Judith D. Schaechter and Richard J. Wurtman- Tryptophan Availability Modulates Serotonin Release from Rat Hypothalamic Slices

8/3/2019 Judith D. Schaechter and Richard J. Wurtman- Tryptophan Availability Modulates Serotonin Release from Rat Hypot

1/9

Journal of NeurochemistryRaven Press, Ltd., New York0 989 International Society for Neurochemistry

Tryptophan Availability Modulates Serotonin Releasefrom Rat Hypothalamic SlicesJudith D. Schaechter and Richard J. Wurtman

Department of Brain and Cognitive Sciences, Massachusetts Institute of Technology, Cam bridge, Massachusetts, U S A .

Abstract: Application o f a novel in vitro experimental systemhas allowed u s to describe the relationship between trypto-pha n availability an d sero tonin release from rat hypothalamicslices. Superfusing hypothalamic slices with a physiologicmedium conta ining / - tryptophan (1, 2, 5 , or 10pM ) causeddose-dependent elevations in t issue tryptophan levels; themagnitude o f the elevations produced by supplementing themedium wi th


2/9

1926 J. D. SCHAECHTER AND R. J . R'URTM4h1976) an d increase dayt im e drowsiness (Smith a ndProckop, 1962; Gree nwoo d e t a l . , 1975) and decreasenight t ime s leep la tency (Har tman and Spinweber .1979) in hum ans . Rats which ra ise thei r b ra in t ryp-tophan levels by consuming a carbohydrate-rich meal(Ferns t rom an d W urtm an, 197 16) choose to ea t lesscarbohydrate, relative to protein, at the subsequentmea l (Wur tm an et al. , 1983). Oral tr ypto pha n also re-portedly reduces food inta ke in lean m en (L eiter et al .,1987), facilitates weight loss in obese patients (Heraiefet al., 1985), an d elevates mo od in depressed persons(Lapin and Oxenkrug , 1969).These behav ioral effects suggest that physiologic an dpharmacologic t rea tmen ts which increase bra in t ryp-tophan levels thereby e nhanc e 5 -HT synthes is and fa-cilitate 5-HT-mediated neurotransmission. Evidencefo r a physiologic coupling between brain tryptophanlevels and 5-H T release has also been provided by so meexperimen ts performed in vivo. Trea tme nts which el-evated bra in t ry ptophan levels increased th e am oun tsof 5-HT released in to the m edia perfusing the rat lateralventricle (Ter nau x et al . , 1976), the surface of th e catcaudate nucleus (Ternaux e t al., 1977) . and t he ra t nu-c leus accumben s (G ua n and McBride, 1987) . In vivovol tammetr ic exper iments , which are now in terpre tedas reflecting the sum me d extracellular levels of 5 - H Tan d 5-hydroxyindoleacetic acid (5-HIA A) (Joseph an dKennet t , 1981 ; Bau ma nn a nd Waldmeier , 1984). in -dicate that increases in brain tryptophan accelerate 5-hydroxyindole re lease in the ra t h ippoca mpus du r ingphys iologic s t ress ( Joseph an d Kenn et t . 1983) and inthe rat s tr iatum with electr ical stimulation of the dorsalraphe nucleus (De Sim oni e t a l . , 1987). Und er basa lcondi t ions , t ryp tophan ad minis t ra t ion has been ob-served to produce elevations in extracellular 5-hy-droxyindole levels in t he r a t h ippocampus ( J os eph andKennet t , 1981) , bu t no t in the ra t s t r ia tum (Marsde net a l. , 1979; De Simo ni e t al., 1987). In vitro s tudiessimilarly have yielded conflicting results. A dep ende nceof 5-HT release on brain tryptophan availabili ty wasobserved in a study using incubated rat hippocampalslices from which neurotransmitter release was evokedby a high potass ium medium (Auerbach and Lip ton ,1985), but not in o ne using incubated whole-brain s liceselectr ically s timulated a t a high ra te (100 H z) (Elks e tal., 19796).In terpre ta t ion of such s tudies som et imes has beenconfounded by the use of exper imenta l approacheswhich ma y no t reflect accurately t he physiologic pro-cesses which control 5-HT release. For example, ele-vations in brain tryptophan levels com mo nly have beenind uce d by giving a pharmacologic dose of t ryptoph an(i.e., 100 mg/kg), even though this dose elevates thelevel of brain tryptop han approximately ninefold (M oirand Eccles ton , 1968; Graha me-S mith , 197 ) , an in -crease that is well beyond its physiologic range, anddepresses the f iring rate of dorsal raphe serotoninergicneurons (Ag hajanian, 1972; Gallager and Aghajanian.1976). Th e characteris tic s low fir ing rate of these ne u-

rons (Aghajanian e t al., 1968: M c G i n t y a n d H a r p e r ,1976) is not well s imu lated by t he application of a highpotass ium med ium or by the delivery of high-frequencyelectr ical pulses , methods often used to evoke 5 -HTrelease from brain tissue in vivo an d in vitro.W e have des igned a novel in vitro exp enm ent al sys-tem which ha s allowed us to reexamine th e relationshipbetween t ryp tophan availab il i ty an d 5-H T re lease un-der experimental conditions perhaps more reflectiveof those occurring physiologically.

MATERIALS AND METHODSSlice preparationMale Sprague-Dawley rats ( I 80-240 g) were killed by de-capitation: two rats were used in each experiment. Each brainwas rapidly removed from t h e skull and immersed in ice-chilled physiologic medium (previously gassed with 95%02/5% COz)of th e following composition (in mM ): NaCI, 130;KCI, 3.5; CaCI2, 1.3: MgS0 4. 1.5; N a H 2 P 0 4 ,1; N a H C 0 3 ,25 : d-glucose. 10. The hypothalamus was dissected out inblock ( 5 X 5 X 2 mm ) and cut in halfalong the third ventricle.T h e demihypothalami were cut simultaneously into 300-pmcoronal slices using a Mcllwain tissue slicer, and then quicklysubmerged in chilled medium. Each slice was isolated se-quentially using a fine sable brush, and placed alternatelyinto one o f t w o glass tubes containing chilled medium. S omeof the slices, approximately every fifth, were retained, frozenover dry ice, stored at -70C. and used subsequently forbiochemical assays. The hypothalamic slices in each of thetubes were transferred into parallel superfusion chambers;thus . each chamber contained two demihypothalami. Thechambers were constructed as described by Milner andWurtman (1984). t hough modified by the placement of anylon disc on th e bottom electrode on which the slices rested.The slices were equilibrated by superfusion (0.6 ml/m in) for50 min at 37C with physiologic medium which was gassedcontinuously with 95% 0 2 / 5 % C02 to maintain pH at 7.4.This superfusion medium included th e drug fluoxetine hy-drochloride ( 2 phe provided by Eli Lilly Laboratories, In-dianapolis. IN , U . S . A . ) o block the reuptake of synaptic 5-HT (Wong et al., 1974). (I n th e absence of fluoxetine, 5-HTreleased spontaneously into th e medium was not detected,and only about 20% of the 5-HT released with electrical fieldstimulation was recovered.)Release experimentsIn experiments designed to examine the relationships be-tween tryptophan availability and 5-HT release, the mediumsupcrfusing th e hypothalamic slices in o n e of the tw o cham-bers was supplemented with /-tryptophan (1-10 pM , SigmaChemical Co., St. Louis, M O , U.S.A.). Five-minute fractionswere collected for 80 min. between 50 and 130 min, beyondthe initial equilibration period. The slices were electricallystimulated for three periods of 4.7 min each, starting at 60,85. and I10 min from the onset of superfusion. Electricalfield stimulation was induced by delivery of bipolar square-wave pulses ( 5 H z, 2 ms, 100 mA/cm 2) using a stimulator(Grass Instruments Stimulator S88: Quincy, M A , U.S .A . ) nseries with a 12-V relay: pulses were monitored continuouslyby an oscilloscope (Hitachi V-2 17; Tokyo, Japan). A t the endof each experiment the slices were removed from their cham-bers, quickly rinsed i n distilled water, frozen over dry ice,and stored at -70C for subsequent biochemical assays.

J . Neurochem., Vol. 53.A??.6, 1989


3/9

TRYPTOPHAN AVAILABILITY MODULATES 5-HT RELEASE 1927In experiments addressing the calcium dependence of 5-

HT release, the slices in one of the two chambers were exposedto calcium-free medium, prepared without CaC12 and con-taining EGTA (1 &, Sigma), from the onset of superfusionuntil 85 min; from 85 to 135 min, the slices were superfusedwith the usual physiologic medium (1.3 mM CaCI2). Slicesin the other chamber were superfused with the physiologicmedium throughout. Fractions were collected every 5 minbetween 50 and 85 rnin and between 100and 135 min. Elec-trical field stimulation (using the same parameters as above)was delivered starting at 65 rnin (SI ) and again at 115min (S2).

To test whether 5-HT release from the rat hypothalamicslices required the activation of voltage-sensitive sodiumchannels, we monitored the effect of the voltage-sensitiveso-dium channel inhibitor, tetrodotoxin (TTX; l pM , Sigma),on basal and electrically evoked 5-HT release. Slices in thetwo chambers were superfused initially with physiologic me-dium; between 30 and 85 min, the slices in one of the cham-bers were superfused with medium to which TTX had beenadded. Slices in both chambers were then washed rapidly (1ml/min) for 1 h (85-145 min) with physiologic medium, andthen superfused at the usual rate for the remainder of theexperiment. Fractions were collected at 5-min intervals be-tween 50 and 85 rnin and between 145 and 180 min. Theslices were electrically stimulated (parameters as above) be-ginning at 65 rnin (S l) and 160 rnin (S2).

Each fraction of superfused medium was collected in 100pI of 7 mM ascorbate containing two internal standards: 5-hydroxy-N-methyltryptamineoxalate (5-HT-CH3) and Sh y-droxy-2-indolecarboxylic acid (5-HICA) (both purchasedfrom Aldrich Chemical Co., Milwaukee, WI, U.S.A.). Oncecollected, each fraction was mixed by vortex and stored inthe dark, on ice, until undergoing further processing.Biochemical analysis

Each fraction of superfused medium was passed througha preparative column to concentrate 5-HT and 5 -H IM ; thesecolumns were prepared by loading 100 mg of dry CI8 everse-phase sorbent (40 pm; Analytichem International, HarborCity, CA, U.S.A.) into glass wool-plugged Pasteur pipettes (9in). The columns were conditioned with I .5 ml of methanolfollowed by 0.75 ml of 0.1 M NaH 2P04 (pH 3.0). Samplefractions and standard fractions, prepared with knownamounts of 5-HT and 5-HIAA, were brought to pH 2.8-3.0with 1 O M HCl. After a sample passed through its column,the aqueous phase in the co lumn was displaced by 125 pl of70% methanol/30% acetic acid. The 5-hydroxyindoles wereeluted into Eppendorf tubes with 300 pl of this organic so-lution, and the solvent in each tube was evaporated under astream of nitrogen. The dried eluates were reconstituted with50 pl of 0.15 M HC1 containing 0.25 mM ascorbate. Recov-eries of 5-HT and 5-HIAA were generally 80-90%, based oncalculations using the internal standards.

The amounts of 5-HT and 5-HIAA in the reconstitutedsamples were assayed by HPLC with electrochemical detec-tion. The samples (45pl of 50 pl) were automatically injected(Waters Intelligent Systems Program; Milford, MA, U.S.A.)over a reverse-phase C18 column (5 pm, 25 cm; BeckmanInstruments, San Ramon, CA, U.S.A.). The mobile phasewas of the following composition (in mM): N aH 2P04 , 00;Na,-EDTA, 0. I ; octyl sodium sulfate, 0.17; with 13% meth-anol (vol/vol) and having a final pH of 4.3. The substanceswere detected electrochemically (Model LC4A; BioanalyticalSystems Inc., West Lafayette, IN, U.S.A.) at 2 nA/V when

the potential of the glassy carbon electrode was set at 0.55 Vagainst the Ag/AgCl reference electrode. Tissue levels oftryptophan, 5-HT, and 5-HIAA were assayed using thischromatographic system, although the applied potential wasset at 0.85 V, in order to detect tryptophan in the samples,and the sensitivity at 5 nA/V. Frozen tissue samples weresonicated in 0.2 M HC104 (approximately 0.4 ml/mg of pro-tein), containing 0.5 mM ascorbate and internal standards,and centrifuged (35,000 g, 10 min). Aliquots (50 p1 in du-plicate) of these tissue supernatants were injected over thereverse-phase column. The tissue pellets were assayed for theirprotein content by the method of Lowry et al. ( 1 95 1) usingbovine serum albumin as the standard protein.Data analysisAmounts of the indoles in each sample of superfusion me-dium and tissue supernatant were estimated by correctingthe recorded peak height for its recovery, using the designatedinternal standard (5-HT-CH3 for 5-HT and tryptophan; 5 -HICA for 5-HIAA), followed by linear regression analysisbased on standard curves run in parallel with each set ofsamples. These amounts were normalized to the amount ofprotein in the tissue pellet, allowing data to be expressed aspicomoles of 5-HT or 5-HIAA per gram of protein per minutefor the rates of release, or nanomoles of tryptophan, 5-HT,or 5-HIAA per gram of protein for levels within the slices.Values are reported here as means * SEM.Calculations of the average rates of 5-HT release duringthe four rest periods (spontaneous 5-HT release) and the threeperiods of electrical stimulation (evoked 5-HT release) weremade for each experiment in which the effect of tryptophanavailability on 5-hydroxyindole release was to be examined.The rate of 5-HIAA efflux was taken as an average across the80-min collection period. Effects of increasing the tryptophanconcentration of the superfusing medium on slice indole levelsand release were evaluated using the two-tailed Studentspaired t test. Statistically significant differences were notedfor these and all other analyses when the probability valuewas less than 0.05. Dose-dependent relationships betweentryptophan availability and slice 5-HT levels and release wereanalyzed by one-way analyses of variance, followed by theDuncans multiple range post hoc test.The calcium dependence and TTX sensitivity of sponta-neous 5-HT release were assessed by comparing the averagerates of 5-HT release from slices superfused with calcium-free or TTX-containing medium versus release occumng inthe presence of the physiologic medium. The calcium de-pendence and TTX sensitivity of evoked 5-HT release wereevaluated by comparing the rates of 5-HT release due to de-livery of S 1, when the slices were being superfused with alteredmedia, versus release in the presence of the usual physiologicmedium. Restoration of evoked 5-HT release was determinedby comparing the amounts of 5-HT released from slices inthe two chambers due to S2. These data were evaluated bythe F test for equality of two variances, and then by the two-tailed Students t test with equal or unequal variances whereappropriate.

RESULTSThe levels of tryptophan, 5-HT, and 5-HIAA in the

hypothalamic slices prior to superfusion were 167.8* 4.8, 73.1 -t 2.3, and 46.0 k 1.6 nmol/g of protein,respectively. Following 130 rnin of superfusion withtryptophan-free medium (including three periods ofJ. Neurochem., Val.53, o. 6 , 1989


4/9

1928 J . D . SCI.4ECHTER AND R. J . WURTMAN

-.-E1 6 0 0 -za

electrical stimulation), these levels were IS 1.7 f .0,6 1. 1 t .9, and 4.6 +-0.2 nmol/g of protein. The actualtryptophan concentration in the tryptophan-freemedium 15 min after the start of superfusion was ap-proximately 0.1 pM, and this concentration declinedsteadily over the next 115 min to approximately 0.05pA4 (data not shown). Endogenous 5-HT was releasedspontaneously from the slices at a rate of 4 I k 2 pmolfg of protein/min (Fig. IA). This rate was increased to2 13 & 7 pmol/g of protein/min for 10 min with eachof the three periods of electrical field stimulation. afterwhich 5-HT release returned to its basal rate. The rateof 5-HIAA efflux was an average of 460 -t 23 pmol/gof protein/min across the 80-min collection period.which included transient poststimulation elevations(Fig. IB). The molar ratio of 5-HT to 5-HIAA in thesuperfusion medium when the slices were at rest wasapproximately 0.09; with electrical stimulation, thisratio rose to approximately 0.40.

The electrically evoked release of 5-HT from thehypothalamic slices was dependent on the presence ofcalcium in the superfusion medium (Fig. 2). Slices su -perfused with calcium-free medium containing EGTA( 1 mM) released only 18 t 2% as much 5-HT withelectrical field stimulation as control slices ( p < 0.01).In contrast, basal 5-HT release was unaffected by the

B

2 5005 400

2 300Im- t0050 60 70 80 90i0

Superfusion t i me (rnin)electr ical stirnulotion

FIG. 1. Time course of 5-HT release (A) and 5-HIAA efflux ( 6 ) romrat hypothalamic slices superfused n tryptophan-free physiologicmedium. Fractions of superfusion medium were collected every 5min after an initial equilibration period, and the amounts of 5-HTand 5-HIAA released (pmol/g of proteinlmin) were monitored. Theslices were electrically field stimulated for three periods. n = 32.

a-: 00 1

+o L0

0 orm01 mediumD medium * Caf+

FIG. 2. Calcium dependence of electrically evoked 5-HT release.Rat hypothalamic slices were superfused nitially with either phys-iologic medium or calcium-free medium containing 1 mM EGTA,and were delivered the first period of electrical field stimulation(Sl )under these conditions. Physiologic medium replaced the cal-cium-free medium prior to the second period of electrical stimulation(S2 ) .The amounts of 5-HT released (pmol/g of protein/min) weremonitored. Values are group means i EM for n = 9. *p< 0.01,differs from control by the Students test with unequal variances.

lowered calcium concentration (43 -t 3 and 35 t 4pmol/g of protein/min with physiologic and calcium-free media, respectively). When slices which had beenexposed to the calcium-free medium were superfusedsubsequently with physiologic medium (calcium con-centration = I .3 mM), the electrically evoked releaseof 5-HT returned to control values.

Evoked 5-HT release was sensitive to the activity ofvoltage-dependent sodium channels (Fig. 3) . Inhibitingtheir activation with TTX ( I pM ) caused a 60 t %reduction ( p < 0.05) in 5-HT release with electricalstimulation, again without changing basal 5-HT release

0 ormal mediumm medium i TxT

* r+m kl l Xs1 s2Stimulat ion periodFIG. 3. TTX sensitivity of electrically evoked 5-HT release. Rathypothalamic slices were superfused nitially with physiologic me-dium; the medium superfusing one of the two parallel chamberswas replaced with TTX-containingmedium (1 pM) prior to deliveryof the first period of electrical field stimulation(SI).lices of bothchambers were then superfused rapidly (1 ml/min) with physiologicmedium before the second period of electrical stimulation (S2).Values are group means k SEM or n = 3. *p < 0.05, differs from

control by the Students test with unequal variances.

J.Neurochem , Vol 53, o . 6 , 1989


5/9

TRYPTOPHAN A VAILABILITY MODULATES 5-HT RELEASE I9 2 9

C

(3 9 k 2 and 38 k 4 pmol/g of protein/min with phys-iologic and TTX-containing media, respectively). Therate of evoked 5-HT release was partially recovered, to80 k 2% of control, after the slices were rapidly super-fused for 1 h with physiologic medium.

Tryptophan levels, measured in the slices at the endof each experiment, increased in a dose-dependentmanner with the addition of tryptophan ( I - 10 yM ) tothe superfusion medium (Fig. 4A). Slice tryptophanlevels increased by 40 f % ( p < 0.01) 105 k 8% ( p< O .O l ) , 223 k 25% ( p < 0.01), and 490 k 30% ( p< 0.01) when 1,2, 5 , or 10 pMtryptophan, respectively,was added to the superfusion medium. Final tissue lev-els of 5-HT rose biphasically when the medium tryp-

1000 .800 .

2 00 .400

200 .

5F

0 1 2 5 10

tophan concentration was increased (Fig. 4B): Lowtryptophan concentrations caused proportionatelygreater incremental changes in tissue 5-HT levels [ Iy M 8 k 3% over control; 2 yM: 13 k 3% ( p < 0.01)over control] than higher tryptophan concentrations[ 5 y M 19 f % ( p < 0.01) over control; 10 pM 34k % ( p < 0.01)over control]. The amount of 5-HIAAremaining in the slices following superfusion also ex-hibited a dose-dependent relationship to the tryptophanconcentration of the superfusion medium (Fig. 4C).Spontaneous and electrically evoked release of 5 -HT from the slices increased in a dose-dependentmanner with the medium tryptophan concentration(Table 1): Addition of 1, 2 , 5, or 10 pM tryptophan tothe medium elevated spontaneous 5-HT release by 8k 4%, 36 f % ( p < 0.01), 47 f % ( p < 0.01), and67 f 3%( p < O.Ol), respectively. Electrically evokedrelease of 5-HT was elevated by 5 k %, 19 f % ( p< O.O l ) , 34 k 6% ( p < O.Ol ) , and 59 f 0% p < 0.01)in the presence of these tryptophan concentrations. Theefflux of 5-HIAA increased by 21 t % ( p < 0.01), 47k9%( p


6/9

1930 J . D. SCHAECHTER AND R. J . WURTMANTABLE 1. Efect of tryptophan availability on release of5-hvdroxvindoles from rat hypothalamic sliees

5-HT release( P W Basal Evoked efflux

Tryptophan 5-HIAA

38 t 219 f 15 46 0 f 941 t 3 23 2 f 8 554 f 6"(108 +- 4) (105 f ) (I21 f )44 t 4 221 f 13 510 f 360 t " 260 f 6" 74 0 f 7"(136 f ) (119 f ) ( I47 t )40 t 207 f 2 44 8 f 958 t " 274 f 1" 91 7 f 16Sb(147 t ) (134 k 6 ) (206 * 17 )44 t 202 f 2 41 3 f 072 f " 319 2 24" 985 f 138"(167 4 13) (159 -t 10) (238 f 19 )

Rat hypothalamic slices were superfused in physiologic mediumwhich either contained no exogenous tryptophan or was supplementedwith tryptophan ( I , 2, 5, or 10 phi'). The amounts of 5-HT released,basally and with electrical field stimulation (5 Hz, 2 ms, 100 mA/cm'), and 5-HIAA emuxed from the slices are expressed in pmol/gof protein/min. Values in parentheses are the relative amounts, inpercent, of the 5-hydroxyindole released from slices superfused withtryptophan-supplemented medium as compared to slices superfusedwith tryptophan-free medium. Data are given as group means f EMfor n = 8- I2 pairs." p< 0.01; ' p < 0.05, differs from control group by Student's pairedI test.

and 3) . This suggests that the influx of calcium ionscaused directly by stimulation-induced activation ofvoltage-sensitive calcium channels may be greater thanthat caused indirectly by sodium influx-induced mem-brane depolarization. Basal 5-HT release was not al-tered by the omission of calcium from the superfusionmedium nor by the addition of TTX to the medium.This differential sensitivity of basal and evoked 5-HTrelease to calcium or to TTX has been observed pre-viously by others (Elks et al., 1 9 7 9 ~ ; othert, 1980;Schlicker et al., 1985), and suggests that spontaneous5-HT release from brain nerve terminals is regulatedby factors other than the influx of calcium ions throughvoltage-sensitive calcium channels.

Slice tryptophan levels were maintained over the su-perfusion period in the absence of exogenous trypto-phan, probably reflecting the fact that only a relativelysmall proportion (approximately 2%) of brain trypto-phan is converted to 5-HT in situ (Pardridge, 1977),and the turnover of tryptophan-containing protein wasat steady state. Final 5-HIAA levels were markedlylower than those measured in presuperfusion tissue(46.0k 1.6 as compared to 4.6* 0.2 nmol/g of protein).The utilization of the 5-HT reuptake inhibitor, fluox-etine, in the superfusion medium presumably pre-vented synaptic 5-HT from being recaptured into se-rotoninergic nerve terminals, and thereby diminished

the contribution to tissue 5-HIAA levels of metabolized5-HT which had been released previously (Reinhardand Wurtman, 1977). Additionally, those 5-HT mol-ecules which were degraded intraneuronally (i.e., afraction of newly synthesized 5-HT molecules) wereprobably transported, via an acid transporter, acrossthe neuronal membrane into the extracellular spaceand removed by the superfusing medium. The activityof this transport process most likely accounts for therelatively low levels of 5-HIAA that remained in theslices and the high concentrations of 5-HIAA detectedin the superfusion medium.

Although the inclusion of a 5-HT reuptake inhibitorin the superfusion medium was necessary to detect re-leased 5-HT reliably, the resulting elevation in synaptic5-HT may have activated serotoninergic autoreceptorson hypothalamic nerve terminals (Cem to and Raiteri,1979), and thus suppressed 5-HT release. Indeed, theaddition of a nerve terminal autoreceptor antagonist,methiothepin (Gothert, 1980), to the medium (in thepresence of fluoxetine) potentiated spontaneous andelectrically evoked 5-HT release (data not shown). Ac-tivation of these autoreceptors may have caused anunderestimation of the magnitude of the effect of tryp-tophan supplementation on 5-HT release.

100 110 120 130 140Final slice 5-HT content (% control)

FIG. 5. Dose-dependent relationships between changes in finaltissue 5-HT levels and in spontaneous (A) and electrically evoked(6 ) -HT release when the slices were superfused with tryptophan-supplemented medium (1 , 2, 5, or 10 fiM)as compared to tryp-tophan-free medium (control). These are replottingsof data fromFig. 46 and Table 1. Data are group means 2 SEM for n = 8-12pairs. Analyses of variance detected significantly different effectswithin levels of tryptophan supplementation for changes in 5-HTlevels (p< 0.01). spontaneous 5-HT release(p< O .O l ) , and evoked5-HT release (p < 0.01).

J.Neurochem., Vol. S3 , No. 6 , 1989


7/9

T R Y P T O P H A N A VAlLABILITY MODULATES 5 - H T RELEASE 1931Electrically stimulating the slices transiently en-

hanced 5-HIAA efflux, but this effect was delayedcompared with the immediate increase in 5-HT release(Fig. I) . The depolarization of serotoninergic neuronshas been shown previously to accelerate 5-HT synthesisin vivo (Shields and Eccleston, 1972; Herr et al., 1975)and in vitro (Elks et al., 1979b; Hamon et al., 1979).Conceivably, this newly synthesized 5-HT was releasedinto the synapse during periods of electrical field stim-ulation, but was metabolized to 5-HIAA after stimu-lation ceased.

The tryptophan-free medium actually containedlow concentrations of tryptophan which declined overthe superfusion period, from approximately 0.1 to 0.05pM . The presence of endogenous tryptophan in themedium possibly reflects the physiologic flux of aminoacids across plasma membranes (Parfitt and Grahame-Smith, 1973) and an artifactual elevation in free aminoacid levels due to protein breakdown associated withpreparation of the slices. The time-dependent decreasein medium tryptophan levels probably resulted fromthe continua! dilution of this tryptophan-free me-dium with fresh (truly tryptophan-free) medium. Add-ing exogenous tryptophan to the medium at concen-trations within the range others have measured in ce-rebrospinal fluid of untreated rats (1-3 pM ) (Sarna etal., 1983; Hutson et al., 1985; Anderson et al., 1987)caused elevations in tissue tryptophan levels quanti-tatively similar to those which occur physiologically inrat brain, e.g., with exercise (Chaouloff et al., 1985,1986), certain stressors (Knott et al., 1973; Kennettand Joseph, 1981; Culman et al., 1984; Dunn, 1988),ingestion of a carbohydrate-rich meal (Fernstrom andWurtman, 1971b;Colmenares et al., I975), or diurnalvariation (Morgan et al., 1975; Hery et al., 1977).

The observed relationship between increased tryp-tophan availability and tissue 5-HT levels (Fig. 4B) isindicative of the particular kinetics of tryptophan hy-droxylase, and probably reflects the sensitivity of 5-HTbiosynthesis to changes in the degree of saturation oftryptophan hydroxylase with its amino acid substrate.When tissue tryptophan levels were elevated by super-fusing the slices with media containing 1 or 2 pMtryp-tophan, 5-HT levels increased proportionately; how-ever, superfusing the slices with 5 or 10pM tryptophancaused smaller incremental rises in 5-HT levels. No-tably, this inflection in the curve relating mediumtryptophan concentration to 5-HT levels occurredwhen tissue tryptophan levels were between approxi-mately 40 and 60 p M , concentrations which boundthe K,,, value of tryptophan hydroxylase for the aminoacid [SO pM (Fnedman et al., 1972)]. [These calcula-tions assume that tryptophan is distributed uniformlythroughout the tissue (Wurtman and Fernstrom,1976).] Such levels were attained when the slices weresuperfused with 2 and 5 pM tryptophan, respectively.

Elevating slice tryptophan levels within their phys-iologic dynamic range caused proportionate increases

in 5-HT levels and in the spontaneous and electricallyevoked release of the neurotransmitter. This observa-tion supports the view that a physiologic coupling doesexist between tryptophan levels, 5-HT synthesis, and5-HT release in the rat hypothalamus. This precursordependence of 5-HT release may allow the brain tosense metabolic processes which alter brain tryptophanlevels. This property of serotoninergic neurons may beutilized, for example, to determine the proportion ofcarbohydrate and protein in a meal (by the effects ofthe meal on brain tryptophan levels) and to decideabout subsequent food choices.

Acknowledgment: This study was supported by grants fromthe National Aeronautics and Space Administration. theUnited States Air Force, the Center for Brain Sciences andMetabolism Charitable Trust, an NIMH training grant T32M H 1976 1-085 1, and the Surdna Predoctoral FellowshipAward to J.D.S.A preliminary report of this work was pre-sented at the a nnua l meeting of the Society for Neurosciencein 1987.

REFERENCESAghajanian G. K. (1972) Influence of drugs on the firing of serotonin-containing neurons in brain. Fed. Proc. 31, 91-96.Aghajanian G. K., Foote W. E., and Sheard M. H. (1968) Lysergicacid diethylamide: sensitive neuronal units in the midbrain raphe.Science 161, 706-708.Anderson G. M., Teff K . L., and Young S. N . ( 1 987) Serotonin incisternal cerebrospinal fluid of the rat: measurement and use asan index of functionally active serotonin. Li fe Sci. 40, 2253-2260.Auerbach S. and Lipton P. (1985) Regulation of serotonin releasefrom the in vitro rat hippocampus: effects of alterations in levelsof depolarization and in rates of serotonin metabolism. J . NPII -

rochem. 44, 1 I 16-1 130.Baumann P. A. and Waldmeier P. C. (1984) Negative feedback controlof serotonin release in vivo: comparison of 5-hydroxyindoleaceticacid levels measured by voltammetry in conscious rats and bybiochemical techniques. Neuroscience 1 1 , 195-204.Cemto F. and Raiteri M. (1979) Serotonin release is modulated bypresynaptic autoreceptors. Eur. J . Phurmucol. 57, 427-430.Chamberlain B. , Ervin F. R. , Pihl R. O., and Young S. N. (1987)The effectof raisingor lowering tryptophan levels on aggressionin vervet monkeys. Phurmacol. Biochem. Behav. 28, 503-5 10.Chaouloff F., Elghozi J. L., Guezennec Y . , and Laude D. (1985)Effects of conditioned running on plasma, liver and brain tryp-tophan and on brain 5-hydroxytryptamine metabolism of therat. Br . J. Phurmucol. 86, 33-41.Chaouloff F., Laude D., Guezennec Y., and Elghozi J . L. (1986)Motor activity increases tryptophan, 5-hydroxyindoleacetic acid.and homovanillic acid in ventricular cerebrospinal fluid of theconscious rat. J . Neurochem. 46, 13 13- I3 16.Colmenares J. L., Wurtman R. J., and Fernstrom J . D. (1975) Effectsof ingestionofa carbohydrate-fat meal on the levels and synthesisof 5-hydroxyindoles in various regions of rat central nervoussystem. J. Neurochem. 25, 825-829.Culman J. , Kvetnanskf R., Zeman P., and Kiss A. ( 1 984) Hypotha-lamic serotonin: its synthesis and degradation under acute andrepeated stress, in Stress: The Role ofCutecholumines and othe rNeurofransmitlers,Val. 1 (Usdin E., Kvetiianskf R., and AxelrodJ., eds), pp. 109-123. Gordon and Breach Science Publishers,New York.Curzon G. (1986) Serotonin neurochemistry revisited: a new look atsome old axioms. Neurochem. In[. 8, 155- 159.

J . Neurochem., Vol.53.No 6 . 1989


8/9

1932 J . D. SCH'4ECHTER A ND R. J . WURTMANDe Simo ni M. G. , Sokola A, , Fodrit to F.. Dal Toso G.. and AIgeri

S . (1987) Functional meaning of tryptophan-induced increaseof 5-H T metabolism a s clarified by in vivo voltamm etry. BrainRes. 411, 89-94.Dun n A. J . (1988) Changes in plasma and brain tryptophan andbrain serotonin a nd 5-hydroxyindoleacetic acid after footshockstress. Life Sci. 42, 1847-1853.Eccleston D., Ashcroft G. W., an d Crawford T. B. 8. (1965) 5-Hy-droxyindole metabolism in rat brain: a study of intermediatemetabolism using the technique of tryptophan loading-11. J .Neurochem. 12,493-503.Elks M. A,, Youngblood W. W., and Kizer J. S . (1979a) Synthesisand release of serotonin by brain slices: effects of ionic m anip-ulations and cationic ionophores. Brain Rex 172, 46 1-469.Elks M. L., Youngblood W. W., and h z e r J . S. (19796) Serotoninsynthesis an d release in brain slices: indepen dence o f tryptoph an.Brain Rex 172, 471-486.Fernstrom J. D. and Wurtm an R. J . (197 la ) Brain serotonin content:physiological dependence o n plasma tryptop han levels. Sc.ienccJ

Fernstrom J. D. and Wur tman R. J . (1971b) Brain serotonin c ontent:increase following ingestion of carbohydrate diet. Science 174,Friedman P. A, , Kappelm an A. H., and Kaufman S. (1972) Part ialpurification an d characterization of trypto phan hydroxylase from

rabbit hindbrain. J . Biol . Ch on . 247, 4165-41 73.Gallager D. W . and Aghajanian G. K. (1976) Inhibition of firing ofraphe neurons by tryptophan and 5-hydroxytryptophan: block-ade by inhibiting serotonin synthesis with Ro-4-4602. Feuro-pharmacology 15, 149-156.Gothert M. (1980) Serotonin-receptor-mediated modulation of Ca"-depende nt 5-hydroxytryptamine release from neurons of the ratbrain cortex. Naun.vn-Schr?iiedeher'~~Arch Pharmacol. 514,223-230.Grahame-Smith D. G. (1971) Studies in vivo on the relationshipbetween brain tryptophan, b rain 5-HT synthesis and hyperac-tivity in rats treated with a m ono am ine oxidase inhibitor an d I-tryptophan. J. Neurochern. IS , 1053-1066.Greenwood M. H. , Lader M. H. , Kantamaneni B. D.. and CurzonG. (1975) Th e acute effects of oral (-)-tryptophan in hum ansubjects. Br . J. Clin. Pharmacol. 2, 165-172.Guan X.-M. and McBride W. J . (198 7) Effects of K '-stimulationand precursor loading on the in vivo release of dopamine. se-rotonin and their metabolites in the nucleus accumbens of therat. LifeSci. 40, 2579-2586.Hamon M., Bourgoin S . , Artaud F. . and Glowinski J . (19 79) Therole of intraneuronal 5 -HT an d of tryptophan hydroxylase ac-tivation in the co ntrol of 5 -HT synthesis in rat brain slices in-cubated in K'enriched medium . J . Xeurochem. 33, I03 1-1042.Hartma n E. and Spinweber C. L. (197 9) Sleep induced by l-trypto-phan: effect of dosages within normal dietary intake. J . IVerr.Ment. Dis. 167, 497-499.Harvey J. A, , Schlosberg A. J., and Y unger L. M. (1975) Behavioralcorrelates of serotonin depletion. Fed. Proc 34, 1796- 1801.HeraiefE., Burckhardt P., Wurtm an J. J . , and W urtman R. J . (1985)Tryptoph an administrat ion may en hance weight loss by some

173, 149-152.

1023-1025.

Jacobs B. L., Trimba ch C . , Eubank s E. E. , and Trulson M . (1975)Hippocampal mediation of raphe lesion- and PCPA-inducedhyperactivity in the rat. Brain Ke.r 94, 253-261.Joseph M. H. and Kennett G. A. ( I 98 I ) In vivo voltammetry in therat hippoca mpus as an index of drug effects on extraneuronal5-HT. N~, i tropharmacolo~y0, I36 I - 1364.Joseph M. H . and Kennet t G. A. (198 3) Stress-induced release of 5-H T in the hip pocam pus and i ts dependence o n increased tryp-toph an availability: an in vivo electrochemical study. Brain R ex

Kantak K. M.. Hegstrand L. R. , Whitman J ., and Eichelman B.(1979) Effects ofdietary supplem ents and a tryptophan-free dieton aggressive behavior in rats. Pharmacol. Biochem. Behuv. 12 ,Kantak K. M., Hegstrand L. R ., and Eichelman B. (1981) Dietarytryptophan reversal of septa1 lesion a nd 5,7-D HT lesion elicitedshock-induced fighting. Pharmacol. Biochem. Behav. 15, 343-350.Kennet t G . A. and Joseph M. H. ( 1 98 ) The functional importanceof increased brain tryptophan in the serotonergic response torestraint stress. ~ e i r ro~ )harm uco logy0, 39-43.Knott P. J . . Joseph M . H., and Curzon G . (1973) Effects of fooddeprivation an d immobilization o n tryptophan an d other am inoacids in rat brain. J . Neurochtvn. 20, 249-25 I .Kozell L.. Sandyk R.. Wagner G. C., and Fisher H. (1987) Th e effects

of /- tryptophan o n haloperidol-induced mov ement disorder inthe rat. Life Sci. 41, 1739-1744.Lapin I . P. and Oxenkrug G. F. ( 1 969) Intensification of the centralserotoninergic processes as a possible determinant of the thy-moleptic effect. Lancer i, 132- 136.Leiter L. A,, Hrboticky N. . and Anderson G. H. (1987) Effects of I-tryptophan o n food intake and selection in lcan men and wom en,in H u m a n Ubesit j , ( W u r t m a n R. J. and Wurtman J. J . , eds),Ann. .V. Y . Asad. Sci.. 449, 327-328.Lowry0.H., Rosebrough N. J., Fan: A. L., and Randall R. J. (1951)Protein measurement with the Fohn phenol reagent. J . Biol.C h e m . 193, 265-275.Lytle L. D., Messing R. B.. Fisher L., and Phebus L. (1975) Effectsof long-term corn consu mption on brain serotonin and the re-sponse to electrical shock. Science 190, 692-694.Marsden C. A. and Curzon G . (1976) Studies on the behavioral effectsof tryptophan and p-chlorophenylalanine. Neuropharmacology

Marsden C. A., Conti J. , S t ro p e E. , C u r z o n G . , a n d A d a m R . N.( 1979) Monitoring 5-hydroxytryp tamine release in the brain ofthe freely moving unanaesthetizcd rat using in vivo voltammetry.Brain Res. 171, 85-99.McCinty D. J. and Harper R. M . (1976) Dorsal raphe neurons:depression of firing during sleep in cats. Brain Rex 101, 569-575.Milner J . D. and Wurtman R. J. (1984) Release of endogenous d o-pam ine from electrically stimulated slices of rat striatum. BrainRes. 301, 139-142.

Moir A. T. B. and Eccleston D. (1968) Th e effects of precursor loadingin the cerebral metabolism of 5-hydroxyindoles. J . Neurochem.

27 0,2 5 1-257.

173-1 79.

15, 164-171.

15. 1093-1 108.moderately obese patients on a protein-spanng modified fast(PSM F) diet. Int. J . Ea f. Disorders 4, 28 1-292.Herr B. E., Gallager D. W., and Roth R. H. (1975) Tryptop hanhydroxylase: activation in vivo following stimulation of centralserotonergic neurons. Biochetn. Fharrnucd 24 ,20 19-2023.Hery F., Chouvet G. ,K an I. P.,ujol J . F., and Glowinski J . (1977)Dailyvariationsof various ofsero ton in metabolismin the rat brain. 11. Circadian variations in serum and cerebraltryptophan levels: lack of correlation with 5-H T turnover. BrainRes. 123, 137-145.Hutson P. H., Sarna G. S. , Kantamaneni B. D.. and Curzon G. ( I 985)Monitoring the effect of tryptophan load on brain indole me-tabolism in freely moving rats by simultaneous cerebrospinalfluid sam plingand braindialysis . J . Neurochem. 44, 1266-1273.

Morgan W . w., sa1dana J , J ., Y n d o c ,A, , and M~~~~~ . F, (1975)Correlations between circadian changes in seru m am ino acidsor brain tryptophan and the contents of serotonin and 5-hy-droxyindoleacetic acid in regions of the rat brain. Brain Res.84, 75-86.

Pardridge W . M. (1977 ) Regulation of amin o acid availabil i ty to thebrain. in Nutrition and the Brain, Vol. I (Wur tman R. J. andW u r t m a n J. J., eds). pp. 141-203. Rav en Press, New York.Parfitt A . and Grahame-Smith D. G . (1973) Th e transfer o f trypto-phan across the synaptosomal m embran e, in Aromatic Amino.-kids in /h e Brain (Wolstenholme G. E. W. and FitzsimonsD. w., eds), pp. 175-196. Elsevier, Ams terda m.Reinhard J. F. a n d W u r t m a n R. J . (1977) Relation between brain

J . Neurochem.. Vol.53.No. 6 , 1989


9/9

TRYPTOPHAN AVAILABILITY MODULATES 5-HT RELEASE I9335-HIAA levels and the release of serotonin into brain synapses.Life Sci. 21, 1741-1746.SarnaG.S.,Hutson P. R. , Tricklebank M. D., and Curzon G. (1983)Determination of brain 5-hydroxytryptamine turnover in freelymoving rats using repeated sampling of cerebrospinal fluid. J .Neurochern.40,383-388.SchlickerE., Brandt F., ClassenK., and Gothert M. (1985) Serotoninrelease in human cerebral cortex and its modulation via serotoninreceptors. Brain Rex 331, 337-341.ShieldsP. J. and Eccleston D. (1972) Effects of electrical stimulationof rat midbrain on 5-hydroxytryptamine synthesisas determinedby a sensitive radioisotope method. J. Neurochern.19,265-272.Smith B. and Prockop D. J. ( 1 962) Central-nervous-system effectsof ingestion of /-tryptophan by normal subjects. N. Engl. J. Med.

Stewart R. M., Growdon J. H., Cancian D., and Baldessarini R. J.(1976) 5-Hydroxytryptophan-induced myoclonus: increasedsensitivity to serotonin after intracranial 5,7-dihydroxytrypta-mine in the adult rat. Neuropharmacology 15,449-455.Taylor M. (1976) Effects of 1-tryptophan and I-methionine on activityin the rat. Br . J. Pharrnacol. 58, 117-1 19.

267, 1338-1341.

Ternaux J. P., Boireau A ,, Bourgoin S ., Hamon M., Hery F., andGlowinskiJ. (1976) In vivo release of 5-HT in the lateral ventricleof the rat: effects of 5-hydroxytryptophan and tryptophan. BrainR ex 101, 533-548.Ternaux J. P., Hery F., Hamon M., Bourgoin S . , and Glowinski J .(1977) 5-HT release from ependymal surface of the caudate nu -cleus in enciphale isolc cats. Brain Rex 132, 575-579.Weber L. J. and Horita A. ( 1 965) A study of 5-hydroxytryptamine

formation from tryptophan in the brain and other tissues.Biochern. Pharrnacol. 14, 114 - 149.Wong D. T., Horng J. S., Bymaster F. P., Hauser K. L., and MolloyB. B. (1974) A selective inhibitor of serotonin uptake: Lilly110141, 3-(ptrifluoromethylphenoxy)-n-methyl-3-phenylpro-pylamine. L$e Sci. 15,47 1-479.Wurtman J. J., Moses P. L., and Wurtman R. J. (1983) Prior car-bohydrate affects the amount of carbohydrate that rats chooseto eat. J. Nutr. 113, 70-78.Wurtman R. J. and Fernstrom J. D. (1976) Control of brain neu-rotransmitter synthesis by precursor availability and nutritionalstate. Biochern. Pharrnacol. 25, 1691-1696.

J. Neurochern., Vol.53,No . 6, 1989

Judith D. Schaechter and Richard J. Wurtman- Tryptophan Availability Modulates Serotonin Release from Rat Hypothalamic Slices

Documents