A microdialysis profile of β-endorphin and catecholamines in the rat nucleus accumbens following alcohol administration
- First Online:
- Cite this article as:
- Marinelli, P.W., Quirion, R. & Gianoulakis, C. Psychopharmacology (2003) 169: 60. doi:10.1007/s00213-003-1490-2
Alcohol stimulates the release of dopamine in the nucleus accumbens (NACB) of rats, mice and humans. There is evidence to suggest that the activation of beta-endorphin (β-EP) in the mesolimbic pathway by alcohol and other drugs of abuse may be associated with the rise in dopamine levels in the NACB.
The present studies investigate whether the release of β-EP in the NACB is (1) dependent on the dose of alcohol that is administered, and (2) associated with changes in the extracellular concentrations of the catecholamines dopamine and norepinephrine, and the dopamine metabolites 3,4-dihydroxyphenylacetic acid (DOPAC) and homovanillic acid (HVA), in the NACB.
Male Sprague-Dawley rats were implanted with a microdialysis probe positioned in the shell region of the NACB. Artificial cerebrospinal fluid was pumped at a rate of 2.3 μl/min in awake and freely moving animals and the dialysate was collected at 30-min intervals. After a baseline period, rats were injected intraperitoneally with either physiological saline or one of three doses of alcohol: 0.8, 1.6, or 2.4 g ethanol/kg body weight. The dialysates collected were analyzed with radioimmunoassay, to estimate the content of β-EP; and high performance liquid chromatography, to estimate the content of dopamine, norepinephrine, DOPAC and HVA.
Alcohol induced a dose-dependent increase in the extracellular levels of β-EP and dopamine. However, elevations in the extracellular levels of norepinephrine, DOPAC and HVA did not reach significance. The largest increase in β-EP and dopamine was observed with the 2.4 g/kg dose.
The alcohol-induced release of β-EP and dopamine in the NACB is dose-dependent, where the highest dose resulted in more pronounced concentrations in the dialysate. Furthermore, the increase in the extracellular levels of dopamine appeared to occur at an earlier time point following alcohol administration, than for β-EP. These results suggest that alcohol stimulates dopamine and β-EP in the NACB, but probably does so via independent mechanisms.
KeywordsDopamineRadioimmunoassayMicrodialysisNorepinephrineHigh performance liquid chromatographyEthanol
Although its precise role is currently under considerable debate (Salamone and Correa 2002), the involvement of the nucleus accumbens (NACB) in drug and alcohol dependence is largely undisputed in the animal literature. Within this brain region, a considerable body of research has focused on dopaminergic innervation from the cell bodies of the ventral tegmental area (VTA) and how it serves in the physiology underlying substance abuse (Bozarth and Wise 1981; DiChiara and Imperato 1988). In the case of alcohol, administration through systemic injection (DiChiara and Imperato 1985) or via oral self-administration (Weiss et al. 1993) results in a dose dependent increase in the concentration of extracellular dopamine at the level of the NACB. Reverse dialysis of alcohol into this region also increases dopamine (Wozniak et al. 1991; Yoshimoto et al. 1991), while microinjecting alcohol into the VTA reinforces operant responding (Gatto et al. 1994). On the other hand, it is also well documented that the pharmacological blockade of dopamine receptors through systemic administration (Files et al. 1998) or intracranial microinjection of specific dopamine receptor antagonists at the level of the NACB (Rassnick et al. 1992) attenuates responding for alcohol.
Although the evidence implicating mesolimbic dopamine in alcohol reinforcement is strong, it is also clear that systems besides dopamine are involved. For example, mesolimbic dopamine cell destruction by means of 6-hydroxydopamine (6-OHDA) does not have a significant effect on responding for alcohol (Rassnick et al. 1993; Fahlke et al. 1994; Ikemoto et al. 1997). This paradox has led researchers to investigate other neurotransmitters that might be affected by alcohol. One candidate is the endogenous opioid, β-endorphin (β-EP). Antagonists to receptors for β-EP and other opioids reliably attenuate responding for alcohol in rats and mice when administered systemically (Hyytiä and Sinclair 1993; Stromberg et al. 1998; Middaugh et al. 1999) or in the NACB (Heyser et al. 1999). Indeed, it has been suggested that β-EP interacts with mesolimbic dopamine in order to mediate alcohol reinforcement, at least in part (Gianoulakis 2001). For example, the non-selective opioid antagonist, naltrexone, reduces self-administration of alcohol in rats and simultaneously inhibits alcohol-stimulated dopamine release (Gonzales and Weiss 1998; Middaugh et al. 2001). However, it is also observed that systemic naltrexone administration suppresses responding for alcohol in animals with dopamine depletions of more than 90% in the NACB (Koistinen et al. 2001). This suggests the existence of additional opioid mechanisms mediating alcohol consumption independent from those acting on mesolimbic dopamine.
Earlier in vitro studies have shown that low doses of alcohol (20–30 mM) can stimulate the release of hypothalamic β-EP (Gianoulakis 1990), but that much higher concentrations of alcohol (80–120 mM) are required to stimulate dopamine release from NACB slices (Snape and Engel 1988). Efforts to use in vitro methods to measure alcohol-stimulated β-EP release in the NACB have been unsuccessful, probably because endorphinergic terminals are severed from their cell bodies in the hypothalamus (Gianoulakis et al. 1999). Hence, investigators have turned to in vivo techniques, such as microdialysis. In a single published report, a 2.0 g/kg dose of alcohol caused a mild increase in β-EP release (Olive et al. 2001). However, this study used only one dose of alcohol and did not examine changes in extracellular concentrations of dopamine. Thus, it was the objective of the present studies to use an in vivo dialysis technique to: a) confirm the ability of alcohol to increase β-EP release in the NACB, b) test the hypothesis that alcohol will increase extracellular β-EP and dopamine concentrations in a dose-dependent manner, and c) determine whether an increase in extracellular dopamine is associated with an increase in norepinephrine, as well as the dopamine metabolites 3,4-dihydroxyphenylacetic acid (DOPAC) and homovanillic acid (HVA). DOPAC and HVA can serve as indicators of dopamine utilization, while norepinephrine was investigated because noradrenergic afferents stimulate β-EP release in the hypothalamus and anterior pituitary (Raymond et al. 1981; Tsingos and Chrousos 1995; Li et al. 1996), and modulate mesolimbic dopamine (Grenhoff et al. 1993). Furthermore, alcohol can increase norepinephrine turnover in the brain (Carlsson et al. 1973; Hunt and Majchrowicz 1974) while lesioning noradrenergic neurons can suppress alcohol intake (Brown and Amit 1977).
Materials and methods
All animals were male Sprague-Dawley rats (Charles River, St Constant, QC, Canada) weighing 415–535 g. Animal treatment was in accordance with McGill University's Policy on the Handling and Treatment of Laboratory Animals and the Canadian Council on Animal Care guidelines, and conforms to the Guide for the Care and Use of Laboratory Animals (Institute of Laboratory Animal Resources Commission Life Sciences National Research Council 1996). Each rat was caged individually with food and water available ad libitum and given at least 2 weeks to settle in the animal housing facility prior to experimentation. Animals were kept in a temperature and humidity controlled environment, and maintained on a 12-h light/dark cycle (lights on at 8:00 a.m.).
Rats were anaesthetised with a drug cocktail (50 mg/kg ketamine, 5 mg/kg xylazine, and 1 mg/kg acepromazine) prior to surgery. A stereotaxic apparatus was used to implant the guide cannula (20 mm shaft length; 1 mm o.d.; S.P.E., Concord, ON, Canada) unilaterally above the shell region of the NACB according to the following coordinates (in relation to bregma): +1.6 mm anteroposterior; +1.0 mm mediolateral; −6.0 mm dorsoventral (to dura mater). The cannula passage was blocked with an obturator that descended approximately 0.5 mm below its depth. The cannula was secured to the skull with dental cement anchored by three stainless steel screws. Animals were allowed 2–4 days to recover prior to habituation.
Habituation and microdialysis cages were identical, round, black, plastic cages measuring 31 cm in diameter and 30 cm high. On three alternate-day occasions, rats were moved to the microdialysis room and placed in their individual cages. During the habituation sessions, rats were not tethered and the cannula-obturator remained intact. On each occasion, rats were placed in their cages for 24 h beginning in the evening. The following day, a single intraperitoneal (IP) injection of sterile saline solution was administered to each rat to gradually acclimatize them to the handling they would experience with testing.
On the evening following their last day of habituation, rats were moved to the microdialysis room where they were lightly anaesthetised with isoflurane (Janssen, Toronto, ON, Canada) before being weighed and having their obturator replaced with the microdialysis probe (20 mm shaft length; 0.6 mm O.D.; 2 mm PES membrane, 15 kD cut-off; S.P.E.). The syringe pump (Harvard Apparatus, Holliston, Mass., USA) was set to infuse artificial cerebrospinal fluid (aCSF; 124 mM sodium chloride, 3 mM potassium chloride, 1 mM magnesium chloride·6H2O, 0.5 mM sodium phosphate monobasic·H2O, 5 mM sodium phosphate dibasic, 1.3 mM calcium chloride·2H2O, 0.2 mM L-ascorbic acid, 0.025% bovine serum albumin) overnight at a rate of 0.2 μl/min. PTFE tubing (0.56 mm i.d.; Cole-Parmer, Vernon Hills, Ill., USA) ran from the pump to a dual channel swivel (Instech, Plymouth Meeting, Pa., USA). The swivel was counter-balanced on an arm that swung laterally and vertically with the animal's movement. FEP tubing (0.12 mm i.d.; CSC, Montréal, QC, Canada) ran to and from the probe through a stainless steel spring tether (Plastics One, Roanoke, Va., USA) and dialysate was collected in a 500 μl polypropylene vial immersed in a container of iced water covered with aluminium foil. At 8:00 a.m. the next morning, the pump rate was increased to 2.3 μl/min. At this time, samples were collected at 30-min intervals for the duration of the study. At around 11:00 a.m., animals were injected IP with 0.0, 0.8, 1.6 or 2.4 g ethanol/kg as a solution of 0, 7, 14 or 21% (v/v) ethanol in 0.9% saline, respectively. This ensured that all animals were injected with identical volumes of fluid. In order to estimate catecholamine content, 13 μl of the dialysate were removed from the collection vial at the end of each interval, and were placed in another 500 μl polypropylene tube. All dialysate samples were promptly frozen in CO2 prior to storage in a –75°C freezer.
Blood alcohol concentration
In a separate cohort of rats from those used for microdialysis, tail blood was collected immediately prior, and at 30, 60, 90, 120, 150, 180 and 240 min following an IP injection of 0.8, 1.6 or 2.4 g ethanol/kg of 7, 14 or 21% (v/v) ethanol, respectively. Aliquots of 50 μl of blood was quickly pipetted into 450 μl of ice-cold trichloroacetic acid (6.25 w/v, Sigma-Aldrich, St Louis, Mo., USA) and vigorously shaken. Samples were then centrifuged for 6 min at 10,000 rpm; the supernatant was removed and frozen at −20°C. Samples were later analyzed using an ethanol assay (Diagnostic Chemicals Limited, Charlottetown, PE, Canada) based on a modified alcohol dehydrogenase spectrophotometric method (Lundquist 1957).
At the end of the microdialysis session, animals were again lightly re-anaesthetised with isoflurane, the probes were removed and animals were decapitated. Brains were quickly removed and snap frozen in isopentane, then stored in a –75°C freezer. After storage, the brains were sliced on a cryostat into 40 μm slices on gelatine-coated slides. Slides were stained for Nissl and inspected for probe placement.
Peptide concentration in the dialysate was determined with a solid phase radioimmunoassay (Maidment and Evans 1991; Olive et al. 2001). Ninety-six removable-well microplates (Dynex Microlite 2, Chantilly, Va., USA) were filled with 0.8 μg protein A (Sigma, St Louis, Mo., USA)/100 μl 0.1 M sodium bicarbonate (pH≈8.4) and incubated for 24 h at 4°C. The next day, wells were emptied and rinsed twice with 200 μl buffer (0.15 M potassium phosphate dibasic, 0.2 mM L-ascorbic acid, 0.1% Tween 20, 0.1% gelatin, pH adjusted to 7.4 with 10 N hydrochloric acid). A 50-μl aliquot of a 1:5000 dilution of antiserum specific for β-EP was placed in each well and incubated for 24 h at 4°C. Specific properties of the antibody have been previously reported (Gianoulakis and Gupta 1986). In brief, the antibody was specific for the C-terminal of β-EP and recognized proopiomelanocortin, β-lipotropin, and both acetylated and non-acetylated forms of β-EP 1–31, 1–27 and 1–26. This antibody did not recognize adrenocorticotropic hormone, α-melanotropin or β-lipotropin fragments 1–65, 62–67, and 80–84. The next day, wells were emptied and rinsed twice with buffer. Aliquots of 50 μl of dialysate samples or of standards (diluted in aCSF and ranging from 0.5 to 1000 pg/50 μl) were added to each well. After a 24-h incubation at 4°C, 50 μl of iodinated β-EP (5000 cpm/50 μl, specific activity ≅74 TBq/mmol; Amersham Pharmacia Biotech, Buckinghamshire, UK) was added to each well and left to incubate for an additional 48 h at 4°C. At the end of the incubation, wells were emptied, rinsed twice with buffer and placed into 5 ml polypropylene culture tubes to be counted by a γ-ray counter (Cobra II; Packard, Meriden, Conn., USA). The detection limit of these assays was 0.5 pg/tube and the IC50 was 51.39±6.49 (mean±SEM) pg/tube.
High performance liquid chromatography
Dialysate samples were thawed on iced water and 10 μl was used for HPLC analysis. The sample was loaded into a Rheodyne micro-injector (Rohnert Park, Calif., USA) with a 20 μl sample loop and injected into an analytical column (Beta-basic C18, 5 μm, 150×4.6 mm). The HPLC system was comprised of an ESA 582 solvent delivery module connected to an ESA Coulochem II (Chelmsford, Mass., USA) electrochemical detector. The infusion rate was set to 1 ml/min. The electrode parameters were as follows: −380 mV for guard cell; +300 mV on electrode one; and –280 mV, with a sensitivity setting of 500 nA, on electrode 2 (the reading electrode). Three dilutions of chemical standards (Sigma, St Louis, Mo., USA), ranging from 100 to 3000 ng/10 μl, were used to estimate the quantities of monoamines and their metabolites. The mobile phase consisted of 90 mM sodium phosphate monobasic·H2O, 50 mM citric acid, 1.7 mM 1-octanesulfonic acid disodium, 50 μM EDTA disodium and 10% acetonitrile (pH=2.8).
The effects of saline and alcohol on the extracellular levels of β-EP, dopamine, DOPAC, HVA and norepinephrine were estimated as a percent change from basal levels (the last two 30 min collection intervals prior to the IP saline or alcohol injection). Data are always represented as mean±SEM. Statistical analysis of the data was performed with a mixed two-way ANOVA, with simple ANOVAs used to interpret interactions and Tukey's Honestly Significant Difference post-hoc test used to interpret main effects. Significance for all tests was regarded at P<0.05.
Effects of various doses of alcohol on dialysate 3,4-dihydroxyphenylacetic acid (DOPAC), homovanillic acid (HVA) and norepinephrine (NE) in the nucleus accumbens, as determined in 30-min collection intervals. Results are expressed as a percentage of baseline levels. Basal levels were the average of the DOPAC, HVA and NE contents in the two 30-min dialysate collections preceding the IP injection of either saline or alcohol. Basal levels were 306±40.1 pg DOPAC/10 μl, 123±11.9 pg HVA/10 μl and 20±1.6 pg NE/10 μl. Each value represents the mean±SEM
Time interval (min) following IP administration
Ethanol (g/kg body wt)
An increase in extracellular dopamine in the NACB following exposure to alcohol has been demonstrated in rats (DiChiara and Imperato 1985; Weiss et al. 1993), mice (Middaugh et al. 2001) and humans (Pihl et al. 2001). The data presented here are in accordance with these findings, showing that a greater dose of alcohol induces a stronger release of dopamine in the NACB. However, the present studies do not show significant elevations with doses of 0.8 and 1.6 g ethanol/kg Some previous studies which have administered alcohol via IP injection have found an increase in dopamine in the NACB with doses as low as 0.5 g ethanol/kg (DiChiara and Imperato 1985; Tizabi et al. 2002). On the other hand, others have noted an overall effect of alcohol dose, but found that doses within the 0.5–2.0 g/kg range did not alter dopamine differently from placebo-injected rats of a selected line (Kiianmaa et al. 1995). Several groups have also noted that alcohol can increase dialysate concentrations of DOPAC and HVA in the NACB, albeit more modestly than dopamine (DiChiara and Imperato 1985). However, other studies did not observe this effect (Yoshimoto et al. 1992), in agreement with the present findings.
There are several factors in the current methodology that may explain the reduced response of dopamine in the NACB to alcohol, relative to other publications. Firstly, all previous studies used more frequent collection intervals, the majority having four or five, but some as many as 12 collections per hour (Yan 1999). Due to optimal conditions for peptide detection, the current procedure has only two. Consequently, these findings have a lower temporal resolution and a greater opportunity for transmitter degradation. Furthermore, the perfusion rate, also based on conditions favorable for β-EP analysis, is faster than the typical range for dopamine detection, which might diminish membrane recovery. This effect could be compensated by using larger, bilateral, transverse probes (Imperato and DiChiara 1986), but these sacrifice tissue integrity to a greater extent than the upright probes used here. Additionally, the current studies compared dose groups with a saline-treated control group while others have compared alcohol-induced changes in transmitter levels only to corrected baseline values (Imperato and DiChiara 1986; Tizabi et al. 2002). Finally, the phase of the circadian cycle of the rat during testing may also affect the response of NACB neurons to alcohol. For example, basal extracellular concentrations and manipulation-induced fluctuations of catecholamines in the NACB are less pronounced during rat's sleep, versus wake period (Feenstra et al. 2000). In behavioral studies, motor activity was found to show a very high correlation with striatal and NACB dopamine levels, and both locomotion and extracellular dopamine are significantly higher during the dark versus light phase of the animal's cycle (O'Neill and Fillenz 1985). Therefore, because animals in the current experiments were tested during their sleep period, the NACB may be less responsive to an IP injection of alcohol. Furthermore, since basal extracellular levels were determined in the morning, an apparent drop in transmitter levels of saline-treated animals (Fig. 3 and Table 1) may signify a normal decrease as sleep period progresses.
The data of the present studies are also in accordance with past studies showing a stimulatory effect of alcohol on β-EP systems. Hypothalami from rats and mice release β-EP in the presence of alcohol in vitro (de Waele et al. 1992, 1994). Also, acute administration of alcohol results in a greater expression of proopiomelanocortin mRNA in the hypothalamus and greater levels of β-EP in the VTA and NACB (Rasmussen et al. 1998). Furthermore, examination of Fig. 3 indicates that in animals administered 2.4 g ethanol/kg, the peak concentration of β-EP occurred at 90 min following the injection. This is in agreement with a recent report using Long Evans rats, showing that a 55±19% increase in β-EP levels in the NACB, relative to baseline, was observed at 90 min following a 2.0 g ethanol/kg injection (Olive et al. 2001). Thus, it appears that with the doses of alcohol used in the present study, the higher the dose of alcohol injected IP, the greater the release of β-EP in the NACB.
The interaction of endogenous opioids and mesolimbic dopamine is corroborated by studies showing that pharmacological stimulation of opioid receptors in the VTA produces an enhanced release of dopamine in the NACB, and that this effect is reversed with the application of opioid-receptor antagonists to the same region (Devine et al. 1993; Noel and Gratton 1995). This action is believed to be the result of the increased activity of endogenous opioids disinhibiting dopamine cell bodies in the VTA from the tonic inhibition imposed on them by neighbouring GABAergic neurons (Johnson and North 1992). In fact, this constitutes one of the proposed mechanisms involved in alcohol reinforcement (Froehlich and Li 1994; Gianoulakis 2001). In experiments examining oral self-administration, a systemic dose of naltrexone was found to abolish alcohol responding along with its associated dopamine elevations in the NACB (Gonzales and Weiss 1998; Middaugh et al. 2001). Presumably, naltrexone acts at the VTA to reduce opioid-inhibition of GABAergic cells, causing a greater suppression of mesolimbic dopamine release. It remains unknown whether systemic naltrexone could also act at the level of the NACB to suppress dopamine release.
The present data show that similar doses of alcohol stimulate the release of both dopamine and β-EP in the NACB. However, in relation to BAC (Fig. 1), there appears to be a temporal incongruity, where dopamine showed peak concentrations during the ascending phase of the BAC curve, while maximum β-EP levels were achieved only after BAC had reached its plateau. In the case of both the present in vivo microdialysis studies and previous in vitro perifusion studies of hypothalamic tissue (de Waele and Gianoulakis 1993), the maximum release of β-EP occurred after the concentration of alcohol in the blood (in vivo microdialysis) or incubation medium (in vitro perfusion) reached its maximum levels. However, with in vitro perfusion studies, the maximum concentration of alcohol and corresponding peak in β-EP release is more immediate than in the living animal. Also, in in-vitro investigations, brain tissue is severed from afferents that modulate the function of endorphinergic perikarya (Boyadjieva and Sarkar 1995; Tsingos and Chrousos 1995).
The temporal dissociation in peak dopamine and β-EP levels, relative to BAC probably suggests that β-EP in the NACB does not modulate dopamine activity as it does in the VTA. Indeed, it has been suggested that opiates acting at the VTA are reinforcing via a dopamine-dependent mechanism, while opiates acting at the NACB are reinforcing via a dopamine-independent mechanism (Hakan and Henriksen 1989; Jiang and North 1992). Thus, it is possible that the same might apply to alcohol-induced β-EP release in these regions. Alternatively, dopamine itself may act in the NACB to modulate other transmitters and peptides. Although neuromodulation is generally considered to be a property of neuropeptides, this is not always the case. Dopamine in the striatum, for example, modulates enkephalins, dynorphins and substance P release (Hökfelt et al. 2000).
Given the evidence to date, it is still difficult to tell whether a direct physiological interaction exists between the alcohol-induced release of dopamine and β-EP in the NACB. Even though the delivery of opioid receptor antagonists or μ-opioid receptor antisense directly into the NACB can reduce the self-administration of alcohol, it was not determined in these studies whether this effect was related to a deficiency in dopamine release (Heyser et al. 1999; Myers and Robinson 1999). At least one study has found that the focal application of a δ-opioid receptor antagonist will prevent an alcohol-stimulated dopamine release in the NACB (Acquas et al. 1993). Finally, in a recent experiment, it was shown that 6-OHDA lesioned rats did not show a reduction in alcohol intake, but did so following the administration of naltrexone (Koistinen et al. 2001). This last finding suggests that there are dopamine-independent opioid mechanisms supporting alcohol consumption.
Alcohol probably affects several other neurotransmitters which may foster dependence. For example, serotonin release from the NACB can be stimulated by alcohol administration (Yan 1999), and serotonin is known to stimulate β-EP release in the NACB (Zangen et al. 1999). Furthermore, there is some evidence to suggest that alcohol has an effect on noradrenergic systems (Carlsson et al. 1973; Hunt and Majchrowicz 1974; Brown and Amit 1977), and that these systems may modulate β-EP (Tsingos and Chrousos 1995) and dopamine (Grenhoff et al. 1993). The findings presented here do not demonstrate a significant alcohol-induced release of norepinephrine in the NACB. However, given the statistical proximity to a significant dose-effect (P=0.067) despite microdialysis conditions optimized for β-EP detection, the effect of alcohol on norepinephrine in the NACB warrants closer examination.
In conclusion, this study demonstrated that alcohol stimulated dopamine and β-EP release in the NACB in a dose dependent manner, while no significant elevations were found for norepinephrine, or for the dopamine metabolites, DOPAC and HVA. Furthermore, an apparent temporal dissociation between the peak increase of dopamine and β-EP, relative to BAC, was also observed. This dissociation may have a functional significance in mediating alcohol drinking.
These experiments were funded through a grant from the Natural Sciences and Engineering Research Council of Canada. Peter Marinelli is supported by a doctoral scholarship from Fonds de la recherche en santé du Québec.