- Email: [email protected]
Comparison of the effect of prolactin on dopamine release from the rat dorsal and ventral striatum and from the mediobasal hypothalamus superfused in vitro
European Journal of Pharmacology, 149 (1988~ 1-8
1
Elsevier ILIP 50235
Comparison of the effect of prolactin on dopamine release from the rat dorsal and ventral striatum and from the mediobasal hypothalamus superfused in vitro J i n - C h u n g C h e n and Victor D. R a m i r e z * Department of Physiology" and Biophysics, Unieersi(v of Illinois, Urbana, IL 61801, U.S.A. Received 2 February1988, accepted 9 February 1988
In this paper, by means of a superfusion technique, we have examined the effect of ovine prolactin on dopaminergic neurons innervating the dorsal striatum, ventral striatum and the mediobasal hypothalamus of male rats. Fragments from dorsal striatum were superfused with ovine prolactin dissolved in normal medium (Krebs-Ringer phosphate) or medium containing tetrodotoxin (TTX, 10 6 M). Ovine prolactin stimulated the in vitro release of dopamine from dorsal and ventral striatal fragments. In dorsal striatal fragments a linear dose-dependent dopamine release was observed only when fragments were superfused with Krebs-Ringer phosphate-TTX medium. In addition, [LeuS]enkephalin (10 -6 and 10 -5 M) decreased the prolactin-induced in vitro dopamine release from dorsal striatal fragments superfused with Krebs-Ringer phosphate-TTX medium. Ovine prolactin (10 9 - 1 0 5 M) did not elicit changes in dopamine, 3,4-dihydroxyphenylacetic acid or 5-hydroxyindoleacetic acid outputs from mediobasal hypothalamic fragments superfused with Krebs-Ringer phosphate medium containing TTX. The possible regulatory mechanisms of ovine prolactin on dopaminergic neurons are discussed. Prolactin: Dopamine; Corpus striatum (dorsal/ventral): Tetrodotoxin: [Leu 5]enkephalin; Hypothalamus (mediobasal)
1. I n t r o d u c t i o n
Prolactin exerts multiple actions on central dopaminergic neurons in mammals (see review Drago, 1984). Neurochemical and pharmacological evidence (Moore et al., 1985; Leong et al., 1983) has indicated that prolactin can increase the synthesis/release of dopamine (DA) from tuberoinfundibular DAergic neurons. Among extrahypothalamic actions, prolactin administration induces grooming (Drago et al., 1984), yawning (Laping and Ramirez. 1986) and potentiates
* To whom all correspondence should be addressed: 524 Burrill Hall, Department of Physiology and Biophysics, University of Illinois, 407 South Goodwin Avenue, Urbana, IL 61801, U.S.A.
amphetamine-induced rotation behavior (Joseph et al., 1986). Prolactin also increases the number of DA binding sites (Hruska, 1986) and DA uptake by striatal synaptosomes (Gregerson and Selmanoff, 1985). Moreover, prolactin enhances electrical-induced DA release in in vitro striatal slices (Perkins and Westfall, 1978). These experiments strongly suggest that prolactin participates in the regulatory mechanisms controlling nigrostriatal DAergic activity. The finding of specific prolactin binding sites in the substantia nigra and striatum (Di Carlo et al., 1985) offers a possible explanation by which activation by prolactin of nigrostriatal DAergic neurons can be mediated. It is also known that prolactin-enhanced grooming behavior can be mediated through changes in DA metabolism of the nucleus accumbens (Drago et al., 1984). In addition, lesions of the nucleus ac-
0014-2999/88/$03.50 :~' 1988 Elsevier Science Publishers B.V. (Biomedical Division)
cumbens impaire lactational performance and maternal behavior (Smith and Holland, 1975). Thus, it is conceivable that prolactin also affects DAergic transmission of the mesolimbic pathway. Previously our group published that in vitro prolactin stimulated DA release from striatal tissue from male but not from female rats (Chen and Ramirez, 1982). In vivo, this hormone increased 3,4-dihydroxyphenylacetic acid (DOPAC) output from the caudate nucleus of male rats along with activation of grooming behavior (Chen et al., 1985). The aim of the present study was to compare the effect of prolactin on the activity of three DAergic systems of the central nervous system: (1) the nigrostriatal (dorsal striatum). (2) the mesolimbic (ventral striatum) and (3) the tuberoinfundibular (mediobasal hypothalamus) neurons. In addition, tetrodotoxin (TTX, 10 -6 M) known to block neurons impulse flow (Narahashi, 1972) was used in the in vitro superfusion system to examine a postulated presynaptic effect of ovine prolactin on DA neurons. Also, the interaction of [LeuS]enkephalin with prolactin on the nigrostriatal dopaminergic system was examined.
2. Materials and methods
2.1. Tissue preparation Adult male Spraque-Dawley rats (Holtzman, Madison, WI) weighing 300-500 g were housed in groups of five per cage in a temperature-controlled room (22-24°C) with lights on from 05:00 to 19:00 h. Water and Purina rat chow were available ad libitum. All rats were allowed to accustom to the environment for at least 7 days prior to the experiments. In all experiments, animals were killed by decapitation between 09:00 and 09:30 h.. The dorsal striatum (both sides, mainly nucleus caud a t e / p u t a m e n ) and mediobasal hypothalamus were dissected out as previously reported (Becket and Ramirez, 1980). The ventral striatal fragments that include the nucleus accumbens were taken from an area defined by a coronal cut about 1 m m anterior to the optic chiasma to expose the ventral striatum at the level of the anterior commissura,
and then four cuts were made at right angles: the dorsal cut was immediately above the anterior commissura near the midline approximately 2 mm long; the lateral cuts were made inside the lateral extension of the corpus callosum and the midline. respectively about 1.5 mm long followed by a ventral cut made at right angles to the first two lateral cuts to remove a fragment of about 1.0 mm thick. Similar cuts were performed in the contralateral side. Tissue fragments were immediately placed in the ice-cold Krebs-Ringer phosphate medium, pH 7.4, containing 10 mM sodium molybdate (Sigma) with or without TTX (1()-6M, Sigma). Prior to placing the tissue in an experimental chamber, the dorsal/ventral striatal fragments were divided into eight pieces by three additional perpendicular cuts and the mediobasal hypothalamus fragments were halved at a midsagittal plane. Sixteen striatal pieces (dorsal striatum: 34.7 + 5.2 rag; ventral striatum: 11.4 +_ 2.6 nag, n = 12 perifusion) or eight mediobasal hypothalamus halves (40.6_+3.2 mg, n = 12 perifusion) were placed in each experimental chamber. These wet weights correspond to values obtained at the termination of the superfusions.
2.2. Superft~'ion system The superfusion system was essentially the same as described previously (Gallardo and Ramirez, 1977) except that a 35 p,l/min flow rate was used. The time interval for the fluid to reach the tissuecontaining chamber (400 ktl dead volume) from the beaker was approximately 10 rain. Following a 45 rain stabilization period (30 rain for mediobasal hypothalamus), perfusate samples were collected on ice for ten or fi~urteen 15 n'fin intervals. In the first experiment, dorsal striatal fragments from the control group were superfused with Krebs-Ringer phosphate whereas for the experimental groups, ovine prolactin (Sigma) at concentrations of l0 v 10 6 or 10 -5 M was administered for 30 min starting from the beginning of collection interval 5. A similar experimental procedure but with Krebs-Ringer phosphate medium containing 10 ~ M TTX throughout the superfusion period was performed using the same doses of ovine prolactin (10-'7_10 5 M, 30 min).
In the second experiment, dorsal striatal fragments were superfused with Krebs-Ringer phosphate containing 10 - 6 M TTX plus either 10 -7, 10- 6 or 10- 5 M [Leu 5]enkephalin acetate (Sigma). T h e n , 10 - 6 M ovine prolactin was administered during intervals 6 and 7 for 30 rain. In experiment 3 and 4, ventral striatal and mediobasal hypothalamus fragments respectively, were superfused with Krebs-Ringer phosphate containing TTX. Ovine prolactin at concentrations of 10-7, 10--6 or 10 .-5 M was then administered for 30 rain during intervals 5 and 6. Twenty microlitres of untreated superfusate samples were then subjected to high performance liquid chromatography with electrochemical detection (HPLC-EC) for catecholamine and indoleamine analyses.
2.3. HPLC-EC system An HPLC-EC system (Bioanalytical System, Inc., West Lafayette, IN) with a C-18 reverse phase column (4.6 × 250 mm, 5 /,tm, Biophase, ODS) coupled to a carbon paste (CP-O) electrode at a potential of +0.65 V relative to the Ag/AgCI reference electrode was used. Mobile phase consisted of 5.75 g citric acid, 4.1 g sodium acetate, 85 mg sodium octyl sulfate, 7% methanol in 1 i of deionized water, adjusted to pH 4.5 with concentrated sodium hydroxide solution. The buffer was then degassed and filtered through a millipore membrane (0.45 #m, Millipore, Bedford, MA). External standards for DA, DOPAC, 5-hydroxyindoleacetic acid (5-HIAA) and homovanillic acid (HVA) were prepared fresh with superfusion medium at proper doses. Details of HPLC operation were described previously (Chen et al., 1984).
groups, at interval 6). ANOVA followed by posthoc comparisons was used to calculate significant differences among controls and groups treated with different doses of ovine prolactin (see fig. 3). A P < 0.05 value was considered significant.
3. Results
3. I. Effect of in vitro ovine prolactin administration on DA release from dorsal striatal fragments superfused with Krebs-Ringer phosphate with or without TTX The basal release ( p g / m g per min) of DA and its metabolite DOPAC from dorsal striatal fragments is illustrated in fig. 1. In general, the rate of in vitro DA and DOPAC release decreased significantly during the first 1.5 h of superfusion to reached a stationary phase thereafter (from about interval 5 to the end of the superfusion). The output of 5-HIAA and HVA became barely detectable after 90 min of superfusion (data not shown). TTX (10 - 6 M ) significantly reduced basal DA release during the stationary phase from a mean value of 7.0 + 0.9 in preparations superfused with Krebs-Ringer phosphate to 2.8 _+ 0.2 in preparations superfused with Krebs-Ringer phosphateTTX medium (last six intervals). The output of
,j -I
z
-
Je
=
OO..-
2.4. Statistical analysis All data in this report are expressed as means _+ S.E.M. A t-test was employed to compare mean values between two groups. To examine the effect of ovine prolactin in these systems, a two-tailed paired Student's t-test was employed to compare the mean release of DA, DOPAC or 5-HIAA corresponding to each interval after ovine prolactin administration with the mean basal level at interval 5 (for [LeuS]enkephalin experimental
=
I:DI~ :~ IOl
s',',~r,t
:_ N.M~
..r~
Fig. 1. In vitro output of DA and DOPAC from dorsal striatal fragments superfused with Krebs-Ringer phosphate medium or Krebs-Ringer phosphate medium containing TTX (10 -6 M). The basal release of DA and DOPAC reached a stationary, phase after approximately 1.5 b control period. Notice thai the DA output in Krebs-Ringer phosphate-TTX medium was in general lower than in normal Krebs-Ringer phosphate medium. See text for details.
--TTX{
~-POt-
"
oPRL IC~7M
F",,
~g o
i
,or k
C~/
:
....
J
J - -
O'6M) - - .
4
oPRL 10"TM "
U.
.
.
.
I
"
"
i .tim
(z
,~:z 0 C3 + % u 20
.
.
.
.
1
J
_ _
[
1
__
oPRL IO~M
1
l
oPRL 0 5M
SAMPLE N U M B E R
Fig. 2. Effect of ovine prolactin (oPRL) on in vitro DA output from dorsal striatal fragments superfused with Kxebs-Ringer phosphate medium or Krebs-Ringer phosphate medium containing TTX (10- ~ M). A dose-dependent stimulatory effect of ovine prolactin (10-%10 -~ M) on DA release was observed in the fragments superfused with Krebs-Ringer phosphate medium containing TTX but not from those superfused with normal Krebs-Ringer phosphate medium (* P < 0.05). See also fig. 3.
The effect of ovine prolactin on DA activity from dorsal striatal fragments is shown in fig. 2. Ovine prolactin tested at three doses: 10 -v, 10 (' or 10 --s M significantly enhanced basal DA release following prolactin administration in the TTX free Krebs-Ringer phosphate as well in the Krebs-Ringer phosphate-TTX preparations when compared to interval 5. However, in the former the response was not linearly related to the doses whereas in the latter a clear dose-response effect was obtained as shown in fig. 3 in which the peak D A responses from the data presented in fig. 2 as function of doses of ovine prolactin as well the values for control preparations are presented.
3.2. Effect of [LeuS /enkephalin on or,me prolactinenhanced DA release from dorsal striatal fragments Fig. 4 portrays the effect of 10 (' M ovine prolactin on in vitro D A and D O P A C release
)euS-EnKidTM
15-I'
DOPAC, however, decreased slightly from a mean value of 6.8 _+ 0.5 to 5.7 _+ 0.2 (last six intervals), respectively.
~
0PRL
l
I
10+6M
k\\\\~\\~'l
n : 4
0 c~
0
i
L
i
leuS- EnKIO'gM
~-.
,..4
,,
:o+ , +o,,_++ o° ° + +
..... + ...... , +
.
o~
+'
L
I u:
o
~.
CoP',c
.
.
.
.
£
'C
I
t-+\\-.\\\\+I
,~ 15
....
I
1_
LeuS-EnK I O ' S M
4
P-.\\\\\\~
"
o 6
Fig. 3. In vitro DA output from dorsal striatal fragments as function of three doses of ovine prolactin (10 v, 10-6, and 10-5 M) in preparations superfused with Krebs-Ringer phosphate medium or Krebs-Ringer phosphate medium containing "I3"X (10 6 M). Control preparations did not receive ovine prolactin. Peak DA outputs for the experimental preparations and basal values corresponding to interval 5 of controls and ovine prolactin-treated preparations are plotted. Notice a clear linear dose-response for the striatal fragments superfused with Krebs-Ringer phosphate medium containing TTX and the absence of a sigmficant response in the preparations superfused with Krebs-Ringer phosphate medium only.
~0
0i SAMPLE
NUMBER
Fig. 4. Effect o f 10 -~' M o'.ine p r o l a c t i n o n in vitro DA and D O P A C outputs from dorsal striatal fragments superfused with Krebs-Ringer phosphate-'ITX (10 6 M) a n d [LeuS]en k e p h a l i n at doses 10 v, 10 ~ o r 10 ~ M. [Leu'S]Enkephalin decreased the ovine prolactin-induced I)A release in a dose-dependent manner.
from dorsal striatal fragments superfused with Krebs-Ringer phosphate containing TTX (10 ..6 M) and [LeuS]enkephalin at doses 10 -7, 10 -6 or 10 -5 M. As shown in the figure, [LeuS]enkephalin tends to decrease basal as well as the ovine prolactin-induced DA release in a dose-dependent manner. Low doses of the ]LeuS]enkephalin (10 -7 M) did not significantly reduce the DA response of the terminals to 10 -6 M ovine prolactin (186% compared with 228% from control experiments without [ Leu 5]enkephalin when peak DA response were compared to interval 6) whereas middle and high doses (10 .-6 and 10 -5 M) partially or completely blocked the ovine prolactin-induced DA release (118 and 102%, respectively). In none of these groups were changes in D O P A C output observed.
e: I.Z
Unlike the basal release pattern of DA and D O P A C from the dorsal and ventral striatal frag-
n = ,4
I0 5 c E
0 o P R L 10"6M
t5
~
g
~o
c~
s
~:
2o
o P R L I0-~' M
i5 _~
3.4. Effect of in vitro ovine prolactin infusion on DA, DOPA C and 5-HIAA release from mediobasal hypothalamus fragments
o P R L 10-TM
15
3.3. Effect of ovine prolactin on DA and DOPAC release from ventral striatal fragments. The effect of ovine prolactin on mesolimbic DAergic terminals with Krebs-Ringer phosphate medium containing TTX (10 -6 M) is illustrated in fig. 5. Low doses of ovine prolactin (10 -7 M) did not affect either DA nor D O P A C release during the 2 h superfusion period following prolactin infusion. However, 10 -6 M ovine prolactin stimulated DA release reaching its peak response 30 min post-ovine prolactin infusion (7.4 + 2.5 in interval 5 vs. 11.2 + 2.5 p g / m g per min in interval 8) without affecting D O P A C output. A high dose of ovine prolactin (10 -5 M) induced a greater response since the stimulated DA release was prolonged (up to 45 min) and reached maximal levels of 16.8 + 3.4 at interval 7 compared to 9.0 + 1.3 p g / m g per min at interval 5. Moreover, a slight increase in D O P A C output was observed at interval 8. The percent changes of the peak DA response induced by ovine prolactin were 151% at 10 - 6 M and 187% at 10 -5 M.
TTX (iO'6M)
20
to 5 !
I
I
J
t
5
IO
15
SAMPLE NUMBER
Fig. 5. Effect of ovine prolactin on in vitro DA and D O P A C outputs from ventral striatal fragments superfused with Krebs-Ringer p h o s p h a t e - l T X (10 6 M) medium ( * P < 0.05). Notice that ovine prolactin was clearly effective at l 0 6 and l0 =5 M.
ments, the DA and D O P A C levels in superfusates from mediobasal hypothalamus fragments remained low and stable from the very beginning to the end of the superfusion, probably a reflection of the integrity of the tubuloinfundibular DA neurons in this tissue preparation (fig. 6). Detectable levels of 5-HIAA were also found throughout the experiment, however, with a significant decrease in levels when initial and final values were compared (3.8 + 0.5 and 1.8 + 0.3 p g / m i n , n = 12, respectively). However, neither of the four doses of ovine prolactin tested had any effect on the release of these three neurochemicals. As shown in fig. 6, ovine prolactin infused either at very low doses of 10 -9 M (n = 4) or at high doses of 10 -5 ( n = 2 ) and 10-6 M ( n = 2 ) did not affect the basal release of DA nor of D O P A C since in all the
TTX (IO'6M) oPRL 10-gM i-.\\\\\N\',I
n=4
o~
[
o P R L I0-;' M .........
#5
IIl
I °
~\\\\\\\x 1
I
5
SAMPLE
I0
I 15
NUMBER
Fig. 6. Effect of ovine prolactin on in vitro DA, I)OPAC and 5-HIAA outputs from mediobasal hypothalamus fragments superfused with Kxebs-Ringerphosphate-TTX medium. S.E.M. bars are depicted only for the DA profiles. Notice that I)A and DOPAC levels were low and stable during the superfusion. The 5-HIAA output showed a 2-fold decrease when the beginning and the end values were compared. Neither of the four doses of ovine prolactin had any effect on the release of these three neurochemicals.
experiments stable levels of DA and D O P A C (2.8 _+ 0.2 and 3.1 + 0.5 p g / m g per min, respectively) were detected during the pre- or post-ovine prolactin superfusion periods.
4. Discussion Considerable evidence has accumulated supporting the concept that prolactin affects extra-hypothalamic DAergic neurons (Drago, 1984; Chen and Ramirez, 1985). In the present study in preparations superfused with TTX ovine prolactin was highly effective in increasing the in vitro DA release from either the dorsal or the ventral striaturn, target sites of the nigrostriatal and meso-
limbic DAergic systems respectively, adding further support to this contention. Of the three doses of ovine prolactin tested (10 -7 , 10 ~ and 10 -~ M) in dorsal striatai tissue superfused with normal Krebs-Ringer phosphate medium, only the 10 -7 and 10 5 M ovine prolactin significantly enhanced DA release above control superfusions (fig. 3) but not linearly. Interestingly, identical doses of ovine prolactin elicited DA release in a linear dose-dependent manner when dorsal striatal fragments were perfused with Krebs-Ringer phosphate medium containing 10 ~ M TTX suggesting the existence of regulatory interneuronal inputs controlling DA release within the fragments which can then be blocked by the TTX treatment. Tetrodotoxin, a sodium channel blocker, that exerts a highly specific action on nerve excitation without any effect on other membrane parameters, has been used as a tool to study the mechanism of transmitter release from nerve terminals without being interfcrcd with by impulse flow in both presynaptic and postsynaptic elements (Narahashi, 1972; Katz and Miledi, 1967). Consequently, the present results clearly indicatc that ovine prolactin acts on the nerve terminals of the dorsal striatal to elicit DA release. TTX (10 ~ M) also reduced the basal relcasc of DA (60%) and D O P A C (20%) from dorsal striatal fragments as compared to control tissues. The reduction by TTX of in vivo basal DA release estimated with push-pull perfusion was previously reported by Voigt et al. (1985) and discussed, however, as duc to a direct action upon some DAergic axons which are spontaneously active. Our in vitro results in preparations in which DA axons were severed from their cell bodies but still contained viable interneurons connections suggest that the action of TTX is due to inhibition of action potentials from these interncurons, resulting in a net decrease of DA release. In support of a presynaptic control of DA release by interneurons is the finding that the decrement of basal DA release under TTX left unimpaired the ovine prolactin-evoked DA response which was even higher in preparations superfused with TTX than in control (e.g. 228, 631% in Krebs-Ringer phosphate-TTX vs. 125,264% in Krebs-Ringer phosphate at 10 -~' and 10 5 M ovine prolactin, respectively).
Furthermore, our studies using [LeuS]enkepha lin, an opioid peptide that can inhibit DA release (Loh et al., 1976) significantly blocked the ovine prolactin-induced DA release from dorsal striatum fragments, confirming the existence of a complex presynaptic regulatory mechanism controlling DA release from the nigrostriatal dopaminergic terminals (Dray, 1979). Considering that prolactin- and ACTH-induced grooming behavior (Drago et at., 1984; Gispen et al., 1975) result from an activation of nigrostriatal and mesolimbic DA transmission, respectively it is significant to note that ACTH-induced grooming in the rat can be suppressed by treatment with opiates (Cools et al., 1978). In the ventral striatal tissue preparation, ovine prolactin also stimulated DA release. There is evidence indicating that prolactin increases DA turnover rate in the nucleus accumbens of aMTtreated rats (Fuxe et al., 1977; 1978). In addition, we have found that local administration of ovine prolactin within the nucleus accumbens increased in vivo DOPAC output further supporting these data (paper submitted for publication). It is known that DAergic terminals in the nucleus accumbens originate from the ventral tegmental area and substantia nigra (Lindvall and Bj~3rklund, 1983). Moreover, the necessary co-existence of nucleus accumbens and caudate nucleus DA neurons for the performance of normal rotation behavior (Pycock and Marsden, 1978; Kelly and Moore, 1977) indicate a functional interaction between these two DA structures. The observation of a similar stimulatory effect of ovine prolactin in these distinct but interrelated DAergic pathways suggest that not only an anatomical contiguity exist between these two systems but also a similar functional response to ovine prolactin. in view of the postulated existence of a short inhibitory feedback loop between tubuloinfundibular DA neurons and prolactin (Gudelsky and Porter, 1980; Moore, 1987), it may appear surprising that neither of the doses of ovine prolactin tested in the present investigation (10 9, 1 0 - 7 , 10 - 6 n o r 1 0 .s M) affected basal DA or DOPAC release from mediobasal hypothalamus fragments. The failure to detect any significant changes in the release of these neurochemicals upon ovine pro-
lactin administration might be explained by the fact that only a 2 h period post-ovine prolactin infusion was studied. Similar observations have been reported by Fuxe et al. (1977) and Perkins and Westfall (1978) since prolactin did not enhance DA turnover in aMT-treated animals elicit basal DA release from the mediobasal hypothalamus in short-term experiments. Although we cannot exclude the possibility that ovine prolactin may activate tubuloinfundibular DA neurons through interneurons blocked by the TTX treatment, we favor the hypothesis of an effect of ovine prolactin on the synthesis of DA through activation of tyrosine hydroxylase. Demarest et al. (1984) clearly showed that there are two components in the prolactin-induced activation of tubuloinfundibular DA tyrosine hydroxylase activity. The relatively rapid 'tonic' component that is activated approximately 4 h post-systemic administration of prolactin and the delayed 'induction' component that occurs 12 h post-prolactin administration. Recently, our own data indicate that there is approximately a 30% increase in tyrosine hydroxylase activity in the mediobasal hypothalamus but not in the dorsal or the ventral striatal 4 h after ovine prolactin administration (submitted for publication). Hence, there appears to be significant temporally related differences as well differences in the mechanism of action of ovine prolactin on DA metabolism between the nigrostriatal-mesolimbic DAergic system and the tubuloinfundibular DA system. In the former ovine prolactin activates preferentially and directly the release mechanism of DA which can occur within 1 h whereas in the latter the 'tonic component' of the synthetic mechanism of DA appears to be initially activated. In summary, the results from this study provide the first indication of a presynaptic effect of ovine prolactin on the release of DA from nigrostriatal and mesolimbic DAergic terminals. The role of ovine prolactin on these systems and clarification of its physiological significance remains to be elucidated.
Acknowledgement This work was in part support by a NIH Grant No. HD-14625 to V.D.R.
References Becket, J. and V.D. Ramirez, 1980, Dynamics of endogenous catecholamine release from brain fragments of male and female rats, Neuroendocrinology 31, 18. Chen, J.C., A.D. Ramirez and V.D. Ramirez, 1985, Prolactin effect on in vivo striatal dopaminergic activity as estimated with push-pull perfusion in male rat, in: Prolactin, Basic and Clinical Correlates, eds. R.M. MacLeod, M.O. Thorner and U. Scapagnini (Liviana Press, Padova) p. 523. Chen, J.C., K.K. Rhee, D.M. Beaudry and V.D. RamireT, 1984, In vivo output of dopamine and metabolites from the rat caudate nucleus as estimated with push-pull perfusion online with HPLC-EC in unrestrained, conscious rats, Neuroendocrinology 38, 362. Chen, Y.F. and V.D. Ramirez, 1982, Prolactin stimulates dopamine release from male but not from female rat striatal tissue superfused in vitro, Endocrinology 111, 1740. Cools, A.R., V.M. Wiegant and W.H. Gispen, 1978, Distinct dopaminergic systems in A C T H - i n d u c e d grooming, European J. Pharmacol. 50, 265. Demarest, K.T.. G.D. Riegle and K.E. Moore, 1984, Prolactininduced activation of tuberoinfundibular dopaminergic neurons: evidence for both a rapid 'tonic' and a delayed "induction" component, Neuroendocrinology 38, 467. Di Carlo, R., G. Muccioli, D. Lando and G. Bellussi, 1985, Further evidence for the presence of specific binding sites for prolactin in the rabbit brain. Preferential distribution in the hypothalamus and substantia nigra, Life Sci. 36, 375. l)rago, It:., 1984, Prolactin and sexual behavior: a review, Neurosci. Biobehav. Rev. 8. 433. Drago, F., G.L. Kov~.cs, G. Continella and U. Scapagnini, 1984, Dopamine in the nucleus accumbens: its role in prolactin-enhanced grooming behavior of the rat, Biogenic Amines 1, 75. Dray, A., 1979, The striatum and substantia nigra: a commentary on their relationships, Neuroscience 4, 1407. Fuxe, K., K. Andersson, T. HiSkfeh and L.F. Agnati, 1978. Prolactin-monoamine interactions in rat brain and their importance in regulation of LH and prolactin secretion, in: Progress in Prolactin Physiology and Pathology, eds. C. Robyn and M. Harter (Elsevier/North-Holland Biomedical Press) p. 95. Fuxe, K., P. Eneroth, J.-A. Gustafsson, A. L~SfstriSm and P. Skett, 1977, Dopamine in the nucleus accumbens: preferential increase of DA turnover by rat prolactin, Brain Res. 122, 177. Gallardo, E. and V.D. Ramirez, 1977, A method for the superfusion of rat hypoth-,dami: secretion of luteinizing hormone-releasing hormone (LH-RI-1), Proc. Soc. Exp. Biol. Med. 155, 79. Gispen, W.H., V.M. Wiegant, H.M. Greven and D. De Wied, 1975, The induction of excessive grooming in the rat by intraventricular application of peptides derived from AC'I'tI: structure-activity studies, Life Sci. 17, 645.
Gregerson, K.A. and M. Selmanoff, 1985, F.ffects of hyperprolactinemia on dopamine uptake into striatal synaptosomes in vitro, in: Prolactin, Basic and Clinical Correlates, eds. R.M. MacLeod, M.O. Thorner and U. Scapagnini (Liviana Press, Padova) p. 625. Gudelsky, G.A. and J.C. Porter, 1980, Release of dopaminc from tuberoinfundibular neurons in pituitary' stalk bhn~d after prolactin or haloperidol administration, l-ndocrinol. 106, 526. Hruska, R.I-.. 1986, Modulatory role for prolactin in the elevation of striatal dopamine receptor density' induced by chronic treatment with dopamine receptor antagonists. Brain Res. Bull. 16, 331. Joseph, J.A., G.S. Roth and A.S. IAppa, 1986, Reduction of motor behavioral deficits in senescent animals via chronic prolactin administration I. Rotational behavior. Ncurobiol. Aging 7, 31. Katz. B. and R. Miledi, 1967. Tetrodotoxin and neuromuscular transmission, Pro,:. R. Soc. London, Set. B 167, 8. Kelly, P.H. and K.F.. Moore, 1977, Mcsolimbic dopamine neurons: effects of 6-hydroxydopamine-induced destruction and receptor blockade on drug-induced rotation of rats, Psychopharmacology 55, 35. l.aping, N.J. and V.D. Ramirez. 1986, Prolactin induces yawning and the stretch-yawning syndrome in young adult male rats, Horm. Behav. 20, 49. Leong, D.A., L.S. Frawley and J.D. Neill, 1983, Neuroendotrine control of prolactin secretion, Ann. Rev. Physiol. 45. 109. Lindvall, O. and A. Bj~rklund, 1983, Dopamine- and norepinephrine-containing neuron system: Their anatomy' in the rat brain, in: Chemical Neuroanatomy, eds. O. Lindvall and A. BjiSrklund (Raven Press. New York) p. 229. Moore. K.E., 1987, Interactions between prolactin and dopaminergic neurons, Biol. Reprod. 36, 47. Moore, K.E., G.D. Riegle and K.T. Demarest, 1985. Regulation of tuberoinfundibular dopaminergic neurons: prolactin and inhibitory neuronal influences, in: Catecholamines as Hormone Regulator, eds. N. Ben-Jonathan, J.M. Bahr and R.I. Weiner (Ravan Press, New York) p. 31. Narahashi, T., 1972, Mechanism of action of tetrodotoxin and saxitoxin on excitable membranes, Fed. Proc. 31, 1124. Perkins, N.A. and T.C. Westfall, 1978, The effect of prolactin on dopamine release from rat striatum and medial basal hypothalamus, Neuroscience 3. 59. Pycock, C.J. and C.D. Marsden, 1978, The rotating rodent: a two component system, European J. Pharmacol. 47, 167. Smith, M.O. and R. Ilolland, 1975, Effects of lesions ¢~f the nucleus accumbens on lactation and postpartum behavior. Physiol. Psychol. 3, 331. Voigt, M.M.. R.Y. Wang and T.C. Westfall, 1985, The effects of cholecystokinin on the in vivo release of newly synthesized [3H]dopamine from the nucleus accumbens of the rat. J. Neurosci. 5, 2744.
1
Elsevier ILIP 50235
Comparison of the effect of prolactin on dopamine release from the rat dorsal and ventral striatum and from the mediobasal hypothalamus superfused in vitro J i n - C h u n g C h e n and Victor D. R a m i r e z * Department of Physiology" and Biophysics, Unieersi(v of Illinois, Urbana, IL 61801, U.S.A. Received 2 February1988, accepted 9 February 1988
In this paper, by means of a superfusion technique, we have examined the effect of ovine prolactin on dopaminergic neurons innervating the dorsal striatum, ventral striatum and the mediobasal hypothalamus of male rats. Fragments from dorsal striatum were superfused with ovine prolactin dissolved in normal medium (Krebs-Ringer phosphate) or medium containing tetrodotoxin (TTX, 10 6 M). Ovine prolactin stimulated the in vitro release of dopamine from dorsal and ventral striatal fragments. In dorsal striatal fragments a linear dose-dependent dopamine release was observed only when fragments were superfused with Krebs-Ringer phosphate-TTX medium. In addition, [LeuS]enkephalin (10 -6 and 10 -5 M) decreased the prolactin-induced in vitro dopamine release from dorsal striatal fragments superfused with Krebs-Ringer phosphate-TTX medium. Ovine prolactin (10 9 - 1 0 5 M) did not elicit changes in dopamine, 3,4-dihydroxyphenylacetic acid or 5-hydroxyindoleacetic acid outputs from mediobasal hypothalamic fragments superfused with Krebs-Ringer phosphate medium containing TTX. The possible regulatory mechanisms of ovine prolactin on dopaminergic neurons are discussed. Prolactin: Dopamine; Corpus striatum (dorsal/ventral): Tetrodotoxin: [Leu 5]enkephalin; Hypothalamus (mediobasal)
1. I n t r o d u c t i o n
Prolactin exerts multiple actions on central dopaminergic neurons in mammals (see review Drago, 1984). Neurochemical and pharmacological evidence (Moore et al., 1985; Leong et al., 1983) has indicated that prolactin can increase the synthesis/release of dopamine (DA) from tuberoinfundibular DAergic neurons. Among extrahypothalamic actions, prolactin administration induces grooming (Drago et al., 1984), yawning (Laping and Ramirez. 1986) and potentiates
* To whom all correspondence should be addressed: 524 Burrill Hall, Department of Physiology and Biophysics, University of Illinois, 407 South Goodwin Avenue, Urbana, IL 61801, U.S.A.
amphetamine-induced rotation behavior (Joseph et al., 1986). Prolactin also increases the number of DA binding sites (Hruska, 1986) and DA uptake by striatal synaptosomes (Gregerson and Selmanoff, 1985). Moreover, prolactin enhances electrical-induced DA release in in vitro striatal slices (Perkins and Westfall, 1978). These experiments strongly suggest that prolactin participates in the regulatory mechanisms controlling nigrostriatal DAergic activity. The finding of specific prolactin binding sites in the substantia nigra and striatum (Di Carlo et al., 1985) offers a possible explanation by which activation by prolactin of nigrostriatal DAergic neurons can be mediated. It is also known that prolactin-enhanced grooming behavior can be mediated through changes in DA metabolism of the nucleus accumbens (Drago et al., 1984). In addition, lesions of the nucleus ac-
0014-2999/88/$03.50 :~' 1988 Elsevier Science Publishers B.V. (Biomedical Division)
cumbens impaire lactational performance and maternal behavior (Smith and Holland, 1975). Thus, it is conceivable that prolactin also affects DAergic transmission of the mesolimbic pathway. Previously our group published that in vitro prolactin stimulated DA release from striatal tissue from male but not from female rats (Chen and Ramirez, 1982). In vivo, this hormone increased 3,4-dihydroxyphenylacetic acid (DOPAC) output from the caudate nucleus of male rats along with activation of grooming behavior (Chen et al., 1985). The aim of the present study was to compare the effect of prolactin on the activity of three DAergic systems of the central nervous system: (1) the nigrostriatal (dorsal striatum). (2) the mesolimbic (ventral striatum) and (3) the tuberoinfundibular (mediobasal hypothalamus) neurons. In addition, tetrodotoxin (TTX, 10 -6 M) known to block neurons impulse flow (Narahashi, 1972) was used in the in vitro superfusion system to examine a postulated presynaptic effect of ovine prolactin on DA neurons. Also, the interaction of [LeuS]enkephalin with prolactin on the nigrostriatal dopaminergic system was examined.
2. Materials and methods
2.1. Tissue preparation Adult male Spraque-Dawley rats (Holtzman, Madison, WI) weighing 300-500 g were housed in groups of five per cage in a temperature-controlled room (22-24°C) with lights on from 05:00 to 19:00 h. Water and Purina rat chow were available ad libitum. All rats were allowed to accustom to the environment for at least 7 days prior to the experiments. In all experiments, animals were killed by decapitation between 09:00 and 09:30 h.. The dorsal striatum (both sides, mainly nucleus caud a t e / p u t a m e n ) and mediobasal hypothalamus were dissected out as previously reported (Becket and Ramirez, 1980). The ventral striatal fragments that include the nucleus accumbens were taken from an area defined by a coronal cut about 1 m m anterior to the optic chiasma to expose the ventral striatum at the level of the anterior commissura,
and then four cuts were made at right angles: the dorsal cut was immediately above the anterior commissura near the midline approximately 2 mm long; the lateral cuts were made inside the lateral extension of the corpus callosum and the midline. respectively about 1.5 mm long followed by a ventral cut made at right angles to the first two lateral cuts to remove a fragment of about 1.0 mm thick. Similar cuts were performed in the contralateral side. Tissue fragments were immediately placed in the ice-cold Krebs-Ringer phosphate medium, pH 7.4, containing 10 mM sodium molybdate (Sigma) with or without TTX (1()-6M, Sigma). Prior to placing the tissue in an experimental chamber, the dorsal/ventral striatal fragments were divided into eight pieces by three additional perpendicular cuts and the mediobasal hypothalamus fragments were halved at a midsagittal plane. Sixteen striatal pieces (dorsal striatum: 34.7 + 5.2 rag; ventral striatum: 11.4 +_ 2.6 nag, n = 12 perifusion) or eight mediobasal hypothalamus halves (40.6_+3.2 mg, n = 12 perifusion) were placed in each experimental chamber. These wet weights correspond to values obtained at the termination of the superfusions.
2.2. Superft~'ion system The superfusion system was essentially the same as described previously (Gallardo and Ramirez, 1977) except that a 35 p,l/min flow rate was used. The time interval for the fluid to reach the tissuecontaining chamber (400 ktl dead volume) from the beaker was approximately 10 rain. Following a 45 rain stabilization period (30 rain for mediobasal hypothalamus), perfusate samples were collected on ice for ten or fi~urteen 15 n'fin intervals. In the first experiment, dorsal striatal fragments from the control group were superfused with Krebs-Ringer phosphate whereas for the experimental groups, ovine prolactin (Sigma) at concentrations of l0 v 10 6 or 10 -5 M was administered for 30 min starting from the beginning of collection interval 5. A similar experimental procedure but with Krebs-Ringer phosphate medium containing 10 ~ M TTX throughout the superfusion period was performed using the same doses of ovine prolactin (10-'7_10 5 M, 30 min).
In the second experiment, dorsal striatal fragments were superfused with Krebs-Ringer phosphate containing 10 - 6 M TTX plus either 10 -7, 10- 6 or 10- 5 M [Leu 5]enkephalin acetate (Sigma). T h e n , 10 - 6 M ovine prolactin was administered during intervals 6 and 7 for 30 rain. In experiment 3 and 4, ventral striatal and mediobasal hypothalamus fragments respectively, were superfused with Krebs-Ringer phosphate containing TTX. Ovine prolactin at concentrations of 10-7, 10--6 or 10 .-5 M was then administered for 30 rain during intervals 5 and 6. Twenty microlitres of untreated superfusate samples were then subjected to high performance liquid chromatography with electrochemical detection (HPLC-EC) for catecholamine and indoleamine analyses.
2.3. HPLC-EC system An HPLC-EC system (Bioanalytical System, Inc., West Lafayette, IN) with a C-18 reverse phase column (4.6 × 250 mm, 5 /,tm, Biophase, ODS) coupled to a carbon paste (CP-O) electrode at a potential of +0.65 V relative to the Ag/AgCI reference electrode was used. Mobile phase consisted of 5.75 g citric acid, 4.1 g sodium acetate, 85 mg sodium octyl sulfate, 7% methanol in 1 i of deionized water, adjusted to pH 4.5 with concentrated sodium hydroxide solution. The buffer was then degassed and filtered through a millipore membrane (0.45 #m, Millipore, Bedford, MA). External standards for DA, DOPAC, 5-hydroxyindoleacetic acid (5-HIAA) and homovanillic acid (HVA) were prepared fresh with superfusion medium at proper doses. Details of HPLC operation were described previously (Chen et al., 1984).
groups, at interval 6). ANOVA followed by posthoc comparisons was used to calculate significant differences among controls and groups treated with different doses of ovine prolactin (see fig. 3). A P < 0.05 value was considered significant.
3. Results
3. I. Effect of in vitro ovine prolactin administration on DA release from dorsal striatal fragments superfused with Krebs-Ringer phosphate with or without TTX The basal release ( p g / m g per min) of DA and its metabolite DOPAC from dorsal striatal fragments is illustrated in fig. 1. In general, the rate of in vitro DA and DOPAC release decreased significantly during the first 1.5 h of superfusion to reached a stationary phase thereafter (from about interval 5 to the end of the superfusion). The output of 5-HIAA and HVA became barely detectable after 90 min of superfusion (data not shown). TTX (10 - 6 M ) significantly reduced basal DA release during the stationary phase from a mean value of 7.0 + 0.9 in preparations superfused with Krebs-Ringer phosphate to 2.8 _+ 0.2 in preparations superfused with Krebs-Ringer phosphateTTX medium (last six intervals). The output of
,j -I
z
-
Je
=
OO..-
2.4. Statistical analysis All data in this report are expressed as means _+ S.E.M. A t-test was employed to compare mean values between two groups. To examine the effect of ovine prolactin in these systems, a two-tailed paired Student's t-test was employed to compare the mean release of DA, DOPAC or 5-HIAA corresponding to each interval after ovine prolactin administration with the mean basal level at interval 5 (for [LeuS]enkephalin experimental
=
I:DI~ :~ IOl
s',',~r,t
:_ N.M~
..r~
Fig. 1. In vitro output of DA and DOPAC from dorsal striatal fragments superfused with Krebs-Ringer phosphate medium or Krebs-Ringer phosphate medium containing TTX (10 -6 M). The basal release of DA and DOPAC reached a stationary, phase after approximately 1.5 b control period. Notice thai the DA output in Krebs-Ringer phosphate-TTX medium was in general lower than in normal Krebs-Ringer phosphate medium. See text for details.
--TTX{
~-POt-
"
oPRL IC~7M
F",,
~g o
i
,or k
C~/
:
....
J
J - -
O'6M) - - .
4
oPRL 10"TM "
U.
.
.
.
I
"
"
i .tim
(z
,~:z 0 C3 + % u 20
.
.
.
.
1
J
_ _
[
1
__
oPRL IO~M
1
l
oPRL 0 5M
SAMPLE N U M B E R
Fig. 2. Effect of ovine prolactin (oPRL) on in vitro DA output from dorsal striatal fragments superfused with Kxebs-Ringer phosphate medium or Krebs-Ringer phosphate medium containing TTX (10- ~ M). A dose-dependent stimulatory effect of ovine prolactin (10-%10 -~ M) on DA release was observed in the fragments superfused with Krebs-Ringer phosphate medium containing TTX but not from those superfused with normal Krebs-Ringer phosphate medium (* P < 0.05). See also fig. 3.
The effect of ovine prolactin on DA activity from dorsal striatal fragments is shown in fig. 2. Ovine prolactin tested at three doses: 10 -v, 10 (' or 10 --s M significantly enhanced basal DA release following prolactin administration in the TTX free Krebs-Ringer phosphate as well in the Krebs-Ringer phosphate-TTX preparations when compared to interval 5. However, in the former the response was not linearly related to the doses whereas in the latter a clear dose-response effect was obtained as shown in fig. 3 in which the peak D A responses from the data presented in fig. 2 as function of doses of ovine prolactin as well the values for control preparations are presented.
3.2. Effect of [LeuS /enkephalin on or,me prolactinenhanced DA release from dorsal striatal fragments Fig. 4 portrays the effect of 10 (' M ovine prolactin on in vitro D A and D O P A C release
)euS-EnKidTM
15-I'
DOPAC, however, decreased slightly from a mean value of 6.8 _+ 0.5 to 5.7 _+ 0.2 (last six intervals), respectively.
~
0PRL
l
I
10+6M
k\\\\~\\~'l
n : 4
0 c~
0
i
L
i
leuS- EnKIO'gM
~-.
,..4
,,
:o+ , +o,,_++ o° ° + +
..... + ...... , +
.
o~
+'
L
I u:
o
~.
CoP',c
.
.
.
.
£
'C
I
t-+\\-.\\\\+I
,~ 15
....
I
1_
LeuS-EnK I O ' S M
4
P-.\\\\\\~
"
o 6
Fig. 3. In vitro DA output from dorsal striatal fragments as function of three doses of ovine prolactin (10 v, 10-6, and 10-5 M) in preparations superfused with Krebs-Ringer phosphate medium or Krebs-Ringer phosphate medium containing "I3"X (10 6 M). Control preparations did not receive ovine prolactin. Peak DA outputs for the experimental preparations and basal values corresponding to interval 5 of controls and ovine prolactin-treated preparations are plotted. Notice a clear linear dose-response for the striatal fragments superfused with Krebs-Ringer phosphate medium containing TTX and the absence of a sigmficant response in the preparations superfused with Krebs-Ringer phosphate medium only.
~0
0i SAMPLE
NUMBER
Fig. 4. Effect o f 10 -~' M o'.ine p r o l a c t i n o n in vitro DA and D O P A C outputs from dorsal striatal fragments superfused with Krebs-Ringer phosphate-'ITX (10 6 M) a n d [LeuS]en k e p h a l i n at doses 10 v, 10 ~ o r 10 ~ M. [Leu'S]Enkephalin decreased the ovine prolactin-induced I)A release in a dose-dependent manner.
from dorsal striatal fragments superfused with Krebs-Ringer phosphate containing TTX (10 ..6 M) and [LeuS]enkephalin at doses 10 -7, 10 -6 or 10 -5 M. As shown in the figure, [LeuS]enkephalin tends to decrease basal as well as the ovine prolactin-induced DA release in a dose-dependent manner. Low doses of the ]LeuS]enkephalin (10 -7 M) did not significantly reduce the DA response of the terminals to 10 -6 M ovine prolactin (186% compared with 228% from control experiments without [ Leu 5]enkephalin when peak DA response were compared to interval 6) whereas middle and high doses (10 .-6 and 10 -5 M) partially or completely blocked the ovine prolactin-induced DA release (118 and 102%, respectively). In none of these groups were changes in D O P A C output observed.
e: I.Z
Unlike the basal release pattern of DA and D O P A C from the dorsal and ventral striatal frag-
n = ,4
I0 5 c E
0 o P R L 10"6M
t5
~
g
~o
c~
s
~:
2o
o P R L I0-~' M
i5 _~
3.4. Effect of in vitro ovine prolactin infusion on DA, DOPA C and 5-HIAA release from mediobasal hypothalamus fragments
o P R L 10-TM
15
3.3. Effect of ovine prolactin on DA and DOPAC release from ventral striatal fragments. The effect of ovine prolactin on mesolimbic DAergic terminals with Krebs-Ringer phosphate medium containing TTX (10 -6 M) is illustrated in fig. 5. Low doses of ovine prolactin (10 -7 M) did not affect either DA nor D O P A C release during the 2 h superfusion period following prolactin infusion. However, 10 -6 M ovine prolactin stimulated DA release reaching its peak response 30 min post-ovine prolactin infusion (7.4 + 2.5 in interval 5 vs. 11.2 + 2.5 p g / m g per min in interval 8) without affecting D O P A C output. A high dose of ovine prolactin (10 -5 M) induced a greater response since the stimulated DA release was prolonged (up to 45 min) and reached maximal levels of 16.8 + 3.4 at interval 7 compared to 9.0 + 1.3 p g / m g per min at interval 5. Moreover, a slight increase in D O P A C output was observed at interval 8. The percent changes of the peak DA response induced by ovine prolactin were 151% at 10 - 6 M and 187% at 10 -5 M.
TTX (iO'6M)
20
to 5 !
I
I
J
t
5
IO
15
SAMPLE NUMBER
Fig. 5. Effect of ovine prolactin on in vitro DA and D O P A C outputs from ventral striatal fragments superfused with Krebs-Ringer p h o s p h a t e - l T X (10 6 M) medium ( * P < 0.05). Notice that ovine prolactin was clearly effective at l 0 6 and l0 =5 M.
ments, the DA and D O P A C levels in superfusates from mediobasal hypothalamus fragments remained low and stable from the very beginning to the end of the superfusion, probably a reflection of the integrity of the tubuloinfundibular DA neurons in this tissue preparation (fig. 6). Detectable levels of 5-HIAA were also found throughout the experiment, however, with a significant decrease in levels when initial and final values were compared (3.8 + 0.5 and 1.8 + 0.3 p g / m i n , n = 12, respectively). However, neither of the four doses of ovine prolactin tested had any effect on the release of these three neurochemicals. As shown in fig. 6, ovine prolactin infused either at very low doses of 10 -9 M (n = 4) or at high doses of 10 -5 ( n = 2 ) and 10-6 M ( n = 2 ) did not affect the basal release of DA nor of D O P A C since in all the
TTX (IO'6M) oPRL 10-gM i-.\\\\\N\',I
n=4
o~
[
o P R L I0-;' M .........
#5
IIl
I °
~\\\\\\\x 1
I
5
SAMPLE
I0
I 15
NUMBER
Fig. 6. Effect of ovine prolactin on in vitro DA, I)OPAC and 5-HIAA outputs from mediobasal hypothalamus fragments superfused with Kxebs-Ringerphosphate-TTX medium. S.E.M. bars are depicted only for the DA profiles. Notice that I)A and DOPAC levels were low and stable during the superfusion. The 5-HIAA output showed a 2-fold decrease when the beginning and the end values were compared. Neither of the four doses of ovine prolactin had any effect on the release of these three neurochemicals.
experiments stable levels of DA and D O P A C (2.8 _+ 0.2 and 3.1 + 0.5 p g / m g per min, respectively) were detected during the pre- or post-ovine prolactin superfusion periods.
4. Discussion Considerable evidence has accumulated supporting the concept that prolactin affects extra-hypothalamic DAergic neurons (Drago, 1984; Chen and Ramirez, 1985). In the present study in preparations superfused with TTX ovine prolactin was highly effective in increasing the in vitro DA release from either the dorsal or the ventral striaturn, target sites of the nigrostriatal and meso-
limbic DAergic systems respectively, adding further support to this contention. Of the three doses of ovine prolactin tested (10 -7 , 10 ~ and 10 -~ M) in dorsal striatai tissue superfused with normal Krebs-Ringer phosphate medium, only the 10 -7 and 10 5 M ovine prolactin significantly enhanced DA release above control superfusions (fig. 3) but not linearly. Interestingly, identical doses of ovine prolactin elicited DA release in a linear dose-dependent manner when dorsal striatal fragments were perfused with Krebs-Ringer phosphate medium containing 10 ~ M TTX suggesting the existence of regulatory interneuronal inputs controlling DA release within the fragments which can then be blocked by the TTX treatment. Tetrodotoxin, a sodium channel blocker, that exerts a highly specific action on nerve excitation without any effect on other membrane parameters, has been used as a tool to study the mechanism of transmitter release from nerve terminals without being interfcrcd with by impulse flow in both presynaptic and postsynaptic elements (Narahashi, 1972; Katz and Miledi, 1967). Consequently, the present results clearly indicatc that ovine prolactin acts on the nerve terminals of the dorsal striatal to elicit DA release. TTX (10 ~ M) also reduced the basal relcasc of DA (60%) and D O P A C (20%) from dorsal striatal fragments as compared to control tissues. The reduction by TTX of in vivo basal DA release estimated with push-pull perfusion was previously reported by Voigt et al. (1985) and discussed, however, as duc to a direct action upon some DAergic axons which are spontaneously active. Our in vitro results in preparations in which DA axons were severed from their cell bodies but still contained viable interneurons connections suggest that the action of TTX is due to inhibition of action potentials from these interncurons, resulting in a net decrease of DA release. In support of a presynaptic control of DA release by interneurons is the finding that the decrement of basal DA release under TTX left unimpaired the ovine prolactin-evoked DA response which was even higher in preparations superfused with TTX than in control (e.g. 228, 631% in Krebs-Ringer phosphate-TTX vs. 125,264% in Krebs-Ringer phosphate at 10 -~' and 10 5 M ovine prolactin, respectively).
Furthermore, our studies using [LeuS]enkepha lin, an opioid peptide that can inhibit DA release (Loh et al., 1976) significantly blocked the ovine prolactin-induced DA release from dorsal striatum fragments, confirming the existence of a complex presynaptic regulatory mechanism controlling DA release from the nigrostriatal dopaminergic terminals (Dray, 1979). Considering that prolactin- and ACTH-induced grooming behavior (Drago et at., 1984; Gispen et al., 1975) result from an activation of nigrostriatal and mesolimbic DA transmission, respectively it is significant to note that ACTH-induced grooming in the rat can be suppressed by treatment with opiates (Cools et al., 1978). In the ventral striatal tissue preparation, ovine prolactin also stimulated DA release. There is evidence indicating that prolactin increases DA turnover rate in the nucleus accumbens of aMTtreated rats (Fuxe et al., 1977; 1978). In addition, we have found that local administration of ovine prolactin within the nucleus accumbens increased in vivo DOPAC output further supporting these data (paper submitted for publication). It is known that DAergic terminals in the nucleus accumbens originate from the ventral tegmental area and substantia nigra (Lindvall and Bj~3rklund, 1983). Moreover, the necessary co-existence of nucleus accumbens and caudate nucleus DA neurons for the performance of normal rotation behavior (Pycock and Marsden, 1978; Kelly and Moore, 1977) indicate a functional interaction between these two DA structures. The observation of a similar stimulatory effect of ovine prolactin in these distinct but interrelated DAergic pathways suggest that not only an anatomical contiguity exist between these two systems but also a similar functional response to ovine prolactin. in view of the postulated existence of a short inhibitory feedback loop between tubuloinfundibular DA neurons and prolactin (Gudelsky and Porter, 1980; Moore, 1987), it may appear surprising that neither of the doses of ovine prolactin tested in the present investigation (10 9, 1 0 - 7 , 10 - 6 n o r 1 0 .s M) affected basal DA or DOPAC release from mediobasal hypothalamus fragments. The failure to detect any significant changes in the release of these neurochemicals upon ovine pro-
lactin administration might be explained by the fact that only a 2 h period post-ovine prolactin infusion was studied. Similar observations have been reported by Fuxe et al. (1977) and Perkins and Westfall (1978) since prolactin did not enhance DA turnover in aMT-treated animals elicit basal DA release from the mediobasal hypothalamus in short-term experiments. Although we cannot exclude the possibility that ovine prolactin may activate tubuloinfundibular DA neurons through interneurons blocked by the TTX treatment, we favor the hypothesis of an effect of ovine prolactin on the synthesis of DA through activation of tyrosine hydroxylase. Demarest et al. (1984) clearly showed that there are two components in the prolactin-induced activation of tubuloinfundibular DA tyrosine hydroxylase activity. The relatively rapid 'tonic' component that is activated approximately 4 h post-systemic administration of prolactin and the delayed 'induction' component that occurs 12 h post-prolactin administration. Recently, our own data indicate that there is approximately a 30% increase in tyrosine hydroxylase activity in the mediobasal hypothalamus but not in the dorsal or the ventral striatal 4 h after ovine prolactin administration (submitted for publication). Hence, there appears to be significant temporally related differences as well differences in the mechanism of action of ovine prolactin on DA metabolism between the nigrostriatal-mesolimbic DAergic system and the tubuloinfundibular DA system. In the former ovine prolactin activates preferentially and directly the release mechanism of DA which can occur within 1 h whereas in the latter the 'tonic component' of the synthetic mechanism of DA appears to be initially activated. In summary, the results from this study provide the first indication of a presynaptic effect of ovine prolactin on the release of DA from nigrostriatal and mesolimbic DAergic terminals. The role of ovine prolactin on these systems and clarification of its physiological significance remains to be elucidated.
Acknowledgement This work was in part support by a NIH Grant No. HD-14625 to V.D.R.
References Becket, J. and V.D. Ramirez, 1980, Dynamics of endogenous catecholamine release from brain fragments of male and female rats, Neuroendocrinology 31, 18. Chen, J.C., A.D. Ramirez and V.D. Ramirez, 1985, Prolactin effect on in vivo striatal dopaminergic activity as estimated with push-pull perfusion in male rat, in: Prolactin, Basic and Clinical Correlates, eds. R.M. MacLeod, M.O. Thorner and U. Scapagnini (Liviana Press, Padova) p. 523. Chen, J.C., K.K. Rhee, D.M. Beaudry and V.D. RamireT, 1984, In vivo output of dopamine and metabolites from the rat caudate nucleus as estimated with push-pull perfusion online with HPLC-EC in unrestrained, conscious rats, Neuroendocrinology 38, 362. Chen, Y.F. and V.D. Ramirez, 1982, Prolactin stimulates dopamine release from male but not from female rat striatal tissue superfused in vitro, Endocrinology 111, 1740. Cools, A.R., V.M. Wiegant and W.H. Gispen, 1978, Distinct dopaminergic systems in A C T H - i n d u c e d grooming, European J. Pharmacol. 50, 265. Demarest, K.T.. G.D. Riegle and K.E. Moore, 1984, Prolactininduced activation of tuberoinfundibular dopaminergic neurons: evidence for both a rapid 'tonic' and a delayed "induction" component, Neuroendocrinology 38, 467. Di Carlo, R., G. Muccioli, D. Lando and G. Bellussi, 1985, Further evidence for the presence of specific binding sites for prolactin in the rabbit brain. Preferential distribution in the hypothalamus and substantia nigra, Life Sci. 36, 375. l)rago, It:., 1984, Prolactin and sexual behavior: a review, Neurosci. Biobehav. Rev. 8. 433. Drago, F., G.L. Kov~.cs, G. Continella and U. Scapagnini, 1984, Dopamine in the nucleus accumbens: its role in prolactin-enhanced grooming behavior of the rat, Biogenic Amines 1, 75. Dray, A., 1979, The striatum and substantia nigra: a commentary on their relationships, Neuroscience 4, 1407. Fuxe, K., K. Andersson, T. HiSkfeh and L.F. Agnati, 1978. Prolactin-monoamine interactions in rat brain and their importance in regulation of LH and prolactin secretion, in: Progress in Prolactin Physiology and Pathology, eds. C. Robyn and M. Harter (Elsevier/North-Holland Biomedical Press) p. 95. Fuxe, K., P. Eneroth, J.-A. Gustafsson, A. L~SfstriSm and P. Skett, 1977, Dopamine in the nucleus accumbens: preferential increase of DA turnover by rat prolactin, Brain Res. 122, 177. Gallardo, E. and V.D. Ramirez, 1977, A method for the superfusion of rat hypoth-,dami: secretion of luteinizing hormone-releasing hormone (LH-RI-1), Proc. Soc. Exp. Biol. Med. 155, 79. Gispen, W.H., V.M. Wiegant, H.M. Greven and D. De Wied, 1975, The induction of excessive grooming in the rat by intraventricular application of peptides derived from AC'I'tI: structure-activity studies, Life Sci. 17, 645.
Gregerson, K.A. and M. Selmanoff, 1985, F.ffects of hyperprolactinemia on dopamine uptake into striatal synaptosomes in vitro, in: Prolactin, Basic and Clinical Correlates, eds. R.M. MacLeod, M.O. Thorner and U. Scapagnini (Liviana Press, Padova) p. 625. Gudelsky, G.A. and J.C. Porter, 1980, Release of dopaminc from tuberoinfundibular neurons in pituitary' stalk bhn~d after prolactin or haloperidol administration, l-ndocrinol. 106, 526. Hruska, R.I-.. 1986, Modulatory role for prolactin in the elevation of striatal dopamine receptor density' induced by chronic treatment with dopamine receptor antagonists. Brain Res. Bull. 16, 331. Joseph, J.A., G.S. Roth and A.S. IAppa, 1986, Reduction of motor behavioral deficits in senescent animals via chronic prolactin administration I. Rotational behavior. Ncurobiol. Aging 7, 31. Katz. B. and R. Miledi, 1967. Tetrodotoxin and neuromuscular transmission, Pro,:. R. Soc. London, Set. B 167, 8. Kelly, P.H. and K.F.. Moore, 1977, Mcsolimbic dopamine neurons: effects of 6-hydroxydopamine-induced destruction and receptor blockade on drug-induced rotation of rats, Psychopharmacology 55, 35. l.aping, N.J. and V.D. Ramirez. 1986, Prolactin induces yawning and the stretch-yawning syndrome in young adult male rats, Horm. Behav. 20, 49. Leong, D.A., L.S. Frawley and J.D. Neill, 1983, Neuroendotrine control of prolactin secretion, Ann. Rev. Physiol. 45. 109. Lindvall, O. and A. Bj~rklund, 1983, Dopamine- and norepinephrine-containing neuron system: Their anatomy' in the rat brain, in: Chemical Neuroanatomy, eds. O. Lindvall and A. BjiSrklund (Raven Press. New York) p. 229. Moore. K.E., 1987, Interactions between prolactin and dopaminergic neurons, Biol. Reprod. 36, 47. Moore, K.E., G.D. Riegle and K.T. Demarest, 1985. Regulation of tuberoinfundibular dopaminergic neurons: prolactin and inhibitory neuronal influences, in: Catecholamines as Hormone Regulator, eds. N. Ben-Jonathan, J.M. Bahr and R.I. Weiner (Ravan Press, New York) p. 31. Narahashi, T., 1972, Mechanism of action of tetrodotoxin and saxitoxin on excitable membranes, Fed. Proc. 31, 1124. Perkins, N.A. and T.C. Westfall, 1978, The effect of prolactin on dopamine release from rat striatum and medial basal hypothalamus, Neuroscience 3. 59. Pycock, C.J. and C.D. Marsden, 1978, The rotating rodent: a two component system, European J. Pharmacol. 47, 167. Smith, M.O. and R. Ilolland, 1975, Effects of lesions ¢~f the nucleus accumbens on lactation and postpartum behavior. Physiol. Psychol. 3, 331. Voigt, M.M.. R.Y. Wang and T.C. Westfall, 1985, The effects of cholecystokinin on the in vivo release of newly synthesized [3H]dopamine from the nucleus accumbens of the rat. J. Neurosci. 5, 2744.