Division of Molecular Psychiatry, Departments of Psychiatry and
Pharmacology, Yale University School of Medicine and Connecticut Mental
Health Center, New Haven, Connecticut (N.S., J.T., S.S.N., J.C.,
M.B.K., R.S.D.); and Department of Psychiatry, the University of Texas
Southwestern Medical Center, Dallas, Texas (C.S., E.J.N.)
To investigate the role of cAMP response element-binding protein (CREB)
in the adaptive responses to psychotropic drugs, we have developed
inducible, brain region-specific CREB transgenic mice using the
tetracycline-regulated gene expression system. The tetracycline
transactivator (tTA) was placed under the control of 1.8-kilobase
neuron-specific enolase (NSE) promoter for this purpose. Different
patterns of CREB overexpression were found in striatum, nucleus
accumbens, and cingulate cortex in different lines of bitransgenic
mice, and CREB expression was blocked by addition of doxycycline, an
analog of tetracycline. Overexpression of CREB influenced the
expression of other members of the CREB/ATF family of transcription
factors, consistent with previous reports. In addition, psychostimulant
induction of dynorphin, a neuropeptide regulated by drugs of abuse, was
up-regulated in striatum. Finally, there was a significant reduction in
cocaine-induced locomotor activity in the CREB bitransgenic mice. These
results are consistent with a role for CREB in mediating adaptive
changes that occur in response to drugs of abuse.
 |
Introduction |
A
variety of neurotransmitter and endocrine signals which elevate
intracellular cAMP levels activate gene transcription via a cAMP
response element (CRE) found in the promoter region of target genes
(Ziff, 1990
). The cAMP response element binding protein (CREB), a
substrate of cAMP-dependent protein kinase A, is able to bind to the
CRE and trans-activate gene expression (Hoeffler et al.,
1988; Yamamoto et al., 1988; Gonzalez and Montminy, 1989). In addition
to CREB, several other CRE-binding proteins make up the CREB/activation
transcription factor (ATF) family (Hai et al., 1989; Maekawa et al.,
1989; Hai and Curran, 1991). Binding to a CRE is dependent on
dimerization at the basic/leucine zipper domain in the carboxyl
terminus of CREB/ATF proteins (Hai et al., 1989; Hai and Curran, 1991).
CREB can form functional homodimers, or heterodimers with other
CREB/ATF family proteins. After phosphorylation of CREB at the
Ser133 residue and binding to a CRE site, the
CREB binding protein, a coactivator of CREB, binds to the
phosphorylated CREB. This leads to activation of a transcription
factor/RNA polymerase II complex that directly
trans-activates target gene expression. Ser133 can also be phosphorylated by a number of
protein kinases other than protein kinase A, including calcium
calmodulin-dependent protein kinase, protein kinase C, and ribosomal
S-6 kinase. This suggests that CREB plays an important role in
integrating intracellular cAMP and calcium signaling as well as
responses to neurotrophic factors (Dash et al., 1991; Matthews et al.,
1994).
CREB has been shown to play a critical role in different types of
neuronal function, including memory formation and circadian rhythm
(Ginty et al., 1993; Milner et al., 1998; Obrietan et al., 1998).
Furthermore, region-specific regulation of CREB has been implicated in
the long-term neuronal plasticity that underlies the actions of
psychotropic drugs (Duman et al., 1997, 2000; Nestler and Aghajanian,
1997). For example, chronic opiate or psychostimulant administration
up-regulates the phosphorylation or expression of CREB in specific
brain regions, such as the locus ceruleus or nucleus accumbens
(Guitart et al., 1992; Widnell et al., 1994; Cole et al., 1995; Widnell
et al., 1996; Lane-Ladd et al., 1997; Nestler, 1997; Nestler and
Aghajanian, 1997). A role for CREB in the behavioral actions of drugs
of abuse is also supported by studies of CREB
mutant mice or
viral mediated gene transfer of CREB (Maldonado et al., 1996; Carlezon
et al., 1998). Another series of studies demonstrates that
administration of antidepressants, but not drugs of abuse, increases
the phosphorylation and expression of CREB in the hippocampus and
cerebral cortex but not in nucleus accumbens or locus ceruleus (Nibuya
et al., 1996; Thome et al., 2000). These findings demonstrate
that CREB is differentially regulated by psychotropic drugs and may
play a role in the cellular and behavioral responses to these agents.
To further elucidate the function of CREB, we have developed inducible,
brain region-specific CREB transgenic mice. Conventional transgenic
mice, which constitutively overexpress a transgene from early
development, can often exhibit compensatory or developmental adaptations that make it difficult to interpret the actions of transgene overexpression. In addition, expression of a transgene throughout the brain, as well as in peripheral tissues, can lead to
effects that indirectly influence the function of specific populations
of neurons. The use of inducible and region-specific transgenic mice
circumvents these problems. We have used the tetracycline-regulated system, in which the expression of CREB is placed under the control of
the tetracycline-operated promoter (TetOp). TetOp-CREB is regulated by
the tetracycline transactivator (tTA), which is under the control of
the neuron-specific enolase (NSE) promoter (Chen et al., 1998; Kelz et
al., 1999). Addition of tetracycline induces a conformational change in
tTA and thereby blocks its ability to bind TetOp and trans-activate CREB expression. This is referred to as the
tetracycline-off system (Fig. 1).
View larger version (35K):
[in this window]
[in a new window]
|
Fig. 1.
Schematic diagram of the tetracycline-regulated gene
expression system. Gene 1 encodes tTA under the control of NSE
promoter. Gene 2 encodes a rat CREB cDNA under the control of TetOp.
Transgenic mice with gene 1 or 2 are developed independently, and these
two lines are crossed with each other to obtain NSE-tTA × TetOp-CREB bitransgenic mice. In NSE-tTA × TetOp-CREB
bitransgenic mice, tTA binds to TetOp and activates the transcription
of CREB. In this way, CREB is expressed under the control of NSE
promoter in a region-specific manner. In the presence of tetracycline
or its derivative, tetracycline binds to tTA and cause a conformational
change that blocks its binding to TetOp and CREB expression is turned
off.
|
|
In this study, we have characterized the expression of CREB in several
lines of inducible and region-specific transgenic mice, and we have
examined the effects of doxycycline, an analog of tetracycline, on the
expression of CREB in vivo. In addition, we report the regulation of
CREB/ATF1 family members, as well as possible target genes, in the CREB
transgenic mice. Moreover, we investigate the behavioral consequences
of CREB overexpression.
| |
Materials and Methods |
Development of Transgenic Mice and Drug Treatments.
The
methods for generating inducible and brain region-specific CREB
transgenic mice are described in our previous reports (Chen et al.,
1998; Kelz et al., 1999). The principle of the Tet-regulated system is
summarized in Fig. 1. In brief, plasmids that can drive tTA under the
control of 1.8-kilobase NSE promoter (designated pNSE-tTA) and an
another that can drive expression of rat CREB under the control of
the TetOp (designated pTetOp-CREB) were constructed. Linearized DNA
fragments from pNSE-tTA or pTetOp-CREB, which contain the promoter,
open reading frame, simian virus 40 intron, and poly(A)+ signal, were microinjected into
pronuclei of oocytes from SJL × C57BL6 mice. Transgenic mice with
a single transgene (either NSE-tTA or TetOP-CREB) were generated
independently. These two lines of mice were then crossed to develop
NSE-tTA × TetOp-CREB bitransgenic mice. Genotyping of generated
transgenic mice was performed by PCR with specific primer sets for the
NSE promoter or rat CREB cDNA using genomic DNA isolated from mouse
tail as template.
To turn off CREB overexpression, the NSE-tTA × TetOp-CREB
bitransgenic mice were given water containing doxycycline (50 µg/ml; Sigma, St. Louis, MO), a derivative of tetracycline, and 5% sucrose. Amphetamine (15 mg/kg), cocaine (10 mg/kg), or saline was administrated i.p. to examine the regulation of dynorphin expression. Animals were
killed 3 h after injection. All transgenic mice used in this study
were strictly maintained according to the guidelines from the National
Institutes of Health and institutional animal care.
Immunohistochemistry.
Immunohistochemistry was conducted as
described previously (Hiroi and Graybiel, 1996; Hiroi et al., 1997). In
brief, mice were perfused with 0.9% NaCl followed by 4%
paraformaldehyde in 0.1 M phosphate-buffered saline (PBS) after
anesthesia with sodium pentobarbital. Brains were removed and further
fixed with 4% paraformaldehyde in 0.1 M PBS for 16 h at 4°C.
The brains were stored in 20% (v/v) glycerol in 0.1 M PBS for 2 days
at 4°C. Coronal brain slices of 20 to 25 µm thickness were cut with
the use of a microtome. For CREB immunohistochemistry, slices were
blocked with 5% BSA in 0.01 M PBS for 30 min at room temperature and
then incubated with a polyclonal CREB antibody (1:500 dilution; Upstate
Biotechnology Incorporated, Lake Placid, NY) for 2 to 3 days at 4°C.
CREB immunoreactivity was detected using standard avidin-biotin
complex-diaminobenzidine method as described previously (Hiroi and
Graybiel, 1996; Hiroi et al., 1997).
Immunoblotting of CREB and Phospho-CREB.
Dissected striatum
and cerebellum from NSE-tTA × TetOp-CREB bitransgenic
(NSE-tTA+/TetOp-CREB+, +/+) mice or NSE-tTA monotransgenic (NSE-tTA+/TetOp-CREB
, +/) mice were homogenized and sonicated in
immunoprecipitation assay buffer (10 mM Tris-HCl, 1% Nonidet P-40,
0.1% sodium deoxycholate, 0.1% SDS, 0.15 M NaCl, 1 mM EDTA, 20 µg/ml leupeptin, and 1 mM phenylmethylsulfonyl fluoride, pH 7.4) and
centrifuged at 19,000g for 15 min at 4°C. Supernatants were subjected to SDS-polyacrylamide gel electrophoresis and
immunoblotting using a polyclonal CREB antibody (1:500 dilution;
Upstate Biotechnology Incorporated) as described previously (Takahashi
et al., 1999).
For phospho-CREB immunoblotting (pCREB), forskolin-stimulated striatal
slices were prepared. Briefly, mouse brains were quickly removed and
immediately immersed in ice-cold HEPES-buffered saline (136.7 mM NaCl,
5 mM KCl, 0.1 mM Na2HPO4,
0.2 mM KH2PO4, 2 mM CaCl2, 1 mM MgSO4, 1 mM
MgCl2, 16.6 mM glucose, 23.8 mM sucrose, and 9.84 mM HEPES, pH 7.4) agitated with
O2-CO2 (95:5). After 1 min
incubation in ice-cold HEPES buffered saline, brains were trimmed to
obtain striatum blocks. Striatum slices of 200 µm thickness were cut
using a vibratome (Technical Products International, Inc., St. Louis,
MO) and then slices were preincubated in HEPES-buffered saline agitated
with O2-CO2 (95:5) at
37°C for 45 min. After the preincubation, slices were treated with
200 µM forskolin (Sigma) for 15 min to activate the cAMP system and
induce CREB phosphorylation. Thereafter, treated slices were
homogenized in electrophoretic mobility shift assay buffer (20 mM
HEPES, 0.4 M NaCl, 20% glycerol, 5 mM MgCl2, 0.5 mM EDTA, 0.1 mM EGTA, 1% Nonidet P-40, 20 µg/ml leupeptin, and 1 mM
phenylmethylsulfonyl fluoride, pH 7.9), containing phosphatase
inhibitors (1 mM Na3VO4, 1 mM NaF, and 100 nM calyculin A), and incubated for 20 min on ice. After
centrifugation at 19,000g for 15 min, supernatants were
subjected to SDS-polyacrylamide gel electrophoresis, and phospho-CREB
immunoblotting was performed using a polyclonal phospho-CREB antibody
(1:500 dilution, New England Biolabs, Beverly, MA), as described
previously (Takahashi et al., 1999). To compare the precise size of the
phospho-CREB band, a lysate of forskolin-treated CATH.a cells was used
as positive control. CREB immunoblotting was carried out simultaneously
to examine the level of total CREB immunoreactivity.
Phospho-CREB Immunohistochemistry.
Immunohistochemistry for
pCREB was performed using a slide-based protocol. Cryocut sections (14 µm thick) were briefly fixed in 4% paraformaldehyde followed by
rinses in PBS. This was followed by a 5-min treatment with 1% hydrogen
peroxide and two 5-min rinses in PBS. Sections were blocked to prevent
nonspecific binding of antibody by a 30-min treatment in a blocking
solution containing 2.5% BSA.
Sections were incubated overnight at 4°C in antibody solution (0.25%
Triton X-100, 1% BSA in PBS) containing anti-phospho-CREB antibody (Upstate Biotechnology) at 1:500 dilution. Antibody-treated sections were given three 5-min washes with PBS to remove unbound primary antibody and were incubated in peroxidase-conjugated goat anti-rabbit IgG at 1:400 dilution for 1 h at room temperature. Unbound secondary antibody was removed by three 5-min rinses in PBS.
Sections were then incubated in a preformed avidin and biotinylated horseradish peroxidase macromolecular complex (ABC reagent; Vector Labs, Burlingame, CA) for 1 h. After this incubation, sections were rinsed three times in PBS and then stained with the
3,3'-diaminobenzidine staining kit (Vector Labs) according to
manufacturer's directions. Slides were air dried, dehydrated in an
alcohol series, and coverslipped with distyrene/plasticizer/xylene mountant.
RNase Protection and in Situ Hybridization Analysis.
Total
RNA was extracted from various regions of mouse brain using the
RNAqueous kit (Ambion, Austin, TX) according to the manufacturer's
protocol. RPA was performed with the RPAII kit (Ambion) according to
the manufacturer's protocol and using total RNA extracts from mouse
brain as templates.
To generate the CREB-specific riboprobes that distinguish the three
major CREB isoforms (CREB, CREB, and CREB
; Ruppert et al.,
1992; Blendy et al., 1996), reverse transcriptase (RT)-polymerase chain
reaction was first carried out to obtain partial mouse CREB cDNA. A
sense primer that recognizes exon 2 (designated primer 6) of the mouse
CREB gene and an antisense primer that recognizes exon 7 (designated
primer 7) were designed for RT-PCR. The sequences of primers 6 and 7 were 5'-TAAATGACCATGGAATCTGGAGCA-3' and 5'-AGTTACACTATCCACAGACTCCTG-3', respectively. RT-PCR was performed using the Access RT-PCR system (Promega, Madison, WI), according to the manufacturer's recommended method. Total striatal RNA was used as template for RT-PCR, followed by
agarose gel separation and ethidium bromide staining to visualize the
products. A DNA band of 318 bp, which corresponds to the CREB isoform, was isolated using a gel extraction kit (QIAGEN, Valencia, CA), subcloned into pGEM-T Easy vector (Promega), and verified by
sequencing. The resulting plasmid was designated pCREB 6-7. A
riboprobe was generated by linearizing pCREB 6-7 with
SpeI and 32P-labeled using SP6 RNA
polymerase (Roche Applied Science, Indianapolis, IN). Protected
fragments were loaded onto an 8% acrylamide-Tris/borate/EDTA gel and the separated bands were detected by autoradiography. Measurement and quantification of protected band density was carried out using the Macintosh version of the NIH image analysis program (version 1.52; http://rsb.info.nih.gov/nih-image/).
For CREM (CRE modulator)-specific RPA analysis, RT-PCR was also
performed, as described above, to obtain partial cDNA encoding the
CREM
isoform (Foulkes et al., 1992). The sequences of the sense and
antisense primers were 5'-GAAACAGTTGAATCACAGCAGGAT-3' and
5'-TGATTGAATAACCGATGGATGTGG-3', respectively. RT-PCR products were
subcloned (designated as pCREM). This riboprobe can distinguish between CREM and -, repressor isoforms of CREM, as well as
CREM, an activator isoform of CREM, by the size of protected
fragments. CREM and - are recognized as a single band of 105 bp
and CREM is a protected fragment of 231 bp.
To generate a probe for RPA of ATF1, RT-PCR was performed using mouse
ATF1 specific primers (Lee et al., 1992). The sense and antisense
primers were 5'-ATAGGCTCCTCACAGAAAGCTCAC-3' and 5'-TAATGTCTGCAGTGCCTGCACTCC-3', respectively. The resulting subcloned plasmid was designated pATF1. RPA using the pATF1 riboprobe recognizes an ATF1 mRNA band of 258 bp. Probes used for RPA of inducible cAMP
early repressor (ICER) was carried out as described previously (Fitzgerald et al., 1996a).
For RPA and in situ hybridization of dynorphin, a riboprobe was
generated from a plasmid encoding the major exon of rat prodynorphin gene, which was provided by Dr. C. R. Gerfen (see Carlezon et al.,
1998). Experimental procedures for in situ hybridization of dynorphin
were as described previously (Nibuya et al., 1996; Takahashi et al.,
1999). After ISH, sections were dipped in nuclear track emulsion (NTB2;
Eastman Kodak, Rochester, NY) after diluting emulsion (1:1) in 600 mM
ammonium acetate at 42°C. Slides were developed after 1 week exposure
in emulsion, counter-stained in cresyl violet (10-15 min), dehydrated
in an alcohol series, cleared in Histochoice (15 s; Amresco, Solon,
OH), and coverslipped with distyrene/plasticizer/xylene mountant.
Analysis of Cocaine-Induced Locomotor Activity in Bitransgenic
Mice.
Locomotor sensitization was conducted in mice as described
previously (Hiroi et al., 1997). Subjects were habituated to the test
apparatus over 5 consecutive days and then received cocaine (10 mg/kg,
i.p.) or saline injections for 5 consecutive days. Horizontal locomotor
activity was quantified by automated beam crossing for 60 min each day.
| |
Results |
Expression Pattern of CREB in NSE-tTA × TetOp-CREB
Bitransgenic Mice.
We have developed six lines of TetOp-CREB mice,
each of which was crossed with NSE-tTA lines A and B generated by our
group (Chen et al., 1998). Previous studies demonstrate that these
lines induce overexpression of a reporter gene
(luciferase) and another target gene
(FosB) under the TetOp promoter to varying degrees in
striatum, nucleus accumbens, cortex, and hippocampus (Chen et al.,
1998). Immunohistochemical analysis demonstrated significant levels of
CREB overexpression in three lines of bitransgenic mice: TetOp-CREB 6 line × NSE-tTA B line (CREB6-B line); TetOp-CREB 6 line × NSE-tTA A line (CREB6-A line); and TetOp-CREB 3 line × NSE-tTA A
line (CREB3-A line).
In the 6B line (NSE-tTA+/TetOp-CREB+, +/+), CREB overexpression
predominated in the dorsal and medial portion of striatum compared with
CREB staining in mice with only NSE-tTA (NSE-tTA+/TetOp-CREB, +/)
(Fig. 2). Higher magnification (400×)
demonstrates that CREB overexpression is restricted to the nucleus of
cells. CREB was also overexpressed in nucleus accumbens, primarily in
the core subdivision (Fig. 2, lower). In the CREB6-A line, CREB was
overexpressed in striatum and nucleus accumbens, similar to the
expression pattern in the CREB6-B line (summarized in Table
1). In addition, CREB was sparsely, but
significantly, overexpressed in deep and superficial layers of
cingulate cortex (Fig. 3). CREB
overexpression in deep layers of parietal cortex was also observed
(data not shown). No CREB overexpression was observed in hippocampus in
either the CREB6-A or -B lines (Table 1). However, low levels of CREB
overexpression were observed in hippocampus of the CREB3-A line (Fig.
3, lower trace). CREB was sparsely but clearly overexpressed in CA1
pyramidal cells and granular cells of dentate gyrus (Fig. 3, lower
trace). CREB overexpression was also observed in the dorsal striatum
and nucleus accumbens but at lower levels than in the CREB6-A or
CREB6-B lines (summarized in Table 1). There was no expression of CREB observed in monotransgenic TetOp-CREB mice (i.e., without the NSE-tTA),
demonstrating that there is no leak of the TetOp-CREB transgene (not
shown).
View larger version (74K):
[in this window]
[in a new window]
|
Fig. 2.
Pattern of CREB overexpression in striatum and
nucleus accumbens of CREB6-B bitransgenic mice. CREB expression was
determined by immunohistochemistry in mice expressing one
transgene (NSE-tTA+/TetOp-CREB, +/) and bitransgenic mice
(NSE-tTA+/TetOp-CREB+, +/+) mice. In bitransgenic mice, intense
staining of CREB is observed in many cells of the dorsal and medial
regions of striatum (upper trace) and in the core of nucleus accumbens
(lower trace). Weak nuclear staining of endogenous CREB is seen
throughout the striatum and nucleus accumbens of bitransgenic mice, as
well as in mice expressing only NSE-tTA. Higher magnification
demonstrates that CREB is localized in the nucleus. Results are
representative of the analysis of at least three animals in each group.
LV, lateral ventricle; cc, corpus callosum; ac, anterior commissure;
Icj, islands of Calleja.
|
|
View this table:
[in this window]
[in a new window]
|
TABLE 1
Pattern of CREB overexpression in three different lines of NSE-tTA × TetOp-CREB bigenic mice
Level of CREB overexpression is indicated by the number of + symbols.
Overexpression of CREB in cerebellum was confirmed by Western blotting
(see text).
|
|
View larger version (72K):
[in this window]
[in a new window]
|
Fig. 3.
Pattern of CREB overexpression in CREB6-A and CREB3-A
bitransgenic mice. In CREB6-A bitransgenic mice, CREB is overexpressed
in deep layers of cingulate cortex (+/+, upper trace). Arrowheads show
CREB-overexpressing cells. A small number of cells also exhibit dense
CREB staining in the superficial layer of cortex. Higher magnification
demonstrate that this staining is localized to the nucleus. In CREB3-A
bitransgenic mice (+/+), there is sparse but distinct CREB
overexpression in CA1 pyramidal and dentate gyrus granule cells of
hippocampus (lower trace). Results are representative of the analysis
of at least three animals in each group. cg, cingulum; DG, dentate
gyrus.
|
|
Overexpression of CREB was further examined by immunoblotting of brain
regions taken from CREB6-B mice. A CREB immunoreactive band of 43 kDa
was overexpressed in NSE-tTA × TetOp-CREB bitransgenic (+/+)
mice. CREB overexpression was found in striatum, as well as in
cerebellum of the 6B line (Fig. 4A). No
additional CREB immunoreactive bands were detected in +/+ mice compared
with the genetic controls (+/) (data not shown).
View larger version (51K):
[in this window]
[in a new window]
|
Fig. 4.
Immunoblot analysis of CREB overexpression in CREB6-B
bitransgenic mice. A, in both striatum and cerebellum, CREB
immunoreactive bands of approximately 43 kDa were detected; the
intensity of this band was greater in CREB6-B bitransgenic mice
relative to the +/ control mice. B, forskolin-stimulated
phosphorylation of CREB (pCREB) was enhanced in striatum of CREB6-B
bitransgenic mice. Striatal slices were freshly prepared and maintained
at 37°C in HEPES-buffered saline as described under Materials
and Methods. Slices were treated with 200 µM forskolin (F)
for 15 min and then subjected to pCREB immunoblot analysis. Samples
from forskolin-stimulated CATH.a cells, which express high levels of
CREB, were used as a positive control (right lane). The pCREB
immunoreactive bands were detected as a doublet just above an intense
nonspecific band (NS). The size of the pCREB bands in the bitransgenic
mice is the same as those in the forskolin-treated CATH.a cells. In
both the +/+ and +/ slices, levels of pCREB staining are greater
after incubation with forskolin, but the intensity of the bands is
greater in the CREB-overexpressing +/+ slices. Immunoblot analysis of
CREB demonstrates similar amounts of CREB regardless of the forskolin
incubation, but higher levels of CREB in the bitransgenic mice as
expected. Results are representative of three separate experiments.
|
|
To determine whether CREB phosphorylation was enhanced in CREB
overexpressing bitransgenic mice, phospho-CREB immunoreactivity in
striatal slices of CREB6-B +/+ and +/ mice was examined. In preliminary time course experiments, the induction of phospho-CREB by
forskolin was found to peak after 15 min and then gradually decrease
with longer periods of incubation. Therefore, a 15-min time point was
chosen for further studies. Stimulation with forskolin increased
phospho-CREB immunoreactivity relative to control, and this effect was
greater in the CREB6-B +/+ mice relative to the +/ mice (Fig. 4B).
Two phospho-CREB immunoreactive bands, which were just above an intense
nonspecific band, were observed in response to forskolin. The
phospho-CREB bands in the bitransgenic mice comigrate with the
phospho-CREB bands prepared from forskolin-treated CATH.a cells (Fig.
4B), which are also detected as a doublet upon shorter exposure (data
not shown). The expression of CREB is relatively high in the CATH.a
cells, and we have used them routinely as markers for CREB and
phospho-CREB (Widnell et al., 1994, 1996). The relatively lower level
of CREB and phospho-CREB immunoreactivity in the mouse brain compared
with CATH.a cells is consistent with the low levels observed in our
previous studies (Widnell et al., 1994, 1996). Figure 4B also
demonstrates that levels of total CREB immunoreactivity are higher in
the +/+ mice relative to the +/ mice.
The influence of doxycycline, a derivative of tetracycline, on the
expression of CREB in the bitransgenic mice was examined next. This was
necessary to confirm that the tTA-TetOp system was functioning
correctly in vivo. Application of doxycycline (50 µg/ml in water) for
25 days completely blocked CREB overexpression in the 6-B line (Fig.
5). Removal of doxycycline for 2 weeks
allowed for a nearly complete recovery of CREB overexpression (Fig. 5, right). These findings demonstrate that the expression of CREB can be
turned off and then back on upon addition and removal of doxycycline.
View larger version (71K):
[in this window]
[in a new window]
|
Fig. 5.
Influence of doxycycline, a derivative of
tetracycline, on the overexpression of CREB in striatum of CREB6-B
bitransgenic mice. Immunohistochemistry of CREB was performed to verify
the overexpression of CREB. Before the treatment with doxycycline, CREB
overexpression was seen in striatum near lateral ventricle, as seen in
Fig. 2 (left). After the treatment with doxycycline (50 µg in water)
for 25 days, CREB overexpression was almost completely abolished and
only staining of endogenous CREB was detected (middle). Overexpression
of CREB was once again observed 2 weeks after the removal of
doxycycline (right). Data are representative of two separate
experiments.
|
|
Regulation of CREB, ATF, and CREM Isoforms in the Bitransgenic
Mice.
Studies were conducted to determine whether overexpression
of CREB alters levels of endogenous CREB or CREB-related transcription factors, including CREM and ATF1. Expression of three major isoforms of
CREB (i.e., , , and ) (Ruppert et al., 1992; Blendy et al., 1996), were analyzed by RPA in NSE-tTA × TetOp-CREB bitransgenic mice. A riboprobe that can distinguish the CREB isoforms by the size of
protected fragments was prepared by RT-PCR (Fig.
6A). The relative levels of the three
isoforms are similar to what has been reported (Blendy et al., 1996).
In wild-type animals, levels of CREB are highest, with approximately
equal amounts of CREB and CREB. In striatum from either CREB6-A
or -B bitransgenic (+/+) mice, CREB mRNA, the isoform used to
develop the transgenic lines, is significantly increased as expected
(Fig. 6, B and C). In addition, levels of CREB and isoforms are
significantly up-regulated. The induction of the , as well as and , isoforms in the CREB6-A line was reversed by addition of
doxycycline to the drinking water (Fig. 6C).
View larger version (34K):
[in this window]
[in a new window]
|
Fig. 6.
Analysis of CREB, -, and - isoform
expression in bitransgenic mice. A, strategy for RNase protection
analysis (RPA) of CREB, -, and - isoforms. The mouse CREB gene
consists of 11 exons (Ruppert et al., 1992). All 11 exons are used for
the transcription of the CREB isoform, whereas CREB lacks exon 5 and CREB lacks exons 2 and 5 compared with CREB. For the
translation of CREB and CREB isoforms, the initiation site in
exon 2 is used, whereas the initiation site in exon 4 is used for
CREB. RT-PCR was carried out to get a partial CREB cDNA that was
used to generate probes for RPA. A sense primer (primer 6) and an
antisense primer (primer 7) were designed to recognize exon 2 and exon
7, respectively. A partial cDNA of CREB isoform (pCREB 6-7) was
obtained and used for making a riboprobe. This probe recognizes CREB
mRNA as two protected fragments of 264 bp and 54 bp, CREB as a
fragment of 318 bp, and CREB as a fragment of 201 bp. B, analysis of
CREB, -, and - isoforms by RPA in striatum of CREB6-A and -B
line mice. In both NSE-tTA × TetOp-CREB bitransgenic mice (+/+)
of CREB6-A and -B lines, all three major CREB isoforms (CREB, -,
and -) were up-regulated compared with the expression levels in
NSE-tTA monotransgenic (+/) mice or doxycycline-treated NSE-tTA × TetOp-CREB bitransgenic mice [+/+, Dox(+)]. A greater
up-regulation of the three CREB isoforms was observed in the CREB6-B
line. Results are expressed as mean percentage of control +/
mice ± S. E. M. (n = 4 or 5 as
indicated). *, P < 0.01, compared with control
(+/) mice (Student's t test).
|
|
Transcription factors in CREB/ATF family modulate transcription via CRE
sites, which are found in the promoter region of their target genes by
forming homo- or heterodimers with one another (Hai et al., 1989; Ziff,
1990; Hai and Curran, 1991). Within the CREB/ATF family of
transcription factors, CREB, CREM, and ATF1 are known to preferentially
form heterodimers with one another in vitro, indicating that these
three transcription factors function as partners for regulation of
CRE-mediated transcription in vivo (Hai and Curran, 1991). To examine
the influence of CREB overexpression on these related proteins, RPA
analysis of CREM and ATF1 was performed.
A riboprobe was generated for CREM, an activator of CRE-mediated
trans-activation that can discriminate between this isoform and CREM and , two repressors of CRE trans-activation,
by the size of the protected fragments (Fig.
7A). Both repressor and activator types
of CREM (CREM, , and , respectively) were down-regulated by
approximately 20% in striatum and cerebellum of CREB6-B bitransgenic mice (Fig. 7). Down-regulation of CREM was not observed in CREB6-A bitransgenic mice (data not shown). This may be related to the greater
induction of CREB in the CREB6-B line relative to the -A line, as
demonstrated in Fig. 6B.
View larger version (50K):
[in this window]
[in a new window]
|
Fig. 7.
Analysis of CREM (, , and ICER) and ATF1
expression in CREB6-B bitransgenic mice. Left, representative RPA
autoradiograms of CREM, ICER, and ATF1 protected fragments. As
described under Materials and Methods, protected
fragments of 231 bp (corresponding to CREM) and 105 bp
(corresponding to CREM) are activator and repressors of CRE
sites, respectively. The and isoforms are very similar and
cannot be distinguished with the probe used for these studies. Bands of
approximately 130 bp were unexpected and may correspond to another
unknown isoform of CREM (top). RPA for ICER demonstrates a single major
band that corresponds to ICER (middle) (Fitzgerald et al., 1996a).
Protected fragments for ATF1 are of 258 bp (bottom). In both striatum
and cerebellum of NSE-tTA × TetOp-CREB bitransgenic (+/+) mice,
CREM and - are down-regulated, compared with the expression
level in NSE-tTA monotransgenic (+/) mice or doxycycline-treated
bitransgenic mice [+/+, Dox(+)]. Expression levels of ICER and ATF1
were unchanged in striatum and cerebellum of CREB6-B bitransgenic mice.
Results are expressed as mean percentage of control +/ mice ± S.E.M. (n = 3 to 10 as indicated). *,
P < 0.05; **, P < 0.01, compared with +/ mice or +/+, Dox(+) mice (Student's
t test).
|
|
We also examined the regulation of ICER, another repressor isoform of
CREM. ICER is generated by use of a different promoter in the
CREM gene than that used for generating CREM, -, and - (Molina et al., 1993). Basal levels of ICER mRNA were not
significantly influenced in striatum of CREB6-B bitransgenic mice (Fig.
7). Because ICER is rapidly induced by activation of the CREB,
induction of this isoform in response to amphetamine was also examined
in control and CREB-6B bitransgenic mice. Induction of ICER in response to amphetamine was observed in both single (NSE-tTA) and bitransgenic (NSE-tTA × TetOp-CREB) mice, but there was no significant
difference between these two groups [273 ± 22.6 and 228 ± 14.6% of saline, for single and bitransgenic CREB6-B mice,
respectively (mean ± S.E.M., n = 10 per group)].
There was also no significant difference in levels of mRNA for ATF1,
another member of the CREB/ATF family, in striatum or cerebellum of the
single CREB6-B bitransgenic line (Fig. 7).
Regulation of Dynorphin Expression in NSE-tTA × TetOp-CREB
Bitransgenic Mice.
Previous reports demonstrate that chronic
administration of a psychostimulant (e.g., cocaine or amphetamine)
up-regulates the cAMP pathway, which leads to activation of
CRE/CREB-mediated transcription in nucleus accumbens and dorsal
striatum (Terwilliger et al., 1991; Konradi et al., 1994; Cole et al.,
1995; Hyman, 1996; Unterwald et al., 1996; Nestler, 1997; Nestler and
Aghajanian, 1997; Turgeon et al., 1997). In addition, dynorphin has
been shown to be a target of CREB-mediated trans-activation
in these regions after chronic cocaine or amphetamine treatment (Hurd
et al., 1992; Daunais et al., 1993; Spangler et al., 1993; Cole et al.,
1995; Carlezon et al., 1998). The promoter region of the rat dynorphin gene has three predicted CRE sites, and promoter activity is increased by stimulation of receptors that activate the cAMP pathway in vitro
(Collins-Hicok et al., 1994). These findings led us to examine the
regulation of dynorphin in the dorsal striatum and nucleus accumbens of
NSE-tTA × TetOp-CREB bitransgenic mice.
Expression of dynorphin was determined under two conditions. In the
first paradigm, CREB6-B bitransgenic mice were raised in the absence of
doxycycline (CREB on) throughout development. This would be similar to
a constitutive transgenic line of mice in which there is no control
over CREB overexpression. In the second paradigm, mice were raised on
doxycycline (CREB off) until they were weaned and then withdrawn from
doxycycline for 6 weeks before analysis (CREB on). In both cases,
monotransgenic mice (+/) were used as a control. Using the first
paradigm, in which CREB overexpression is on throughout development, we
found that basal dynorphin expression was depressed in striatum of
CREB6-B bitransgenic mice relative to the +/ control mice (Fig.
8). In NSE-tTA × TetOp-CREB
bitransgenic (+/+) mice, dynorphin mRNA was decreased by 40 to 60%,
determined by either RPA or in situ hybridization, compared with
control +/ mice (Fig. 8A). Basal expression of dynorphin was also
significantly reduced in the striatum of CREB6-A bitransgenic mice
(data not shown). Induction of dynorphin mRNA in response to a
relatively high dose of amphetamine (15 mg/kg, i.p.) was also examined.
Levels of dynorphin mRNA were determined 3 h after amphetamine
administration. In both CREB6-B +/+ and +/ mice, amphetamine
increased dynorphin expression, but the induction relative to the basal
state was greater in the CREB6-B +/+ mice than in the control +/ mice
(Fig. 8B).
View larger version (39K):
[in this window]
[in a new window]
|
Fig. 8.
Regulation of dynorphin expression in bitransgenic
mice expressing CREB since birth. Both monotransgenic
(NSE-tTA+/TetOp-CREB) (+/) and bitransgenic ((NSE-tTA
+/TetOp-CREB+) (+/+) mice were raised in the absence of doxycycline
from the time of birth. A, the expression of dynorphin mRNA in striatum
was examined by RPA and in situ hybridization. A representative
autoradiogram of the RPA shows two major protected dynorphin fragments
of approximately 150 bp. In situ hybridization analysis demonstrates
high levels of dynorphin mRNA in striatum. Levels of dynorphin
expression in +/+ mice are reduced by 60 and 40% when determined by
RPA and in situ hybridization, respectively, compared with the control
+/ mice. Results are expressed as mean percentage of control +/
mice ± S.E.M. (n = 9 or 10 as indicated).
*, P < 0.01, compared with the density in +/
mice (Student's t test). B, regulation of dynorphin by
a high dose of amphetamine. Amphetamine (15 mg/kg, i.p.) or saline was
administered and animals were killed 3 h later. Amphetamine
administration increased dynorphin expression in both lines of mice but
the effect was more prominent in the CREB bitransgenic animals. Results
are expressed as mean percentage of control +/ mice ± S.E.M.
(n = 5). *, P < 0.1; **,
P < 0.05, compared with the level in
saline-treated mice (Student's t test).
|
|
The results obtained with the second paradigm, in which CREB expression
is not induced until after weaning, resulted in a somewhat different
profile. For these studies, expression of dynorphin was visualized by
examination of grains over individual cells. This allowed us to look
specifically in regions in which there is a high level of CREB
overexpression, the medial lateral nucleus accumbens (see Fig. 2). In
the naive CREB overexpressing mice, there was an increase in dynorphin
expression in this region of the nucleus accumbens (Fig.
9, A and B). Moreover, administration of
amphetamine resulted in a greater induction of dynorphin in the CREB
overexpressing mice relative to the monotransgenic control mice (Fig.
9, C and D). Levels of pCREB were also determined in adjacent sections
by immunohistochemistry. Amphetamine-induction of pCREB was
significantly elevated in the CREB overexpressing mice relative to the
monotransgenic control mice expressing endogenous levels of CREB (Fig.
9, E and F). This demonstrates that increased expression of CREB
results in increased levels of pCREB and increased expression of
dynorphin. There was also a tendency for levels of pCREB to be
increased in the vehicle treated CREB overexpressing mice (data not
shown).
View larger version (152K):
[in this window]
[in a new window]
|
Fig. 9.
Regulation of dynorphin expression in bitransgenic
mice expressing CREB after weaning. Both monotransgenic
(NSE-tTA+/TetOp-CREB) (CREB Off) and bitransgenic
(NSE-tTA+/TetOp-CREB+) (CREB On) mice were raised on doxycycline
until weaning. Six weeks later, the mice were treated with vehicle (A,
B) or amphetamine (15 mg/kg) (C-F) as indicated. A to D, dynorphin mRNA
was determined by in situ hybridization. The sections were then dipped
in emulsion and counterstained with cresyl violet. E and F, levels
of pCREB were determined by immunohistochemistry using a
phosphospecific antibody. Four separate animals were examined in
duplicate for each condition and representative figures are shown for
each.
|
|
Regulation of Cocaine-Induced Locomotor Activity in CREB
Bitransgenic Mice.
Because of the high level of CREB
overexpression in the dorsal striatum and nucleus accumbens and the
regulation of dynorphin expression by cocaine, we examined
cocaine-induced locomotor activity in the CREB bitransgenic mice. The
responses to a single dose of cocaine, as well as to repeated
administration of cocaine, which is known to produce locomotor
activation, were examined. For these studies, bitransgenic mice
maintained on doxycycline in the drinking water (CREB off) were
compared with bitransgenic littermates withdrawn from doxycycline at
weaning (CREB on). This paradigm was chosen for the behavioral studies
because all animals have exactly the same genotype. First, locomotor
activity after administration of saline was determined on 5 consecutive
days. This allows for analysis of baseline locomotor activity and
provides a period of time when the animals can habituate to the test
chambers. There was no significant difference between the CREB-on and
CREB-off bitransgenic mice when they were first exposed to the chamber (day 1) or upon subsequent exposures (Fig.
10). Administration of cocaine (10 mg/kg) increased locomotor activity on day 1 and this effect was
similar in both the CREB-on and -off groups. Repeated cocaine
administration on subsequent days resulted in a greater increase in
locomotor activity as expected in both groups. However, this effect was
significantly lower in the CREB-on group relative to the CREB-off
animals (Fig. 10). These findings are consistent with previous reports
demonstrating that viral-mediated CREB expression in the nucleus
accumbens decreases behavioral responses to cocaine (Carlezon et al.,
1998).
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 10.
Analysis of cocaine-induced locomotor sensitization
in CREB bitransgenic mice. CREB bitransgenic mice were maintained
either in the presence (CREB Off) or absence (CREB On) of doxycycline
in the drinking water. The mice were then tested for baseline locomotor
activity for 5 days, which also allowed the animals to become
habituated to the activity chambers. There was no significant
difference between the CREB-on and -off groups upon either the initial
or subsequent exposures to the activity chamber. Animals were then
administered cocaine (10 mg/kg, i.p.) each day for 5 days. The initial
response to cocaine was identical in both the CREB-on and -off groups.
However, the enhanced response to cocaine on each consecutive day was
lower in the CREB-on group relative to the CREB-off group. The results
are expressed as mean activity counts ± S.E.M.
(n = 25 per group). *, P < 0.05 compared with CREB off (analysis of variance and Fisher's post
hoc test. **, P < 0.07.
|
|
| |
Discussion |
Characterization of NSE-tTA × TetOp-CREB Bitransgenic
Mice.
In the present study, we characterized three lines of
NSE-tTA × TetOp-CREB bitransgenic mice (CREB6-A, CREB6-B, and
CREB3-A lines). In all three lines, CREB was overexpressed in dorsal
striatum and nucleus accumbens (Table 1). In the CREB6-A line, CREB
overexpression was also observed in deep layers of cortex, whereas in
the CREB3-A line, CREB was overexpressed, albeit sparsely, in
hippocampal CA1 pyramidal and dentate gyrus granule cell layers (Figs.
2 and 3, Table 1). In the CREB6-B line, immunoblotting also
demonstrated overexpression of CREB in striatum and in cerebellum (Fig.
4A). Analysis of CREB mRNA and protein revealed that levels of
expression in dorsal striatum and nucleus accumbens are more prominent
in CREB6-B than in CREB6-A mice. These patterns of CREB expression in A
and B NSE-tTA lines are similar to those seen in TetOp-luciferase reporter mice crossed with these lines (Chen et al., 1998). This indicates that the expression pattern of the TetOp-driven target gene
is determined largely by the expression pattern of NSE-tTA. The results
also demonstrate that CREB overexpression can be turned off and on by
addition or removal, respectively, of doxycycline, an analog of
tetracycline. It is notable that a relatively low dose of doxycycline
(50 µg/ml in water) can inhibit CREB overexpression. The low dose
used could also explain the relatively short washout period required
for CREB expression after the removal of doxycycline. These findings
demonstrate that the tTA-TetOP system functions correctly in the CREB
overexpressing bitransgenic mice.
High-power magnification of brain sections from bitransgenic mice
demonstrates that CREB is localized to the cell nucleus, as would be
expected for this transcription factor (Fig. 2 and 3). In addition,
activation of the cAMP pathway results in increased CREB
phosphorylation; this effect is greater in the CREB bitransgenic mice
(Fig. 4B). These findings demonstrate that CREB is localized to the
appropriate cellular compartment and that it can be regulated by
activation of the cAMP cascade. However, it is important to point out
that overexpression of transgenic CREB may not result in up-regulation
of CRE-mediated gene expression, because endogenous CREB is already
expressed in most cells. The functional outcomes of CREB overexpression
must be examined by analysis of target genes that are regulated by
CREB, as well as by analysis of the behavioral phenotype of the
bitransgenic CREB mice. In addition, it will be important to develop
additional lines of mice to further examine the function of endogenous
CREB. This could include transgenic lines that overexpress a form of
CREB that is an active or a dominant negative form of CREB. With regard
to the latter, we have made a bitransgenic line that expresses a
dominant negative phosphorylation mutant and we are currently
characterizing the neurochemical and behavioral phenotype of these animals.
Regulation of CREB/ATF-Like Transcription Factors in CREB
Bitransgenic Mice.
Within the CREB/ATF family of transcription
factors, CREB, CREM, and ATF1 are known to heterodimerize with one
another in vitro (Hai and Curran, 1991). In addition, these
transcription factors are regulated by the cAMP pathway and share
sequence homology (Brindle and Montominy, 1992; Lee and Masson, 1993).
Based on homology and sequence conservation, the CREB and
CREM genes are thought to have been formed by duplication of
an ancestral gene (Ruppert et al., 1992). Previous reports have
revealed that the CREB isoform and CREM are up-regulated in mice
that have a null mutation of CREB and - isoforms (Hummler et al.,
1994; Blendy et al., 1996). These results indicate that null mutation
of CREB/CREB results in compensatory expression of other members
of this transcription factor family. Here we studied what the
compensatory responses of CREB-related proteins (CREB, CREM, ICER, and
ATF1 isoforms) might be in response to overexpression of CREB in
specific brain regions, which has not yet been examined. The results
demonstrate that CREB and CREB isoforms are up-regulated in
CREB-overexpressing mice. Levels of ATF1 and ICER were not altered
significantly in the bitransgenic mice.
The up-regulation of CREB and CREB is consistent with previous
reports that activation of the cAMP pathway in most cell types results
in up-regulation of CREB gene expression (Meyer et al.,
1993; Walker et al., 1995; Coven et al., 1998). Sequence analysis
demonstrates that the CREB promoter contains three CRE elements that mediate the up-regulation of CREB gene
expression in response to stimulation of the cAMP system (Coven et al.,
1998). A similar mechanism could underlie the up-regulation of CREB and CREB observed in CREB-overexpressing mice. These findings seem
to contradict the up-regulation of CREB in the CREB/ null mutant mice (Blendy et al., 1996). However, increased CREB
expression in the latter study was reported to result from alternative
splicing and increased mRNA stability, not increased CREB
gene expression.
The mechanisms underlying the down-regulation of CREM (CREM, ,
and isoforms) in CREB-overexpressing bitransgenic mice seem to
involve another mechanism. There are no CRE elements in the promoter of
the CREM gene that control the expression of these isoforms (Molina et
al., 1993). However, previous studies have reported that CREM
expression during spermatogenesis is controlled by post-transcriptional
mechanisms, including alternative splicing and increased mRNA stability
(Foulkes et al., 1993). It is possible that down-regulation of the CREM
isoforms observed in the present study results from similar mechanisms.
There is a CRE in an intronic promoter in the CREM gene that
controls the induction of ICER (Molina et al., 1993). ICER contains
only the DNA binding domain and acts as a repressor of CRE-mediated
gene expression. It is surprising that expression of ICER is not
up-regulated in the CREB-overexpressing mice. However, ICER has been
shown to rapidly counter-regulate its own induction (Molina et al.,
1993), and it is possible that a similar mechanism occurs in the
CREB-overexpressing mice. The induction of ICER in response to
amphetamine, although not different between the single and bitransgenic
CREB mice, demonstrates the relatively normal responsiveness of this
inducible isoform in the transgenic animals.
Regulation of Dynorphin in CREB Bitransgenic Mice.
In
bitransgenic mice in which CREB is overexpressed throughout
development, we found that dynorphin expression was reduced relative to
the littermate monotransgenic control mice. In contrast, in
bitransgenic mice in which CREB overexpression is turned on only after
weaning, there is an increase in basal dynorphin expression and the
induction of dynorphin by amphetamine is increased relative to
littermate control mice expressing normal levels of CREB. The repression of basal dynorphin expression could represent a compensatory response to long-term overexpression of CREB. For example,
overexpression of CREB in cultured cells is reported to decrease
dynorphin promoter activity (Collins-Hicok et al., 1994). One possible
explanation is that in the absence of phosphorylation, the
overexpressed CREB could compete with endogenous, phosphorylated CREB
for the CRE sites and could thereby act as a repressor of dynorphin
gene expression. CREB could also compete for other activator proteins
(e.g., ATF-2, AP-1) and thereby repress dynorphin expression. Yet
another possibility is that overexpressed CREB could interact with
other regulatory elements in the prodynorphin gene, such as a
downstream regulatory element (DRE) and its binding protein
(DRE-antagonist modulator), which represses basal prodynorphin
transcription (Carrion et al., 1998, 1999).
Activation of the cAMP pathways is reported to increase dynorphin
promoter activity in cultured cells (Collins-Hicok et al., 1994).
Studies conducted in primary striatal cultures confirm the induction of
dynorphin gene expression by activation of the cAMP pathway and
demonstrate that this effect is mediated by three CRE elements in the
prodynorphin promoter (Cole et al., 1995). Administration of
amphetamine or cocaine is reported to increase the expression of
dynorphin in the striatum and nucleus accumbens (Hurd et al., 1992;
Daunais et al., 1993; Spangler et al., 1993; Cole et al., 1995;
Carlezon et al., 1998). In this study, we demonstrate that
administration of amphetamine increases dynorphin expression and that
this effect is greater in the CREB-overexpressing mice. Moreover, the
induction of dynorphin expression was observed in an area of the
nucleus accumbens in which the expression of CREB, as well as pCREB, is
greatest in the CREB bitransgenic mice. Taken together, the results
indicate that overexpression of CREB in bitransgenic mice increases
psychostimulant-induced expression of dynorphin. Moreover, the
repression of basal levels of dynorphin expression after long-term
expression of CREB, and then psychostimulant induction is consistent
with studies of the dynorphin promoter in vitro (Collins-Hicok et al.,
1994).
Cocaine-Induced Locomotor Activity Is Decreased in
CREB-Overexpressing Mice.
The dorsal striatum and nucleus
accumbens are targets of the mesolimbic dopamine system and are known
to play a prominent role in mediating behavioral responses to drugs of
abuse (Nestler and Aghajanian, 1997). Moreover, adaptations of CREB and
its target genes, such as dynorphin, in these regions have been
implicated in the chronic actions of cocaine and other drugs of abuse
(Guitart et al., 1992; Hurd et al., 1992; Spangler et al., 1993;
Widnell et al., 1994; Cole et al., 1995; Nestler and Aghajanian, 1997; Carlezon et al., 1998). The results of the present study are consistent with these reports and demonstrate that behavioral responses to cocaine
are altered in CREB bitransgenic mice. We found that repeated cocaine
treatment for three days or longer caused significantly less locomotor
activation in bitransgenic mice relative to monotransgenic littermate
control mice. In contrast, there was no difference in the response to
the first day of cocaine administration, or in response to saline
administration over 5 days of exposure to the test chambers. These
finding indicate that CREB overexpression in bitransgenic mice alters
the response to repeated cocaine administration, but not to acute
cocaine or basal levels of locomotor activity.
The mechanism underlying the down-regulation of cocaine-induced
locomotor activation in the CREB bitransgenic mice could involve expression of dynorphin. Dynorphin acts at 
-opioid receptors on
dopaminergic terminals in striatum to decrease dopamine release and
thereby serve as a feedback inhibitor of dopamine transmission (see
Carlezon et al., 1998). Repeated cocaine administration results in a
greater induction of dynorphin, which could lead to a greater feedback
inhibition of dopamine transmission. Thus, the reduced locomotor
response could result from a greater up-regulation of dynorphin in
response to repeated cocaine treatment in the CREB bitransgenic mice.
This hypothesis must be further tested by analysis of dynorphin peptide
expression in the CREB bitransgenic mice. The reason for the delay in
the response is not clear, but it may be related to the requirement for
up-regulation of other components of the cAMP cascade (e.g., adenylyl
cyclase or cAMP-dependent protein kinase) that are necessary for full
activation of the overexpressed CREB.
Conclusions.
The results outlined in this article represent
the initial characterization of inducible CREB overexpressing mice, and
additional studies will be needed to further characterize the
biochemical and behavioral phenotype of these animals. Studies are
currently underway to identify other gene targets regulated by CREB,
including glutamate and dopamine receptor subtypes, other
neuropeptides, and neurotrophic factors. The behavioral phenotype of
these mice will be further characterized in other models of drug abuse
(e.g., place preference and self-administration), as well as in
behavioral models of depression and anxiety. The results presented here
indicate that the CREB over-expressing bitransgenic mice will be
beneficial and unique tools to investigate the mechanisms underlying
neural plasticity in response to a variety of behavioral and
pharmacological stimuli.
This work was supported by United States Public Health Service
grants MH45481 and 2-PO1-MH25642, a Veterans Administration National
Center for Post Traumatic Stress Disorder grant, and the Connecticut
Mental Health Center.
CRE, cAMP responsive element;
CREB, cAMP
response element binding protein;
ATF, activation transcription factor;
tTA, tetracycline transactivator;
TetOp, tetracycline-operated
promoter;
NSE, neuron-specific enolase;
PCR, polymerase chain reaction;
PBS, phosphate-buffered saline;
BSA, bovine serum albumin;
pCREB, phospho-cAMP response element binding protein;
RPA, RNase protection
analysis;
RT, reverse transcriptase;
bp, base pair(s);
CREM, cAMP
response element modulator;
ICER, inducible cAMP early repressor;
DRE, downstream regulatory element.
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Top
Abstract
Introduction
