- Email: [email protected]
Effects of monoamine oxidase inhibitors on levels of catechols and homovanillic acid in striatum and plasma
~eur~p~~~coiagy Vol. 28, No. 8. pp. 791-797, 1989 Printed in Great Britain
~28.3~8/89 %3.00+ 0.00 Maxwell Pergamon Macmiilan pIc
EFFECTS OF MONOAMINE OXIDASE INHIBITORS ON LEVELS OF CATECHOLS AND HOMOVAN~L~IC ACID IN STRIATUM AND PLASMA D, HOVEVEY-SION’, I. J. KOPIN’, R. W. STULL’ and D. S. GOLDSTEIN~* ‘Intramural Research Program, National Institute of Neurolo~caI Disorders and Stroke, and 2Hypertension-Endocrine Branch, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland 20892, U.S.A. (Accepted 4 February 1989)
Summary-Levels of homovanillic acid (HVA), dihydroxyphenylacetic acid (DOPAC) and dihydroxyphenyiglyco1 (DHPG) in plasma and the striatium were measured after inhibition of monoamine oxidase type A (MAO-A) by clorgyline (4 mgikg i.p.), MAO-B by (-fdeprenyl (1 mg/kg i.p.), both MAO-A and MAO-B by niatamide (75 mg/kg i.p.) or peripheral neuronal MAO by debrisoquin (40 mg/kg i.p.). Levels of HVA in plasma decreased by about 60% after single doses of nialamide or clorgyline, by about 80% after repeated doses of niaiamide, by about 40% after a single dose of debrisoquin and by about 50% after repeated doses of debrisoquin. The administration of clorgyline, niaiamide or debrisoquin significantly decreased concentrations of DOPAC and DHPG in plasma, whereas (-)deprenyl did not affect levels of DHPG or HVA. None of the MAO inhibitors produced more than about 80% depression of levels of any of the deaminated metabolites. The results suggest that most of the HVA in plasma is derived from de~ination of DA by MAO-A in peripheral neurons; that DOPAC in plasma is derived from cells outside the central nervous system; that DHPG in plasma is derived virtually exclusively from the metabolism of norepinephrine in sympathetic nerve endings and that residual levels of HVA after treatment with debrisoquin provide an improved but limited indication of central dopaminer~c activity. Kqv words-homovanillic norepinephrine
acid, monoamine oxidase, 3,4-dihydroxyphenylglycol,
Homovanillic acid (HVA), produced from the enzymatic deamination of dopamine (DA) by monoamine oxidase (MAO) to form 3,4-dihydroxyphenylacetic acid (DOPAC) and from the enzymatic 0-methylation of DOPAC by cat~hol-~-methyltransferase (COMT), is the major final metabolite of DA in humans and rats (Kopin, 1985). Two subtypes of MAO have been described, MAUA and MAO-B. The distribution of these subtypes in tissue appears to differ in that MAO-B is not prominent in nervous tissue, whereas MAO-A is present in both neural and non-neural tissue (Waldmeier, 1987). The relative contributions of these subtypes of MAO to levels of HVA in plasma have been poorly understood. In the present study, these contributions were assessed using subtype-specific and nonspecific MAO inhibitors: nialamide, which inhibits both subtypes of MAO (MAO-A and MAO-B); clorgyline, which selectively inhibits MAO-A; and (-)deprenyl, which selectively inhibits MAO-B. Alterations of levels of HVA in plasma in psychiatric disorders and in response to drugs have been used to assess changes in dopaminergic activity in the brain (Davis, Davidson, Mohs, Kendler, Davis, Johns, DeNigris and Horvath, 1985; Roy, Agren, Pickar,
*To whom correspondence
should be addressed.
DOPAC, dopamine,
Linnoila, Doran, Cutler and Paul, 1986; Roy-Byrne, Uhde, Sack, Linnoila and Post, 1986). Whereas the concentration of HVA in cerebrospinal fluid (CSF) is decreased in Parkinson’s disease (Burns, LeWitt, Ebert, Pakkenberg and Kopin, 1985), in primates that have been treated with 1-methyl-4-phenyl- 1,2,3,6tetrahydropyridine (MPTP), a neurotoxin which induces a Parkinsonian-like syndrome (Burns, Chiueh, Markey, Ebert, Jacobowitz and Kopin, 1983), the level of HVA in plasma is affected little, consistent with only a small contribution from the central nervous system (CNS) to HVA in plasma. Using decrements in methoxyhydroxyphenylglycol (MHPG) to estimate decrements in HVA derived from noradrenergic neurons after the administration of debrisoquin, an MAO inhibitor which does not penetrate the Moodbrain barrier, Kopin, Bankiewicz and Harvey-White (1988) recently estimated that only about 25% of the concentration of HVA in plasma was derived from dopaminergic neurons in the brain. Because of this substantial peripheral source of HVA in plasma, the residual level of HVA in plasma after the ad~nistration of debrisoquin, which inhibits MAO but does not appear to traverse the blood-brain barrier, has been proposed as providing a better method for estimating dopaminergic activity in the central nervous system (Maas, Contreras, Seleshi and Bowden, 1988; Riddle, Leckman, Cohen, Anderson, 791
192
D.
HOVEVEY-SIONet
Ort, Caruso and Shaywitz, 1986; Sternberg, Heninger and Roth, 1983). Accurate interpretation of the results, however, requires knowledge about the relative efficacy of debrisoquin and globally-acting MAO inhibitors in reducing the concentration of HVA in plasma. For instance, if the administration of a large dose of a globally-acting MAO inhibitor resulted in the same proportional fall in the concentration of HVA in plasma as obtained after debrisoquin, then it could be concluded that all of the HVA in plasma was derived from peripheral tissue. Therefore, in the present study, responses of levels of HVA in plasma after the administration of debrisoquin were compared with responses after the administration of the globally-acting MAO inhibitor, nialamide. Incomplete suppression of levels of HVA in plasma after the administration of debrisoquin could be interpreted as indicating insufficient efficacy of the drug as an MAO inhibitor at doses used in vivo, alternative pathways in the formation of HVA, besides through deamination of dopamine, or buildup of substrate catecholamines in the axoplasm. These explanations could have different consequences for estimates of central dopaminergic activity based on levels of HVA after treatment with debrisoquin. To evaluate these possibilities, changes in the levels in plasma of DOPAC and 3,4-dihydroxyphenylglycol (DHPG), the main deaminated metabolites of DA and norepinephrine (NE), were compared with changes in the levels of HVA in plasma. By comparing effects of globallyacting MAO inhibitors with those of debrisoquin on DHPG and DOPAC in plasma, central and peripheral sources of DHPG and DOPAC in plasma also could be identified. To verify the central or peripheral sites of action of the MAO inhibitors, levels of HVA and DOPAC in the striatum were measured in addition to levels of HVA, DOPAC, and DHPG in plasma. Since central neural inhibition of MAO would be expected to increase DA in tissue, it was determined whether levels of DA in the striatum would increase after the administration of drugs which inhibit MAO in brain. Time-response data were obtained for nialamide and clorgyline, and the effects of single and multiple doses were examined. METHODS
al.
Test procedure
Animals were transferred to clean cages at least 1 hr before the administration of drugs. At various times after treatment with drugs, the rats were anesthetized with pentobarbital (Somnifer, Richmond Veterinary Supply Co., Richmond, Virginia) and blood from the vena cava collected into heparinized tubes and the tubes put on ice. All blood samples were obtained between 12: 30 and 2: 30 p.m. in order to avoid the diurnal variation in levels of HVA in plasma (Sack, James, Doran, Sherer, Linnoila and Wehr, 1988). The animals were also deprived of food for at least 14 hr before blood sampling. The plasma was separated within 1 hr by refrigerated (4°C) centrifugation at 3000g. In one experiment involving repeated withdrawal of blood from the same conscious animals, blood was collected by tail-cut (initial tail-cut under light ether anesthesia). Plasma was stored at -70°C. In some experiments, immediately after blood sampling, the rat was decapitated and the brain removed and placed on a chilled glass plate covered with Whatman No 1 paper, soaked with ice-cold saline. The left striatum was dissected within 2min and frozen on dry ice. Samples of brain tissue were kept at -70°C until assayed. In 4 rats, a baseline blood sample was obtained by tail-cut and then debrisoquin was administered repeatedly (6 doses at 12 hr intervals); a second sample of blood was obtained 4 hr after the last dose. The same animals were then injected with 4 doses of nialamide (at 12 hr intervals) and blood was obtained by tail-cut 6 hr after the last injection. Assay methods
Dopamine, DOPAC and HVA in the striatum were assayed using high pressure liquid chromatography with electrochemical detection (LC-ED), as described by Chiueh, Markey, Burns, Johannessen, Pert and Kopin (1984), NE, DOPAC and DHPG in plasma, as described by Eisenhofer, Goldstein, Stull, Keiser, Sunderland, Murphy and Kopin (1986) and HVA in plasma, as described by Hovevey-Sion, HarveyWhite, Kopin and Goldstein (1988). The limits of detection were 10-25 pg per sample for each of these substances.
Animals
Sources of drugs and chemicals
Male Sprague-Dawley rats (Taconic Farm, Germantown, New York) weighing 200-350 g were housed with a 12 hr light/dark cycle and room temperature 24°C.
Dopamine, DOPAC, DHPG, HVA, NE and nialamide were obtained from Sigma (St Louis, Missouri), clorgyline from Research Biochemicals Incorporated (Wayland, Massachusetts), (-)deprenyl from Dr J. Knoll, Semmelweiss University of Medicine, Budapest, Hungary and debrisoquin sulfate from HoffmannLa-Roche (Nutley, New Jersey).
Administration of drugs
Drugs were administered intraperitoneally (i.p.) in a volume of 4ml/kg (8 ml/kg when nialamide was injected). Doses and schedules of administration for the various drugs are presented in the legends of the figures and table.
Statistical analyses
Results were expressed as mean values f SEM for the indicated number of animals. Statistical signifi-
MAO inhibitors and plasma HVA
793
cance was determined using Student’s t-tests or oneway analyses of variance (ANOVA) as appropriate. RESULTS
After the administration of a large, single dose of the globally-acting, subtype-nonspecific MAO inhibitor, nialamide, the levels of the deaminated DA metabolites, DOPAC and HVA in plasma and the striatum were reduced (Fig. 1). Levels of DOPAC in the striatum were reduced by about 70% at 1.5 hr and continued to decline to 13% and 10% of control at 6 and 16 hr. Similarly, levels of HVA in the striatum progressively decreased during the 16 hr after the administration of nialamide (Fig. 2). Levels of DOPAC and HVA in plasma fell more slowly than did levels of HVA in the striatum. Levels of DOPAC and HVA in plasma were not significantly decreased at 1.5 hr, but by 16 hr the HVA in plasma had decreased to about 40% of control (Fig. 2). Levels of DHPG in plasma also were decreased after the administration of nialamide (Table 1). Levels of DA in the striatum nearly doubled within 1.5 hr of the injection of nialamide, and the levels of DA were increased by 40% and 50% at 6 and 16 hr. The administration of (-)deprenyl did not significantly affect levels of DOPAC or HVA in the striatum or in plasma (Fig. I), nor levels of DHPG in plasma or levels of DA in the striatum (Table 1). In contrast, the globally-acting MAO-A inhibitor, clorgyline, significantly reduced levels of HVA, DOPAC and DHPG in plasma (Fig. 3) with the maximum decrease in HVA to about 40% of control at 6 hr. Levels of DHPG and DOPAC decreased to about l/4 to l/3 of control between 6 and 8 hr.
n
04. 0
I~,.,.,.,.,.,.,,,
2
4
6
Hours after
8
10
injection
12
14
16
18
of niolomide
Fig. 2. Levels of HVA in plasma and the striatum after the injection of nialamide. Nialamide (75 mg/kg) or saline (8 ml/kg) was injected intraperitoneally. Blood samples and striatal tissue were obtained at 1S, 6 and 16 hr after injection of drug. Results are percentages of saline-treated controls, sacrificed at the same time as the nialamide-treated animals.
After the administration of debrisoquin to inhibit peripheral neuronal MAO, levels of DOPAC and DHPG in plasma also were significantly decreased (Table 1 and Fig. 4). Levels of DOPAC and DHPG in plasma were reduced to about l/3 of control, whereas HVA was reduced to 57% of control (Fig. 4). Debrisoquin had no effect on levels of DA or DA metabolites in the striatum. By 4 hr after the last of 6 repeated doses of debrisoquin over 3 days, the level of HVA in plasma had decreased significantly to 47% of control (Fig. 5). In contrast, by 6 hr after the last of the repeated doses of nialamide in the same animals, the level of HVA in plasma had decreased to 20% of control. The additional reduction of HVA after the administration of nialamide, compared with that of debrisoquin,
NIALAMIDE
Fig. 1. Effects of MAO inhibitors on levels of DOPAC and HVA in plasma and in the striatum of rats. Saline (4 ml/kg), deprenyl (1 mg/kg), or nialamide (75 mg/kg) was injected intraperitoneally, and blood samples and striatal tissues were obtained 1.5 hr after the injection of the drugs. *P < 0.05(ANOVA) compared with saline.
D. HOVEVEY-SION et al.
194
Table I. Effects of MAO inhibitors on levels of norepinephdnc and DHPG in plasma and on levels of dopamine in the striatium 1.5 hr after the administration of drugs Treatment
rm7 0
I
1
1
4
Hours after
r 6
injection
-
s
’ 10
of clorgyline
NE in plasma (pg/ml) DHPG in plasma (pg/ml) DA in striatum (rig/g tissue)
Saline
Deprenyl (1 mg/kg)
Nialamide (75 mg/kg)
371 + 136
416 k 150
360 * 122
673 f 82
771 i 139
9475 it 259
9550 f 577
218* k 43 18,360. f 964
Treatment
NE in plasma (pg/ml) DHPG in plasma (pg/ml) DA in striatum (rig/g tissue)
Saline
Debrisoquin (40 mg/kg)
455 * 153
608 i I85
803 + 130 11,810~680
268* + 56 13,022 i 555
‘Different from saline, P < 0.05. NE = norepinephrine; DHPG = dihydroxyphenylglycol; DA = dopamine. Upper data analyzed by ANOVA, lower by r-test.
0
2
Hours after
4
injection
6
I)
10
of clorgyline
Fig. 3. Time course of inhibition of levels of HVA, DOPAC and DHPG in plasma by clorgyline. Upper panel, HVA in plasma, lower panel, DOPAC, and DHPG in plasma. Inset results are shown as percentages of control. Clorgyline (4 mg/kg) or saline (4 ml/kg) was injected intraperitoneally, with blood samples collected at 4, 6, and 8 hr after the injection.
led to the estimate that the central nervous system contributed by at most 27% to HVA in plasma. DISCUSSION
Nialamide, which inhibits MAO-A and MAO-B and clorgyline, which inhibits MAO-A, both cross
the blood-brain barrier and therefore inhibit MAO in the brain and periphery. Nialamide and clorgyline produced marked decreases of similar magnitude in the levels of the deaminated metabolites of catecholamines in plasma. In particular, levels of HVA in plasma decreased to about 20% of control. Consistent with previous observations (Westerink, Bosker and Wirix, 1984), levels of dopamine in the striatum were approximately doubled after inhibition of MAO-A, whereas inhibition of MAO-B with (-)deprenyl did not alter levels of the deaminated catecholamine metabolites in plasma or in the striatum. Therefore, MAO-B appears to be unimportant in the metabolism of dopamine in rats (Waldmeier, Delini-Stula and Maitre, 1976; Waldmeier, 1987).
Fig. 4. Effects of debrisoquin on levels of DOPAC and HVA in plasma and the striatum. Saline (4 ml/kg) or debrisoquin (DEBRISO, 40 mg/kg) was injected intraperitoneally, and blood samples and striatal tissue were obtained 4 hr after injection of drug. Data analyzed by ANOVA.
MAO inhibitors and plasma HVA
Fig. 5. Effects of repeated doses of debrisoquin, followed by repeated doses of nialamide, on levels of HVA in plasma. Pre-treatment blood samples were withdrawn by tail-cut. Debrisoquin (DEBRISO, 40 mg/kg, i.p.) was injected 6 times every 12 hr for 3 days. Four hr after the sixth injection of debiisoquin, blood was collected by tail-cut. Animals were then treated with 4 doses of nialamide (NIAL, 75 mg/kg, i.p.), 12 hr between doses. Six hr after the 4th injection of nialamide, a final blood sample was obtained. Blood samples were collected between 12: 30 and 2 : 30 p.m.
Even after repeated administration of effective doses of the MAO inhibitors, levels of HVA in plasma were not reduced to zero; maximum reductions of HVA in plasma were to about 20% of control. This could be explained by alternative routes of formation of HVA, other than through oxidative deamination of DA, by incomplete inhibition of MAO, or by buildup of cytoplasmic levels of catecholamines. Transamination of 3,4_dihydroxyphenylalanine (DOPA) (Sandler, Johnson, Ruthven, Reid and Calne, 1974) to form dihydroxyphenylpyruvic acid, with subsequent decarboxylation and 0 -methylation, could provide an alternate source for the formation of HVA; however, this would not explain why the levels of two other deaminated metabolites of catecholamines, DOPAC and DHPG in plasma, were not reduced to zero after administration of the MAO inhibitors. Failure to eliminate the formation of deaminated metabolites could have been due to incomplete inhibition of MAO, either because of limited efficacy of the drug, or because of rapid synthesis of new molecules of MAO (Egashira and Yamanaka, 1981; Finberg and Tal, 1985). However, in a previous study where clorgyline was administered at half the dose used in the present study, conversion of tracer-labelled norepinephrine to tracer-labelled DHPG was virtually abolished (Eisenhofer, Goldstein, Ropchak, Nguyen, Keiser and Kopin, 1988), implying that the doses of MAO inhibitors used in the present study probably did effectively suppress MAO activity. Unlike the tracer, the endogenous substrates for MAO are continuously formed in the axoplasm and would be expected to accumulate if MAO were inhibited. Thus, the failure of the levels of the deaminated metabolites in plasma to fall to less than 20% of baseline after any of the MAO inhibitors was most likely due to a buildup of cytoplasmic levels of the substrates in the presence of a markedly decreased, but still present, MAO activity.
195
Since levels of DOPAC in plasma were decreased to a similar extent after peripheral inhibition of MAO by debrisoquin as after global inhibition of MAO by nialamide, most of the DOPAC in plasma appeared to be derived from the metabolism of DA outside the central nervous system. Since clorgyline significantly decreased levels of DOPAC, whereas (-)deprenyl did not, this peripheral source probably included neuronal elements; however, deprenyl did decrease levels of DOPAC somewhat, and a previous report noted significant decreases in concentrations of DOPAC in humans treated chronically with deprenyl (Eisenhofer et al., 1986). Dopamine can be produced in non-neural cells by uptake and decarboxylation of circulating DOPA (Zimlichman, Levinson, Kelly, Stull, Keiser and Goldstein, 1988) and so it is possible that the levels of DOPAC in plasma are derived from both neuronal and non-neuronal cells. In contrast, NE is produced only in sympathetic nerve endings, the adrenal medulla and the central nervous system, and recent experiments in this laboratory have indicated that adrenal demedullation produces little if any decrease in levels of DHPG in plasma (unpublished observations). The pattern of similar decreases in the levels of DHPG in plasma after the administration of nialamide, debrisoquin and clorgyline, and the absence of an effect after (-)deprenyl, therefore supports the view that levels of DHPG in plasma are derived predominantly if not exclusively from the metabolism of NE in sympathetic nerve endings. Attempts to estimate the proportion of HVA derived from brain have been based on residual levels of HVA in plasma after inhibition of peripheral neuronal MAO by debrisoquin (Giachetti and Shore, 1967; Medina, Giachetti and Shore, 1969; Pettinger, Korn, Spiegle, Solomon, Pocelinko and Abrams, 1969), which does not cross the blood-brain barrier. After the administration of debrisoquin to animals, levels of deaminated catecholamine metabolites in plasma are reduced to a smaller extent than after the global inhibition of MAO (Bacopoulos, Hattox and Roth, 1979). Kendler, Heninger and Roth (1981) and Kendler and Davis (1984) reported reductions of 30-50% in levels of HVA in plasma after the administration of debrisoquin to rats and Edwards, Ravitch and Knopf (1985) and Edwards, Ravitch, Knopf and Sedlock (1985) estimated that 39-64% of HVA in the urine was derived from the brain. If the administration of debrisoquin only incompletely suppressed the peripheral formation of HVA, even when peripheral MAO was effectively suppressed, then the central neural contribution to HVA in plasma would be overestimated. In the present study, the repeated administration of debrisoquin produced a maximum decreased in HVA in plasma of about 50%; the residual level of HVA, about SO%, would have been thought to reflect the central neural contribution to HVA in plasma. Because the repeated administration of nialamide reduced levels of HVA
796
D. HOVEVEY-SION et d.
in plasma by only another 27% however, it was estimated that, at most, 25-30% of HVA in plasma was derived from DA in brain, in rats. Although the percentage reductions in the levels of deaminat~ metabolites probably underestimated the extent of inhibition of MAO, the amount of error of the estimates could not be determined from the present results. The estimated central neural contribution to levels of HVA in plasma could have remained valid if the increases in the cytoplasmi~ concentrations of dopamine were independent of the type of MAO inhibitor used. These considerations lead to the suggestion that the use of residual levels of HVA in plasma after the administration of debrisoquin is an improved but nevertheless imperfect method for evaluating central dopaminergic activity, especially in comparing groups of patients with psychiatric disorders. The technique may be more valid in assessing changes within individuals in response to various treatments. Levels of HVA in plasma exceeded those of the precursor, DOPAC, by over IO-fold. The large ratio of HVA to DOPAC in plasma probably reflected the rapid conversion of DOPAC to HVA, before and after DOPAC reached the circulation, as well as the relatively slow removal of HVA from plasma. The small estimated contribution of DA in brain to HVA in plasma in rats in the present study, confirmed the small estimated cont~bution in monkeys, using a different method (Kopin et al., 1988). Reductions in the formation of the metabolite of norepinephrine, MHPG, after the adminstration of debrisoquin were positively linearly related to reductions in the formation of HVA, consistent with the view that the formation of HVA in peripheral sympathetic neurons results mainly from deamination of axoplasmic DA before translocation of DA into vesicles and conversion to norepinephrine. The small extent to which the y-intercept of this relationship was above the origin, led to the suggestion that there was only a small central neural contribution to HVA in plasma. This y-intercept method is attractive in that it avoids the limitations of measurements of levels of HVA alone in a setting where the peripheral production of deaminated metabolites is not completely suppressed by debrisoquin. Nevertheless, even the y-intercept method may not be ideal, because it assumes that decreases in MHPG and HVA are due only to inteference with MAO and tyrosine hydroxylation in sympathetic neurons, Since levels of DOPA in plasma are decreased only slightly during inhibition of MAO (Eisenhofer et al., 1986), it is possible that some of the HVA in plasma can be derived from DA formed extraneuronally after the uptake of circulating DOPA. If so, then since there is no extraneuronal formation of norepinephrine and since debrisoquin inhibits neuronal MAO, the extent of debrisoquininduced inhibition of the formation of metabolites of norepinephrine could exceed the extent of inhibition
of the formation of HVA, resulting in the y-intercept value described above overestimating the central neural contribution to HVA in plasma. To summarize, the present results lead to the conclusions that most of HVA in plasma is derived from deamination of DA in peripheral neurons; that levels of DOPAC in plasma are derived from peripheral neural and non-neural cells; that levels of DHPG in plasma are derived virtually exclusively from the metabolism of norepinephrine in sympathetic nerve endings and that residual levels of HVA after treatment with debrisoquin provide an improved method to indicate central dopaminergic activity but may lead to overestimation because of the buildup of axoplasmic DA in peripheral neurons.
REFERENCES Bacopoulos N. G., Hattox S. E. and Roth R. H. (1979) 3,4-Dihydroxyphenylacetic acid and homovanillic acid in rat plasma: Possible indicators of central dopaminergic activity. Eur. J. Pharmac. 56: 225-236. Burns R. S., Chiueh C. C., Markey S. P., Ebert M. H., Jacobowitz D. M. and Kopin I. J. (1983) A primate model of parkinsonism: Selective destruction of dopaminergic neurons in the pars compacta of the substantia nigra by N-methyl-Cphenyl-1,2,3,6_tetrahydropyridine. Proc. natn. Acad. Sci. 80: 4546-4550.
Burns R. S., LeWitt P. A., Ebert M. H., Pakkenberg H. and Kooin I. J, 119851 The clinical svndrome of striatal dopami~e deficiency. ~arkinsonism induced by I-methyl4-phenyl-1,2,3,6_tetrahydropy~dine (MPTP). N. Engl. J. Med. 312: 1418-1421. Chiueh C. C., Markey S. P., Burns R. S., Johannessen J. N.. Pert A. and Kopin I. J. (1984) Neurochemical and behavioural effects of systemic and intranigral administration of N-methyl-Cphenyl-1,2,3,6_tetrahydropyridine in the rat. Eur. J. Pharmac. 100: 189-194.
Davis K. L., Davidson M., Mohs R. C., Kendfer K. S., Davis B. M., Johns C. A., DeNigris Y. and Horvath T. G. (1985) Plasma homovanillic acid concentration and the severity of schizophrenic illness. Science 227: 1601-1602. Edwards D. J., Ravitch J. and Knopf S. (1985) Effects of debrisoquin on the excretion of catecholamine and octopamine metabolites in the rat and guinea pig. &o&em. Phurmac. 34: 29 I l-29 16.
Edwards D. J.. Ravitch J.. Knoof S. and Sedlock M. L. _ -(1985) Effects of intraventricul~r injections of 6-hydroxydopamine on amine metabolites in the rat brain and urine. Eiochem. Pharmac. 34: 1255-1263.
Egashira T. and Yamanaka Y. (1981) Further studies on the synthesis of A-form of monoamine oxidase. Jap. J. P~armuc. 31: 763-770.
Eisenhofer G., Goldstein D. S., Ropchak T. G., Nguyen H. Q., Keiser H. R. and Kopin I. J. (1988) Source and physiological significance of plasma 3,4dihydroxyphenylglycol and 3-methoxy-4-hydroxyphenylglycol. J. &ton. Nerv. Sys. 24: l-14. Eisenhofer G.. Goldstein D. S.. Stull R.. Keiser H. R., Sunderland ‘T., Murphy D. L. and Kopin I. J. (1986) Simultaneous liquid~hromatographic determination of 3,4-dihyroxyphenylglycol, catecholamines, and 3,4-dihydroxyphenylanine in plasma, and their responses to inhibition of monoamine oxidase. Clin. Chem. 31: 2030-2033. Finberg J. P. M. and Tal A. (1985) Reduced peripheral presynaptic adrenoceptor sensitivity following chronic antidepressant treatment in rats. Br. J. Pharmac. 84: 609-617.
MAO inhibitors and plasma HVA Giachetti A. and Shore P. A. (1967) Monoamine oxidase inhibition in the adrenergic neuron by bretylium, debrisoquin, and other adrenergic neuronal blocking agents. Biochem. Pharmac. 16: 237-238.
Hovevey-Sion D., Haney-White J., Kopin 1. J. and Goldstein D. S. (1988) Measurement of homovani~lic acid (HVA) in small volumes of plasma using liquid chromatography with electrochemical detection. J. Chromat. 426: 141-147.
Kendler K. S. and Davis K. L. (1984) Acute and chronic effects of neuroleptic drugs on plasma and brain homovanillic acid in the rat. Psychiat. Rex. 13: 5138. Kendler K. S., Heninger G. R. and Roth R. H. (1981) Brain contribution to the halo~~doi-induced increase in plasma homovanillic acid. Eur. J. Phurmac. 71: 321-326. Kopin I. 3. (1985) Catecholamine metabolism: Basic aspects and clinical significance. Pharmac. Rev. 37: 333-364. Kopin I. J., Bankiewicz K. S. and Harvey-White J. (1988) Assessment of brain dopamine metabolism from plasma HVA and MHPG during debrisoquin treatment: Validation in monkeys treated with MPTP. N~ropsychoph~ma~ofogy 1: 119-12s. Maas J. W., Contreras S. A., Seleshi E. and Bowden C. L. (f988) Dopamine metabolism and disposition in schizophrenic patients. Studies using debrisoquin. Arch. Gen. Psychiar. 45: 553-559. Medina M. A., Giachetti A. and Shore P. A. (1969) On the physiological disposition and possible mechanism of the antihypertensive action of debrisoquin. Biochem. Pharmac. 1s: 891-901. Pettinger W. A., Korn A., Spiegle H., Solomon H. M., Pocehnko R. and Abrams W. B. (1969) Debrisoquin, a selective inhibitor of intraneuronai monoamine oxidase in man. C/in. Pharmac. Ther. 10: 667674. Riddle M. A., Leckman J. F., Cohen D. J., Anderson M., Ort S. I., Caruso K. A., Shaywitz 8. A. (1986) Assessment of central dopaminergic function using plasma-free homovanillic acid after debrisoquin a~inist~tion. J. Neural Trans. 67: 31-43.
N.P. 2SlS-C
797
Roy A., Agren H., Pickar D., Linnolia M., Doran A. R., Cutler N. R. and Paul S. M. (1986) Reduced CSF concentrations of homovanillic acid and homavanillic acid to 5hydrox~ndolindol~tic acid ratios in depressed patients: relations~p to suicidal behaviour and dexame~asone nonsuppression. Am. J. Psychiat. 143: 1539-1545. Roy-Byrne P. P., Uhde T. W., Sack D. A., Linnoila M. and Post R. M. (1986) Plasma HVA and anxiety in patients with panic disorder. Biol. Psychiat. 21: 847-849. Sack D. A.. James S. P.. Doran A. R.. Sherer M. A.. Linnoila M. and Wehr T. A. (1988) The diurnal variation &Iplasma homovanillic acid level persists but the variation in 3methoxy-4-hydroxyphenyI~yco1 level is abolished under constant conditions. Arch. gen. Psych&. 4% 162-166. Sandler M., Johnson R. D., Ruthven C. R. J., Reid J. L. and Calne D. B. (1974) Transamination is a major pathway of L-DOPA metabolism following peripheral decarboxylase inhibition. Nature 24% 364-366. Sternberg D. E., Heninger G. R. and Roth R. H. (1983) Plasma homovanillic acid as an index of brain dopamine metabolism: enhancement with debrisoquin. Life Sci. 32: 2447-2452. Waldmeier P. C. (1987) Amine oxidases and their endogenous substrates (with special reference to monoamine oxidase and the brain). J. Neural Trans. Suppl. U: 55-72. Waldmeier P. C., Delini-Stula A. and Maitre L. (1976) Preferential deamination of dopamine by an A type monoamine oxidase in rat brain. Naunyn-Schmiedebergs Arch. Pharmac. 292: 9-14.
Westerink B. H. C., Bosker F. J. and Wirix E. (1984) Formation and metabolism of dopamine in nine areas of the rat brain: Modifications by haloperidol. J. Neurochem. 42: 1321-1327.
Zimlichman R., Levinson P. D., Kelly G., Stull R., Keiser H. R. and Goldstein D. S. (1988) Derivation of urinary dopamine from plasma dihydroxyphenylalanine. Ciin. Sci. 75: 515-520
~28.3~8/89 %3.00+ 0.00 Maxwell Pergamon Macmiilan pIc
EFFECTS OF MONOAMINE OXIDASE INHIBITORS ON LEVELS OF CATECHOLS AND HOMOVAN~L~IC ACID IN STRIATUM AND PLASMA D, HOVEVEY-SION’, I. J. KOPIN’, R. W. STULL’ and D. S. GOLDSTEIN~* ‘Intramural Research Program, National Institute of Neurolo~caI Disorders and Stroke, and 2Hypertension-Endocrine Branch, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland 20892, U.S.A. (Accepted 4 February 1989)
Summary-Levels of homovanillic acid (HVA), dihydroxyphenylacetic acid (DOPAC) and dihydroxyphenyiglyco1 (DHPG) in plasma and the striatium were measured after inhibition of monoamine oxidase type A (MAO-A) by clorgyline (4 mgikg i.p.), MAO-B by (-fdeprenyl (1 mg/kg i.p.), both MAO-A and MAO-B by niatamide (75 mg/kg i.p.) or peripheral neuronal MAO by debrisoquin (40 mg/kg i.p.). Levels of HVA in plasma decreased by about 60% after single doses of nialamide or clorgyline, by about 80% after repeated doses of niaiamide, by about 40% after a single dose of debrisoquin and by about 50% after repeated doses of debrisoquin. The administration of clorgyline, niaiamide or debrisoquin significantly decreased concentrations of DOPAC and DHPG in plasma, whereas (-)deprenyl did not affect levels of DHPG or HVA. None of the MAO inhibitors produced more than about 80% depression of levels of any of the deaminated metabolites. The results suggest that most of the HVA in plasma is derived from de~ination of DA by MAO-A in peripheral neurons; that DOPAC in plasma is derived from cells outside the central nervous system; that DHPG in plasma is derived virtually exclusively from the metabolism of norepinephrine in sympathetic nerve endings and that residual levels of HVA after treatment with debrisoquin provide an improved but limited indication of central dopaminer~c activity. Kqv words-homovanillic norepinephrine
acid, monoamine oxidase, 3,4-dihydroxyphenylglycol,
Homovanillic acid (HVA), produced from the enzymatic deamination of dopamine (DA) by monoamine oxidase (MAO) to form 3,4-dihydroxyphenylacetic acid (DOPAC) and from the enzymatic 0-methylation of DOPAC by cat~hol-~-methyltransferase (COMT), is the major final metabolite of DA in humans and rats (Kopin, 1985). Two subtypes of MAO have been described, MAUA and MAO-B. The distribution of these subtypes in tissue appears to differ in that MAO-B is not prominent in nervous tissue, whereas MAO-A is present in both neural and non-neural tissue (Waldmeier, 1987). The relative contributions of these subtypes of MAO to levels of HVA in plasma have been poorly understood. In the present study, these contributions were assessed using subtype-specific and nonspecific MAO inhibitors: nialamide, which inhibits both subtypes of MAO (MAO-A and MAO-B); clorgyline, which selectively inhibits MAO-A; and (-)deprenyl, which selectively inhibits MAO-B. Alterations of levels of HVA in plasma in psychiatric disorders and in response to drugs have been used to assess changes in dopaminergic activity in the brain (Davis, Davidson, Mohs, Kendler, Davis, Johns, DeNigris and Horvath, 1985; Roy, Agren, Pickar,
*To whom correspondence
should be addressed.
DOPAC, dopamine,
Linnoila, Doran, Cutler and Paul, 1986; Roy-Byrne, Uhde, Sack, Linnoila and Post, 1986). Whereas the concentration of HVA in cerebrospinal fluid (CSF) is decreased in Parkinson’s disease (Burns, LeWitt, Ebert, Pakkenberg and Kopin, 1985), in primates that have been treated with 1-methyl-4-phenyl- 1,2,3,6tetrahydropyridine (MPTP), a neurotoxin which induces a Parkinsonian-like syndrome (Burns, Chiueh, Markey, Ebert, Jacobowitz and Kopin, 1983), the level of HVA in plasma is affected little, consistent with only a small contribution from the central nervous system (CNS) to HVA in plasma. Using decrements in methoxyhydroxyphenylglycol (MHPG) to estimate decrements in HVA derived from noradrenergic neurons after the administration of debrisoquin, an MAO inhibitor which does not penetrate the Moodbrain barrier, Kopin, Bankiewicz and Harvey-White (1988) recently estimated that only about 25% of the concentration of HVA in plasma was derived from dopaminergic neurons in the brain. Because of this substantial peripheral source of HVA in plasma, the residual level of HVA in plasma after the ad~nistration of debrisoquin, which inhibits MAO but does not appear to traverse the blood-brain barrier, has been proposed as providing a better method for estimating dopaminergic activity in the central nervous system (Maas, Contreras, Seleshi and Bowden, 1988; Riddle, Leckman, Cohen, Anderson, 791
192
D.
HOVEVEY-SIONet
Ort, Caruso and Shaywitz, 1986; Sternberg, Heninger and Roth, 1983). Accurate interpretation of the results, however, requires knowledge about the relative efficacy of debrisoquin and globally-acting MAO inhibitors in reducing the concentration of HVA in plasma. For instance, if the administration of a large dose of a globally-acting MAO inhibitor resulted in the same proportional fall in the concentration of HVA in plasma as obtained after debrisoquin, then it could be concluded that all of the HVA in plasma was derived from peripheral tissue. Therefore, in the present study, responses of levels of HVA in plasma after the administration of debrisoquin were compared with responses after the administration of the globally-acting MAO inhibitor, nialamide. Incomplete suppression of levels of HVA in plasma after the administration of debrisoquin could be interpreted as indicating insufficient efficacy of the drug as an MAO inhibitor at doses used in vivo, alternative pathways in the formation of HVA, besides through deamination of dopamine, or buildup of substrate catecholamines in the axoplasm. These explanations could have different consequences for estimates of central dopaminergic activity based on levels of HVA after treatment with debrisoquin. To evaluate these possibilities, changes in the levels in plasma of DOPAC and 3,4-dihydroxyphenylglycol (DHPG), the main deaminated metabolites of DA and norepinephrine (NE), were compared with changes in the levels of HVA in plasma. By comparing effects of globallyacting MAO inhibitors with those of debrisoquin on DHPG and DOPAC in plasma, central and peripheral sources of DHPG and DOPAC in plasma also could be identified. To verify the central or peripheral sites of action of the MAO inhibitors, levels of HVA and DOPAC in the striatum were measured in addition to levels of HVA, DOPAC, and DHPG in plasma. Since central neural inhibition of MAO would be expected to increase DA in tissue, it was determined whether levels of DA in the striatum would increase after the administration of drugs which inhibit MAO in brain. Time-response data were obtained for nialamide and clorgyline, and the effects of single and multiple doses were examined. METHODS
al.
Test procedure
Animals were transferred to clean cages at least 1 hr before the administration of drugs. At various times after treatment with drugs, the rats were anesthetized with pentobarbital (Somnifer, Richmond Veterinary Supply Co., Richmond, Virginia) and blood from the vena cava collected into heparinized tubes and the tubes put on ice. All blood samples were obtained between 12: 30 and 2: 30 p.m. in order to avoid the diurnal variation in levels of HVA in plasma (Sack, James, Doran, Sherer, Linnoila and Wehr, 1988). The animals were also deprived of food for at least 14 hr before blood sampling. The plasma was separated within 1 hr by refrigerated (4°C) centrifugation at 3000g. In one experiment involving repeated withdrawal of blood from the same conscious animals, blood was collected by tail-cut (initial tail-cut under light ether anesthesia). Plasma was stored at -70°C. In some experiments, immediately after blood sampling, the rat was decapitated and the brain removed and placed on a chilled glass plate covered with Whatman No 1 paper, soaked with ice-cold saline. The left striatum was dissected within 2min and frozen on dry ice. Samples of brain tissue were kept at -70°C until assayed. In 4 rats, a baseline blood sample was obtained by tail-cut and then debrisoquin was administered repeatedly (6 doses at 12 hr intervals); a second sample of blood was obtained 4 hr after the last dose. The same animals were then injected with 4 doses of nialamide (at 12 hr intervals) and blood was obtained by tail-cut 6 hr after the last injection. Assay methods
Dopamine, DOPAC and HVA in the striatum were assayed using high pressure liquid chromatography with electrochemical detection (LC-ED), as described by Chiueh, Markey, Burns, Johannessen, Pert and Kopin (1984), NE, DOPAC and DHPG in plasma, as described by Eisenhofer, Goldstein, Stull, Keiser, Sunderland, Murphy and Kopin (1986) and HVA in plasma, as described by Hovevey-Sion, HarveyWhite, Kopin and Goldstein (1988). The limits of detection were 10-25 pg per sample for each of these substances.
Animals
Sources of drugs and chemicals
Male Sprague-Dawley rats (Taconic Farm, Germantown, New York) weighing 200-350 g were housed with a 12 hr light/dark cycle and room temperature 24°C.
Dopamine, DOPAC, DHPG, HVA, NE and nialamide were obtained from Sigma (St Louis, Missouri), clorgyline from Research Biochemicals Incorporated (Wayland, Massachusetts), (-)deprenyl from Dr J. Knoll, Semmelweiss University of Medicine, Budapest, Hungary and debrisoquin sulfate from HoffmannLa-Roche (Nutley, New Jersey).
Administration of drugs
Drugs were administered intraperitoneally (i.p.) in a volume of 4ml/kg (8 ml/kg when nialamide was injected). Doses and schedules of administration for the various drugs are presented in the legends of the figures and table.
Statistical analyses
Results were expressed as mean values f SEM for the indicated number of animals. Statistical signifi-
MAO inhibitors and plasma HVA
793
cance was determined using Student’s t-tests or oneway analyses of variance (ANOVA) as appropriate. RESULTS
After the administration of a large, single dose of the globally-acting, subtype-nonspecific MAO inhibitor, nialamide, the levels of the deaminated DA metabolites, DOPAC and HVA in plasma and the striatum were reduced (Fig. 1). Levels of DOPAC in the striatum were reduced by about 70% at 1.5 hr and continued to decline to 13% and 10% of control at 6 and 16 hr. Similarly, levels of HVA in the striatum progressively decreased during the 16 hr after the administration of nialamide (Fig. 2). Levels of DOPAC and HVA in plasma fell more slowly than did levels of HVA in the striatum. Levels of DOPAC and HVA in plasma were not significantly decreased at 1.5 hr, but by 16 hr the HVA in plasma had decreased to about 40% of control (Fig. 2). Levels of DHPG in plasma also were decreased after the administration of nialamide (Table 1). Levels of DA in the striatum nearly doubled within 1.5 hr of the injection of nialamide, and the levels of DA were increased by 40% and 50% at 6 and 16 hr. The administration of (-)deprenyl did not significantly affect levels of DOPAC or HVA in the striatum or in plasma (Fig. I), nor levels of DHPG in plasma or levels of DA in the striatum (Table 1). In contrast, the globally-acting MAO-A inhibitor, clorgyline, significantly reduced levels of HVA, DOPAC and DHPG in plasma (Fig. 3) with the maximum decrease in HVA to about 40% of control at 6 hr. Levels of DHPG and DOPAC decreased to about l/4 to l/3 of control between 6 and 8 hr.
n
04. 0
I~,.,.,.,.,.,.,,,
2
4
6
Hours after
8
10
injection
12
14
16
18
of niolomide
Fig. 2. Levels of HVA in plasma and the striatum after the injection of nialamide. Nialamide (75 mg/kg) or saline (8 ml/kg) was injected intraperitoneally. Blood samples and striatal tissue were obtained at 1S, 6 and 16 hr after injection of drug. Results are percentages of saline-treated controls, sacrificed at the same time as the nialamide-treated animals.
After the administration of debrisoquin to inhibit peripheral neuronal MAO, levels of DOPAC and DHPG in plasma also were significantly decreased (Table 1 and Fig. 4). Levels of DOPAC and DHPG in plasma were reduced to about l/3 of control, whereas HVA was reduced to 57% of control (Fig. 4). Debrisoquin had no effect on levels of DA or DA metabolites in the striatum. By 4 hr after the last of 6 repeated doses of debrisoquin over 3 days, the level of HVA in plasma had decreased significantly to 47% of control (Fig. 5). In contrast, by 6 hr after the last of the repeated doses of nialamide in the same animals, the level of HVA in plasma had decreased to 20% of control. The additional reduction of HVA after the administration of nialamide, compared with that of debrisoquin,
NIALAMIDE
Fig. 1. Effects of MAO inhibitors on levels of DOPAC and HVA in plasma and in the striatum of rats. Saline (4 ml/kg), deprenyl (1 mg/kg), or nialamide (75 mg/kg) was injected intraperitoneally, and blood samples and striatal tissues were obtained 1.5 hr after the injection of the drugs. *P < 0.05(ANOVA) compared with saline.
D. HOVEVEY-SION et al.
194
Table I. Effects of MAO inhibitors on levels of norepinephdnc and DHPG in plasma and on levels of dopamine in the striatium 1.5 hr after the administration of drugs Treatment
rm7 0
I
1
1
4
Hours after
r 6
injection
-
s
’ 10
of clorgyline
NE in plasma (pg/ml) DHPG in plasma (pg/ml) DA in striatum (rig/g tissue)
Saline
Deprenyl (1 mg/kg)
Nialamide (75 mg/kg)
371 + 136
416 k 150
360 * 122
673 f 82
771 i 139
9475 it 259
9550 f 577
218* k 43 18,360. f 964
Treatment
NE in plasma (pg/ml) DHPG in plasma (pg/ml) DA in striatum (rig/g tissue)
Saline
Debrisoquin (40 mg/kg)
455 * 153
608 i I85
803 + 130 11,810~680
268* + 56 13,022 i 555
‘Different from saline, P < 0.05. NE = norepinephrine; DHPG = dihydroxyphenylglycol; DA = dopamine. Upper data analyzed by ANOVA, lower by r-test.
0
2
Hours after
4
injection
6
I)
10
of clorgyline
Fig. 3. Time course of inhibition of levels of HVA, DOPAC and DHPG in plasma by clorgyline. Upper panel, HVA in plasma, lower panel, DOPAC, and DHPG in plasma. Inset results are shown as percentages of control. Clorgyline (4 mg/kg) or saline (4 ml/kg) was injected intraperitoneally, with blood samples collected at 4, 6, and 8 hr after the injection.
led to the estimate that the central nervous system contributed by at most 27% to HVA in plasma. DISCUSSION
Nialamide, which inhibits MAO-A and MAO-B and clorgyline, which inhibits MAO-A, both cross
the blood-brain barrier and therefore inhibit MAO in the brain and periphery. Nialamide and clorgyline produced marked decreases of similar magnitude in the levels of the deaminated metabolites of catecholamines in plasma. In particular, levels of HVA in plasma decreased to about 20% of control. Consistent with previous observations (Westerink, Bosker and Wirix, 1984), levels of dopamine in the striatum were approximately doubled after inhibition of MAO-A, whereas inhibition of MAO-B with (-)deprenyl did not alter levels of the deaminated catecholamine metabolites in plasma or in the striatum. Therefore, MAO-B appears to be unimportant in the metabolism of dopamine in rats (Waldmeier, Delini-Stula and Maitre, 1976; Waldmeier, 1987).
Fig. 4. Effects of debrisoquin on levels of DOPAC and HVA in plasma and the striatum. Saline (4 ml/kg) or debrisoquin (DEBRISO, 40 mg/kg) was injected intraperitoneally, and blood samples and striatal tissue were obtained 4 hr after injection of drug. Data analyzed by ANOVA.
MAO inhibitors and plasma HVA
Fig. 5. Effects of repeated doses of debrisoquin, followed by repeated doses of nialamide, on levels of HVA in plasma. Pre-treatment blood samples were withdrawn by tail-cut. Debrisoquin (DEBRISO, 40 mg/kg, i.p.) was injected 6 times every 12 hr for 3 days. Four hr after the sixth injection of debiisoquin, blood was collected by tail-cut. Animals were then treated with 4 doses of nialamide (NIAL, 75 mg/kg, i.p.), 12 hr between doses. Six hr after the 4th injection of nialamide, a final blood sample was obtained. Blood samples were collected between 12: 30 and 2 : 30 p.m.
Even after repeated administration of effective doses of the MAO inhibitors, levels of HVA in plasma were not reduced to zero; maximum reductions of HVA in plasma were to about 20% of control. This could be explained by alternative routes of formation of HVA, other than through oxidative deamination of DA, by incomplete inhibition of MAO, or by buildup of cytoplasmic levels of catecholamines. Transamination of 3,4_dihydroxyphenylalanine (DOPA) (Sandler, Johnson, Ruthven, Reid and Calne, 1974) to form dihydroxyphenylpyruvic acid, with subsequent decarboxylation and 0 -methylation, could provide an alternate source for the formation of HVA; however, this would not explain why the levels of two other deaminated metabolites of catecholamines, DOPAC and DHPG in plasma, were not reduced to zero after administration of the MAO inhibitors. Failure to eliminate the formation of deaminated metabolites could have been due to incomplete inhibition of MAO, either because of limited efficacy of the drug, or because of rapid synthesis of new molecules of MAO (Egashira and Yamanaka, 1981; Finberg and Tal, 1985). However, in a previous study where clorgyline was administered at half the dose used in the present study, conversion of tracer-labelled norepinephrine to tracer-labelled DHPG was virtually abolished (Eisenhofer, Goldstein, Ropchak, Nguyen, Keiser and Kopin, 1988), implying that the doses of MAO inhibitors used in the present study probably did effectively suppress MAO activity. Unlike the tracer, the endogenous substrates for MAO are continuously formed in the axoplasm and would be expected to accumulate if MAO were inhibited. Thus, the failure of the levels of the deaminated metabolites in plasma to fall to less than 20% of baseline after any of the MAO inhibitors was most likely due to a buildup of cytoplasmic levels of the substrates in the presence of a markedly decreased, but still present, MAO activity.
195
Since levels of DOPAC in plasma were decreased to a similar extent after peripheral inhibition of MAO by debrisoquin as after global inhibition of MAO by nialamide, most of the DOPAC in plasma appeared to be derived from the metabolism of DA outside the central nervous system. Since clorgyline significantly decreased levels of DOPAC, whereas (-)deprenyl did not, this peripheral source probably included neuronal elements; however, deprenyl did decrease levels of DOPAC somewhat, and a previous report noted significant decreases in concentrations of DOPAC in humans treated chronically with deprenyl (Eisenhofer et al., 1986). Dopamine can be produced in non-neural cells by uptake and decarboxylation of circulating DOPA (Zimlichman, Levinson, Kelly, Stull, Keiser and Goldstein, 1988) and so it is possible that the levels of DOPAC in plasma are derived from both neuronal and non-neuronal cells. In contrast, NE is produced only in sympathetic nerve endings, the adrenal medulla and the central nervous system, and recent experiments in this laboratory have indicated that adrenal demedullation produces little if any decrease in levels of DHPG in plasma (unpublished observations). The pattern of similar decreases in the levels of DHPG in plasma after the administration of nialamide, debrisoquin and clorgyline, and the absence of an effect after (-)deprenyl, therefore supports the view that levels of DHPG in plasma are derived predominantly if not exclusively from the metabolism of NE in sympathetic nerve endings. Attempts to estimate the proportion of HVA derived from brain have been based on residual levels of HVA in plasma after inhibition of peripheral neuronal MAO by debrisoquin (Giachetti and Shore, 1967; Medina, Giachetti and Shore, 1969; Pettinger, Korn, Spiegle, Solomon, Pocelinko and Abrams, 1969), which does not cross the blood-brain barrier. After the administration of debrisoquin to animals, levels of deaminated catecholamine metabolites in plasma are reduced to a smaller extent than after the global inhibition of MAO (Bacopoulos, Hattox and Roth, 1979). Kendler, Heninger and Roth (1981) and Kendler and Davis (1984) reported reductions of 30-50% in levels of HVA in plasma after the administration of debrisoquin to rats and Edwards, Ravitch and Knopf (1985) and Edwards, Ravitch, Knopf and Sedlock (1985) estimated that 39-64% of HVA in the urine was derived from the brain. If the administration of debrisoquin only incompletely suppressed the peripheral formation of HVA, even when peripheral MAO was effectively suppressed, then the central neural contribution to HVA in plasma would be overestimated. In the present study, the repeated administration of debrisoquin produced a maximum decreased in HVA in plasma of about 50%; the residual level of HVA, about SO%, would have been thought to reflect the central neural contribution to HVA in plasma. Because the repeated administration of nialamide reduced levels of HVA
796
D. HOVEVEY-SION et d.
in plasma by only another 27% however, it was estimated that, at most, 25-30% of HVA in plasma was derived from DA in brain, in rats. Although the percentage reductions in the levels of deaminat~ metabolites probably underestimated the extent of inhibition of MAO, the amount of error of the estimates could not be determined from the present results. The estimated central neural contribution to levels of HVA in plasma could have remained valid if the increases in the cytoplasmi~ concentrations of dopamine were independent of the type of MAO inhibitor used. These considerations lead to the suggestion that the use of residual levels of HVA in plasma after the administration of debrisoquin is an improved but nevertheless imperfect method for evaluating central dopaminergic activity, especially in comparing groups of patients with psychiatric disorders. The technique may be more valid in assessing changes within individuals in response to various treatments. Levels of HVA in plasma exceeded those of the precursor, DOPAC, by over IO-fold. The large ratio of HVA to DOPAC in plasma probably reflected the rapid conversion of DOPAC to HVA, before and after DOPAC reached the circulation, as well as the relatively slow removal of HVA from plasma. The small estimated contribution of DA in brain to HVA in plasma in rats in the present study, confirmed the small estimated cont~bution in monkeys, using a different method (Kopin et al., 1988). Reductions in the formation of the metabolite of norepinephrine, MHPG, after the adminstration of debrisoquin were positively linearly related to reductions in the formation of HVA, consistent with the view that the formation of HVA in peripheral sympathetic neurons results mainly from deamination of axoplasmic DA before translocation of DA into vesicles and conversion to norepinephrine. The small extent to which the y-intercept of this relationship was above the origin, led to the suggestion that there was only a small central neural contribution to HVA in plasma. This y-intercept method is attractive in that it avoids the limitations of measurements of levels of HVA alone in a setting where the peripheral production of deaminated metabolites is not completely suppressed by debrisoquin. Nevertheless, even the y-intercept method may not be ideal, because it assumes that decreases in MHPG and HVA are due only to inteference with MAO and tyrosine hydroxylation in sympathetic neurons, Since levels of DOPA in plasma are decreased only slightly during inhibition of MAO (Eisenhofer et al., 1986), it is possible that some of the HVA in plasma can be derived from DA formed extraneuronally after the uptake of circulating DOPA. If so, then since there is no extraneuronal formation of norepinephrine and since debrisoquin inhibits neuronal MAO, the extent of debrisoquininduced inhibition of the formation of metabolites of norepinephrine could exceed the extent of inhibition
of the formation of HVA, resulting in the y-intercept value described above overestimating the central neural contribution to HVA in plasma. To summarize, the present results lead to the conclusions that most of HVA in plasma is derived from deamination of DA in peripheral neurons; that levels of DOPAC in plasma are derived from peripheral neural and non-neural cells; that levels of DHPG in plasma are derived virtually exclusively from the metabolism of norepinephrine in sympathetic nerve endings and that residual levels of HVA after treatment with debrisoquin provide an improved method to indicate central dopaminergic activity but may lead to overestimation because of the buildup of axoplasmic DA in peripheral neurons.
REFERENCES Bacopoulos N. G., Hattox S. E. and Roth R. H. (1979) 3,4-Dihydroxyphenylacetic acid and homovanillic acid in rat plasma: Possible indicators of central dopaminergic activity. Eur. J. Pharmac. 56: 225-236. Burns R. S., Chiueh C. C., Markey S. P., Ebert M. H., Jacobowitz D. M. and Kopin I. J. (1983) A primate model of parkinsonism: Selective destruction of dopaminergic neurons in the pars compacta of the substantia nigra by N-methyl-Cphenyl-1,2,3,6_tetrahydropyridine. Proc. natn. Acad. Sci. 80: 4546-4550.
Burns R. S., LeWitt P. A., Ebert M. H., Pakkenberg H. and Kooin I. J, 119851 The clinical svndrome of striatal dopami~e deficiency. ~arkinsonism induced by I-methyl4-phenyl-1,2,3,6_tetrahydropy~dine (MPTP). N. Engl. J. Med. 312: 1418-1421. Chiueh C. C., Markey S. P., Burns R. S., Johannessen J. N.. Pert A. and Kopin I. J. (1984) Neurochemical and behavioural effects of systemic and intranigral administration of N-methyl-Cphenyl-1,2,3,6_tetrahydropyridine in the rat. Eur. J. Pharmac. 100: 189-194.
Davis K. L., Davidson M., Mohs R. C., Kendfer K. S., Davis B. M., Johns C. A., DeNigris Y. and Horvath T. G. (1985) Plasma homovanillic acid concentration and the severity of schizophrenic illness. Science 227: 1601-1602. Edwards D. J., Ravitch J. and Knopf S. (1985) Effects of debrisoquin on the excretion of catecholamine and octopamine metabolites in the rat and guinea pig. &o&em. Phurmac. 34: 29 I l-29 16.
Edwards D. J.. Ravitch J.. Knoof S. and Sedlock M. L. _ -(1985) Effects of intraventricul~r injections of 6-hydroxydopamine on amine metabolites in the rat brain and urine. Eiochem. Pharmac. 34: 1255-1263.
Egashira T. and Yamanaka Y. (1981) Further studies on the synthesis of A-form of monoamine oxidase. Jap. J. P~armuc. 31: 763-770.
Eisenhofer G., Goldstein D. S., Ropchak T. G., Nguyen H. Q., Keiser H. R. and Kopin I. J. (1988) Source and physiological significance of plasma 3,4dihydroxyphenylglycol and 3-methoxy-4-hydroxyphenylglycol. J. &ton. Nerv. Sys. 24: l-14. Eisenhofer G.. Goldstein D. S.. Stull R.. Keiser H. R., Sunderland ‘T., Murphy D. L. and Kopin I. J. (1986) Simultaneous liquid~hromatographic determination of 3,4-dihyroxyphenylglycol, catecholamines, and 3,4-dihydroxyphenylanine in plasma, and their responses to inhibition of monoamine oxidase. Clin. Chem. 31: 2030-2033. Finberg J. P. M. and Tal A. (1985) Reduced peripheral presynaptic adrenoceptor sensitivity following chronic antidepressant treatment in rats. Br. J. Pharmac. 84: 609-617.
MAO inhibitors and plasma HVA Giachetti A. and Shore P. A. (1967) Monoamine oxidase inhibition in the adrenergic neuron by bretylium, debrisoquin, and other adrenergic neuronal blocking agents. Biochem. Pharmac. 16: 237-238.
Hovevey-Sion D., Haney-White J., Kopin 1. J. and Goldstein D. S. (1988) Measurement of homovani~lic acid (HVA) in small volumes of plasma using liquid chromatography with electrochemical detection. J. Chromat. 426: 141-147.
Kendler K. S. and Davis K. L. (1984) Acute and chronic effects of neuroleptic drugs on plasma and brain homovanillic acid in the rat. Psychiat. Rex. 13: 5138. Kendler K. S., Heninger G. R. and Roth R. H. (1981) Brain contribution to the halo~~doi-induced increase in plasma homovanillic acid. Eur. J. Phurmac. 71: 321-326. Kopin I. 3. (1985) Catecholamine metabolism: Basic aspects and clinical significance. Pharmac. Rev. 37: 333-364. Kopin I. J., Bankiewicz K. S. and Harvey-White J. (1988) Assessment of brain dopamine metabolism from plasma HVA and MHPG during debrisoquin treatment: Validation in monkeys treated with MPTP. N~ropsychoph~ma~ofogy 1: 119-12s. Maas J. W., Contreras S. A., Seleshi E. and Bowden C. L. (f988) Dopamine metabolism and disposition in schizophrenic patients. Studies using debrisoquin. Arch. Gen. Psychiar. 45: 553-559. Medina M. A., Giachetti A. and Shore P. A. (1969) On the physiological disposition and possible mechanism of the antihypertensive action of debrisoquin. Biochem. Pharmac. 1s: 891-901. Pettinger W. A., Korn A., Spiegle H., Solomon H. M., Pocehnko R. and Abrams W. B. (1969) Debrisoquin, a selective inhibitor of intraneuronai monoamine oxidase in man. C/in. Pharmac. Ther. 10: 667674. Riddle M. A., Leckman J. F., Cohen D. J., Anderson M., Ort S. I., Caruso K. A., Shaywitz 8. A. (1986) Assessment of central dopaminergic function using plasma-free homovanillic acid after debrisoquin a~inist~tion. J. Neural Trans. 67: 31-43.
N.P. 2SlS-C
797
Roy A., Agren H., Pickar D., Linnolia M., Doran A. R., Cutler N. R. and Paul S. M. (1986) Reduced CSF concentrations of homovanillic acid and homavanillic acid to 5hydrox~ndolindol~tic acid ratios in depressed patients: relations~p to suicidal behaviour and dexame~asone nonsuppression. Am. J. Psychiat. 143: 1539-1545. Roy-Byrne P. P., Uhde T. W., Sack D. A., Linnoila M. and Post R. M. (1986) Plasma HVA and anxiety in patients with panic disorder. Biol. Psychiat. 21: 847-849. Sack D. A.. James S. P.. Doran A. R.. Sherer M. A.. Linnoila M. and Wehr T. A. (1988) The diurnal variation &Iplasma homovanillic acid level persists but the variation in 3methoxy-4-hydroxyphenyI~yco1 level is abolished under constant conditions. Arch. gen. Psych&. 4% 162-166. Sandler M., Johnson R. D., Ruthven C. R. J., Reid J. L. and Calne D. B. (1974) Transamination is a major pathway of L-DOPA metabolism following peripheral decarboxylase inhibition. Nature 24% 364-366. Sternberg D. E., Heninger G. R. and Roth R. H. (1983) Plasma homovanillic acid as an index of brain dopamine metabolism: enhancement with debrisoquin. Life Sci. 32: 2447-2452. Waldmeier P. C. (1987) Amine oxidases and their endogenous substrates (with special reference to monoamine oxidase and the brain). J. Neural Trans. Suppl. U: 55-72. Waldmeier P. C., Delini-Stula A. and Maitre L. (1976) Preferential deamination of dopamine by an A type monoamine oxidase in rat brain. Naunyn-Schmiedebergs Arch. Pharmac. 292: 9-14.
Westerink B. H. C., Bosker F. J. and Wirix E. (1984) Formation and metabolism of dopamine in nine areas of the rat brain: Modifications by haloperidol. J. Neurochem. 42: 1321-1327.
Zimlichman R., Levinson P. D., Kelly G., Stull R., Keiser H. R. and Goldstein D. S. (1988) Derivation of urinary dopamine from plasma dihydroxyphenylalanine. Ciin. Sci. 75: 515-520