| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Institute of Molecular Pharmacology, Berlin, Germany;
* Department of Pharmacology, Erasmus Medical Center, Rotterdam, The Netherlands;
# Department of Cardiology, Charité Berlin, Campus Benjamin Franklin, Berlin, Germany; and
Max Delbrück Center for Molecular Medicine, Berlin, Germany
1Correspondence: FMP (Forschungsinstitut für Molekulare Pharmakologie), Robert-Rössle-Strasse 10, D-13125 Berlin, Germany. E-mail: siems{at}fmp-berlin.de
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: angiotensin • dopamine • alcohol • knockout mice • transgenic rats
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2
3)
.
Interactions between the RAS and the endogenous reward system have frequently been discussed, especially in terms of alcohol intake and dependency. The RAS-related actions on alcohol intake were found to be mainly related to Ang II, but despite intense research (4
5
6
7)
the results remained contradictory and pharmacological interventions often failed.
We recently found that voluntary alcohol consumption is directly connected to Ang II generation (8)
. Mice harboring an angiotensinogen transgene of the rat consumed significantly more alcohol than their wild-type littermates. This effect was blunted by the ACE inhibitor Spirapril. Accordingly, mice lacking the angiotensinogen gene exhibited a lower alcohol intake (8)
. These results indicated a role for the angiotensinogen gene in alcohol consumption via modulation of endogenous Ang II levels. However, these experiments left open whether specifically central or peripheral Ang II triggered these effects and which subtypes of the participating receptors and downstream signaling pathways are involved. We have attempted to solve these problems using genetically modified animals. We used several receptor-deficient mice as well as transgenic rats that expressed an angiotensinogen-specific antisense RNA in the central nervous system (CNS). Consequently, these rats exhibited a >90% reduced angiotensinogen level in the brain, resulting in a reduction of the central Ang II formation. These animal models thus enabled the certain description of the role of central Ang II in the modulation of alcohol consumption. Moreover, studies with the knockout and transgenic animals provided clear insight into participating receptors and definite hints on downstream signal transduction processes.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
were obtained from breeding stocks of M.B. at the Max Delbrück Center for Molecular Medicine (MDC) in Berlin, Germany. Angiotensin receptor AT1A knockout mice (10)
and AT2 knockout mice (11)
were obtained from breeding stocks of T.W. at the Campus Benjamin Franklin (CBF) in Berlin, Germany.
Bradykinin B2 receptor-deficient mice of a mixed genetic background (J129Sv(/Ev) x C57BL/6J) (12)
were purchased from Jackson-Laboratories (Jacksonville, FL USA). A genetic backcross to a C57BL/6J background was performed over seven generations before animals were used in experimentation. During backcross breeding, the genotype of the animals was checked by PCR according to a specific protocol.
Preference tests
Animals were housed in groups of two per cage but separated by a transparent plastic divider with several holes. This setting allowed the animals to have social contact. However, they were prevented from inflicting bodily harm on each other in ranking fights. Animals were housed at 22 ± 1°C in a 12 h/12 h light/dark cycle with food and beverage available ad libitum. All animals were habituated to drink from two tap water bottles for 2 wk. Eventually they were constantly given the free choice between a bottle of tap water and a bottle with a solution of either sucrose, saccharin, quinine, or alcohol. The influence of the ACE inhibitor Spirapril (AWD, Dresden, Germany) was tested by adding an individually adjusted amount to the drinking water (5 mg/kg body weight/day). The intake of water and alcoholic solution was recorded separately for each animal every 24 h. Bottle positions were changed every day. Water, alcoholic, and Spirapril solutions were renewed every second day.
Peptidase activities
After 3 wk of treatment with 5 mg Spirapril per kilogram body weight, animals were killed by decapitation. The brains were rapidly removed and stored until use at –80°C. Homogenates were prepared in a 50 mM Tris buffer, pH 7.4, using a glass-Teflon potter at 4°C, filtered through nylon gauze, and stored until use at –80°C. ACE activity was measured using hippuryl-histidyl-leucine as substrate and histidyl-leucine as standard reagent with a fluorimetric method as described by Maul et al. (8)
. Specificity controls were carried out with 10–6 M of the specific ACE inhibitor Lisinopril (Sigma-Aldrich, Taufkirchen, Germany). The total protein contents of the homogenates were determined by a Bradford assay (13)
.
NEP activities were measured in the homogenates using [D-Ala2, Leu5]enkephalin (DALEK) (Sigma-Aldrich, Taufkirchen, Germany) as substrate. The degradation was monitored by HPLC as described by Winkler et al. (14)
. To control the specificity of the reaction, we carried out experiments under specific NEP suppression with 10–6 M Candoxatrilat (Pfizer, Kent, UK).
Catecholamine measurements
Brains were quickly removed and dissected on chilled Petri dishes according to Popov et al. (15)
. Each of the dissected regions was added to 0.5 mL of cold perchloric acid (0.4 M). After homogenization in a glass-Teflon potter, samples were centrifuged at 20,000 x g for 10 min. Supernatants were analyzed for catecholamines, especially dopamine and its metabolite 3.4-dihydroxyphenylacetic acid (DOPAC), by HPLC (125 mmx3 mm I.D. column, packed with Nucleosil 100 C 18; 3 µm particle size) and electrochemical detection (INTRO, ANTEC Leyden, The Netherlands; cell potential=800 mV). The mobile phase consisted of 5% acetonitril, 10 g/L citric acid, 4 g/L KH2PO4, 0,1 g/L EDTA, and 0,175 g/L octanesulfonic acid; pH = 3.0. All values in means with SE.
Tyrosine hydroxylase activity
Tyrosine hydroxylase (TH) activity was detected by measuring the production of 3,4-dihydroxyphenylalanine (DOPA) from L-tyrosine. Brain regions (obtained as described above) were homogenized in a 30-fold volume of phosphate buffer (10 mM, pH 6.0). 50 µL of homogenate were added to 50 µL incubation mixture containing 100 mM phosphate buffer (pH 6.0), 0.6 mM tyrosine, 20 mM dithiothreitol, 3.75 mM 3-hydroxybenzylhydrazine (inhibitor of DOPA decarboxylase), and 0.6 mg/mL catalase. The reaction was started by addition of 50 µL 3 mM 6-methyl-5,6,7,8-tetrahydropteridine (cofactor). After 30 min of incubation at 37°C the reaction was terminated by adding of 50 µL perchloric acid (0.8 M) and 1 µM 
-methyl-DOPA as internal standard. After centrifugation the supernatant was added to a mixture of 20 mg Al2O3, 500 µL TRIS (3.0 M, pH 8.6) and EDTA (0.15 M) to adsorb the catecholamines to Al2O3. After stirring the mixture for 10 min, it was washed twice with 1 mL distilled water. Then the catecholamines were desorbed from Al2O3 by adding 200 µL of perchloric acid (0.4 M). The amount of formed DOPA was analyzed by HPLC with electrochemical detection (as described above) and corrected by the recovery rate of -methyl-DOPA.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 16)
and in their wild-type controls (each n=8). Wild-type rats were Sprague Dawley animals of the same age, the strain in which the transgenics have been generated. Both groups of rats are bred in the same unit of the animal house. At alcohol concentrations of 10% or higher, the animals increasingly avoided the alcoholic solution. At concentrations of 5% or less, the rats did not distinguish the alcoholic solution from water (data not shown). Thus we performed two-bottle, free choice experiments with permanent access to 7.5% alcohol and tap water. In this experimental design, wild-type rats consumed 2.57 ± 0.17 and TGR(ASrAOGEN)680 rat 1.84 ± 0.18 g per kg body weight per day (mean values for each of 16 animals over 10 consecutive days).
TGR(ASrAOGEN)680 rats drank significantly less alcohol solution than wild-type controls in all sessions. The preference ratio, expressed as proportion of ethanolic solution per total fluid consumption, of four consecutive drinking sessions with 7.5% alcohol and tap water in a two-bottle, free-choice experiment, is summarized in Fig. 1
. The total fluid consumption of the transgene rats was slightly elevated. This is due to a very moderate and age-dependent form of a diabetes insipidus-like syndrome with slightly elevated fluid intake, up to finally 25% elevation in the second half of life (9)
. In the first period without any treatment TGR(ASrAOGEN)680 rats consumed 85.2 ± 4.5, and wild-type controls 82.9 ± 3.5 g beverages /kg body weight.
|
To test whether the TGR(ASrAOGEN)680 would exhibit a generally altered affinity to any fluid other than water, we assayed their intake of sweet and bitter tastants in three different free choice paradigms. In a first experiment, TGR(ASrAOGEN)680 (n=8) and wild-type controls (n=8) were given the choice between tap water and 1.70% (w/v), followed by 4.25% sucrose in tap water over a period of 4 days. TGR(ASrAOGEN)680 were indistinguishable from controls in their preference for the sucrose solutions (Fig. 2
a). These findings make it unlikely that the altered alcohol consumption in TGR(ASrAOGEN)680 was calorie driven. We also did not find significant differences between TGR(ASrAOGEN)680 (n=8) and their wild-type controls (n=8) in the preference to water sweetened with saccharin (0.033% or 0.066% w/v) (Fig. 2b
). When offered quinine-embittered solutions (0.03 and 0.10 mM) vs. tap water, animals did not distinguish the smaller concentration, but both groups of animals equally avoided drinking from the bottle with 0.10 mM quinine (Fig. 2c
).
|
Spirapril, a lipophilic nonpeptidic substance (17)
, has been found to be an orally effective inhibitor of ACE activity in mice after application via feeding needles (8
, 18)
. The substance reduces peripheral as well as central ACE activity and leads to a decrease in Ang II concentration. We applied Spirapril (5 mg/kg body weight/day) via the drinking water, a procedure not accompanied by fixation stress. Lung and brain homogenates from rats administered with Spirapril displayed a significant reduction in ACE activity (Fig. 3
). The activity of the ACE-related enzyme neutral endopeptidase 24.11 (NEP) was not altered by Spirapril treatment (data not shown).
|
We tested whether Spirapril application resulted in a significant reduction of voluntary alcohol consumption. Spirapril significantly reduced alcohol consumption in wild-type animals (Fig. 4
a), but did not induce any changes in TGR(ASrAOGEN)680 (Fig. 4b
). A comparison of the alcohol intake ratio in TGR(ASrAOGEN)680 (Fig. 4b
) with that of wild-type rats under Spirapril treatment (Fig. 4a
) demonstrates that the ACE inhibitor could not reduce the intake in wild-type animals to the level of the transgenic rats.
|
Spirapril may act via decreasing the level of Ang II or increasing the level of bradykinin. To clarify the mechanism of Spirapril-mediated reduction of voluntary alcohol intake, we tested the alcohol preference ratio in mice lacking the receptors for these peptides—AT1A, AT2, and B2—and their corresponding wild-type littermates in a two-bottle, free choice paradigm using 10% alcohol in tap water. Exclusively, the ablation of AT1A receptors significantly reduced the alcohol intake (Fig. 5
a); the deficiencies in AT2 or B2 receptors were without any effect (Fig. 5b, c
).
|
The reduced availability of central Ang II in TGR(ASrAOGEN)680 and the known influence of this peptide on the central dopamine system (for a review, see ref 19
) prompted us to test catecholamine levels in brain areas important for reward. By HPLC and electrochemical detection, we found a significantly lower level of dopamine and its main metabolite DOPAC in the brain region tegmentum/colliculi, which includes the ventral tegmental area (VTA), of TGR(ASrAOGEN)680 rats compared with wild-type littermates (Fig. 6
a, b). The levels of dopamine and DOPAC were not significantly different in other regions of the two groups of rats (data not shown). Furthermore, we measured the activity of TH, a key enzyme of the dopamine synthesis. Differences as described for the dopamine or DOPAC levels were not found for the TH in the investigated regions of the CNS. We measured the brain region tegmentum/colliculi 66.4 ± 3.2 (TGR(ASrAOGEN)680) vs. 54.9±4.3 (wt) pg DOPA/mg/min.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
21
22
23)
.
Although the involvement of the RAS in alcohol-consuming behavior had been suggested (4
5
6
7)
, the specific function of its main peptide, Ang II, remained unclear. We recently demonstrated a direct correlation between endogenous Ang II levels and voluntary alcohol intake in genetically altered mice (8)
. Alcohol consumption in TGM123 mice, which show elevated Ang II levels resulting from additional expression of an angiotensinogen transgene (24)
, was significantly increased. On the other hand, the alcohol preference ratio in TLM mice lacking the angiotensinogen gene was less than that in their wild-type littermates. In spite of an excessive total fluid consumption due to disturbances in fluid homeostasis (10)
, these TLM mice drank even less alcohol than the controls (8)
. Moreover, it has been found that the ACE inhibitor Spirapril (17)
, known to cross the blood-brain barrier of mice (8
, 18)
and consequently lower the Ang II levels in brain and blood circulation, suppressed alcohol intake in animals with elevated Ang II levels (TGM123) (8)
. However, all these experiments could not distinguish between the peripheral and central effects of Ang II. Nevertheless, our own observation in TGM123 mice on increased alcohol consumption agrees with former findings of Fitts (6)
, who described an increased alcohol intake upon intracerebroventricular infusion of Ang II in rats. Fitts hypothesized on the importance of central Ang II for alcohol intake. Although assaying directly the contribution of central Ang II to alcohol consumption (6)
, these experiments are based on invasive techniques with unpredictable effects on drinking behavior (25
26
27)
. Moreover, the question whether peripheral Ang II also contributes to alcohol consumption was still elusive.
TGR(ASrAOGEN)680 is a transgenic rat that expresses an antisense RNA targeted against the 5' region of angiotensinogen mRNA under the control of the glial cell specific fibrillary acidic protein promoter (9)
. This leads to strongly reduced angiotensinogen levels in all regions of the brain (<<10%), whereas all peripheral components of the RAS were found to be completely unaffected (9
, 28
, 29)
. These rats represent a unique model to study the effects of central Ang II without any surgical interventions (9)
, thereby avoiding all external, perturbing influences, e.g., on voluntary alcohol consumption. In this model, we identified a key role of central Ang II in alcohol drinking behavior applying two-bottle, free choice experiments, and finally discriminated it from a possible influence of peripheral Ang II alterations. The difference between genetically altered rats and wild-type controls shown in Fig. 1
is not due to the known Ang II-induced strengthened thirst and increases in total fluid consumption. In spite of their slightly increased total fluid consumption (9; this study), TGR(ASrAOGEN)680 drank even substantially less alcohol than the controls. Because the genetically altered rats otherwise display normal behavior, we conclude that they selectively avoid ingesting the alcoholic beverage.
We proposed that suppression of Ang II formation and consequently its downstream interactions with the receptors AT1 and/or AT2 is the mechanism that triggers the alterations in alcohol drinking behavior (8)
. To define the receptor responsible for the ACE inhibitor-mediated effects, we used mice lacking genes for the AT1A or AT2 receptors. While loss of the AT1A receptor significantly reduced alcohol intake, loss of AT2 did not lead to substantial differences in alcohol consumption in mice. Obviously, the effects of central Ang II on voluntary alcohol consumption are mediated exclusively by the AT1 receptor.
Oral application of the ACE inhibitor Spirapril resulted in a reduction of alcohol consumption in wild-type rats. The ACE activity in their brain homogenates was reduced by about one-third. A similar decrease was observed in mice (8
, 18)
. Takai et al. (18)
showed that Spirapril also leads to changes in other behavioral parameters. They found that Spirapril-treated animals have an increased pain sensitivity (18)
. This study unequivocally showed that the mentioned effect was triggered exclusively by central ANG II, as the observed alterations in hot plate latency did not occur in mice treated with Enalapril or Captopril. These ACE inhibitors do not cross the blood-brain barrier. Accordingly, we conclude that Spirapril in our experiments leads to a reduction of alcohol consumption by inhibiting formation of central Ang II.
In contrast to the wild-types, Spirapril did not reduce the alcohol consumption of TGR(ASrAOGEN)680. This affirms in two ways the exclusive importance of central Ang II on alcohol consumption. 1) The already low level of central Ang II in these transgenic animals (<<10% of wild-types) is possibly further limited by Spirapril treatment, but this obviously cannot further veritably reduce the small alcohol intake in TGR(ASrAOGEN)680. 2) The peripheral ACE is unequivocally inhibited by Spirapril and consequently peripheral Ang II levels are decreased, but this does not have an impact on alcohol consumption. Consequently, only treatment of wild-types with the blood-brain-barrier crossing ACE inhibitor Spirapril reduces alcohol consumption, as a sufficient reduction of central Ang II formation is only possible in these animals.
Alcohol consumption has been suggested to depend on bradykinin actions (30
, 31)
on the constitutively expressed receptor B2. The absence of any difference between the preference ratios of B2-deficient mice and their wild-type littermates questions this hypothesis. This notion is further strengthened by the finding that Spirapril did not alter alcohol consumption of TGR(ASrAOGEN)680. In these animals, the substantial reduction of ACE activity by Spirapril treatment doubtlessly influenced bradykinin levels. However, this did not lead to any consequences in drinking behavior.
Ang II might lead to consumption of alcohol as a result of motivations other than pharmacological effects. Altered taste sensitivity and caloric contents of the alcoholic solution are the most probable factors of influence. However, since TGR(ASrAOGEN)680 were indistinguishable from their wild-type littermates in their response to saccharin, quinine, and sucrose, taste reactivity, and caloric drives are unlikely to have a decisive role in Ang II-modulated alcohol consumption.
Ang II belongs to a group of neuropeptides known to stimulate dopamine release in the brain (18
, 32)
. Moreover, angiotensin receptors are known to be expressed in brain areas such as the nucleus accumbens, where dopaminergic transmission is implicated in alcohol self-administration and sensitivity. The concept of a peptidergic modulation of reward, including alcohol consumption, via a mesolimbic dopamine-mediated transmission is frequently discussed (33
, 34)
. The dopamine system is believed to be the critical component in the reward circuitry of the mammalian CNS (35)
. According to this, we recently found that voluntary alcohol consumption of the angiotensinogen-overexpressing TGM123 mice was significantly reduced upon oral administration of the dopamine receptor antagonist fluphenazine, suggesting an interplay of Ang II and the dopaminergic system in alcohol drinking behavior (8)
. Indeed, D2 receptor-deficient mice have been shown to drink less alcohol in a two-bottle, free choice paradigm (36)
.
If the angiotensin-mediated dopamine release plays an essential role in Ang II-triggered regulation of alcohol intake, the dopamine concentrations should be altered in relevant brain areas of TGR(ASrAOGEN)680, animals in which angiotensin generation is drastically reduced in the CNS. Indeed, concentrations of dopamine as well as DOPAC were found to be strongly reduced in a region covering the VTA of TGR(ASrAOGEN)680 rats. These results clearly support our hypothesis that pharmacological manipulations of the central RAS, e.g., after oral application of ACE inhibitors reaching the CNS, alter the alcohol intake via AT1-mediated dopamine modulation (Fig. 7
).
|
Angiotensin has been shown to cause structural changes in the brain. For example, angiotensinogen-deficient mice exhibit a reduced cell number in the hippocampus (37)
, and mice lacking the AT2 receptor show an altered composition of cells in the amygdala and other parts of the brain (38)
, which may be the structural correlate to the X-linked mental retardation observed in patients with a mutation in this receptor (39)
. Furthermore, TGR(ASrAOGEN)680 differ in the development of the catecholaminergic neurons from respective controls (40)
. Thus, we cannot completely exclude that structural alterations in the brain of the genetically modified mice and rats may cause the observed alterations in dopamine release and alcohol consumption. However, the immediate effect of ACE inhibitors in our experiments with mice and rats, and the exclusive involvement of the AT1 receptor do not support developmental defects as cause for the observed phenotypes.
Measurement of the activity of tyrosine-hydroxylase, frequently regarded as a key enzyme of the dopamine biosynthesis, in several brain regions did not result in any relevant differences between wild-type and TGR(ASrAOGEN)680. The nature of the Ang II-induced modulation of the dopamine levels in the VTA-straddling brain regions remains unclear.
Up to now, effective causal therapeutic strategies for treatment of alcohol addiction are rare. Pharmacological interventions, including application of Acamprosate or Naltrexone, were developed mainly to help patients maintain the procedures of withdrawal. More detailed description of genetic factors involved in alcohol consumption and in the development of alcohol addiction might be helpful for prevention and therapy. Studies with genetically modified animal models, such as the one presented here, can unravel connections between gene products (e.g., components of the RAS), and the genetic risk to develop this devastating disease and can help to refine biochemical and molecular diagnostics. These experiments may lead to novel targets for rational drug design. According to this report, drugs interfering with the RAS could be of particular advantage because the available compounds (as ACE inhibitors and AT1 antagonists) are nearly free of severe side effects and widely used in the treatment of hypertension and in cardioprotection. As a consequence, future animal studies with the aim of developing drugs that influence drinking behavior by interfering with the RAS should be performed with lipophilic substances that effectively cross the blood-brain barrier. Finally, epidemiological studies should reveal how effective long-time treatment of modern ACE inhibitors or AT1 blockers can be in the treatment of alcohol addiction.
|
ACKNOWLEDGMENTS |
|---|
Received for publication February 7, 2005. Accepted for publication April 27, 2005.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
