|Home | About | Journals | Submit | Contact Us | Français|
The endocannabinoid (eCB) system is involved in pathways that regulate drug addiction and eCB‐mediated synaptic plasticity has been linked with addictive behaviours. Here, we investigated the molecular mechanisms underlying the changes in eCB‐dependent synaptic plasticity in the nucleus accumbens core (NAcc) following short‐term withdrawal from repeated morphine treatment.
Conditioned place preference (CPP) was used to evaluate the rewarding effects of morphine in rats. Evoked inhibitory postsynaptic currents of medium spiny neurons in NAcc were measured using whole‐cell patch‐clamp recordings. Changes in depolarization‐induced suppression of inhibition (DSI) in the NAcc were assessed to determine the effect of short‐term morphine withdrawal on the eCB system. To identify the potential modulation mechanism of short‐term morphine withdrawal on the eCB system, the expression of diacylglycerol lipase α (DGL‐α) and monoacylglycerol lipase was detected by Western blot analysis.
Repeated morphine administration for 7 days induced stable CPP. Compared with the saline group, the level of DSI in the NAcc was significantly increased in rats after short‐term morphine withdrawal. Furthermore, this increase in DSI coincided with a significant increase in the expression of DGL‐α.
Short‐term morphine withdrawal potentiates eCB modulation of inhibitory synaptic transmission in the NAcc. We also found that DGL‐α expression was elevated after short‐term morphine withdrawal, suggesting that the eCB 2‐arachidonyl‐glycerol but not anandamide mediates the increase in DSI. These findings provide useful insights into the mechanisms underlying eCB‐mediated plasticity in the NAcc during drug addiction.
This article is part of a themed section on Endocannabinoids. To view the other articles in this section visit http://onlinelibrary.wiley.com/doi/10.1111/bph.v173.7/issuetoc
Tables of Links
These Tables list key protein targets and ligands in this article which are hyperlinked to corresponding entries in http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS Guide to PHARMACOLOGY (Pawson etal., 2014) and are permanently archived in the Concise Guide to PHARMACOLOGY 2013/14 (a,b,cAlexander etal., 2013a, 2013b, 2013c).
The endocannabinoid (eCB) system regulates a wide variety of functions in the CNS, where eCBs often act retrogradely to activate presynaptic CB1 receptors at both excitatory and inhibitory synapses (Hillard etal., 2012; Starowicz and Di Marzo, 2013). CB1 receptors are expressed throughout the brain, but their density is particularly high in brain regions associated with reward circuitry and drug addiction, including the prefrontal cortex, amygdala, nucleus accumbens (NAc), striatum and hippocampus (Manzanares etal., 1999; Fattore etal., 2005; Carai etal., 2006; Robledo etal., 2008). It has been reported that mice lacking CB1 receptors (CB1 knockouts) fail to exhibit conditioned place preference (CPP) with morphine (Martin etal., 2000) and do not acquire morphine self‐administration (Ledent etal., 1999; Cossu etal., 2001). In addition, activation of CB1 receptors has been shown to reinstate heroin‐seeking behaviour (De Vries etal., 2003; Fattore etal., 2011), whereas blockade of CB1 receptors in a related region attenuates cue‐induced heroin‐seeking behaviour (Alvarez‐Jaimes etal., 2008). However, how the eCB system affects drug‐related addiction behaviours remains unclear.
Synaptic plasticity is defined as the ability of synapses to strengthen or weaken over time, following an increase or decrease in synaptic activity. It is an important process in the CNS as it is considered to be a cellular correlate of learning and memory, and addiction (Luscher, 2013). The eCB system has been implicated in several different forms of short‐ and long‐term synaptic plasticity in many brain areas important in drug reward and addiction (Chevaleyre etal., 2006; Sidhpura and Parsons, 2011). Of these areas, the NAc, in particular, plays a critical role in the development of morphine dependence and behavioural sensitization (Di Chiara etal., 1999). Depolarization‐induced suppression of inhibition (DSI) is a form of eCB‐dependent short‐term synaptic plasticity (Kreitzer and Regehr, 2001; Ohno‐Shosaku etal., 2001; Wilson and Nicoll, 2001), which has been shown to be involved in a variety of processes in vivo (Chevaleyre etal., 2006). In striatal medium spiny neurons (MSNs), DSI was abolished with the diacylglycerol lipase α (DGL‐α) inhibitor tetrahydrolipstatin, suggesting that 2‐arachidonoyl‐glycerol (2‐AG) is the major eCB mediating retrograde suppression at inhibitory synapses of MSNs (Uchigashima etal., 2007).
The eCB system in the NAc core (NAcc) has been linked to drug‐seeking behaviour and implicated mechanistically in processes underlying relapse to drug addiction (Alvarez‐Jaimes etal., 2008; Yuan etal., 2013). Although eCB‐mediated synaptic plasticity has been hypothesized to contribute to maladaptive behaviours caused by drug exposure (Sidhpura and Parsons, 2011), there is little evidence available regarding morphine‐induced alterations in eCB‐mediated synaptic plasticity in the NAcc. Therefore, we investigated changes in the modulation of synaptic transmission by the eCB system in the NAcc after repeated morphine exposure or short‐term withdrawal from repeated morphine treatment. Furthermore, we explored potential molecular mechanisms that underlie these processes.
Adult male Sprague Dawley rats (150–250g) were obtained from the Animal Care Committee of the Fourth Military Medical University (Xi'an, China). Rats were housed in a colony room under standard room temperature (20–23°C) and humidity (approximately 50%) conditions with a 12–12h light–dark cycle. The rats were housed conventionally in individual stainless steel cages, with food and tap water available ad libitum. All experimental protocols and housing arrangements were approved by the Ministry of Health of China and had received ethical approval from the institutional ethical committee of the Fourth Military Medical University. Either saline (0.9% NaCl) or morphine (10mg·kg−1) s.c. injections were given for 7 consecutive days followed by a 3 day short‐term withdrawal. Saline was obtained from Disai Biological Pharmaceutical Co. (Xi'an, China) and morphine was purchased from Shenyang No.1 Medical Drugs Co. (Shenyang, China). All studies involving animals are reported in accordance with the ARRIVE guidelines for reporting experiments involving animals (Kilkenny etal., 2010; McGrath etal., 2010).
Morphine‐rewarding effects were evaluated in a two‐chamber CPP apparatus as described previously (Randall etal., 1998). Briefly, the experiment consisted of three phases: preconditioning, conditioning and post‐conditioning.
In the preconditioning phase (days 1–3), rats were placed in the box for 15min and allowed to explore both compartments freely once a day. On day 3, the time spent in the white and black compartments was recorded and this was considered the initial ‘unconditioned preference’ of each rat for both compartments.
In the conditioning phase (day 3 or days 4–10), the rats received morphine (10mg·kg−1) s.c. daily and were then placed in the ‘non‐preferred’ compartment for 30min. After approximately 4h, the rats were administered saline and placed in the other compartment. For the saline group, rats were administered saline in both compartments. Morphine or saline was paired seven times to a specific compartment.
In the post‐conditioning phase (day 11), 15min post‐conditioning tests were conducted 24h after the last conditioning session. The rats in a drug‐free state were placed in the shuttle box with the door opened for 15min. The difference between post‐conditioning and preconditioning time spent in the drug‐paired compartment was recorded.
Both the saline‐treated and morphine‐treated rats (saline, n = 38; morphine, n = 46) were killed 4 or 72h after the last drug injection for electrophysiological recordings. Rats were anaesthetized with an i.p. injection (15mg·kg−1) of chloral hydrate (Aoxin Chemical Factory, Yangzhou, China) and decapitated. The brains were rapidly removed and submerged in ice‐cold artificial CSF containing (in mM): 126 NaCl, 25 NaHCO3, 2.5 KCl, 1.2 NaH2PO4, 2.4 CaCl2, 7 MgCl2 and 11D‐glucose (pH7.4 adjusted by HCl, saturated with 95% O2/5% CO2). All electrophysiological reagents were obtained from Sigma‐Aldrich. Coronal slices (300μM thick) containing NAc were cut with a Vibratome (Leica, Wetzlar, Germany) and placed in an incubating chamber at 32°C where they remained for at least 1h before the recording.
Whole‐cell recordings were made at 32°C from MSNs in NAcc or dorsal striatum using an upright microscope (BX50WI, Olympus, Tokyo, Japan) equipped with an infrared‐CCD camera system (Hamamatsu Photonics, Hamamatsu, Japan). MSNs were identified visually by their medium‐size and spindle‐like cell bodies. Resistance of the patch pipette (Upwards Teksystems, Ltd., Beijing, China) was 3–5MΩ when filled with the standard intracellular solution containing (in mM): 50 CsCl, 90 Cs‐gluconate, 10 HEPES, 1 EGTA, 0.1 CaCl2, 4.6 MgCl2, 4 Na‐ATP and 0.4 Na‐GTP (pH7.2, adjusted with CsOH). MSNs were usually held at a membrane potential of −80 mV and the pipette access resistance was compensated by 50–70%. During baseline recording, inhibitory inputs to MSNs in the NAcc or dorsal striatum were stimulated by a bipolar tungsten stimulating electrode at 0.2Hz for at least 1min. For induction of DSI, a depolarizing pulse (1, 3, 5 or 10s duration from −80 to 0mV) was applied through the recording electrode to the MSN. The paired‐pulse ratio (PPR) was monitored by applying a paired pulse with an interstimulus interval of 50ms. The magnitude of the DSI was calculated as the percentage difference between the mean amplitude of three consecutive IPSCs after depolarization relative to that of the mean amplitude of 10 consecutive IPSCs prior to depolarization.
Seventy‐two hours after the last drug injection, rats were killed and the NAcc tissues were obtained on an ice‐cold platform from a 1mm thick coronal section using a 14 gauge punch as described previously (Robison etal., 2013). Proteins were lysed in buffer containing: 50mM Tris‐HCl, 150mM NaCl, 1% Triton X‐100, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate, pH8.0, supplemented with 1% protease inhibitor cocktail (Sigma‐Aldrich). The protein concentration was determined using a bicinchoninic assay kit (Beyotime, Ltd., Haimen, China) according to the manufacturer's protocol. Equal amounts of protein from the NAcc or dorsal striatum were resolved on 8% SDS‐PAGE and electrophoretically transferred to PVDF membranes (Millipore, Billerica, MA, USA). The membranes were blocked for 2h in 5% skim milk diluted in PBS/tween (PBST, 0.01M PBS with 0.1% Tween 20) at 37°C with gentle shaking. Membranes were then incubated overnight with either polyclonal goat anti‐DGL‐α antibodies (1:1000; Abcam, Cambridge, UK) or polyclonal rabbit anti‐monoacylglycerol lipase (MGL) antibodies (1:1000; Abcam) at 4°C. After being washed in PBST, membranes were incubated with the designated secondary antibody for 2h at room temperature. Bands were visualized using an enhanced chemiluminescence system (Perkin Elmer, Waltham, MA, USA). Band intensity was quantified using ImageJ 1.46r (National Institutes of Health, Bethesda, MD, USA), where the expression of DGL‐α and MGL was normalized to β‐actin (1:5000, TA‐09; ZSGB‐BIO Co., Beijing, China).
Data are presented as mean ± SD and were analysed with SPSS 16.0 software (SPSS, Inc., Chicago, IL, USA). Statistical significance was determined by anova followed by least significant difference test. P < 0.05 was considered statistically significant.
All drugs were dissolved to their final desired concentration before use and were bath‐applied in the in vitro brain slice experiments. The concentrations of various drugs were as follows: bicuculline (10μM; Sigma‐Aldrich, Arklow, Ireland), 6‐cyano‐7‐nitroquinoxaline‐2,3‐dione (CNQX, 10μM; Sigma‐Aldrich), DL‐2‐amino‐5‐phosphonopentanoic acid (APV also known as AP‐5, 50μM; Sigma‐Aldrich), N‐(piperidin‐1‐yl)‐5‐(4‐lodophenyl)‐1‐(2,4‐dichlorophenyl)‐4‐methyl‐1H‐pyrazole‐3‐carboxamide (AM251, 4μM; Tocris Bioscience, Minneapolis, MN, USA), WIN 55,212‐2 (5μM; Tocris Bioscience), URB597 (1μM; Sigma‐Aldrich) and orlistat (10μM; Sigma‐Aldrich). CNQX, AM251, WIN 55,212‐2, URB597 and orlistat were made up as stock solutions in DMSO (final total concentration of DMSO was <0.05%). All other drugs were prepared as stock solutions in deionized water. CNQX was used as an AMPA receptor antagonist; APV was used as an NMDA receptor antagonist; orlistat was used as a DGL‐α inhibitor; URB597 was used as a fatty acid amide hydrolase (FAAH) inhibitor; AM251 was used as selective CB1 receptor antagonist and WIN55,212‐2 was used as a non‐selective CB1 receptor agonist (Alexander etal., 2013a, 2013b).
To investigate addictive behaviour in rats with morphine treatment, we administered morphine (10mg·kg−1) for 7 days and utilized the CPP paradigm. Figure1A shows that morphine‐treated adult Sprague Dawley rats exhibited significantly greater CPP than saline‐treated controls (P < 0.05, n = 8).
MSNs in NAcc are GABAergic projection neurons that comprise the vast majority (90–95%) of the cells in the NAc (Meredith, 1999; Zhou etal., 2002). Whole‐cell voltage‐clamp was performed in MSNs and IPSCs were evoked and recorded every 5s in the presence of the AMPA receptor antagonist CNQX (10μM) and the NMDA receptor antagonist APV (50μM). These IPSCs were abolished by GABAA receptor antagonist bicuculline (10μM, n = 3; data not shown), demonstrating that they were mediated primarily by the activation of GABAA receptors. There was no significant difference in the frequency or amplitude of spontaneous IPSCs in NAcc between control (1.46 ± 0.51 and 21.6 ± 4.1pA, respectively) and morphine‐ (1.62 ± 0.47 and 23.40 ± 3.9pA, respectively) treated groups (P > 0.05). Depolarization steps, from −80 to 0mV, induced a transient depression of IPSC amplitude in a time‐dependent manner (Figure1D). Maximal DSI was observed with a 5 and 10s depolarization step with no significant difference between these two duration times. Therefore, we used a 5s depolarization step for all other experiments. Furthermore, DSI was accompanied by a significant increase in the PPR (eIPSC2/eIPSC1) (Figure1E), suggesting this form of plasticity is expressed presynaptically. The reduction in IPSC amplitude and the change in PPR following the 5s depolarization were abolished by pretreatment with the CB1 receptor antagonist AM251 (4μM, n = 6), while AM251 by itself had no effect on IPSC amplitudes (95.4 ± 4.4%, n = 5). These findings indicated that DSI was mediated by activation of presynaptic CB1 receptors and that there was no tonic inhibition of basal inhibitory synaptic transmission by eCBs.
The eCB system has been implicated in the development of morphine‐induced CPP in NAcc (Azizi etal., 2009; Yuan etal., 2013; Haghparast etal., 2014). Thus, we measured DSI immediately after the CPP test to investigate possible alterations in eCB‐mediated DSI in NAcc after repeated morphine treatment. As shown in Figure2, the amplitude of the DSI was not significantly different between rats repeatedly treated with saline or morphine (P > 0.05).
Although the DSI was stably induced in NAcc neurons from both short‐term saline and morphine withdrawal rats, the magnitude of the DSI was significantly enhanced in the morphine withdrawal rats (n = 8) relative to the saline withdrawal rats (n = 7) (Figure3D). In both morphine and saline withdrawal groups, the transient suppression of eIPSCs induced by a 5s depolarization was associated with a significant increase in PPR. This increase in PPR was significantly greater in morphine withdrawal rats than saline withdrawal rats (P < 0.05) (Figure3C).
To further explore the potential mechanisms underlying the enhancement of DSI in rats during morphine withdrawal, we investigated the relative contribution of the two major eCBs in brain, 2‐AG and AEA (Figure4). DGL‐α is the main enzyme necessary for the synthesis of 2‐AG and FAAH degrades AEA. We used the DGL‐α inhibitor orlistat and the FAAH inhibitor URB597. Neither the magnitude nor the time course of DSI were affected by URB597, whereas orlistat blocked the DSI a concentration‐dependent manner in both groups. Full blockade of DSI occurred at an orlistat concentration of 10−5M (Figure4). After confirming that DSI was blocked by orlistat, we treated slices with the CB1 receptor agonist WIN 55,212‐2 and showed that CB1 receptor‐mediated inhibition of IPSCs was intact after orlistat treatment. This finding demonstrates that orlistat did not prevent DSI by acting on or inhibiting CB1 receptors.
Next, we measured the levels of DGL‐α and MGL (enzymes responsible for the synthesis and degradation of 2‐AG, respectively) in rats from saline and morphine withdrawal groups. As shown in Figure5A and C, the expression of DGL‐α was significantly increased in the NAcc of the morphine withdrawal group compared with the saline withdrawal group (P < 0.05). However, there was no significant difference in the expression of MGL between the two groups (Figures5B and D).
In addition, we investigated DSI after short‐term withdrawal from repeated morphine treatment in the dorsal striatum, another important region for drug addiction (Hyman etal., 2006). Unlike the NAcc, we found no significant difference in the DSI in the dorsal striatum between the short‐term morphine withdrawal and short‐term saline withdrawal groups (P > 0.05) (Figure6). Additionally, there was no significant difference in the expression of DGL‐α and MGL in this brain region between the two groups (P > 0.05). Based on our observations, the changes in eCB modulation of inhibitory transmission induced by a short‐term morphine withdrawal is probably limited to and specific for only specific brain regions.
Here, we investigated eCB‐mediated alterations in synaptic transmission in NAcc after short‐term withdrawal from repeated morphine treatment. Using behavioural and electrophysiological paradigms, we demonstrated that although repeated morphine administration did not cause any changes in DSI, short‐term morphine withdrawal significantly potentiated DSI in the NAcc. Furthermore, this increase in DSI was probably mediated by 2‐AG, as it was associated with a significant increase in the expression of DGL‐α.
The eCB system is an important regulatory system involved in physiological homeostasis and eCBs often act retrogradely on presynaptic CB1 receptors (Hillard etal., 2012; Starowicz and Di Marzo, 2013). According to our results, a significant CPP was induced in the rats after repeated morphine administration. This is consistent with results from previous studies, which showed that morphine is effective at reactivating opiate‐seeking behaviour in mice and that the CPP paradigm is useful for investigating mechanisms underlying relapse of drug abuse (Martin etal., 2000; Ribeiro Do Couto etal., 2003). Interestingly, drug‐seeking behaviour associated with a variety of abused substances can be reinstated or attenuated following CB1 receptor activation or antagonism respectively (De Vries etal., 2003; Alvarez‐Jaimes etal., 2008; Fattore etal., 2011). However, how the eCB system affects drug‐related addiction behaviours was unclear.
eCBs act as retrograde messengers at synapses in many different brain areas and their role in both short‐term and long‐term forms of synaptic plasticity is well documented (Sidhpura and Parsons, 2011). Modulating enzymatic pathways that regulate eCB levels have been shown to modify the expression of DSI (Hashimotodani etal., 2007; Gao etal., 2010), and thus, investigating changes in DSI can be used to clarify potential modulation mechanisms of the eCB system during drug addiction (Narushima etal., 2006; Uchigashima etal., 2007). In this study, we investigated alterations in the eCB system in the NAcc following repeated morphine exposure or short‐term withdrawal from repeated morphine treatment. We found that the magnitude of DSI in the NAcc is much greater in the morphine withdrawal group than the saline withdrawal group. Moreover, this enhancement in suppression was blocked by the application of orlistat, a DGL‐α inhibitor. These results suggest that the modulation of inhibitory synaptic transmission by eCB signalling is significantly potentiated in the rat NAcc after withdrawal from repeated morphine administration and this may explain why treatment with a CB1 receptor antagonist can reduce opiate withdrawal syndrome (Rubino etal., 2000; Wills etal., 2014). The major source of inhibitory GABAergic input to MSNs in the NAcc is derived from a small population of intrinsic CB1 receptor‐expressing interneurons (Koos and Tepper, 1999; Hohmann and Herkenham, 2000). These interneurons receive a strong glutamatergic input from the cortex and can induce feed‐forward inhibition of MSNs (Bennett and Bolam, 1994). Therefore, one function of eCBs in the NAc may be to disinhibit MSNs and to increase the output of the NAc to the ventral tegmental area and other targets. In other words, the alterations in the modulation of inhibitory synaptic transmission by eCBs observed in the present study may result in a greater disinhibition of MSNs and consequently the signal throughput from MSNs in the NAcc to their projection areas is increased. Given the important role of the NAc in reward and motivation behaviours, it follows that such changes in the NAc output may have profound effects on the reward circuitry and drug‐related behaviours.
Recently, activation of dopamine D2 receptors, increased CB1 receptor expression and/or signal transduction were all shown to be involved the plasticity of the eCB system in morphine‐dependent rats (Gonzalez etal., 2002; Centonze etal., 2007; Yuan etal., 2013). In addition, DGL‐α is an important enzyme responsible for the synthesis of 2‐AG and, by using the DGL‐α inhibitor tetrahydrolipstatin (orlistat), 2‐AG was shown to be the primary eCB mediating retrograde suppression at excitatory and inhibitory synapses on MSNs (Uchigashima etal., 2007). In the present study, we demonstrated that the eCB 2‐AG, but not AEA, regulates DSI in the NAcc of rats that experience short‐term withdrawal from repeated morphine treatment. Additionally, DSI was blocked in a concentration‐dependent manner in both groups by the DGL‐α inhibitor orlistat and the increase in DSI coincided with a significant increase in the expression of DGL‐α. These data reveal that an up‐regulation of DGL‐α may be an additional molecular mechanism underlying morphine withdrawal‐induced plasticity of the eCB system. In contrast, Vigano etal. showed that repeated morphine treatment decreased 2‐AG levels, whereas AEA levels were unaffected (Vigano etal., 2003; 2004). This disparity may be due to differences in the morphine administration protocol. In addition, we measured the changes in 2‐AG in living cells following an exogenous stimulation, whereas the other study examined basal levels in dissected tissues. Despite these differences, these findings and ours suggest that morphine treatment influences eCB homeostasis in different ways. The function of these two eCBs might differ depending on the phases of drug‐dependence and neuronal activity.
Although we have shown that short‐term morphine withdrawal was accompanied by an increase in DGL‐α expression, the mechanism underlying it remains to be determined. One possible mechanism is direct or indirect activation of cannabinoid signalling pathways induced by morphine that leads to a modification in the expression of DGL‐α. Alternatively, chronic morphine exposure may change the inhibitory inputs onto NAc neurons that then promotes feedback plasticity to increase both DGL‐α and 2‐AG. This latter scenario would explain the finding that 2‐AG successfully attenuated withdrawal signs in morphine‐dependent mice (Yamaguchi etal., 2001). Furthermore, we found that the magnitude of eCB‐mediated DSI was significantly increased only after a short‐term withdrawal but not by repeated morphine treatment alone. We suspect that DSI in the NAc, at least in our model, may not be involved in the development of CPP. GABA receptors play a modulatory role in the mechanism of action of different drugs of abuse (Xi etal., 2003; Yoon etal., 2010; Liang etal., 2014) and can alter eCB‐mediated plasticity of GABAergic transmission. Previously, activation of GABAA receptors was shown to inhibit the morphine withdrawal syndrome (Zarrindast and Mousa‐Ahmadi, 1999; Lee etal., 2011). This may explain the potentiated DSI during morphine withdrawal, a condition where there is less activation of GABAA receptors. Further studies using a combination of optogenetic and operant behavioural techniques are required to further elucidate the extent by which these morphine withdrawal‐induced alterations contribute to the withdrawal syndrome.
In conclusion, we demonstrated that short‐term morphine withdrawal potentiates eCB modulation of inhibitory synaptic transmission in the NAcc. We also demonstrated that 2‐AG, but not AEA, regulates synaptic transmission by increasing the expression of DGL‐α after short‐term morphine withdrawal. Our study provides important insights into the mechanisms underlying eCB‐mediated plasticity in the NAcc during drug addiction.
L‐J. H., G‐D. G., X‐Q. W. and J. M. participated in the research design. X‐Q. W., J. M., W. C., W‐X. Y. and G. Z. conducted the experiments. X‐Q. W., J. M. and Q. Y. performed the data analysis. X‐Q. W., J. M., L‐J. H. and W. C. wrote or contributed to the writing of the manuscript.
This work was funded by grants from the National Natural Science Foundation of China (No. 31070940 to G‐D. G. and No. 31100778 to L‐J. H.).
Wang X.‐Q., Ma J., Cui W., Yuan W.‐X., Zhu G., Yang Q., Heng L.‐J., and Gao G.‐D. (2016) The endocannabinoid system regulates synaptic transmission in nucleus accumbens by increasing DAGL‐α expression following short‐term morphine withdrawal. Br J Pharmacol, 173: 1143–1153. doi: 10.1111/bph.12969.