PSMC5, a 19S Proteasomal ATPase, Regulates Cocaine Action in the Nucleus Accumbens (2015)

PLoS One. 2015 May 11;10(5):e0126710. doi: 10.1371/journal.pone.0126710. eCollection 2015.

Ohnishi YH1, Ohnishi YN2, Nakamura T3, Ohno M4, Kennedy PJ5, Yasuyuki O6, Nishi A7, Neve R8, Tsuzuki T4, Nestler EJ5.

Abstract

ΔFosB is a stable transcription factor which accumulates in the nucleus accumbens (NAc), a key part of the brain’s reward circuitry, in response to chronic exposure to cocaine or other drugs of abuse. While ΔFosB is known to heterodimerize with a Jun family member to form an active transcription factor complex, there has not to date been an open-ended exploration of other possible binding partners for ΔFosB in the brain. Here, by use of yeast two-hybrid assays, we identify PSMC5-also known as SUG1, an ATPase-containing subunit of the 19S proteasomal complex-as a novel interacting protein with ΔFosB. We verify such interactions between endogenous ΔFosB and PSMC5 in the NAc and demonstrate that both proteins also form complexes with other chromatin regulatory proteins associated with gene activation. We go on to show that chronic cocaine increases nuclear, but not cytoplasmic, levels of PSMC5 in the NAc and that overexpression of PSMC5 in this brain region promotes the locomotor responses to cocaine. Together, these findings describe a novel mechanism that contributes to the actions of ΔFosB and, for the first time, implicates PSMC5 in cocaine-induced molecular and behavioral plasticity.

Citation: Ohnishi YH, Ohnishi YN, Nakamura T, Ohno M, Kennedy PJ, Yasuyuki O, et al. (2015) PSMC5, a 19S Proteasomal ATPase, Regulates Cocaine Action in the Nucleus Accumbens. PLoS ONE 10(5): e0126710. doi:10.1371/journal.pone.0126710

Academic Editor: James Edgar McCutcheon, University of Leicester, UNITED KINGDOM

Received: December 10, 2014; Accepted: April 7, 2015; Published: May 11, 2015

Copyright: © 2015 Ohnishi et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited

Data Availability: All relevant data are within the paper.

Funding: This work was supported by grants from the National Institutes of Health, the National Institute on Drug Abuse, and by the Ishibashi Foundation and the Japan Society for the Promotion of Science (KAKENHI numbers: 24591735, 26290064, 25116010). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

ΔFosB, a truncated product of the FosB gene, belongs to the Fos family of transcription factors, which also include c-Fos, full-length FosB, Fra-1, and Fra-2. ΔFosB, like other Fos proteins, heterodimerizes with a Jun family protein—c-Jun, JunB, or JunD—to form an active AP-1 (activator protein-1) transcription factor complex, which induces or represses the expression of specific target genes [1,2].

ΔFosB has been shown to play a key role in drug addiction [2]. Uniquely among Fos family proteins, it accumulates in nucleus accumbens (NAc) and other reward-related brain regions after repeated drug administration due to its high level of stability [3,4], which is mediated by the lack of C-terminal degron domains and by phosphorylation by several proteins kinases [57]. Such induction of ΔFosB in the NAc mediates increased behavioral responses to drugs of abuse. Thus, overexpression of ΔFosB in this brain region of adult animals, either by use of viral vectors or inducible bitransgenic mice, increases an animal’s sensitivity to the locomotor-activating and rewarding effects of cocaine and opiates as well as an animal’s motivation to self-administer cocaine [711]. Conversely, overexpression of dominant negative antagonists of ΔFosB causes the opposite behavioral phenotypes [1012].

We and others have previously confirmed, by use of gel shift assays, that JunD and perhaps other Jun family proteins are the major binding partners of ΔFosB in the brain in vivo [1315]. However, to date, there has not been an open-ended, unbiased evaluation of ΔFosB’s binding partners in brain. Here, we sought to identify novel binding partners for ΔFosB by use of a yeast two-hybrid assay [16,17]. Our data revealed that PSMC5, also known as SUG1, is a robust partner of ΔFosB both in vitro and in the NAc in vivo, where it joins ΔFosB as part of a cocaine-induced transcription activation complex, which also contains CBP (CREB binding protein) and p300—both histone acetyltransferases (HATs)—as well as BRG1 (a chromatin remodeling protein). We go on to show that chronic exposure to cocaine alters the nuclear levels of PSMC5, an ATPase-containing subunit of the 19S proteasomal complex, in the NAc and that PSMC5 in turn controls behavioral responses to cocaine.

Material and Methods

Yeast two-hybrid screening

MaV203 yeast cells (Invitrogen Life Technologies) were co-transfected with pDBLeu driving different fragments of the ΔFosB protein, and a mouse brain library was subcloned in pPC86 (Invitrogen Life Technologies). Transformed cells were grown on SC-medium lacking leucine, tryptophan, and histidine, and containing 10 mM of 3-aminotriazole. Binding between FosB fragments and a candidate partner induces three reporter genes (His3, Ura3, and LacZ), and the induction makes transformants able to survive under the cultured conditions used. Positive clones were retested with fresh pDBLeu-ΔFosB fragments by retransformation assays in MaV203 cells.

Cell lines

Mouse Neuro 2A neuroblastoma cells (ATCC) were maintained in Eagle’s minimum essential medium (EMEM) (ATCC), supplemented with 10% fetal bovine serum (FBS) at 37°C and 5% CO2. Rat 1A cells were a gift from Yusaku Nakabeppu (Fukuoka, Japan) [18] and maintained in Dulbecco’s MEM (DMEM) (Life Technologies), supplemented with 10% FBS at 37°C and 5% CO2. Transfection of cells with plasmid DNA was accomplished using Effectene (Qiagen) according to the manufacturer’s instructions.

PSMC5 and ΔFosB constructs

Several mutant forms of PSMC5, each FLAG-tagged at their N-terminus, were generated for use in immunoprecipitation or viral-mediated gene transfer experiments. These included: PSMC5-K196M, PSMC5-Δcoiled-coil domain (PSMC5-ΔCC, lacking amino acids 27–68), PSMC5-NT (consisting of the N-terminal fragment of the protein, amino acids 1–151), and PSMC5-CT (consisting of the C-terminal fragment of the protein, 172 amino acids) (see Fig 1). We also utilized N-terminal MYC-tagged forms of wildtype ΔFosB as well as ΔFosB with a mutation in its leucine-zipper domain (mutation of amino acids 182 to 205 which is known to obliterate heterodimerization with Jun family proteins [6].

thumbnail
Fig 1. ΔFosB binds to PSMC5 in vitro.

 

A. Schematic of ΔFosB, ΔFosB-LZM in which the leucine zipper domain is mutated to obliterate ΔFosB heterodimerization with Jun proteins, and Δ2ΔFosB which lacks the first 78 amino acids of the ΔFosB N-terminus. B. Schematic of PSMC5, PSMC5-NT which comprises the first 151 amino acids of PSMC5, PSMC5-CT which lacks the first 235 amino acids of PSMC5, and PSMC5-ΔCC which lacks the coiled-coil domain (amino acids 28–68). The AAA domain corresponds to a motif, ATPases Associated with diverse cellular Activities, present in many ATPases. C. 2.4 μg of pcDNA3.1-ΔFosB (lanes 1–4) or ΔFosB-LZM (lane 5) was co-transfected with 2.4 μg of FLAG-tagged PSMC5 or various deletion mutants into Neuro2a cells. Two days after transfection, cells were lysed and subjected to immunoprecipitation with an anti-FLAG antibody and then Western blotted with anti-ΔFosB or anti-FLAG antibody. Note that ΔFosB, but not ΔFosB-LZM, binds robustly to PSMC5 or PSMC5-NT, but not PSMC5-CT or PSMC5-ΔCC. The data shown in the figure were replicated in triplicate in each of three separate experiments.

doi:10.1371/journal.pone.0126710.g001

Animals

Nine- to 11-week-old C57BL/6J male mice (The Jackson Laboratory) were used for all experiments. Animals were housed on a 12-h light-dark cycle with access to food and water ad libitum and were habituated 1 week before experimentation. Two cocaine treatment regimens were used. To study the biochemical effects of cocaine, animals were given 7 daily doses of cocaine (20 mg/kg) or saline, and killed by decapitation 24 hr after the last injection. This is a standard protocol, which has been shown to produce numerous molecular and cellular responses to the drug [7]. To study the influence of PSMC5 in nucleus accumbens on behavioral responses to cocaine, we used a subthreshold dose of the drug (7.5 mg/kg; see Locomotor sensitization below) based on the hypothesis that PSMC5 would, like ΔFosB, increase an animal’s sensitivity to cocaine [8]. All animal experiments were approved by the Institutional Animal Care and Use Committee at Mount Sinai.

Immunoprecipitation and Western blotting

Neuro 2A cells were transfected with wildtype or mutant forms of PSMC5. Two days after transfection, cells were washed in PBS, lysed in RIPA buffer (50 mM Tris pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% NP-40, 0.25% sodium deoxycholate, 10 mM sodium butyrate, protease inhibitor cocktail). Lysates were split and incubated with either nonimmune IgG (Sigma) or anti-FLAG antibodies (Sigma) for 3 hr at 4°C. Immunoprecipitation was performed with Protein G beads (Invitrogen) as described [19]. Briefly, immunoprecipitated proteins were subjected to SDS-PAGE and analyzed by Western blotting using anti-FosB/ΔFosB antibody (Cell Signaling Technology) based on published protocols [7]. For in vivo protein binding assays, we used purified nuclear fractions from punch-dissected NAc of mice after chronic cocaine treatment (20 mg/kg IP daily for 7 days, with mice used 24 hr after the last injection). Co-immunoprecipitation from nuclear fractions was performed using the Nuclear Complex Co-IP kit (Active Motif) following the manufacturer’s instructions. The following antibodies were used: MYC or ß-actin, Cell Signaling Technology (Danvers, MA), PSMC5 and histone H3, Abcam (Cambridge, MA), CBP, p300 and BRG1, Santa Cruz Biotechnology (Santa Cruz, CA), and FLAG M2, Sigma.

Immunohistochemistry

Immunohistochemistry was performed according to published procedures [20]. Mice were anesthetized and perfused intracardially with 4% paraformaldehyde in PBS. Brain were cryoprotected with 30% sucrose, and then frozen and stored at -80°C until use. Coronal sections (40 μm) were cut on a cryostat and processed for immunohistochemistry. Free-floating sections were pre-incubated in a blocking buffer containing 0.3% Triton and 3% normal donkey serum. ΔFosB was detected using a goat polyclonal antibody raised against the N-terminal portion of the protein (1/1000 Santa Cruz Biotechnology). PSMC5 was detected using a rabbit polyclonal antibody (1/100 Abcam, Cambridge, MA). Images were taken with a confocal microscope (60x magnification; Zeiss).

Locomotor sensitization

All mice received daily IP injections of saline for 3 days to habituate them to the stress of the injections. The next day, mice were injected IP with saline or a subthreshold dose of cocaine (7.5 mg/kg; see under Animals above) and placed immediately into novel locomotor boxes. The locomotor activity of the mice was recorded using a photobeam system as ambulatory beam breaks for 30 min. These treatments were repeated daily for 3 days.

Viral-mediated gene transfer

We used extensively published methods for viral-mediated gene transfer [7,8,11,19]. Briefly, expression plasmids for PSMC5 or for several of its mutants (see PSMC5 and ΔFosB constructs above) were subcloned into the bicistronic p1005(+) HSV plasmid expressing GFP under the control of the CMV promoter, and PSMC5 or its mutants under that of the IE4/5 promoter. Under ketamine (100 mg/kg)/xylazine (10 mg/kg) anesthesia, mice were positioned in a small-animal stereotaxic instrument, and the cranial surface was exposed. Thirty-three gauge syringe needles were used to bilaterally infuse 0.5 μl of an HSV vector into the NAc at a 10° angle (AP +1.6; ML +1.5; DV -4.4) at a rate of 0.1 μl/min. Animals receiving HSV injections were allowed to recover for 2 days following surgery before experimentation.

Statistics

ANOVAs and student’s t-tests were used, corrected for multiple comparisons, with significance set at p<0.05.

Results

PSMC5: novel binding partner of ΔFosB

We performed preliminary experiments to identify an appropriate fragment of ΔFosB that served as bait in a yeast two-hybrid assay without autoactivating the system. Holo-ΔFosB induced reporter gene activity on its own, as did the N-terminal 1–78 amino acid fragment of the protein. However, an N-terminal truncated ΔFosB (Fig 1A), termed Δ2ΔFosB, which lacks the first 78 amino acids of the protein, did not have this effect. Therefore, we used Δ2ΔFosB as the bait protein.

To screen for potential binding partners, we used a mouse brain library subcloned in pPC86. We identified 11 candidates for binding partners. Although these proteins included ΔFosB’s known heterodimerization partners, c-Jun and JunD (Table 1), the most prevalent candidate by far was PSMC5. While surprising, this was an interesting finding, since PSMC5 was shown in a single report years ago to bind to c-Fos in vitro [21]. However, there are no prior reports of PSMC5 involvement in cocaine action. Nevertheless, because of the strength of the PSMC5 signal in the yeast two-hybrid assay, we set out to further study possible ΔFosB-PSMC5 interactions.

thumbnail
Table 1. Results of Yeast Two-Hybrid Screening with Δ2ΔFosB.

doi:10.1371/journal.pone.0126710.t001

First, to confirm the physical interaction between ΔFosB and PSMC5, we performed in vitro co-immunoprecipitation experiments. We found that FLAG-tagged PSMC5 (Fig 1B), transfected into Neuro 2A cells, effectively pulled down ΔFosB (Fig 1C). Second, to identify the region in PSMC5 that is responsible for its binding to ΔFosB, we generated several FLAG-tagged PSMC5 mutants (Fig 1B) and repeated the co-immunoprecipitation experiment. ΔFosB was pulled down effectively with the N-terminal 151 amino acids of PSMC5 (PSMC5-NT), but not with the C-terminal 172 amino acid fragment of the protein (PSMC5-CT) (Fig 1C). PSMC5 lacking its coiled coil domain (PSMC5-ΔCC) was also ineffective in precipitating ΔFosB. These findings suggest that PSMC5 binds ΔFosB via its coiled-coil domain (amino acids 27–68). Moreover, FLAG-tagged PSMC5 did not precipitate a mutant form of ΔFosB with a mutated leucine zipper domain (ΔFosB-LZM) (Fig 1C), indicating that ΔFosB either binds PSMC5 through this domain or, more likely, that ΔFosB heterodimerization is required for PSMC5 binding.

PSMC5-ΔFosB binding in NAc after chronic cocaine administration

Based on these findings in vitro, we studied whether PSMC5 levels in the NAc are altered in response to chronic cocaine administration. We found by subcellular fractionation and Western blotting that chronic cocaine increases nuclear levels of PSMC5 in this brain region without a change in cytoplasmic levels (Fig 2A). This effect was not seen after single doses of cocaine (data not shown). We next examined the localization of PSMC5 and ΔFosB in NAc by confocal immunofluorescence microscopy. We analyzed mice 24 hr after the last repeated dose of cocaine, a time point when ΔFosB is the only detectable FosB gene product (see Nestler 2008). We found strong PSMC5 immunoreactivity in the NAc, including a strong nuclear signal. ~85% of ΔFosB+ nuclei co-stained for PSMC5 (Fig 2B). Additionally, we performed co-immunoprecipitation experiments on NAc extracts and found that, after chronic cocaine treatment, ΔFosB was pulled down effectively by an anti-PSMC5 antibody (Fig 2C). In contrast, analysis of drug-naïve NAc (after repeated saline injections) revealed no detectable ΔFosB pull down (data not shown). These data are consistent with our findings in cell culture and confirm that ΔFosB and PSMC5 interact in the NAc in vivo.

thumbnail
Fig 2. PSMC5 regulation in mouse NAc.

 

A. Western blotting of nuclear and cytosolic fractions of NAc of mice treated daily with saline or cocaine (20 mg/kg) for 7 days, with animals analyzed 24 hr after the last injection. Cocaine increases nuclear but not cytosolic levels of PSMC5. Histone H3 and ß-actin, which were not affected by cocaine, were used as loading controls. Data are mean ± SEM (n = 8–10/group, *p<0.05). B. Co-localization of endogenous PSMC5 (green) and ΔFosB (blue) in NAc of mice treated chronically with cocaine as in A. C. Nuclear lysates of mouse NAc after chronic cocaine treatment were subjected to immunoprecipitation with anti-PSMC5 antibody or mouse IgG as control, and then Western blotted with anti-FosB/ΔFosB antibody. The figure demonstrates PSMC5-ΔFosB interactions in the NAc in vivo. Data in B and C were replicated in triplicate in each of three separate experiments.

doi:10.1371/journal.pone.0126710.g002

PSMC5 enhances ΔFosB expression in vitro

Since PSMC5 is a known member of the proteasome complex, we tested whether it regulates ΔFosB levels using Rat 1A cells. PSMC5 overexpression had no effect on basal levels of ΔFosB, but caused a small but significant enhancement of ΔFosB induction upon serum stimulation of the cells (Fig 3A). A similar trend was seen for full-length FosB but the effect did not achieve statistical significance. Conversely, suppression of endogenous PSMC5 expression in Rat 1A cells, achieved by use of siRNAs that target PSMC5, did not affect basal ΔFosB levels but strongly inhibited ΔFosB induction by serum stimulation (Fig 3B). Similar effects were seen for full-length FosB. These data suggest that PSMC5 does not promote the proteasomal degradation of ΔFosB, as might be expected as a core subunit of the proteasome, but instead is required for maximal accumulation of FosB gene products in vitro, perhaps through stabilizing the proteins.

thumbnail
Fig 3. PSMC5 regulation of FosB/ΔFosB expression in Rat 1A cells.

 

A. Rat 1A cells were transfected with 4 μg of PSMC5 or control DNA. PSMC5 overexpression had no effect on basal expression levels of FosB or ΔFosB protein as determined by Western blotting, but produced a small but significant increase in the induction of ΔFosB by serum stimulation (F(2,21) = 9.75, p = 0.001). B. Rat 1A cells were transfected with 5 pmol of either of two siRNAs or scrambled RNA (control). Both siRNAs effectively decreased PSMC5 protein levels compared to control conditions (siRNA #1, 23 ± 5% of control; siRNA #2, 18 ± 6%; p<0.05; n = 4). PSMC5 knockdown had no effect on basal levels of FosB or ΔFosB but attenulated the induction of both FosB and ΔFosB by serum stimulation (FosB: F(2,6) = 20.99, p = 0.002; ΔFosB: F(2,6) = 22.83, p = 0.002).

doi:10.1371/journal.pone.0126710.g003

ΔFosB and PSMC5 form complexes with CBP, p300, and BRG1 in NAc

To better understand the transcriptional mechanisms by which PSMC5 might influence ΔFosB function, we investigated possible additional binding partners for the two proteins in the NAc under chronic cocaine-treated conditions. There is one report that PSMC5 binds to CBP—a HAT—and increases histone H3 acetylation at the MHC-II proximal promoter in HeLa cells [22]. Moreover, mice that are deficient in CBP display reduced behavioral sensitivity to cocaine as well as reduced histone acetylation at the FosB promoter [23]. We thus tested whether PSMC5 might bind with ΔFosB as part of complexes that also contain CBP and perhaps other transcriptional activators.

We first demonstrated that ΔFosB effectively pulled down both CBP and p300, a related HAT, in Neuro2A cells (Fig 4A). In contrast, the leucine zipper mutant form of ΔFosB, as expected, did not exhibit this activity. Similarly, PSMC5 effectively pulled down CBP and p300 (Fig 4B). Interestingly, this effect was also seen for PSMC5-ΔCC, which did not pull down ΔFosB, indicating that PSMC5 interacts with CBP and p300 via other domains of the protein and independently of its binding to ΔFosB.

thumbnail
Fig 4. ΔFosB and PSMC5 interact with CBP, p300, and BRG1 in vitro and in vivo.

 

A. Neuro2A cells were transfected with 2.4 μg of MYC-tagged ΔFosB or MYC-tagged ΔFosB-LZM. Cell extracts were immunoprecipitated with anti-CBP or anti-p300 antibody, and precipitates were Western blotted with the same antibody or with anti-MYC antibody. Both CBP and p300 interact with ΔFosB and such interactions require an intact leucine zipper. B. Neuro2A cells were transfected with 2.4 μg of FLAG-tagged PSMC5 or FLAG-tagged PSMC5-ΔCC. Cell extracts were immunoprecipitated with anti-CBP or anti-p300 antibody, and precipitates were Western blotted with the same antibody or with anti-FLAG antibody. Both CBP and p300 interact with PSMC5 and such interactions do not require the CC domain. C. Nuclear lysates of mouse NAc after chronic cocaine treatment were subjected to immunoprecipitation with anti-CBP or anti-p300 antibody. Subsequent Western blotting of resulting precipitates with anti-FosB/ΔFosB antibody showed endogenous interactions between ΔFosB and CBP/p300. D. Aliquots of the same nuclear lysates were subjected to immunoprecipitation with anti-BRG1 or anti-PSMC5 antibody, followed by Western blotting of precipitates with anti-FosB/ΔFosB or anti-BRG1 antibody. The results show endogenous interactions between ΔFosB and BRG1, and BRG1 and PSMC5. E. Schematic illustration of transcriptional activation complex composed of ΔFosB:JunD heterodimers interacting with CBP/p300, BRG1, and PSMC5.

doi:10.1371/journal.pone.0126710.g004

To confirm that these interactions also occur in vivo, we administrated cocaine chronically to induce ΔFosB and nuclear PSMC5 levels, then immunoprecipitated NAc extracts with anti-CBP or anti-p300 antibodies. Consistent with our cell culture data, immunoprecipitation of CBP or of p300 effectively pulled down ΔFosB (Fig 4C). We tested whether BRG1, a core subunit of the SWI-SNF chromatin remodeling complex, might also bind to ΔFosB and PSMC5, based on our earlier finding that BRG1 is recruited to certain ΔFosB target genes in concert with their activation in NAc after chronic cocaine [24]. We found that immunoprecipitation of BRG1 pulled down ΔFosB in NAc extracts, and that immunoprecipitation of PSMC5 likewise coprecipitated endogenous BRG1 (Fig 4D). Taken together, these results suggest that ΔFosB-PSMC5 form complexes in NAc which also include CBP/p300 and BRG1 (Fig 4E).

PSMC5 overexpression increases locomotor responses to cocaine

The prominent binding of PSMC5 with ΔFosB in NAc prompted us to investigate whether increasing PSMC5 levels in this brain region regulates behavioral responses to cocaine. We generated a Herpes Simplex Virus (HSV) vector that overexpresses either wildtype PSMC5 or one of its mutants and validated the vectors in NAc in vivo (Fig 5A). Viral-mediated expression of PSMC5 predominates in the cell nucleus (Fig 5B). Mice overexpressing wild-type PSMC5 did not show altered responses to initial doses of cocaine, but displayed increased locomotor activation in response to repeated cocaine doses compared to GFP-expressing control mice. In contrast, mice overexpressing a mutant form of PSMC5 that lacks its coiled-coil domain (PSMC5-ΔCC) did not exhibit this effect (Fig 5B). Interestingly, overexpression of PSMC5-K196M, which lacks the ATPase activity of the wildtype protein, also potentiated cocaine’s locomotor responses.

thumbnail
Fig 5. PSMC5 overexpression in NAc increases locomotor responses to cocaine.

 

A. Representative HSV-mediated transgene expression in medial NAc. AC, anterior commissure. NAc core and shell subregions are noted on the figure. B. Representative higher magnifications (60x) of immunohistochemical staining of PSMC5 in NAc neurons after HSV-PSMC5 injection showing that the protein is predominantly nuclear as marked by DAPI staining. C. Mice received bilateral HSV injections into NAc followed by daily IP injections of subthreshold doses of cocaine (7.5 mg/kg). Locomotor responses are shown in response to the first and last of 3 daily doses of the drug. Overexpression of PSMC5 or PSMC5-K196M increases locomotor responses to repeated cocaine, an effect not seen with PSMC5-ΔCC. There was no significant effect of the transgenes on locomotor responses to initial cocaine doses. ANOVA F(3,125) = 4.163, *p<0.05 by Dunnett’s posthoc test.

doi:10.1371/journal.pone.0126710.g005

Discussion

The results of the present study reveal a new mechanism by which ΔFosB mediates its effects on the brain and a new mechanism involved in cocaine action. By use of an unbiased approach, a yeast two-hybrid assay, we identified PSMC5 as a novel binding partner for ΔFosB. We validated this finding both in cultured cells in vitro and in the NAc in vivo by demonstrating robust PSMC5-ΔFosB binding. Importantly, nuclear levels of PSMC5 are induced in NAc by chronic cocaine administration. We showed further that PSMC5-ΔFosB binding occurs in concert with several other transcriptional activator proteins, namely CBP and p300 (two HATs) and BRG1 (a key constituent of SWI-SNF chromatin remodeling complexes). Together, our findings support the hypothesis that PSMC5 is part of the transcriptional activation complex that is recruited to at least certain ΔFosB-induced genes during a course of chronic cocaine administration (Fig 4E). Consistent with this hypothesis is the additional finding that overexpression of PSMC5 in NAc, like the overexpression of ΔFosB itself, promotes behavioral responses to cocaine. It would be interesting in future studies to follow up these in vivo observations with characterization of PSMC5-ΔFosB-HAT-BRG1 interactions by use of in vitro reporter assays.

The involvement of PSMC5 in cocaine action is completely novel. Identified initially as a member of a large family of ATPases that comprise the proteasome, PSMC5 has been shown over the years to interact with several transcription factors, including c-Fos, p53, nuclear hormone receptors, and constituents of the basal transcription complex [25], however, few functional studies have been performed over the years [26]. Its best established action is to promote the activity of MYC transcription factors in cultured cells [27]. The implication of PSMC5 in transcriptional mechanisms has suggested a potential role for ubiquitination-proteasomal activity in the regulation of gene transcription, but the involvement of PSMC5 in such regulation remains virtually untested to date [28,29].

Very little is known about PSMC5 function in brain. An earlier study demonstrated widespread expression of PSMC5 mRNA throughout brain [30], but its functional activity has remained unstudied. Our findings now prompt further investigations of this interesting protein to better understand its role in regulating gene transcription and its relationship to ubiquitination-proteasomal function in brain. The binding of PSMC5 to ΔFosB is mediated by PSMC5’s coiled-coil domain. Moreover, the ability of PSMC5 to promote locomotor responses to cocaine, while requiring the coiled-coil domain, does not require the ATPase activity that is intrinsic to the protein. These results raise the possibility that, at least in our system, the main activity of PSMC5 might be mediated through its binding to ΔFosB and other transcriptional regulatory proteins and not through its proteasomal-related activity per se. Further work is needed to directly test this and alternative possibilities. The hypothesis that viral-mediated overexpression of PSMC5 increased locomotor responses to cocaine via interactions with ΔFosB is plausible, despite the use of a 3 day cocaine treatment regimen, because it is known that appreciable levels of ΔFosB accumulate in brain within this time frame [3].

The present findings further substantiate the utility of using unbiased, open-ended experimental approaches in studying the molecular basis of brain regulation. Our initial attention to PSMC5 was based solely on its prominent binding to ΔFosB in a yeast two-hybrid assay, yet it appears to be an important component of transcriptional changes that are recruited in NAc by repeated cocaine administration. A better understanding of the detailed mechanisms by which nuclear levels of PSMC5 are induced by cocaine and, in turn, by which PSMC5 then contributes to cocaine-induced transcriptional activation complexes are the focus of current investigations. Meanwhile, our yeast two-hybrid assay revealed several additional putative binding partners of ΔFosB (Table 1) which also now warrant direct examination in cocaine models. Together, this work is contributing to an increasing understanding of the complex molecular mechanisms through which cocaine alters NAc function.

Acknowledgments

This work was supported by grants from the National Institute on Drug Abuse, and by the Ishibashi Foundation and the Japan Society for the Promotion of Science (KAKENHI numbers: 24591735, 26290064, 25116010).

Author Contributions

Conceived and designed the experiments: YHO YNO EJN. Performed the experiments: YHO YNO PJK RN. Analyzed the data: YHO YNO EJN. Contributed reagents/materials/analysis tools: TN MO OY AN RN TT. Wrote the paper: YHO EJN.

References

  1. 1. Morgan JI, Curran T (1995) Immediate-early genes: ten years on. Trends Neurosci 18: 66–67. pmid:7537412 doi: 10.1016/0166-2236(95)80022-t
  2. 2. Nestler EJ (2008) Transcriptional mechanisms of addiction: role of deltaFosB. Philos Trans R Soc London B Biol Sci 363: 3245–3255. doi: 10.1098/rstb.2008.0067. pmid:18640924
  3. View Article
  4. PubMed/NCBI
  5. Google Scholar
  6. View Article
  7. PubMed/NCBI
  8. Google Scholar
  9. View Article
  10. PubMed/NCBI
  11. Google Scholar
  12. View Article
  13. PubMed/NCBI
  14. Google Scholar
  15. View Article
  16. PubMed/NCBI
  17. Google Scholar
  18. View Article
  19. PubMed/NCBI
  20. Google Scholar
  21. View Article
  22. PubMed/NCBI
  23. Google Scholar
  24. View Article
  25. PubMed/NCBI
  26. Google Scholar
  27. View Article
  28. PubMed/NCBI
  29. Google Scholar
  30. View Article
  31. PubMed/NCBI
  32. Google Scholar
  33. View Article
  34. PubMed/NCBI
  35. Google Scholar
  36. View Article
  37. PubMed/NCBI
  38. Google Scholar
  39. View Article
  40. PubMed/NCBI
  41. Google Scholar
  42. View Article
  43. PubMed/NCBI
  44. Google Scholar
  45. View Article
  46. PubMed/NCBI
  47. Google Scholar
  48. View Article
  49. PubMed/NCBI
  50. Google Scholar
  51. View Article
  52. PubMed/NCBI
  53. Google Scholar
  54. View Article
  55. PubMed/NCBI
  56. Google Scholar
  57. View Article
  58. PubMed/NCBI
  59. Google Scholar
  60. View Article
  61. PubMed/NCBI
  62. Google Scholar
  63. View Article
  64. PubMed/NCBI
  65. Google Scholar
  66. View Article
  67. PubMed/NCBI
  68. Google Scholar
  69. View Article
  70. PubMed/NCBI
  71. Google Scholar
  72. View Article
  73. PubMed/NCBI
  74. Google Scholar
  75. View Article
  76. PubMed/NCBI
  77. Google Scholar
  78. View Article
  79. PubMed/NCBI
  80. Google Scholar
  81. View Article
  82. PubMed/NCBI
  83. Google Scholar
  84. View Article
  85. PubMed/NCBI
  86. Google Scholar
  87. View Article
  88. PubMed/NCBI
  89. Google Scholar
  90. 3. Hope BT, Nye HE, Kelz MB, Self DW, Iadarola MJ, Nakabeppu Y, et al. (1994) Induction of a long-lasting AP-1 complex composed of altered Fos-like proteins in brain by chronic cocaine and other chronic treatments. Neuron 13: 1235–1244. pmid:7946359 doi: 10.1016/0896-6273(94)90061-2
  91. 4. Hiroi N, Brown J, Haile C, Ye H, Greenberg ME, Nestler EJ (1997) FosB mutant mice: Loss of chronic cocaine induction of Fos-related proteins and heightened sensitivity to cocaine’s psychomotor and rewarding effects. Proc Natl Acad Sci USA 94: 10397–10402. pmid:9294222 doi: 10.1073/pnas.94.19.10397
  92. 5. Ulery PG, Rudenko G, Nestler EJ (2006) Regulation of ΔFosB stability by phosphorylation. J Neurosci 26: 5131–5142. pmid:16687504 doi: 10.1523/jneurosci.4970-05.2006
  93. 6. Carle TL, Ohnishi YN, Ohnishi YH, Alibhai IN, Wilkinson MB, Kumar A, et al. (2007) Absence of conserved C-terminal degron domain contributes to ΔFosB’s unique stability. Eur J Neurosci 25: 3009–3019. pmid:17561814
  94. 7. Robison AJ, Vialou V, Mazei-Robison M, Feng J, Kourrich S, Collins M, et al. (2013) Behavioral and structural responses to chronic cocaine require a feed-forward loop involving ΔFosB and CaMKII in the nucleus accumbens shell. J Neurosci 33: 4295–4307 doi: 10.1523/JNEUROSCI.5192-12.2013. pmid:23467346
  95. 8. Kelz MB, Chen J, Carlezon WA Jr, Whisler K, Gilden L, Beckmann AM, et al. (1999) Expression of the transcription factor ΔFosB in the brain controls sensitivity to cocaine. Nature 401: 272–276. pmid:10499584
  96. 9. Colby CR, Whisler K, Steffen C, Nestler EJ, Self DW (2003) ΔFosB enhances incentive for cocaine. J Neurosci 23: 2488–2493. pmid:12657709
  97. 10. McClung CA, Nestler EJ (2003) Regulation of gene expression and cocaine reward by CREB and DFosB. Nat Neurosci 11: 1208–1215. pmid:14566342 doi: 10.1038/nn1143
  98. 11. Zachariou V, Bolanos CA, Selley DE, Theobald D, Cassidy MP, Kelz MB, et al. (2006) ΔFosB: An essential role for ΔFosB in the nucleus accumbens in morphine action. Nature Neurosci 9: 205–211. pmid:16415864 doi: 10.1038/nn1636
  99. 12. Peakman MC, Colby C, Perrotti LI, Tekumalla P, Carle T, Ulery P, et al. (2003) Inducible, brain region specific expression of a dominant negative mutant of c-Jun in transgenic mice decreases sensitivity to cocaine. Brain Res 970: 73–86. pmid:12706249 doi: 10.1016/s0006-8993(03)02230-3
  100. 13. Chen J, Nye HE, Kelz MB, Hiroi N, Nakabeppu Y, et al. (1995) Regulation of delta FosB and FosB-like proteins by electroconvulsive seizure and cocaine treatments. Mol Pharmacol 48: 880–889. pmid:7476919
  101. 14. Hiroi N, Marek GJ, Brown J, Ye H, Saudou F, Vaidya VA, et al. (1998) Essential role of the fosB gene in molecular, cellular, and behavioral actions of electroconvulsive seizures. J Neurosci 18: 6952–6962. pmid:9712664
  102. 15. Perez-Otano I, Mandelzys A, Morgan JI (1998) MPTP Parkinsonism is accompanied by persistent expression of a D-FosB-like protein in dopaminergic pathways. Mol Brain Res 53: 41–52. pmid:9473580 doi: 10.1016/s0169-328x(97)00269-6
  103. 16. Ma J, Ptashne M (1988) Converting a eukaryotic transcriptional inhibitor into an activator. Cell 55: 443–446. pmid:3180218 doi: 10.1016/0092-8674(88)90030-x
  104. 17. Chien CT, Bartel PL, Sternglanz R, Fields S (1991) The two-hybrid system: a method to identify and clone genes for proteins that interact with a protein of interest. Proc Natl Acad Sci USA 88: 9578–9582. pmid:1946372 doi: 10.1073/pnas.88.21.9578
  105. 18. Nakabeppu Y 1, Oda S, Sekiguchi M (1993) Proliferative activation of quiescent Rat-1A cells by delta FosB. Mol Cell Biol 13:4157–4166. pmid:8321220
  106. 19. Scobie KN, Damez-Werno D, Sun H, Shao N, Gancarz A, Panganiban CH, et al. (2014) Essential role of poly(ADP-ribosyl)ation in cocaine action. Proc Natl Acad Sci USA 111: 2005–2010. doi: 10.1073/pnas.1319703111. pmid:24449909
  107. 20. Perrotti LI, Weaver RR, Robison B, Renthal W, Maze I, Yazdani S, et al. (2008) Distinct patterns of ΔFosB induction in brain by drugs of abuse. Synapse 62: 358–369. doi: 10.1002/syn.20500. pmid:18293355
  108. 21. Wang WL, Chevray PM, Nathans D (1996) Mammalian Sug1 and c-Fos in the nuclear 26S proteasome. Proc Natl Acad Sci USA 93: 8236–8240. pmid:8710853 doi: 10.1073/pnas.93.16.8236
  109. 22. Koues OI 1, Dudley RK, Truax AD, Gerhardt D, Bhat KP, McNeal S, et al. (2008) Regulation of acetylation at the major histocompatibility complex class II proximal promoter by the 19S proteasomal ATPase Sug1. Mol Cell Biol 28: 5837–5850. doi: 10.1128/MCB.00535-08. pmid:18662994
  110. 23. Levine AA, Guan Z, Barco A, Xu S, Kandel ER, Schwartz JH (2005) CREB-binding protein controls response to cocaine by acetylating histones at the fosB promoter in the mouse striatum. Proc Natl Acad Sci USA 102: 19186–19191. pmid:16380431 doi: 10.1073/pnas.0509735102
  111. 24. Kumar A, Choi KH, Renthal W, Tsankova NM, Theobald DEH, Truong HT, et al. (2005) Chromatin remodeling is a key mechanism underlying cocaine-induced plasticity in striatum. Neuron 48: 303–314. pmid:16242410 doi: 10.1016/j.neuron.2005.09.023
  112. 25. St-Arnaud R (1999) Dual functions for transcriptional regulators: myth or reality. J Cell Biochem Suppl 32/33: 32–40. doi: 10.1002/(sici)1097-4644(1999)75:32+<32::aid-jcb5>3.0.co;2-x
  113. 26. Ferrell K, Wilkinson CRM, Dubiel W, Gordon C (2000) Regulatory subunit interactions of the 26S proteasome, a complex problem. Trends Biochem Sci 25:83–88. pmid:10664589 doi: 10.1016/s0968-0004(99)01529-7
  114. 27. von der Lehr N, Johansson S, Larson L-G (2003) Implication of the ubiquitin/proteasome system in myc-regulated transcription. Cell Cycle 2–5: 403–407. doi: 10.4161/cc.2.5.484
  115. 28. Geng FQ, Wenzel S, Tansey WP (2012) Ubiquitin and proteasomes in transcription. Annu Rev Biochem 81: 177–201. doi: 10.1146/annurev-biochem-052110-120012. pmid:22404630
  116. 29. Collins GA, Tansey WP (2006) The proteasome: a utility tool for transcription? Curr Op Genet Dev 16: 197–202. pmid:16503126
  117. 30. Sun DH, Swaffield JC, Johnston SA, Milligan CE, Zoeller RT, Schwartz LM (1997) Identification of a phylogenetically conserved Sug1 CAD family member that is differentially expressed in the mouse nervous system. Dev Neurobiol 33: 877–890. doi: 10.1002/(sici)1097-4695(199712)33:7<877::aid-neu2>3.0.co;2-5