hpbmeliaptimcwrb

h

NeuroImage 14, 87–94 (2001)doi:10.1006/nimg.2001.0793, available online at http://www.idealibrary.com on

Measurement of the Striatal Dopamine Transporter Density andHeterogeneity in Type 1 Alcoholics Using Human

Whole Hemisphere AutoradiographyErkki Tupala,*,† Jyrki T. Kuikka,‡ Håkan Hall,§ Kim Bergstrom,‡ Terttu Sarkioja,¶ Pirkko Rasanen,\

Tuija Mantere,*,† Jukka Hiltunen,** Jouko Vepsalainen,†† and Jari Tiihonen*,‡*Department of Forensic Psychiatry, Niuvanniemi Hospital, †Department of Pharmacology and Toxicology, and ††Department of

Chemistry, University of Kuopio, Finland; ‡Department of Clinical Physiology and Nuclear Medicine, Kuopio University Hospital,Finland; §Department of Clinical Neuroscience, Psychiatry Section, Karolinska Institute and Hospital, Stockholm, Sweden;

¶Department of Forensic Medicine and \Department of Psychiatry, University of Oulu, Finland; and**MAP-Medical Technologies Ltd., Helsinki, Finland

Received August 28, 2000; published online May 11, 2001

(spcd

Dopaminergic mechanisms are involved in the pos-itive reinforcing and addicting effects of alcohol.Positron emission tomography (PET) and single pho-ton emission tomography (SPET) studies have indi-cated alterations in striatal dopamine transporters(DAT) and in presynaptic dopamine (DA) function inalcoholics, although also contradictory results havebeen reported. Normal variations in blood flow, me-tabolism, and receptor densities are apparently im-portant to brain function. Such variations are knownto decrease during pathophysiological processes, suchas epilepsy, whereas normal receptor distributionsare broadly heterogenous. We evaluated the densitiesand heterogeneities of striatal DAT in 8 adult-onset,Cloninger type I alcoholics and 10 controls using[125I]N-(3-iodoprop-2E-enyl)-2b-carbomethoxy-3b-(4*-methylphenyl)nortropane ([125I]PE2I) as a ligand for

uman postmortem whole hemisphere autoradiogra-hy, which provided high resolution images of therain when compared with in vivo PET and SPET. Theean density and heterogeneity of DAT were mark-

dly lower in the alcoholics. A significant linear corre-ation existed between DAT density and heterogene-ty, as well as between DAT densities in the nucleusccumbens and in the dorsal striatum (caudate andutamen) in alcoholics, but not consistently in con-rols. The observed low DAT density and heterogene-ty in the dorsal striatum suggest that type 1 alcoholics

ay have a dysfunctional DA system. These data indi-ate that human whole hemisphere autoradiographyith the analysis of binding heterogeneity may be a

elevant tool to measure pathological processes in therain. © 2001 Academic Press

Key Words: dopamine; alcoholism; autoradiography;uman brain; transporters; binding heterogeneity.

87

INTRODUCTION

A neurobiological explanation for the positive rein-forcing effects of most drugs of abuse has focused onmesocorticolimbic dopamine (DA) system and its con-nections in the basal forebrain (Koob, 1992; Koob andLe Moal, 1997; Robbins and Everitt, 1999). It has beenshown that alcohol dependence is associated with theDA function in the striatum and in the limbic system(Yoshimoto et al., 1991; Kalivas, 1993; Dyr et al., 1993),and that addictive drugs, such as cocaine and amphet-amine, exert their rewarding effects through their in-fluence on dopamine transporters (Rossetti et al., 1991;Sorensen et al., 1992). In 1974 Seeman and Lee showedthat high brain ethanol concentrations could releaseDA in the striatum, although low to moderate concen-trations were found to have no effect. Lower concentra-tions of DA and its metabolites have been found in thenucleus accumbens (Nacc) of ethanol preferring P(Murphy et al., 1987) and HAD (high-alcohol-drinking)rats (Gongwer et al., 1989) when compared to their NP(alcohol nonpreferring) and LAD (low alcohol-drinking)counterparts. Therefore, it follows that much effort hasbeen devoted to determining whether or not DA alsounderlies ethanol preference (Nevo and Hamon, 1995).

The neuronal dopamine transporter (DAT) is a pre-synaptically located protein responsible for the re-uptake, and thus removal, of dopamine from the syn-aptic cleft, and it is almost exclusively located in thebasal ganglia of the human brain (Boja et al., 1994;Hall et al., 1999). Both positron emission tomographyPET) and single photon emission tomography (SPET)tudies have indicated alterations in striatal DAT andresynaptic dopamine function in Cloninger type 1 al-oholics (characterized by late onset, social depen-ency, and anxiety), whereas type 2 alcoholics (charac-

1053-8119/01 $35.00Copyright © 2001 by Academic Press

All rights of reproduction in any form reserved.

1baahta

aageei11i

C

wegot

tcsofpcfa

88 TUPALA ET AL.

terized by early onset, impulsive and novelty seekingbehavior, and social hostility) seem to have a seroto-nergic deficit with normal or perhaps slightly increasedDAT density in the striatum (Cloninger, 1995; Ti-ihonen et al., 1995, 1998). In addition, we have recentlyreported low DAT densities in the Nacc of type 1 alco-holics (Tupala et al., 2000). However, also contradic-tory results have been reported from an unselectedalcoholic population regarding striatal DAT (Volkow etal., 1996). Concerning heterogeneity, one previousSPET study found a higher heterogeneity of striatalDAT binding among type 2 alcoholics when comparedwith controls (Kuikka et al., 1998a).

The normal variations in blood flow, metabolism andreceptor densities are apparently important to brainfunction. Such variations are known to decrease duringpathophysiological processes, such as epilepsy (Gold-berger et al., 1990; Goldberger, 1996; Tiihonen et al.,997), whereas normal receptor distributions areroadly heterogenous, as illustrated by postmortemutoradiographic mapping (Dillon et al., 1991). Theim of this study was to evaluate the densities andeterogeneities of dorsal striatal (caput nucleus cauda-us and putamen) DAT in adult onset type I alcoholicsnd controls using [125I]N-(3-iodoprop-2E-enyl)-2b-car-

bomethoxy-3b-(49-methylphenyl)nortropane ([125I]PE2I)s a ligand for human postmortem whole hemisphereutoradiography (Hall et al., 1998, 1999). This radioli-and has proven to be a selective DAT ligand withxtremely low affinity for other monoamine transport-rs and has been used successfully in both in vitro andn vivo imaging of DAT (Emond et al., 1997; Hall et al.,999; Kuikka et al., 1998b, 1998c; Guilloteau et al.,998; Tupala et al., 2000). We also compared the bind-ng of [125I]PE2I to DAT between the Nacc and dorsal

striatum (caudate and putamen) in both alcoholics andcontrols. The in vitro method provided high resolutionimages of the brain, corresponding to in vivo PET andSPET studies, which enabled a more detailed study ofvarious fine structures (Hall et al., 1996, 1998, 1999;Tupala et al., 2000).

MATERIAL AND METHODS

Brain Sampling

Human brains were obtained during clinical nec-ropsy at the Department of Forensic Medicine, Univer-sity of Oulu (Finland), and this portion of the study wasapproved by the Ethics Committee of the University ofOulu and the National Institute of Medicolegal Affairsin Helsinki (Finland). The brains were removed,cleaned of the dura, and divided at the midsagittalplane. The left hemisphere was placed with the mid-sagittal plane on a glass-plate before freezing (275°C).Medical records on the cause of death, previous dis-eases, and medical treatments were also collected. Di-

agnoses were made by two physicians independently ofeach other. Mental disorders were coded according toDSM-III-R diagnostic criteria and the alcoholics weresubcategorized into type 1 according to Cloninger(1995). Diagnoses regarding type 1 alcoholism and con-trols were unanimous. The subjects were classified ascontrols when necropsy or anamnestic data revealed nosigns of alcohol or other substance abuse.

Alcoholism was diagnosed when alcohol dependenceor harmful use was evident on the basis of autopsy,medical records, and anamnestic data. Subjects havingpsychotic disorders or any diseases (such as epilepsy)or taking medications that affect the central nervoussystem (such as neuroleptics or antidepressants) wereexcluded. All 18 cases were white Caucasians. A post-mortem analysis for drugs, including alcohol, and nor-mal necropsy protocol were also performed. Gross neu-ropathological evaluation for the right hemisphere wasperformed during necropsy and a more detailed neuro-anatomical evaluation was made on the left hemi-sphere sections used in the study from an adjacentNissl stained section. None of the study hemispheresexhibited damage or neuroanatomical abnormalities.

Study Subjects

The study groups consisted of 8 type 1 alcoholics (7males, 1 female; mean age 52.9 years; post mortemdelay 12.0 6 4.8 h; mean 6 SD) and 10 controls (8males, 2 females; mean age 53.5 years; postmortemdelay 14.8 6 9.2 h; mean 6 SD). Seven of the alcoholicshad alcohol in their blood at the time of death, and onealcoholic had an abstinence period of 10 h. One of thecontrols had a small amount of alcohol in his blood atthe time of death (0.036%). Two of the alcoholics hadtraces of diazepam in their blood samples. Blood alco-hol and radioligand binding had no correlation witheach other (r 5 0.43, P 5 0.29 in the putamen; r 520.12, P 5 0.77 in the caudate).

ryosectioning

The object glasses (size 100 3 220 3 1.0–1.2 mm)ere washed in a laboratory dishwasher, dipped inthanol for 15 min, dried, and then dipped briefly inelatine (Hall et al., 1998). The glasses were then driedvernight at room temperature under a cover to protecthem against dust.

For cryosectioning, the glass plate was removed fromhe frozen hemisphere, which was then put in a spe-ially designed rubber box with a specimen holder. Aemiliquid gel (approximate concentration 5 g/l, 15°C)f carboxymethylcellulose (CMC) was added and therozen hemisphere was inserted with the midsagittallane of the hemisphere facing up, while carefullyhecking the horizontal level of the midsagittal sur-ace. Positioning of the hemisphere was achieved bydjusting the hemisphere parallel to a ruler over the

bosd

w[

(edttii(

i

1mns

89DOPAMINE TRANSPORTER DENSITY AND HETEROGENEITY IN ALCOHOLICS

upper margin of the anterior and lower margin of theposterior commissura and the surface of the cryostatspecimen holder (the cutting plane) to obtain parallelsections to the bicommisural line of Talairach (Niewen-huys et al., 1988). When the hemisphere-CMC blockwas totally frozen, the rubber box was removed.

The serial sectioning of the block was performed witha heavy-duty cryomicrotome (LKB 2250, LKB, Stock-holm, Sweden). The temperature of the hemisphere-CMC block was allowed to achieve the cutting temper-ature (215°C optimal) before cryosectioning. The brainwas sectioned into 100-mm horizontal (cantomeatal)sections. A thin tissue paper was put on the section,after which a transparent tape (3M, type 810) wasfastened to the block by gentle rubbing. Tissue sectionscut with the cryotome adhered to the tissue paper, andwere transferred to the gelatinized glass plates using asoft paintbrush. The tape and the paper were carefullyremoved, with the sections thawed onto the glassplates. The sections were allowed to air dry before theywere stored with dehydrating agents at 225°C untiluse.

Autoradiography

[125I]PE2I was obtained from MAP Medical Technol-ogies Oy, (Helsinki, Finland). The radiolabeling wasperformed by using a stannylated precursor with[125I]Nal, as previously described (Hall et al., 1999).

-CIT (2b-carbomethoxy-3b-(4-iodophenyl)tropane) wasbtained from the Department of Chemistry, Univer-ity of Kuopio, and synthesized according to previouslyescribed methods (Swahn et al., 1996).The first series of sections (7 controls, 7 alcoholics)ere incubated in a solution (vol 2.5 liters) containing

125I]PE2I (approximately 20 pM) in Tris–HCl buffer,pH 7.4, 0.05 M. b-CIT (2 mM) was used as a specificdisplacer to determine nonspecific binding for all sec-tions, as previously described (Hall et al., 1999; Tupalaet al., 2000). Washing was performed in cold buffer for3 3 10 min, followed by a brief and final cold wash ofdipping the sections into distilled water. The sectionswere dried for 10 min with a gentle stream of warm airand then for 3 h at room temperature, exposured toradiation-sensitive film (3H-Hyperfilm, Amersham,London, UK) for 5 days development (developer, KodakD19; fixation, Kodak Fixer 3000). The autoradiogramswere analysed by computerized densitometry, using aMikrotek Scanmaker E6 connected to an Osborne PCPentium II. Software packages included Adobe Photo-shop 4.01 (Mountain View, CA) and Scion Image forWindows, Version Beta 3b (Frederick, MD). The result-ing pixel values of the binding data were mathemati-cally transformed by an exponential calibration equa-tion into relative radioactivity values by the use of125I-calibrating scales (Cat. No. RPA 523L, Amersham),after background subtraction. All analyses were made

blind to the clinical classification of the subjects. Anadjacent section from the respective specimen’s levelwas stained with cresyl violet (Nissl staining) to serveas an anatomical correlate to the autoradiography. Thesecond autoradiographic series (three controls and onealcoholic) with [125I]PE2I) was made according to thesame protocol, but only in an incubation volume of 750ml.

To correlate the DAT densities between the Nacc anddorsal striatum, we used the results of our formerstudy regarding DAT data from the Nacc of the samesubjects that were used in the present study (Tupala etal., 2000).

Heterogeneity of Dorsal Striatal DAT Binding

The pixel sizes of the autoradiograms were reorga-nized so that the whole dorsal striatal (caudate andputamen) area consisted of 512 pixels. Mean binding(m) of [125I]PE2I and its standard deviation (SD) werecomputed for the dorsal striatum. The observed dorsalstriatal heterogeneity (RDobserved) was expressed as thecoefficient of variation (RDobserved 5 SD/m). When theaccumulation of a radioactive tracer is used to measurethe heterogeneity of transporter binding, the observeddispersion (RDobserved) is a composite of at least twodispersive processes, the spatial dispersion (RDspatial)and the methodological dispersion (RDmethod)Bassingthwaighte et al., 1989). Assuming that thexponential calibration equation correctly transfersensitometric pixel values into radioactivity values,he methodological dispersion can be estimated fromhe heterogeneity of nonspecific binding. A region ofnterest was drawn from the dorsal prefrontal cortex,ts pixel values and heterogeneity was computedRDmethod). Thus the “true” spatial heterogeneity can be

estimated as RDspatial 5 (RDobserved2 2 RDmethod

2 )1/2, andRDspatial was used as a measure of the dorsal striatalDAT heterogeneity (Bassingthwaighte et al., 1994;Kuikka et al., 1998b).

RESULTS AND DISCUSSION

Binding Study

[125I]PE2I bound with high intensity in the nucleuscaudatus and putamen indicating high densities ofDAT in these structures (Fig. 1). The labeling of DATwith [125I]PE2I in the nucleus caudatus and putamenwas totally (.90%) abolished by the addition of DATnhibitor b-CIT (2 mM) (data not shown). [125I]PE2I also

accumulated in the white matter slightly (Fig. 1), pos-sibly due to its relatively high lipohilicity (Hall et al.,999). This accumulation corresponded to approxi-ately 24% of the specific binding in the putamen and

o differences were found between the groups (data nothown). Addition of b-CIT did not affect this accumu-

sacPni

v

90 TUPALA ET AL.

lation of [125I]PE2I in white matter. No specific bindingwas observed in the cortical areas or globus pallidus(Fig. 1).

The mean binding of [125I]PE2I to DAT in the caputnucleus caudatus was 35% lower in the alcoholic group(380 6 85 Bq/mg; mean 6 SD) when compared with thecontrol group (584 6 94 Bq/mg), and this differencewas statistically significant (P 5 0.0003; the univariateanalysis of covariance (ANCOVA) with the experimentseries (the first or second) as a covariate) (Fig. 2).[125I]PE2I-binding (DAT) in the putamen was reducedby 28% in the alcoholic group (471 6 114 Bq/mg) whencompared with the control-group (658 6 84 Bq/mg; P 50.002; ANCOVA).

Our results of lower DAT densities among alcoholicsare in line with previous reports from in vivo studies onstriatal DAT (Tiihonen et al., 1995; Repo et al., 1999;Laine et al., 1999), although one study (Volkow et al.,1996) did not find a similar reduction in unselected(i.e., type 1 plus type 2) alcoholics. However, only fivesubjects were measured in that study, which couldexplain the negative result. One previous SPET study(Laine et al., 1999) found considerably lower DAT den-sities in the striatum of alcoholics (about one half ofthem had a history of criminal offenses (P. Laine, per-sonal communication), suggesting type 2 alcoholism)after admission for detoxification, and after a 4-weekabstinence the densities achieved levels of the healthycontrols. There is also some evidence suggesting thatthe decreased DAT density in type 1 alcoholics is not

e dorsal striatum of alcoholics and controls (m indicates the meanicate cases with traces of diazepam in the blood.

FIG. 1. Color coded autoradiograms of human postmortem brainections showing the binding of [125I]PE2I to DAT in a control (A) andn alcoholic (B) brain. One section of each brain was stained withresyl violet (Nissl staining) to serve as an anatomical correlate (C).ut, putamen; Nc, caput nucleus caudatus; Gp, globus pallidus. Theumber of the brain and the level of the section from the vertex is

ndicated on the panels.

FIG. 2. Scatter plots showing the binding of [125I]PE2I to DAT in thalue). Circles indicate males and triangles females. Open symbols ind

ha

Cl

fi

CntKSndPrbffp

awHaaont1

sc

91DOPAMINE TRANSPORTER DENSITY AND HETEROGENEITY IN ALCOHOLICS

merely a consequence of recent alcohol abuse but, atleast to some extent, a trait marker for alcoholism(Repo et al., 1999).

Genetic studies on DAT gene polymorphism (VNTR)ave suggested a connection between polymorphismnd alcoholism (Muramatsu et al., 1995; Ueno et al.,

1995) or in the severity of withdrawal symptoms(Sander et al., 1997), although some studies have failedto find this connection (Parsian et al., 1995; Heinz etal., 1995), perhaps depending on the region studiedand possibly the ethnic heterogeneity of the studiedpopulations. Additionally, one study has suggestedthat VNTR polymorphism of the DAT gene (SLC6A3)effects translation of the DAT protein, where a 22%reduction was measured in DAT availability by SPET(Heinz et al., 1995), similar to what was found in thepresent study. Although these genetic studies on poly-morphisms have received criticism and have their pit-falls (Nevo and Hamon, 1995), they provide furtherevidence indicating that our findings may reflect ge-netic or developmental variability related to addictionvulnerability.

Heterogeneity of the Dorsal Striatal DAT Binding

In the alcoholic group the dorsal striatal heterogene-ity of the [125I]PE2I-binding was decreased by 18%(38 6 8% vs 45 6 5%; mean 6 SD, P 5 0.039; AN-

OVA) (Fig. 3). A statistically significant linear corre-ation (r 5 0.82, P 5 0.014; Pearson’s correlation coef-

FIG. 3. Scatter plots showing the heterogeneity of [125I]PE2I btriatum) of alcoholics and controls and the correlation between hetorrelation coefficient). See Fig. 2 legend for symbol explanations.

cient) existed between the [125I]PE2I-binding andheterogeneity in alcoholics, but not in controls (Fig. 3).The concept of fractal analysis originated in 1983 byMandelbrot, and has since been adapted to character-ize a variety of anatomic and physiological phenomena(Bassingthwaighte et al., 1989, 1994; West et al., 1987).

hanges in the brain functions in patients with variouseuropsychiatric disorders are often very small, andypically only a few percent (Schlegel et al., 1994;uikka et al., 1998a, 1998b; Kaschka et al., 1995).mall changes in receptor occupancies between theormal heterogeneity and pathological states are oftenifficult to measure with high accuracy by conventionalET and SPET techniques due to the limited spatialesolution (Tiihonen et al., 1997). Therefore, it has noteen possible to measure differences in receptor uptakerom regions of interest smaller than 7 3 7 mm, andractal analysis might be a useful tool to overcome thisroblem.By contrast, postmortem human whole hemisphere

utoradiography allows the analysis of subregionsithin a millimeter scale or less (Dillon et al., 1994;all et al., 1998), while PET and SPET methodologiesre 1–2 orders coarser for the same ligand (Tiihonen etl., 1997). As the observed heterogeneity is a functionf sample size, and the greatest estimate of heteroge-eity will be obtained from the smallest subregions ofhe organ (Bassingthwaighte et al., 1994; Kuikka et al.,997), the in vitro autoradiographic procedure gives a

ing to DAT in the caput nucleus caudatus and putamen (dorsalgeneity of the binding and total binding in the alcoholics (Pearson’s

indero

aict

92 TUPALA ET AL.

more realistic estimate of the actual heterogeneity. Ithas been suggested that the multi-scaled fractal prop-erty of neuronal functions has obvious adaptive advan-tages, because a system with multiscale variability ismore capable of coping with an inherently unpredict-able environment (Goldberger et al., 1990; Goldberger,1996; Kuikka et al., 1997). The observed low DAT den-sity and heterogeneity in the striatum add evidence tothe assumption that type 1 alcoholics may have a hy-pofunctioning dopamine system, in contrast to type 2alcoholics, who may have overactive and poorly con-trolled dopaminergic transmission (Tiihonen et al.,1995; Kuikka et al., 1998a). It is suggested, herein, thatthese two variables are connected with each other, i.e.,high DAT density is connected with high heterogene-ity, whereas low DAT density is connected with lowheterogeneity.

Correlation of Binding between Nacc and DorsalStriatum

Behavioral genetic studies on rodents indicate thatindividual differences in the number of dopamine re-leasing neurons, within a dopamine cell group, aremaintained across all dopamine cell groups in thebrain (Depue et al., 1994; Farde et al., 1997). This isreflected with a strong correlation observed betweenthe reactivity of hormonal and behavioral indicators indifferent dopamine cell groups. Our finding of de-creased DAT among alcoholics in both Nacc (Tupala et

l., 2000) and dorsal striatum (caudate and putamen)s consistent with this assumption. Moreover, statisti-ally significant positive correlation was observed forhe binding of [125I]PE2I to DAT between Nacc and

dorsal striatum in the alcoholics, but not consistently

FIG. 4. The correlation of the binding of [125I]PE2I to DAT betwcontrols (Pearson’s correlation coefficient). See Fig. 2 legend for sym

in the controls (Figs. 4 and 5). The fact that the bindingof [125I]PE2I to DAT correlated strongly between Naccand dorsal striatum in the alcoholics but not betweenputamen and Nacc in the controls is somewhat puz-zling. However, a similar observation in a SPET studyhas recently been made regarding striatal (striatum-to-cerebellum ratio) and extrastriatal (temporal pole tocerebellum ratio) DA D2 receptors (Repo et al., 1999).Perhaps this could be explained by the overall higherheterogeneity of DAT in the healthy controls, whichalso shows as larger variability of DAT densities be-tween different brain structures.

Conclusions

Further studies are needed to establish the extent ofthis observed low binding and heterogeneity of DAT inalcoholics, and if this relates to altered neurodevelop-ment, or to chronic alcohol exposure. However, ourresults suggest that a selected group of alcoholics (i.e.,type 1) could benefit from long term treatment withdrugs that enhance dopaminergic function, whereasthis may not be the case regarding type 2 alcoholics.Analysis of binding heterogeneity seems to be a rele-vant tool in measuring brain abnormalities, and couldbe especially useful in PET and SPET studies to detectsmaller changes observed in some disorders, such asbenzodiazepine receptor binding in generalized anxietydisorder (Tiihonen et al., 1997).

Furthermore, although also other neurotransmittersystems play an active role in alcoholism, only someneuronal receptors and transporters have been exten-sively studied by both in vitro postmortem whole hemi-sphere autoradiography and in vivo imaging tech-niques (i.e., PET and SPET). Of great importance for

nucleus accumbens and caput nucleus caudatus in alcoholics andexplanations.

eenbol

B

B

C

D

E

H

93DOPAMINE TRANSPORTER DENSITY AND HETEROGENEITY IN ALCOHOLICS

our understanding of psychiatric and neurological dis-eases is to characterize the distribution of receptorsand transporters in normal and affected brains as wellas to characterize the detailed regional pharmacology.Human whole hemisphere autoradiography provides auseful tool in exploring brain regions, which are noteasily defined in vivo and, especially when using 11C-and 18F-labeled compounds, also in early phase of drugor radioligand development.

ACKNOWLEDGMENTS

We thank Pirjo Halonen, M.Sc., for her excellent assistance withthe statistical analyses. Radiolabeling of [125I]PE2I was performedunder the EUREKA Dopimag-program and was financially sup-ported by TEKES. We also thank Kari Karkola, MD, for providingtwo brains for the study. This study was supported by the FinnishCultural Foundation of Northern Savo and the Swedish MedicalResearch Council (11640).

REFERENCES

Bassingwaighte, J. B., King, R. B., and Roger, S. A. 1989. Fractalnature of regional myocardial blood flow heterogeneity. Circ. Res.65: 578–590.asingwaighte, J. B., Liebovitch, L. S., and West, B. J. 1994. FractalPhysiology. Oxford Univ. Press, New York.oja, J. W., Vaughan, R., Patel, A., Shaya, E. K., and Kuhar, M. J.1994. The dopamine transporter. In Dopamine Receptors andTransporters (H. B. Niznik, Ed.), pp. 611–644. Marcel Dekker,New York.loninger, C. R. 1995. The psychobiological regulation of socal co-operation. Nature Med. 1, 623–624.epue, R. A., Luciana, M., Arbisi, P., Collins, P., and Leon, A. 1994.Dopamine and the structure of personality: Relation of agonist-induced dopamine activity to positive emotionality. J. Perspect.Soc. Psychol. 67: 485–498.

FIG. 5. The correlation of the binding of [125I]PE2I to DAT bet(Pearson’s correlation coefficient). See Fig. 2 legend for symbol expla

Dyr, W., McBride, W. J., Lumen, L., Li, T-K., and Murphy, J. M.1993. Effects of D1 and D2-dopamine receptor agents on ethanolconsumption in the high alcohol drinking (HAD) line of rats. Al-cohol 10: 207–212.

Dillon, K. W., Gross-Isseroff, R., Israeli, M., and Biegon, A. 1991.Autoradiographic analysis of serotonin 5-HT1A receptor binding inthe human brain postmortem. Brain Res. 554: 56–64.dmond, P., Garreau, L., Chalon, S., Boazi, M., Caillet, M.,Bricard, J., Frangin, Y., Maucalire, L., Besnard, J.-C., andGuilloteau, D. 1997. Synthesis and ligand binidng of nortro-pane derivatives: N-substituted-2b-carbomethoxy-3b-(3949-disubsti-tuted phenyl)nortropane. New high affinity and selective com-pounds for the dopamine transporter. J. Med. Chem. 40: 1366–1372.

Farde, L., Gustavsson, J. P., and Jonsson, E. 1997. D2 dopaminereceptors and personality traits. Nature 385: 590.

Goldberger, A. L., Rigney, D. R., and West, B. J. 1990. Chaos andfractals in human phsiology. Sci. Am. 262: 42–49.

Goldberger, A. L. 1996. Fractals and the birth of Gothic: Reflectionson the biologic basis of creativity. Mol. Psychiatry 1: 99–104.

Gongwer, M. A., Murphy, J. M., McBride, W. J., Lumeng, L., and Li,T.-K. 1989. Regional brain contents of serotonin, dopamin andtheir metabolites in the selectively bred high- and low-alcoholdrinking lines of rats. Alcohol 6: 317–320.

Guilloteau, D., Emond, P., Baulieu, J. L., Garreau, L., Frangin, Y.,Pourcelot, L., Mauclaire, L., Besnard, J. C., and Chalon, S. 1998.Exploration of the dopamine transporter: in vitro and in vivocharacterization of a high-affinity and high specificity iodinatedtropane derivative (E)-N-(3-iodoprop-2enyl)-2 beta-carboxyme-thoxy-3beta-(49-methylphenyl)nortropane (PE2I). Nucl. Med. Biol.25: 331–337.

Hall, H., Farde, L., Halldin, C., Hurd, Y. L., Pauli, S., and Sedvall, G.1996. Autoradiographic localization of extra-striatal D2-dopaminereceptors in the human brain using [125]Epidepride. Synapse 23:115–123.all, H., Halldin, C., Farde, L., and Sedvall, G. 1998. Whole hemi-sphere autoradiography of the postmortem human brain. Nucl.Med. Biol. 25: 715–719.

n nucleus accumbens and putamen in the alcoholics and controlstions.

weena

K

K

K

K

K

K

K

L

M

M

M

N

N

P

R

R

R

S

S

S

S

S

T

T

T

T

U

V

W

Y

94 TUPALA ET AL.

Hall, H., Halldin, C., Guilloteau, D., Chalon, S., Emond, P., Besnard,J., Farde, L., and Sedvall, G. 1999. Visualization of the dopaminetransporter in the human brain with the new selective ligand[125]PE2I. NeuroImage 1: 108–116.

Heinz, A., Goldman, D., Jones, D. W., Palmour, R., Hommer, D.,Gorey, J. G., Lee, K. S., Linnoila, M., and Weinberger, D. R. 2000.Genotype influences in vivo dopamine transporter availability inhuman striatum. Neuropsychopharmacology 22: 133–139.

Kaliva, P. W. 1993. Neurotransmitter regulation of dopamine neu-rons in the ventral tegmental area. Brain. Res. Rev. 18: 75–113.aschka, W., Feistel, H., and Ebert, B. 1995. Reduced benzodiaz-epine binding in panic disorders measured by iomazenil study.J. Psychiatr. Res. 29: 427–434.oob, G. F. 1992. Drugs of abuse: Anatomy pharmacology and func-tion of reward pathways. Trends. Pharmacol. Sci. 13: 177–184.oob, G. F., and Le Moal, M. 1997. Drug abuse: hedonic homeostaticdysregulation. Science 278: 52–58.uikka, J. T., Tiihonen, J., Karhu, J., Bergstrom, K. A., andRasanen, P. 1997. Fractal analysis of striatal dopamine re-uptakesites. Eur. J. Nucl. Med. 24: 1085–1090.uikka, J. T., Tiihonen, J., Bergstrom, K. A., Karhu, J., Rasanen, P.,Eronen, M. 1998a. Abnormal structure of human striatal dopa-mine re-uptake sites in habitually violent offenders: A fractalanalysis. Neurosci. Lett. 253: 195–197.uikka, J. T., Yang, J., Karhu, J., Laitinen, J., Tupala, E., Halli-kainen, T., Tiihonen, J. 1998b. Imaging the structure of the stri-atum: A fractal approach to SPECT image interpretation. Physiol.Meas. 19: 367–374.uikka, J. T., Baulieu, J. L., Hiltunen, J., Halldin, C., Bergstrom,K. A., Farde, L., Emond, P., Chalon, S., Meixiang, Y., Nikula, T.,Laitinen, T., Karhu, J., Tupala, E., Hallikainen, T., Kolehmainen,V., Mauclaire, L., Maziere, B., Tiihonen, J., and Guilloteau, D.1998c. Pharmacokinetics and dosimetry of [I-123]PE2I: A Radioli-gand for dopamine transporter imaging. Eur. J. Nucl. Med. 25:531–534.

aine, T. P., Ahonen, A., Torniainen, P., Heikkila, J., Pyhtinen, J.,Rasanen, P., Niemela, O., and Hillbom, M. 1999. Dopamine trans-porters increase in human brain after alcohol withdrawal. Mol.Psychiatry 4: 189–191.andelbrot, B. B. 1983. The Fractal Geometry of Nature. Freeman,San Francisco.uramatsu, T., and Higuchi, S. 1995. Dopamine transporter genepolymorphism and alcoholism. Biochem. Biophys. Res. Commun.211: 28–32.urphy, J. M., McBride, W. J., Lumeng, L., and Li, T.-K. 1987.Contents of monoamines in forebrain regions of alcohol preferring(P) and non-preferring (NP) lines of rats. Pharmacol. Biochem.Behav. 26: 389–392.evo, I., and Hamon, M. 1995. Neurotransmitter and neuromodula-tory mechanisms involved in alcohol abuse and alcoholism. Neu-rochem. Int. 26: 305–336.iewenhuys, R., voogd, J., and vanHuijzen, C. 1988. The HumanCentral Nervous System, 3rd ed. Springer-Verlag, Berling Heidel-berg.

arsian, A., and Zhang, Z. H. 1997. Human dopamine transportergene polymorphism (VNTR) and alcoholism. Am. J. Med. Genet.74: 480–482.epo, E., Kuikka, J. T., Bergstrom, K. A., Karhu, J., Hiltunen, J., andTiihonen, J. 1999. Dopamine transporter and D2-receptor densityin late-onset alcoholism. Psychopharmacology 147: 314–318.

obbins, T. W., and Everitt, B. J. 1999. Drug addiction: Bad habitsadd up. Nature 398: 567–570.osetti, Z. L., D’Aguila, P. S., Hmaidan, Y., Gessa, G. L., and Serra,G. 1991. Repeated treatment with imipramine potentiates cocaine-induced dopamine release and motor stimulation. Eur. J. Pharma-col. 201: 243–245.

ander, T., Harms, H., Podschus, J., Finckh, U., Nickel, B., Rolfs, A.,Rommelspacher, H., and Schmidt, L. G. 1997. Allelic association ofa dopamine transporter gene polymorphism in alcohol dependencewith withdrawal seizures or delirium. Biol. Psychiatry 41: 299–304.

chlegel, S., Steinert, H., Bockish, A., Hahn, K., Schloesser, R., andBenkert, O. 1994. Decreased benzodiazepine binding in panic dis-order measured bu lomazenil SPECT. Eur. Arch. Psychiatry Clin.Neurosci. 244: 49–51.

eeman, P., and Lee, T. 1974. The dopamine releasing actions ofneuroleptics and ethanol. J. Pharmacol. Exp. Ther. 190: 131–135.

orensen, S. M., Juphreys, T. M., Taylor, V. L., and Schmidt, C. J.1992. 5-HT-receptor antagonists reverse amphetamine-inducedslowing of dopaminergic neurons by interfering with stimulateddopamine synthesis. J. Pharmacol. Exp. Ther. 260: 872–878.

wahn, C.-G., Halldin, C., Farde, L., and Sedvall, G. 1996. Synthesisof unlabelled, 3H- and 125I-labelled b-CIT and its Ã-fluoroalkylanalogues b-CIT-FE and b-CIT-FP, including synthesis of precur-sors. J. Labeled Comp. Radiopharmacol. 38: 675–685.

iihonen, J., Kuikka, J., Bergstrom, K., Hakola, P., Karhu, J.,Ryynanen, O-P., Fohr, J. 1995. Altered striatal dopamine re-up-take site densities in habitually violent and non-violent alcoholics.Nature Med. 7: 654–657.

iihonen, J., Kuikka, J., Rasanen, P., Lepola, U., Koponen, H., Li-uska, A., Lehmusvaara, A., Vainio, P., Kononen, M., Bergstrom,K., Yu, M., Kinnunen, I., Åkerman, K., and Karhu, J. 1997. Cere-bral benzodiazepine receptor binding and distribution in general-ized anxiety disorder: A fractal analysis. Mol. Psychiatry 2: 463–471.

iihonen, J., Vilkman, H., Rasanen, P., Ryynanen, O-P., Hakko, H.,Bergman, J., Hamalainen, T., Laakso, A., Haaparanta-Solin, M.,Solin, O., Kuoppamaki, M., Syvalahti, E., and Hietala, J. 1998.Striatal presynaptic dopamine function in type 1 alcoholics mea-sured with positron emission tomography. Mol. Psychiatry 4: 156–161.

upala, E., Hall, H., Sarkioja, T., Rasanen, P., and Tiihonen, J. 2000.Dopamine transporter density in nucleus accumbens of type-1alcoholics. Lancet 355: 380.eno, S., Nakamura, M., Mikami, M., Kondoh, K., Ishiguro, H.,Arinami, T., Komiyama, T., Mitsushio, H., Sano, A., and Tanabe,H. 1999. Identification of a novel polymorphism of the humandopamine transporter (DAT1) gene and the significant associationwith alcoholism. Mol. Psychiatry 4: 552–557.olkow, N. D., Wang, G.-J., Fowler, J. S., Logan, J., Hitzemann, R.,Ding, Y-S., Pappas, N., Shea, C., and Piscani, K. 1996. Decreasesin dopamine receptors but not in dopamine transporters in alco-holics. Alcohol Clin. Exp. Res. 9: 1594–1598.est, B. J., and Goldberger, A. L. 1987. Physiology in fractal dimen-sions. Am. Sci. 75: 354–365.

oshimoto, K., McBride, W. J., Lumeng, L., and Li, T-K. 1991.Alcohol stimulates the release of dopamine and serotonin in thenucleus accumbens. Alcohol 9: 17–22.