Molecular and Cellular Neuroscience 17, 303–316 (2001)
doi:10.1006/mcne.2000.0942, available online at http://www.idealibrary.com on MCN
A
Cocaine-Induced Expression of the TetraspaninCD81 and Its Relation to Hypothalamic Function
Maria S. Brenz Verca,* David A. J. Widmer,*George C. Wagner,† and Jean-Luc Dreyer* ,1
*Institute of Biochemistry, University of Fribourg, CH-1700 Fribourg, Switzerland; and†Department of Psychology, Rutgers University, New Brunswick, New Jersey 08903
CD81, a tetraspanin transmembrane protein involved incell adhesion, was found by differential display to be up-regulated in the nucleus accumbens of rat brain followingacute cocaine treatment (four injections of 30 mg/kg ev-ery 2 h followed by 24 h withdrawal). Cocaine-inducedexpression of CD81 in adult rat brain was confirmed byquantitative real-time RT-PCR. Its expression in neuronsand its function in the brain are unknown. In situ hybrid-zation shows a neuron-specific expression pattern inrain regions functionally related to the regulation of car-iovascular function and fluid homeostasis. CD81 dis-lays codistribution to galanin and, to a lesser extent, toasopressin. These findings add to data that suggest aonnection between the brain reward pathway and theenters regulating endocrine and autonomic functions, inelation to neurochemical, behavioral, and somatic con-equences of drug abuse.
INTRODUCTION
CD81 belongs to the tetraspanin family and containsfour transmembrane domains with intracellular C- andN-terminal regions and one large extracellular loop.Tetraspanins form a multiprotein net on the cell surfacethat modulates integrin signaling and favors the forma-tion of complexes of extracellular matrix proteins, li-gands, and receptors, thus facilitating particular cellularresponses to external stimuli (Maecker et al., 1997;Yanez-Mo et al., 1998; Berditchevski and Odintsova,1999).
1 To whom correspondence should be addressed at the Institute ofiochemistry, University of Fribourg, Rue du Musee 5, CH-1700
CD81, or TAPA-12, was first identified as the target ofan antiproliferative antibody to a B-lymphoma cell line(Oren et al., 1990). Subsequent studies showed thatCD81 is involved in a large number of cellular func-tions, mostly established for cells of the immune sys-tem, including the regulation of cell shape, motility, orgrowth (Lin et al., 1992; Boismenu et al., 1996). CD81 ispart of a complex on B lymphocytes, which includesCD19, CD21, and Leu13. This complex amplifies signaltransduction through membrane-bound immunoglobu-lin, enabling B cells to respond to low concentrations ofantigen by homotypic cellular aggregation. The CD81component brings together independently functioningsubunits, probably through integrin activation (Matsu-moto et al., 1993). Many tetraspanin multimolecularcomplexes contain a subset of integrins, but only CD81and CD151 interact directly with integrins (Serru et al.,1999). CD81 associates with integrins a3b1, a4b1, anda6b1 (Mannion et al., 1996), a functional relationshipthat may be cell type-specific (Behr and Schriever, 1995;Maecker et al., 1997). CD81 also associates with MHCclass II molecules in human B cells (Angelisova et al.,1994), together with CD82, CD4, and CD8 (Imai andYoshie, 1993). This provides a costimulatory signal withCD3 on human thymocytes (Todd et al., 1996; Lagaud-riere-Gesbert et al., 1997). Recent studies found that
2 Abbreviations used: AD, anterodorsal thalamic nucleus; AVP,Arg-vasopressin; BST, bed nucleus of the stria terminalis; CART,cocaine- and amphetamine-regulated transcript; ChP, choroid plexus;CRF, corticotropin-releasing factor; DA, dopamine; DB, nucleus of thediagonal band of Broca; DIG, digoxigenin; Gal, galanin; Hipp, hip-pocampus; NAcc, nucleus accumbens; TO, oxytocin; PBS, phosphate-buffered saline; PVH, paraventricular hypothalamic nucleus; SN, sub-stantia nigra; SO, supraoptic nucleus; SSC, saline sodium citrate;
TAPA-1, target of the antiproliferative antibody-1; TH, tyrosine hy-droxylase; VTA, ventral tegmental area.
303
roOdatticsPaatdt
304 Brenz Verca et al.
FIG. 1. (A) Map of CD81 and positions of fragments used in this study. MD3 and TAPA are the mouse and rat CD81 mRNA sequences,respectively, as designated in GenBank. Fragments are presented as submitted to GenBank, but both are incomplete, although full-lengthmRNAs from both species most probably have the same length and structure. Homology is shown between coding sequences of both mRNAs.The fragment P51 was found in the differential display study, with the larger part homologous to rat and mouse 39-untranslated region (UTR)and the smaller one homologous only to mouse 39-UTR, since the sequence for rat 39-UTR is not yet available in GenBank. U and L showespectively the positions of the upper and lower primers used for end-point quantitative RT-PCR. F and R respectively indicate the positionsf forward and reverse primers used for real-time PCR. The in situ probe was made from a clone containing the coding sequence of mouse CD81.RF, open reading frame. (B) Differential display of mRNAs isolated from three brain regions of control or cocaine-treated rats (see text foretails). s, saline; c, cocaine; H, hippocampus; A, nucleus accumbens; LS, lateral striatum. The band corresponding to CD81 is indicated by anrrow. (C) End-point quantitative RT-PCR with CD81-specific primers on cDNA produced from mRNA of saline- or cocaine-treated rat accumbalissue (sAcc and cAcc, respectively). Four serial dilutions of cDNA were used (from 13 to 10003), and the PCR products were loaded on gelogether with 100 bp ladder. The CD81 amplified band of 243 bp disappears when PCR is performed on 1003 sAcc dilution, while is still presentn cAcc, indicating upregulation of CD81 upon cocaine treatment in NAcc. (D) Real-time quantitative RT-PCR with CD81-specific primers onDNA produced from NAcc, tegmentum, or hippocampus. Total RNA was extracted, using an RNAqueous kit (Ambion), from the NAcc ofaline- or cocaine-treated rats. Two independent RT reactions were performed and the cDNA templates used for PCR. Real-time quantitativeCR was carried out on an iCycler (Bio-Rad) using the SYBR green method, with appropriate primers for CD81 and for 28S rRNA (as a control),s described under Experimental Methods. Up to five independent PCRs were performed for each condition. The threshold cycle was defineds the fractional cycle number at which the fluorescence passed the fixed threshold. The measures of relative fluorescence were made during thehreshold cycle, determined for each PCR. The measures for CD81 were normalized using the numbers for 28S rRNA. The average with standard
eviation is presented. Although the comparison of absolute values between different regions is not possible, these results clearly demonstrate
he CD81 upregulation in the cocaine-treated NAcc, as opposed to tegmentum and hippocampus.
sCADnmns ; SO,n utamc pal
305Cocaine-Induced CD81 and Hypothalamic Function
human CD81 binds to hepatitis C virus via the envelope
FIG. 2. In situ hybridization of CD81 mRNA distribution throughections hybridized with DIG-labeled CD81 probe. (A–H) Sections aD81 probe are marked on the right side. For orientation, some unstarc, arcuate nucleus; BST, bed nucleus of the stria terminalis; CG, cenEn, dorsal endopiriform nucleus; DG, dentate gyrus; Ep, ependymucleus; Hipp, hippocampal formation; LOT, nucleus of the lateral oledian preoptic hypothalamic nucleus; MPO, medial preoptic hyp
protein E2 and is, therefore, a possible key molecule invirus pathogenicity (Pileri et al., 1998). Anti-CD81
en
blocks T cell maturation (Boismenu et al., 1996); how-
e adult rat brain. Photomicrographs are shown for coronal 25-mmown in rostral-to-caudal direction. Regions specifically labeled withregions are indicated on sections. AD, anterodorsal thalamic nucleus;ray; ChP, choroid plexus; DB, nucleus of the diagonal band of Broca;yer of the ventricles; IG, induseum griseum; IPR, interpeduncular
supraoptic nucleus; TT, tenia tecta; VMH, ventromedial hypothalamicen; fmj, forceps major of the corpus callosum; gcc, genu of the corpuscommissure.
out thre shinedtral g
al lafactoothaVP, p
ver, in CD81-deficient mice the thymocytes developormally (Tsitsikov et al., 1997). These mice show, in
1p(iutrGo
pptttCnnlTp
lexusells b
306 Brenz Verca et al.
general, a subtle phenotype, suggesting the extensivecompensatory processes that occur during develop-ment (Levy et al., 1998).
CD81 is expressed in virtually all types of cells, withthe exception of red blood cells and platelets (Levy et al.,998), and has been thoroughly characterized in lym-hocytes. In the brain CD81 is expressed by glial cells
Sullivan and Geisert, 1998), anti-CD81 causing changesn astrocyte morphology (Geisert et al., 1996). CD81 ispregulated during early postnatal development, at the
ime of glial birth and maturation, and in the case ofeactive gliosis (Irwin and Geisert, 1993; Sullivan and
FIG. 3. CD81 in situ hybridization in the ependymal layer of the ventSeveral layers of CD81-positive ependymal cells are indicated by an arhypothalamus. Inset magnified in (C): The ependymal lining of the thirdCD81-negative zones is indicated by arrow. (D) CD81 signal in choroid pthe ventricular cavity are indicated by arrows, internal choroid plexus c
eisert, 1998). This upregulation is evidence for a rolef tetraspanins in brain plasticity.
wr
In this study we have performed a differential dis-lay screening to identify molecular cues induced bysychomotor stimulants that revealed CD81 upregula-
ion in the nucleus accumbens (NAcc) upon cocainereatment. We have confirmed this finding by quanti-ative RT-PCR and undertaken a detailed description ofD81 distribution in rat brain. This shows that CD81 isot exclusively glial and is expressed in selective brainuclei, mainly confined to regions important for regu-
ation of cardiovascular function and fluid homeostasis.he pattern of CD81 localization is consistent with autative role in relation to effects observed after drug
and choroid plexus. (A) The ependymal lining of the lateral ventricle.(B) The ependymal lining of the third ventricle at the level of anteriorricle. The border between CD81-positive chain-like ependymal cells and. (E) Nissl staining in choroid plexus of section adjacent to D. Cells facingy dotted arrows. Scale bars, 50 mm (in A, D, E) and 25 mm (in C).
riclesrow.vent
ithdrawal. This adds to data linking together the neu-ochemical, behavioral, and somatic consequences of
et(
fNoirm
ined
307Cocaine-Induced CD81 and Hypothalamic Function
drug abuse. It provides additional evidence for a rela-tion between the primary targets of psychostimulantsin the brain reward circuits and centers regulating au-tonomic or neuroendocrine functions.
RESULTS
CD81 Upregulation in Nucleus Accumbens
To examine the changes in gene expression in thedopaminergic brain system upon acute cocaine treat-
FIG. 4. CD81 in situ hybridization in the olfactory bulb (A) and theIGr, internal granular layer; IPl, internal plexiform layer; Mi, mitral cethe bigger diffusely distributed cells of layers IV–V are marked withBroca. (C) Low magnification; (D) high magnification. (E) A Nissl-stain both stainings. Scale bars, 250 mm (in A–C) and 25 mm (in D, E).
ment, we used the method of differential display (Liangand Pardee, 1992) according to the modifications de-
sr
scribed recently (Brenz Verca et al., 1998). Patterns ofxpression were compared for mRNAs isolated fromhe NAcc, the lateral striatum, and the hippocampusHipp) (as control) of saline- or cocaine-treated rats.
Among differentially expressed tags one cDNA wasound to be induced upon cocaine treatment in theAcc (Fig. 1B). The candidate was not detectable in
ther parts, neither in the NAcc of control animals norn any other brain region examined. The result waseproduced in triplicate differential display experi-ents starting from three independent reverse tran-
x (B) and the diagonal band of Broca. EPl, external plexiform layer;er. The small densely packed cells of layer II are indicated by arrows;ed arrows. (C–E) CD81 in situ hybridization in the diagonal band ofadjacent section. Big cells positive for CD81 are indicated by arrows
cortell laydott
cription (RT) reactions. The band was cut out andeamplified with the set of universal reamplification
308 Brenz Verca et al.
primers described in the protocol (Brenz Verca et al.,1998) and the PCR product was sequenced directly,yielding a sequence with 98% homology with theTAPA-1 (Geisert et al., 1996), rat CD81-homologous se-quence.
To further assess the induction of CD81, primerswere designed for quantitative PCR based on rat(TAPA-1) and mouse (MD3) CD81 homologous se-quences in GenBank (Fig. 1A). Serial 10-fold dilutions ofRT product from mRNA of saline- or cocaine-treatedrats were amplified with these CD81-specific primers.As a control, the product of the RT reaction was usedbut without RTase added. The expected band of 243 bpwas found in undiluted or 10-fold-diluted cDNAs fromboth treated and untreated animals; however, the signalfrom 100-fold-diluted cDNA was evident only for co-caine-treated brain, suggesting an increased quantity ofCD81 mRNA in this region compared to control (Fig.1C). At the 1000-fold dilution, the band was absentunder both conditions.
The quantification of RT-PCR end-product does notallow for precise evaluation of concentration differ-ences between the samples. Therefore the real-timeRT-PCR was used to confirm the upregulation ofCD81 mRNA in the NAcc, which allows for quanti-fication of PCR product during the exponential phaseof reaction. Primers were chosen based on the rat(TAPA-1) CD81-homologous sequence in GenBank(Fig. 1A). Quantitative PCR was performed in quin-tuplicate from two different RT reactions and 28SrRNA was used as a normalization standard in allexperiments. As shown in Fig. 1D, CD81 is upregu-lated after cocaine treatment and its expression raises4.6-fold in the NAcc. No changes in CD81 expressionwere found in the Hipp and the tegmentum. Controlsin which RTase had been omitted or in which just oneprimer had been used in the PCR were always neg-ative. These data clearly indicate that CD81 upregu-lation is specific for certain brain regions and thatCD81 may play a major role in the response tococaine.
Localization of CD81 mRNA in the Rat Brain
We performed in situ hybridization with a riboprobecorresponding to the coding sequence of CD81 mRNA(Fig. 1A). Controls with sense riboprobe consistentlyyielded no signals. CD81 distribution throughout thebrain is shown in Fig. 2 on coronal sections in a rostral-to-caudal direction.
In general CD81 expression is abundant in typicallyglial structures, confirming the immunocytochemical
observations made by Geisert and colleagues (Sullivanand Geisert, 1998). The glia limitans, the layer of astro-cytic cells lying beneath the pial surface, was positivefor CD81 (not shown). However, only the minor part ofthese cells was stained, as could be seen by comparisonwith Nissl-stained consequent sections (not shown).The same was true for the ependymal lining of ventri-cles. In many cases one or more layers of ependymalcells were stained at the inner surface of the ventricles(Figs. 3A and 3B). However, we have found some ex-ceptions to this general rule. As can be clearly seen inFigs. 3B and 3C, the extensive staining is present in theventral part of the third ventricle where the tightlyjoined, labeled ependymal cells form a chain-like struc-ture, though in the more dorsal part of the ventricle, thestaining disappears completely. We also found theCD81 signal in circumventrical organs, namely in thechoroid plexus (ChP) (Fig. 2E) and the subfornical or-gan (Fig. 2D). While comparing the two subsequentsections, stained for CD81 mRNA and Nissl substance,respectively (Figs. 3D and 3E), we observed that thestaining for CD81 in ChP was especially strong for theexternal layer of cells facing the ventricular cavity filledwith cerebrospinal fluid. However, the internal cells ofthe ChP were unstained for CD81.
In addition to these purely glial structures, we couldfind other positively stained regions, not described thusfar, where the staining was not equivocally confined toglial or neuronal cells.
Olfactory areas. In the olfactory bulb, the internalgranular layer was most extensively labeled with CD81,whereas both the external and the internal plexiformlayers as well as the mitral cell layer were moderatelystained (Fig. 4A). Of all other regions related to theolfactory system, the piriform cortex and the induseumgriseum were the most intensively labeled (see, forexample, Fig. 2A). Less CD81 staining was found in theanterior olfactory nucleus (not shown), tenia tecta (Fig.2A), dorsal endopiriform nucleus (Fig. 2B), and nucleusof the lateral olfactory tract (Fig. 2E), and a very faintsignal was seen in the olfactory tubercle.
Cortex. The parietal cortex was the least intensivelymarked zone, compared to the other cortical regions(see, for example, Fig. 2B). At the cellular level, CD81expression was absent in cortical layer I, most pro-nounced in layer II with small, quite densely packagedcells, and then equally distributed in lower corticallayers, where the cells were in general larger and morediffusely distributed (Fig. 4B).
Septum. The posterior septum was not labeled with
CD81 probe. The lateral septum was moderately la-beled throughout, including its dorsal, intermediate,
309Cocaine-Induced CD81 and Hypothalamic Function
and ventral parts (Fig. 2C). However, the most specificCD81 labeling was found in the medial septal division.The nucleus of the diagonal band of Broca (DB) wasextensively labeled (Figs. 2B and 5C). Only a very spe-cific restricted subset of cells was positive for CD81, asseen by comparison with a Nissl-stained subsequentsection (Figs. 4D and 4E). Nissl staining clearly shows atleast two different cell types, small and large ones (Fig.4E), and CD81 staining is almost exclusively confined tolarge cells, probably cholinergic and/or peptidergicneurons. Small cells are either GABAergic interneuronsor astrocytes and are CD81-negative, so it is very im-probable that expression is confined to glial cells in thatregion.
Hippocampus. Moderate CD81 staining was foundin the pyramidal cell layer of the hippocampal forma-tion, as well as in the granule cell layer of dentate gyrus(Fig. 2G).
Amygdala and extended amygdala. Moderate CD81expression was found in the bed nucleus of the striaterminalis (BST) (Fig. 2D) and in the central amygdaloidnucleus (Fig. 2F). Other amygdala subdivisions con-tained very few CD81 message-positive cells.
Thalamus. Faint CD81 staining of mainly medium-sized neurons was found in medial habenular (Fig. 2G),paraventricular (Figs. 2E–2G), and reuniens (notshown) thalamic nuclei. However, more specific andextensive labeling was in the anterodorsal nucleus (AD)(Figs. 2F and 6A). These neurons were bigger androunded and contained large nuclei (Fig. 5C). Interest-ingly, the AD contains the highest density of large-sizeneurons compared to other thalamic subregions. As aresult of this morphological particularity the borders ofthe nucleus can be easily appreciated in Nissl-stainedsections even at low magnification (Fig. 5B). Some cellsshow a specific triangular pattern of CD81 labeling,possibly reflecting mRNA concentration in a region ofthe axon hillock (Fig. 5C).
Hypothalamus. In the preoptic regions, the medianpreoptic nucleus was highly positive for CD81 mRNAlabeling (Fig. 2C). Both the medial preoptic area and themedial preoptic nucleus were moderately stained (Figs.2C–2E). In the periventricular zone, the suprachias-matic nucleus was moderately labeled (Fig. 2C), andarcuate (Fig. 2G) and paraventricular (PVH) (Figs. 2Fand 7B) nuclei were both very extensively stained. Thespecificity was pronounced in PVH with much higherlabeling in the lateral magnocellular part compared tomedial parvocellular and ventral parts (Fig. 6B). Themassive CD81 staining in the lateral part is probably
confined to magnocellular neurosecretory cells (Fig.6D). Again some cells show the triangular pattern of
CD81 labeling in the area of the axon hillock. In thesupraoptic nucleus (SO), expression is also probablyconfined to magnocellular neurosecretory cells (Figs.7A and 7B). In the tuberal region, the ventromedialnucleus is slightly labeled (Fig. 2F). In the mammillaryregion, the supramammillary and lateral mammillarynuclei are slightly labeled (not shown). In general, dis-perse staining was present throughout the entire hypo-thalamus.
Brain stem. Several regions were moderately la-beled in brain stem, including the central gray area, theinterpeduncular nucleus, the substantia nigra (SN), andthe linear raphe nucleus (Fig. 2H).
Colocalization of CD81 and Tyrosine HydroxylaseMessages in Rat Brain as Shownby in Situ Hybridization
The modified expression of CD81 message followingcocaine administration and in regions related to dopa-minergic brain system suggested that, under normalconditions, CD81 might be localized in dopaminergicneurons. Therefore we compared the CD81 distributionwith one for tyrosine hydroxylase (TH) in subsequentbrain sections. Both CD81 and TH staining were oftenfound in the same brain regions, including olfactorybulb, hypothalamic nuclei, and SN. However, TH ismore specific and touches only very limited subgroupsof cells, for example in SO (Figs. 7B and 7C) and SN(data not shown). Without double-labeling experi-ments, it cannot be said as a general rule that bothmessages are colocalized, especially with our observa-tions that CD81 is not at all expressed in the A10 andA11 groups of dopaminergic cells, nor is TH expressedin many CD81-positive areas.
CART Overexpression in the Nucleus Accumbensof Cocaine-Treated Rat Brain
The CART (cocaine and amphetamine-regulatedtranscript) peptide is a putative neurotransmitter andtrophic factor. CART is one of the messages known tobe overregulated in the striatum upon acute psycho-stimulant application. More precise localization andfunctional significance of such overexpression have yetto be demonstrated. However, such a signal could bestill present 24 h after acute cocaine treatment, when theexpression of acutely induced, immediate early genes isalready terminated. Thus, we decided to use the CARTprobe as a positive control of cocaine-induced gene
expression in our in situ hybridization studies. As ex-pected, CART message was overexpressed in the stria-
310 Brenz Verca et al.
tum 24 h after acute cocaine (data not shown). More-over, we have shown that this overexpression isspecifically confined to the ventral part of striatum, orthe NAcc. The signal seemed to be particularly strong inthe shell of the NAcc; however, this statement needsadditional experimental support. Concerning the gen-
FIG. 5. CD81 in situ hybridization in the anterodorsal nucleus of thethalamus. (A) Low magnification; (B) an adjacent Nissl-stained sec-tion; (C) CD81 staining at high magnification. Staining of axonhillocks is indicated by arrows. AD, anterodorsal thalamic nucleus;sm, stria medullaris of the thalamus. Scale bars, 250 mm (in A, B) and25 mm (in C).
eral expression of CART message in rat brain, we ob-served localization similar to that found in previously
published results (Douglass et al., 1995), namely in ol-factory areas, PVH (Fig. 6C), and SO, but we could notfind any signal in the central amygdala region (Fager-gren and Hurd, 1999).
CD81 and CART Colocalization in Brain
Because both CD81 and CART were found to beoverexpressed upon cocaine treatment in the NAcc, itwas interesting to compare their general distributionpatterns in adult rat brain. In fact, we could observeboth signals in different olfactory regions, includingtenia tecta, induseum griseum, anterior olfactory nu-cleus, and piriform cortex. In addition, we found bothmessages in particular hypothalamic regions, includingthe PVH, arcuate, and SO nuclei, which are highlyheterogeneous at the cellular level, even though theyhave a common function in the regulation of endocrineand autonomic functions. However, whereas CD81 lo-calization was primarily confined to the magnocellularpart of the PVH (Fig. 6B), CART signal was preferen-tially concentrated in the parvocellular region (Fig. 6C).Another striking difference in hybridization signal dis-tribution of the two probes was in the basal gangliaregion. Whereas CART mRNA is found there in veryhigh concentration, especially in the NAcc, ventral pal-lidum, and globus pallidus, CD81 was not expressed inany of these regions.
DISCUSSION
CD81 Is Involved in Cocaine Action
In the present study we have shown the upregulationof CD81 in the NAcc following cocaine treatment. Theexpression was region-specific and not observed inother regions examined (lateral striatum, tegmentum,and Hipp). The absence of CD81 induction in the teg-mentum suggests that the effect is mediated by cocaineaction on postsynaptic part of the ventral tegmentalarea (VTA)–NAcc axis, probably involving dopamine(DA) receptors in the NAcc. The DA hypothesis ofaction is fortified by the fact that DA and its metabolitelevels are increased specifically in the NAcc of CD81-deficient mice, as opposed to other neurotransmitters(Michna et al., 1999; submitted for publication).
Although we found CD81 in classical glial structures(ChP, ependyma, glia limitans) in accordance with pre-
viously published data (Sullivan and Geisert, 1998), it isnot exclusively glial, as many other groups of CD81-
ocellu
311Cocaine-Induced CD81 and Hypothalamic Function
positive cells in the PVH, SO, or DB show a typicallyneuronal morphology.
CD81 Is Associated with Circuits ControllingCardiovascular Responses and Fluid Homeostasis
Our study shows high expression of CD81 in braincircuits involved in cardiovascular responses and fluidhomeostasis. These circuits consist of two parts: (a) thesensory part composed of ChP, subfornical organ, vas-
FIG. 6. CD81 in situ hybridization in the paraventricular hypothalaof anterior hypothalamus, with inset represented in CD81 (B) and CAmagnification. PaLM, lateral magnocellular part; PaMP, medial parvand 25 mm (in D).
cular organ of lamina terminalis, and median preopticnucleus and containing receptors for insulin, vasopres-
sin (AVP), atrial natriuretic factor, and angiotensin IIand (b) the motor or effector part composed of themagnocellular cells of PVH and SO in hypothalamussecreting AVP and oxytocin (TO), two hormones influ-encing body fluid homeostasis through regulation ofblood pressure as well as through water retention bythe kidney (Tohyama and Takatsuji, 1998). In thesehypothalamic nuclei, but also in other AVP-positiveareas (the suprachiasmatic nucleus, DB, BST, and Hipp)or in regions projecting to PVH and SO (DB, BST, and
ucleus. (A) Schematic view of coronal rat brain section in the regionC) hybridized micrographs. (D) CD81-positive cells in PaLM at highlar part; PaV, ventral part. Scale bars, 100 mm (in B), 250 mm (in C),
mic nRT (
medial preoptic nucleus), CD81 is expressed at higherlevels. Together these results suggest a functional role
Cdgh(
O
CS
HT
A
H
B
arl
312 Brenz Verca et al.
for CD81 in the neuroendocrine regulation of cardio-vascular and fluid homeostasis.
CD81 Is Associated with Galanin Pathways
We found striking similarities in brain localization ofCD81 and galanin (Gal) (Table 1). CD81 and Gal colo-calization specificity is maintained within the same re-gion: in PVH, for example, both are strongly expressedin the lateral part and weakly in the medial and ventralparts (see Table 1). The specificity was striking in thethalamus where high expression of both Gal and CD81is found in only one thalamic nucleus, AD of anteriorgroup, a part of limbic Papez’s circuit (Heimer, 1995).AD inhibits the hypophysoadrenal system under basaland acute stress conditions (Suarez et al., 1998), againproviding a link of CD81 to neuroendocrine function.
From our study it is difficult to say if CD81 and Galare colocalized at the cellular level, without double-labeling experiments. However, the expression of CD81in a subset of DB and AD neurons with the character-istic size of peptidergic neurons strongly favors thishypothesis.
Gal modulates the dopaminergic function. Direct in-teraction between Gal and the DA pathway leads torapid increase of prolactin and growth hormone secre-tion following intraventricular injections of Gal (Me-lander et al., 1987). Gal release from noradrenergic neu-rons of locus coeruleus (Hokfelt et al., 1999) inhibits theactivity of VTA dopaminergic cell bodies, thus indi-rectly interfering with the reward pathway. This resultsin two of the principal symptoms seen in depression,decreased motor activation and anhedonia (Weiss et al.,1998). The codistribution of CD81 and Gal indicatesfunctional relations between the two molecules andcould explain CD81 involvement in drug addiction.CD81, for example, may provide a link between the firstsite of cocaine action (NAcc) and the Gal pathway,ultimately altering dopaminergic responses and influ-encing the reward pathway. The involvement of CD81in long-lasting cocaine effect is supported by the obser-vation of modified cocaine-elicited behavior in CD81-deficient mice (Michna et al., 1999; submitted for publi-cation).
Cocaine Interacts with Neuroendocrine Function
The cocaine-induced expression of CD81 and its pu-tative role in neuroendocrine function provide addi-tional evidence of an interaction between hormonal
function and the actions of drugs of abuse (Torres andRivier, 1992). The best examples of such interaction
iM
include reciprocal cocaine regulation of the HPA axisand OT–AVP system.
Cocaine induces changes in the HPA axis, often in“binge” protocols very similar to the one used in thepresent study. Acute binge cocaine alters hypothalamicand extrahypothalamic/limbic corticotropin-releasingfactor (CRF) concentrations in rats (Sarnyai, 1998), lead-ing to the release of adrenocorticotropic hormone, b-en-dorphin, and corticosterone and activation of the HPAaxis (Moldow and Fischman, 1987; Zhou et al., 1999).
hronic binge cocaine administration is coupled to CRFecrease in the hypothalamus and activation of CRFene expression in extrahypothalamic regions, whichas implications in the behavioral responses to cocaineZhou et al., 1996). Finally, cocaine withdrawal elicits
TABLE 1
The Codistribution of CD81 and Galanin in Rat Brain
ortex 1 1eptumNucleus of the diagonal band 111 111 11ippocampus 11 1halamusParaventricular nucleus 1 1Anterodorsal nucleus 111 11mygdala and extended amygdalaBed nucleus of the stria terminalis 11 11ypothalamusSuprachiasmatic nucleus 1 1Arcuate nucleus 11 11Supraoptic nucleus 111 111Periventricular nucleus 1 11Paraventricular nucleus, lateral
magnocellular part 111 111Paraventricular nucleus, medial
parvocellular part 1 1Paraventricular nucleus, ventral part 1 1Medial preoptic area 1/2 1Medial preoptic nucleus 11 11Anterior hypothalamic area 1 1 1Lateroanterior nucleus 1 1Tuber cinereum area 1 1 11Accessory neurosecretory nuclei 111 111
rain stemInterpeduncular nucleus 1 1
Note. CD81 expression was assessed in the present study. We haverbitrarily defined four grades of in situ hybridization signal in oureactions (from “1/2” to “111”, according to its intensity underight microscopy at high magnification). The data on galanin local-
zation were taken from the published sources (Hamill et al., 1986;
elander et al., 1986).
t(1tsrfb
cfflb
libasaaaiafca
[
313Cocaine-Induced CD81 and Hypothalamic Function
anxiety-like behavior and alterations of CRF concentra-tion in the hypothalamus, amygdala, and basal fore-brain (Sarnyai, 1998).
Exposure to stress increases vulnerability to self-ad-minister psychostimulants. Plasma corticosterone isnecessary for cocaine reinforcement. The mechanism bywhich corticosteroids interact with the mesocorticolim-bic dopaminergic system remains to be determined, butthe medial prefrontal cortex is probably one brain re-gion where DA and adrenocorticosteroids interact toaffect cocaine reinforcement (Goeders, 1997).
Both acute and chronic cocaine administration in-duces changes in the OT and AVP contents in thehypothalamus and in limbic structures (Van de Kar etal., 1992). AVP inhibits, whereas OT facilitates, the de-velopment of behavioral sensitization to cocaine (Sarn-yai, 1998). OT also inhibits acute cocaine-induced loco-motor hyperactivity, exploratory activity, andstereotyped behavior in rodents. Interaction of OT withdopaminergic neurotransmission in the NAcc has beenpostulated as a mechanism of action to modulate neu-roadaptation to cocaine (Sarnyai, 1998).
A more recent example concerns the CART upregu-lation in the striatum after acute cocaine treatment(Douglass et al., 1995). CART is involved in the regula-ion of feeding behavior, sensory processing, and stressCouceyro et al., 1997, 1998; Kuhar and Dall Vechia,999). CART expression was also observed in our pro-ocol of cocaine treatment. We have found the expres-ion of CART and CD81 mRNAs in functionally sepa-ate hypothalamic cell groups, suggesting a distinct roleor the two proteins in normal and cocaine-treatedrain.These studies strongly support the hypothesis that
ocaine action affects expression of proteins importantor hypothalamic functions, such as cardiovascular anduid homeostasis control, stress response, and feedingehavior.In the present study CD81 was found to be upregu-
ated in the NAcc upon cocaine treatment. DA level isncreased in the NAcc of CD81-deficient mice, and theirehavioral responses to cocaine are modified (Michna etl., 1999; submitted for publication). In addition wehow the expression of CD81 in brain regions function-lly related to the regulation of cardiovascular functionnd fluid homeostasis and its codistribution with Galnd AVP. These findings add to previous data suggest-ng a connection between the brain reward pathwaysnd the centers regulating endocrine and autonomicunctions and may show a relation of the neurochemi-
al, behavioral, and somatic consequences of drugbuse.
EXPERIMENTAL METHODS
Drug treatment. Eight-week-old, male Sprague–Dawley rats (Taconic Farms, Germantown, NY) wereinjected subcutaneously with saline or 30 mg/kg co-caine–HCl (Sigma Chemical Co., St. Louis) every 2 h forfour injections and were sacrificed 24 h after the lastinjection (Gawin, 1991). Brains were extracted and stri-atum, hippocampus, and tegmentum were dissected asfollows: a coronal slice through the caudal third of theolfactory tubercle was made. A second coronal slice wasmade through the anterior hypothalamus. Lateral stri-atum and accumbens were isolated from the resultingcoronal section. Accumbens tissue around the anteriorcommissure was first dissected, followed by striataltissue to the corpus callosum. On exposure of the dorsalside of the remaining undissected mid- and hindbrain,the telencephalon was pushed forward and the Hipplifted free. At the leading edge of the superior collicu-lus, a diagonal coronal incision was made in the rostral-to-caudal direction. Similarly a caudal-to-rostral cutwas made at the trailing edge of the inferior colliculus.The resulting wedge of tissue contains the colliculi onits dorsal face and a section of the cerebral peduncle onits ventral face. Recumbent on its posterior plane, thepeduncle was removed and a vertical incision wasmade ventral to the cerebral aqueduct. The resultingstrip of tegmentum contained the cell bodies of themidbrain dopaminergic system.
RNA isolation and differential display analysis.Total RNA was extracted, with the RNAqueous kit(Ambion), from the selected brain regions of control orcocaine-treated rats. Differential display was then per-formed essentially as described (Brenz Verca et al.,1998). Briefly, 1 mg total RNA was reverse transcribedwith the modified oligo(dT) primer sets. Reaction con-ditions were set as described (Liang and Pardee, 1992).Subsequent PCR was performed with the same sets ofoligo(dT) primers in addition to arbitrary primers XAp1to XAp20. PCR was performed as follows: 5 cycles atlow stringency (94°C for 60 s, 40°C for 120 s, 72°C for60 s) followed by 35 cycles at high stringency (94°C for45 s, 56°C for 60 s, 72°C for 60 s). A typical PCR reactionwas carried out in 20 ml final volume with the followingfinal reagent concentrations: dNTPs, 2 mM; 39-anchor-ing primer, 1 mM; arbitrary primer, 1 mM; MgCl2, 1 mM.One-fiftieth of the cDNA obtained from RT, 1–2 mCia-32P]dATP at 3000 Ci/mmol (Hartmann Analytic),
and 1.5 U Taq polymerase (Gibco BRL) were used for
each PCR. Reaction products were resolved on 6% de-naturing sequencing gels which were dried and ex-
etowPwqsTG
314 Brenz Verca et al.
posed to Fuji X-ray film overnight. Excised gel sliceswere rehydrated in 100 ml 13 TE at 80°C for 10 min.Reamplifications were carried out in 40 ml total volumewith the following final reagent concentrations: dNTPs,250 mM; reamplification primers, 1 mM each; MgCl2, 2.5mM; Taq DNA polymerase, 1.5 U/reaction; DNA, 5 mleluate/reaction. Cycling conditions were 5 cycles (94°Cfor 45 s, 40°C for 60 s, 72°C for 60 s), followed by 35cycles at high stringency (94°C for 60 s, 62°C for 60 s,72°C for 60 s) and a 15-min extension step at 72°C. PCR
FIG. 7. CD81 in situ hybridization in the supraoptic magnocellularcells at (A) low magnification and (B) high magnification. (C) THhybridization at the same magnification as in (B). Scale bars, 100 mm(in A) and 25 mm (in B, C).
products were sequenced by Microsynth GmbH eitherdirectly or after cloning into pBluescript KS(1) vector.
Quantitative end-point RT-PCR. Total RNA wasxtracted using an RNAqueous kit (Ambion, AMS Bio-echnology, Ltd., Lugano, Switzerland) from the NAccf control or cocaine-treated rats. Reverse transcriptionas performed essentially as described (Liang andardee, 1992; Brenz Verca et al., 1998). Primers for PCRere designed based on CD81 mRNA GenBank se-
uence for rat (TAPA-1) and mouse (MD3). The up-tream primer was 59-ACT TTT CCT GTC ACC TTTGG-39; the downstream primer was 59-TCT TGA AATGA GGC AGG AAA-39. PCR was carried out in 20 ml
final volume with the following final reagent concen-trations: dNTPs, 250 mM; both primers, 1 mM; MgCl2,1.5 mM. Four 10-fold serial dilutions of cDNA from RTreaction and 2 U Taq polymerase (Gibco BRL) wereused for each PCR. PCR was performed as follows:initial denaturation step at 94°C for 4 min followed by35 reamplification rounds (94°C for 30 s, 60°C for 30 s,72°C for 60 s).
Quantitative real-time RT-PCR. Total RNA wasextracted using an RNAqueous kit (Ambion, AMS Bio-technology Ltd.) from the selected brain regions of con-trol or cocaine-treated rats. RT was performed essen-tially as described (Liang and Pardee, 1992; Brenz Vercaet al., 1998). Primers for quantitative PCR were designedusing PrimerExpress software (PE Applied Biosystems,Foster City, CA) based on CD81 GenBank sequences forrat (TAPA-1). The 28S ribosomal RNA (rRNA) wasused as an endogenous control. The forward and re-verse primers for CD81 were, respectively, 59-GGCTAG GCT CCA ACC CTT CT-39 and 59-GAC TGC GGTGTG GCT ATT GAG-39. Primers for the control 28SrRNA were 59-CAG CGA AAC CAC AGC CAA G-39and 59-CTT CTT TCC CCG CTG ATT CC-39. Up to fiveindependent real-time PCRs were carried out on aniCycler (Bio-Rad). The measure of SYBR green fluores-cence was used for quantification of PCR product. Thereaction was carried out in 50 ml final volume contain-ing 25 ml SYBR Green PCR Master Mix (PE AppliedBiosystems), 300–900 nM each primer, and variableamount of cDNA template. PCR was performed asfollows: initial denaturation step at 94°C for 10 minfollowed by 40 annealing–extension cycles (94°C for15 s, 60°C for 60 s). The relative fluorescence at thethreshold cycle was used to calculate the relative ex-pression of CD81, using the 28S rRNA as endogenouscontrol.
In situ hybridization. In situ hybridization was per-formed essentially as described (Braissant and Wahli,1998). Briefly, DIG- or fluorescein-labeled sense and
antisense probes were produced using T3, T7, or SP6promoters from plasmids containing the clones of in-
mrpcpCpk((dN
pbtcto
ecrhpiddhS
it3sftNMwsSp(m
H
H
315Cocaine-Induced CD81 and Hypothalamic Function
terest. Before the in vitro transcription reaction the plas-ids were digested nearest the insert with suitable
estriction enzymes to prevent the contamination ofrobe with vector sequences. The mouse MD3 (CD81)omplete cds clone in pCRII vector, the rat CART com-lete cds clone corresponding to the short form ofART peptide in pBSII SK(1) vector, and the rat THartial cDNA clone ('1500 bp) in pSPT18 vector wereindly provided by E. Tsitsikov (Harvard), S. HastrupNovo Nordisk, Denmark), and N. Faucon BiquetCNRS, Paris, France), respectively. TH probe was hy-rolyzed for 3 min in 0.2 M carbonate buffer (80 mMaHCO3/120 mM Na2CO3, pH 10.2) at 60°C to the final
fragment length of '500 bp. Serial dilutions of ribo-robes were spotted on positively charged nylon mem-rane together with serial dilutions of DIG-labeled an-isense neo-RNA (Boehringer Mannheim) of knownoncentration as standard, to estimate the efficiency ofhe labeling reaction. In addition, probes were loadedn 2% agarose gel to check for RNA size and integrity.Rat brains were rapidly frozen in isopentane upon
xtraction and stored dry at 280°C. Twenty-five-mi-rometer sections were cut on a cryostat, then dried atoom temperature, and stored at 220°C. For in situybridization, sections were postfixed for 10 min in 4%araformaldehyde in phosphate-buffered saline (PBS),
ncubated 2 3 15 min in PBS containing 0.1% activeiethylpyrocarbonate, rinsed briefly in 53 saline so-ium citrate (SSC), and prehybridized for 1–2 h inybridization buffer containing 50% formamide, 53SC, and 40 mg/ml salmon sperm DNA at 58–61°C.
Hybridization was then carried out in the same bufferand at the same temperature with 50–100 ng/ml DIG-or fluorescein-labeled probe overnight. After hybridiza-tion sections were washed for 30 min at room temper-ature, then for 1 h at 65°C in 23 SSC, and for 1 h at 65°Cn 0.13 SSC. For in situ hybridization with TH probe,he additional step of RNase treatment (30 mg/ml at0°C) after hybridization was necessary to remove non-pecific staining. After washing the detection was per-ormed with anti-DIG or anti-fluorescein Abs coupledo alkaline phosphatase and color development with
BT/BCIP substrate (all reagents from Boehringerannheim). Nonspecific staining was then removed byashing for 1 h in 95% EtOH. The images of entire
ections were digitalized using videocamera and Imagetore 5000 hardware (UVP). Photomicrographs wererepared by capturing the images with a Spot Camera
Diagnostic Instruments, Inc.) on a Zeiss Axiophot lighticroscope. I
ACKNOWLEDGMENTS
This work was supported, in part, by NIDA Grant R01 DA11480, aCharles and Johanna Bush Memorial award, a Johnson and Johnsonaward (G.C.W.), and Swiss National Foundation Grant 31-53206.97(J.L.D.). J.L.D. and G.C.W. should be considered co-senior authors.
REFERENCES
Angelisova, P., Hilgert, I., and Horejsi, V. (1994). Association of fourantigens of the tetraspans family (CD37, CD53, TAPA-1, and R2/C33) with MHC class II glycoproteins. Immunogenetics 39: 249–256.
Behr, S., and Schriever, F. (1995). Engaging CD19 or target of anantiproliferative antibody 1 on human B lymphocytes inducesbinding of B cells to the interfollicular stroma of human tonsils viaintegrin alpha 4/beta 1 and fibronectin. J. Exp. Med. 182: 1191–1199.
Berditchevski, F., and Odintsova, E. (1999). Characterization of inte-grin–tetraspanin adhesion complexes: Role of tetraspanins in inte-grin signaling. J. Cell Biol. 146: 477–492.
Boismenu, R., Rhein, M., Fischer, W. H., and Havran, W. L. (1996). Arole for CD81 in early T cell development. Science 271: 198–200.
Braissant, O., and Wahli, W. (1998). A simplified in situ hybridisationprotocol using non-radioactively labeled probes to detect abundantand rare mRNAs on tissue sections. Biochemica 1: 10–16.
Brenz Verca, M. S., Brenz Verca, S., Rusconi, S., and Dreyer, J. L.(1998). Modification of primer design facilitates the use of differ-ential display. Biotechniques 24: 374–376, 378–380.
Couceyro, P., Paquet, M., Koylu, E., Kuhar, M. J., and Smith, Y. (1998).Cocaine- and amphetamine-regulated transcript (CART) peptideimmunoreactivity in myenteric plexus neurons of the rat ileum andco-localization with choline acetyltransferase. Synapse 30: 1–8.
Couceyro, P. R., Koylu, E. O., and Kuhar, M. J. (1997). Further studieson the anatomical distribution of CART by in situ hybridization.J. Chem. Neuroanat. 12: 229–241.
Douglass, J., McKinzie, A. A., and Couceyro, P. (1995). PCR differen-tial display identifies a rat brain mRNA that is transcriptionallyregulated by cocaine and amphetamine. J. Neurosci. 15: 2471–2481.
Fagergren, P., and Hurd, Y. L. (1999). Mesolimbic gender differencesin peptide CART mRNA expression: Effects of cocaine. NeuroReport10: 3449–3452.
Gawin, F. H. (1991). Cocaine addiction: Psychology and neurophysi-ology. Science 251: 1580–1586. [Published erratum appears in Sci-ence, 1991, 253: 494]
Geisert, E. E., Jr., Yang, L., and Irwin, M. H. (1996). Astrocyte growth,reactivity, and the target of the antiproliferative antibody, TAPA.J. Neurosci. 16: 5478–5487.
Goeders, N. E. (1997). A neuroendocrine role in cocaine reinforce-ment. Psychoneuroendocrinology 22: 237–259.
Hamill, G. S., Skofitsch, G., and Jacobowitz, D. M. (1986). Immuno-cytochemical localization of atrial natriuretic factor, galanin andcalcitonin gene-related peptide within the rat interpeduncular nu-cleus. Brain Res. Bull. 17: 83–93.eimer, L. (1995). The Human Brain and Spinal Cord: Functional Neu-roanatomy and Dissection Guide, 2nd ed. Springer-Verlag, Berlin.okfelt, T., Broberger, C., Diez, M., Xu, Z. Q., Shi, T., Kopp, J., Zhang,X., Holmberg, K., Landry, M., and Koistinaho, J. (1999). Galaninand NPY, two peptides with multiple putative roles in the nervous
system. Horm. Metab. Res. 31: 330–334.
mai, T., and Yoshie, O. (1993). C33 antigen and M38 antigen recog-
I
K
L
L
L
L
M
M
M
M
M
M
M
O
P
S
S
S
S
T
T
T
T
V
W
Y
Z
Z
316 Brenz Verca et al.
nized by monoclonal antibodies inhibitory to syncytium formationby human T cell leukemia virus type 1 are both members of thetransmembrane 4 superfamily and associate with each other andwith CD4 or CD8 in T cells. J. Immunol. 151: 6470–6481.
rwin, M. H., and Geisert, E. E., Jr. (1993). The upregulation of a glialcell surface antigen at the astrocytic scar in the rat. Neurosci. Lett.154: 57–60.
uhar, M. J., and Dall Vechia, S. E. (1999). CART peptides: Noveladdiction- and feeding-related neuropeptides. Trends Neurosci. 22:316–320.
agaudriere-Gesbert, C., Le Naour, F., Lebel-Binay, S., Billard, M.,Lemichez, E., Boquet, P., Boucheix, C., Conjeaud, H., and Rubin-stein, E. (1997). Functional analysis of four tetraspans, CD9, CD53,CD81, and CD82, suggests a common role in costimulation, celladhesion, and migration: Only CD9 upregulates HB-EGF activity.Cell. Immunol. 182: 105–112.
evy, S., Todd, S. C., and Maecker, H. T. (1998). CD81 (TAPA-1): Amolecule involved in signal transduction and cell adhesion in theimmune system. Annu. Rev. Immunol. 16: 89–109.
iang, P., and Pardee, A. B. (1992). Differential display of eukaryoticmessenger RNA by means of the polymerase chain reaction. Science257: 967–971.
in, S. L., Derr, D., and Hildreth, J. E. (1992). A monoclonal antibodyagainst a novel 20-kDa protein induces cell adhesion and cytoskel-eton-dependent morphologic changes. J. Immunol. 149: 2549–2559.aecker, H. T., Todd, S. C., and Levy, S. (1997). The tetraspaninsuperfamily: Molecular facilitators. FASEB J. 11: 428–442.annion, B. A., Berditchevski, F., Kraeft, S. K., Chen, L. B., andHemler, M. E. (1996). Transmembrane-4 superfamily proteins CD81(TAPA-1), CD82, CD63, and CD53 specifically associated with in-tegrin alpha 4 beta 1 (CD49d/CD29). J. Immunol. 157: 2039–2047.atsumoto, A. K., Martin, D. R., Carter, R. H., Klickstein, L. B.,Ahearn, J. M., and Fearon, D. T. (1993). Functional dissection of theCD21/CD19/TAPA-1/Leu-13 complex of B lymphocytes. J. Exp.Med. 178: 1407–1417.elander, T., Fuxe, K., Harfstrand, A., Eneroth, P., and Hokfelt, T.(1987). Effects of intraventricular injections of galanin on neuroen-docrine functions in the male rat. Possible involvement of hypo-thalamic catecholamine neuronal systems. Acta Physiol. Scand. 131:25–32.elander, T., Hokfelt, T., and Rokaeus, A. (1986). Distribution ofgalaninlike immunoreactivity in the rat central nervous system.J. Comp. Neurol. 248: 475–517.ichna, L., Brenz Verca, M. S., Chen, S., Lee, J., Rogrove, J., Zhou, R.,Tsitsikov, E., G. C., M., J.-L., D., and G. C., W. (1999). Cocaine-induced place preference in CD81 deficient mice. In Society forNeuroscience, 29th Annual Meeting, Miami Beach, FL.oldow, R. L., and Fischman, A. J. (1987). Cocaine induced secretionof ACTH, beta-endorphin, and corticosterone. Peptides 8: 819–822.ren, R., Takahashi, S., Doss, C., Levy, R., and Levy, S. (1990).TAPA-1, the target of an antiproliferative antibody, defines a new
family of transmembrane proteins. Mol. Cell. Biol. 10: 4007–4015.
R., Weiner, A. J., Houghton, M., Rosa, D., Grandi, G., and Abrig-nani, S. (1998). Binding of hepatitis C virus to CD81. Science 282:938–941.
arnyai, Z. (1998). Oxytocin and neuroadaptation to cocaine. Prog.Brain Res. 119: 449–466.
erru, V., Le Naour, F., Billard, M., Azorsa, D. O., Lanza, F., Boucheix,C., and Rubinstein, E. (1999). Selective tetraspan–integrin com-plexes (CD81/alpha4beta1, CD151/alpha3beta1, CD151/alpha6beta1) under conditions disrupting tetraspan interactions.Biochem. J. 340: 103–111.
uarez, M., Maglianesi, M. A., and Perassi, N. I. (1998). Involvementof the anterodorsal thalami nuclei on the hypophysoadrenal re-sponse to chronic stress in rats. Physiol. Behav. 64: 111–116.
ullivan, C. D., and Geisert, E. E., Jr. (1998). Expression of rat target ofthe antiproliferative antibody (TAPA) in the developing brain.J. Comp. Neurol. 396: 366–380.
odd, S. C., Lipps, S. G., Crisa, L., Salomon, D. R., and Tsoukas, C. D.(1996). CD81 expressed on human thymocytes mediates integrinactivation and interleukin 2-dependent proliferation. J. Exp. Med.184: 2055–2060.
ohyama, M., and Takatsuji, K., Eds. (1998). Atlas of NeuroactiveSubstances and Their Receptors in the Rat. Oxford Univ. Press, Ox-ford/New York/Tokyo.
orres, G., and Rivier, C. (1992). Differential effects of intermittent orcontinuous exposure to cocaine on the hypothalamic–pituitary–adrenal axis and c-fos expression. Brain Res. 571: 204–211.
sitsikov, E. N., Gutierrez-Ramos, J. C., and Geha, R. S. (1997). Im-paired CD19 expression and signaling, enhanced antibody re-sponse to type II T independent antigen and reduction of B-1 cellsin CD81-deficient mice. Proc. Natl. Acad. Sci. USA 94: 10844–10849.
an de Kar, L. D., Rittenhouse, P. A., O’Connor, P., Palionis, T.,Brownfield, M. S., Lent, S. J., Carnes, M., and Bethea, C. L. (1992).Effect of cocaine injections on the neuroendocrine response to theserotonin agonist MK-212. Biol. Psychiatry 32: 258–269.eiss, J. M., Bonsall, R. W., Demetrikopoulos, M. K., Emery, M. S.,and West, C. H. (1998). Galanin: A significant role in depression?Ann. N. Y. Acad. Sci. 863: 364–382.
anez-Mo, M., Alfranca, A., Cabanas, C., Marazuela, M., Tejedor, R.,Ursa, M. A., Ashman, L. K., de Landazuri, M. O., and Sanchez-Madrid, F. (1998). Regulation of endothelial cell motility by com-plexes of tetraspan molecules CD81/TAPA-1 and CD151/PETA-3with alpha3 beta1 integrin localized at endothelial lateral junctions.J. Cell Biol. 141: 791–804.
hou, Y., Schlussman, S. D., Ho, A., Spangler, R., Fienberg, A. A.,Greengard, P., and Kreek, M. J. (1999). Effects of chronic ‘Binge’cocaine administration on plasma ACTH and corticosterone levelsin mice deficient in DARPP-32. Neuroendocrinology 70: 196–199.
hou, Y., Spangler, R., LaForge, K. S., Maggos, C. E., Ho, A., andKreek, M. J. (1996). Corticotropin-releasing factor and type 1 corti-cotropin-releasing factor receptor messenger RNAs in rat brain and
pituitary during “binge”-pattern cocaine administration and
