
gene expression to generate coordinated
responses to environmental changes, how-
ever, whether it could be directly regulated
by signals mediated by G protein-coupled
receptors (GPCRs), the largest surface re-
ceptor family, is not known. Here, we show
that stimulation of delta-opioid receptor,
a member of the GPCR family, induces
nuclear translocation of b-arrestin 1 (barr1),
which was previously known as a cytosolic
regulator and scaffold of GPCR signaling.
In response to receptor activation, barr1
translocates to the nucleus and is selec-
tively enriched at specific promoters such
as that of p27 and c-fos, where it facilitates
p300, resulting in enhanced local histone
H4 acetylation and transcription of these
genes. Our results reveal a novel function
in GPCR signaling and elucidate an epige-
netic mechanism for direct GPCR signaling
from cell membrane to the nucleus through
signal-dependent histone modification.
Beta-arrestins (barrs, consisting of barr1 and barr2) are cyto-
solic proteins. Activation of G protein-coupled receptors
the translocation of barrs from the cytoplasm to cell mem-
brane and the interaction of barrs with the activated recep-
tor, and hence results in receptor endocytosis and attenua-
tion of receptor signaling (Claing et al., 2002). barrs were
initially known merely as negative regulators of GPCRs, but
new roles of barrs in receptor trafficking and signaling have
been discovered recently (Lefkowitz and Whalen, 2004).
barrs also serve as scaffolds and adapters in receptor endo-
cytosis and signal transduction. They recruit endocytic
proteins including AP-2, clathrin, ARF6, and NSF to the re-
necting the receptors to various cytoplasmic effector path-
ways such as MAPK cascades (Shenoy and Lefkowitz,
2003). Recent studies from Lefkowitz and our laboratories
demonstrate that barrs bind to IkBa in the cytoplasm in an
agonist-dependent manner and regulate NF-kB signaling
(Gao et al., 2004; Witherow et al., 2004). These studies re-
veal important roles of barrs acting as key scaffold proteins
to guide the receptor signals from cell membrane to various
target cascades and thus to different destinations in cell.
Receptor-mediated extracellular signals are transmitted
through the cytoplasm to the nucleus by a complicated sig-
naling network through a series of protein-protein interac-
tions and protein kinase cascades. Agonist-stimulated re-
ceptor phosphorylation and internalization have been long
thought to be solely a negative feedback regulatory mecha-
nism of this process. However, recent evidence suggests
that these receptor-activation-dependent signal regulatory
in the signaling functions and serve as an important pathway
to transmit signals from the cell membrane to the cytoplasm
and the nucleus (Benmerah, 2004; Shi and Massague, 2003).
that endocytic protein Rab5 interacts with APPL proteins
shuttling between the cytoplasm and the nucleus. Upon ac-
tivation of Rab5 by extracellular stimuli, APPL1 translocates
from the membranes to the nucleus where it interacts with
the nucleosome remodeling and histone deacetylase multi-
protein complex NuRD/MeCP1, an established regulator of
chromatin structure and gene expression (Miaczynska
reveal that barrs, mediators of endocytosis of seven mem-
brane-spanning receptors, are also able to shuttle between
Cell 123, 833–847, December 2, 2005 ª2005 Elsevier Inc. 833
Figure 1. Activation of DOR Induces Translocation of barr1 to the Nucleus
(A) Confocal visualization of HA-DOR and barr1-GFP, barr2-GFP or barr1Q394L-GFP in HEK293 cells incubated without (Ctrl) or with 1 mM DPDPE (DP) for
5 min before fixation.
(B) Confocal real-time visualization of barr1-GFP distribution in living HEK293 cells incubated with 1 mM DP. The pictures shown are representative of
five independent experiments. The scale bars represent 20 mm. Values are expressed as the mean ± SD, *p < 0.05, **p < 0.01 versus the 0 min group.
Sp-1 were also detected to show, if any, crosscontamination in the cytosolic and nuclear fractions. The bands of barr1 in nuclear fractions were quantified
and normalized to Sp1, and the data shown are the means ± SD of three independent experiments, *p < 0.05, **p < 0.01 versus the 0 min group.
834 Cell 123, 833–847, December 2, 2005 ª2005 Elsevier Inc.
the cytoplasm and the nucleus, and barr1 is present in both
the nucleus and the cytoplasm at steady state (Scott et al.,
2002; Wang et al., 2003b). These observations suggest
that barr may have an important, yet unknown function in
the nucleus, most likely, to regulate gene transcription
through a novel mechanism.
cells via cascades that lead to appropriate gene and cellular
responses to the environmental changes. Growing evidence
a fundamental mechanism of regulating gene expression to
integrate environmental signals and generate a coordinated
ever, whether epigenetic events could be directly regulated
by GPCR-mediated signal transduction remains unknown.
The current study explored the effects of activation of
GPCRs on nuclear distribution of barr1 and reported that
stimulation of delta-opioid receptor (DOR), a member of
GPCR family, induces nuclear translocation of barr1 and
barr-dependent histone H4 acetylation. We further demon-
strated that nuclear translocation of barr1 leads to its accu-
mulation and H4 hyperacetylation at the p27 and c-fos pro-
moter regions, stimulating transcription of these genes.
These results highlight an epigenetic mechanism for GPCR
signaling from the cell membrane to the nucleus through
signal-dependent histone modification and a novel function
of barr in the nucleus as a GPCR cytoplasmic-nuclear mes-
senger to control transcription of the target genes.
DOR Activation Induces Translocation of barr1
to the Nucleus
HEK293 cells transiently expressing DOR and barr-GFP
were challenged with DPDPE, a specific agonist of DOR,
and the effect of DOR activation on the level of barr in the nu-
cleus was examined by confocal microscopy. As shown in
Figure 1A, before agonist stimulation, barr1-GFP fluores-
cence was distributed in both the nucleus and cytoplasm,
while barr2-GFP fluorescence was mainly distributed in the
cytoplasm. DPDPE stimulation induced trafficking of barr1-
GFP and barr2-GFP to the cell membrane. However, an ap-
parentaccumulation ofbarr1-GFP, butnotbarr2-GFP, inthe
nucleus was also observed under the same conditions. In-
troduction of the nuclear export signal of barr2 into barr1
by a single point (Q394L) mutation eliminated nuclear distri-
bution of barr1, which is consistent with what was previously
observed (Scott et al., 2002; Wang et al., 2003b), and also
abolished DPDPE-induced barr1 nuclear accumulation (Fig-
ure 1A). DOR activation-stimulated barr nuclear transloca-
tion was confirmed by monitoring real-time barr1-GFP fluo-
rescence in living cells during agonist treatment. As shown
in Figure 1B, accumulation of barr1-GFP in the nucleus of
HEK293 cells could be observed within 1 min of DPDPE
challenge. Significant increase of barr1-GFP fluorescence
in the nucleus was observed in 51% of the barr1-GFP ex-
pressing cells challenged with DPDPE (102 out of 200 cells
analyzed). Western analysis of HA-barr1 exogenously ex-
in HeLa cells indicated that concentration of barr1 in the nu-
clear fraction increased 50%–100% after 5 min of DPDPE
incubation (Figures 1C and 1D), confirming the results of
fluorescence microscopy and suggesting that agonist-
stimulated nuclear translocation of barr1 is not an artifact of
receptor (KOR) also induced nuclear translocation of barr1,
however, activation of m-opioid receptor (MOR) or b2-adren-
ergic receptor (b2AR) only caused barr1 translocation to the
plasma membrane but not to the nucleus (Figures S1A and
S1B in the Supplemental Data available with this article on-
line). These data indicate that activation of certain GPCRs in-
duces barr1 trafficking into and accumulation in the nucleus.
DOR Activation Stimulates barr1-Mediated
p27 and c-fos Transcription
Our preliminary microarray analysis using Affymetrix gene-
chip U133A showed that inhibition of expression of barr1
by barr1 siRNA in HeLa cells downregulated transcription
of apoptosis- or cell cycle-related genes such as p27 and
c-fos. Thus, the effect of barr1 on the expression of p27,
tant roles in regulation of cell proliferation, was examined. As
shown in Figures 2A and S2A, overexpression of barr1 (by
?8-fold) resulted in a statistically significant increase in p27
mRNA and protein levels in HeLa cells, whereas overexpres-
sion of barr2 or barr1Q394L (by ?8-fold over endogenous
barr2 or barr1) had no such effect. Furthermore, expression
of barr1 siRNA, but not barr2 siRNA (both reduced the level
of the corresponding isoform by 80%), significantly de-
creased mRNA levels of p27 and c-fos and protein level of
p27 (Figures 2A and S2A). The effect of barr1 siRNA on
c-Fos protein level was not observed in Western assays,
possibly due to the instability and rapid degradation of
c-Fos protein (Curran et al., 1985). In contrast, expression
of barr1 or its siRNA had no influence on the mRNA level
of c-jun (Figure 2A), cyclin A, or cyclin D1 (data not shown).
The effect of barr1 on gene transcription was also examined
in barr1 and barr2 double knockout (barrs?/?) MEF cells
(Kohout et al., 2001). Consistent with what observed in
HeLa cells, expression of barr1 significantly increased p27
mRNA and protein levels, but expression of barr2 or
barr1Q394L failed to cause any change in p27 transcription
(Figures 2B and S2B). These results indicate that barr1 can
regulate expression ofp27 and c-fos genes and this function
of barr1 is apparently correlated with its presence in the
nucleus. Coincidently, DPDPE treatment, which induces ac-
expression of p27, but not that of c-jun (Figure 2C). The ef-
fect of DPDPE on p27 transcription could be blocked by
either DOR antagonist naltridole or barr1 siRNA (Figure
2C), but not by inhibitors of Gi/Go (pertussis toxin), PI3K
(wortmannin), p38 (SB203580), JNK (SP600125), or ERK
(PD98059) (data not shown). These results indicate that
activation of DOR can affect gene expression at transcrip-
tion level and such an effect is mediated by barr, likely via
receptor-activation-induced barr1 nuclear translocation.
Cell 123, 833–847, December 2, 2005 ª2005 Elsevier Inc. 835
Nuclear Accumulation of barr1 Promotes Acetylation
of Histone H4 at p27 and c-fos Promoters
Epigenetic regulation, especially acetylation modification of
histones, has been found to play critical roles in regulation
of eukaryotic gene transcription (Grewal and Moazed,
2003; Grunstein, 1997). Thus the potential influence of
barr1 on histone acetylation was investigated. The Western
data (Figure S3A) showed that the acetylation of histone
Figure 2. DOR Activation Stimulates barr1-Mediated p27 and c-fos Transcription
(A and B) HeLa (A) and barrs?/?MEF (B) cells transiently expressing the indicated plasmids were subjected to RT-qPCR and Western blotting for detection
of p27, c-fos, and c-jun expression.
(C)HeLacellstransientlyexpressingDORor DORandbarr1siRNA(as indicated)were pretreated with orwithout100nMnaltridole (Nal)for5min,incubated
with 1 mM DP for the times as indicated (left) or 60 min (middle and right), and subjected to RT-qPCR and Western analysis. Hypoxanthine phosphoribosyl
transferase (HPRT) was used as acontrolinRT-qPCR. Theproteinlevels of p27 and c-Jun were quantified and normalized toactin protein inWestern. Data
shown are means ± SD of three independent experiments, *p < 0.05, **p < 0.01 versus the corresponding control.
836 Cell 123, 833–847, December 2, 2005 ª2005 Elsevier Inc.
H4, but not H3, increased in HeLa cells following overex-
pression of barr1 and decreased after barr1 siRNA applica-
tion. Out of the four conserved lysine residues (Lys-5, Lys-8,
Lys-12, and Lys-16) susceptible to acetylation in histone H4,
barr1 selectively affected the acetylation of Lys-12 and
Lys-16. Moreover, the acetylation of H4 at Lys-12 and Lys-16
in barrs?/?MEFs was significantly lower than that in the
wild-type MEFs, and reintroduction of barr1 in barrs?/?
MEFs significantly increased the acetylation of H4 at these
sites (Figure S3B). As expected, no significant difference
was observed in acetylation of H3 or H4 after expression
of barr2, barr1Q394L, or barr2 siRNA in HeLa or barrs?/?
MEF cells (Figure S3). These data suggest that nuclear
barr1 regulates acetylation level of H4 and may thus affect
transcription of a number of genes including p27 and c-fos.
Transcription of a particular gene is dependent on the sta-
tus of histone acetylation in close proximity to this gene, es-
pecially within its promoter region (Fry and Peterson, 2002;
Vermaak et al., 2003). As shown by chromatin immunopre-
cipitation (ChIP), the acetylation levels of histone H4 in the
p27 and c-fos promoter regions were decreased by expres-
sion of barr1 siRNA and increased by overexpression of
barr1 in HeLa cells (Figure 3A). Reintroduction of barr1 in
barrs?/?MEFs also significantly increased the amount of
acetylated H4 in the p27 and c-fos promoter regions.
Whereas, overexpression of barr2 or barr1Q394L had no ef-
fect on acetylation of the histones associated with any of the
promoters tested (Figure 3A), suggesting that nuclear accu-
mulation of barr1 is necessary for the altered H4 acetylation
at these promoters. Expression of barr1 siRNA or overex-
pression of barr1 did not show any significant effect on the
moter and acetylation of H3 around all five promoters tested
(Figure 3A, and data not shown for cyclin A, or cyclin D1), in-
dicating that barr1 induces a gene-specific H4 acetylation.
Coincidently, DPDPE treatment led to a barr1- and receptor-
activation-dependent increase of H4 acetylation at p27 (Fig-
ures 3B and 3C) and c-fos (data not shown) promoters. Our
data from various barr1 mutants indicate that their abilities
to promote H4 acetylation at p27 promoter and to stimulate
p27 transcription correlate well (Figure 3E), suggesting that
H4 acetylation alteration is probably one of the mechanisms
for barr1-mediated regulation of gene transcription.
Protein-chromatin binding assay was used to explore the
potential association of barr1 and chromatin and the mech-
anismofregulation onhistone acetylationbybarr1.Similarto
the distribution pattern of histone H4 and p300 (Figure 3D),
the endogenous and exogenous barr1 could be detected
in the crude chromatin pellet (lane 3), the supernatant after
micrococcal nuclease (MNase) digestion (lane 4), and the
pellet after ultra-centrifugation of the supernatant of MNase
digestion (lane 7), indicating the binding of barr1 with chro-
matin. The accumulation of barr1 at p27 and c-fos promoter
regionswasalsoshown byChIPassay,where barr1-specific
antibody could immunoprecipitate the genomic DNA frag-
ments containing the promoter sequences of p27 and
c-fos, but not that of c-jun (Figure 3A), cyclin A, or cyclin D1
(data not shown) in HeLa cells. As expected, barr1 antibody
could not recover any DNA fragments containing the tested
promoter sequences in barrs?/?MEFs (Figure 3A and data
not shown for cyclin A and cyclin D1). Consistent with the
stimulation effect of DOR activation on H4 acetylation, accu-
mulation of barr1 at p27 and c-fos promoters was also sig-
nificantly increased in response to DOR activation (Figure 3B
and data not shown for c-fos). These results indicate a pos-
sible linkage between barr1 enrichment and the elevated H4
acetylation at these promoter regions.
barr1 Recruits p300 to p27 and c-fos
transferase (HAT) and histone deacetylase (HDAC). Our
in vitro acetylation and deacetylation assays using purified
GST-barr1 and immunoprecipitated HA-barr1 showed that
barr1 possesses neither the activity of HAT nor the capability
to affect the catalytic activity of HATs or HDACs in vitro (Fig-
ures S4A–S4C). Furthermore, trichostatin A (TSA) and sirti-
nol, the specific inhibitor of Class I, II, and III HDACs, respec-
tively (Grozinger et al., 2001; Yoshida et al., 1990), had no
significant effect on barr1-induced H4 hyperacetylation in
p27 and c-fos promoter regions (Figure S4D). These results
suggest that barr1 may promote H4 acetylation through re-
cruiting HAT proteins to the specific genomic regions.
p300 and the cAMP response element binding protein
(CREB) binding protein (CBP) are potent HATs possibly re-
lated with the histone acetylation within c-fos promoter
(Usenko et al., 2003), while the potential interaction between
barr1 and HAT protein Tip60 has been suggested (Salwinski
et al., 2004). Our data showed that reducing nuclear barr1
by its siRNA decreased, while overexpression of barr1 in-
creased the accumulation of p300 at p27 and c-fos pro-
moters (Figure 4A), but the accumulation of CBP and
Tip60 at these regions was not affected. In addition, the level
lation of these HAT proteins in the promoter regions of c-jun
(Figure 4A), cyclin A, and cyclin D1 (data not shown) was not
affected by barr1. Coincidently, DPDPE stimulation also in-
creased p300 accumulation at p27 and c-fos promoters,
which was temporally parallel to the enrichment of barr1
and the increase of H4 acetylation in these two promoter re-
gions (Figures 3B and 4B, data not shown for c-fos). Coim-
munoprecipitation using HeLa nuclear extracts showed the
presence of endogenous barr1 in the p300 immunocomplex
and the endogenous p300 in the barr immunocomplex, indi-
cating that barr1 may interact with p300 in the nucleus
(Figure 4C). Collectively, these results suggest that barr1
may promote gene-specific H4 hyperacetylation through re-
cruiting p300 to the target genomic regions.
p300 Plays a Role in barr1-Mediated Histone H4
Hyperacetylation and Gene Transcription
The potential role of p300 in barr1-mediated gene-specific
H4 acetylation was then investigated. H4 acetylation at
p27 and c-fos promoter regions was strongly increased by
overexpression of p300 in HeLa and barrs?/?MEF cells,
and this effect of p300 was augmented by coexpression of
Cell 123, 833–847, December 2, 2005 ª2005 Elsevier Inc. 837
Figure 3. Nuclear Accumulation of barr1 Promotes Acetylation of Histone H4 at p27 and c-fos Promoters
(A–C) ChIP experiments were carried out using antibodies against acetylated H4 (H4Ac), H3 (H3Ac), barr1, and human or mouse IgG (as a negative control)
and the presence of the p27, c-fos, and c-jun promoter sequences in the input DNA and that recovered from antibody-bound chromatin segments were
analyzed by qPCR. The data were normalized to the corresponding input controls. Primer sets covering different regions of the same promoter were used
and produced similar results. The data shown are the means ± SD of three independent experiments of one set of primers. (A) HeLa (left) and barrs?/?MEF
(right) cells expressing the indicated plasmids. NSD, no signal detected. (B) HeLa cells expressing DOR alone were incubated with 1 mM DP for the time
indicated. (C) HeLa cells expressing DOR or DOR and barr1 siRNA as indicated were treated with or without 100 nM Nal for 5 min before incubated with
1 mM DP for 60 min.
838 Cell 123, 833–847, December 2, 2005 ª2005 Elsevier Inc.
(D) Chromatin-protein binding assay. The whole-cell lysates (WCE) were centrifuged and the pellet (Crude Pel) was treated with micrococcal nuclease
(MNase) and centrifuged again. The supernatant obtained (MNase Sup) was subjected to ultracentrifugation to obtain pellet (Ultra Pel) and supernatant
(Ultra Sup). Samples were analyzed in Western using the antibodies indicated.
(E) Nuclear distribution of the wild-type and mutant barr1 transiently expressed in HeLa cells and their effects on H4 acetylation and gene transcription.
Nuclear localization of barr1-HA was observed under confocal microscope. H4 acetylation in p27 promoter regions was determined in ChIP assay and
the transcription of p27 was analyzed by RT-qPCR. Data shown are the means ± SD from three independent experiments, *p < 0.05, **p < 0.01 versus
the corresponding control.
Figure 4. barr1 Recruits p300 to p27 and c-fos Promoter Regions
(AandB)ChIP experiments weredoneusingantibodiesagainstp300,CBP,Tip60,orhumanormouseIgG(as anegativecontrol).Thepresence ofthep27,
c-fos, and c-jun promoter sequences in the input DNA and antibody bound chromatin segments were analyzed by qPCR. The data were normalized to the
corresponding input. The data shown are the means ± SD of three independent experiments of one set of primers. *p < 0.05, **p < 0.01 versus the cor-
responding control. (A) HeLa (left) and barrs?/?MEF (right) cells expressing the indicated plasmids. (B) HeLa cells expressing DOR were incubated in 1 mM
DP for the time indicated (upper). HeLa cellsexpressing DOR orDOR and barr1siRNA as indicated were treated with or without 100nM Nal for 5 minbefore
incubated with 1 mM DP for 60 min (middle and lower).
(C) HeLa nuclear extracts were immunoprecipitated with p300 or barr antibody and the immunocomplexes were analyzed in Western using antibodies
against p300 or barr. 5% of total nuclear extracts was loaded as a control.
Cell 123, 833–847, December 2, 2005 ª2005 Elsevier Inc. 839
barr1 and blocked by coexpression of barr1 siRNA. More-
over, p300 DN, a dominant-negative mutant of p300,
strongly attenuated the effect of barr1 on H4 acetylation at
p27 and c-fos promoters (Figure 5 for p27 and data not
shown for c-fos). These data indicate that p300 plays a criti-
cal role in barr1-mediated H4 hyperacetylation in these pro-
moter regions. Similar to its effect on barr1-mediated H4
acetylation, overexpression of p300 DN also diminished
the effect of barr1 on p27 and c-fos transcription (Figures
6A and 6B), suggesting that gene-specific H4 hyperacetyla-
tion promoted by barr1 and p300 contributes to transcrip-
tional activation of these genes.
CREB is the known transcription factor that recruits the
coactivators CBP and p300 (Mayr and Montminy, 2001)
and regulates transcription of p27 and c-fos (Garcia et al.,
2004; Mayr and Montminy, 2001). Recent studies indicate
the importance of stimulus-induced histone acetylation on
CREB-dependent transcription of genes such as c-fos (Jo-
hannessen et al., 2004). Our coimmunoprecipitation data
munocomplex and vice versa (Figure 6C). Further studies
showed that the expression of CREB siRNA significantly at-
(Figure 6D), although expression of barr1 and its siRNA had
no effect on the binding of CREB (Figure 6D) or phosphory-
lated CREB (data not shown) at these regions. In addition,
overexpression of barr1 showed no significant effect on
CREB-mediated transcriptional activity in both GAL4-CREB
Figure 5. p300 Plays a Role in barr1-Mediated Gene-Specific H4 Acetylation
ChIP experiments were carried out in HeLa (A) or barrs?/?MEF (B) cells expressing the indicated plasmids using antibodies against H4Ac and H3Ac. The
presence of the p27 and c-jun promoter sequences in the input DNA and antibody-bound chromatin segments was analyzed by qPCR. The data obtained
were normalized on the basis of the corresponding input control, and means ± SD from three independent experiments were plotted. **p < 0.01 versus the
cells transfected with NS siRNA or bgal in the corresponding bgal and p300 group, respectively.
840 Cell 123, 833–847, December 2, 2005 ª2005 Elsevier Inc.
reporter system and the system with reporter plasmid carry-
ing a CREB binding site (Figure S4E). These results suggest
that barr1 unlikely activates transcription of these genes
through promoting CREB phosphorylation or in the absence
andCREBmaybeinvolved inGPCRsignal-stimulated accu-
mulation of barr1 and histone acetylation in specific chromo-
somal regions and the transcription of specific genes.
Activation of Endogenous DOR in Neural
Cells Promotes barr1-Dependent
Histone H4 Hyperacetylation,
p27 Transcription, and Growth Inhibition
DOR is an important neurotransmitter receptor widely ex-
pressed in the central nerve system. To confirm whether
the DOR activation-stimulated H4 acetylation and p27 tran-
scription occur under more physiological conditions, human
brain neuroblastoma SK cells, which express DOR endoge-
nously (Yu et al., 1986), were challenged with DPDPE. As
shown in Figures 7A and 7B, DPDPE treatment time depen-
dently increased the nuclear concentration of endogenous
barr1, the H4 acetylation at p27 promoter, and the transcrip-
tridole and barr1 siRNA. Similarly, intracerebroventricular in-
jections of DPDPE also DOR specifically increased the H4
acetylation at p27 promoter and the transcription of p27 in
mouse hippocampus (Figure 7C) and cerebral cortex (data
not shown), demonstrating the regulation of DOR activation
on histone acetylation and p27 transcription occurs in vivo.
Since p27 is known to be involved in suppression of cell
Figure 6. p300 Plays a Role in barr1-Mediated Gene Transcription
(AandB) HeLa(A)andbarrs?/?MEF(B)cellsexpressingthe indicated plasmids were usedand RT-qPCR wasperformed toevaluate the transcriptionlevels
of p27, c-fos and c-jun. HPRT was used to normalize the amounts of cDNA template. The data shown are the means ± SD from three independent experi-
ments. **p < 0.01 versus the cells transfected with NS siRNA or bgal in the corresponding bgal and p300 group, respectively.
(D) ChIP experiments were carried out to detect the binding of CREB or barr1 at the p27, c-fos, and c-jun promoter regions. The data were normalized on the
Cell 123, 833–847, December 2, 2005 ª2005 Elsevier Inc. 841
Figure 7. barr1-Mediated Epigenetic Regulation of Gene Transcription in Neural Cells
(A) SK cells expressing DOR and barr1 endogenously were incubated with 1 mM DP, 1 mM DADLE (DA), or 1 mM deltorphin-I (DEL) for the time indicated. In
the right panel, cells were pretreated with or without 100 nM Nal for 5 min before incubated with 1 mM DP for 20 min. barr1 content in nuclear extracts was
analyzed by Western and quantified as described in the legend for Figure 1.
842 Cell 123, 833–847, December 2, 2005 ª2005 Elsevier Inc.
growth (Kiyokawa et al., 1996; Nakayama et al., 1996),
whether barr1 and DOR activation could p27-dependently
tion in neuroblastoma cells. As shown in Figure 7E, overex-
pression of barr1 inhibited, while its siRNA promoted the
growth of SK cells. The effect of barr1 on cell growth was
p27 dependent since it could be blocked by p27 siRNA. Co-
incidently, DOR activation by DPDPE obviously inhibited the
growth of SK cells in a barr1- and p27-dependent manner.
Moreover, treatment with other agonists of DOR including
DADLE (structure similar to DPDPE) and deltorphin-I (struc-
ture different from DPDPE), capable of stimulating nuclear
translocation of barr1 (Figure 7A), p27 promoter H4 hyper-
acetylation, and p27 transcription (Figure 7D) in SK cells,
also inhibited growth of these cells (Figure 7E). These data
implicate that activation of endogenous DOR in neural cells
may exert impact on physiological functions of these cells
through a barr1-mediated epigenetic mechanism.
barr1 and barr2 are previously known as cytosolic signaling
regulatory and scaffold proteins. Recent studies revealed
that barr1 is distributed in both the cytoplasm and the nu-
cleus, but the potential function of barr1 in the nucleus is un-
known. The present study showed that activation of DOR
could induce barr1 translocation to the nucleus and stimu-
ing a novel function of barr as a messenger carrying receptor
signals to the nucleus. Overexpression of barr1 and nuclear
translocation of barr1 promoted histone H4 hyperacetylation
at the p27 and c-fos promoters and hence activated their
transcription, indicating that epigenetic regulation of gene
expressionisone function ofbarr inthe nucleus.Collectively,
this study demonstrates that chromatin is a direct target of
GPCR-mediated signal transduction and reveals an epige-
netic mechanism for GPCR signaling from cell membrane
of the arrestin family in the nucleus as a GPCR messenger.
Previous studies demonstrated that activation of GPCR
recruits bothbarr1 and barr2to the cellmembrane and inter-
actions of the phosphorylated GPCR and barrs induce re-
ceptor endocytosis and signal inhibition. However, accumu-
lating evidence also revealed potential functional differences
between the two barr subtypes as well as their receptor
specificity. For example, barr1 binds some GPCRs such as
b2AR, MOR, and endothelin type receptor with lower affinity
than barr2 and is less efficient in membrane translocation
upon agonist stimulation, while barr1 and barr2 bind other
receptors, including angiotensin II type 1A receptor, neuro-
tensin 1 receptor, and substance P receptor with similar
high affinities, and translocate to the membrane with similar
efficiencies (Oakley et al., 2000). Furthermore, sequestration
of the b2AR was compromised in the barr2—but not barr1—
knockout cells (Kohout et al., 2001). Our current study
showed that nuclear accumulation of barr1 occurred follow-
ing stimulation of DOR and KOR, but not after activation of
b2AR or MOR, suggesting that this barr1-mediated epige-
netic pathway may be preferentially used by certain recep-
tors. Although the underlying mechanisms are not under-
stood, differential interactions of barr1 with these receptors
may contribute to the receptor specificity observed. More-
over, our data also showed that DOR activation stimulated
nuclear trafficking responses of barr1 and barr2 to DOR ac-
tivation may be partially attributed to the difference in their
structure. Previous studies showed that both barrs are
able toshuttlebetween cytoplasm andnucleus.But different
frombarr1,barr2possesses astrong nuclearexport signalin
its C terminus, which hinders its retention in the nucleus
(Scott et al., 2002; Wang et al., 2003b). Our data showing
that agonist stimulation failed to induce nuclear accumula-
tion of Q394L, a mutant barr1 with barr2 nuclear export sig-
nal, demonstrated the importance of the C-terminal domain
in regulating nuclear concentration of different barrs. These
tant role in GPCR-mediated nuclear signaling.
The classical GPCR pathways involve the activation of G
proteins and the hydrolysis of GTP, the regulation of cAMP
formation as well as various signaling molecules such as
PKA and MAPKs, and the alteration of transcription of the
target genes (Neves et al., 2002; Shaywitz and Greenberg,
1999). Signal transduction initiated by DOR stimulation acti-
vates Gi/Go proteins and ERK1/2, JNK, P38, and PI3K cas-
cades (EisingerandSchulz, 2004;Perssonetal.,2003;Sha-
the cell surface to the nucleus mediated by barr1 nuclear
translocation. The precise molecular mechanism by which
(B) SK cells transfected with or without barr1siRNA as indicated were incubated in 1 mM DP for the time indicated (left) or 60 min (middle and right) after
pretreated with or without 100 nM Nal for 5 min. The level of H4Ac at p27 and c-jun promoters and the transcription levels of p27 and c-fos were analyzed
by ChIP and RT-qPCR. HPRT was used in RT-qPCR to normalize the input cDNA. Values are expressed as the mean ± SD, *p < 0.05, **p < 0.01 versus the
p27 and c-jun by ChIP and RT-qPCR. Left, samples were taken after different intervals of DP injection. Right, Nal was injected before DP injection and the
hippocampi were taken 60 min after DP injection. Data are from three independent experiments. TATA box binding protein (TBP) was used in RT-qPCR to
normalize the input cDNA. Values are expressed as the mean ± SD, **p < 0.01 versus mice treated without DP.
(D) SK cells were incubated with 1 mM DA or 1 mM DEL for the indicated times and then H4Ac at p27 and c-jun promoters (left) and the transcription of p27
and c-jun (right) were determined by ChIP or qPCR. *p < 0.05, **p < 0.01 versus 0 min control in the corresponding group.
(E)SK cells were transfected with bgal,barr1,barr1Q394L,NS siRNA,barr1 siRNA,p27 siRNA,or indicated combinations andincubated inthe presence or
absence of 1 mM of DP and DA for 60 min or 1 mM of DEL for 30 min. The [3H]thymidine incorporation was determined, and data shown are the means ± SD
from three independent experiments, **p < 0.01 versus the corresponding control.
Cell 123, 833–847, December 2, 2005 ª2005 Elsevier Inc. 843
tobeelucidated, buttheprocessappears tobe independent
of the activation of Gi/Go, PI3K, P38, JNK, and ERK. In the
JAK-STAT signaling pathway, receptor phosphorylation trig-
gered by cytokine and growth factors induces phosphoryla-
tion of STATs, which subsequently translocate to the nu-
cleus to regulate transcription (Levy and Darnell, 2002).
The phosphorylation and nuclear translocation of Smad
proteins also play a critical role in the propagation of TGF-b
signaling from cell membrane to the nucleus (Shi and Mas-
sague, 2003). BothDOR and barr1could be phosphorylated
at their C termini and the changes in DOR and barr1 phos-
phorylation in response to agonist treatment have been ob-
possibility is that, similar to the JAK-STAT and TGF-b signal-
ing pathways, GPCR phosphorylation and posttranslational
modification of barr1 might play a role in the receptor
signal-induced barr1 nuclear translocation. Interestingly,
a recent study has demonstrated that upon extracellular
stimulation, endocytic protein APPL1 translocates into the
nucleus and interacts with the nucleosome remodeling and
histone deacetylase multiprotein complex NuRD/MeCP1
(Miaczynska et al., 2004). barr1 is also a well known impor-
tant regulator of receptor endocytosis. The potential roles
of barr1 membrane trafficking, receptor endocytosis, and
the endocytosis machinery in GPCR activation-induced
barr1 nuclear translocation await future investigations.
Two possible mechanisms could be involved in barr1-
mediated histone hyperacetylation. barr1 may function as a
targeted chromatin regions. Alternatively, barr1 may inhibit
tin. The data herein showed that HDAC activity was not re-
did not have any HAT activity. Moreover, barr1 did not affect
the activity of HAT or HDAC proteins. Therefore we believe
that the increased levels of histone H4 acetylation reflect
the enhanced recruitment of HAT to chromatin mediated by
mulation of barr1 and p300, a HAT protein, in the p27 and
c-fos promoter regions was detected, and the level of p300
in these regions was regulated by barr1. Furthermore, p300
DN inhibited barr1-dependent regulation of H4 acetylation
and gene activation and an interaction between p300 and
barr1wasdetected inimmunoprecipitation. Thesedatasup-
port the hypothesis that inresponseto DOR activation, barr1
p300 to these locations to induce the H4 hyperacetylation
transcription cofactor, barr1 thus could potentially influence
Among the five genes tested, DOR activation increased
transcription of p27 and c-fos, but not c-jun, cyclin A, and
cyclin D1 genes, suggesting that GPCR stimulation-
induced,barr1-mediated histonemodification andtranscrip-
tional activation occur at a defined set(s) of genes. It has
been shown that certain chromatin-remodeling enzymes,
scription factors to ensure that chromatin remodeling is
targeted to the correct gene (Fry and Peterson, 2002). Fur-
thermore, growing studies indicate that in addition to tran-
scription factors, the preexisting nucleoprotein architecture
of specific promoters also plays critical roles in chromatin re-
modeling of specific gene loci (Urnov and Wolffe, 2001).
Consistent with this view, our results showed that agonist-
stimulated accumulation of barr1 and p300 and H4 acetyla-
tion occurred specifically at p27 and c-fos, but not c-jun pro-
moter regions, and that expression of CREB siRNA strongly
inhibited barr1 accumulation at p27 and c-fos promoter re-
gions. It is generally believed that CREB-dependent tran-
scription is stimulated by signal-induced CREB phosphory-
lation at Ser-133 and the subsequent recruitment of the
coactivators p300 and CBP. Interestingly, our data suggest
that while both CREB and p300 are required in barr1-medi-
does not appear to be a change in CREB phosphorylation at
target promoter regions upon DOR stimulation (unpublished
data). Thus, analogous to what happens in the cytoplasm
fold molecule in the nucleus by interacting with transcription
factors and other nuclear proteins to recruit p300 to the tar-
get chromatin regions. Our preliminary mass spectrum anal-
ysis of Flag-barr1 immunoprecipitation complex from the
nuclear or total cell extracts of HEK293 cells suggests the
presence of other nuclear protein components in addition
to p300 and CREB. Thus, in addition to CREB, other se-
quence-specific factor(s) may also contribute to the recruit-
ment of barr1 to target promoter regions such as that of
p27 and c-fos, which could provide a molecular basis for
gene-specific transcriptional regulation by barr1.
Epigenetic regulation is an important pathway to induce
a coordinated transcriptional response to environmental sig-
nals and the balance of epigenetic networks contributes to
the normal processes of human development, while the dis-
ruption of this balance can cause aberrant disease states
such as cancer and mental retardation (Levenson and
Sweatt, 2005; Sutherland and Costa, 2003). GPCRs trans-
and play vital roles in regulation of various cellular functions.
This study shows that activation of DOR induces trafficking
of barr1 to the nucleus and results in histone modification
and gene activation, revealing that epigenetic events such
as histone modification are subjected to direct regulation by
The physiological significance of this barr-mediated epi-
genetic regulatory pathway is implicated by the results that
activation of DOR led to barr1- and p27-dependent growth
inhibition in human neuroblastoma cells. Further research
will be needed to elucidate mechanistic details of this barr-
insight into the physiology regulated by this novel pathway.
Antibodies, Reagents, Plasmids, and siRNAs
Antibodies and reagents are shown in the Supplemental Data. Plasmid
844 Cell 123, 833–847, December 2, 2005 ª2005 Elsevier Inc.
GFP, and HA-barr1 truncation mutants were generated as described
(Wang et al., 2003a, 2003b). Construction of barr2 siRNA, pBS/U6/barr1
siRNA, and pBS/U6/nonspecific siRNA plasmids were as described pre-
viously (Sun et al., 2002; Wang et al., 2003b). The nucleotide sequences
carried in the plasmids are 50-GGAAGCTCAAGCACGAAGACAA-30
(siRNA against both splicing isoforms of barr1 (Parruti et al., 1993)). The
wild-type and dominant negative (C/H1 deletion) p300 pCMVb plasmids
were purchased from Upstate Biotechnology. siRNAs for p27, CREB,
and non-silencing were synthesized by Shanghai GeneChem Inc (Shang-
hai, China). The target sequences of siRNA oligonucleotides are: for p27,
Cell Culture and Transfection
HEK293, HeLa, and SK-N-SH (SK) cells were obtained from the Ameri-
can Type Culture Collection (Rockville, MD) and maintained in MEM me-
dium (Gibco-BRL, Gaithersburg, MD). The wild-type and barr1/2 double
knockout (barrs?/?) murine embryonic fibroblast (MEF) cells, provided by
Dr. Robert J. Lefkowitz (Duke University Medical Center, Durham, NC),
were maintained in DMEM medium (Gibco-BRL). HEK293 cells were
transfected using calcium phosphate coprecipitation and HeLa, SK,
and barrs?/?MEF cells were transfected by Lipofectamine (Invitrogen,
After a 2 hr starvation with serum-free MEM, HEK293 cells transfected
with the indicated plasmids were treated with or without the receptor ag-
onist for 5 min, fixed, and incubated with 12CA5 and then Texas Red-
conjugated anti-mouse IgG. The fluorescence signals were observed un-
Germany). For observation of distribution of barr1-GFP in live cells in real
time, the fluorescence of barr1-GFP in HEK293 cells transfected with re-
ceptor and barr1-GFP was observed at 37ºC using TCS NT equipped
with a temperature controller. The cells were scanned in a time series.
The fluorescence density of barr1-GFP in the entire compartment of the
nucleus, plasma membrane, or cytoplasm of the same cell wasquantified
using Image-Pro Plus 5.1 software (Media Cybernetic, Silver Spring, MD).
Nuclear Extract Preparation
Nuclear extracts were prepared as described previously (Dignam et al.,
1983) with minor modifications. After a 12 hr serum starvation, cells
were incubated with 1 mM DPDPE for different times, washed and resus-
pended in 400 ml of hypotonic buffer. After incubation on ice for 10 min, 3
ml of 1% NP-40 for HEK293, 4 ml of 10% NP-40 for SK, and 30 ml of 10%
resuspended in hypertonic buffer and shaken for 1 hr at 4ºC. After centri-
fugation, the supernatant (nuclear extracts) were saved.
In Western blotting analysis, the protein bands visualized by enhanced
chemiluminescence method were quantified by Scion Image Beta 4.02
software (SynGene, Cambridge, Great Britain). For more quantitative
measurement(as in Figures 1and 7), the blots were incubated withIRDye
800CW-conjugated secondary antibody, the infrared fluorescence image
was obtained using Odyssey infrared imaging system (Li-Cor Bioscience,
Lincoln, NE), and the bands were quantified by Image-Pro Plus 5.1 soft-
2 pmol DPDPE (DP, 1.5 ml/mouse) were injected into the third cerebral
ventricle of 3-week-old C57 mice. In some experiments, saline or
20 fmol naltridole (1.5 ml/mice) were injected 15 min before DP injection.
Mice were sacrificed at different time points after DP injection and hippo-
campi were immediately separated, snap-frozen in liquid nitrogen, and
stored at ?80ºC. All animal treatments were carried out strictly in accor-
dance with the National Institutes of Health Guide for the Care and Use of
Reverse Transcription Quantitative Real-Time PCR
Total RNAs were extracted from cultured cells or mouse hippocampi with
TRIzol (Invitrogen) according to the manufacturer’s instructions. Reverse
transcription of purified RNA was performed using oligo(dT) priming and
superscript II reverse transcriptase (Invitrogen). The quantification of all
gene transcripts was done by qPCR, using Brilliant SYBR Green QPCR
Master Mix and a Light Cycler apparatus (Stratagene). The primer pairs
used are described in the Supplemental Data.
Chromatin Immunoprecipitation (ChIP) assay was performed according
to the protocol for the ChIP assay kit (Upstate Biology). The presence of
ered DNA immunocomplexes was detected by qPCR. The primer pairs
for specific promoter regions (within ?1000 to +100 region, correspond-
Data.Thedataobtainedwerenormalized tothecorresponding DNAinput
The assay was performed as described previously (Donovan et al., 1997;
Liang and Stillman, 1997). Briefly, HeLa cells were lysed in extraction
buffer containing protease inhibitor cocktail (Roche Molecular Biochemi-
cals), and centrifuged. The pellet (Crude Pel) was digested for 20 s with
5 units of micrococcal nuclease (MNase, Takara Biotechnology). The
supernatant after MNase digestion (MNase Sup) was centrifuged at
500,000 ?g for 1 hr again to yield ultracentrifugation pellet (Ultra Pel)
and supernatant (Ultra Sup). All pellet fractions were resuspended in ex-
traction buffer,andthe volumes ofall fractionswere adjusted toreflect the
same cell equivalent before Western analysis.
Cells and the nuclear extracts were lysed in buffer containing 1% Triton
X-100, 10% glycerol, 10 mM Tris-HCl (pH 7.5), 1 mM EDTA, 150 mM
NaCl, 20 mM NaF, and 1 mM phenylmethylsulfonyl fluoride for 4 hr at
4ºC. After centrifugation at 15,000 ?g for 10 min, the supernatant was in-
cubated at 4ºC with indicated antibodies for 12 hr. The immunocom-
plexes were captured by rotating for 1 hr with protein G Sepharose.
SK cells were transfected with barrs for 42 hr or barr1 siRNA for 90 hr. For
estimation of DOR activation effect, SK cells were transfected with barr1
siRNA or p27 siRNA for 90 hr, starved with serum-free MEM for 2 hr, and
treated with 1 mM of DP and DA for 60 min, or 1 mM of DEL for 30 min.
Then differently treated cells were incubated with fresh MEM containing
1 mCi/ml of [3H]thymidine (24 Ci/mmol; Amersham) for 6 hr, and the
[3H]thymidine incorporation in DNA was determined using a Beckman
scintillation S6500 counter.
Quantitative data are expressed as the means ± standard deviation (SD).
The statistical significance was determined by ANOVA followed by Bon-
ferroni post-hoc test for multiple comparisons or Student’s t test.
Supplemental Data include Supplemental Experimental Procedures, four
figures, and Supplemental References and can be found with this article
online at http://www.cell.com/cgi/content/full/123/5/833/DC1/.
We thank Dr. R.J. Lefkowitz for providing antibody A1CT and barrs?/?
MEFs, Drs. Y. Sun and H. Gao for helpful discussions, and S. Xin,
Y. Huang, G. Ding, Y. Li, B. Zhao, and Y. Wu for technical assistance.
Cell 123, 833–847, December 2, 2005 ª2005 Elsevier Inc. 845
This research was supported by grants from the Ministry of Science
and Technology (2003CB515405, 2005CB522406), the National Natural
Academy of Sciences (KSCX1-SW, KSCX2-SW), the Ministry of Educa-
tion, Shanghai Municipal Commission for Science and Technology
(03DZ19213, 02DJ14020), China Post Doctoral Science Foundation,
and the K.C. Wong Education Foundation.
Received: March 29, 2005
Revised: July 7, 2005
Accepted: September 12, 2005
Published: December 1, 2005
Benmerah, A. (2004). Endocytosis: signaling from endocytic membranes
to the nucleus. Curr. Biol. 14, R314–R316.
Claing, A., Laporte, S.A., Caron, M.G., and Lefkowitz, R.J. (2002). Endo-
cytosis of G protein-coupled receptors: roles of G protein-coupled recep-
tor kinases and beta-arrestin proteins. Prog. Neurobiol. 66, 61–79.
Curran, T., Bravo, R., and Muller, R. (1985). Transient induction of c-fos
and c-myc in an immediate consequence of growth factor stimulation.
Cancer Surv. 4, 655–681.
Dignam, J.D., Lebovitz, R.M., and Roeder, R.G. (1983). Accurate tran-
scription initiation by RNA polymerase II in a soluble extract from isolated
mammalian nuclei. Nucleic Acids Res. 11, 1475–1489.
Donovan, S., Harwood, J., Drury, L.S., and Diffley, J.F. (1997). Cdc6p-
dependent loading of Mcm proteins onto pre-replicative chromatin in
budding yeast. Proc. Natl. Acad. Sci. USA 94, 5611–5616.
Eisinger, D.A., and Schulz, R. (2004). Extracellular signal-regulated
kinase/mitogen-activated protein kinases block internalization of delta-
opioid receptors. J. Pharmacol. Exp. Ther. 309, 776–785.
Fry, C.J.,and Peterson, C.L.(2002).Transcription.Unlockingthe gatesto
gene expression. Science 295, 1847–1848.
Gao, H., Sun, Y., Wu, Y., Luan, B., Wang, Y., Qu, B., and Pei, G. (2004).
Identification of beta-arrestin2 as a G protein-coupled receptor-
stimulated regulator of NF-kappaB pathways. Mol. Cell 14, 303–317.
Garcia, C., Paez, P., Davio, C., Soto, E.F., and Pasquini, J.M. (2004).
Apotransferrin induces cAMP/CREB pathway and cell cycle exit in imma-
ture oligodendroglial cells. J. Neurosci. Res. 78, 338–346.
Grewal, S.I., and Moazed, D. (2003). Heterochromatin and epigenetic
control of gene expression. Science 301, 798–802.
Grozinger, C.M., Chao, E.D.,Blackwell,H.E.,Moazed,D.,and Schreiber,
S.L. (2001). Identification of a class of small molecule inhibitors of the sir-
tuin family of NAD-dependent deacetylases by phenotypic screening.
J. Biol. Chem. 276, 38837–38843.
Grunstein, M. (1997). Histone acetylation in chromatin structure and tran-
scription. Nature 389, 349–352.
Johannessen, M., Delghandi, M.P., and Moens, U. (2004). What turns
CREB on? Cell. Signal. 16, 1211–1227.
Kiyokawa, H., Kineman, R.D., Manova-Todorova, K.O., Soares, V.C.,
Hoffman, E.S., Ono, M., Khanam, D., Hayday, A.C., Frohman, L.A.,
and Koff, A. (1996). Enhanced growth of mice lacking the cyclin-
dependent kinase inhibitor function of p27(Kip1). Cell 85, 721–732.
Kohout, T.A., Lin, F.S., Perry, S.J., Conner, D.A., and Lefkowitz, R.J.
(2001). beta-Arrestin 1 and 2 differentially regulate heptahelical receptor
signaling and trafficking. Proc. Natl. Acad. Sci. USA 98, 1601–1606.
Lefkowitz, R.J., and Whalen, E.J. (2004). beta-arrestins: traffic cops of
cell signaling. Curr. Opin. Cell Biol. 16, 162–168.
Levenson, J.M., and Sweatt, J.D. (2005). Epigenetic mechanisms in
memory formation. Nat. Rev. Neurosci. 6, 108–118.
Levy, D.E., and Darnell, J.E., Jr. (2002). Stats: transcriptional control and
biological impact. Nat. Rev. Mol. Cell Biol. 3, 651–662.
Liang, C., and Stillman, B. (1997). Persistent initiation of DNA replication
and chromatin-bound MCM proteins during the cell cycle in cdc6 mu-
tants. Genes Dev. 11, 3375–3386.
Lin, F.T., Krueger, K.M., Kendall, H.E., Daaka, Y., Fredericks, Z.L.,
Pitcher, J.A., and Lefkowitz, R.J. (1997). Clathrin-mediated endocytosis
of the beta-adrenergic receptor is regulated by phosphorylation/
dephosphorylation of beta-arrestin1. J. Biol. Chem. 272, 31051–31057.
Mayr, B., and Montminy, M. (2001). Transcriptional regulation by the
phosphorylation-dependent factor CREB. Nat. Rev. Mol. Cell Biol. 2,
Miaczynska, M., Christoforidis, S., Giner, A., Shevchenko, A., Utten-
weiler-Joseph, S., Habermann, B., Wilm, M., Parton, R.G., and Zerial,
M. (2004). APPL proteins link Rab5 to nuclear signal transduction via an
endosomal compartment. Cell 116, 445–456.
Nakayama, K., Ishida, N., Shirane, M., Inomata, A., Inoue, T., Shishido,
N., Horii, I., and Loh, D.Y. (1996). Mice lacking p27(Kip1) display in-
creased body size, multiple organ hyperplasia, retinal dysplasia, and pitu-
itary tumors. Cell 85, 707–720.
Neves, S.R., Ram, P.T., and Iyengar, R. (2002). G protein pathways. Sci-
ence 296, 1636–1639.
Oakley, R.H., Laporte, S.A., Holt, J.A., Caron, M.G., and Barak, L.S.
(2000). Differential affinities of visual arrestin, beta arrestin1, and beta
arrestin2 for G protein-coupled receptors delineate two major classes
of receptors. J. Biol. Chem. 275, 17201–17210.
Parruti, G., Peracchia, F., Sallese, M., Ambrosini, G., Masini, M., Rotilio,
D., and De Blasi, A. (1993). Molecular analysis of human beta-arrestin-1:
cloning, tissue distribution, and regulation of expression. Identification of
two isoforms generated by alternative splicing. J. Biol. Chem. 268, 9753–
Persson, A.I., Thorlin, T., Bull, C., and Eriksson, P.S. (2003). Opioid-
induced proliferation through the MAPK pathway in cultures of adult hip-
pocampal progenitors. Mol. Cell. Neurosci. 23, 360–372.
Salwinski, L., Miller, C.S., Smith, A.J., Pettit, F.K., Bowie,J.U., and Eisen-
berg, D. (2004). The Database of Interacting Proteins: 2004 update. Nu-
cleic Acids Res. 32, D449–D451.
Scott, M.G., Le Rouzic, E., Perianin, A., Pierotti, V., Enslen, H., Benichou,
S., Marullo, S., and Benmerah, A. (2002). Differential nucleocytoplasmic
shuttling of beta-arrestins. Characterization of a leucine-rich nuclear ex-
port signal in beta-arrestin2. J. Biol. Chem. 277, 37693–37701.
Shahabi, N.A., McAllen, K., and Sharp, B.M. (2003). Phosphorylation of
activating transcription factor in murine splenocytes through delta opioid
receptors. Cell. Immunol. 221, 122–127.
Shaywitz, A.J., and Greenberg, M.E. (1999). CREB: a stimulus-induced
transcription factor activated by a diverse array of extracellular signals.
Annu. Rev. Biochem. 68, 821–861.
Shenoy, S.K., and Lefkowitz, R.J. (2003). Multifaceted roles of beta-
arrestins inthe regulation ofseven-membrane-spanningreceptor traffick-
ing and signalling. Biochem. J. 375, 503–515.
Shi, Y., and Massague, J. (2003). Mechanisms of TGF-beta signaling
from cell membrane to the nucleus. Cell 113, 685–700.
Sun, Y., Cheng, Z., Ma, L., and Pei, G. (2002). Beta-arrestin2 is critically
involved in CXCR4-mediated chemotaxis, and this is mediated by its en-
hancement of p38 MAPK activation. J. Biol. Chem. 277, 49212–49219.
Sutherland,J.E., and Costa, M. (2003). Epigenetics and the environment.
Ann. N Y Acad. Sci. 983, 151–160.
Urnov, F.D., and Wolffe, A.P. (2001). Chromatin remodeling and tran-
scriptional activation: the cast (in order of appearance). Oncogene 20,
Usenko, T., Kukushkin, A., Pospelova, T., and Pospelov, V. (2003). Tran-
sient expression of E1A and Ras oncogenes causes downregulation of
c-fos gene transcription in nontransformed REF52 cells. Oncogene 22,
846 Cell 123, 833–847, December 2, 2005 ª2005 Elsevier Inc.
Vermaak, D., Ahmad, K., and Henikoff, S. (2003). Maintenance of
chromatin states: an open-and-shut case. Curr. Opin. Cell Biol. 15,
G. (2003a). Beta-arrestin 2 functions as a G-protein-coupled receptor-
activated regulator of oncoprotein Mdm2. J. Biol. Chem. 278, 6363–
Wang, P., Wu, Y., Ge, X., Ma, L., and Pei, G. (2003b). Subcellular lo-
calization of beta-arrestins is determined by their intact N domain and
the nuclear export signal at the C terminus. J. Biol. Chem. 278,
Witherow, D.S., Garrison, T.R., Miller, W.E., and Lefkowitz, R.J. (2004).
beta-Arrestin inhibits NF-kappaB activity by means of its interaction
with the NF-kappaB inhibitor IkappaBalpha. Proc. Natl. Acad. Sci. USA
Yoshida, M., Kijima, M., Akita, M., and Beppu, T. (1990). Potent and spe-
cific inhibition of mammalian histone deacetylase both in vivo and in vitro
by trichostatin A. J. Biol. Chem. 265, 17174–17179.
Yu, V.C., Richards, M.L., and Sadee, W. (1986). A human neuroblastoma
cell line expresses mu and delta opioid receptor sites. J. Biol. Chem. 261,
Zhang, Z., Xin, S.M., Wu, G.X., Zhang, W.B., Ma, L., and Pei, G.
(1999). Endogenous delta-opioid and ORL1 receptors couple to phos-
phorylation and activation of p38 MAPK in NG108–15 cells and this is
regulated by protein kinase A and protein kinase C. J. Neurochem. 73,
Zhao, J., Pei, G., Huang, Y.L., Zhong, F.M., and Ma, L. (1997). Carboxyl
tor phosphorylation. Biochem. Biophys. Res. Commun. 238, 71–76.
Cell 123, 833–847, December 2, 2005 ª2005 Elsevier Inc. 847