andrew j. bean et al- effects of dopamine depletion on striatal neurotensin: biochemical and...

8/3/2019 Andrew J. Bean et al- Effects of Dopamine Depletion on Striatal Neurotensin: Biochemical and lmmunohistochemical

1/9

The Journal of Neuroscience, December 1999, 9(12): 44304439

Effects of Dopamine Depletion on Striatal Neurotensin: Biochemicaland lmmunohistochemical StudiesAndrew J. Bean, Matthew J. During, Ariel Y. Deutch, and Robert H. RothDepartments of Pharmacology and Psychia try, Yale University Schoo l of Medicine, New Haven, Connecticut 06510

Interactions between striatal dopamine (DA) and neuroten-sin (NT) have been suggested by anatomical, behavioral,and biochemical studies. Nigrostriatal DA neurons, in con-trast to mesocotticolimbic DA neurons, do not appear tocontain NT. Thus, distinct neuronal elements subserve in-teractions between DA and NT within the striatum. We havepreviously demonstrated that reserpine-induced depletionof striata l D A is accompanied by a dose- and time-dependentincrease in striata l NT concentrations. In order to furthercharacterize the effects of reserpine and to define the mech-anism by which reserpine acts to increase striatal NT con-centrations, we have used immunohistochemical and bio-chemical approaches. lmmunohistochemical examination ofrats pretreated with reserpine revealed marked increases inthe density of NT-like immunoreactive (NT-Ii) perikarya andfibers, and the development of NT-Ii patches. Pretreatmentwith reserpine had no apparent effect on NT synthesis, asassessed by examination of cycloheximide-induced inhibi-tion of protein synthesis. However, reserpine administrationresulted in a significant decrease in the release of both DAand NT into the striata l extrace llular fluid, as measured byin viva microdialysis. These data suggest that the increasein striata l NT concentrations observed after reserpine trea t-ment resu lts from decreased release, rather than increasedsynthesis of the peptide.Neurotensin (NT) is a tridecapeptide originally isolated frombovine hypothalamus (Carraway and Leeman, 1973,1976) whichis heterogeneously distributed through the CNS (Carraway andLeeman, 1976; Uhl et al., 1977; Jennes et al., 1982; Uhl, 1982).The demonstration that NT perikarya, fibers, and receptors existwithin the striatum (Young and Kuhar, 198 1; Jennes et al., 1982;Uhl et al., 1982), a brain region known to contain high concen-trations of dopamine (DA), has led to studies aimed at eluci-dating the possible interactions between these transmitters/modulators in this brain region. Nigrostriatal DA system func-tion i s influenced by NT. For example, NT enhances both basal

Received Feb. 10, 1989; revised Apr. 27, 1989; accepted June 1, 1989.This work was supported in part by NIGMS-GM 07324, MH-14092, TheScottish Rite Schizophrenia Research Program, and The Department of Psychia-try, Yale University School of Medicine, and the Ribicoff Research Fac ilities,Connecticut Mental Health Center. State of Connecticut Denartment of MentalHealth. M.J.D. is the recipient of k Inte rnational Fogarty Fillowship.We wish to thank Dr. Lothar Jennes for supplies of NT antiserum and Dr.Clinton Kilts for sharing unpublished data.Correspondence should be addressed to Dr. R. H. Roth, Department of Phar-macology, Yale University School of Medicine, 333 Cedar Street, New Haven,CT 06510.

Copyright 0 1989 Society for Neuroscience 0270-6474/89/1244 30-09$02.00/0

and stimulated DA release from striatal slices; these effects maybe mediated by NT receptors localized to the terminals of do-paminergic axons (De Quidt and Emson, 1983; Quirion et al.,1985; Hetier et al., 1988). Striatal NT levels have been shownto be influenced by DA; dopamine receptor antagonists havebeen reported to alter striatal NT levels and distribution (Go-voni et al., 1980; Goedert et al., 1985; Frey et al., 1986; Nem-eroff, 1986; Letter et al., 1987; Eggerman and Zahm, 1988).We have recently demonstrated that reserpine, which depletesmonoamines by disrupting their vesicular storage, produces adose- and time-related increase in striatal NT concentrations(Bean et al., 1989). The influence of reserpine on striatal NTlevels might be anticipated on the basis of a postulated tonicinhibitory influence of DA on striatal NT levels (Merchant etal., 1988a). The increase in striatal NT tissue concentrationsresulting from treatment with either reserpine or DA receptorblockers is presumably due to increases in NT synthesis and/or to decreases in NT release. In the present study, we havefurther characterized the increase in striatal NT produced byreserpine using immunohistochemical methods and have in-vestigated the mechanism by which this increase is produced.Materials and MethodsAnimals. Male Sprague-Dawley rats (225-325 gm, Cam m, Wayne, NJ)were used in al l experiments. Rats were housed 2-3 per cage and wereallowed free access to food and water at al l t imes.Experimental design. Immunohistochemical analysis of str iatal NTwas carried ou t following the procedure outlined below , 18-20 hr fol-lowing the administration of reserpine (5 mg /kg, i.p.).NT synthesis was est imated fol lowing the method of Frey et al. (1988).Brief ly, rats were injected with cycloheximide (1 mg/kg, s .c.) fol lowed60 min later by 0.9% saline or reserpine (5 mg /kg, i.p.). Nineteen hoursfol lowing cycloheximide inject ion, the rats were sacrif iced by decapi-tat ion and the str iatum was punch dissected as described previously(Bean et al. , 1989). Samples were frozen at -70C unt i l assayed for NTby radioimmunoassay (see below). Proteins were measured followingthe method of Lowry et al. (195 1).

The e ffects of reserpine on NT and DA release in the rat str iatumwere determined by est imation of DA and NT content in the str iatalextracellular fluid using in vivo microd ialysis both before and after re-serpine (5 mg/kg, i .p.) or 0.9% saline injections. We examined the releaseof DA and NT for 6 hr followin g reserpine administration, a time atwhich a marked increase in NT and decrease in DA t issue concentrat ionsoccnrs (Bean e t al. , 1989).In vivo microdialysis. Dialysis probes were prepared as fol lows. Brief-ly, the probes w ere of conce ntric design using a hollow dialysis fiber(5000 MW cute@ Cuprophan, Hospal, Edison, NJ) sealedat one endwith epoxy resin (De&oh, Danvei, MA). A length of hollow fusedsil ica fiber (0.170 m m O.D.. Ansnec Inc.. MI) w as inserted into thedialysis tu& to its end. The hialyiis tubing coitaining the fused silicawas then inserted through a length of 22 gauge stainles s steel tubinginto which another piece of fused sil ica tubing (outlet tube) had beenplaced. B oth ends of the 22 gauge tubing were sealed with epoxy resin,leaving 4 mm of exposed dialysis tubing protruding from one end of


2/9

The Journal of Neuroscience, December 1989, S(12) 4431

the probe and the fused silica inlet and outlet tube s extending from theother end. A sma ll piece of 23 gauge tubing was glued onto the end ofthe inlet tube connected to a length of PE-50 tubing, which was con-nected to a 1 ml syringe mounted on a syringe pump (Harvard Instru-ments, S. Nat ick, MA). A rt i fic ial cerebrospinal f luid modif ied after Mer-l is (1940) [consist ing of ( in mM) NaCl, 135; KCl, 3.0; CaC I,, 1.2; MgCl, ,1.0; phosphate, 2.0; and 100 ELMascorbic acid, DH 7.41 was oumoedthrough the dialysis probe at a rate of 2.1 @m&t: At this f low-rate-therecovery of dopamine was between lO-20h, while the recovery of NTwas 2-8%. In the reserpine experiment, samples collected during each60 min period were divided such that a port ion was used for HPLCanalysis of dopamine content, while the remaining sample was frozenat - 70C unt i l assayed for NT content bv radioimmunoassav.In viva dialysis experiments were accomplished fol lowing ihe place-ment of probes into the str iatum of chloral hydrate anesthet ized rats(400 mgk g, i.a) Icoordin ates: AP = +O.S. L = +2.5. V = -7.0 (Paxinosand Wa tson, 1986 )] and initiating perfusion. Alter a 120 min periodfor equil ibrat ion, samples were collected every 20 min (calcium exper-iment) or 60 min (reserpine exp eriment) for the duration of the exper-iment. Determinat ion of a baseline was made by assaying samples forDA content (NT sa mples were assayed subsequently) and a stable mea-surement (5 20% variation in 3 consecut ive samples) was typically ob-tained 60-100 min after sample collect ion was begun (i.e. , 180-220 minfollowing initiation of perfusion). The effec t of calcium on NT releasewas determined by measuring the abil ity of potassium (60 mM , addedto the artificial cerebrospinal fluid while altering the sodium concen -trat ion to maintain osmolarity) to elevate NT levels in the presence andabsen ce of calcium in the artificial cerebrospinal fluid. We did not correctthe level of NT or DA per sample for the recovery of the individualprobe; thus, only relat ive changes in ECF concentrat ions are reported.Rudioimmunoussuy ofNT. The radioimmunoassay (RIA) for NT wasperformed under nonequilibrium conditions in 12 x 75 borosilicatetubes employing sodium phosphate buffer (50 mM, pH 7.4) containing10 mM EDTA . 0.1% bovine serum albumin. and 0.02% sodium azide.The 500 ~1 reaction mixture s contained 100 ~1 of diluted primary anti-serum raised in the rabbit (final dilution 1:250,00 0, which producedapproxim ately 25% binding), 50 gl of Y-NT containing approxim ately10,000 cpm, and 25 ~1 of NT standard or unknown. To faci l i tate in-terassay comparisons, the standards for approximately 50 assays werealiquoted into sets of microcentrifuge tubes which were kept frozen at-70C unt i l used in an assay. The a ssay was set up at room temperatureby combining the standard or unknown, primary ant ibody, and bufferfol lowed by vortex mixing. The tubes were then incubated at 4C for24 hr at which t ime lz51-NT was added, fol lowed by an additional 24hr incubation period at 4C. Subse quently, the bound (B) and free (F)12I-NT were separated using a goat anti-rabbit second ary antibddy(Ant ibodies Inc., Davis. CA), which was incubated with the assav mix-ture for 4 hr, af ter which polyethylene glycol (10% in 0.1 M phosphatebuffer) containing normal rabbit serum (Ant ibodies Inc., Davis, CA)was added. This was fol lowed immediately by centri fugat ion for 15 minat 3000 rpm in a refrigerated centri fuge (Beckman Instruments). Thesupem atant (free fraction) was decan ted, and radioac tivity in thepelletwas counted. The B/F rat ios were corrected for nonspecific binding,which was est imated by examination of the B/F rat io in the absence ofant ibody. Nonspecif ic binding was typically l-2% of total. The IC,, forsynthet ic NT was approximately 4 fmol, while the detect ion l imit ofthe assay was 0.15 fmo l. Intra- and interassay variation was 5 and 8%,respect ively (n = 20).

Chromatography ofNT. NT was separated chromatographically usinga HPLC system comprised of 2 Altex model 1OOA pumps, an Altexmodel 420 gradient programmer, and a Beckman model 163 variable-wavelength UV detector se t at 210 nm. S amples were injected onto aC-18 column where elut ion was achieved with a l inear gradient (15-4046, 1%/min) of 100% acetonitr ile in 1 M sodium phosphate buffer anda f low rate of 2 ml/min. One minute fract ions were collected, and theirNT content was measured by radioimmunoassay. Synthet ic NT stan-dards were assayed under the same condit ions.Chromatography of DA. DA concentrat ions were measured accordingto the technique of Church and Just ice (1987) with minor modif icat ions.Briefly, 20 ~1 dialysate or standard was injected onto a 3 pm narrow-bore C- 18 column ( 10 x 2 .1 mm ). Se lect ive retent ion of DA was achievedwith a mobile phase consist ing of 0.05 M sodium phosphate buffer (pH5.8) with 0.1 mM EDTA , 2.2 mM sodium octyl sulfate, 5 mM triethyl-amine, and 15% methanol at a f low rate of0.35 ml/min. Electrochemicaldetect ion was accomplished using an LC-3 am perometric detector

Figure 1. S triatal NT-li follow ing pre treatmen t with colchicine . NT-li fibers are present in high density in the periventricular striatum . SomeNT-li perikarya can be seen in the dorsal striatu m and in the periven-tricular as well as subcallosal regions (arrow s). In the ventral striatum ,NT-l i patches which mainly consist of NT-11 f ibers can be seen (arrow-head). V, ventric le; UC,anterior commissure.(Bioanalyt ical System s, Wes t Lafayette, IN) with a glassy carbon elec-trode (+0.7 V vs. Ag/AgCl reference). Chromatograms were recordedon a chart recorder and peak heights of unknowns were compared withstandards for quantitation.Zmmu nohistochemistry. Rats used for immunohistochem istry re-ceived either 200 tia colchicine into the lateral ven tricle 24 hr mior tosacrifice or 5 mg/kg -(i.p.) reserpine 18-20 hr before sacrifice (re&pine-treated rats did not receive colchicine). Animals were deeply anesthe-tized with chloral hydrate (600 mg /kg, i.p.) and were transcardiallyperfused using a modif icat ion of the pH-shif t method of Berod et al.( 198 1). Brief ly, animals were perfused with normal saline a t room tem-perature, follow ed by ice-cold 4% paraformalde hyde in 0.1 M sodiumacetate (pH 6.5), fol lowed by 4% paraformaldehyde-O.08% glutaral-dehyde in 0.05 M sodium borate (pH 9.5). Tissue was postf ixed for 4-6 hr, transferred to 0.1 M sodium phosphate buffer (PH 7.4), and coronalsect ions (50 rm) cut on a vibrating microtome. Sect ions were thenprocessed for immunohistochem istry using the avidin-biotin method.Free-floating section s were rinsed in 0.05 M Tris-buffered saline (TBS)and subseq uently incub ated in methano lic-peroxide for 10 min. Follow-ing repeated washes in TBS , t issue was incubated in TBS containing0.3% Triton X-100 and 2% normal serum (TBS+) for 60 min. Sect ionswere then incubated in the same NT ant ibody used for the radioim-mun oassa y (1:600 0) for 48 hr. Section s were incubated in biotinylatedsecondary ant ibody for 90-l 20 min, washed with TBS, then incubatedin the avidin solution (Vector La boratories) for 60-90 min . Tissue sec-tions were then developed in diaminobenzidine with hydrogen peroxide,washed in TBS, and mounted on sl ides before being dehydrated, cleared,and coversl ipped.Materials. The NT ant iserum used in these studies was kindly donatedby Dr. Lothar Jennes, Department ofAnatomy, Wright S tate University


3/9

4432 Bean et al. l Striatal Neurotensin

Figure 2. A, Localization of NT-li eleme nts in the striatum of a reserpine-treated rat. NT-li fibers are organized into discrete pa tches (arrows)throughout both the dorsal and ventral str iatum. NT-l i perikarya are present both within and outside of the patches of high f iber density and aremost prominent in the periphery of the str iatum. B, Patches of NT-E f ibers (marked by arrows in A). The patches of high f iber density also containNT-l i perikarya. Howeve r, cel l bodies are also observed outside of the NT-h f iber patches (arrow). The NT-l i perikarya outside of the f iber patchesare frequent ly seen to be arranged in clusters. C, An example of an NT-l i str iatal neuron. This neuron has clearly def ined processes (arrows) whichextend for a considerable distance; however, most str iatal N T-l i neurons do not elaborately branch. V, ventric le; UC ,anterior commissure.

School of Medicine, Dayton, OH. Synthet ic N T and al l NT fragments,as well as somatostat in and cort icotropin-releasing factor (CRP) w erepurchased from Bachem Inc . (Torrance, C A). Neuromedin N was pur-chased from Peninsula Labs (Belmont, CA). Y-NT was purchased fromAmersham Inc. (Arl ington Heights, IL). Cycloheximide was purchasedfrom Sigma (St. Louis, MO), and reserpine (Serpasil) was a gif t of Ciba-Geigy Co. (Summ it, NJ).Statistics. he NT turnover experiments were analyzed by means ofANOVA with post hoc Dunnett s test, while the dialysis experimentswere analyzed by ANOVA with repeated measures.

ResultsImmunohistochemistryImmunohistochemical examination of the striatum of rats pre-treated with colchicine revealed the presence of moderately denseNT-like immunoreactive (NT-li) fibers in the medial subcallosaland periventricular regions (Fig. 1). In the precommissural stria-turn, the highest density of fibers was observed in the medialstriatum and nucleus accumbens, while in the postcommissuralstriatum fiber label ing was concentrated mainly in the medialand dorsal regions. NT-l i neurons were observed in the medialstriatum and within the nucleus accumbens. Scattered NT-l ineurons adjacent to the callosum were observed at all levels ofthe striatum (see Fig. 1).The number of striatal NT-B neurons in rats pretreated with

reserpine was markedly increased, as was the density of NT-lifibers (Fig. 2A). Both NT-li perikarya and fibers appeared to bepresent in discrete patches, surrounded by regions relativelydevoid of immunoreactive elements (Fig. 2, A-C). The patchesof NT-l i neurons and fibers were prominent in the dorsal andlateral aspects of the striatum and in the ventromedial striatum,including the nucleus accumbens; the central core of the pre-commissural striatum contained few patches of NT-l i fibers andcells.NT synthesisA modification of the techniques of Frey et al. (1988) and Ban-non and Goedert (1984) was used to determine whether thereserpine-induced increase in striatal NT levels was attributableto an increase in peptide synthesis. An increase in tissue NTlevels was observed following reserpine administration (see Fig.3). Following administration of the protein synthesis inhibitorcycloheximide, NT levels were significantly decreased from con-trol levels (Fig. 3). However, when cycloheximide was admin-istered in addition to reserpine, levels of NT observed in thestriatum were significantly increased versus those seen followingcontrol and cycloheximide treatments, but did not differ sig-nificantly from the increased levels seen after reserpine alone.


4/9

The Journal of Neuroscience, December 1989,9(12) 4433

Vehicle

Figure 3. Effectsof reserpineon NT synthesis.Reserpine 5 mgfltg,i-p.) was administered 18 hr prior to sacrifice.Cycloheximide (1 mglkg, i.p.) was administered 19 hr prior to sacrifice, .e., 1 hr prior toreserpineor vehicle.BasalNT concentrations n the striatum were 30.02 2.48 fmoYmg protein. Data are expressed smean + SEM percentageof vehicle-injected ontrols (n = 8). Cycloheximideplus reserpinegroupdiU not differ sign ifican tly from reserpine group alone. * p 5 0.05,relative to the vehiclecontrol group.

DA and NT measurementsIn order to determine whether reserpine altered striatal NTrelease, we developed and characterized a sensitive NT radioim-munoassay which was used in conjunction with brain micro-dialysis sampling techniques. A radioimmunoassay utilizing anantibody directed against the COOH-terminal of the NT mol-ecule was used to measure NT in the same striatal dialysissample used for DA measurement. The specific ity of this an-tibody for both the tridecapeptide, as well as fragments of theparent peptide, is illustrated in Table 1 (also see Jennes et al.,1982). As can be seen n Table 1, the COOH-terminal fragmentsNT (8-l 3) and NT (9-l 3) exhibited 5 1 and 68% cross-reactivity,respectively, with the full length peptide. However, the C-ter-minal ly related peptide neuromedin N and amino terminal frag-ments NT (l-11) and NT ( l-8) cross-reacted less than 0.1%with NT. Furthermore, neither somatostatin, a peptide of sim-ilar molecular weight to NT, or CRF was observed to cross-react (~0.1%) with the antibody.To characterize the material collected through the dialysisprobe as authentic NT, dialysis samples were separated by HPLC,and eluted fractions were assayed using our radioimmunoassay.Elution profiles of tractions were compared with that of syntheticNT. HPLC analysis revealed an absorbance peak in the dialysissample which coeluted with synthetic NT (Fig. 4). Furthermore,when fractions of HPLC eluate were measured by RIA, themajor NT-l i peak from the dialysis sample occurred at the sametime point as the peak containing synthetic NT (Fig. 5).We quantified the DA found in our dia lysis samples by com-parison with DA standards assayed according to the same meth-

Figure 4. HPLC analysisof striatal dialysate or the presenceof NT.Syntheticstandardsand dialysissamples seeMaterials and Methods)were separatedwith reverse-phase PLC usinga linear gradient of 15-

I9

L

t0 2 4 6 8 1Omin

Inn.0 2 4 6 8 lOmin40% aceton itrile in 0.1 M phosphate buffer, and UV detection at 2 10nm. A, Synthet ic NT standard; B, dialys issample.Arrows refer to NTsignal n A and B.


5/9


odology (HPLC-EC). Characterization of the DA peak in thechromatographs was accomplished by both modification of theanalytical (I-IPLC-EC) method (altering the buffer and deter-mining the profile ofthe dialysis sample in relation to a standard)and by determining the response to pharmacological manipu-lation. cu-Methyl-para-tyrosine, apomorphine, and reserpine alldecreased the sizeof the peak which coeluted with DA standardswith magnitudes and time courses comparable to those observedpreviously (data not shown, see Fig. 7; Zetterstrom and Unger-stedt, 1984; Butcher et al., 1988). Furthermore, amphetamineand electrical stimulation of the median forebrain bundle bothincreased the size of the peak which coeluted with DA standardsas measured by HPLC-EC (data not shown; Zetterstrom et al.,1983; Imperato and DiChiara, 1984; Butcher et al., 1988).In vivo MicrodialysisWe have examined the dependence of the release of NT oncalcium. As can be seen in Figure 6, substitution of 60 mMpotassium for an equal amount of sodium in the artif icial cere-brospinal fluid resulted in an elevation of NT levels (epoch 4).However, when calcium was omitted from the artif icial cere-brospinal fluid, potassium did not evoke an increase in NT levelssignif icantly different from baseline (Fig. 6, epoch 7). Followingcharacterization of the dialysate for the presence of DA and NTand the calcium dependence of such release, studies on theeffects of reserpine were begun. Basal DA and NT values in thestriatal dialysate were 2.47 + 0.16 and 0.040 + 0.004 fmol/min, respectively. Reserpine administration produced decreasesin both DA and NT concentration in the extracel lular fluid (ECF,Fig. 7). The magnitude of the decline in DA at its nadir was79.0 + 4.9% (p 5 0.05 at all points after reserpine), while themaximal decline in NT levels was 77.2 f 5.0% (p 5 0.05 at allpoints exmpt 180 min).DiscussionThe present data suggest that the reserpine-induced increase inNT levels may be attributable to a decrease in peptide releaserather than an increase in NT synthesis. Moreover, the increasein tiatal NT levels was paralleled by an increase in the densityof NT-l i neurons and fibers, including the emergence of a patchyorganization of NT-l i elements. Thus, the reserpine-induceddecrease in release of NT appeared to elevate intraneuronalconcentrations of the peptide.In order to test the hypothesis that increases in striatal NTproduced by reserpine are due to alterations in NT release, wedeveloped an in vivo microdialysis method coupled with sen-sitive RIA and HPLC-EC analytical assays. While the mea-surement of femtomole quantities of DA has been accomplished

04::::; :;:I ::::: I : :: :I0 1 2 3 4 5 6 7 8 9 1011121314151617181920TIME (MN)

Figure 5. Analysisof NT-Ii in fractions of HPLC-separatedNT stan-dard and dialys issamples.One minute fractions of HPLC eluate werecollected rom the HPLC on which a standard NT solution and dialysissampleswere separated.The fractions were quantitated using our RIA(seeMaterials and Methods), and data are expressedas radioactivity@pm) per sample.by many laboratories (Ungerstedt and Hallstrom, 1987), themeasurement of dialyzed peptides from brain has proven to bemore difficult. Using a nonequilibrium RIA, the sensi tivity limitof our assay s approximately 0.15 fmol NT. We were thereforeable to divide 20 min samples and measure both DA and NTwithin such samples. The NT antibody used appears to bind tothe COOH-terminal region of the NT molecule (based on thecross-reactivity profile of various NT fragments, Table l), con-sistent with the previous immunohistochemical characteriza-tion by Jennes et al. (1982). It is interesting to note that theantibody recognizes neuromedin N, a COOH-terminally relatedpeptide which is encoded by the rat NT gene (Kislauskis et al.,1988), to a limited degree (


6/9

The Journal of Neuroscience, December 1989,9(12) 4435

200

ii'yj 150::n

5w 100

50 I2 3 4 5 6 7

L

IK+, K+/no Ca2+OK+, K+

*

Tr r igure 6. Dependence of NT releaseon calcium.Dialysisprobeswere placedinto the striatum, and after a stablebaselineperiod (epochs -3) 60 mM po-tassium was substituted for an equalamount of sodium (epoch4). Following2 further baseline amples epochs and6), 60 mM potassium wasagain substi-tuted for sodium in the presence openbars)or absence filled bars) of calciumin the ar tificial cerebrospind fluid per-fusion buffer (epoch 7). Bars representmean * SEM percentageof baseline n= 4). * p < 0.05 .supported by the anatomical findings in the present report andby our previous biochemical data (Bean et al., 1989).The approaches used to study the turnover of nonpeptidetransmitters, such as inhibition of specific synthesizing or de-grading enzymes, would not he appropriate for the estimationof peptide turnover hecause of the differences in the modes ofsynthesis and degradation of these 2 classesof neuroactive sub-stances. NT levels were examined after cycloheximide-inducedinhibition of protein synthesis in order to assessNT synthesis,This method has heen used previously to examine the effects

of haloperidol on substance P and NT synthesis (Bannon andGoedert, 1984; Frey et al., 1988). The lack of selectivity ofcycloheximide for the synthesis of particular proteins could bepotentially confounding if, for example, one neuropeptide mod-ulates the synthesis of another. However, the turnover of otherpeptides, such as substance P, estimated using the cycloheximidemethod agrees well with other indirect measures of their tum-over (Bannon and Goedert, 1984). In the present study, cy-cloheximide pretreatment had no significant effect on reserpine-induced elevation of striatal tissue levels of NT (Fig. 3). These

26..125 v 0. : : : : : : : I-1 0 1 2 6 4 6 6 7T nw 0-4

25

TIME (hours)

Figure 7. DA and NT release ollow-ing mserpineor vehicleadministration.Dialysis probes were placed into thestriatum of chloral hyclmte-anesthe-tized rats. Following establishment ofa stablebaseline or DA release insert),either reserpme (5 mg/kg, i.pJ or ve-hicle was administered, and sampleswere collected subsequently or 6 hr.BaselineDA levels were 2,47 + 0.16fmol/min, while NT levelswere 0.040f 0.004 fmol/min. V, time of vehicleor drug injection. * p -z 0.05.


7/9


data are therefore consistent with the observations of Frey etal. (1988), who noted that haloperidol pretreatment did not alterNT turnover. Thus, lowering the synaptic efficacy of DA, eitherthrough blocking DA receptors or inducing DA depletion, doesnot appear to alter striatal NT synthesis.Using immunohistochemica l techniques we have observedthat the reserpine-induced increase in striatal NT levels corre-sponds to an increase in the apparent density of striatal NT-liperikarya and fibers (Fig. 2), suggesting an increase in intraneu-ronal stores of the peptide. These data may explain why striatalNT levels, as measured by RIA, are moderately high, yet it isdifficult to demonstrate striatal NT-li with immunohistochemis-try (even after colchic ine pretreatment, see Fig. 1). Furthermore,these data may suggest that under normal circumstances NTmay be released from striatal neurons at such a rapid rate as toprevent the accumulation of sufficient peptide in either the cel lbody or neuronal processes o be visualized using immunohis-tochemical methods. Thus, the inhibition of neuronal NT re-lease by reserpine and the resultant accumulation of NT withinneurons may allow a more representative estimate of the actualdensity of striatal NT neurons and fibers.

NT-li perikarya were observed in the striata of animals treatedwith reserpine, even though these animals did not receive in-traventricular colchicine. The presence of a large number ofstriatal NT-li cell bodies was observed in every animal treatedwith reserpine, in contrast to the relative lack of striatal NT-lineurons (except for the medial periventricular stria) in animalssubjected only to intraventricular colchicine. It is difficult tocompare the distribu tions of NT- li fiber staining in the striataof animals treated with reserpine and those receiving only col-chicine pretreatment since colchicine frequently reduces fiberand terminal staining. Whi le it would be desirable to comparemore fully the characteristics of NT-li fiber and terminal stainingin animals treated with colchicine with those treated with re-serpine and colchicine, the latter treatment is invariably fatal.However, the cri tical difference between the 2 conditions issimply that striatal NT-li perikarya become apparent only underconditions of reduced dopaminergic tone (such as that elicitedby reserpine treatment) and are not seen when they are pre-treated even with large doses of colchicine (up to 250 clg in-traventricularly).The striatum is organized into histochemically defined com-partments in which some efferent and afferent striatal pathwaysbegin or terminate (Olson et al., 1972; Fuxe et al., 1979; Gray-biel, 1984; Gerfen et al., 1987). In this context, it was intriguingto note that the increase in density of striatal N T-l i followingreserpine treatment was accompanied by the appearance ofpatches of NT-li fibers (Fig. 2). These patches of dense NT-lifibers also contained perikarya and were found in both the dorsaland ventral striatum. The NT- li patches were surrounded byareas relat ively devoid of NT-B fibers, although some peptideperikarya were present. Previous histochemical observations byZahm and coworkers (Zahm, 1987; Eggerman and Zahm, 1988;Zahm and Heimer, 1988), Goedert et al. (1983), and Sugimotoand Mizuno (1987) have described the compartmentalizationof ventral s triatal NT-li into clus ters; these striatal NT-B patchesappear to correspond to zones which are acetylcholinesterasepoor and opiate receptor rich. Our data confirm the presenceof NT- li patches in the ventral striatum and extend these ob-servations to the dorsal striatum, a region in which only scat-tered NT-li neurons have been observed previously (Jennes etal., 1982; Uhl et al., 1982; Goedert et al., 1983; Zahm and

(+)@ T-PFigure 8. Theoret ical syn apt ic arrangement which may underlie str ia-tal DA-NT interact ions. For discussion, see text.Heimer, 1988). Additional study is required to define the dorsalstriata l NT patches in terms of other prev iously defined neu-rochemical markers and in order to begin to understand howto define NT with respect to the overa ll s triatal infrastructure.

Recent data concerning the alteration of striatal NT concen-trations by antipsychotic agents has prompted renewed exam-ination of the interactions of striatal DA and NT (Nemeroff,1986). The abil ity of acute administration of both typical andatypical antipsychotic agents to increase striatal NT levels isconsistent with our observations on reserpine. Reserpine is in-deed an effective antipsychotic agent (Slotkin, 1974) and thusthe similar ities between reserpine and the modem antipsychoticagents are interesting and suggest a common effect of these drugson striatal NT. However, the implications of these data forantipsychotic efficacy or side effects are not yet clear.Several lines of evidence support the hypothesis that DA maymodulate the release of striata l NT. Anatomica l studies haveshown that dopaminergic axons terminate on the dendrites andcell bodies of medium spiny striatal cells; among striatal me-dium spiny cell s are those containing GABA or NT (Ribak etal., 1979; Sugimoto and Mizuno, 1987). Pharmacological stud-ies have shown that administration of D2 antagonists increase,while D2 agonists decrease, striatal NT concentrations. Con-versely, D 1 antagonists decrease and D 1 agonists increase stria-tal NT tissue leve ls (Frey et al., 1986; Letter et al., 1987). Thedistinct regulatory roles of D 1 and D2 receptors on striatal NTneurons suggest certain ways through which DA transmissionmay influence the release of NT from striatal neurons.A working model of striatal DA-NT interactions which takesinto account previous anatomical and biochemical data as wellas the present data can be envisioned (Fig. 8). Thus, reductionof the synaptic efficacy of DA by blockade of D2 receptors mayinitiate the following sequence of events (refer to Fig. 8): (1) aloss of the inhibition exerted by DA on excitatory corticostriatalafferents. This may occur by a direct effect on D2 receptorslocated on corticostriatal neurons (Brown et al., 1983; Mercuriet al., 1985; Fill oux et al., 1988). Alternatively, DA afferentssynapse onto the shafts of dendritic spines, thereby inhibitingthe glutamatergic input which arrives onto spine heads (Freundet al., 1984). This would lead to (2) an increase in the influence


8/9

The Journal of Neuroscience, December 1999, 9(12) 4437

of excitatory corticostriatal afferents onto inhibitory GABAergicinterneurons (Cospito and Kultas-Ll insky, 1981), effecting (3)inhibition of the activity of NT cells, which in turn would resultin (4) a decrease in NT release into the ECF and (5) an increasein tissue levels of NT. Conversely, blockade of Dl receptorsmay be envisioned to release the inhibitory influence of DA onNT cells directly (Mercuri et al., 1985; Calabresi et al., 1987;Hu and Wang, 1988), leading to an increase in their activity;this would produce an increase in NT release, and therefore adecrease in tissue levels, consistent with previous data (Frey etal., 1986; Letter et al., 1987). The effects of reserpine (presentstudy) on striatal NT appear to be more similar to the increasein striatal NT levels evoked following D2 blockade, possiblyindicating a predominance of D2 tone on striata l NT neurons.Additionally, the effects of Dl and D2 agonist treatment areconsistent with the increase (Dl) and decrease (D2) in NT tissueleve ls predicted by this model (Merchant et al., 1988b). Al-though alternative mechanisms can be envisioned, this theo-retical scheme may serve as a convenient model and generateappropriate empir ical studies designed to test the hypothesis.

In the present study, we have examined the effect of reserpine-induced DA depletion on striata l NT. The data herein suggestthat DA depletion produces a decrease in NT release, resultingin the increase in striatal tissue levels of NT, as well as theapparent increase in striatal NT-li as measured by immunohis-tochemistry. We have proposed a working model which takesinto account the dissimilar ways in which DA, through Dl andD2 receptors, produces inhibition of striata l neurons. Such anarrangement may provide a framework with wh ich to examine(either at the receptor level or at the level of signal transductionsystems) the mechanisms that enable differential sensitiv ity ofstriata l NT neurons to occur following acute and chron ic treat-ment with selec tive D 1 and D2 receptor b lockers, effects whichmay be related to some effects of antipsychotic drugs (Frey etal., 1986; Nemeroff, 1986).

ReferencesBannon, M. J. , and M. Goedert (1984) Changes in substance P con-centration s afte r protein synth esis inhibition provide an index of sub-stance P ut i l izat ion. Brain Res. 301: 184-186.Bean, A. J. , T. E. Adrian, I . M. Modlin, and R. H. Roth (1989) Do-pamine and neurotensin storage in colocalized and noncolocalizedneuronal populat ions. J. Pharmacol. Exp. Ther. 249: 68 l-687.Berod, A., B. K. Hartman, and J. F. Pujol (198 1) Importance of f ixat ionin immunoh istochemistrv: Use of formaldehyde solut ions at variablepH for the localizat ion of tyrosine hydroxyiase. J. Histochem. Cy-tochem. 29: 844-850.Brown, J. R., and G. W. Arbuthnott (1983) The electrophysiology of

dopamine (D2) receptors: A study of the act ions of dopamine oncort icostriatal t ransmission. Neuroscience 10: 349-355.Butcher, S. P., I . S. Fairbrother, J. S. Kelly, and G . W. Arbuthnott(1988) Amphetamine-induced dopamine release in the rat str iatum:An invivo microdialysis study. J. Neurochem. 50: 346-355.Calabresi. P .. N. Mercuri. P. Stanzione. A. Stefani. and G. Bemardi(1987) Intracellular studies on the dopamine-induced firing inhibi-tion of neostriatal neurons in vitro: Ev idence for D 1 receptor involve -ment. Neuroscience 20: 757-77 1.Carraway, R ., and S. E. Leeman (1973) The isolation of a new hy-potensive pept ide, neurotensin, f rom bovine hypothalami. J. Biol.Chem. 248: 6854-6861.Carraway, R., and S. E. Leeman (1976) Characterization of radioim-munoassayable neurotensin in the rat. J. Biol. Chem . 251: 7045-7052.Church, W . H., and J. B. Just ice, Jr. (1987) Rapid sampling anddetermination of extracellular dopamine in vivo. Anal. Chem . 59:712-716.

Cospito, J. A., and K. Kultas-Ll insky (198 1) Synapt ic organization ofmotor corticostriatal projections in the rat. Exp . Neurol. 72: 257-266.De Quidt, M., and P. C. Emson (1983) Neurotensin faci l i tates do-pamine release in vitro from rat striatal slices. Brain Res . 274: 376380.Eggerman, K. W., and D. S. Z&m (1988) Numbers of neurotensin-immunoreact ive neurons selectively increased in rat ventral str iatumfollowing acute haloperidol admin istration. Neurope ptides I I: 125-II ?1 JL.Fil loux, F., T. H. Liu, C . Y. Hsu, M. A. Hunt, and J. K. Wamsley(1988) Selective cortical infarction reduce s [3H]sulpiride binding inrat caudate-putamen: Autoradiographic evidence for presynapt ic D2receptors on cort icostriate terminals. Synapse 2: 52 l-53 1.Freund, T. F., J. F. Powell, and A. D. Smith (1984) Tyrosine hy-droxylase-immunoreact ive boutons in synapt ic contact with ident i-fied striatonigral neurons with particular reference to dendritic spines .Neuroscience 13: 1189-1215.Frey, P., K. Fuxe, P. Eneroth, and L. F. Agnat i (1986) Effects of acuteand long-term treatme nt with neurolep tics on regional telencephalicneurotensin levels in the male rat. Neurochem. Int. 8: 429-434.Frey, P., M. Lis, and D. M. Coward (1988) Neurotensin concentrat ionsin rat str iatum and nucleus accumbens: Further study on their reg-ulat ion. Neurochem. Int. 12: 33-38.Fuxe, K., K. Andersson, R. Schwartz, M. Perez de la Mora, T. Hakfelt ,M. Ferland, and R. Tapia (1979) Studies on dif ferent type s of do-pamine nerve termina ls in the forebrain and their possible interactionswith hormones and with neurons containing GABA, glutamate, andopioid pept ides. Adv. Neurol. 24: 197-215.Gerfen, C. R., M. Herkenham, and J. Thibault (1987) The neostriatalmosaic: I I . Patch- and matrix-directed mesostxiatal dopaminergic andnondopaminergic system s. J. Neurosci. 7: 39 15-3934.Goedert, M., R . W. Mantyh, S. P. Hunt, and P. C. Emson (1983)Mos aic distribution of neurotensin-like imm unore activity in the catstr iatum. Brain Res. 274: 176-179.Goedert, M., S. D. Iversen, and P. C. Emson (1985) The effects ofchronic neuroleptic treatm ent on neurotensin-like immu norea ctivityin the rat central nervous system . Brain R es. 335: 334-336.Govoni, S., J. S. Hong, H. Y. T. Yang, and E. Costa (1980) Increaseof neurotensin content el ic ited by neurolept ics in nucleus accumbens.J. Pharmacol. Exp. Ther. 215: 413-417.Graybiel, A. M. (1984) Correspondence between the dopamine islandsand striosomes of the mamma lian str iatum. Neuroscience 13: 1157-1187.Hetier, E., A. Boireau, P. Dubedat, and J. C. Blanchard (1988) Neu-rotensin effects on evoked release ofdopamine in slices from striatum,nucleus accumbens, and prefrontal cortex in rat. Naunyn Schmie-debergs Arch. Pharmacol. 337: 13-l 7.Hu, X.-T., and R. Y. Wang (1988) Comparison of effects of D-l andD-2 dopamine receptor agon ists on neurons in the rat caudate pu-tamen: An electrophysiological study. J. Neurosci. 8: 4340-4348.Imperato, A., and G. DiChiara (1984) Trans-striatal dialysis coupledto reverse phase high performance l iquid chromatography with elec-trochemical detect ion: A new method for the study of the in vivorelease ofendogenous dopamine and metabolites. J. Neurosci. 4 : 966-977.Jennes, L., W. E. Stumpf, and P. W. Kalivas (1982) Neurotensin:Topographical distribut ion in rat brain by immunohistochem istry. J.Comp . Neurol. 210: 21 l-224.Kislauskis, E., B. Bullock, S. McNeil, and P. R. Dobner (1988) Therat gene encoding neurotensin and neuromedin N: Structure, t issuespecif ic expression, and evolut ion of exon sequ ences. J. Biol. Chem.263: 4963-4968.Letter, A. A., L. A. Matsuda, K. M. M erchant, J. W. Gibb, and G. R.Hanso n (1987) Chara cterization of dopam inergic influence on stria-tal-nigral neurotensin system s. Brain Res. 422: 200-203.Lindefors, N., E. Brodin, and U. Ungerstedt (1987) Microdialysiscombined with a sensit ive radioimmunoassay: A technique for study-ina in vivo release of neuroDeptides. J. Pharmacol. Methods 17: 305-3i2. _ _Lowry, 0. H ., N. J. Rosebrough, A. L. Farr, and R. J. Randall (195 1)Protein measurement with the fol in nhenol reaaent. J. Biol. Chem.193: 265-275.Merchant, K. M., L. Bush, J. W. Gibb, and G. R. Hanson (1988a)Rece ptor-sp ecific in teractions of the nigro-striatal dopamine projec-


9/9


tions with striatal neurotensin systems n rat brain. Sot. Neurosci.Abstr. 14: 114.Merchant, K. M., A . A. Letter, J. W. Gibb, and G. R. Hanson (1988b)Changes n the limbic neurotensin systemsnduced by dopaminergicdrugs.Eur. J. Pharmacol. 153: l-9.Mercuri, N., G. Bemadi, P. Calabresi,A. Contugno, G. Levi, and P.Stanzione (1985) Dopamine decreasesell excitability n rat striatalneuronsby pre- and postsynapticmechanisms.Brain Res.358: 1 O-121.Merlis, J. K. (1940) The influence of cembrospinal luid calcium ona spinal flexion reflex. Am. J. Phys iol. 129: 422423.Nemeroff, C. B. (1986) The interaction of neurotensin with dopami-nergic pathways in the central nervous system:Basic neurobiologyand mplications for the pathogenesis nd treatment of schizophrenia.Psychoneuroendocrinology 1: 15-37.Olson, L., A. Se iger, nd K. Fuxe (1972) Heterogeneityof striatal andlimbic .dopam& innervation: Highly fluorescent slands in devel-onina and adult rats. Brain Res. 44: 283-288.Pa&o;, G., and C. Watson (1986) The Rat Brain in StereotaxicCoordinates,Academic, New York .Quirion , R., C. C. Chiueh, H. D. Eve&t, and A. Pert (1985) Com-parative localization of neurotensin receptors on nigrostriatal andmesolimbic dopaminergic terminals. Brain Res. 327: 385-389.Ribak, C. E., J. E. Vaughn,and E. Roberts (1979) The GABA neuronsand their axon terminals in rat corpus striatum as demonstrated byGAD immunohistochemistry. J. &imp. Neurol. 187: 261-284. -Slotkin.T. A. (1974) Resemine. n Neuroooisons:Their Pathoohvsio-log&lActio&, L. i. Simkon and D. R:Curtis, eds.,pp. l-60,.Ple-num, New York.Sugimoto, T., and N. Mizuno (1987) Neurotensin in pro jection neu-

rons of the striatum and nucleusaccumbens,with particular referenceto coexistence ith enkephalinand GAB A An immunohistochemicalstudy in the cat. J. Comp. Neurol. 257: 383-395.Uhl, G. R. (1982) Distribu tion of neurotensin and its receptor n thecentral nervous system.Ann. NY Acad. Sci . 400: 132-149.Uhl, G. R., M. J. Kuhar, and S. H. Snyder (1977) Neurotensim Im-munohistochemical ocalization in the rat central nervous system.Proc. Natl. Acad. Sci. USA 74: 405wO63.Unaerstedt. U.. and A. Hallstrom 11987) In vivo microdialvsis-A&w approach to the analysisof neurotransmitters n the b&n. LifeSci . 41: 861-864.Young III, W. S., and M. J. Kuhar (1981) Neurotensm receptor o-calization by light m icroscopicautoradiography in rat brain. BrainRes. 206: 273-285.Zahm, D. S. (1987) Concentration of neurotensin immunoreactiveneurons in the ventral striatum of the rat: Ventromedial caudate-putamen, nucleusaccumbens,and olfactory tubercle.Neurosci. Lett.81: 41-47.Zahm, D. S., and L. Heimer (1988) Ventral striatopal lidal parts of thebasalganglia in the rat: I. Neurochemical compartment&ion as re-flected by the distributions of neurotensin and substanceP immu-noreactivitv. J. Como. Neurol. 272: 5 16-535.Zetterstrom, T., and U: Ungerstedt (1984) Effectsof apomorphine onthe in vivo eleaseof dopamine and its metabolites, studied by braindialysis.Eur. J. Pharmacol. 97: 29-36.Zetterstrom, T., T. Sharp, C. A. Marsden, and U. Ungerstedt (1983)In vivomeasurementof dopamine and its metabolitesby intracerebraldialvsis : Chanaesafter d-amnhetamine. J. Neurochem. 41: 1769-1773.

andrew j. bean et al- effects of dopamine depletion on striatal neurotensin: biochemical and...

Documents