Selective Increase of Dopamine D3 Receptor Gene Expression As A

Home , Amineptine, Oxaprotiline

Selective increase of dopamine D3 receptor gene expression as a common effect of chronic antidepressant treatments C-H Lammers1,2*, J Diaz1*, J-C Schwartz3 and P Sokoloff3

1Laboratoire de Physiologie, Universite´ Rene´ Descartes, 4 Avenue de l’Observatoire, 75006 Paris, France; 3Unite´ de Neurobiologie et Pharmacologie Mole´culaire (U. 109) de l’Institut National de la Sante´ et Recherche Me´dicale, Centre Paul Broca, 2ter rue d’Ale´sia, 75014 Paris, France

The mesolimbic dopaminergic system is a neuroanatomical key structure for reward and motivation upon which previous studies indicated that antidepressant drugs exert a stimu- latory inﬂuence, via still unknown neurobiological mechanisms. Here we examined the effects of chronic administration of antidepressants of several classes (amitriptyline, desipramine, imipramine, ﬂuoxetine and tranylcypromine) and repeated electroconvulsive shock treatments

(ECT) on dopamine D3 receptor expression in the shell of the nucleus accumbens, a major projection area of the mesolimbic dopaminergic system. Short-term drug treatments had vari-

able effects on D3 receptor mRNA expression. In contrast, treatments for 21 days (with all drugs except fluoxetine) significantly increased D3 receptor mRNA expression in the shell of nucleus accumbens; D3 receptor binding was also significantly increased by amitriptyline or fluoxetine after a 42-day treatment. ECT for 10 days increased D3 receptor mRNA and binding in the shell of nucleus accumbens. D1 receptor and D2 receptor mRNAs were increased by imipramine and amitriptyline, but not by the other treatments. The time-course of altered D3 receptor expression, in line with the delayed clinical efficiency of antidepressant treatment, and the fact that various antidepressant drugs and ECT treatments eventually produced the

same effects, suggest that increased expression of the D3 receptor in the shell of nucleus accumbens is a common neurobiological mechanism of antidepressant treatments, resulting in enhanced responsiveness to the mesolimbic dopaminergic system. Molecular Psychiatry (2000) 5, 378–388. Keywords: depression; antidepressant drug; electroconvulsive shock; mesolimbic dopaminergic system

Introduction to an idiopathic manic episode7,8 and the discontinu- ation of such drugs9 or the acute administration of Considerable emphasis has been placed upon the puta- dopamine receptor antagonists can result in a psycho- tive role of nucleus accumbens dopamine systems in pathological state similar to a depressive episode.10 appetitive motivation and positive reinforcement.1,2 Clinical studies in depressed patients using [123I]IBZM- Hence, mesolimbic dopaminergic neurons projecting to SPECT have found increased D receptor binding in the the nucleus accumbens have been suggested to be 2 striatum probably reflecting reduced dopamine func- involved in the neurobiology of depression and the tion in depression.11,12 Furthermore, dopaminergic therapeutic actions of some antidepressant drugs.3–6 drugs such as the dopamine-uptake inhibitors This hypothesis postulates that decreased dopamine amineptine, nomifensin and buproprion have been activity is involved in depression, while increased successfully used for treating major depression.13–15 dopamine function contributes to mania. Accordingly, Chronic treatment with antidepressant drugs pro- dopaminergic drugs (eg amphetamine or cocaine) can duces in rats a variety of changes in dopaminergic produce effects in humans that are remarkably similar neurotransmission, most notably a sensitization of behavioral responses to agonists acting at dopamine D /D receptors within the nucleus accumbens,6,16 an Correspondence: P Sokoloff, Unite´ de Neurobiologie et Pharmaco- 2 3 logie Molećulaire (U.109) de l’Institut National de la Santeét increased electrophysiological activity of mesolimbic 17,18 Recherche Me´dicale, Centre Paul Broca, 2ter rue d’Ale´sia, 75014 dopaminergic neurons, an increase in D2/D3 recep- Paris, France tor binding sites19 and an increase in interstitial dopa- See also the Image page, Molecular Psychiatry 2000; 5: 229 mine concentrations in the nucleus accumbens.20,21 *These two authors have contributed equally to this paper 2Present address: Zentrum fur Nervenheilkunde, Klinik fur Neur- Similar results have been obtained after chronic ¨ ¨ 22–24 ologie, Rudolf-Bultmann Str 8, 35039 Marburg, Germany ECT. Received 3 January 2000; revised and accepted 17 March 2000 The D3 receptor is expressed mainly in the limbic Increase of dopamine D3 receptor expression C-H Lammers et al 379 ventral part of the striatal complex, particularly in the wet weight using 14Cor3H standard stripes shell of the nucleus accumbens,25,26 thought to be (Amersham). involved in reward, emotional, and cognitive processes.27 A recent study indicates that antidepressant Statistical analysis ± drugs administered repeatedly enhance D3 receptor Values reported are means SEM. Differences between binding.28 groups of animals treated with saline or antidepressant In this study we have examined the changes in cer- drugs were analyzed by one-way ANOVA followed by ebral expression of dopamine receptors, as well as a Dunnett’s multiple comparison post-hoc test when P Ͻ those in dynorphin and substance P, two neuropep- 0.05. Analysis of variations of D3 receptor mRNA tides whose expression is regulated by dopamine, in with treatment duration was performed by two-way rats receiving various kinds of antidepressant treat- ANOVA followed by a Least Significant Difference test. ment, including monoamine uptake inhibitors, a Effects of electroconvulsive shocks were analysed by monoamineoxidase inhibitor and ECT. Since responses the Mann–Whitney U test. to antidepressant drugs emerge in humans after several weeks of treatment, we have compared the effects of Results these drugs produced by either acute or repeated administrations. A daily regimen of antidepressant drugs of various chemical classes with different pharmacological pro- files (see Table 1) was administered for 21 days, and

Materials and methods dopamine D1,D2 and D3 receptor mRNA levels measured by quantitative in situ hybridization studies 24 h Drug treatments after the last injection. Hybridization signals were ana- Male Wistar rats (IFFA Credo, l’Arbresles, France (180– lyzed in several brain regions as specified in Figure 1e. 200 g)) were housed in groups of five in a temperature- As exemplified with amitriptyline (Figure 1a, b), D controlled room on a 12 h : 12 h light : dark schedule, 3 receptor mRNA level increased prominently in the light starting at 7.00 am, with free access to food and shell part of nucleus accumbens (particularly in the water. They received daily intraperitoneal injections ventral part) and ventromedial part of the striatum, (between 4.00 and 6.00 pm) of either desimipramine − − whereas lesser effects were produced in the core of (15 mg kg 1) or imipramine (15 mg kg 1), amitriptyline − − nucleus accumbens. Analysis of in situ hybridization (15 mg kg 1), fluoxetine (15 mg kg 1) or tranylcyprom- − signals at high magnification (Figure 1c, d) indicated ine (7.5 mg kg 1) for 1, 5, 10, 21 or 42 days. Rats were an increased D receptor mRNA per cell and not an killed by decapitation 24 h after the last injection and 3 increased number of D receptor mRNA-expressing the brain rapidly removed and frozen by dipping in 3 cells. isopentane maintained at −30°C. After 21-day treatments, other mixed norepinephrine/ serotonin uptake inhibitors, ie desip- Electroconvulsive shocks ramine and imipramine, as well as the monoaminoxid- Electroconvulsive treatment (85 V, 50 Hz, 0.05 s) was ase inhibitor tranylcypromine all increased, like ami- administered using ear clip electrodes in non anes- triptyline, D3 receptor mRNA by 35–54% in the shell thetised rats once a day for 10 days. Each treatment of nucleus accumbens, but fluoxetine produced an caused a typical tonic clonic convulsion for 20–35 s. opposite effect (Table 1). In addition, some compounds Sham treatment consisted in application of electrodes also increased D3 receptor mRNA in some other without current. Rats were killed 24 h after the last regions, ie desipramine in the fronto-parietal cortex, treatment and their brain rapidly dissected and frozen. septum and olfactory tubercle and islands of Calleja. No changes were observed in the dorso-lateral In situ hybridization and receptor autoradiography striatum. Coronal tissue sections (10 ␮m in striatal region and On the contrary, when imipramine, amitriptyline 20 ␮m in mesencephalic region) were cut on a cryostat and tranylcypromine were administered only once, and thaw-mounted on RNAse-free slides. Sections they elicited a decrease in the D3 receptor mRNA level were used either unfixed (in receptor autoradiography in the shell of nucleus accumbens; imipramine also experiments) or fixed (in in situ hybridization decreased D3 receptor mRNA in the islands of Calleja experiments) in 4% paraformaldehyde. In situ (Table 2). Analysis of the changes produced during hybridization experiments were performed with spe- prolonged treatment by amitriptyline (Figure 2) indi- 33 cific [ P] cRNA probes for D1,D2 and D3 receptor, cated that after an initial reduction, D3 receptor mRNA dynorphin or substance P mRNAs as described.25,29 in the shell of nucleus accumbens was rapidly nor-

D3 receptor binding autoradiography was performed malized within 5 days, was progressively and maxi- as described26 using 7-[3H]hydroxy-N,N-di-n-propyl- mally enhanced after a 21-day treatment, and had 2-aminotetralin ([3H]7-OH-DPAT, 130 Ci mmol−1, returned to normal levels after a 42-day treatment.

Amersham, Little Chalfont, Bucks, UK) as ligand. Auto- Changes in D3 receptor mRNA followed the same evol- radiographic signals were quantiﬁed on 2–3 slices per ution in the islands of Calleja, and were more pre- animal using an image analyser (IMSTAR, Paris, cocious in the core of nucleus accumbens (Figure 2). France). Gray values were converted to ␮Ci per gram In contrast, ﬂuoxetine produced changes in opposite

Molecular Psychiatry Increase of dopamine D3 receptor expression C-H Lammers et al 380 35 180 380* 210 380** 250 ± ± ± ± ± ± 0.0001 Ͻ 63 1,680 3084 1,460 2,900 46112* 1.620 5,740 56 1,200 ± ± ± ± ± ± 5 456 148.4 471 607 4.83* 393 967 7.4 308 ± ± ± ± ± ± 3.5 109 2.92.8* 101 95.5 1.54.7 70.5 1.5 135 47.2 ± ± ± ± ± ± 1.2 24.1 3.03.0 18.3 25.3 3.06.2 15.8 21.6 2.1 9.2 ± ± ± ± ± ± 4.0*6.1* 10.6 4.2 7.04.6 24.4 2.4* 7.4 10.0 3.3 8.2 ± ± ± ± ± ± 0.0001 0.094 0.0018 0.0001 0.0003 Ͻ receptor mRNA levels in various brain regions 3 8* 58.7 6* 15.3 5 32.7 7** 47.6 9**9** 35.6 38.6 ± ± ± ± ± ± shell striatum striatum 0.0001 Ͻ 5.67.31.3 135 70 138 10.1*4.6 154 135 5.6 100 ± ± ± ± ± ± cortex accumbens core lateral medial tubercle Calleja Fronto-parietal Ventral Accumbens Dorso- Ventro- Septum Olfactory Islands of NA 7 25.2 5HT 4 53.5 5HT 15 24.0 5HT 7 13.2 Ͼ = = saline. Ͼ proﬁle vs 0.01 Ͻ P Effects of antidepressant drug treatments for 21 days on D 0.05; ** value 0.043 Ͻ F[5,43]P 2.529 14.35 8.301 2.025 4.627 6.524 5.810 14.59 P ANOVA AmitryptilineFluoxetine NA 5HT DesipramineImipramine NA Tranylcypromine NA IMAO 3 17.4 Table 1 Treatment Pharmacological nSaline –Data are expressed 13 as* percentages of mean values obtained 20.1 in the ventral accumbens shell of saline-treated animals. Brain region

Molecular Psychiatry Increase of dopamine D3 receptor expression C-H Lammers et al 381

Figure 1 In situ hybridization pictures obtained with a D3 receptor cRNA probe in coronal brain sections of rats receiving repeated administration for 21 days of saline (a) or amitriptyline (b). Microphotographs of hybridization signals at the level of the shell of nucleus accumbens in animals treated for 21 days with saline (c) or amitriptyline (d). Bar = 10 ␮m. In (e) is shown a picture of a rat brain section taken from the atlas70 at 1.7 mm anterior to the bregma and displaying the various regions where hybridization signals and receptor binding were quantified by densitometry. Abbreviations: AcCo, core of nucleus accumbens; AcShV, shell of nucleus accumbens, ventral part; AcShD, shell of nucleus accumbens, dorsal part; DISt, dorsal striatum; FrPaCx, Fronto-parietal cortex; ICj, islands of Calleja; PvSt, periventricular striatum; Sp, septum; Tu, olfactory tubercle, Vm St, ventromedial striatum. directions as compared with the other antidepressant pressant drugs slowed down the time-dependent drugs (Tables 1, 2 and Figure 2). Thus, D3 receptor decrease in D3 receptor binding. Post-hoc analysis indi- mRNA in the shell of nucleus accumbens was initially cates that both amitriptyline and fluoxetine had sig- enhanced after a single administration, returned to nor- nificant effects after a 42-day treatment (P Ͻ 0.05). In mal levels within few days, was decreased after a 21- the core of nucleus accumbens, the effects were essen- day treatment and had returned to normal levels after tially the same as in the shell of nucleus accumbens, a 42-day treatment. There were similar changes in the although much limited in amplitude, the effects of core of nucleus accumbens, but no significant changes fluoxetine being not statistically significant. in the islands of Calleja (Figure 2). We also examined whether the observed changes In order to assess whether the changes observed produced by antidepressant drugs were specific to the in D3 receptor mRNA levels were accompanied by D3 receptor (Table 3). Imipramine and amitriptyline 3 changes in the receptor protein, binding of [ H]7-OH- increased the D1 receptor mRNA level in the shell of 26 DPAT, a selective D3 receptor ligand was measured. nucleus accumbens and the caudate-putamen, but These experiments were designed in such a way that desipramine and tranylcypromine were inactive. Imip- they allowed us to quantify D3 receptor binding in ramine and amitriptyline increased the D2 receptor absolute amounts, and thus to assess the changes after mRNA level, but only in the caudate-putamen. Fluox- repeated handling and/or injections with saline (Figure etine was unable to change D1 or D2 receptor mRNA 3). In fact, such treatment with either saline or antide- levels in either direction. pressant drugs produced a marked time-dependent We next examined the effects of a 21-day treatment reduction of [3H]7-OH-DPAT binding in the shell of with amitriptyline or fluoxetine on substance P and = Ͻ nucleus accumbens (F2,26 30.76, P 0.0001 by two- dynorphin mRNA levels known to be regulated by D1 29,30 way ANOVA, see Figure 3), core of nucleus accumbens and D3 receptors. Both amitriptyline and fluoxetine = = + + (F2,26 4.053, P 0.029, see Figure 3) and islands of significantly increased substance P ( 67% and 94%, = = Calleja (F2,26 4.270, P 0.0497, not shown). For respectively), but not dynorphin mRNA level, and this example, [3H]7-OH-DPAT binding after a 42-day treat- effect was restricted to the ventral part of shell of Ͻ ment by saline was reduced by 44% (P 0.001) and nucleus accumbens, the region where the D3 receptor 33% (P Ͻ 0.01) in the shell and core of nucleus accum- is expressed (Table 4). bens, respectively, compared with the levels after a 10- Finally, we investigated the effects of repeated ECT day treatment. In the shell of nucleus accumbens, there for 10 days, as another antidepressant treatment. ECT = = + + was an overall effect of treatment (F2,26 11.58, P increased D3 receptor mRNA ( 49%) and binding ( 0.00025 by two-way ANOVA) by amitriptyline or 42%) in the shell of nucleus accumbens and D3 recep- fluoxetine, and significant time × treatment interaction tor mRNA (+ 24%) in the islands of Calleja (Figure 4). = = (F4,26 3.465, P 0.021), which means that the antide- D1 and D2 receptor mRNA levels in the shell of nucleus

Molecular Psychiatry Increase of dopamine D3 receptor expression C-H Lammers et al 382 35 24 140 52* 120 ± ± ± ± ± 39 744 15 1,400 15 705 23 586 35 1,035 ± ± ± ± ± 1.5 227 10.9 440 6.6* 280 5.0* 274 9.8 314 ± ± ± ± ± 1.3 28.9 1.4 27.6 1.3 39.9 2.9* 85.6 2.3 69.8 ± ± ± ± ± 0.4 6.2 3.0 14.3 0 13.7 3.4 25.0 0 6.5 ± ± ± ± ± 1.8 0 1.78.5* 0.4 15.5 2.5 0 3.1 8.5 ± ± ± ± ± receptor mRNA levels in various brain regions 3 9.4* 19.3 1.0**9 13.1 39.4 9.9* 14.8 4 21.9 ± ± ± ± ± 2.5 78.2 2.24.0 68.9 124 2.2 76.0 4.2 100 ± ± ± ± ± cortex shell striatum striatum Fronto- Ventral Accumbens Dorso- Ventro- Septum Olfactory Islands of parietal accumbens core lateral medial tubercle Calleja saline. vs 0.01 Ͻ P Effects of a single injection of antidepressant drugs on D 0.05; ** value 0.139 0.0005 0.008 0.009 0.001 0.003 0.023 0.003 Ͻ F[4,15]P 2.048 9.800 5.181 4.985 8.223 6.399 3.905 6.407 P ANOVA Tranylcypromine 3 7.4 AmitriptylineFluoxetine 3 3 7.1 18.2 Imipramine 4 4.3 Table 2 TreatmentSaline n 7Data are expressed as* percentages of mean values obtained 15.9 in the ventral accumbens shell of saline-treated animals. Brain region

Molecular Psychiatry Increase of dopamine D3 receptor expression C-H Lammers et al 383

Figure 2 Time-course of changes in D3 receptor mRNA in the ventral shell of nucleus accumbens (left), the core of nucleus accumbens (middle) and the islands of Calleja (right) after repeated administration of amitriptyline or ﬂuoxetine. Results are expressed as percent of mRNA levels measured in animals receiving saline at the same time. *PϽ0.05 and **PϽ0.01 vs saline- treated animals at respective times by one-way analysis of variance.

Figure 3 Changes in D3 receptor binding in the nucleus accumbens after repeated administration of saline, amitriptyline or ﬂuoxetine for 42 days. (a) autoradiographic pictures; (b) quantitative analysis in the ventral shell of nucleus accumbens (left) and the core of nucleus accumbens (right). *PϽ0.05 and **PϽ0.01 vs saline-treated animals at respective times, §PϽ0.05; §§PϽ0.01 vs saline-treated animals at 10 or 21 days by two-way analysis of variance. accumbens were not signiﬁcantly changed by ECT swimming test31 and the tail suspension test.32 How- treatment (not shown): in treated animals, mRNA lev- ever, the antidepressant effect emerges in patients after els were 96 ± 6% (P = 0.45 by the Mann–Whitney U chronic but not acute treatment, enlightening the cru- test) and 84 ± 14% (P = 0.28) of the control values for cial importance of adaptive changes in promoting this

D1 and D2 receptors, respectively. activity. Moreover, the various antidepressant drugs have different primary pharmacological targets, which makes it difficult to define a final common mechanism Discussion of action, and the situation is even less clear regarding In spite of the long-standing clinical use of antidepress- other antidepressant treatments such as ECT. Our data ant drugs, the neural mechanisms underlying the thera- show that selective increase in D3 receptor binding or peutic effect of these drugs are still incompletely mRNA expression in the shell of nucleus accumbens understood. Most antidepressant drugs are active after accompanies all chronic antidepressant treatments that acute administration in mouse behavioral tests pre- were studied here. Some features of the response to dicting antidepressant activity, such as the forced these antidepressant treatments suggest that enhanced

Molecular Psychiatry Increase of dopamine D3 receptor expression C-H Lammers et al 384 26 25 4* 21 12 20* R 2 ± ± ± ± ± ± 14 196 17 147 4 192 9** 196 17** 211 6 140 RD 1 ± ± ± ± ± ± 26 135 35 120 6 139 29 150 14 139 15 93 RD 2 ± ± ± ± ± ± 10 173 11 170 9 174 11 223 11 178 7 138 RD 1 ± ± ± ± ± ± 11 111 15 91 11 119 18 118 10 102 589 RD 2 ± ± ± ± ± ± R) receptor mRNA levels in various brain regions 2 (D 2 17 96 17 100 6* 133 5 115 7 100 24** 141 RD 1 ± ± ± ± ± ± R) and D 1 (D 1 18 119 20 110 11 140 12 166 22 135 6 100 RD 2 ± ± ± ± ± ± 10 126 15 103 6 129 25 135 3 112 7 102 RD Accumbens core Ventral accumbens shell Dorsal accumbens shell Dorso-lateral striatum 1 ± ± ± ± ± ± D 0.208 0.319 0.0108 0.071 0.0871 0.1147 0.0089 0.0125 saline. vs 0.01 Ͻ P Effects of antidepressant drug treatments for 21 days on D 0.05; ** Ͻ F[5,25]P 1.558 1.243 3.794 2.357 2.193 2.002 3.948 3.672 P ANOVA Tranylcypromine 4 87 Fluoxetine 4 95 Amitriptyline 8 116 Imipramine 4 117 Desipramine 3 119 Table 3 TreatmentSaline n 8Data are expressed as* percentages of mean values obtained 90 in the ventral accumbens shell of saline-treated animals. Brain region

Molecular Psychiatry Increase of dopamine D3 receptor expression C-H Lammers et al 385 Table 4 Effects of a 21-day treatment with amitrityline and ﬂuoxetine on substance P and dynorphin mRNA levels in the shell of nucleus accumbens and caudate-putamen

Treatment n Substance P mRNA Dynorphin mRNA

Ventral Dorsal Ventromedial Ventral Dorsal Ventromedial accumbens accumbens striatum accumbens accumbens striatum shell shell shell shell

Saline 4 170 ± 37 141 ± 11 116 ± 10 173 ± 10 114 ± 678± 2 Amitriptyline 6 284 ± 31** 184 ± 21 128 ± 12 213 ± 28 151 ± 25 82 ± 5 Fluoxetine 6 329 ± 12* 200 ± 11 143 ± 11 275 ± 25 175 ± 21 99 ± 6 ANOVA F[2,13] 8.121 2.991 0.137 3.696 1.548 3.786 P 0.0051 0.085 0.873 0.059 0.2556 0.056

Data are expressed in ␮Ci per gram wet weight. *PϽ0.05; **PϽ0.01 vs saline.

action in rats as compared with other antidepressant drugs. A short-term exposure to fluoxetine decreases dopamine extracellular levels, whereas other antidepressant drugs have an opposite effect;20 this also exemplifies the fact that the acute effects of antidepressant drugs may not be relevant to their clinical efficacy. The discrepancies observed between the

effects of antidepressant drugs on D3 receptor binding and mRNA levels could be due to different turnover rates of mRNA and membrane proteins. In fact, adaptive mechanisms increasing degradation of short-lived mRNA could have progressively masked increased mRNA levels, when measured 24 h after the last drug administration. On the contrary, slowly increasing and Figure 4 Effects of electroconvulsive shocks on D3 receptor long-lived receptor protein, together with a delayed mRNA (left) and binding (right) in the ventral shell of nucleus action of fluoxetine, could account for the fact that sig- Ͻ accumbens (AcShV) or islands of Calleja (ICj). *P 0.05 nificant effects of antidepressant drugs could be and **PϽ0.01 vs shams by the Mann–Whitney U test. observed only after a long-term treatment. Thus, our results also suggest that receptor binding may be a bet- dopamine neurotransmission through this receptor ter indicator than receptor mRNA for the effects of anti- participates in the adaptive changes leading to antidepressant drugs. depressant activity. Whereas the various treatments used in the present One important feature is that antidepressant drugs, study, including inhibitors with different selectivity except fluoxetine, enhanced D3 receptor mRNA after for monoamine uptakes, a monoamine oxidase inhibi- chronic (21-day), but not acute treatment; in addition, tor and ECT, have most likely different short-term out-

fluoxetine and amitriptyline enhanced D3 receptor comes, they all increased D3 receptor expression upon binding after a 42-day treatment, which is in agreement long-term repeated administration. These observations, with the delayed therapeutic action of these drugs in together with those of a previous study showing that depressed patients. Several antidepressant drugs had other antidepressant drugs increase D3 receptor bind- 28 actually an opposite effect when tested after a single ing, show that enhanced D3 receptor expression is a administration: imipramine, amitriptyline and tranyl- common secondary action of antidepressant treat- cypromine significantly reduced D3 receptor mRNA in ments. Moreover, this common effect is restricted to the shell of nucleus accumbens. Several other antide- the shell of nucleus accumbens, that various studies pressant drugs were previously shown to enhance D3 have shown to subserve the role of dopamine in appeti- receptor binding after chronic treatment (14 days, two tive motivation, positive reinforcement, pleasurable injections per day), but to have much limited and vari- effect of reinforcing stimuli1,2 and therapeutic actions able effects after an acute injection.28 Interestingly, of some antidepressant drugs.3–6 Interestingly enough, fluoxetine, which markedly differed from other antide- ECT produced the most rapid and robust increase in D3 pressant drugs in its effects on D3 receptor mRNA, receptor protein and mRNA expression, as compared to especially after short chronic treatment, also increased antidepressant drugs, which is in agreement with their

D3 receptor protein in the shell of nucleus accumbens therapeutic efﬁcacy in refractory depression. after a 42-day chronic treatment, suggesting a delayed Although some antidepressant drugs elicit changes

Molecular Psychiatry Increase of dopamine D3 receptor expression C-H Lammers et al 386 in D1 and D2 receptor expression (this study and Dzied- to counteract this downregulation, thus opposing the zicka-Wasylewska et al),33,34 this effect is not common effects of stress. Accordingly, antidepressant drugs to all antidepressant drugs (see also Ainsworth et al),35 reverse chronic mild stress-induced decrease in the or occurs in different brain regions. Several previous consumption of palatable sweet solutions,44 an effect 19 studies have reported that ECT treatment increases D1 accompanied by increased D2/D3 receptor binding 34 and D2 receptor mRNA and D2 receptor protein in the and D3 receptor mRNA. These results point to a role 22,24 nucleus accumbens, but these changes are limited of the D3 receptor in the adaptive changes induced by in amplitude or short-lasting. Here we could conﬁrm chronic stress, a major risk factor in depression,45–48

that there were no changes in D1 or D2 receptor mRNAs that antidepressant treatments may correct by their in the nucleus accumbens 24 h after termination of effects on D3 receptor expression. chronic ECT. In contrast, chronic ECT induced a robust One can speculate about possible mechanisms by

and long-lasting increase in both D3 receptor binding which chronic stress and antidepressant treatments and mRNA in the shell of nucleus accumbens. A selec- regulate the D3 receptor gene in opposite directions. tive enhanced D3 receptor expression in this brain Acute stress activates the mesocortical dopaminergic region thus appears as a common consequence of vari- system,49 but chronic stress triggers adaptive processes ous antidepressant treatments. leading to decreased dopaminergic transmission in the Furthermore, our results show that antidepressant shell of nucleus accumbens.50,51 On the contrary, anti- 52 53 treatments not only increase D3 receptor expression, depressant drugs and ECT induce a subsensitivity but also elicit changes related to D3 receptor function. of dopamine autoreceptors, leading to a persistent In the shell of nucleus accumbens this receptor is enhancement of dopamine neuron electrical found expressed mainly in substance P- and dynor- activity.17,18 Although antidepressant drug treatments phin-expressing neurons30,36 and both neuropeptide have variable effects on basal extracellular dopamine expressions are inﬂuenced by pharmacological in the nucleus accumbens,54 they increase amphetam- 21,55 manipulations of D1 and D3 receptors in various experi- ine-evoked dopamine release and ECT treatments mental circumstances.29,30 In agreement with a pre- enhance spontaneous dopamine release.23 There is also vious observation,37 we found that amitriptyline and some evidence for a decreased dopamine metabolism ﬂuoxetine increased substance P mRNA in the shell of and turnover in depressed patients.3,11,12 Hence, the nucleus accumbens, which might be the result of an data in the literature are rather consistent with a

increased D3 receptor function. Since antidepressant decreased activity of mesolimbocortical dopamine treatments decrease substance P concentrations,38 they neurons by chronic stress and in depression, which is may accelerate substance P turnover and release. reversed by chronic antidepressant treatments. The Nevertheless, this hypothesis is hardly reconcilable convergent effects of antidepressant drugs with differ- with the antidepressant effect of a substance P receptor ent serotonin/norepinephrine pharmacological selec- antagonist recently reported.39 tivities on dopamine neuron activity can be explained It has been repeatedly shown that antidepressant as an adaptation of these neurons to the inhibitory treatments in rats signiﬁcantly inﬂuence dopamine actions of both serotonin and norepinephrine, sug- function in the mesolimbic system by increasing the gested by various anatomical and functional studies responsiveness of postsynaptic dopaminergic receptors (reviewed in Prisco and Esposito; and Grenhoff et 6 56,57 (see Willner for a review). Notably, the locomotor al). We have previously shown that D3 receptor stimulants effects of the D2/D3 receptor agonist quinpir- expression, unlike that of other dopamine receptor sub- ole are enhanced by repeated treatments with various types, depends critically upon dopamine neuron antidepressant drugs,40,41 an effect attributable to activity, through the release of an anterograde factor

increased D3 receptor gene expression. Moreover, different from dopamine itself and its known co-trans- enhancement of dopamine function by repeated ECT mitters.58 Thus, the opposite variations of dopamine

requires concomitant activation of D1 and D2/D3 recep- neuron activity in response to chronic stress or antide- tors,24 which is in line with observations showing that pressant drug treatments may determine opposite par-

stimulation of the D3 receptor potentiates D1 receptor- allel variations in D3 receptor expression in the shell 42 mediated responses. Finally, the role of D3 receptor of nucleus accumbens. Recently, we provided evidence overexpression in one form of behavioral sensitiz- indicating that brain-derived neurotrophic factor ation29 is consistent with the idea that a similar process (BDNF), a neurotrophin synthesized by dopamine neu- participates in the behavioral changes triggered by anti- rons,59 anterogradely transported and released by neu- 60 depressant treatments. rons upon depolarization, is the factor controlling D3 Importantly, our data also show a progressive and receptor expression.61 In agreement with this conten- 62 strong down-regulation of D3 receptor binding by tion, BDNF expression is reduced by stress and elev- repeated handling and injections, even though no drug ated by chronic antidepressant treatments.63

was administered. This change is not related to In conclusion, our study suggests that D3 receptor increasing age, which actually enhances D3 receptor expression and function are down-regulated in stress expression43 and might be interpreted as a stress- and, possibly, depression, and that these changes are induced effect. Actually the action of ﬂuoxetine and reversed by antidepressant treatments. As a corollary, 42 amitriptyline were not strictly to enhance D3 receptor selective D3 receptor agonists may represent a new binding in the shell of nucleus accumbens, but rather class of antidepressant drugs, a proposal supported by

Molecular Psychiatry Increase of dopamine D3 receptor expression C-H Lammers et al 387 activation of D1-like and D2-like receptors. Psychopharmacology data showing that pramipexole, a preferential D3 receptor agonist,64,65 displays antidepressant-like effects in 1997; 133: 77–84. 66,67 25 Sokoloff P, Giros B, Martres M-P, Bouthenet M-L, Schwartz J-C. animals and is efﬁcacious as an antidepressant in Molecular cloning and characterization of a novel dopamine recep- 68,69 humans. tor (D3) as a target for neuroleptics. Nature 1990; 347: 146–151. 26 Le´vesque D, Diaz J, Pilon C, Martres M-P, Giros B, Souil E et al.

Identiﬁcation, characterization and localization of the dopamine D3 receptor in rat brain using 7-[3H]-hydroxy-N, N di-n-propyl-2-amin- References otetralin. Proc Natl Acad Sci USA 1992; 89: 8155–8159. 1 Koob GF. Drugs of abuse: anatomy, pharmacology and function of 27 Heimer L, Alheid GF, de Olmos JS, Groenewegen HJ, Haber SN, reward pathways. Trends Pharmacol Sci 1992; 13: 177–184. Harlan RE et al. The accumbens: beyond the core-shell dichotomy. 2 Salamone JD. The involvement of nucleus accumbens dopamine J Neuropsychiatry Clin Neurosci 1997; 9: 354–381. in appetitive and aversive motivation. Behav Brain Res 1994; 61: 28 Maj J, Dziedzicka-Wasylewska M, Rogoz E, Rogoz Z. Effect of anti- 117–133. depressant drugs administered repeatedly on the dopamine D3 3 Kapur S, Mann J. Role of the dopaminergic system in depression. receptors in the rat brain. Eur J Pharmacol 1998; 351: 31–37. Biol Psychiatry 1992; 32: 1–17. 29 Bordet R, Ridray S, Carboni S, Diaz J, Sokoloff P, Schwartz J-C. 4 Brown AS, Gershon S. Dopamine and depression. J Neural Transm Induction of dopamine D3 receptor expression as a mechanism of 1993; 91: 75–109. behavioral sensitization to levodopa. Proc Natl Acad Sci USA 1997; 5 Fibiger HC. Neurobiology of depression: focus on dopamine. Adv 94: 3363–3367. Biochem Psychopharmacol 1995; 49: 1–17. 30 Ridray S, Griffon N, Souil E, Mignon V, Carboni S, Diaz J et al. 6 Willner P. The mesolimbic dopamine system as a target for rapid Coexpression of dopamine D1 and D3 receptors in rat ventral stria- antidepressant action. Int Clin Psychopharmacol 1997; 12 (Suppl tum: opposite and synergistic functional interactions. Eur J Neuro- 3): S7–S14. sci 1998; 10: 1676–1686. 7 Silverstone T. Dopamine in manic depressive illness. A pharmaco- 31 Porsolt RD, Anton G, Blavet N, Jalfre M. Behavioral despair in rats: logical synthesis. J Affect Dis 1985; 8: 225–231. a new model sensitive to antidepressant treatments. Eur J Pharma- 8 Gessa GL, Pani L, Serra G, Fratta W. Animal models of mania. Adv col 1978; 47: 379–391. Biochem Psychopharmacol 1995; 49: 43–66. 32 Steru L, Chermat R, Thierry B, Simon P. The tail suspension test: 9 Markou A, Koob GF. Postcocaine anhedonia. An animal model of a new method for screening antidepressants in mice. Psychopharm- cocaine withdrawal. Neuropsychopharmacology 1991; 4: 17–26. acology (Berl) 1985; 85: 367–370. 10 Belmaker RH, Wald D. Haloperidol in normals. Br J Psychiatry 33 Dziedzicka-Wasylewska M, Rogoz R, Klimek V, Maj J. Repeated 1977; 131: 222–223. administration of antidepressant drugs affects the levels of mRNA 11 D’Haenen HA, Bossuyt A. Dopamine D2 receptors in depression coding for D1 and D2 dopamine receptors in the rat brain. J Neural measured with single photon emission computed tomography. Biol Transm 1997; 104: 515–524. Psychiatry 1994; 35: 128–132. 34 Dziedzicka-Wasylewska M, Willner P, Papp M. Changes in dopa- 12 Shah PJ, Ogilvie AD, Goodwin GM, Ebmeier KP. Clinical and mine receptor mRNA expression following chronic mild stress and psychometric correlates of dopamine D2 binding in depression. chronic antidepressant treatment. Behav Pharmacol 1997; 8: 607– Psychol Med 1997; 27: 1247–1256. 618. 13 Zung WW. Review of placebo-controlled trials with bupropion. J 35 Ainsworth K, Smith SE, Zetterstrom TS, Franklin M, Sharp T. Clin Psychiatry 1983; 44: 104–114. Effects of antidepressant drugs on dopamine D1 and D2 receptor 14 Kinney JL. Nomifensine maleate: a new second-generation antide- expression and dopamine release in the nucleus accumbens of the pressant. Clin Pharmacol 1985; 4: 625–636. rat. Psychopharmacology (Berl) 1998; 140: 470–477. 15 Garattini S. Pharmacology of amineptine, an antidepressant agent 36 Le Moine C, Bloch B. Expression of the D3 dopamine receptor in acting on the dopaminergic system: a review. Int Clin Psychophar- peptidergic neurons of the nucleus accumbens: comparison with macol 1997; 12 Suppl. 3: S15–S19. the D1 and D2 dopamine receptors. Neuroscience 1996; 73: 131– 16 Maj J, Rogoz Z, Skuza G, Sowinska H. Repeated treatment with 143. + antidepressant drugs potentiates the locomotor response to ( )- 37 Walker PD, Riley LA, Hart HP, Jonakait GM. Serotonin regulation amphetamine. J Pharm Pharmacol 1984; 36: 127–130. of tachykinin biosynthesis in the rat neostriatum. Brain Res 1991; 17 Chiodo LA, Bunney BS. Typical and atypical neuroleptics: differ- 546: 33–39. ential effects of chronic administration on the activity of A9 and 38 Shirayama Y, Mitsushio H, Takashima M, Ichikawa H, Takahashi A10 midbrain dopaminergic neurons. J Neurosci 1983; 3: 1607– K. Reduction of substance P after chronic antidepressants treatment 1619. in the striatum, substantia nigra and amygdala of the rat. Brain Res 18 White FJ, Wang RY. Differential effects of classical and atypical 1996; 739: 70–78. antipsychotic drugs on A9 and A10 dopamine neurons. Science 39 Kramer MS, Cutler N, Feighner J, Shrivastava R, Carman J, Sramek 1983; 221: 1054–1057. JJ et al. Distinct mechanism for antidepressant activity by blockade 19 Papp M, Klimek V, Willner P. Parallel changes in dopamine D2 receptor binding in limbic forebrain associated with chronic mild of central substance P receptors. Science 1998; 281: 1640–1645. stress-induced anhedonia and its reversal by imipramine. Psycho- 40 Maj J, Papp M, Skuza G, Bigajska K, Zazula M. The inﬂuence of + − pharmacology 1994; 115: 441–446. repeated treatment with imipramine, ( )- and ( )-oxaprotiline on 20 Ichikawa J, Meltzer HY. Effect of antidepressants on striatal and behavioural effects of dopamine D-1 and D-2 agonists. J Neural accumbens extracellular dopamine levels. Eur J Pharmacol 1995; Trans 1989; 76: 29–38. 281: 255–261. 41 Serra G, Collu M, D’Aquila PS, De Montis MG, Gessa GL. Possible 21 Stewart J, Rajabi H. Initial increases in extracellular dopamine in role of dopamine D1 receptor in the behavioural supersensitivity the ventral tegmental area provide a mechanism for the develop- to dopamine agonists induced by chronic treatment with antide- ment of desimipramine-induced sensitization within the midbrain pressants. Brain Res 1990; 527: 234–243. dopamine system. Synapse 1996; 23: 258–264. 42 Pilla M, Perachon S, Sautel F, Garrido F, Mann A, Wermuth CG et 22 Barkai Al, Durkin M, Nelson HD. Localized alterations of dopamine al. Selection inhibition of cocaine-seeking behaviour by a partial receptor binding in rat brain by repeated electroconvulsive shock: dopamine D3 receptor agonist. Nature 1999; 400: 371–375. an autoradiographic study. Brain Res 1990; 529: 208–213. 43 Gurevich EV, Himes JW, Joyce JN. Developmental regulation of 23 Nomikos GG, Zis AP, Damsma G, Fibiger HC. Electroconvulsive expression of the D3 dopamine receptor in rat nucleus accumbens shock produces large increases in interstitial concentrations of and islands of Calleja. J Pharmacol Exp Ther 1999; 289: 587–598. dopamine in the rat striatum: an in vivo microdialysis study. Neur- 44 Willner P, Muscat R, Papp M. Chronic mild stress-induced opsychopharmacology 1991; 4: 65–69. anhedonia: a realistic animal model of depression. Neurosci & 24 Smith SE, Sharp T. Evidence that the enhancement of dopamine Biobehav Rev 1992; 16: 525–534. function by repeated electroconvulsive shock requires concomitant 45 Anisman H, Zacharko RM. Depression as a consequence of inad-

Molecular Psychiatry Increase of dopamine D3 receptor expression C-H Lammers et al 388 equate neurochemical adaptation in response to stressors. Br J Psy- 58 Le´vesque D, Martres M-P, Diaz J, Griffon N, Lammers CH, Sokoloff

chiatry 1992; 15: 36–43. P et al. A paradoxical regulation of the dopamine D3 receptor 46 Risch SC. Recent advances in depression research: from stress to expression suggests that involvement of an anterograde factor from molecular biology and brain imaging. J Clin Psychiatry 1997; 58 dopamine neurons. Proc Natl Acad Sci USA 1995; 92: 1719–1723. Suppl 5: 3–6. 59 Seroogy KB, Lundgren KH, Tran T, Guthrie KM, Isackson PJ, Gall 47 Agid O, Shapira B, Zislin J, Ritsner M, Hanin B, Murad H et al. CM. Dopaminergic neurons in rat ventral midbrain express brain- Environment and vulnerability to major psychiatric illness: a case derived neurotrophic factor and neurotrophin-3 mRNAs. J Comp control study of early parental loss in major depression, bipolar Neurol 1994; 342: 321–334. disorder and schizophrenia. Mol Psychiatry 1999; 4: 163–172. 60 Altar CA, Di Stefano PS. Anterograde transport of neurotrophins 48 Harkness KL, Monroe SM, Simons AD, Thase M. The generation in the nervous systems. Trends Neurosci 1998; 21: 433–437. of life events in recurrent and non-recurrent depression. Psychol 61 Guillin O, Damier L, Griffon N, Diaz J, Carroll P, Schwartz J-C et

Med 1999; 29: 134–144. al. Role of brain-derived neurotrophic factor in the control of D3 49 Thierry A-M, Tassin J-P, Blanc G, Glowinski J. Selective activation receptor expression. Eur J Neuropsychopharmacol 1999; 9: S184– of mesocortical DA systems by stress. Nature 1976; 263: 242–244. S185. 50 Finlay JM, Zigmond MJ. The effects of stress on central dopami- 62 Smith MA, Makino S, Kvenansky R, Post RM. Stress and glucocort- nergic neurons: possible clinical implications. Neurochem Res icoids affect the expression of brain-derived neurotrophic factor 1997; 22: 1387–1394. and neurotrophin-3 mRNAs in the hippocampus. J Neurosci 1995; 51 Gambarana C, Masi F, Tagliamonte A, Scheggi S, Ghiglieri O, De 15: 1768–1777. Montis GM. A chronic stress that impairs reactivity in rats also 63 Nibuya M, Morinobu S, Duman RS. Regulation of BDNF and TrkB decreases dopaminergic transmission in the nucleus accumbens: a mRNA in rat brain by chronic electroconvulsive seizure and antide- microdialysis study. J Neurochem 1999; 72: 2039–2046. pressant drug treatments. J Neurosci 1995; 15: 7539–7547. 52 Chiodo LA, Antelman SM. Repeated tricyclics induce a progressive 64 Sautel F, Griffon N, Le´vesque D, Pilon C, Schwartz JC, Sokoloff P.

dopamine autoreceptor subsensitivity. Nature 1980; 287: 451–454. A functional test identiﬁes dopamine agonists selective for D3 ver-

53 Chiodo LA, Antelman SM. Electroconvulsive shock: progressive sus D2 receptors. NeuroReport 1995; 6: 329–332. dopamine autoreceptor subsensitivity independent of repeated 65 Mierau J, Schneider FJ, Ensinger HA, Chio CL, Lajiness ME, Huff treatment. Science 1980; 210: 799–801. RM. Pramipexole binding and activation of cloned and expressed

54 Ainsworth K, Smith SE, Zetterstrom TS, Pei Q, Franklin M, Sharp dopamine D2,D3,D4 receptors. Eur J Pharmacol 1995; 290: 29–36.

T. Effects of antidepressant drugs on dopamine D1 and D2 receptor 66 Maj J, Rogoz Z, Skuza G, Kolodziejczyk K. Antidepressant effects expression and dopamine release in the nucleus accumbens. Psych- of pramipexole, a novel dopamine receptor agonist. J Neural opharmacology 1998; 140: 470–477. Transm 1997; 104: 525–533. 55 Brown EE, Nomikos GG, Wilson C, Fibiger HC. Chronic desimipra- 67 Willner P, Lappas S, Cheeta S, Muscat R. Reversal of stress-induced mine enhances the effect of locally applied amphetamine on inter- anhedonia by the dopamine receptor agonist, pramipexole. Psycho- stitial concentrations of dopamine in the nucleus accumbens. Eur pharmacol (Berl) 1994; 115: 454–462. J Pharmacol 1991; 202: 125–127. 68 Goldberg JF, Frye MA, Dunn RT. Pramipexole in refractory bipolar 56 Prisco S, Esposito E. Differential effects of acute and chronic ﬂuox- depression. Am J Psychiatry 1999; 156: 798. etine administration on the spontaneous activity of dopaminergic 69 Szegedi A, Wetzel J, Hillert A, Kleiser E, Gaebel W, Benkert O. neurones in the ventral tegmental area. Br J Pharmacol 1995; 116: Pramipexole, a novel selective dopamine agonist in major 1923–1931. depression. Mov Disord 1996; 11 (Suppl.1): 266. 57 Grenhoff J, Nisell M, Ferre S, Aston-jones G, Svensson TH. Norad- 70 Paxinos G, Watson C. The Rat Brain in Stereotaxic Coordinates, renergic modulation of midbrain dopamine cell ﬁring elicited by 2nd edn. Academic Press: New York, 1986. stimulation of the locus coeruleus in the rat. J Neural Transm 1993; 93: 11–25.

Molecular Psychiatry