SYVN1, an ERAD E3 Ubiquitin Ligase, Is Involved in GABAAα1 Degradation Associated with Methamphetamine-Induced Conditioned Place Preference

Abuse of methamphetamine (METH), a powerful addictive amphetamine-type stimulants (ATS), is becoming a global public health problem. The gamma-aminobutyric acid (GABA)ergic system plays a critical role in METH use disorders. By using rat METH conditioned place preference (CPP) model, we previously demonstrated that METH-associated rewarding memory formation was associated with the reduction of GABAAα1 expression in the dorsal straitum (Dstr), however, the underlying mechanism was unclear. In the present study, we found that METH-induced CPP formation was accompanied by a significant increase in the expression of Synovial apoptosis inhibitor 1 (SYVN1), an endoplasmic reticulum (ER)-associated degradation (ERAD) E3 ubiquitin ligase, in the Dstr. The siRNA knockdown of SYVN1 significantly increased GABAAα1 protein levels in both primary cultured neurons and rodent Dstr. Inhibition of proteasomal activity by MG132 and Lactacystin significantly increased GABAAα1 protein levels. We further found that SYVN1 knockdown increased GABAAα1 in the intra-ER, but not in the extra-ER. Accordingly, endoplasmic reticulum stress (ERS)-associated Glucose-regulated protein 78 (GRP78) and C/EBP homologous protein (CHOP) increased. Thus, this study revealed that SYVN1, as the ERAD E3 ubiquitin ligase, was associated with Dstr GABAAα1 degradation induced by METH conditioned pairing.


INTRODUCTION
Amphetamine-Type Stimulants (ATS), the second-most prevalent illicit drugs after cannabis/marijuana in the world (United Nations Office on Drugs and Crime (UNODC), 2011), are highly addictive and can lead to psychiatric illnesses or cognitive defect (Batki and Harris, 2004;Curran et al., 2004). ATS include amphetamine, methamphetamine (METH), 3,4-methylenedioxyamphetamine (MDA), and 3,4-methylenedioxymethamphetamine (MDMA), among which METH is the most potent amphetamine derivative and widely used substance. The mechanism underlying METH dependence is poorly understood.
Emerging increasing evidence shows that GABAergic dysfunction plays a critical role in the development of ATS use disorders (Addolorato et al., 2012;Kumar et al., 2013;Jiao et al., 2015). The GABA A receptors are the major inhibitory receptors in the central nervous system. Use of ATS will damage the normal functions of GABA A receptors (Hondebrink et al., 2013), and GABA A receptor modulators are reported to be effective in treating ATS use disorders (Mintzer and Griffiths, 2003;Rush et al., 2004;Johnson et al., 2007;Spence et al., 2016;Berro et al., 2017). The GABA A receptors are ligandgated and made up of five protein subunits that belong to different subunit classes, including α1-6, β1-3, γ1-3, δ, ε, θ1-3, π, ρ1-3. Our previous study demonstrated that the level of GABA A α1 subunits was significantly decreased in the dorsal striatum (Dstr). We also found that the GABA A α1 expression level in the Dstr decreased only in METH pairing group, but remained unchanged in the unpaired group, indicating that the changes of GABA A α1 receptors were related to METHpairing, rather than the pure drug effects. Intra-Dstr injection of the specific α1-containing GABA A receptor agonist before pairing abolished METH-induced conditioned place preference (CPP) formation. Therefore, our findings demonstrated that the decrease of GABA A α1 proteins in the Dstr was correlated to METH rewarding memory formation (Jiao et al., 2016). So far, little was known about the underlying mechanism of this decreased GABA A α1 expression associated with METH pairing.
Endoplasmic reticulum (ER)-associated degradation (ERAD) is the process by which the ER directs the degradation of misfolded or inappropriate proteins. It has been reported that METH or cocaine abuse is able to increase the misfolded or inappropriate proteins in ER and induce endoplasmic reticulum stress (ERS; Jayanthi et al., 2004Jayanthi et al., , 2009Beauvais et al., 2011;Pavlovsky et al., 2013). Moreover, it has been demonstrated that ERAD is involved in GABA A α1 degradation (Kang and Macdonald, 2004;Gallagher et al., 2005). Thus, we hypothesized that Dstr GABA A α1 protein was degraded via ERAD in METH dependent rat. In this study, by utilizing METH-induced CPP model, we tested this hypothesis by investigating the role of ERAD in the regulation of GABA A α1 expression, and found that Synovial apoptosis inhibitor 1 (SYVN1), the ERAD E3 ubiquitin ligase, modulated the downregulation of GABA A α1 protein associated with METH CPP.

Animals
Sprague Dawley male rats (weighting 220-300 g) were purchased from the Experimental Animal Center, Chinese Academy of Sciences (Shanghai, China). Rats were housed 2-3 per cage on a 12/12-h light/dark cycle with free access to food and water. All the animals used in this study were maintained in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals (NIH Publications No. 80-23, revised 1996) and experiments were approved by the Institutional Animal Care and Use Committee of Chinese Academy of Sciences (Shanghai, China).

Conditioned Place Preference
The CPP apparatus (Jiliang Software and Instruments) consists of two square compartments of the same size [40 cm (length) × 40 cm (width) × 60 cm (height)], separated by removable doors (10 × 10 cm), allowing rats free access to each compartment. One compartment has black and white horizontal stripes walls with an iron wire floor and the other has black and white vertical stripes walls with a steel bar floor. CPP procedure has been described in our previous studies (Jiao et al., 2016). The place conditioning procedure consists of the following phases: Habituation, Preconditioning, Conditioning and Testing. During Habituation phase, rats were allowed access to the entire apparatus for 30 min. In the Preconditioning phase, rats were allowed access to both compartments of the apparatus for 15 min. The duration spent in each compartment was recorded. The animals showing strong unconditioned aversion (one compartment >720 s) for one of the compartments were excluded. Following the Preconditioning, rats went through 8 days of conditioning. On the first day, the rats were intraperitoneally injected with either METH (1 mg/kg,) or saline (1 mL/kg.) and then confined to the non-preferred compartment for 45 min. On alternate days, rats were received saline and placed immediately in the preferred compartment. The procedure was repeated four times in the conditioning phase. During Testing phase, 24 h after the conditioning trial, the doors were opened and the rats were allowed to explore the entire apparatus for 15 min. The time spent in each compartment was recorded. The CPP score was the time spent in the drug-paired compartment during the testing phase minus the time spent in that compartment during the preconditioning phase. To minimize the basal stress, all rats were gentled by handling twice a day till the start of behavioral experiment. METH treated rats and saline-treated rats were individually housed in different cages with same standard conditions.

Subcellular Fractionation
Rats were sacrificed and the brains were removed immediately after CPP conditioning. The brain was dissected into coronal slices (1 mm thick) using a rat brain slicer (Braintree Scientific), and the dorsal striatum was punched by a bluntend, 17-gauge syringe needle. The homogenate was centrifuged at 1000× g for 10 min at 4 • C, and the supernatant was collected for analysis.

Immunoblotting
Equal amounts of protein were loaded on 10% sodium dodecyl sulfate polyacrylamide gels for electrophoresis. Separated proteins were then transferred on nitrocellulose FIGURE 2 | Increase in Synovial apoptosis inhibitor 1 (SYVN1) protein expression in the Dstr of METH-CPP rat. (A) SYVN1 protein expression level increased in the Dstr. (B) Glycoprotein 78 (Gp78) protein expression level remained unchanged. Error bars represent mean ± SEM, n = 8 in each group. * p < 0.05 compared with control group, two tailed Student's t-test.
Lysates from Dstr were subjected to co-immunoprecipitation (Co-IP) using an anti-GABA A α1 antibody followed by western blotting with SYVN1 or GABA A α1 antibody. In reverse IP, lysates were immunoprecipiated with an anti-SYVN1 antibody followed by western blotting with GABA A α1 or SYVN1 antibody. membrane. GABA A α1, GABA A α2, GABA A α3, GABA A α5, GABA A β2, Gp78, SYVN1, C/EBP homologous protein (CHOP) or GRP78 was detected with primary antibody at a 1:1000 dilution in TBS containing 5% non-fat dried milk and 0.05% Tween-20. After incubation with secondary antibody (HRP-conjugated goat anti-rabbit IgG or goat anti-mouse IgG) at a 1:2000 dilution, immunoreactive bands were detected with chemiluminescent substrate (RPN2232, GE Healthcare). The immunoreactive signals were quantified by quantity analysis software (Bio-Rad).

Co-Immunoprecipitation (Co-IP)
Brain homogenate was incubated overnight with antibodies (anti-GABA A α1 or anti-SYVN1; dilution 1:100) and with Protein A or Protein G agarose beads (Sigma) at 4 • C. Beads were washed three times by centrifugation and bound proteins were analyzed by Western blotting.

Quantitative Reverse Transcriptase PCR (qRT-PCR)
Total RNA was isolated from rat Dstr using a commercially available kit (RNeasy Plus Mini Kit, Qiagen). QRT-PCR was performed on a ABI7500 Real-Time PCR system (ABI) using SYBR Premix Ex Taq TM kit (TaKaRa Bio Group, Japan). A typical reaction of a total volume of 20 µl consisted of 2 µl Template DNA, 10 µl 2× SYBR Green Reaction Mix, 0.4 µl PCR Forward Primer (10 µM), 0.4 µl PCR Reverse Primer (10 µM), 0.4 µl ROX Reference Dye II and 6.8 µl DEPC treated water. PCR amplification was done with an initial incubation at 95 • C for 30 s, and then followed by 40 cycles of 95 • C for 5 s, 60 • C for 34 s, and final melting curve from 95 • C for 5 s, 60 • C 60 s. Primer specificity was confirmed by melting curve analysis. The mRNA for GABA A α1 and SYVN1 was normalized to a control gene (GAPDH). Primers utilized were as follows: GABA A α1 (GABA A α1, Fwd-CCT GGACCCTCATTCTGAGCA, Rev-ATCCTCGTGAAGACA GTGGTGTTG; GAPDH, Fwd-AACTTTGGCATTGTG GAAGG, REV-ACACATTGGGGGTAGGAACA; Genewiz, China).

Intracerebral Microinjection
Rats were anesthetized using sodium pentobarbital (50 mg/kg, i.p.) under aseptic conditions, and then a stereotaxic instrument (Narishige) with the incisor bar set at 3.3 mm was used for Cannula implantation and microinjection. For dorsal striatum infusion, guide cannula (26 gauge) was bilaterally implanted in the dorsal striatum (anteroposterior: +0.6 mm; mediolateral: ±2.7; dorsoventral: −2.0 mm). After recovering from surgery and anesthesia for 1 week, animals were placed in a stereotaxic apparatus again for drug infusion. Bilateral microinfusions were made through 31-gauge injection needles (2 mm below the tip of the guide cannulae), which was connected to a 10 µl microsyringe mounted in the microinfusion pump (Harvard Apparatus). During the injection, the animals were gently restrained by hand.
Mg132 and Lactacystin were dissolved in dimethylsulfoxide (DMSO) to stock concentration (50 µg/µl) and then diluted in PBS to a work concentration (0.25 µg/µl, DMSO concentration ≤20%). During the conditioning session, 10 min before METH administration (1 mg/kg, i.p.), Mg132 or Lactacystin (0.5 µg/side) was bilaterally microinjected into the dorsal striatum. Each infusion volume was 2 µl per side infused at a rate of 0.4 µl/min. After injection, rats were given an additional 2 min to allow drug diffusion.

Histology
Animals were deeply anesthetized with sodium pentobarbital and received a transcardial perfusion with 0.9% saline followed by 4% paraformaldehyde in PBS. The brains were removed and post fixed in 4% paraformaldehyde for 24 h and then kept in a 30% sucrose/PBS solution for 3-5 days to dehydration. The brains were cut coronally in 30 µm thickness on a cryostat (Leica). The tissue was stained with cresyl violet, and examined by light microscopy to detect the injection sites. Only animals with right injection sites were included for data analysis.

Lentivirus and Adeno-Associated Virus (AAV)
The SYVN1 lentivirus and adeno-associated virus (AAV) were purchased from Obio Technology (Shanghai) Co., Ltd. The lentiviral stocks (1.5-2.0 × 10 8 integration units (IU)/ml) were used to infect cultured neurons. The fluorescence was examined and the lysates were collected at 96 h after infection.
The AAV stocks (3.0-3.5 × 10 12 IU/mL) were used to infect rodent Dstr. Rats were anesthetized and the guide cannula (26 gauge) was bilaterally implanted in the dorsal striatum (anteroposterior: +0.6 mm; mediolateral: ±2.7; dorsoventral: −2.0 mm). Bilateral microinfusions were made through 31-gauge injection needles (2 mm below the tip of the guide cannulae), connecting to a 10 µl microsyringe mounted in the microinfusion pump (Harvard Apparatus). AAV (1.5 µl/site) was microinjected into the dorsal striatum over 15 min. The microinjection needle was retained for another 5 min for drug diffusion. Two to three weeks after AAV-injection, animals were deeply anesthetized and received a transcardial perfusion. The brains were collected, dehydrated and cut for fluorescence examining.

Endoplasmic Reticulum Isolation
The ER was isolated from rat dorsal striatum using a commercially available kit (Endoplasmic Reticulum Isolation Kit, Sigma) according to the supplier's instructions. Isolated ER was stored at −20 • C for analysis.

Data Analysis
The data were analyzed with one-way or two-way analysis of variance (ANOVA), and then followed by Newman-Keuls FIGURE 5 | Knockdown of Dstr SYVN1 increased the level of Glucose-regulated protein 78 (GRP78) and C/EBP homologous protein (CHOP) in METH-CPP rat. (A,B) Conditioned METH pairing did not affect Dstr GRP78 and CHOP expression level. Error bars represent mean ± SEM, n = 6 in each group. p > 0.05 compared with the control group, two tailed Student's t-test. (C,D) Knockdown of SYVN1 increased Dstr GRP78 and CHOP levels in METH-CPP rat. Error bars represent mean ± SEM, n = 6 in each group. * * * p < 0.001, * p < 0.05 compared with the corresponding control groups, one-way ANOVA with Newman-Keuls post hoc test.
or Bonferroni post hoc tests. Differences with p < 0.05 were considered statistically significant. The results are presented as mean ± SEM.

Decrease in GABA A α1 Protein Expression in the Dstr of METH-CPP Rat
We previously found that conditioned METH rewarding reduced GABA A α1 protein expression in the Dstr. This result was confirmed in the present study (p < 0.01 vs. control group; Figure 1A). And we did not find any significant changes in the other subunits protein levels of GABA A α2, GABA A α3, GABA A α5 and GABA A β2 in Meth-CPP rat (p > 0.05; Figures 1B-E). We further detected the GABA A α1 mRNA expression in the Dstr, and we found that there were no expression changes of the GABA A α1 mRNA (p > 0.05; Figure 1F). These results indicated that GABA A α1 levels were lower in the Dstr after METH pairing, and it occurred at the post-translational level.

Increased SYVN1 Protein Expression in the Dstr of METH-CPP Rat
ERAD has been reported to be involved in the degradation of un-or misfolded GABA receptors (Kang and Macdonald, 2004;Gallagher et al., 2005). The mammalian ERAD systems are organized primarily by two E3 ubiquitin ligases: SYVN1 and Glycoprotein 78 (Gp78; Christianson et al., 2011). We thus examined the protein expression levels of SYVN1 and Gp78, and we found that the SYVN1 level, but not Gp78 level, was significantly increased (p < 0.05 vs. control group; Figure 2). These results suggested that the SYVN 1 might be a key component that associated with degradation of GABA A α1.
The Ubiquitin-Proteasome System (UPS) Was Involved in the Degradation of GABA A α1 SYVN1, as an E3 ubiquitin ligase, significantly increased in the Dstr, suggesting that ubiquitin-proteasome system (UPS) might be involved in METH-induced CPP formation. We thus tested the level of ubiquitin protein in Dstr of METH-CPP rat, and we found that the expression level of ubiquitin protein was FIGURE 7 | SYVN1, as the endoplasmic reticulum associated degradation (ERAD) E3 ubiquitin ligase was involved in Dstr GABA A α1 degradation associated with METH conditioned pairing. Normally, GABA A α1 proteins was correctly folded in ER, and then exported to cytosol or assembled to the cell membrane. METH treatment caused misfolded or inappropriate GABA A α1 accumulated in the ER, induced ERS and led to increase of GRP78 to help to modifying misfolded GABA A α1 proteins. Misfolded GABA A α1 proteins are then delivered to SYVN1 and degraded by UPS.

DISCUSSION
The present study demonstrated that METH CPP formation was accompanied by a significant reduction of GABA A α1 expression in the Dstr. The change in GABA A α1 expression occurs at the post-transcription level. SYVN1, as an ERAD E3 ligase, was associated with UPS-mediated GABA A α1 degradation.
The GABA A receptors are the major inhibitory receptors in the central nervous system and can mediate fast post synaptic inhibitory effects. The α1 subunit-typed GABA A receptors are the most abundant composition subtype (Wisden et al., 1992;Fritschy et al., 1994;Liu and Wong-Riley, 2004) and appear to be correlated with drug addiction. Human genetic research demonstrated that the single nucleotide polymorphism (SNP) at rs2279020 of GABA A α1 subunit gene was associated with ATS-dependence (Lin et al., 2003). Liu et al. (2007) reported that continuous 7-days treatment with cocaine decreased GABA A α1 expression in mouse brain induced by cocaine . Consistently, we found that the expression level GABA A α1, but not the other subunits GABA A α2, GABA A α3, GABA A α5 and GABA A β2, was decreased in the Dstr of rats after conditioned METH pairing. We previously demonstrated that the GABA A α1 expression level in the Dstr remained unchanged in the unpaired group, indicating the changes of GABA A α1 were related to METH-pairing, rather than the pure drug effects (Jiao et al., 2016). Thus, the decrease of GABA A α1 contributed to METH-associated rewarding memory formation. We found that the GABA A α1 mRNA level did no change, demonstrating that the change in GABA A α1 expression occurred at the post-transcription level. METH-induced down-regulation of GABA A α1 might cause decreased GABAergic inhibition, which triggered DA release in the Dstr, leading to a motivation for drug seeking.
ERAD is the process by which the ER directs the degradation of misfolded or inappropriate proteins (Olzmann et al., 2013;Ruggiano et al., 2014). When misfolded proteins accumulated in ER, the ER resident chaperone GRP78 increased for modifying misfolded or inappropriate proteins (Yu et al., 1999;Walter and Ron, 2011), and increased CHOP would trigger cell apoptosis if ER dysfunction was severe or prolonged. Furthermore, misfolded and inappropriate proteins would be delivered to E3 ubiquitin ligases and then ubiquitinated and degraded by the UPS. The mammalian ERAD systems were organized primarily by two E3 ubiquitin ligases: SYVN1 and Gp78 (Christianson et al., 2011). In the present study, we demonstrated that the ubiquitin-proteasome degradation via ERAD contributed to the regulation of GABA A α1 receptor expression during conditioned METH pairing. Several lines of evidence supported this conclusion. First, we demonstrated that conditioned METH pairing decreased GABA A α1 receptor expression and increased E3 ubiquitin ligase SYVN1 expression. The interactions between GABA A α1 receptor and SYVN1 were presented in rodent Dstr. We further showed that knockdown of SYVN1 in primary striatum neurons and rodent Dstr both increased GABA A α1 receptor expression level. Moreover, we found that conditioned METH pairing enhanced expression of ubiquitin protein in Dstr, and intra-Dstr injection of proteasomal inhibitor Lactacystin and MG132 reversed the changes of GABA A α1 expression. Thus, our data provided solid evidence supporting that UPS, as one part of ERAD, was associated with the degradation of GABA A α1. Li et al. (2008) and Mao et al. (2009) reported that repeated METH administration could increase ubiquitin conjugation in striatal neurons of rat (Li et al., 2008;Mao et al., 2009).
Different pathogenic factors, such as abnormal calcium regulation (Pyrko et al., 2007), viral infection (Isler et al., 2005), high-fat feeding (Deldicque et al., 2010), various mutations (Chen et al., 2013) as well as METH or cocaine abuse (Jayanthi et al., 2004(Jayanthi et al., , 2009Beauvais et al., 2011;Pavlovsky et al., 2013) may disrupt ER homeostasis and cause ER stress. ER stress is associated with increased expression of the ER resident chaperone GRP78 (Yu et al., 1999) and the nuclear protein CHOP (Welihinda et al., 1999). It has been found that repeated METH administration can increase GRP78 and CHOP expression in the striatum (Jayanthi et al., 2004(Jayanthi et al., , 2009. However, in the present study, the expression level of GRP78 and CHOP was not altered in the Dstr. One possibility was that the dose and frequency of METH administration used in Jayanthi et al. (2004Jayanthi et al. ( , 2009) was much higher (40 mg/kg, once a day, for seven times) than we used in the present work (1 mg/kg, once 2 days, for four times). The low dose and low frequency of METH treatment may only induce GABA A α1 degradation via ERAD, without affecting the level of GRP78 and CHOP. Although conditioned CPP pairing was not able to change the expression level of GRP 78 and CHOP, our study showed that in vivo knockdown of SYVN1 significantly increased the level of GRP 78 and CHOP in the Dstr. In vivo knockdown of Dstr SYVN1 increased the expression level of total GABA A α1 and GABA A α1 in the intra-ER, without affecting GABA A α1 level in the extra-ER. Furthermore, knockdown of Dstr SYVN1 was not able to affect METH CPP formation. It was reasonable to speculate that knockdown of Dstr SYVN1 might reduce degradation of misfolded and inappropriate GABA A α1, and accumulation of these misfolded protein aggregates in ER would result in ERS. Thus, increase of GABA A α1 induced by knockdown of Dstr SYVN1 was not able to prevent METH CPP behavior.

CONCLUSION
In summary, the present study demonstrated that the UPS via ERAD contributed to GABA A α1 receptor degradation during conditioned METH pairing. SYVN1, an ERAD E3 ubiquitin ligase, was the key component regulating GABA A α1 receptor expression (Figure 7).