Joiiriinl c ! / ~ ~ , i i r n c h r i n i . ~ l r ~ ~

Raven Press. Ltd.. New York C) 1992 International Society for Neurochernistry

Endogenous Noradrenaline and Dopamine in Nerve Terminals of the Hippocampus: Differences in Levels and Release Kinetics Matthijs Verhage, Wim E. J. M. Ghijsen, *Frans Boomsma, and Fernando H. Lopes da Silva Deparltntw f of Experitnental Zoology, University of Amsterdam, Amsterdam: and *Department of Internal Medicine, Erasmiis University, Rotterdam, The Netherlands

Abstract: The presence and release of endogenous catecholamines in rat and guinea pig hippocampal nerve terminals was studied by fluorimetric HPLC analysis. In isolated nerve terminals (synaptosomes) the levels and breakdown of endogenous catecholamines were determined and the release process was characterized with respect to its kinetics and Ca2+and ATP dependence. Endogenous noradrenaline and dopamine, but not adrenaline, were detected in isolated hippocampal nerve terminals. For dopamine both the levels and the amounts released were more than 100-fold lower than those for noradrenaline. In suspension. released endogenous catecholamines were rapidly broken down. This could effectively be blocked by monoamine oxidase inhibitors, Ca2+-freeconditions, and gluthatione. The release of both noradrenaline and dopamine was highly CaZ+and ATP dependent. Marked differences were observed in the kinetics of release between the two catecholamines. Noradrenaline showed an initial burst of release within 10 s after K+ depolarization. The release of noradrenaline was

terminated after approximately 3 min of K+ depolarization. In contrast, dopamine release was more gradual, without an initial burst and without clear termination of release within 5 min. It is concluded that both catecholamines are present in nerve terminals in the rat hippocampus and that their release from (isolated) nerve terminals is exocytotic. The characteristics of noradrenaline release show several similarities with those of other classical transmitters, whereas dopamine release characteristics resemble those of neuropeptide release in the hippocampus but not those of dopamine release in other brain areas. It is hypothesized that in the hippocampus dopamine is released from large, dense-cored vesicles, probably colocalized with neuropeptides. Key Noradrenaline-Dopamine-HippocampusWords: Nerve terminals-Release kinetics-Rat-Guinea pig. Verhage M. et al. Endogenous noradrenaline and dopamine in nerve terminals of the hippocampus: Differences in levels and release kinetics. J . Neurochem. 59, 88 1-887 ( 1 992).

The catecholamines are among the first substances recognized as neurotransmitters. Still, their exact role in signal transduction has remained relatively indistinct compared with the clear-cut role of the fast-acting “ionotropic” neurotransmitters such as glutamate, y-aminobutyric acid (GABA), and acetylcholine. Depending on the system, the release of catecholamines can exhibit characteristics of both the rapid signaling pathway of fast-acting neurotransmitters present in small synaptic vesicles and the slower signaling pathway of neuropeptides contained in large, dense-cored vesicles. Catecholamines are believed to be stored in both types of vesicles or even in separate, morphologically distinct categories (for a review see Thureson-Klein and Klein, 1990).

In the hippocampus the levels of catecholamines seem rather low as judged by immunocytochemical studies (see Lopes da Silva et al., 1990). It has been estimated that probably less than 1% of the total input in the rat hippocampus contains noradrenaline (Storm-Mathisen, 1977) and it has been suggested that dopamine actions in the hippocampus are due to a cross-reactivity with adrenergic p receptors, rather than to a functional dopamine system (Malenka and Nicholl, 1986). Nevertheless, catecholamine transmitters do exert clear effects on cells of the hippocampus (for a review see Lopes da Silva et al., 1990) and have been implicated in long-term potentiation in the hippocampus (Bliss et al., 1983; Stanton and Sarvey, 1985; Frey et al., 1991). Immunological detection

Received October 22, 1991; final revised manuscript received February 10, 1992: accepted February 24, 1992. Address correspondence and reprint requests to Dr. M. Verhage at R. Magnus Institute, Institute for Molecular Biology (IMB), Uni-

versity of Utrecht. Padualaan 8. 3584 CH Utrecht. The Netherlands. ;Ihhrrviations n.sc.d.. GABA. y-aminobutyric acid: MAO. monoarnine oxidase.

88 1

M . VERHAGE ET AL.

882

methods with light microscopy may underestimate the (local) catecholamine synthesis capacity. Alternatively, the relatively long extracellular half-life of catecholamine transmitters suggests that they may act beyond the synaptic clefts and that catecholamine signals may therefore reach cells of the hippocampus after release from nerve terminals outside the hippocampus. In this study the question we asked is whether the catecholamines dopamine and noradrenaline are present in and released from the nerve terminals of the hippocampus and whether this release resembles that of other “classical,” fast-acting neurotransmitters or that of slow neuromodulators. Evidence for a specific transmitter/modulator function of certain catecholamines, especially dopamine, in the hippocampus cannot be based on studies using the uptake and release of radioisotopes (Schroeder et al., 1989; Arias et al., 1990),as it is well established that both the plasma membrane and the synaptic vesicle catecholamine carriers do not exhibit great specificity among different monoamines (for reviews see Smith, 1972;Njus et al., 1986; see also Maycox et al., 1990).Thus nerve terminals can transport and accumulate various catecholamines when present at millimolar concentrations. Radiolabeled dopamine may thus be taken up into noradrenergic and serotonergic synapses and be metabolized inside synaptic vesicles to form noradrenaline. Radiolabeled noradrenaline may be taken up in serotonergic and dopaminergic synapses and vesicles. Experimental evidence is presented to support these possibilities (Kelly et al., 1985; Carboni et al., 1990). Therefore, in this study the release of endogenous dopamine and noradrenaline was analyzed from isolated nerve terminals by a novel technique using HPLC after precolumn extraction and derivatization, and fluorimetric detection. EXPERIMENTAL PROCEDURES Materials Percoll was obtained from Pharmacia (Uppsala, Sweden); ionomycin, from Calbiochem (La Jolla, CA, U.S.A.); pargyline. from Sigma; and dinonylphthalate from BDH. All other chemicals, of the purest grade available, were from Merck (Darmstadt, F.R.G.).

Synaptosomal preparation Synaptosomes were prepared from the hippocampi of one or two male Wistar rats (1 80-240 g) or from male Dunkin-Hartley guinea pigs (250-300 g). Synaptosomes were purified using a modification of the methods of Nagy and Delgado-Escueta (1984) and Dunkley et al. (1986) as described before (Verhage et al., 1988, 1989).with 7.5/10/23’% discontinuous Perco11/0.32 M sucrose gradients. Synaptosomes were collected from the interface 10-23% and diluted in artificial CSF of the following composition (mM): 132 NaC1. 3 KCI, 2 MgSO,, 1.2 NaH,PO,, 10 D-glucose, 10 HEPES, and 0.02 CaCl,. The pH was adjusted at 7.4 with Tris. Synaptosomes were pelleted and resuspended in CSF with 1.5 m M CaCI, and kept on ice. Protein content was J. Neurocliem.. Vol. S9, N(J.3. 1992

measured according to Bradford (1976) with bovine serum albumin as a standard.

Release assay Synaptosomal aliquots (0.5 mg of protein) were spun down before release experiments in a BHG table centrifuge (12.000 g. 1 min) and resuspended in 500 pl of CSF with, routinely. 100 p M pargyline (see Results) and either 1.5 mM CaCI, or 50 pAd EGTA to remove contaminant CaZ+. Synaptosomes (0.5 mg of protein in 500 pl of medium) were preincubated for 10 min in a shaking water bath at 36°C. After 10 min synaptosomes were depolarized with 30 m M K+. After addition. samples weregently mixed. For the estimation of initial release (5-30 s). incubations were terminated by the addition of an equal volume of ice-cold 10 m M EGTA (in water, pH 8.2 with NaOH). Synaptosomes were immediately pipetted onto a layer (200 pl) of silicone oil (Dow Corning 550). diluted 5050% (vol/vol) with dinonylphthalate to increase fluidity and to facilitate migration of synaptosomes through the layer. The suspension and the oil layer were spun for 90 s in a BHG table centrifuge (12.000 g), separating the synaptosomes from the incubation medium. For longer incubations synaptosomes were incubated at 36°C and were spun through oil after 1,3, or 5 min. Samples were taken from the supernatant and pipetted on 1 ml of ice-cold 100% methanol and stored in liquid nitrogen before extraction (see below). Pellets were resuspended in a similar volume and synaptosomes were lysed with sonification (2X 30 s on ice).

Analysis of endogenous catecholamines Synaptosomal catecholamine content was determined after lysis with sonification (2X 30 s o n ice) in CSF with 100 p M pargyline, 100 pMglutathione. or 5 M E G T A . Breakdown was estimated by incubation with 100 pmol of dopamine and I nmol of noradrenaline per mg of protein in the lysate. During release experiments extracellular breakdown of catecholamines was minimized with 100 pM pargyline, 100 p M glutathione (see Results). Catecholamines were extracted by liquid/liquid extraction according to van der Hoorn et al. ( I989), using ammonium/ammonia chloride buffer (pH 8.6) and n-heptanelloctanol after complex formation of the catecholamines with a large hydrophobic tail (diphenylborate/ethanolamine complex, 8.9 mM). Extracted catecholamines were derivatized by incubation with 1,2-diphenylethylenediamineand separated on a MicroSpher C,*column (100 mm X 4.6 mm I.D.; Chrompack, Middelburg, The Netherlands) with a mobile phase consisting of 50 mM sodium acetate (pH 7.0)/acetonitrile/methanol (50:40:8, vol/vol/vol), at a flow rate of 1 ml/min. Fluorescence detection was performed by excitation at 350 nm and monitoring emission at 480 nm.

ATP/ADP ratios ATP levels were estimated with the luciferase ATP assay kit (LKB/ITL, Heemstede. The Netherlands), with the extraction of ATP and ADP and estimation of ADP according to Kauppinen and Nicholls (1986) with perchloric acid/ EDTA. Luciferase luminescence was sampled at 2-s intervals with a CFS interface program (provided by Perkin Elmer). To deplete synaptosomes from ATP, synaptosomes (0.5 mg of protein in 0.75 ml) were incubated for 15 min at 36°C in normal medium. but without glucose and in the presence of 2 pg/ml oligomycin to block ATP synthesis and hydrolysis. and in 1 m M EGTA to deplete synaptosomes of Ca’’. Following this incubation. synaptosomes were

883

CA TECHOLAMINES IN HIPPOCAMPAL NER VE TERMINALS washed, resuspended in a similar volume, and equilibrated for 5 rnin at 36°C before the addition of 5 pM ionomycin and 2.5 mM CaCI,. Oligomycin did not interfere with the luciferase assay. Each analysis was directly calibrated by standard additions of ATP in the assay.

EXOGENOUS NORADRENALINE

r

I

EXOGENOUS DOPAMINE

RESULTS Analysis of hippocampal catecholamines The release of endogenous noradrenaline and dopamine from isolated and semipurified nerve terminals from rat hippocampus could be detected after liquid/liquid extraction and derivatization. The detection limit, at a signal-to-noise ratio of 3, was 1.78 fmol (0.3 pg) ml-’ for noradrenaline and 3.27 fmol (0.5 pg) ml-’ for dopamine as the derivatized compound. Figure 1 shows chromatograms of extracted catecholamines from samples representing basal efflux (middle) and release evoked by chemical depolarization with 30 mMK+ in the presence of Ca2+(right). Both dopamine release and noradrenaline release were significantly higher than background ( p < 0.0 I ) and were at least threefold higher than the detection limit (also in the case of basal efflux). Release of both catecholamines increased on K+ depolarization in the presence of Ca2+(Fig. 1). Because a small, broad fluorescence signal appeared near the retention time of dopamine, the peak height was taken into account for the quantification of dopamine levels, rather than the surface under the peak (as with noradrenaline and the internal standard). No detectable amounts of adrenaline were released from or present inside synaptosomes from rat or guinea pig hippocampus. Synaptosomal catecholamine content For quantification of the amounts of catecholamines present in the nerve terminal fraction of the

,m 2 m

a

standar

-L

-

K+-evoked release

5 rnin

FIG. 1. Typical chromatograms of fluorimetrically detected endogenous catecholamines separated by HPLC after precolumn extraction and derivatization. Left: Standard solutions, containing 250 pg/ml of noradrenaline and adrenaline, 100 pg/ml of dopamine, and 500 pg/ml of isoprenaline (internal standard). L indicates the “lump” mentioned in the text. Middle and right: Extracted catecholamines from synaptosomal release. Middle, basal efflux; right, after 3 min of K+ depolarization in the presence of Ca2+.

+ + - - + - - + I

I

PARGYLINE GLUTATHIONE EXCESSEGTA

- + + - - + . - - - +

FIG. 2. Recovery of exogenous catecholaminesin synaptosomal lysates. Synaptosomes were lysed by sonification (see Experimental Procedures) in normal CSF with the additions indicated. Columns represent means k SEM of four experiments. Recovery of exogenous catecholamineswas estimated after 3-min incubations at 36°C in the presence of 100 pmol of dopamine and 1 nmol of noradrenaline per mg of protein in the lysate.

hippocampus, the extracellular stability of catecholamines was tested first by testing the recovery of exogenous catecholamines incubated with lysed synaptosomes. Figure 2 shows the retrieval of exogenous dopamine and noradrenaline. When Ca2+was present in the suspension, the concentration of both catecholamines declined rapidly. Within 3 min at 36°C the duration of most release experiments, picomolar concentrations of exogenous catecholamines declined to approximately 50-60% of the added amounts. This breakdown was effectively blocked by the addition of excess EGTA (Fig. 2). However, when studying the release process it is necessary to incubate the synaptosomes in the presence of Ca2+.In this case, the extracellular breakdown of catecholamines could be blocked by the monoamine oxidase (MAO) inhibitor pargyline ( 100 pM, Fig. 2). Dopamine was especially sensitive to metabolism through endogenous M A 0 activity. The addition ofthe nonspecific oxidase inhibitor glutathione (100 p M ) had a small additional effect in protecting dopamine from breakdown in lysed synaptosomes (Fig. 2). In the presence of EGTA and both inhibitors the total amounts of endogenous dopamine and noradrenaline in rat and guinea pig hippocampal synaptosomes were analyzed (Table 1). In both species the TABLE 1. Total atnolints of endogenoils catecholarnines per tnilligrum of protein in pur(jied nerve terminuls jrom rat and guinea pig hippocampus Hippocampus

Noradrenaline

Dopamine

Adrenaline

Rat Guinea pig

52 f 5.1 pmol 24 k 3.1 pmol

200 +- 86 fmol 710 t 220 fmol