Cocaine reduces cytochrome oxidase activity in the prefrontal cortex and modifies its functional connectivity with brainstem nuclei

Available online at www.sciencedirect.com

www.elsevier.com/locate/brainres

b r a i n r e s e a r c h 1 5 4 2 ( 2 0 1 4 ) 5 6 – 6 9

0006-8993/$ - see frohttp://dx.doi.org/10.

nCorresponding autE-mail address: c1These authors c

Research Report

Cocaine reduces cytochrome oxidase activity in theprefrontal cortex and modifies its functionalconnectivity with brainstem nuclei

M.E. Velez-Hernandeza, E. Padillab, F. Gonzalez-Limab,1,C.A. Jimenez-Riveraa,n,1

aDepartment of Physiology, University of Puerto Rico, Medical Sciences Campus, San Juan, PR 00936, USAbDepartments of Psychology, Pharmacology and Toxicology, University of Texas, Austin, TX 78712, USA

a r t i c l e i n f o

Article history:

Accepted 10 October 2013

Cocaine-induced psychomotor stimulation may be mediated by metabolic hypofrontality

and modification of brain functional connectivity. Functional connectivity refers to the

Available online 24 October 2013

Keywords:

Functional connectivity

Cytochrome oxidase

Cocaine

Prefrontal networks

Hypofrontality

nt matter & 2013 Elsevie1016/j.brainres.2013.10.01

hor. Fax: þ1 787 753 [email protected] (ontributed equally to th

a b s t r a c t

pattern of relationships among brain regions, and one way to evaluate this pattern is using

interactivity correlations of the metabolic marker cytochrome oxidase among different

regions. This is the first study of how repeated cocaine modifies: (1) mean cytochrome

oxidase activity in neural areas using quantitative enzyme histochemistry, and (2)

functional connectivity among brain regions using inter-correlations of cytochrome

oxidase activity. Rats were injected with 15 mg/kg i.p. cocaine or saline for 5 days, which

lead to cocaine-enhanced total locomotion. Mean cytochrome oxidase activity was

significantly decreased in cocaine-treated animals in the superficial dorsal and lateral

frontal cortical association areas Fr2 and Fr3 when compared to saline-treated animals.

Functional connectivity showed that the cytochrome oxidase activity of the noradrenergic

locus coeruleus and the infralimbic cortex were positively inter-correlated in cocaine but

not in control rats. Positive cytochrome oxidase activity inter-correlations were also

observed between the dopaminergic substantia nigra compacta and Fr2 and Fr3 areas

and the lateral orbital cortex in cocaine-treated animals. In contrast, cytochrome oxidase

activity in the interpeduncular nucleus was negatively correlated with that of Fr2, anterior

insular cortex, and lateral orbital cortex in saline but not in cocaine groups. After repeated

cocaine specific prefrontal areas became hypometabolic and their functional connectivity

changed in networks involving noradrenergic and dopaminergic brainstem nuclei. We

suggest that this pattern of hypofrontality and altered functional connectivity may

contribute to cocaine-induced psychomotor stimulation.

& 2013 Elsevier B.V. All rights reserved.

r B.V. All rights reserved.7

C.A. Jiménez-Rivera).e study.

1. Introduction

Cocaine is a powerful psychomotor stimulant and its abuseand subsequent addiction are persistent public health pro-blems. Human studies have shown a hypofrontality produced

b r a i n r e s e a r c h 1 5 4 2 ( 2 0 1 4 ) 5 6 – 6 9 57

by chronic use of cocaine (Volkow et al., 1988; London et al.,1990; Matochik et al., 2003; Bolla et al., 2004). In animalmodels, repeated exposure to cocaine results in a progressiveand enduring enhancement in locomotion (Post, 1980; Wiseand Bozarth, 1987; Stewart and Badiani, 1993) and changes invarious brain circuits, especially lower metabolic activity infrontal cortical areas and ventral striatum (Robinson andBerridge, 1993; Porrino et al., 2007). This study was conductedto investigate whether cocaine-enhanced locomotion mayinvolve systems-level alterations in the interactivity orfunctional connectivity of specific prefrontal areas. Whileanatomical connectivity refers to patterns of structural rela-tionships among brain regions, functional connectivityrefers to patterns of relationships in metabolic activityamong brain regions (McIntosh and Gonzalez-Lima, 1994a,b;Nair et al., 1999). If the enhancement of locomotion byrepeated cocaine exposure is an emergent property ofaffected prefrontal areas interacting with subcortical regions,understanding it requires a network analysis of the patternsof interaction between brain regions.

Network functional connectivity uses covariance analysesthat cannot determine directionality but can describe thepatterns of interaction between brain regions, as has beenevaluated by inter-regional correlation changes in cyto-chrome oxidase activity (Sakata et al., 2000; Padilla et al.,2011). This particular functional connectivity method usinginter-correlations of cytochrome oxidase activity describesstable metabolic relationships among areas, and it alsodescribes how the regions are modified across sustainedbehavioral paradigms (Puga et al., 2007; Conejo et al., 2010;Fidalgo et al., 2012) or drug treatments (Padilla et al., 2011;Riha et al., 2011). Characterizing which specific neural sys-tems modify their metabolic capacity and functional con-nectivity as a result of repeated cocaine exposure mayadvance our understanding of cocaine-enhanced locomotion.

Cytochrome oxidase (also called cytochrome c oxidase,ferrocytochrome c: O2 oxidoreductase, EC 1.9.3.1, cytochromeaa3, or the respiratory enzyme) is a ubiquitous mitochondrialmembrane integral protein responsible for the last step of theelectron transport chain that catalyzes the transfer of elec-trons to oxygen, which serves to generate ATP via oxidativephosphorylation (Wong-Riley, 1989; Gonzalez-Lima and Gar-rosa, 1991). Neurons depend mostly on oxidative metabolismas an energy source. For this reason, the enzymatic activity ofcytochrome oxidase is used as a metabolic marker forneuronal activity (Wong-Riley, 1989) and cytochrome oxidaseenzyme histochemistry serves to map sustained changes inbrain energy metabolism (Wong-Riley, 1989; Gonzalez-Limaand Garrosa, 1991; Hevner et al., 1993; Sakata et al., 2005).In particular, we have not seen acute effects on cytochromeoxidase histochemistry one hour after a single drug injection,but the longer-term oxidative capacity for energy meta-bolism (protein-synthesis-dependent enzyme induction overhours or days) of brain regions can be investigated usingquantitative cytochrome oxidase histochemistry (Gonzalez-Lima and Cada, 1994; Padilla et al., 2011; Riha et al., 2011).However, to date, there has not been any cytochrome oxidasestudy in animals exposed to cocaine. Therefore, we wereinterested in using cytochrome oxidase to investigate alteredrelationships between neural areas after 5 days of cocaine

exposure, rather than monitoring acute effects of cocaineexposure.

Quantitative enzyme histochemistry of cytochrome oxi-dase (Gonzalez-Lima and Cada, 1994; Gonzalez-Lima, 1998)has been used successfully in over a hundred previousstudies to map alterations in brain oxidative metabolism innumerous learning tasks and drug treatments (Porembaet al., 1997, 1998; Villarreal et al., 2002; Hu et al., 2006;Gonzalez-Pardo et al., 2008; O’Reilly et al., 2009; Conejoet al., 2010; Padilla et al., 2011; Rojas et al., 2012). Analysisof inter-regional correlations of cytochrome oxidase activity(Sakata et al., 2000; Padilla et al., 2011) between cortical andsubcortical regions after cocaine administration, especiallybetween the prefrontal cortex and monoaminergic nuclei,may also identify underlying initial brain effects of repeatedcocaine.

2. Results

2.1. Cocaine enhanced locomotion from days 1–5

The behavioral protocol showed that rats treated with cocaine(15 mg/kg i.p.) for five days had an increase in total locomotionrelative to saline-injected rats, two way ANOVA F(9,110)¼3.20(po0.001). Additionally on day 5, subjects injected withcocaine had a significantly (po0.05) increased total locomotoractivity (56217533 pcc) when compared to day 1 (29667552pcc), two-way ANOVA F(9,110)¼3.20 (po0.05). There was nosignificant change in total locomotor activity between day 1(655788) and day 5 (542782) in saline-treated rats (Fig. 1).

2.2. Prefrontal regions became hypometabolic afterrepeated cocaine

Mean regional cytochrome oxidase effects of cocaine werefocused on the prefrontal cortex. Cytochrome oxidase activitywas significantly (po0.05) decreased in cocaine-treated ani-mals in the superficial layers of dorsal (Fr2) and lateral (Fr3)frontal cortex regions (DFS mean¼21778 and LFSmean¼24279) when compared to saline-treated animals(DFS mean¼24478 and LFS mean¼26576) (Fig. 2). Meansand standard errors for all regions measured are reported inTable 1, which showed that the hypometabolic effect ofrepeated cocaine (15 mg/kg i.p. for 5 days) was specific toprefrontal cortical areas.

2.3. Prefrontal regions increased their functionalconnectivity with noradrenergic and dopaminergic subcorticalnuclei after repeated cocaine

Specific prefrontal-subcortical nuclei inter-regional cytochromeoxidase correlations were significantly different betweencocaine- and vehicle-treated animals (absolute value ofZabs41.96, po0.05), indicating that cocaine had significanteffects on the functional connectivity of these regions, asillustrated in Fig. 3. Inter-regional correlations of cytochromeoxidase activity showed significant cocaine effects focused onprefrontal regions and noradrenergic and dopaminergic nuclei

Fig. 1 – Locomotor activity. (A) Graph shows mean total locomotor activity (photocell counts/60 min, S.E.M. error bars) after15 mg/kg cocaine (n¼12) or saline (n¼12) daily injections (i.p.) for days 1–5. (B) Graph shows comparison of day 1 and day 5time course of total locomotor activity. Asterisks (n) denote a significant mean group difference as compared to day 1 (po0.05).

b r a i n r e s e a r c h 1 5 4 2 ( 2 0 1 4 ) 5 6 – 6 958

listed in Table 2. No significant cocaine-induced inter-correla-tions differences were found among other regions. For simpli-city, correlations with no significant effects were not listed inTable 2. Three types of significant effects were found:

First, the noradrenergic locus coeruleus (LC) and deeplayers of the infralimbic medial frontal cortex (ILD) werepositively correlated in animals treated with cocaine(r¼0.802) but not in saline-treated animals (r¼�0.076, sig-nificant group difference p¼0.013) (Fig. 4).

Second, significant positive correlations were observed incocaine subjects between the dopaminergic substantia nigracompacta (SNc) and the superficial layer of the prelimbicmedial frontal cortex (MFS, r¼0.809), the deep layer of thelateral (Fr3) frontal cortex (LFD, r¼0.770) and the superficiallayer of the lateral orbital cortex (LOS, r¼0.717). These positivecorrelations were absent in saline-treated animals, revealingsignificant group differences (MFS, r¼0.109 p¼0.031, LFD,r¼�0.118 p¼0.015, and LOS r¼�0.157 p¼0.025). A positivecorrelation between the substantia nigra reticulata (SNr) andthe medial septum (MS) was present in saline (r¼0.602) butnot in cocaine (r¼�0.296 p¼0.033) treated animals.

Third, in cholinergic pathways, interpeduncular nucleus(IP) cytochrome oxidase activity was negatively correlatedwith the activities of the deep layer of the dorsal (Fr2) frontalcortex (DFD, r¼-0.851), the superficial and deep layers of theanterior insular cortex (AIS, r¼�0.672 and AID, r¼�0.954),the superficial and deep layers of the lateral orbital cortex(LOS, r¼�0.697 and LOD, r¼0.847) in saline treated animalsbut not in cocaine treated animals (DFD, r¼0.000 p¼0.008,AIS, r¼0.164 p¼0.032, AID, r¼�0.173 p¼0.0001, LOS, r¼0.238p¼0.018, and LOD, r¼0.300 p¼0.001).

3. Discussion

Here we demonstrated two sets of metabolic changes incytochrome oxidase activity in the brains of rats after a 5-day cocaine exposure paradigm. The rich datasets obtainedfrom this mapping revealed not only mean cytochromeoxidase activity decreases in prefrontal regions (hypofrontal-ity) but also modified functional connectivity between spe-cific prefrontal regions and subcortical noradrenergic anddopaminergic nuclei. This cytochrome oxidase analysis

demonstrated novel effects in subcortical brainstem nucleithat are too small to be reliably measured with noninvasiveneuroimaging studies in humans. We speculate that thesemetabolic alterations produced by repeated cocaine impairprefrontal networks for inhibitory psychomotor control andmight contribute to the observed enhanced locomotion.

3.1. Difference between mean activity and functionalconnectivity

Means provide measures of the average metabolic activity ofindividual regions (univariate measure), whereas functionalconnectivity provides measures of the interactivity amongregions (covariance measure) (McIntosh and Gonzalez-Lima,1994a, 1994b). These two statistical approaches have beenembraced by functional neuroimaging studies because theyare complementary, as the nervous system can be viewed asa complex network of interacting individual regions (Gonzalez-Lima and McIntosh, 1994). The analysis of regional means ofcytochrome oxidase activity focuses on how the baseline meta-bolic activity of specific brain regions change after repeatedcocaine, whereas the analysis of functional connectivity focuseson how baseline interactions among brain regions change afterrepeated cocaine. Two regions can have similar means acrossgroups, but different functional connectivity across groups (Nairet al., 1999; Sakata et al., 2000). Fig. 4 helps to explain how twobrain areas may have similar means across groups but differentfunctional connectivity. For example, the mean cytochromeoxidase activity in the infralimbic deep (ILD) cortical area andthe locus coeruleus (LC) nucleus are not different betweencocaine and saline groups, but the functional connectivitybetween cocaine and saline groups is substantially different(LC-ILD inter-regional correlations¼0.80 cocaine vs. �0.08saline). Therefore, in addition to frontal hypometabolism,it is plausible that cocaine-enhanced locomotor behaviormay be mediated by different patterns in the interactivity orfunctional connectivity among particular frontal-subcorticalbrain regions.

3.2. Cytochrome oxidase vs. other metabolic markers

Elegant studies with the 2-deoxyglucose (2-DG) method inrhesus monkeys have mapped regional decreases in metabolic

Fig. 2 – Cytochrome-oxidase stained sections indicating regions of interest by Bregma level. The purpose of this figure is to showrepresentative control sections at the Bregma levels where the regions of interest were investigated in each brain and toillustrate schematically which were the affected regions. The quantitative densitometric cytochrome oxidase differences andfunctional connectivity changes cannot be seen with the naked eye by comparing sections, and are thus presented in Tables 1and 2. To map these changes schematically in one figure, cocaine effects were illustrated so that regions appearing withboldfaced asterisks showed significant mean differences (po0.05) in cytochrome oxidase (μmol/min/g tissue wet weight)between saline (n¼12) and cocaine (n¼12) groups (Table 1); and boxed regions showed significant differences (po0.05) inpair-wise inter-regional correlations between saline and cocaine groups (Table 2).

b r a i n r e s e a r c h 1 5 4 2 ( 2 0 1 4 ) 5 6 – 6 9 59

activity of frontal cortical areas 32 (prelimbic), 25 (infralimbic)and 24 (anterior cingulate) and the ventral striatum after 5days of cocaine (Porrino et al., 2007). Similar 2-DG studies inrats have shown glucose utilization decreases only in thenucleus accumbens after 5 days of cocaine self-administra-tion, while there were decreases in infralimbic and prelimbiccortical regions after 30 days of cocaine (Macey et al., 2004).

Other studies using Fos immunohistochemistry and c-fosmRNA have been used to measure neuronal activity afterrepeated cocaine administration. These investigations differ

in terms of number of injections given, doses and time ofmeasurement after the last injection, and most importantlythe fact that c-fos is a transient activity marker as opposed tocytochrome oxidase histochemistry that is a maker forchronic, sustained metabolic demand over days. Due to thetransient effects inherent to immediate early gene expres-sion, different results in c-Fos activity were found dependingon whether repeated cocaine administration was given in thehome cage or in the locomotor activity chambers. Home cageinjections diminished Fos expression and c-Fos mRNA in the

Table 1 – Means and standard errors of cytochrome oxidase activity units (μmol/min/g tissue w/w) for all regionsmeasured.

Region Bregma Cocaine Saline

Fr2 Dorsal Frontal Cortex Superficial Layer (DFS) 3.7 216.878.3n 243.578.4n

Fr3 Lateral Frontal Cortex Superficial Layer (LFS) 241.578.7n 264.876.3n

Prelimbic Medial Frontal Cortex Superficial (MFS) 227.374.6 225.774.7Fr2 Dorsal Frontal Cortex Deep Layer (DFD) 218.677.0 236.477.6Prelimbic Medial Frontal Cortex Deep (MFD) 221.874.3 228.074.7Fr3 Lateral Frontal Cortex Deep Layer (LFD) 242.477.8 259.075.9Anterior Insular Cortex Superficial Layer (AIS) 210.178.1 219.676.9Anterior Insular Cortex Deep Layer (AID) 231.977.8 248.076.4Lateral Orbital Cortex Superficial Layer (LOS) 238.376.9 239.675.2Lateral Orbital Cortex Deep Layer (LOD) 239.476.9 253.378.0Medial Orbital Cortex Superficial Layer (MOS) 226.777.2 230.276.0Medial Orbital Cortex Deep Layer (MOD) 235.676.3 244.478.9

Infralimbic Cortex Superficial Layer (ILS) 2.2 229.876.5 226.476.9Infralimbic Cortex Deep Layer (ILD) 232.5 77.1 233.877.1

Anterior Cingulate Cortex Superficial Layer (CGS) 0.7 266.977.7 256.1 78.3Anterior Cingulate Cortex Deep Layer (CGD) 257.578.7 247.477.9Lateral Septum Nucleus (LS) 245.876.1 242.473.1Medial Septum Nucleus (MS) 198.475.2 194.3 72.1Accumbens Nucleus Shell (ACS) 220.776.1 219.775.5Accumbens Nucleus Core (ACC) 245.077.9 246.776.4

Caudate Putamen (CPU) �0.3 234.879.4 231.576.7Globus Pallidus (GP) 158.575.2 150.773.2Bed Nucleus Stria Terminalis (BST) 213.275.7 214.472.4Anterior Commissure (AC) 55.673.6 53.473.9

Lateral Hypothalamic Area (LH) �1.3 186.476.3 178.874.6

Paraventricular Hypothalamic Nucleus (PVH) �1.8 194.2711.1 204.4712.2

Basolateral Amygdala (BLA) �2.3 221.377.2 209.2711.5Central Amygdala (CAM) 228.576.8 234.378.4Medial Amygdala (MAM) 198.977.1 212.174.2Posterior Cingulate/Retrosplenial Cortex (PCN) 284.677.5 282.478.7

Field CA1 of Hippocampus (CA1) �3.8 208.578.6 212.879.3Field CA2 of Hippocampus (CA2) 226.0711.3 225.0712.1Dentate Gyrus (DG) 337.4715.7 327.5715.0Medial Habenula (MHB) 221.3712.6 215.5714.4Lateral Habenula (LHB) 293.8713.4 288.6717.9

Central Gray (CG) �6.3 256.876.3 248.975.2Deep Mesencephalic Nucleus (DPME) 168.075.8 183.7713.0Interpeduncular Nucleus (IP) 378.4713.3 353.5720.6Ventral Tegmental Area (VTA) 102.879.8 95.9712.3Substantia Nigra Reticulata (SNR) 234.777.4 234.3711.3Substantia Nigra Compacta (SNC) 202.075.2 197.777.8

Locus Coeruleus (LC) �9.8 250.4719.3 267.3710.5

n Cytochrome oxidase activity was significantly (po0.05) decreased in cocaine-treated animals (n¼12) as compared to saline (n¼12) in thesuperficial layers of dorsal (Fr2) and lateral (Fr3) frontal cortex regions.

b r a i n r e s e a r c h 1 5 4 2 ( 2 0 1 4 ) 5 6 – 6 960

nAcc and caudate–putamen (Hope et al., 1992; Steiner andGerfen, 1993). In contrast, repeated cocaine administrationinside the activity chambers increased Fos expression in thenAcc without changes in the caudate–putamen (Crombaget al., 2002). Experiments using a similar cocaine administra-tion paradigm to ours found a significant decrease in c-fosdensity in the medial prefrontal cortex and two subdivisionsof the nAcc in animals challenged after a 2-days withdrawalperiod (Todtenkopf et al., 2002). These findings suggest that

chronic cocaine administration induces a decrease in pre-frontal activity.

Interestingly, regions related to self-administration reward,such as the nucleus accumbens, showed no changes in meancytochrome oxidase activity or in functional connectivity inour experimenter-administered cocaine paradigm (15 mg/kgi.p. for 5 days). It is possible that involvement of nucleusaccumbens may require a different behavioral protocolsuch as self-administration to reveal self-stimulation effects

Fig. 3 – Schematic diagram illustrating the significant inter-regional correlations of cytochrome oxidase activitycalculated in both saline (n¼12) and cocaine (n¼12) groups.Abbreviations are the same as in Table 1. Solid linesrepresent pair-wise Pearson's correlations significantlydifferent from zero (po0.05) and dotted lines represent nosignificant correlation. The value and sign of the correlationswith significant group differences are listed in Table 2.

Table 2 – Value and sign of pair-wise correlation coeffi-cients (r) for the saline and cocaine groups and p values ofZabs for the group comparison of each regional pair.Abbreviations are the same as in Table 1.

Saline (r) Cocaine (r) p value

LC/ILD �0.08 0.8 0.01LC/CA1 �0.56 0.31 0.05SNC/MFS 0.11 0.81 0.03SNC/LFD �0.12 0.77 0.02SNC/LOS �0.16 0.72 0.03SNR/MS 0.6 �0.3 0.03IP/DFD �0.85 0 0.01IP/AIS �0.67 0.16 0.03IP/AID �0.95 �0.17 0.001IP/LOS �0.7 0.24 0.02IP/LOD �0.85 0.3 0.001LFS/DG 0 0.74 0.05MOS/AIS 0.4 0.89 0.04

b r a i n r e s e a r c h 1 5 4 2 ( 2 0 1 4 ) 5 6 – 6 9 61

of cocaine. But lack of effects on the midbrain ventral teg-mental area (VTA) are consistent with the 2-DG mappingstudies of cocaine-treated rats by Koch et al. (1997) andPorrino and Kornetsky (1988), although here we providefurther alternative explanations (Porrino and Kornetsky,1988; Koch et al., 1997). For example, after cocaine exposurethe VTA changes from tonic to phasic neuronal firing (Porrinoand Kornetsky, 1988; Koch et al., 1997; Schultz, 1998). Thistransition does not necessarily result in a mean change inneuronal metabolic profile. Thus, it is possible that totalneuronal activity in the VTA is not changed, only the patternof firing. Furthermore, we have recently shown that a decrease

in VTA dopaminergic neuronal cell size occurs after cocainesensitization (Arencibia-Albite et al., 2012). This diminutionmay compensate for the increase in metabolic activity thatone may expect with the long-term potentiation known tooccur after chronic cocaine treatment (Ungless et al., 2001).Therefore, some changes in neuronal activity patterns in areaslike the VTA and the nucleus accumbens may not be detectedwith cytochrome oxidase histochemistry or with the cocaineprotocol we used. These regions may show alterations only if adifferent methodology for mapping neuronal activity or adifferent behavioral protocol such as self-administrationare used.

The cytochrome oxidase technique, like 2-DG and fluor-odeoxyglucose (FDG), offers a functional marker for neuronalenergy metabolism (Gonzalez-Lima 1998). Cytochrome oxi-dase is the final step in the electron transport chain; there-fore, its catalytic activity is critical for glucose oxidation andthe creation of ATP (Wong-Riley, 1989; Wong-Riley et al.,1998). But cytochrome oxidase mapping was used to assessthe cumulative effects of repeated drug manipulations overdays (Padilla et al., 2011). While 2-DG/FDG and immediateearly genes as c-fos evaluate on-going changes in neuronalactivity during the period of tracer uptake or gene expression(usually 45–90 min), cytochrome oxidase measures longer-term alterations in enzyme levels that are induced by days ofsustained metabolic changes on the implicated brain region(Gonzalez-Lima, 1998). These markers provide measures ofneuronal metabolism, but some (2-DG, FDG, c-fos) measuremore the on-going neuronal activation while the other(cytochrome oxidase) gives information on longer-termchanges on neuronal oxidative metabolic capacity. The oxi-dative capacity for energy metabolism, as measured byquantitative cytochrome oxidase histochemistry, reflectsthe energy demand history of different brain regions afterrepeated cocaine exposure because cytochrome oxidaselevels change by enzymatic induction (a process that requiresprotein synthesis) in a cumulative way after repeated dailychanges in energy demand (Gonzalez-Lima and Cada, 1994).Thus, we found that cytochrome oxidase served to determinecumulative effects on prefrontal neural oxidative metaboliccapacity and functional connectivity after a 5-day cocainetreatment, although it should be acknowledged that a moreextended cocaine treatment or a different protocol using awithdrawal period could produce different changes.

3.3. Prefrontal cortical involvement

Mean cytochrome oxidase was decreased in cocaine treatedanimals when compared to saline injected rats in the super-ficial layers of dorsal (Fr2) and lateral (Fr3) frontal cortexregions (Table 1). Hypometabolic effects circumscribed to thesuperficial cortical layers of Fr2 and Fr3 imply an alteration ofintercortical communication of association fibers betweenthe superficial layers, as opposed to ascending or descendingprojection fibers that innervate deeper cortical layers. TheseFr2 and Fr3 areas of Zilles and Wree shown in the atlas ofPaxinos and Watson (1986), correspond to the most anteriorsecondary and tertiary association areas of the rat frontalcortex that are regarded as prefrontal cortical areas (Paxinosand Watson, 1986; Paxinos, 1995; Shumake and Gonzalez-Lima,

Fig. 4 – Inter-regional correlations patterns among nucleus locus coeruleus (LC) and deep infralimbic frontal cortex (ILD).Plots show significantly (p¼0.01) different inter-regional correlations of cytochrome oxidase activity between the LC and theILD in cocaine and saline treated animals. The line shows the best fit linear regression (r¼0.80 in cocaine and r¼�0.08 insaline treated animals). Box shows the mean cytochrome oxidase activity units (μmol/min/g tissue wet weight), which are notsignificantly different for both regions in saline (n¼12) and cocaine treated rats (n¼12).

b r a i n r e s e a r c h 1 5 4 2 ( 2 0 1 4 ) 5 6 – 6 962

2003). They are not part of the rat primary motor cortex (Fr1) butmay be prefrontal cortical areas involved in descending inhibi-tory control of psychomotor behavior. Therefore, cocaine-induced hypometabolism of Fr2 and Fr3 areas may lead todiminished inhibitory control that may manifest as enhancedlocomotion.

Other prefrontal cortical areas in the rat showed cocaine-modified functional connectivity, including the prelimbicmedial frontal cortex (area 32), the infralimbic medial frontalcortex (area 25), and the medial and lateral orbital frontalcortex (Fig. 2). These prefrontal regions receive projectionsfrom most of the mesocorticolimbic system such as the VTA,substantia nigra, amygdala, lateral hypothalamus, hippocam-pus, and other areas of the cortex (Groenewegen et al., 1997;Dalley et al., 2004b). They also project back to most of theseregions (Groenewegen et al., 1997; Dalley et al., 2004a). Butthese mesocorticolimbic regions did not show significantchanges in mean cytochrome oxidase or functional connec-tivity differences with the prefrontal areas after our cocaineparadigm. This may be related to the fact that cocaine wasnot self-administered by the rats and thus may not engagethe mesocorticolimbic self-stimulation system.

As discussed below, there were only specific noradrenergicand dopaminergic brainstem nuclei that showed cytochromeoxidase inter-regional correlation differences with prefrontalareas when saline-treated animals are compared to cocaine-treated rats (Fig. 3). Since cocaine not only increases synapticlevels of noradrenaline and dopamine, but also that ofserotonin and glutamate, simply increasing transmitterlevels is unlikely to explain the specific cocaine-inducedmodification of functional connectivity observed in thisstudy. We speculate that 5 days of cocaine leads to neuro-metabolic adaptations that are more specific for prefrontalcortical areas and their functional connectivity with noradre-nergic and dopaminergic brainstem nuclei. Future studies

would be needed to further evaluate the mechanisms ofthese changes.

3.4. Noradrenergic pathways

Our studies found that the noradrenergic locus coeruleus (LC)and deep layers of the infralimbic cortex were positivelycorrelated in cocaine-treated animals but not in control rats(Table 2 and Fig. 4). But inter-regional correlations do notprovide information on the direction of the influence, so theyneed to be interpreted based on other known anatomical andfunctional data (Gonzalez-Lima and McIntosh, 1994). Thenoradrenergic system contributes to control stress responses,arousal, mood and alters learning and memory (Huether,1996; Sved et al., 2001) and plays a key role in mediatingreward (Poschel and Ninteman, 1963; Stein, 1964, 1975; Wise,1978). The LC is the principal site for norepinephrine (NE)synthesis in the brain and it projects to extensive areas thatmay help modulate the observed cocaine-enhanced locomo-tion (projections reviewed by Foote et al. (1983), Grzanna andFritschy (1991), Holstege and Bongers (1991), Jones (1991),Westlund et al. (1991) and Berridge and Waterhouse (2003)).It is also possible that a modified pattern of LC functionalconnectivity might underlie cocaine-induced behaviors notinvestigated in this study, such as self-stimulation, andfuture studies manipulating behavior or brain mechanismswould be important to test this possibility. Regions of themesolimbic dopamine (DA) system, like the VTA, nucleusaccumbens and amygdala receive noradrenergic inputs(Ungerstedt, 1971; Alheid and Heimer, 1988; Liprando et al.,2004; Mejias-Aponte et al., 2009). Also, noradrenergic neuronsin the LC receive afferents from the infralimbic prefrontalcortex (Heidbreder and Groenewegen, 2003).

The links between the LC and the infralimbic cortex withthe VTA are so strong that activation of alpha-1 receptors in

b r a i n r e s e a r c h 1 5 4 2 ( 2 0 1 4 ) 5 6 – 6 9 63

the LC or in the infralimbic cortex by themselves increaseVTA DA neuronal activity (Lategan et al., 1990; Blanc et al.,1994). In addition, our lab has shown that glutamate releaseonto VTA DA neurons is modulated by pre-synaptic alpha-1receptors (Velasquez-Martinez et al., 2012); and systemicinhibition of alpha-1 and activation of alpha-2 receptors blockthe development and expression of cocaine sensitization(Jimenez-Rivera et al., 2006). Together with the observedchange in functional connectivity between the LC and theinfralimbic cortex, these data suggest the hypothesis that NEpathways to the infralimbic cortex play a critical role in thedevelopment of cocaine-induced neuroadaptations. Based onthis hypothesis, the observed significant pair-wise positivecorrelation in cytochrome oxidase activity may be interpretedas repeated cocaine modifying the noradrenergic LC influenceon the infralimbic cortex. This hypothesis deserves testing infuture studies.

3.5. Dopaminergic pathways

There were significant positive correlations in cocaine-treated subjects between the substantia nigra compacta(SNc) and the superficial layer of the prelimbic medial frontalcortex (MFS), the deep layer of the Fr3 lateral frontal cortex(LFD) and the superficial layer of the lateral orbital cortex.These correlations were not present in saline-treated rats(Fig. 3). A positive correlation between the SN reticulata andthe medial septum was present in saline and not cocainetreated animals. The SN is a predominantly dopaminergicarea most commonly known for its involvement in Parkin-son's disease and the extra-pyramidal motor system linked toinvoluntary movement control. The presence of a positivecorrelation between the SN and specific prefrontal regionssuggests a role of DA in the control of these areas afterrepeated cocaine exposure. Augmentation of DA release withrepeated drug exposure is the basis of theories that suggestdrugs of abuse impair adaptive circuitry to become hyper-responsive to drug stimuli (Berridge and Robinson, 1998;Goldstein and Volkow, 2002; Everitt and Robbins, 2005;Kalivas, 2008). In addition, there is a direct interaction fromprefrontal cortex to SN (Watabe-Uchida et al., 2012).

The orbital frontal cortex projects to central parts ofcaudate–putamen and to the lateral part of nucleus accum-bens shell (Schilman et al., 2008). This area also receivesafferents from regions like the VTA, ventral pallidum and themedial temporal lobe (Krettek and Price, 1977; Groenewegen,1988; Ray and Price, 1992) and it is thought to be important indrug vulnerability (Schoenbaum and Shaham, 2008). Hyper-activity of the orbitofrontal cortex in humans results inimpulsive behavior (Baxter et al., 1987, 1989; Zametkinet al., 1990; Andreason et al., 1994) and this area compensatesthis hyperactivity by an inhibitory mechanism (Winstanley,2007). Monkeys and humans show orbital cortex dysfunctionafter cocaine exposure (Franklin et al., 2002; Jentsch et al.,2002; Olausson et al., 2007). Increases in metabolic ratesin the orbitofrontal cortex and basal ganglia in humans correlatenegatively with the duration of abstinence (Volkow et al., 1991).Further studies suggest that orbital cortex abnormalities are aconsequence of drug exposure and not a predisposing factor fordrug addiction (Perry et al., 2011). The observed changes in

cytochrome oxidase correlations between specific regions of theprefrontal cortex and the SNc may be relevant to a role of DApathways in the loss of inhibitory control and increasedlocomotion seen after cocaine intake.

3.6. Cholinergic pathways

Cytochrome oxidase activity in the interpeduncular nucleus(IP) was negatively correlated with the activities of the deeplayer of the Fr2 dorsal frontal cortex, the superficial and deeplayers of the anterior insular cortex, the superficial and deeplayers of the lateral orbital cortex in saline but not in cocaine-treated animals (Fig. 3). Although IP neurons are mainlyGABAergic, its major inputs are cholinergic projections fromthe medial habenula. Indeed, the IP receives more acetylcho-line input than any other region in the mammalian brain(Herkenham and Nauta, 1979; Villani et al., 1983; Artymyshynand Murray, 1985; Contestabile et al., 1987; Eckenrode et al.,1987; Fasolo et al., 1992). Studies have shown that interpe-duncular pathways (i.e. habenula-ip) and the mesolimbicpathways are mutually inhibitory, (Sutherland andNakajima, 1981; Nishikawa et al., 1986). In addition, it hasbeen reported that cocaine injections decrease the extracel-lular levels of acetylcholine in the IP (Hussain et al., 2008).Nicotininc AChR subunits in the IP have been also beenshown to play a crucial role in somatic withdrawal symptomsin drugs of abuse like nicotine (Salas et al., 2009). Acetylcho-line is an integral component of the mesolimbic system(Hoebel et al., 2007). Cholinergic projections may be alteredin learning and memory processes like those changed in drugabuse (Vorel et al., 2001; See et al., 2003). For example,nicotinic receptor inactivation decreases sensitivity tococaine whereas nicotine exposure enhances cocaine seeking(Zachariou et al., 2001; Bechtholt and Mark, 2002). On theother hand, changes in the neurochemistry of the IP and itsprojections induce changes in motor behavior (Shannon andPeters, 1990; Salas et al., 2004; Taraschenko et al., 2007).For example, muscarinic antagonists enhance cocaine andamphetamine-induced locomotor effects (Shannon andPeters, 1990; Bymaster et al., 1993; Ichikawa et al., 2002),whereas mice lacking muscarinic M5 receptors self-administer less cocaine and show reduced conditioned placepreference (Fink-Jensen et al., 2003). These studies suggestthat cholinergic receptors mediate some component ofcocaine-induced changes. Recent evidence has shown thatblockade of nicotinic cholinergic receptors in the VTA pre-vents the cocaine-induced DA increase (Mark et al., 2011).There is also evidence that supports that disruption of IPsignaling can lead to increased drug intake (Picciotto, 1998;Cui et al., 2003; Tapper et al., 2004; Fowler et al., 2011). Theloss of connectivity of the IP observed in our experimentssuggest that cocaine exposure produces an abnormality inacetylcholine signaling that may affect the connectionsbetween the IP and frontal cortical structures that couldfurther enable cocaine sensitization.

3.7. Conclusion

First, the hypofrontality results presented here are in agree-ment with previous glucose metabolic mapping studies

b r a i n r e s e a r c h 1 5 4 2 ( 2 0 1 4 ) 5 6 – 6 964

in cocaine-exposed animals (Porrino and Kornetsky, 1988;Koch et al., 1997). They also agree with previous humanstudies that suggest a hypofrontality produced by chronic useof drugs of abuse such as cocaine (Volkow et al., 1988; Londonet al., 1990; Matochik et al., 2003; Bolla et al., 2004). Second,our findings also indicate, for the first time, that repeatedcocaine modifies the functional connectivity between specificprefrontal regions and subcortical noradrenergic, dopaminer-gic and cholinergic pathways. Taken together these patternsof functional connectivity suggest the general hypothesisthat prefrontal networks change from cholinergic influencesto networks driven by noradrenergic and dopaminergic nucleiafter repeated cocaine administration. Finally, we speculatethat when these prefrontal networks for inhibitory controlare modified by repeated cocaine, enhanced locomotion mayarise, which suggests that treatments that increase themetabolic capacity of these prefrontal networks may antag-onize the psychomotor effects of cocaine (Chen et al., 2013).

4. Experimental procedures

4.1. Subjects

All procedures were performed according to the US PublicHealth Service publication “Guide for the Care and Use ofLaboratory Animals” and were approved by the Animal Careand Use Committee at the University of Puerto Rico MedicalSciences Campus. Twenty-four Sprague-Dawley male rats(Taconic Farms) weighing between 250–300 g were housedtwo per cage and maintained at constant temperature andhumidity with a 12-h light/dark cycle. All behavioral experi-ments were carried out during the light period. Water andfood were provided ad libitum.

4.2. Cocaine protocol and behavior

Animals were randomly divided into two groups (saline n¼12and cocaine n¼12). Prior to any drug treatment, all animals(n¼24) were habituated to locomotor chambers (AccuScanInstruments Inc., Columbus, OH) for two daily 1 h sessions.Each cage frame houses sensor panels and a row of 16infrared beams. When a rodent intersects the beams, thesoftware detects a photocell count. On experimental day 1,animals were placed for 15 min in the photocell box. After15-min habituation, animals were treated with either 15 mg/kg i.p. of cocaine (Sigma, St. Louis, MO, USA) or isovolumetricsaline injections. This particular dose repeated for five injec-tions was chosen because it is a well-established protocol toinduce progressive cocaine-enhanced locomotion that iscommonly used in our laboratory (Jimenez-Rivera et al.,2006; Arencibia-Albite et al., 2012; Santos-Vera et al., 2013)and others (Pierce et al., 1996; Robinson and Berridge, 2001;Yoon et al., 2007). Therefore, using this protocol makes thisstudy more relevant to compare with similar cocaine ratstudies in the literature. Immediately after injections, totallocomotor activity was assessed for 1 h. The total number ofbeam breaks (photocell counts) determined total locomotion.This procedure was repeated for 5 consecutive days. On day

5, after locomotor activity was assessed for 1 h, rats' brainswere collected for tissue processing.

4.3. Tissue processing and cytochrome oxidase staining

Brains were quickly removed and frozen in isopentane. Usinga cryostat microtome (Leica CM3000, Germany) at �20 1C,brains were sectioned at 40 μm, mounted on slides and keptfrozen at �40 1C until they were processed using quantitativecytochrome oxidase enzyme histochemistry (Gonzalez-Lima,1998). Series of coronal sections from each brain were used toperform cytochrome oxidase histochemistry following thevalidated quantitative protocol described by Gonzalez-Limaand collaborators (Gonzalez-Lima and Cada, 1994; Gonzalez-Lima and Jones, 1994) that has been used in hundreds ofother cytochrome oxidase studies. The staining reactionsignal is produced when diaminobenzidine is oxidized to avisible indamine polymer. The reaction product is furtherintensified by the addition of cobalt to the preincubationsolution. Since continuous reoxidation of cytochrome c bycytochrome oxidase is needed for the accumulation of thevisible product, this reaction under linear conditions serves tovisualize cytochrome oxidase reactivity. Enzymatic activityunits are calculated using calibration standards made ofbrain paste which showed a linear relationship (r¼0.99)between cytochrome oxidase activity units measured spec-trophotometrically and cytochrome oxidase reactivity mea-sured with densitometry. Briefly, frozen slides were fixed for5 min using a 10% sucrose phosphate buffer (0.1 M pH 7.6)containing 0.5% glutaraldehyde to adhere the sections to theslides. Next, slides were rinsed in three changes of a 10%sucrose phosphate buffer (0.1 M) for 5 min each to remove redblood cells and warm tissue to room temperature. Then theslides were pre-incubated in 0.05 M Tris buffer (pH 7.6), with275-mg/l cobalt chloride, 10% sucrose, and 0.5% dimethylsulf-oxide, for 10 min to enhance staining contrast (metal inten-sification). Slides were then rinsed for 5 min in phosphatebuffer and incubated in 700 ml of oxygen-saturated 0.1 Mphosphate buffer (350 mg diaminobenzidine tetrahydrochlor-ide, 52.5 mg cytochrome c, 35 g sucrose, 14 mg catalase, and1.75 ml dimethylsulfoxide) at 37 1C for 1 h. The tissue wasfixed in buffered formalin to stop the last reaction (for 30 minat room temperature with 10% sucrose and 4% formalin).Finally, the slides were dehydrated in a series of ethanolbaths (increasing from 30% to 100% ethanol), cleared byimmersion in xylene, and cover slipped with Permount(Fisher Scientific, Pittsburgh, PA, USA).

Each batch of slides was accompanied by a set of brainpaste standards to quantify enzymatic activity and to controlfor staining variability across batches. For these standards,the brains from 12 additional Sprague-Dawley male rats wereremoved after decapitation, stored at 4 1C (in sodium phos-phate buffer, pH 7.6), and then homogenized at 4 1C(Gonzalez-Lima, 1998). Cytochrome oxidase activity of thebrain paste was spectrophotometrically assessed as describedby Gonzalez-Lima and Cada (1998), and activity units weredefined at pH 7 and 37 1C, where 1 unit oxidizes 1 μmol ofreduced cytochrome c per min (μmol/min/g tissue wetweight). Remaining paste was frozen in the same manneras the experimental brains and stored at �40 1C. Immediately

b r a i n r e s e a r c h 1 5 4 2 ( 2 0 1 4 ) 5 6 – 6 9 65

prior to each cytochrome oxidase staining procedure, cryostatsections of different thickness (10, 20, 40, 60 and 80 μm) wereobtained from the rat brain paste and mounted on a slide.These sets of sections of known cytochrome oxidase activitywere used as calibration standards in each cytochromeoxidase staining bath.

4.4. Cytochrome oxidase activity mapping

Using a stereotaxic atlas of the rat brain (Paxinos andWatson, 1986) as well as a cytochrome oxidase atlas of therat brain (Gonzalez-Lima, 1998), cytochrome oxidase-stainedsections were carefully selected for both the appropriatelevels of brain regions of interest and the integrity of thesections. The regions of interest examined are illustrated byBregma level in Fig. 2. An image-processing system consistingof a high-gain video camera, Targa-M8 image capture board,Everex computer, Sony color monitor, DC-powered illumina-tor, and JAVA software (Jandel Scientific, San Rafael, CA, USA)was used to sample optical density (OD) from each ROI. Thissystem was calibrated before each measurement sessionusing an OD tablet (Kodak, Rochester, NY, USA). The filmhad a known set of absolute OD units in seven standardsranging from 0 to 0.92 OD units. Background subtraction ofthe clear part of the slide without sections was used tocorrect possible optical artifacts from the camera.The histochemical reaction product from cytochrome oxidasestaining was measured in OD units. In each measured region,four readings of each section were taken on each of threeadjacent sections to yield 12 readings per region per brain.For each region measured, size of the square-shaped sam-pling window was adjusted so that it was as large as possiblewhile still allowing two, non-overlapping readings to be takenbilaterally (four total). The size of the window was heldidentical across subjects, as was the number of readings foreach ROI. The OD values of these readings were thenconverted to cytochrome oxidase activity units (μmol/min/gtissue w/w) using a regression curve (r240.90) that wasobtained from the mean OD values and enzymatic activityof the tissue standards stained in the same batch and imagedin the same measurement session (Gonzalez-Lima, 1998).

4.5. Statistical analysis of locomotion, mean cytochromeoxidase activity and functional connectivity

Total locomotor activity, expressed as photocell counts,between groups were analyzed using Two-way ANOVA forrepeated measures followed by Bonferroni post-test in orderto establish behavioral locomotor effects (numbers are pre-sented as mean 7 standard error). A significant day 1 to day 5difference in mean photocell counts at po0.05 was consid-ered a successful enhanced locomotor effect. Group differ-ences in mean cytochrome oxidase activity measured in eachbrain region were evaluated by one-way ANOVA. Functionalconnectivity was assessed by computing separate pair-wisePearson correlation matrices of cytochrome oxidase activityacross all regions of interest for each group (within-groupanalysis) (Puga et al., 2007). In this data-driven approach, allthe brain regions are evaluated and the brain effects deter-mine which are the relevant inter-regional correlations, as

opposed to using a more restricted theory-driven or arbitraryselection of regions for analysis (McIntosh and Gonzalez-Lima, 1994a, 1994b). A “jackknife” procedure was used toensure the reliability of significant correlations and to protectagainst the effects of outliers to avoid inflated correlationestimates which sometimes result from small sample sizes.In this procedure, an individual is removed from the data setand correlations are computed on the data set with n�1subjects. The individual is then replaced into the data set,and another individual is removed. Correlations and signifi-cance tests are recomputed on the data set with n-1 subjects.This procedure is repeated until all individuals have beenremoved once. Based on the calculation of all possible pair-wise correlations resulting from removing one subject at atime, this procedure takes into consideration only thosecorrelations that remain significant (po0.01) across all possi-ble combinations. These correlations were then tested forsignificant differences between groups (between-group ana-lysis). The Fisher Z transformation was used to convert eachcorrelation to a Z score to test differences in inter-regionalcorrelations between groups (Jones and Gonzalez-Lima, 2001;Bruchey and Gonzalez-Lima, 2006; Puga et al., 2007). Signifi-cant group differences of pair-wise inter-regional correlationsbetween cocaine- and vehicle-treated animals were calcu-lated as absolute value of Zabs41.96 (po0.05). As with allfunctional connectivity methods, inter-regional correlationsdo not provide information on the direction of the influence,so they need to be interpreted based on known anatomicalpathways. For example, if region A that influences region Bvia an anatomical path showed a significant pair-wise posi-tive correlation in cytochrome oxidase activity after treat-ment, it may be inferred that a change in activity in region Awas functionally related to a corresponding alteration inregion B (Puga et al., 2007). The term functional connectionwas used to refer to jackknife-tested reliably significantcytochrome oxidase activity correlations between two brainregions (Padilla et al., 2011).

Acknowledgments

This work was supported by grants from NIGMSSC1GM084854 to CAJR, MH65728 to FGL and MBRS-RISE R25-GM061838 to MVH. The authors thank Maria C. Velazquez,Bermary Santos and Rafael Vazquez for their help duringbrain collection. We also thank Dr. Jason Shumake for hishelp with statistical analysis and Dr. Douglas Barrett for hisassistance with figure making. Dr. Vélez-Hernández con-ducted this study in partial fulfillment of the requirementsfor the Ph.D. degree at the University of Puerto Rico, MedicalSciences Campus.

r e f e r e n c e s

Alheid, G.F., Heimer, L., 1988. New perspectives in basal forebrainorganization of special relevance for neuropsychiatricdisorders: the striatopallidal, amygdaloid, and corticopetalcomponents of substantia innominata. Neuroscience 27, 1–39.

b r a i n r e s e a r c h 1 5 4 2 ( 2 0 1 4 ) 5 6 – 6 966

Andreason, P.J., Zametkin, A.J., Guo, A.C., Baldwin, P., Cohen, R.M.,1994. Gender-related differences in regional cerebral glucosemetabolism in normal volunteers. Psychiatry Res. 51, 175–183.

Arencibia-Albite, F., Vazquez, R., Velasquez-Martinez, M.C.,Jimenez-Rivera, C.A., 2012. Cocaine sensitization inhibits thehyperpolarization-activated cation current Ih and reduces cellsize in dopamine neurons of the ventral tegmental area. J.Neurophysiol. 107, 2271–2282.

Artymyshyn, R., Murray, M., 1985. Substance P in theinterpeduncular nucleus of the rat: normal distribution andthe effects of deafferentation. J. Comp. Neurol. 231, 78–90.

Baxter Jr., L.R., Phelps, M.E., Mazziotta, J.C., Guze, B.H., Schwartz, J.M.,Selin, C.E., 1987. Local cerebral glucose metabolic rates inobsessive-compulsive disorder. A comparison with rates inunipolar depression and in normal controls. Arch. Gen.Psychiatry 44, 211–218.

Baxter Jr., L.R., Schwartz, J.M., Phelps, M.E., Mazziotta, J.C.,Guze, B.H., Selin, C.E., Gerner, R.H., Sumida, R.M., 1989.Reduction of prefrontal cortex glucose metabolism commonto three types of depression. Arch. Gen. Psychiatry 46, 243–250.

Bechtholt, A.J., Mark, G.P., 2002. Enhancement of cocaine-seekingbehavior by repeated nicotine exposure in rats.Psychopharmacology (Berl) 162, 178–185.

Berridge, C.W., Waterhouse, B.D., 2003. The locus coeruleus-noradrenergic system: modulation of behavioral state andstate-dependent cognitive processes. Brain Res. Brain Res.Rev. 42, 33–84.

Berridge, K.C., Robinson, T.E., 1998. What is the role of dopaminein reward: hedonic impact, reward learning, or incentivesalience?. Brain Res. Brain Res. Rev. 28, 309–369.

Blanc, G., Trovero, F., Vezina, P., Herve, D., Godeheu, A.M.,Glowinski, J., Tassin, J.P., 1994. Blockade of prefronto-corticalalpha 1-adrenergic receptors prevents locomotorhyperactivity induced by subcortical D-amphetamineinjection. Eur. J. Neurosci. 6, 293–298.

Bolla, K., Ernst, M., Kiehl, K., Mouratidis, M., Eldreth, D.,Contoreggi, C., Matochik, J., Kurian, V., Cadet, J., Kimes, A.,Funderburk, F., London, E., 2004. Prefrontal corticaldysfunction in abstinent cocaine abusers. J. NeuropsychiatryClin. Neurosci. 16, 456–464.

Bruchey, A.K., Gonzalez-Lima, F., 2006. Brain activity associatedwith fear renewal. Eur. J. Neurosci. 24, 3567–3577.

Bymaster, F.P., Heath, I., Hendrix, J.C., Shannon, H.E., 1993.Comparative behavioral and neurochemical activities ofcholinergic antagonists in rats. J. Pharmacol. Exp. Ther. 267,16–24.

Chen, B.T., Yau, H.J., Hatch, C., Kusumoto-Yoshida, I., Cho, S.L.,Hopf, F.W., Bonci, A., 2013. Rescuing cocaine-inducedprefrontal cortex hypoactivity prevents compulsive cocaineseeking. Nature 496, 359–362.

Conejo, N.M., Gonzalez-Pardo, H., Gonzalez-Lima, F., Arias, J.L.,2010. Spatial learning of the water maze: progression of braincircuits mapped with cytochrome oxidase histochemistry.Neurobiol. Learn. Mem. 93, 362–371.

Contestabile, A., Villani, L., Fasolo, A., Franzoni, M.F., Gribaudo, L.,Oktedalen, O., Fonnum, F., 1987. Topography of cholinergicand substance P pathways in the habenulo-interpeduncularsystem of the rat. An immunocytochemical andmicrochemical approach. Neuroscience 21, 253–270.

Crombag, H.S., Jedynak, J.P., Redmond, K., Robinson, T.E., Hope, B.T.,2002. Locomotor sensitization to cocaine is associated withincreased Fos expression in the accumbens, but not in thecaudate. Behav. Brain Res. 136, 455–462.

Cui, C., Booker, T.K., Allen, R.S., Grady, S.R., Whiteaker, P.,Marks, M.J., Salminen, O., Tritto, T., Butt, C.M., Allen, W.R.,Stitzel, J.A., McIntosh, J.M., Boulter, J., Collins, A.C.,Heinemann, S.F., 2003. The beta3 nicotinic receptor subunit: acomponent of alpha-conotoxin MII-binding nicotinic

acetylcholine receptors that modulate dopamine release andrelated behaviors. J. Neurosci. 23, 11045–11053.

Dalley, J.W., Cardinal, R.N., Robbins, T.W., 2004a. Prefrontalexecutive and cognitive functions in rodents: neural andneurochemical substrates. Neurosci. Biobehav. Rev. 28,771–784.

Dalley, J.W., Theobald, D.E., Bouger, P., Chudasama, Y., Cardinal, R.N.,Robbins, T.W., 2004b. Cortical cholinergic function and deficits invisual attentional performance in rats following 192 IgG-saporin-induced lesions of the medial prefrontal cortex. Cereb. Cortex 14,922–932.

Eckenrode, T.C., Barr, G.A., Battisti, W.P., Murray, M., 1987.Acetylcholine in the interpeduncular nucleus of the rat:normal distribution and effects of deafferentation. Brain Res.418, 273–286.

Everitt, B.J., Robbins, T.W., 2005. Neural systems of reinforcementfor drug addiction: from actions to habits to compulsion. Nat.Neurosci. 8, 1481–1489.

Fasolo, A., Virgili, M., Panzica, G.C., Contestabile, A., 1992.Immunohistochemistry and neurochemistry of the habenulo-interpeduncular connection after partial developmentaldepletion of habenular cholinergic neurons in the rat. Exp.Brain Res. 90, 297–301.

Fidalgo, C., Conejo, N.M., Gonzalez-Pardo, H., Arias, J.L., 2012.Functional interaction between the dorsal hippocampus andthe striatum in visual discrimination learning. J. Neurosci.Res. 90, 715–720.

Fink-Jensen, A., Fedorova, I., Wortwein, G., Woldbye, D.P.,Rasmussen, T., Thomsen, M., Bolwig, T.G., Knitowski, K.M.,McKinzie, D.L., Yamada, M., Wess, J., Basile, A., 2003. Role forM5 muscarinic acetylcholine receptors in cocaine addiction. J.Neurosci. Res. 74, 91–96.

Foote, S.L., Bloom, F.E., Aston-Jones, G., 1983. Nucleus locusceruleus: new evidence of anatomical and physiologicalspecificity. Physiol. Rev. 63, 844–914.

Fowler, C.D., Lu, Q., Johnson, P.M., Marks, M.J., Kenny, P.J., 2011.Habenular alpha5 nicotinic receptor subunit signallingcontrols nicotine intake. Nature 471, 597–601.

Franklin, T.R., Acton, P.D., Maldjian, J.A., Gray, J.D., Croft, J.R.,Dackis, C.A., O’Brien, C.P., Childress, A.R., 2002. Decreased graymatter concentration in the insular, orbitofrontal, cingulate,and temporal cortices of cocaine patients. Biol. Psychiatry 51,134–142.

Goldstein, R.Z., Volkow, N.D., 2002. Drug addiction and itsunderlying neurobiological basis: neuroimaging evidence forthe involvement of the frontal cortex. Am. J. Psychiatry 159,1642–1652.

Gonzalez-Lima, F., 1998. Cytochrome oxidase in neuronalmetabolism and Alzheimer's disease. Plenum Press, NewYork.

Gonzalez-Lima, F., Cada, A., 1994. Cytochrome oxidase activity inthe auditory system of the mouse: a qualitative andquantitative histochemical study. Neuroscience 63, 559–578.

Gonzalez-Lima, F., Garrosa, M., 1991. Quantitative histochemistryof cytochrome oxidase in rat brain. Neurosci. Lett. 123,251–253.

Gonzalez-Lima, F., Jones, D., 1994. Quantitative mapping ofcytochrome oxidase activity in the central auditory system ofthe gerbil: a study with calibrated activity standards andmetal-intensified histochemistry. Brain Res. 660, 34–49.

Gonzalez-Lima, F.A., McIntosh, A.R., 1994. Neural networkinteractions related to auditory learning analyzed withstructural equation modeling. Hum. Brain Mapp. 2, 23–44.

Gonzalez-Lima, F., Cada, A., 1998. Quantitative histochemistry ofcytochrome oxidase activity: theory, methods and regionalbrain vulnerability. In: Gonzalez-Lima, F. (Ed.), CytochromeOxidase in Neuronal Metabolism and Alzheimer's Disease,pp. 55–90.

b r a i n r e s e a r c h 1 5 4 2 ( 2 0 1 4 ) 5 6 – 6 9 67

Gonzalez-Pardo, H., Conejo, N.M., Arias, J.L., Monleon, S., Vinader-Caerols, C., Parra, A., 2008. Changes in brain oxidativemetabolism induced by inhibitory avoidance learning andacute administration of amitriptyline. Pharmacol. Biochem.Behav. 89, 456–462.

Groenewegen, H.J., 1988. Organization of the afferent connectionsof the mediodorsal thalamic nucleus in the rat, related to themediodorsal-prefrontal topography. Neuroscience 24, 379–431.

Groenewegen, H.J., Wright, C.I., Uylings, H.B., 1997. Theanatomical relationships of the prefrontal cortex with limbicstructures and the basal ganglia. J. Psychopharmacol. 11,99–106.

Grzanna, R., Fritschy, J.M., 1991. Efferent projections of differentsubpopulations of central noradrenaline neurons. Prog. BrainRes. 88, 89–101.

Heidbreder, C.A., Groenewegen, H.J., 2003. The medial prefrontalcortex in the rat: evidence for a dorso-ventral distinctionbased upon functional and anatomical characteristics.Neurosci. Biobehav. Rev. 27, 555–579.

Herkenham, M., Nauta, W.J., 1979. Efferent connections of thehabenular nuclei in the rat. J. Comp. Neurol. 187, 19–47.

Hevner, R.F., Liu, S., Wong-Riley, M.T., 1993. An optimized methodfor determining cytochrome oxidase activity in brain tissuehomogenates. J. Neurosci. Methods 50, 309–319.

Hoebel, B.G., Avena, N.M., Rada, P., 2007. Accumbens dopamine-acetylcholine balance in approach and avoidance. Curr. Opin.Pharmacol. 7, 617–627.

Holstege, J.C., Bongers, C.M., 1991. Ultrastructural aspects of thecoeruleo-spinal projection. Prog. Brain Res. 88, 143–156.

Hope, B., Kosofsky, B., Hyman, S.E., Nestler, E.J., 1992. Regulationof immediate early gene expression and AP-1 binding in therat nucleus accumbens by chronic cocaine. Proc. Natl. Acad.Sci. USA 89, 5764–5768.

Hu, D., Xu, X., Gonzalez-Lima, F., 2006. Vicarious trial-and-errorbehavior and hippocampal cytochrome oxidase activityduring Y-maze discrimination learning in the rat. Int. J.Neurosci. 116, 265–280.

Huether, G., 1996. The central adaptation syndrome: psychosocialstress as a trigger for adaptive modifications of brain structureand brain function. Prog. Neurobiol. 48, 569–612.

Hussain, R.J., Taraschenko, O.D., Glick, S.D., 2008. Effects ofnicotine, methamphetamine and cocaine on extracellularlevels of acetylcholine in the interpeduncular nucleus of rats.Neurosci. Lett. 440, 270–274.

Ichikawa, J., Chung, Y.C., Li, Z., Dai, J., Meltzer, H.Y., 2002.Cholinergic modulation of basal and amphetamine-induceddopamine release in rat medial prefrontal cortex and nucleusaccumbens. Brain Res. 958, 176–184.

Jentsch, J.D., Olausson, P., De La Garza 2nd, R., Taylor, J.R., 2002.Impairments of reversal learning and response perseverationafter repeated, intermittent cocaine administrations tomonkeys. Neuropsychopharmacology 26, 183–190.

Jimenez-Rivera, C.A., Feliu-Mojer, M., Vazquez-Torres, R., 2006.Alpha-noradrenergic receptors modulate the developmentand expression of cocaine sensitization. Ann. N.Y. Acad. Sci.1074, 390–402.

Jones, B.E., 1991. Noradrenergic locus coeruleus neurons: theirdistant connections and their relationship to neighboring(including cholinergic and GABAergic) neurons of the centralgray and reticular formation. Prog. Brain Res. 88, 15–30.

Jones, D., Gonzalez-Lima, F., 2001. Associative effects of Pavloviandifferential inhibition of behaviour. Eur. J. Neurosci. 14,1915–1927.

Kalivas, P.W., 2008. Addiction as a pathology in prefrontal corticalregulation of corticostriatal habit circuitry. Neurotox. Res. 14,185–189.

Koch, S., Piercey, M.F., Galloway, M.P., Svensson, K.A., 1997.Interactions between cocaine and (� )-DS 121: studies with

2-deoxyglucose autoradiography and microdialysis in the ratbrain. Eur. J. Pharmacol. 319, 173–180.

Krettek, J.E., Price, J.L., 1977. The cortical projections of themediodorsal nucleus and adjacent thalamic nuclei in the rat.J. Comp. Neurol. 171, 157–191.

Lategan, A.J., Marien, M.R., Colpaert, F.C., 1990. Effects of locuscoeruleus lesions on the release of endogenous dopamine inthe rat nucleus accumbens and caudate nucleus asdetermined by intracerebral microdialysis. Brain Res. 523,134–138.

Liprando, L.A., Miner, L.H., Blakely, R.D., Lewis, D.A., Sesack, S.R.,2004. Ultrastructural interactions between terminalsexpressing the norepinephrine transporter and dopamineneurons in the rat and monkey ventral tegmental area.Synapse 52, 233–244.

London, E.D., Cascella, N.G., Wong, D.F., Phillips, R.L., Dannals, R.F.,Links, J.M., Herning, R., Grayson, R., Jaffe, J.H., Wagner Jr., H.N.,1990. Cocaine-induced reduction of glucose utilization inhuman brain. A study using positron emission tomographyand [fluorine 18]-fluorodeoxyglucose. Arch. Gen. Psychiatry 47,567–574.

Macey, D.J., Rice, W.N., Freedland, C.S., Whitlow, C.T., Porrino, L.J.,2004. Patterns of functional activity associated with cocaineself-administration in the rat change over time.Psychopharmacology (Berl) 172, 384–392.

Mark, G.P., Shabani, S., Dobbs, L.K., Hansen, S.T., 2011. Cholinergicmodulation of mesolimbic dopamine function and reward.Physiol. Behav. 104, 76–81.

Matochik, J.A., London, E.D., Eldreth, D.A., Cadet, J.L., Bolla, K.I.,2003. Frontal cortical tissue composition in abstinent cocaineabusers: a magnetic resonance imaging study. Neuroimage 19,1095–1102.

McIntosh, A.R., Gonzalez-Lima, F., 1994a. Network interactionsamong limbic cortices, basal forebrain, and cerebellumdifferentiate a tone conditioned as a Pavlovian excitor orinhibitor: fluorodeoxyglucose mapping and covariancestructural modeling. J. Neurophysiol. 72, 1717–1733.

McIntosh, A.R., Gonzalez-Lima, F., 1994b. Structural equationmodeling and its application to network analysis in functionalbrain imaging. Hum. Brain Mapp. 2, 2–22.

Mejias-Aponte, C.A., Drouin, C., Aston-Jones, G., 2009. Adrenergicand noradrenergic innervation of the midbrain ventraltegmental area and retrorubral field: prominent inputs frommedullary homeostatic centers. J. Neurosci. 29, 3613–3626.

Nair, H.P., Collisson, T., Gonzalez-Lima, F., 1999. Postnataldevelopment of cytochrome oxidase activity in fiber tracts ofthe rat brain. Brain Res. Dev. Brain Res. 118, 197–203.

Nishikawa, T., Fage, D., Scatton, B., 1986. Evidence for, and natureof, the tonic inhibitory influence of habenulointerpeduncularpathways upon cerebral dopaminergic transmission in the rat.Brain Res. 373, 324–336.

O’Reilly, K.C., Shumake, J., Bailey, S.J., Gonzalez-Lima, F., Lane, M.A.,2009. Chronic 13-cis-retinoic acid administration disruptsnetwork interactions between the raphe nuclei and thehippocampal system in young adult mice. Eur. J. Pharmacol. 605,68–77.

Olausson, P., Jentsch, J.D., Krueger, D.D., Tronson, N.C., Nairn, A.C.,Taylor, J.R., 2007. Orbitofrontal cortex and cognitive-motivationalimpairments in psychostimulant addiction: evidence fromexperiments in the non-human primate. Ann. N.Y. Acad. Sci.1121, 610–638.

Padilla, E., Shumake, J., Barrett, D.W., Sheridan, E.C., Gonzalez-Lima, F., 2011. Mesolimbic effects of the antidepressantfluoxetine in Holtzman rats, a genetic strain with increasedvulnerability to stress. Brain Res. 1387, 71–84.

Paxinos, G., 1995. The Rat Nervous System. Academic Press, SanDiego.

b r a i n r e s e a r c h 1 5 4 2 ( 2 0 1 4 ) 5 6 – 6 968

Paxinos, G., Watson, C., 1986. The Rat Brain in StereotaxicCoordinates. Academic Press, Sydney, Orlando.

Perry, J.L., Joseph, J.E., Jiang, Y., Zimmerman, R.S., Kelly, T.H.,Darna, M., Huettl, P., Dwoskin, L.P., Bardo, M.T., 2011.Prefrontal cortex and drug abuse vulnerability: translation toprevention and treatment interventions. Brain Res. Rev. 65,124–149.

Picciotto, M.R., 1998. Common aspects of the action of nicotineand other drugs of abuse. Drug Alcohol Depend. 51, 165–172.

Pierce, R.C., Bell, K., Duffy, P., Kalivas, P.W., 1996. Repeatedcocaine augments excitatory amino acid transmission in thenucleus accumbens only in rats having developed behavioralsensitization. J. Neurosci. 16, 1550–1560.

Poremba, A., Jones, D., Gonzalez-Lima, F., 1997. Metabolic effectsof blocking tone conditioning on the rat auditory system.Neurobiol. Learn. Mem. 68, 154–171.

Poremba, A., Jones, D., Gonzalez-Lima, F., 1998. Classicalconditioning modifies cytochrome oxidase activity in theauditory system. Eur. J. Neurosci. 10, 3035–3043.

Porrino, L.J., Kornetsky, C., 1988. The effects of cocaine on localcerebral metabolic activity. NIDA Res. Monogr 88, 92–106.

Porrino, L.J., Smith, H.R., Nader, M.A., Beveridge, T.J., 2007. Theeffects of cocaine: a shifting target over the course ofaddiction. Prog. Neuropsychopharmacol. Biol. Psychiatry 31,1593–1600.

Poschel, B.P., Ninteman, F.W., 1963. Norepinephrine: a possibleexcitatory neurohormone of the reward system. Life Sci. 10,782–788.

Post, R.M., 1980. Intermittent versus continuous stimulation:effect of time interval on the development of sensitization ortolerance. Life Sci. 26, 1275–1282.

Puga, F., Barrett, D.W., Bastida, C.C., Gonzalez-Lima, F., 2007.Functional networks underlying latent inhibition learning inthe mouse brain. Neuroimage 38, 171–183.

Ray, J.P., Price, J.L., 1992. The organization of the thalamocorticalconnections of the mediodorsal thalamic nucleus in the rat,related to the ventral forebrain-prefrontal cortex topography.J. Comp. Neurol. 323, 167–197.

Riha, P.D., Rojas, J.C., Gonzalez-Lima, F., 2011. Beneficial networkeffects of methylene blue in an amnestic model. Neuroimage54, 2623–2634.

Robinson, T.E., Berridge, K.C., 1993. The neural basis of drugcraving: an incentive-sensitization theory of addiction. BrainRes. Brain Res. Rev. 18, 247–291.

Robinson, T.E., Berridge, K.C., 2001. Incentive-sensitization andaddiction. Addiction 96, 103–114.

Rojas, J.C., Bruchey, A.K., Gonzalez-Lima, F., 2012. Low-level lighttherapy improves cortical metabolic capacity and memoryretention. J. Alzheimers Dis. 32, 741–752.

Sakata, J.T., Coomber, P., Gonzalez-Lima, F., Crews, D., 2000.Functional connectivity among limbic brain areas: differentialeffects of incubation temperature and gonadal sex in theleopard gecko, Eublepharis macularius. Brain Behav. Evol. 55,139–151.

Sakata, J.T., Crews, D., Gonzalez-Lima, F., 2005. Behavioralcorrelates of differences in neural metabolic capacity. BrainRes. Brain Res. Rev. 48, 1–15.

Salas, R., Cook, K.D., Bassetto, L., De Biasi, M., 2004. The alpha3and beta4 nicotinic acetylcholine receptor subunits arenecessary for nicotine-induced seizures and hypolocomotionin mice. Neuropharmacology 47, 401–407.

Salas, R., Sturm, R., Boulter, J., De Biasi, M., 2009. Nicotinicreceptors in the habenulo-interpeduncular system arenecessary for nicotine withdrawal in mice. J. Neurosci. 29,3014–3018.

Santos-Vera, B., Vazquez-Torres, R., Marrero, H.G., Acevedo, J.M.,Arencibia-Albite, F., Velez-Hernandez, M.E., Miranda, J.D.,Jimenez-Rivera, C.A., 2013. Cocaine sensitization increases I h

current channel subunit 2 (HCN 2) protein expression instructures of the mesocorticolimbic system. J. Mol. Neurosci.50, 234–245.

Schilman, E.A., Uylings, H.B., Galis-de Graaf, Y., Joel, D.,Groenewegen, H.J., 2008. The orbital cortex in ratstopographically projects to central parts of the caudate-putamen complex. Neurosci. Lett. 432, 40–45.

Schoenbaum, G., Shaham, Y., 2008. The role of orbitofrontalcortex in drug addiction: a review of preclinical studies. Biol.Psychiatry 63, 256–262.

Schultz, W., 1998. Predictive reward signal of dopamine neurons.J. Neurophysiol. 80, 1–27.

See, R.E., McLaughlin, J., Fuchs, R.A., 2003. Muscarinic receptorantagonism in the basolateral amygdala blocks acquisition ofcocaine-stimulus association in a model of relapse to cocaine-seeking behavior in rats. Neuroscience 117, 477–483.

Shannon, H.E., Peters, S.C., 1990. A comparison of the effects ofcholinergic and dopaminergic agents on scopolamine-inducedhyperactivity in mice. J. Pharmacol. Exp. Ther. 255, 549–553.

Shumake, J., Gonzalez-Lima, F., 2003. Brain systems underlyingsusceptibility to helplessness and depression. Behav. Cogn.Neurosci. Rev. 2, 198–221.

Stein, L., 1964. Self-stimulation of the brain and the centralstimulant action of amphetamine. Fed. Proc. 23, 836–850.

Stein, L., 1975. Norepinephrine reward pathways: role of self-stimulation, memory consolidation, and schizophrenia. Nebr.Symp. Motiv. 22, 113–159.

Steiner, H., Gerfen, C.R., 1993. Cocaine-induced c-fos messengerRNA is inversely related to dynorphin expression in striatum.J. Neurosci. 13, 5066–5081.

Stewart, J., Badiani, A., 1993. Tolerance and sensitization to thebehavioral effects of drugs. Behav. Pharmacol. 4, 289–312.

Sutherland, R.J., Nakajima, S., 1981. Self-stimulation of thehabenular complex in the rat. J. Comp. Physiol. Psychol. 95,781–791.

Sved, A.F., Cano, G., Card, J.P., 2001. Neuroanatomical specificityof the circuits controlling sympathetic outflow to differenttargets. Clin. Exp. Pharmacol. Physiol. 28, 115–119.

Tapper, A.R., McKinney, S.L., Nashmi, R., Schwarz, J., Deshpande, P.,Labarca, C., Whiteaker, P., Marks, M.J., Collins, A.C., Lester, H.A.,2004. Nicotine activation of alpha4n receptors: sufficient forreward, tolerance, and sensitization. Science 306, 1029–1032.

Taraschenko, O.D., Rubbinaccio, H.Y., Shulan, J.M., Glick, S.D.,Maisonneuve, I.M., 2007. Morphine-induced changes inacetylcholine release in the interpeduncular nucleus andrelationship to changes in motor behavior in rats.Neuropharmacology 53, 18–26.

Todtenkopf, M.S., Mihalakopoulos, A., Stellar, J.R., 2002.Withdrawal duration differentially affects c-fos expression inthe medial prefrontal cortex and discrete subregions of thenucleus accumbens in cocaine-sensitized rats. Neuroscience114, 1061–1069.

Ungerstedt, U., 1971. Stereotaxic mapping of the monoaminepathways in the rat brain. Acta Physiol. Scand. 367 (Suppl.),1–48.

Ungless, M.A., Whistler, J.L., Malenka, R.C., Bonci, A., 2001. Singlecocaine exposure in vivo induces long-term potentiation indopamine neurons. Nature 411, 583–587.

Velasquez-Martinez, M.C., Vazquez-Torres, R., Jimenez-Rivera, C.A.,2012. Activation of alpha1-adrenoceptors enhances glutamaterelease onto ventral tegmental area dopamine cells.Neuroscience 216, 18–30.

Villani, L., Contestabile, A., Fonnum, F., 1983. Autoradiographiclabeling of the cholinergic habenulo-interpeduncularprojection. Neurosci. Lett. 42, 261–266.

Villarreal, J.S., Gonzalez-Lima, F., Berndt, J., Barea-Rodriguez, E.J.,2002. Water maze training in aged rats: effects on brainmetabolic capacity and behavior. Brain Res. 939, 43–51.

b r a i n r e s e a r c h 1 5 4 2 ( 2 0 1 4 ) 5 6 – 6 9 69

Volkow, N.D., Fowler, J.S., Wolf, A.P., Hitzemann, R., Dewey, S.,Bendriem, B., Alpert, R., Hoff, A., 1991. Changes in brainglucose metabolism in cocaine dependence and withdrawal.Am. J. Psychiatry 148, 621–626.

Volkow, N.D., Mullani, N., Gould, K.L., Adler, S., Krajewski, K.,1988. Cerebral blood flow in chronic cocaine users: a studywith positron emission tomography. Br. J. Psychiatry 152,641–648.

Vorel, S.R., Liu, X., Hayes, R.J., Spector, J.A., Gardner, E.L., 2001.Relapse to cocaine-seeking after hippocampal theta burststimulation. Science 292, 1175–1178.

Watabe-Uchida, M., Zhu, L., Ogawa, S.K., Vamanrao, A., Uchida,N., 2012. Whole-brain mapping of direct inputs to midbraindopamine neurons. Neuron 74, 858–873.

Westlund, K.N., Zhang, D., Carlton, S.M., Sorkin, L.S., Willis, W.D.,1991. Noradrenergic innervation of somatosensory thalamusand spinal cord. Prog. Brain Res. 88, 77–88.

Winstanley, C.A., 2007. The orbitofrontal cortex, impulsivity, andaddiction: probing orbitofrontal dysfunction at the neural,neurochemical, and molecular level. Ann. N.Y. Acad. Sci. 1121,639–655.

Wise, R.A., 1978. Catecholamine theories of reward: a criticalreview. Brain Res. 152, 215–247.

Wise, R.A., Bozarth, M.A., 1987. A psychomotor stimulant theoryof addiction. Psychol. Rev. 94, 469–492.

Wong-Riley, M.T., 1989. Cytochrome oxidase: an endogenousmetabolic marker for neuronal activity. Trends Neurosci. 12,94–101.

Wong-Riley, M.T., Huang, Z., Liebl, W., Nie, F., Xu, H., Zhang, C.,1998. Neurochemical organization of the macaque retina:effect of TTX on levels and gene expression of cytochromeoxidase and nitric oxide synthase and on theimmunoreactivity of Naþ Kþ ATPase and NMDA receptorsubunit I. Vision Res. 38, 1455–1477.

Yoon, H.S., Kim, S., Park, H.K., Kim, J.H., 2007. Microinjection ofCART peptide 55–102 into the nucleus accumbens blocks boththe expression of behavioral sensitization and ERKphosphorylation by cocaine. Neuropharmacology 53, 344–351.

Zachariou, V., Caldarone, B.J., Weathers-Lowin, A., George, T.P.,Elsworth, J.D., Roth, R.H., Changeux, J.P., Picciotto, M.R., 2001.Nicotine receptor inactivation decreases sensitivity to cocaine.Neuropsychopharmacology 24, 576–589.

Zametkin, A.J., Nordahl, T.E., Gross, M., King, A.C., Semple, W.E.,Rumsey, J., Hamburger, S., Cohen, R.M., 1990. Cerebral glucosemetabolism in adults with hyperactivity of childhood onset.N. Engl. J. Med. 323, 1361–1366.