Psychopharmacology (2012) 219:699–713 DOI 10.1007/s00213-011-2406-1
ORIGINAL INVESTIGATION
CREB1 and CREB-binding protein in striatal medium spiny neurons regulate behavioural responses to psychostimulants Heather B. Madsen & Srigala Navaratnarajah & Jessica Farrugia & Elvan Djouma & Michelle Ehrlich & Theo Mantamadiotis & Jan Van Deursen & Andrew J. Lawrence
Received: 11 January 2011 / Accepted: 18 June 2011 / Published online: 16 July 2011 # Springer-Verlag 2011
Abstract Rationale The transcription factor cAMP responsive element-binding protein 1 (CREB1) has a complex influence on behavioural responses to drugs of abuse which varies depending on the brain region in which it is expressed. In response to drug exposure, CREB1 is
H. B. Madsen : S. Navaratnarajah : J. Farrugia : A. J. Lawrence Florey Neuroscience Institutes, Parkville, VIC, Australia J. Farrugia : E. Djouma Department of Human Physiology and Anatomy, La Trobe University, Bundoora, Melbourne, VIC, Australia M. Ehrlich Departments of Neurology, Pediatrics, and Genetics and Genomic Sciences, Mount Sinai School of Medicine, New York, NY, USA T. Mantamadiotis Laboratory of Physiology, University of Patras, 26500 Patras, Greece J. Van Deursen Department of Pediatric and Adolescent Medicine, Mayo Clinic College of Medicine, Rochester, MN, USA A. J. Lawrence Centre for Neuroscience, University of Melbourne, Parkville, VIC, Australia A. J. Lawrence (*) Addiction Neuroscience, Florey Neuroscience Institutes, Royal Parade, Parkville, VIC 3010, Australia e-mail:
[email protected]
phosphorylated in the striatum, a structure that is critically involved in reward-related learning. Objective The present study assessed the role of striatal CREB1 and its coactivator CREB-binding protein (CBP) in behavioural responses to psychostimulants. Methods Using the ‘cre/lox’ recombination system, we generated mice with a postnatal deletion of CREB1 or CBP directed to medium spiny neurons of the striatum. qRT-PCR and immunohistochemistry were used to confirm the deletion, and mice were assessed with respect to their locomotor response to acute cocaine (20 mg/kg), cocaine sensitization (10 mg/kg), amphetamine-induced stereotypies (10 mg/kg) and ethanol-induced hypnosis (3.5 g/kg). Results Here we show that CREB1 mutant mice have increased sensitivity to psychostimulants, an effect that does not generalise to ethanol-induced hypnosis. Furthermore, in the absence of CREB1, there is rapid postnatal upregulation of the related transcription factor CREM, indicating possible redundancy amongst this family of transcription factors. Finally striatal deletion of CBP, a coactivator for the CREB1/CREM signalling pathway, results in an even more increased sensitivity to psychostimulants. Conclusions These data suggest that striatal CREB1 regulates sensitivity to psychostimulants and that CREM acting via CBP is able to partially compensate in the absence of CREB1 signalling. Keywords CBP . CREB . DARPP-32 . Psychostimulants . Striatum
Introduction Drug addiction is a chronic, relapsing disease which represents a major problem in modern society, costing
700
billions of dollars each year (Rice 1999). Characterised as compulsive drug use despite serious negative consequences, a major challenge in the treatment of drug addiction lies in preventing relapse, a risk that remains high even following protracted abstinence (O'Brien 1997). The transition from casual to compulsive drug use and the enduring susceptibility to relapse is thought to be underpinned by long-lasting neuroplasticity as a result of repeated drug exposure (Kalivas and O'Brien 2008; Nestler 2001; Thomas et al. 2008). The transcription factor cAMP responsive element-binding protein 1 (CREB1) has been identified as a key molecular substrate involved in drug-induced plasticity (Carlezon et al. 1998; McPherson and Lawrence 2007; Walters and Blendy 2001). CREB1 is a member of the basic domain leucine zipper (bZIP) family of transcription factors which also includes the closely related cAMP response element modulator (CREM, Shaywitz and Greenberg 1999). Activation of both CREB1 and CREM usually involves dimerisation, phosphorylation and the recruitment of the transcriptional coactivator CREB-binding protein (CBP) which initiates the transcription of various downstream genes (De Cesare et al. 1999). CBP is able to enhance transcription via a number of mechanisms including recruitment and stabilisation of RNA polymerase II to the binding complex and histone acetylation (Bannister and Kouzarides 1996; Nakajima et al. 1997a, b). The exact role of CREB1 in mediating behavioural responses to drugs of abuse is complex and varies according to the drug and also the brain region. Viral-mediated gene transfer has been used to selectively inhibit or overexpress CREB1 in specific brain regions associated with drug reward. Blocking CREB1 activity in the nucleus accumbens (NAc) has been found to enhance the rewarding effects of cocaine, whereas the opposite occurs in response to CREB1 overexpression (Carlezon et al. 1998). Disruption of CREB1 within the ventral tegmental area (VTA) can both enhance or dampen drug reward depending on whether the disruption is rostral or caudal (Olson et al. 2005). To study the effects of prolonged CREB1 disruption, several transgenic mouse lines have been engineered. An attempt to generate a CREB1 null mouse lacking all the functional isoforms of CREB1 both centrally and peripherally resulted in early perinatal death, highlighting the importance of CREB1 in early development (Rudolph et al. 1998). In contrast, mice specifically lacking the α and Δ CREB1 isoforms have no developmental or growth impairments but exhibit an ubiquitous upregulation of the β CREB1 isoform and also CREM, demonstrating the inherent plasticity within this family of transcription factors (Blendy et al. 1996; Hummler et al. 1994). Germline knockouts such as the CREB1α/Δ mutant have their limitations because loss of the targeted gene in all cells and tissues from the earliest time of expression invariably leads to compensatory changes during development.
Psychopharmacology (2012) 219:699–713
Composed primarily of medium spiny GABAergic neurons, the striatum represents a major integrative centre of motivational and sensorimotor input (Gerfen 1992), and plays an important role in addiction to psychostimulants (Belin and Everitt 2008; Vanderschuren et al. 2005). Functional differences between the roles of the dorsal vs ventral striatum have also been noted, the ventral striatum implicated in the acquisition of drug taking and the dorsal striatum more involved once drug taking is established (Porrino et al. 2004; Vanderschuren et al. 2005). CREB1 activity in the striatum is altered in response to drug exposure, with rapid induction of pCREB1 observed after administration of both amphetamine (Choe et al. 2002; Konradi et al. 1994) and cocaine (Kano et al. 1995; Walters and Blendy 2001). The aim of the present study was to investigate the role of striatal CREB1 in behavioural responses to psychostimulants. Using the DARPP-32 promoter to drive expression of Cre recombinase (Bogush et al. 2005), we generated mice with a postnatal deletion of CREB1 specific to medium spiny neurons of the striatum, and characterised them with respect to their locomotor sensitization to cocaine and amphetamine-induced stereotypy. Because an upregulation of the related transcription factor CREM was observed in the CREB1DARPP-32Cre/loxlox mutants, we also generated mice with a striatal deletion of CBP, a coactivator for the CREB1/CREM signalling pathway.
Materials and methods Animals All mice used in this study were backcrossed onto a C57BL/6J background and were typically 8–10 weeks old at the commencement of experimentation. Two separate breeding colonies of double transgenics were established by crossing mice that express Cre recombinase driven by the DARPP-32 promoter (DARPP-32Cre/+, Bogush et al. 2005) with mice expressing either “floxed” Creb1 (CREB1lox/lox, Mantamadiotis et al. 2002) or “floxed” cbp (CBPlox/lox, Kang-Decker et al. 2004). This was to produce F2 progeny with the genotype CREB1DARPP-32Cre/loxlox or CBPDARPP-32Cre/loxlox. Remaining mice generated from this breeding were used as littermate controls. Only male mice were used for behavioural experiments and in situ hybridisation. Female mice were used for all other in vitro studies in order to maximise the use of double transgenic mice. All experiments described were performed in accordance with the Prevention of Cruelty to Animals Act 1986, under the guidelines of the National Health and Medical Research Council of Australia Code of Practice for the Care and Use of Animals for Experimental Purposes in Australia. During the course of the study, all mice were group housed in standard
Psychopharmacology (2012) 219:699–713
701
mouse boxes with nesting material available. Animals were kept in a constant 12-h light/dark cycle (light 0700–1900 hours) with ad libitum access to water and standard mouse chow. Where behavioural experiments took place in an environ separate to where the animals were housed, they were allowed at least 30 min to habituate to the experimental environment.
C r e m ( F - A G T C C C C A G C A A C TA G C A G A , R GATTTTCAAGCACAGCCACA) C r e b 1 ( F - C C A A A C TA G C A G T G G G C A G T, R GAATGGTAGTACCCGGCTGA) Creb1 excised exon (F-CCTCAGGCGATGTACAAACA, R-CTCTCTTCCGTGCTGCTTCT)
PCR genotyping All mice were genotyped using PCR as previously described. The following forward and reverse primers were used: Cre recombinase 5′-GGACATGTTCAGG GATCGCCAGGCG-3′, 5′-GCATAACCAGTGAAACAGCATTGCTG-3′ (Kochilas et al. 2002), Creb1lox 5′-TATGTA AA GC AA GG GA AG ATA ATG -3′, 5′-TAG ACA TACTTGACCCATAGCATT-3′ (Mantamadiotis et al. 1998) and CBPlox 5′-GGGGAAATTTTGGCTGGCAAG-3′, 5′CTGCTCTACCTAAATTCCCAG-3′ (Kang-Decker et al. 2004). All primers were purchased from Geneworks, Hindmarsh, Australia.
β - a c t i n ( F - G AT C T G G C A C C A C A C C T T C T, R GGGGTGTTGAAGGTCTCAAA) H p r t 1 ( F - C T T T G C T G A C T T G C T G G AT T, R TATGTCCCCCGTTGACTGAT) Mthfd 1 (F-AAGGAAAGTCGTGGGTGAT G, RGCTGTGCGCTCTCTACTGTG) T b p ( F - T T C G T G C A A G A A AT G C T G A A , R TCCTGTGCACACCATTTTTC)
Quantitative PCR Immunohistochemistry mRNA expression of CREB1, CBP and related bZIP transcription factors including ATF1, ATF2 and CREM was examined using quantitative reverse transcriptase PCR (qRTPCR) as previously described (McPherson et al. 2010). Cortex, hippocampus, striatum and cerebellum from brains of freshly killed CREB1DARPP-32Cre/loxlox mice and controls were microdissected on ice. Total RNA was extracted using RNeasy Mini Kit (QIAGEN, Doncaster, Australia), and reverse transcription was subsequently performed using the SuperScript III Platinum Two-Step qRT-PCR Kit (Invitrogen, Mulgrave, Australia) according to the manufacturer's instructions. Quantitative PCR reactions were performed using SYBR Green I intercalating dye (Invitrogen) on an Applied Biosystems ABI 7500 fast thermocycler. Data were analysed using the ΔΔCt method (Livak and Schmittgen 2001). Primers for the genes of interest were designed as previously described (McPherson et al. 2010), and one endogenous control gene for each study was selected from four candidates (β-actin, hprt1, mthfd1 and tpb) using geNORM (Vandesompele et al. 2002). The primers were purchased from Geneworks, Hindmarsh, Australia: A t f 1 ( F - G G C T G G C A A G T G A G G A G TA A , R GAACCAGGCTGAGATGCAGT) A f t 2 ( F - C A A G A A G G C T T C C G A A G AT G , R AGGTAAAGGGCTGTCCTGGT) CBP (F-GTCTTTGCCTTTTCGTCAGC, CCACATACTGCCAGGGTTCT)
R-
Mice were anesthetised (sodium pentobarbitone, 80 mg/kg, i.p.) and transcardially perfused with 30 ml phosphate buffered saline (PBS; 0.1 M, pH 7.4) followed by fixation with 30 ml 4% paraformaldehyde (PFA; Sigma-Aldrich, Castle Hill, Australia) in PBS. Mice were immediately decapitated and brains dissected and postfixed in 10% sucrose in PFA overnight. Immunohistochemistry was performed on free-floating brain slices (40 μm). Sections were quenched for 15 mins with 10% methanol and 10% hydrogen peroxide (30% stock) solution in PBS. Following washing (PBS, 3× 5 mins), sections were incubated with rabbit monoclonal CREB1 antibody (Cell Signalling Technology, Boston, MA; #9197, 1:1,000) or rabbit polyclonal CBP antibody (Santa Cruz Biotechnology, Santa Cruz, CA; #sc-583, 1:1,000) in PBS containing 1:200 normal goat serum (NGS) and 0.5% Triton X-100 (TX-100) overnight at room temperature. Sections were subsequently washed and pre-blocked in PBS containing 1:200 NGS and 0.5% TX-100 before being incubated with biotinylated goat anti-rabbit secondary antibody (Vector Laboratories, East Brisbane, Australia, 1:500) in PBS containing 1:200 NGS and 0.5% TX-100 for 1 h. Sections were washed and then incubated with Vectastain Elite ABC Kit (Vector Laboratories) in PBS for 1 h, then washed and incubated with 3,3′-diaminobenzidine tetrahydrochloride chromagen (Sigma-Aldrich) solution containing 25% 0.4 M PBS and 0.004% w/v ammonium chloride/ammonium nickel (II) sulphate hexahydrate for 15 min. Immunoreactivity was developed by addition of
702
hydrogen peroxide, and the reaction was terminated by washing in PBS. Sections were then slide mounted with 0.5% gelatin, and once dry, the sections were dehydrated, cleared and coverslipped with Depex Mounting Medium (BDH Laboratory Supplies, Poole, Dorset, UK). Tissue was visualised using a Leica DM LB-2 microscope, and unilateral CREB1 positive cell counts were performed on individual coronal sections of the NAc and caudate putamen (bregma 1.34 mm) and sagittal sections of the hippocampus and cerebellum (lateral 1.68 mm, n=4 mice per genotype). Care was taken to ensure sections were matched at the same anatomic level for each mouse (Paxinos and Franklin 2001). In situ hybridisation Mice were killed by decapitation and their brains immediately removed and frozen on liquid nitrogen. Using a cryostat, 14μm sagittal sections were cut and thaw mounted onto slides previously baked at 180°C overnight and coated in poly-Llysine. Before hybridisation, slides were fixed and delipidated as previously described (Cowen et al. 2005; Cowen and Lawrence 2001). Briefly, slides were dipped in ice-cold 4% PFA in 0.1 M PBS for 5 min before being rinsed in PBS for 3 min at room temperature, and then serially dehydrated in ethanol. Sections were then delipidated in chloroform (20 min) and slides stored at 4°C in 100% ethanol. An established protocol was used to analyse mRNA expression in CREBDARPP-32 Cre/loxlox mice and littermate controls (Lawrence et al. 1996). Antisense nucleotides complementary to the mRNA encoding preprotachykinin ( 5 ′ - T C G G G C G AT T C T C T G C A G A A G AT G C T CAAAGGGCTCCGGCATTGCCTC-3′), enkephalin (5′A T C T G C A T C C T T C T T C AT G A A G C C G C C A TACCTCTTGGCAAGGATCTC-3′), DARPP-32 (5′AGGTGAAAGACAGGGTACAAAGGAGGGTGG-3′), dopamine D2 receptor (5′-GGCGATCATGACAGTAACT CGGCGCTTGGAGCTGTAGCGTGTGTT-3′) and dopamine D1 receptor (a cocktail of three different probes: 5′TCCGCTGGTCCCTAGATTCCCCAAGGAATGCA TAGGCTTTTAAGC-3′, 5′-AGCCAGACTTCCCCCAAT CACTCACCACCCAGCCCCTTCTCCAG-3′ and 5′C C T T C G G A G T C AT C T T C C T C T C ATA C T G G A AAGGGCTGGAGATAGCCC-3′) were diluted to a working stock of 0.3 pmol/μl, and the 3′- end labelled with [α33P]deoxyadenosine 5′-triphosphate (3,000 Ci/mmol; PerkinElmer, Boston, MA, USA) in the presence of terminal deoxynucleotidyl transferase (Roche Diagnostics, Mannheim, Germany), yielding an activity of 100,000– 500,000 cpm/μl. Labelled probes were spun on Sephadex G-25 medium spin columns at 2,000 rpm to separate any unincorporated nucleotides. The labelled oligonucleotide probes (1 pg/μl, 100 μl per slide) were applied to mouse brain sections in a hybrid-
Psychopharmacology (2012) 219:699–713
isation buffer containing 50% formamide, 4× saline sodium citrate (SSC, pH 7.0) and 10% dextran sulphate. Nonspecific signals for each probe were determined by using 100-fold molar excess of unlabelled oligonucleotide. Hybridisation was allowed to proceed overnight in a humidified (50% deionised formamide and 4× SSC) atmosphere at 42°C. After hybridisation, slides were washed in 1× SSC at 55°C for 1 h, rinsed, serially dehydrated in ethanol and allowed to air-dry. Once dry, slides were apposed to Kodak BioMax XAR film (Eastman Kodak, Rochester, NY, USA) with a set of standard [14C] microscales. Films were manually developed and analysed using ImageJ (v.1.32J, National Institutes of Health, USA) with comparison to standard microscales. Stereology In order to assess for neurodegeneration within the striatum of CREB1DARPP-32 Cre/loxlox mice, stereology was performed on neutral red-stained coronal brain sections as previously described (McPherson et al. 2010). Tissue was visualised using a Leica DM LB-2 microscope, and volumetric analysis was undertaken using the Cavalieri tool from Stereo Investigator 7.0× (MicrobrightField Inc., Williston, VT, USA) to estimate area and volume. Stereotaxic coordinates were established according to a mouse atlas (Paxinos and Franklin 2001), and four separate levels within the striatum were assessed (bregma 1.42, 1.18, 0.86 and 0.62 mm). Ethology In order to assess for anxiety-like behaviour, the elevated plus maze and light/dark locomotor test were employed. To initiate the plus maze experiment, each subject was placed in the centre of the maze facing an open arm, and using EthoVision 3.0, parameters including time spent in the closed or open arms, frequency of entries into the closed or open arms and the latency to leave the central zone were recorded over a 5-min period (Short et al. 2006). For the light–dark test, each subject was introduced into the dark side of the apparatus and was allowed access to the light side via a small rectangular entry. Locomotor activity was measured using Tru Scan Photobeam Activity Monitors and associated software (Coulbourn Instruments, Whitehall, PA, USA). Time spent in the light side of the chamber as well as latency to the first entry into the light side were used as measures of anxiety. Behavioural sensitization to cocaine The effect of genotype and cocaine on locomotor activity was examined using Tru Scan Photobeam Activity Monitors (Coulbourn Instruments) in a low-luminosity environment as
Psychopharmacology (2012) 219:699–713
previously described (Brown et al. 2009; McPherson and Lawrence 2006). Locomotor activity was recorded for 30 min each session using the Truscan 2.0 software. For three consecutive days, mice were placed in the locomotor cells to determine their habituation to a novel environment. Following this, mice from each genotype were randomly divided into two groups which received either cocaine (20 mg/kg, i.p.; Macfarlan Smith Limited, Edinburgh, UK) or saline (0.9% w/v saline 10 ml/kg, i.p.) immediately prior to being placed in the locomotor cells. This protocol was repeated for five consecutive days. A withdrawal period of 7 days followed, during which time the mice were confined to their home cages with free access to food and water. On challenge day, cocaine and saline-pretreated mice were subdivided into two groups, one which received cocaine (10 mg/kg, i.p.) and the other saline (0.9% w/v saline 10 ml/kg, i.p.) prior to being placed in the locomotor cells. Amphetamine-induced stereotypy Mice were placed in Tru Scan Photobeam Activity Monitors (Coulbourn Instruments) and their locomotor activity recorded for 15 min to allow them to habituate to the equipment. Following this, the mice were removed from the chambers and injected with amphetamine (10 mg/kg, i.p.; Sigma-Aldrich) before being returned to the same chamber. Locomotor activity was recorded for a further 45 min post-injection, and the mice were also recorded using a video camera placed above the locomotor cell. The videos were later analysed by a blinded experimenter who manually recorded stereotypies including vertical head dips and lateral head weaves. Ethanol-induced loss of righting reflex A hypnotic dose of ethanol (3.5 g/kg in a volume of saline equal to 10 ml/kg, i.p.) was administered to CREB1DARPP-32Cre/loxlox mice and littermate controls to induce loss of righting reflex (LORR, Bird et al. 2008). Once the righting reflex was lost, mice were oriented supinely and the time taken for the mice to recover was recorded. Regain of righting reflex was considered to have occurred when the mouse could right itself twice within 1 min. The latency between ethanol injection and LORR was also recorded. Statistical analyses All statistical tests were performed using SigmaStat 3.5. Striatal volume and CREB1 cell counts were analysed using two-tailed t tests. qRT-PCR, in situ hybridisation and amphetamine-induced stereotypies were all assessed using one-way analysis of variance (ANOVA). Locomotor habituation data were analysed using two-way ANOVAs with
703
genotype and day as factors. All time-course locomotor data were analysed by two-way RM ANOVAs with either time bin and genotype or time bin and treatment as factors. All post hoc tests were Tukey, and in all analyses, statistical significance was accepted at p
