Ž .Brain Research 771 1997 163–166

Short communication

Electron-microscopic study of MAOB-containing structures in the nucleusaccumbens shell: using MAOA-deficient transgenic mice

Keiko Ikemoto a,) , Kunio Kitahama a, Toshihiro Maeda b, Yoshimitsu Tokunaga b,Jean-Louis Valatx a, Edward De Maeyer c, Isabelle Seif c

a Departement de Medecine Experimentale, INSERM U52, CNRS ERS 5645, UniÕersite Claude Bernard, 8 aÕenue Rockefeller, 69373 Lyon, France´ ´ ´ ´b Department of Anatomy, Shiga UniÕersity of Medical Science, Otsu, Japan

c CNRS, UMR 177, Institut Curie, Orsay, France

Accepted 15 July 1997

Abstract

MAOB-containing structures in the nucleus accumbens were ultrastructurally studied for the first time, using MAOA-deficienttransgenic mice and MAO enzyme histochemistry. Among the striatal structures, the nucleus accumbens, and in particular its dorsal shell,showed the strongest MAOB activity. MAOB-active cell bodies were embedded in a dense MAOB-active fiber plexus. MAOB-positiveterminals formed axo-dendritic synapses which were exclusively of the asymmetric type. It is suggested that dopamine in the nucleusaccumbens shell is transported into MAOB-positive fibers where it is degraded by MAOB. q 1997 Elsevier Science B.V.

Keywords: MAOB; Enzyme histochemistry; Transgenic mouse; Ultrastructure; Nucleus accumbens shell; Dopamine

Ž .The degradative enzyme monoamine oxidase MAOmaintains low cytosolic concentrations of monoaminessuch as serotonin, noradrenaline, and dopamine. There aretwo MAO subtypes, MAOA and MAOB. In vitro experi-ments with brain extracts have shown that MAOA andMAOB have different but overlapping substrate specifici-ties. Dopamine and tyramine are good substrates for bothtypes of MAO. Serotonin is a much better substrate for

w xMAOA than for MAOB 13 . In vivo, the substrate speci-ficities of the MAO subtypes might be regionally modu-lated, and thus, the relevance of the in vitro results is notclear. The highest level of MAOA is localized in noradren-

Ž . w xergic cells such as the locus coeruleus LC 6,20 . Inw xcontrast, MAOB is abundant in serotonergic 6,8 and

w x w xhistaminergic neurons 9 , and in astrocytes 8 .Ž .Recently, we found that the nucleus accumbens NAC

presented the strongest MAOB enzymatic activity in thestriatum, using MAOA-deficient transgenic mice of the

w xTg8 line 2,4 . In this mutant line, MAO activity is notobserved in LC cells, though it is obvious in the dorsal

) Corresponding author. Department of Anatomy, Fujita Health Uni-versity, School of Medicine, Toyoake, Aichi 470-11, Japan. Fax: q81Ž .562 93-2649; E-mail: kikemoto@fujita-hu.ac.jp

Ž . w xraphe DR cells 4 . Interestingly, the NAC has beenshown to be involved in the pathogenesis of psychosesw x w x7,10 and to be an antipsychotic acting site 12,17 . Toexplore the functions of MAOB in this nucleus, we studiedultrastructurally the MAOB-containing structures in theNAC, using a histochemical technique.

Twelve normal C3H mice and seven mice of the Tg8line of both sexes and 8–12 months old were used. Todistinguish the types of MAO in the NAC of normal C3Hmice, preceding MAO histochemistry, we resorted to phar-macological suppression either by the specific MAOB

w y4 y9inhibitor deprenyl 10 –10 M in 0.1 M phosphate-Žbuffered saline 0.1 M phosphate bufferq0.9% NaCl, pH

. x7.4 for 30 min , or by the specific MAOA inhibitorŽ y4 y9clorgyline 10 –10 M in 0.1 M phosphate-buffered

. w xsaline for 30 min 6 . For histochemistry, a modifiedw xcoupled peroxidation method was applied 1 . Mice were

Ždeeply anesthetized with sodium pentobarbital Nembutal,.Dainippon Pharm., 10 mgrkg, i.v. and perfused through

the cardiac ventricle with 30 ml of 0.01 M phosphate-buffered saline, and subsequently with 100 ml of ice-coldfixative containing 2% paraformaldehyde and 2% glu-

Ž .taraldehyde in 0.1 M phosphate buffer pH 7.4 at a rate of20 mlrmin. The brains were immediately cut into 50–60

Ž .mm sections with a vibratome Lancer, 1000 and free-

0006-8993r97r$17.00 q 1997 Elsevier Science B.V. All rights reserved.Ž .PII S0006-8993 97 00909-8

( )K. Ikemoto et al.rBrain Research 771 1997 163–166164

floating sections were incubated overnight with reactionmedium under gentle agitation at 48C. The medium con-sisted of 7.5 mg tyramine hydrochloride, 2 mg 3,3X-di-

Ž .aminobenzidine tetrahydrochloride Sigma , 5 mgŽ .horseradish peroxidase Toyobo, grade III, Japan , 30 mg

nickel ammonium sulfate, and 6.5 mg sodium azide in 10Ž .ml of 0.05 M Tris-HCl buffer pH 7.6 . The reaction was

terminated by placing the sections in 0.05 M Tris-HClbuffer. The stained sections were mounted on gelatin-coated glass slides, dehydrated, cleared, and coverslippedwith Entellan.

The sections for electron microscopy were postfixed in1% OsO , dehydrated, and flat-embedded in Luveak-8124

between silicon-coated slide glass and coverslip. Ultrathinsections from the medial NAC were studied extensively

w xusing an H-7100 electron microscope 11 .Following deprenyl treatment of C3H brain sections at

the concentration of 10y6 M, MAO activity was hardlyŽ .observed in the NAC Fig. 1A . MAO activity in LC cells

and fibers was clear and that in DR cells was extinguished.After clorgyline treatment at the concentration of 10y6 M,a dense MAO-positive fiber plexus and cells were ob-

Žserved in the NAC, in particular in the medial NAC Fig..1B . MAO reactivity was less in the dorsal striatum. At

this concentration of clorgyline, MAO activity in LC cellswas not discriminated, and that in DR cells was clear.These studies confirm that MAO in the NAC of normal

w xmice is mostly type B 14 .In the striatum of Tg8 mice, the medial NAC was most

Ž .strongly MAOB active Fig. 2A . At the light microscopiclevel, many MAOB-positive neurons and fibers were ob-served throughout the NAC. More MAOB-positive fibers

Ž .were condensed in the shell than in the core Fig. 2A . We

noted an oval-shaped area with an extremely dense accu-mulation of MAOB-positive fibers in the dorsal part of the

Ž .shell the area surrounded by arrows . In this fiber accumu-lation, there were many MAOB-positive neuronal cell

Ž .bodies Fig. 2B . They were ovoid and small or medium inŽ .size 10–20 mm in diameter .

At the electron-microscopic level, MAOB-positive neu-ronal cell bodies, dendrites and terminals were observed.The reaction product was on the membranes of synapticvesicles and filled the axoplasm of labeled axon terminals.As shown in Fig. 3A, in MAOB-positive terminals, the

Ž .mitochondrial outer membrane arrows presented strongMAOB activity. The MAOB-positive terminals showedaxo-dendritic contacts with dendritic shafts as well asspines, and axo-somatic contacts. The MAO-positive ter-minals formed asymmetric synapses without exceptionŽ .Fig. 3A and B .

This study demonstrates that the dorsal shell of theNAC possesses the most dense MAOB fiber plexus in theTg8 striatum. Most of these fibers may not be serotonergicor histaminergic fibers because both fibers are not so much

w xconcentrated in this nucleus 18 . It is probable that theyoriginate from intrinsic neurons. The existence of MAOB-positive cells in the mouse NAC has previously been

w xreported by Nakamura et al. 15 .The nucleus is heavily innervated by midbrain

dopaminergic neurons; corresponding to the homologousregion of the MAOB-rich area in the rodent shell, there is

w xa dense dopamine fiber plexus 19 . Dopamine may betransported into MAOB-positive fibers and be degraded.This mechanism may act also on trace amines. It should beemphasized that MAOB terminals form asymmetricsynapses without exception in the NAC.

Ž .Fig. 1. Monoamine oxidase staining in coronal sections through the nucleus accumbens NAC in normal C3H mouse. A: after pretreatment with theŽ y6 .specific MAOB inhibitor deprenyl 10 M , MAO activity is hardly observed in the NAC. B: after pretreatment with the specific MAOA inhibitor

Ž y6 .clorgyline 10 M , a dense MAO-positive fiber plexus was observed in the NAC. Abbreviations: ac, anterior commissure; CO, core of the NAC; S,septum; SH, shell of the NAC; Tu, olfactory tubercle. Barss500 mm.

( )K. Ikemoto et al.rBrain Research 771 1997 163–166 165

Fig. 2. A: monoamine oxidase staining in the coronal section through the NAC in MAOA-deficient Tg8 mouse. Note that there is an oval-shaped area withŽ .especially strong MAOB activity in the shell of the NAC area surrounded by arrows . B: high-power photomicrograph of the shell of the NAC. Many

Ž . Ž .MAOB-positive cells arrows are embedded in the dense MAOB fiber plexus. The cells are ovoid and small to medium sized 10–20 mm . Abbreviations:Ž . Ž .ICj, major island of Calleja; CP, caudate-putamen; LS, lateral septum; lv, lateral ventricle. Bars in A s500 mm; in B s100 mm.

Fig. 3. A: MAOB-positive terminal in the NAC shell. The reactionproduct is on the membranes of synaptic vesicles and filled the axoplasmof labeled axon terminals. The outer membrane of the mitochondria

Ž .shows strong MAO activity thin arrows . The MAOB-positive terminalŽ . Ž .forms an asymmetric synapse thick arrow with a dendritic shaft D . B:

Ž .MAOB-positive terminal forming asymmetric synapses arrows with aŽ .dendritic shaft D . Barss0.25 mm.

In fact, the rat NAC shell has been shown to be anw xactive site of many antipsychotics 3 . The monkey NAC

medial subdivision, a putative homologous area of the ratdorsal shell, shows marked c-fos expression after adminis-

w xtration of the D blocker haloperidol 5 . The abundance of2

MAOB cells and fibers in the mouse NAC shell leads us toconsider this area also as a potentially critical target for

w xdepression therapy with MAOB inhibitors 16 .

Acknowledgements

This study was supported by INSERM U52, CNRSŽERS5645, Institut Curie, and CIES Centre International

.des Etudiants et Stagiaires .

References

w x1 R. Arai, H. Kimura, T. Maeda, Topographic atlas of monoamineoxidase-containing neurons in the rat brain studied by an improved

Ž .histochemical method, Neuroscience 19 1986 905–925.

( )K. Ikemoto et al.rBrain Research 771 1997 163–166166

w x2 O. Cases, I. Seif, J. Grimsby, P. Gaspar, K. Chen, S. Pournin, U.Muller, M. Aguet, C. Babinet, J.C. Shih, E. De Maeyer, Aggressive¨behavior and altered amounts of brain serotonin and norepinephrine

Ž .in mice lacking MAOA, Science 268 1995 1763–1766.w x3 A.Y. Deutsch, A.J. Bourdelais, D.S. Zahm, The nucleus accumbens

core and shell: accumbal compartments and their functional at-Ž .tributes, in: P.W. Kalivas, C.D. Barnes Eds. , Limbic Motor Cir-

cuits and Neuropsychiatry, CRC Press, Broca Raton, 1993, pp.45–88.

w x4 K. Ikemoto, K. Kitahama, I. Seif, T. Maeda, E. De Maeyer, J.L.Valatx, MAOB-containing structures in MAOA-deficient transgenic

Ž .mice, Brain Res. in press .w x5 K. Ikemoto, K. Satoh, T. Maeda, H.C. Fibiger, Neurochemical

heterogeneity of the primate nucleus accumbens, Exp. Brain Res.Ž .104 1995 177–190.

w x6 K. Kitahama, R. Arai, T. Maeda, M. Jouvet, Demonstration ofmonoamine oxidase type B in serotonergic and type A in noradren-ergic neurons in the cat dorsal pontine by an improved histochemical

Ž .technique, Neurosci. Lett. 71 1986 19–24.w x7 T. Lee, P. Seeman, Elevation of brain neurolepticrdopamine recep-

Ž .tors in schizophrenia, Am. J. Psychiatry 137 1980 191–197.w x8 P. Levitt, J.E. Pintar, X.O. Breakefield, Immunohistochemical

demonstration of monoamine oxidase B in brain astrocytes andŽ .serotonergic neurons, Proc. Natl. Acad. Sci. USA 79 1982 6385–

6389.w x9 J.S. Lin, K. Kitahama, P. Fort, P. Panula, R.M. Denney, M. Jouvet,

Histaminergic system in the cat hypothalamus with reference to typeŽ .B monoamine oxidase, J. Comp. Neurol. 330 1993 405–420.

w x10 A.V.P. Mackay, L.L. Iversen, M. Rossor, E. Spokes, E. Bird, A.Arregui, I. Creese, S.H. Snyder, Increased brain dopamine anddopamine receptors in schizophrenia, Arch. Gen. Psychiatry 39Ž .1982 991–997.

w x11 T. Maeda, H. Imai, R. Arai, H. Tago, T. Nagai, T. Sakumoto, K.Kitahama, B. Onteniente, H. Kimura, An improved peroxidaticoxidation method of MAO histochemistry for neuroanatomical re-search at light and electron microscope levels, Cell Mol. Biol. 33Ž .1987 1–11.

w x12 K.M. Merchant, P.R. Dobner, D.M. Dorsa, Differential effects ofhaloperidol and clozapine on neurotensin gene transcription in rat

Ž .neostriatum, J. Neurosci. 12 1992 652–663.w x13 D.L. Murphy, Substrate-selective monoamine oxidases, Biochem.

Ž .Pharmacol. 27 1996 1889–1893.w x14 S. Nakamura, I. Akiguchi, J. Kimura, A subpopulation of mouse

striatal cholinergic neurons show monoamine oxidase activity, Neu-Ž .rosci. Lett. 161 1993 41–144.

w x15 S. Nakamura, I. Akiguchi, J. Kimura, Topographic distributions ofmonoamine oxidase-B-containing neurons in the mouse striatum,

Ž .Neurosci. Lett. 184 1995 29–31.w x16 A. Nies, D.S. Robinson, Monoamine oxidase inhibitors, in: E.S.

Ž .Paykel Ed. , Handbook of Affective Disorders, Churchill Living-stone, Edinburgh, 1982, pp. 246–261.

w x17 G.S. Robertson, H.C. Fibiger, Neuroleptics increase c-fos expressionin the forebrain: contrasting effects of haloperidol and clozapine,

Ž .Neuroscience 46 1992 315–328.w x18 E.J. Van Bockstaele, V.M. Pickel, Ultrastructure of serotonin-

immunoreactive terminals in the core and shell of the rat nucleusaccumbens: cellular substrates for interactions with catecholamine

Ž .afferents, J. Comp. Neurol. 334 1993 603–617.w x19 P. Voorn, B.J. Byham, C. Van Disk, R.M. Buijs, The dopamine

innervation of the rat: a light- and electron-microscopical study withŽ .antibodies against dopamine, J. Comp. Neurol. 251 1986 84–99.

w x20 K.N. Westlund, R.M. Denney, L.M. Kochersperger, R.M. Rose,C.W. Abell, Distinct monoamine oxidase A and B populations in

Ž .primate brain, Science 230 1985 181–183.