L - DOPA Facilitates the Release of Endogenous Norepinephrine

In rat hypothalamic slices, L-aromatic amino acid decarboxylase (AADC) was assayed, and the actions of L-DOPA on impulse (2 Hz)-evoked norepinephrine (NE) and dopamine (DA) release were studied under inhibition of AADC. Slices were incubated with L-DOPA, and DA and NE produced by conversion of the precursor were analyzed by high performance liquid chromatography with electro chemical detection (HPLC-ECD). In the slices, the Km and Vmax of AADC were 131 ƒÊM and 122 pmol/min/mg protein, respectively. NSD-1015, an AADC inhibitor, caused a noncompetitive type of inhibition, and the Ki value was 0.086 ƒÊ M. In the presence of 20 ƒÊM NSD-101 5, which was expected to cause 99.6% inhibition of AADC, L-DOPA (0.01-100 nM) concentration-dependently facilitated the release of NE from the superfused slices, and the L-DOPA (10 nM)-induced facilitation was antagonized by 100 nM ICI 89,406 and 100 nM ICI 118,551, a selective ƒÀ1and ƒÀ2-adrenoceptor antagonist, respectively. This action of LDOPA was not modified by 30 ƒÊM tropolone, an inhibitor of catechol-0-methyltransferase. L-DOPA at 0.01-1 nM similarly facilitated the release of DA. A quantitative analysis revealed that the L-DOPA-induced increase in NE and DA release was much higher by a factor of 3 to 4 ordres than was the amount of DA and NE converted from L-DOPA. These results add further support to the hypothesis that L-DOPA itself acts as a neuroactive substance in the rat central nervous system. It has been generally accepted that L-3,4 dihydroxyphenylalanine (L-PO PA) is merely a precursor of dopamine (DA) and that the effects of L-DOPA are mediated through its conversion to DA by the catalyzing action of L-aromatic amino acid decarboxylase (AA DC) (EC. 4.1.1.28). However, we found that endogenous DOPA was released from rat striatal slices by depolarizing stimuli in a transmitter-like manner (1, 2). Furthermore, exogenously applied L-DOPA produced dual pharmacological actions on the release of norepinephrine (NE) and DA from rat hy pothalamic slices, facilitation via presynaptic (3-adrenoceptors at nanomolar concentra tions and inhibition via presynaptic DA receptors at micromolar concentrations (3). Similar biphasic actions of L-DOPA were also evident in rat striatal slices (4). The present work concerns further studies on L-DOPA with regards to its facilitation of NE and DA release via presynaptic (3-adreno ceptors in rat hypothalamic slices. As this action was observed in the presence of an AADC inhibitor (3), this did not appear to be as a result of a conversion of L-DOPA to DA and/or NE (5), but to be mainly due to L DOPA itself. Indeed, the level of NE produced from the conversion of L-DOPA is very low compared with that of DA (6-8). However, there remains the possibility that even in the presence of an AADC inhibitor, DA and/or NE may be synthesized by residual AADC ac tivities and may be released into the synaptic cleft to produce facilitation of the release of NE and DA: AADC is most active (5, 9-13); the degree of this enzyme inhibition achieved by an inhibitor varies with experimental con ditions (10, 14); even a nonenzymatic decar boxylation of L-DOPA was noted (15); and DA or NE could be an agonist for presynaptic Q-adrenoceptors (16-18). In addition, there has been no report evaluating the activities of AADC or the efficacy of AADC inhibitors in tissue slice preparations. We examined the ac tivities of AADC in rat hypothalamic slices to obtain evidence that in the presence of AADC inhibitors, L-DOPA facilitates the release of NE and DA via presynaptic 8-adrenoceptors in much higher amounts than those of these catecholamines synthesized from this pre cursor. Materials and Methods AADC assay: Male Sprague-Dawley rats (200-250 g) were given food pellets and water ad libitum and kept on a regular day and night schedule (light on at 6:00 A.M. and off at 8:00 P.M.). The animals were decapi tated, the brains placed on ice and the hy pothalami dissected out and sliced sagittally, 0.3 mm in thickness, using a Mcllwain tissue chopper. One to four slices (0.5-2 mg protein) were transferred to 175 /cl of Krebs-Henseleit medium in air-tight polypropylene test tubes (1.5 ml) and preincubated at 37°C for 1 5 min. The composition of the medium was as fol lows: 113 mM NaCI, 25 mM NaHCO3, 4.75 mM KCI, 1.18 mM KH2PO4, 2.52 mM CaC12, 1.19 MM MgSO4, 11.1 mM glucose, 0.029 mM disodium EDTA and 0.29 mM ascorbic acid. The medium was bubbled with 5% CO2 in 02, for at least 30 min immediately before use. In some experiments, an AADC inhibitor, NSD-1 055 or NSD-1015, was applied to the preincubation medium and was present throughout the following incubation. After preincubation, 175 ,al-aliquots were removed, and AADC assays were started by adding the same volume of medium contain ing L-DOPA or ring-labeled [2,5,63H]-L DOPA (specific activity 47.3 Ci/mmole, New England Nuclear, Boston, MA). The assay samples were incubated for 5 to 30 min. [3H]-L-DOPA was previously purified by high performance liquid chromatography (HPLC) on a ,a-Bondapack C18 reverse-phase column using 1 mM phosphoric acid as the mobile phase. The pH of the incubation medium re mained unchanged by the addition of L DOPA or [3H]-L-DOPA solution. In routine assays, incubation was for 15 min with one slice of about 0.5 mg protein. Incubation was stopped by adding to each assay tube 1 ml of ice-cold 0.1 N HCIO4 containing 50 pmol isoproterenol, as an internal standard. The slices were homogenized, centrifuged and 1 ml samples of the supernatant were used for the determination of DA, 3,4-dihydorxy phenylacetic acid (DOPAC) and NE. Un labeled or tritiated DA, DOPAC and NE in the tissue extracts were partially purified by the alumina adsorption method, as described by Goshima et al. (3); and a 20 icl-aliquot of the eluate (100 pl) was analyzed by HPLC with electrochemical detection (-ECD) (Yanagi moto Co., Ltd., Kyoto, Japan). In the assay with [3H]-L-DOPA, radioac tivities of [3H]-DOPAC and [3H]-NE syn thesized were analyzed in each corresponding fraction of a 1 ml-eluate collected after pas sage through the ECD. Ten milliliters of ACS II solution (Amersham/Searle Corporation, Des Plaines, IL) were added to each fraction, and the tritium content was determined using a liquid scintillation spectrometer (Beckman LS 5800). Sensitivity of the assay system for [3H]-DA, [3H]-DOPAC and [3H]-NE was 10 fmol. Details on the chromatographic data are as follows: column, Cosmosil 5C18 (4.6 x 150 mm), mobile phase, 0.1 M phosphate potassium buffer (pH 3.0) containing 10% methanol, 0.02% I-octanesulfonate sodium and 1 mM disodium EDTA; applied potential, 600 mV vs. Ag/AgCI and flow rate, 0.7 ml/ min. Kinetics of the two AADC inhibitors were studied using Dixon plots (19). The degree of % inhibition (i%) by these non-competitive inhibitors (see Results) was determined using the equation: i%=100 ([I]/(Ki+[I])), where I is the concentration of the inhibitor. Superfusion experiments: Superfusion ex periments of brain slices were performed as described (1 ). In brief, four pieces of hypo thalamic slices prepared as described above were superfused with Krebs-Henseleit me dium in the presence of 20 jiM cocaine. Electrical field stimulation (25 V, 2 Hz, al ternative polarity) was performed for 3 min through platinum spiral electrodes at the two ends of the chamber, using an electrical stimulator with an isolator (SEN-3021 and SS-201J, Nihon Kohden), at 30 (So as a test stimulation), 60 (S,) and 90 (S2) min after the start of superfusion. Superfusates were collected every 3 min. The evoked release (S) of NE and DA during the S, and S2 periods of stimulation was calculated as the total minus the basal release. The amount of NE and DA in 3-min samples immediately before stimula tion was regarded as the spontaneous release (Sp). L-DOPA was added to the medium 15 min before S2, and the effects were evaluated by the release ratio, S2/S1, and Sp2/Sp1. Pretreatment with 20 /tM NSD-1015 or 30 ,uM tropolone and with the 8-antagonists used was initiated 20 min and 60 min before S,, respectively, and was continued through out the experiments. N E and DA released and their contents in the tissues at the end of superfusion were measured by HPLC-ECD, as described (3). Drugs used were 4-bromo-3-hydroxy benzyloxyamine (NSD-1 055) (Nacalai Tesque Ltd., Kyoto, Japan); 3-hydroxybenzyl hydrazine (NSD-1015) (Aldrich, Inc., Mil waukee, WI); L-DOPA (Wako Pure Chemical, Ltd., Osaka, Japan); ICI 89,406, erythro-DL 1 (7 methylindan 4 yloxy) 3 isopropyl aminobutan-2-ol (ICI 118,551) (Imperial Chemical Industries, Macclesfield, Cheshire, England); and tropolone (Sigma Chemical Co., St. Louis, MO). The peaks of NE and DA were not interfered with by application of these drugs. Protein concentration was determined by the method of Lowry et al. (20) with bovine serum albumin as the standard. Statistical significance of difference was calculated using the unpaired Student's t test (two-tailed). Fig. 1. Synthesis of DA (DA plus DOPAC) and NE from L-DOPA in rat hypothalamic slices. A, content of DA (0), DOPAC (A) and NE (0) in a one slice sample (0.5 mg protein) incubated with 100 uM L DOPA for the indicated times. B, the synthesis of DA in one to four slices incubated with 100 aM L-DOPA for 15 min, as a function of protein concentrations of the slices (abscissa). C, double reciprocal Line weaver-Burk plot of L-DOPA concentration against/ Results Determination of Km and Vmax value of AADC in rat hypothalamic slices: Formation of DA and its metabolite DOPAC was time dependent (Fig. 1A) and proportional to ,rate of DA formation. The line of best fit of the data was obtained by linear regression analysis. In panel C, incubation was for 15 min with one slice as enzyme. The apparent Km and VmaX values were 131 ,uM and 122 pmol/min/mg protein, respectively. Prior to the addition of L-DOPA, the slices were preincubated for 15 min with oxygenated Krebs medium. Content of DA plus DOPAC in slices incubated in the absence of L-DOPA was 35.1 +4.3 pmol/mg protein (n=1 5), and this was regarded as the blank value. In panels A and C, each value was determined in triplicate, tissue protein concentrations (Fig. 1 B), with 100 itM L-DOPA as the substrate. In contrast, NE contents in the slices appeared to be con stant during incubation with L-DOPA, at least for 30 min (Fig. 1A). Lineweaver-Burk plots (21) of L-DOPA concentraion against the rate of DA (DA plus DOPAC) formation are shown in Fig. 1 C. The apparent Km toward L-DOPA and the VmaX value were 131 AIM and 122 pmol/min/mg protein, respectively. In this assay, however, it was difficult to evaluate the low levels of NE synthesized under normal conditions and DA synthesized under conditions of AADC inhibition. Fig. 2. Synthesis of [3H]-DA ([3H]-DA plus [3H]-DOPAC) (9) and [3H]-NE ([l) as a function of [3H]-L-DOPA concentration in hypothalamic slices. Assay conditions \/\,,ere as in the legend in Fig. 1C, except that [3H]-L-DOPA (10-100 nM) was used as the substrate. Synthesized [3H]-DA, [3H] DOPAC and [3H]-NE were separated by HPLC-ECD. The sensitivity of the assay system for 3H-DA, 3H -DOPAC and 3H-NE was 10 fmol . Each point was determined in duplicate. Synthesis of [3H]-DA, [3H]-DOPAC and [3H1-NE from [3H]-L-DOPA in rat hypothala mic slices: When the substrate, [3H]-L-DOPA (10 nM), was incubated for 30 min at 37°C without slices, the radioactivities recovered in [3H]-L-DOPA, [3H]-DA, [3H]-DOPAC and [3H]-NE fractions were 5357±544 dpm, 51 ± 2 dpm, 63±7 dpm and 67±6 dpm (n=3), respectively, the latter three being the same as the blank value. This indicates that non enzymatic decarboxylation shown in in vitro assay tubes (15) was negligible in this assay system. With the slices as the enzyme source, synthesis of [3H]-DA ([3H]-DA plus [3H] DOPAC) was dependent on the concen trations of [3H]-L-DOPA (Fig. 2) and was also time dependent and proportional to tissue protein concentrations (data not shown). The synthesis of [3H]-NE was detectable even when the slices were in cubated with nanomolar concentrations of [3H]-L-DOPA (Fig. 2) and was also de pendent on the concentration of [3H]-L DOPA. The mean endogenous content of DA plus DOPAC and NE for all tissue samples in Fig. 2 was 46.2±11.3 and 73.2±5.3 pmol/ mg protein, respectively (n=8). With 100 nM [3H]-L-DOPA as substrate, the amounts of [3H]-DA and [3H]-NE converted per 15 min were 5.8% and 0.094% of the final tissue con tent, respectively. Dixon plots analysis of the effects of NSD 1055 and NSD-1015 on AADC in hypo thalamic slices: The inhibitory effects of the two hydrazine derivatives, NSD-1055 and NSD-1015 on AADC were studied using [3H]-L-DOPA as the substrate. Both the in hibitors, when applied for 15 min in the ab sence of [3H]-L-DOPA, had definite inhibi tory effects on AADC in rat hypothalamic slices. Dixon plots (Fig. 3) showed that NSD 1055 and NSD-1015 caused a non-competi tive type of inhibition, and the respective K; was 1.14 ,tM and 0.086 /iM. Ten iiM NSD 1055 and 20 /tM NSD-1 015 are expected to Give 89.8% and 99.6% inhibition, respectively. Fig. 3. Dixon plots analysis of the effects of NSD-1055 (A) and NSD-101 5 (B) on AADC in rat hypo thalamic slice. Experimental conditions were as in Fig. 2, except that an appropriate concentration of NSD-1055 or NSD-101 5 was applied for 1 5 min of the preincubation periods, and it was present through out the assays. The Ki value for NSD-1055 and NSD-1015 was determined at three to seven NSD concentrations. Each value is the mean of the intercepts generated from two separate lines. The lines of best fit of the data were obtained by linear regression analysis. Each point was determined in duplicate. Effects of L-DOPA on impulse (2 Hz) evoked release of N E and DA from superfused rat hypothalamic slices in the presence of 20 pM NSD-1015: The spontaneous release (Spy) of NE and DA in superfused rat hy pothalamic slices in the presence of 20 riM NSD-1015 was 0.1 23±0.01 pmol/mg protein and 0.077±0.016 pmol/mg protein (n=29), respectively. The evoked release (S,) of NE and DA at 2 Hz was 0.306±0.037 pmol/mg protein and 0.234±0.039 pmol/mg protein (n=29), respectively. After superfusion, the respective tissue content of N E and DA was 121.5±8.0 and 102.8±11.7 pmol/mg protein (n=29). The fractional evoked release of NE and DA was 0.264±0.034% and 0.272± 0.040% (n=29), respectively. L-DOPA at 0.01 to 100 nM concentration-dependently facilitated the evoked release of NE: 10 to 100 nM L-DOPA increased the S2/S, ratios by approximately 50% in the presence of 20 itM NSD-1015 (Fig. 4). Similarly, L-DOPA at 0.01 to 1 nM concentration-dependently facilitated the release of DA (Fig. 4), the peak seen at 1 nM. On the other hand, L-DOPA at 10 and 100 nM was without effect, an ob servation which can be explained by the balance between the facilitation of the DA Fig. 4. Effects of L-DOPA on impulse (2 Hz)-evoked release of NE and DA from superfused rat hypo thalamic slices in the presence of 20 pM NSD-101 5. Ordinates: ratios of NE and DA released by S2 and S, periods of stimulation. Abscissae: concentrations of L-DOPA. L-DOPA (0.01-100 nM) was added to the superfusion medium 15 min before S2. NSD-1015 (20,uM) was applied 20 min before S, and was present throughout the experiments. Data represent the mean±S.E.M. of at least three estimations. Statistical significance: *P<0.05, **P<0.01, compared to the corresponding control at 0 nM. Table 1. Effects of ICI 89,406, ICI 118,551 and tropolone on L-DOPA (10 nM)-induced facilitation of evoked NE release from superfused rat hypothalamic slices in the presence of 20 uM NSD-1015 release via presynaptic i9-adrenoceptors and the inhibition via presynaptic DA receptors with higher concentrations of L-DOPA under conditions of AADC inhibition (4). The L DOPA produced no effect on the spontaneous release (Sp2/Sp1) of NE and DA, and their tissue content in slices after the superfusion experiments. Antagonism by ICI 89,406 and ICI 118,551 on L-DOPA (10 nM)-induced facilitation of the evoked release of N E in the presence of 20 aM NSD-1015: L-DOPA (10 nM) induced facilitation of the evoked release of NE was completely antagonized by 100 nM ICI 89,406 and 100 nM ICI 118,551, selective 31 and 92-antagonists, respectively. Tropol one, a catechol-0-methyl-transferase in hibitor, at 30 ItM had no effect on the L DOPA-induced facilitation of the release of NE (Table 1). These pretreatment had no effects on Sp,, S, or on the ratio Sp2/Sp1 of NE and DA. Discussion Under the condition that the activities of AADC in rat hypothalamic slices were es sentially completely inhibited, nanomolar con centrations of exogenously applied L-DOPA facilitated the impulse-evoked release of NE and DA, and the facilitation of NE release was completely antagonized by selective j91 and Q2-adrenoceptor antagonists. The L DOPA-induced facilitation was seen even in the presence of tropolone, a catechol-0 methyl transferase inhibitor, suggesting that this action was not mediated through its con version to 3-0-methyl-DOPA (22). These findings suggest that L-DOPA itself facilitates the release of NE via presynaptic 81 and J 2 adrenoceptors. These results extend our pre vious findings (3, 4) and further support the view that L-DOPA is a neuroactive substance in the rat brain (1, 2). The Km and Vmaxvalues of AADC estimated with micromolar concentrations of unlabeled L-DOPA as substrate parallel findings de scribed for rat brain homogenate (9, 11-13), indicating the validity of our method for the first demonstration of an AADC assay in brain slice preparations. Although the Vmax value was somewhat lower compared to those found by Sims et al. (13) and Hahman et al. (12); these discrepancies are probably due to different experimental conditions such as pH and the presence or absence of the added coenzyme pyridoxal-5-phosphate (9, 10, 13). In the assay with the nanomolar concen trations of [3H]-L-DOPA used as the sub strate, converted [3H]-NE as well as [3H]-DA were detectable; and here, the ratio of NE and DA was about 1:40, a value consistent with that observed in the brains of rats ad ministered with L-DOPA in vivo (7, 8). This radioassay also permitted accurate measurement of synthesized [3H]-DA in the rat hypothalamic slices, under conditions of AADC inhibition. With 100 nM [3H]-L-DOPA as substrate, the amount of [3H]-DA syn thesized in the presence of 20 tcM NSD-1 015 was calculated to be only 0.3% of the total [3H]-L-DOPA, an amount much less than that of DA formed through non-enzymatic decar boxylation, 6% of total L-DOPA, in in vitro assay tubes (15). The present results of a negligible amount of nonenzymatic DA forma tion are in line with those shown by Macko wiak et al. (23). When the slices were pretreated with NSD 1055 or NS D-101 5, either of these inhibitors caused a noncompetitive type of inhibition with a low K; value, thereby indicating potent efficacies on AADC. These results are con sistent with those by other authors (10, 14). NSD-1 015 is much more potent than NSD 1055 in inhibiting AADC. In the superfusion of rat brain slices, we used 10 W NSD-1015 and 20 i M NSD 101 5, the agents expected to give a 89.8% and 99.6% inhibition of AADC, respectively. Even when AADC in the rat hypothalamic slices was essentially completely inhibited by 20 gM NSD-1015, the exogenously applied L-DOPA facilitated the evoked release of NE and DA. This finding further supports the idea that L DOPA itself facilitates the release of these catecholamines (3, 4). L-DOPA (0.01 to 1 nM) concentration dependently facilitated the release of DA in the presence of 20 /LM NSD-1 015, and the concentration-release-curve was shifted to the left by a factor of approximately 2 orders from that seen in the presence of 10 pM NSD-1055 (3). This finding is not consistent with the idea that L-DOPA facilitates DA release through its conversion to DA, thereby producing the resultant increase in the amount of DA available for release (5). The difference between the two concentration release curves might be explained by the degree of AADC inhibition achieved. In the presence of 10 i M NSD-1055, DA converted by residual AADC activity (10%) might inhibit the release of DA via presynaptic DA autoreceptors (24) and then might interfere with the facilitatory action of L-DOPA itself, to a greater degree compared to that seen with 20j M NSD-1015. On the other hand, L-DOPA (0.01 to 100 nM) also concentration dependently facilitated the release of NE in the presence of 20 i M NSD-1015, and this concentration-release-curve was much the same as that obtained with 10 flM NSD-1 055 (3). This similarity between the two curves may be related to our previous findings that tonic regulatory functions of presynaptic dopamine receptors on NE release are minor compared to those of autoreceptors on DA release (17, 24) and that exogenously applied DA only slightly inhibits the NE release (17). DA converted from L-DOPA in the presence of 10 uM NSD-1055 (3) could hardly inter fere with the facilitatory action of L-DOPA on the N E release. releasable pool of these catecholamines. The L-DOPA (10 nM)-induced facilitation of NE release was antagonized by either the ,31-selective antagonist ICI 89,406 (25) or the 132-selective antagonist ICI 118,551 (26). The antagonism by propranolol against the L-DOPA action was stereoselective (3). These results obtained by 3 structurally different /9 adrenoceptor antagonists indicate that L DOPA facilitates NE release selectively via presynaptic a1 and i92-adrenoceptors. These findings are in agreement with the evidence of coexistence of both types of presynaptic 13 adrenoceptors in the same rat hypothalamic slices (27). L-DOPA could not be a direct agonist for these 8-adrenoceptors, because L-DOPA could not displace [3H]-dihydro alprenolol binding in the membrane prep arations of rat brain (28). Some unknown mechanisms must underlie the L-DOPA induced facilitation. In fact, there are tonically functioning facilitatory presynaptic (3-adre noceptors on the noradrenergic and dopamin ergic nerve endings in the same slice prep arations, 132-adrenoceptors are predominant over (3t -adrenoceptors, and an endogenous agonist is epinephrine (27, 29, 30). One possibility is that exogenously applied L DOPA itself potentiates the facilitatory actions of endogenous (3-agonists such as epi nephrine, which are released in response to nerve stimulation (31). On the other hand, there remains the possibility that even in the presence of AADC inhibitors, DA and/or NE converted from L-DOPA (<100 nM) by residual AADC activities is released into the synaptic cleft and acts on presynaptic 13 We compared L-DOPA-induced increases in NE and DA release with the amount of NE and DA converted from L-DOPA in the presence of NSD-1 015 (Table 2). The former was much higher by a factor of 3 to 4 orders, lending further support to the idea that the facilitatory effects of L-DOPA on the DA and NE release are not due to increases in the Table 2. L-DOPA-induced increases in DA and NE release (AS) and amount of [3H]-DA and [3H]-NE converted from [3H]-L-DOPA (Converted) in the presence of 20 ,uM NSD-1015 adrenoceptors to facilitate the release of NE. However, this appears to be unlikely because L-DOPA at 10 nM continued to facilitate the release of NE without facilitating the DA release (Fig. 4). Furthermore, the concen tration of exogenously applied DA, which was required to induce facilitation of NE release via presynaptic 8-adrenoceptors, is higher than 300 nM (17). Involvement of NE seems to be more unlikely because the amount of NE converted from L-DOPA is extremely low compared to that of DA (6-8). In conclusion, micromolar concentrations of hydrazine derivatives led to a marked inhibition of AADC in rat hypothalamic slices; and under nearly complete inhibition of this enzyme, nanomolar concentrations of L-DOPA facilitated the impulse-evoked release of N E via presynaptic 81 and Q2 adrenoceptors, in a much higher amount than that synthesized from this precursor. Acknowledgments: The authors thank K. Ohno for excellent technical assistance. We also thank M. Ohara for reading the manuscript. ICI 89,406 and ICI 118,551 were kindly provided by Imperial Chemical Industries (Macclesfield, Cheshire, Eng land).