- Email: [email protected]
Acute and chronic effects of neuroleptic drugs on plasma and brain homovanilic acid in the rat
P.s>,chiatr>, Rr.vrarch. 13, 5 I-58
51
Elsevier
Acute and Chronic Effects of Neuroleptic Drugs on Plasma and Brain Homovanillic Acid in the Rat Kenneth
S. Kendler
Received Jaman,
and Kenneth
L. Davis
9, 1984; revised version received June 5, 1984; accepred July 24, 1984.
Abstract. In the rat. the acute administration of the neuroleptic drug, haloperidol. produces a parallel increase in brain and plasma concentrations of the dopamine metabolite. homovanillic acid (HVA). The effect of other neuroleptic drugs, which may differ from haloperidol in their central and peripheral actions, on brain and plasma HVA has not been systematically investigated. Therefore, in this report we examine the acute effects of six different neuroleptic drugs. representing most major chemical classes of these drugs, on plasma and brain concentrations of HVA. Metoclopramide, fluphenazine, loxapine, and molindone produce parallel increases in brain and plasma HVA which closely resemble those produced by haloperidol. Compared to these neuroleptics. chlorpromazine produces a much greater increase in plasma HVA but a similar effect on brain HVA. The large chlorpromazine-induced increase in plasma HVA suggests that this drug alters peripheral production or clearance of HVA, perhaps via blockade of peripheral a-receptors. Of the available neuroleptics. sulpiride is one of the most specific and potent at blocking the dopamine vascular receptor. Administration of high doses of sulpiride produces only modest increases in both plasma and brain HVA, suggesting that blockade of peripheral dopamine receptors does not substantially alter peripheral clearance of HVA. After chronic administration of haloperidol for up to 21 days. plasma HVA continued to reflect the brain HVA response to drug administration. Key Words.
Plasma homovanillic
acid, neuroleptics.
In the rat, acute administration of the neuroleptic drug, haloperidol, produces a parallel increase in the concentration of the dopamine metabolite. homovanillic acid (HVA), in brain and plasma (Bacopoulos et al., 1979). This haloperidol-induced increase in plasma HVA appears to result from alterations in brain HVA production and not from changes in the peripheral production or clearance of HVA (Kendler et al., 1981). Haloperidol administration also increases plasma HVA in the monkey (Bacopoulos et al., 1978) and possibly in man (Heninger et al., 1979). Based on these and other studies, plasma HVA is now used in clinical settings as a possible measure of the response of brain dopamine systems to a variety of stimuli including neuroleptic
Kenneth 5 Kendler. M.D.. was t’ormerly Assistant Professor of Psychiatry, Mount Sinai School of Medicine. and is now in the Departments of Psychiatry and Human Genetics. Medical College of Virginia, Rxhmond. VA. Kenneth L. Davis. M.D.. is Professor of Psychiatry and Pharmacology, Mount Sinai School of Medicine, and Chief of Psychiatry. Bronx Veterans Administration Medical Center. (Reprint requests to Dr. K.L. Davis. Psychiatry Service( I 16A). Bronx VA Medical Center. I30 W. Kingsbridge Rd., Bronx. XY 1046X. USA.) 0165-1781
X4 $03.00 % 1984 Elsevier Science Publishers B.V
52 drugs (Bowers et al.. 1980; Naber et al.. 1981; Cutler et al., 1982). The acute effects on plasma HVA of neuroleptics other than haloperidol has not, to our knowledge. been investigated. Neuroleptics differ considerably in their pharmacologic properties both in the brain (Peroutka and Snyder, 1980) and the periphery (Goldberg et al.. 1978). For example. some neuroleptic drugs are considerably more potent than haloperidol at blocking a-adrenergic receptors (Peroutka and Snyder, 1980). The hypotension that often accompanies cu-adrenergic blockade could alter the peripheral production or clearance of HVA. Furthermore. some neuroleptics are more potent than haloperidol at blocking dopamine receptors in the vascular tree. Since administration of at least one such neuroleptic, sulpiride (Chapman et al., 1980). has been shown to alter renal blood flow, it is possible that blockade of renal vascular dopamine receptors could alter HVA clearance. Therefore. the acute effect on plasma HVA of certain neuroleptics might differ considerably from that produced by haloperidol. The chronic effects of neuroleptic drugs on brain dopamine systems differ from their acute effects (Julou et al., 1977). In the monkey, Bacopoulos et al. (1980) demonstrated only a partial correlation between plasma and brain HVA concentrations after chronic administration of fluphenazine and haloperidol. This study was limited because brain samples could be obtained at only one time point (3 weeks) after chronic drug administration. In this report. the acute effects of six different neuroleptic drugs. representing most of the major chemical classes of these compounds. on brain and plasma HVA in the rat were investigated. These drugs include a neuroleptic with a-adrenergic blocking potency, chlorpromazine. and one that is potent at blocking dopamine vascular receptors. sulpiride. Furthermore, the response of plasma and brain HVA to chronic haloperidol administration in the rat was also examined. The results obtained were of potential relevance to the clinical use of plasma HVA in man. However. although similarities exist in the dopamine system of man and rat. sufficient differences have been found to indicate caution in the extrapolation of results from one species to another. Methods Male. Sprague-Dawley rats. habituated to the laboratory environment for at least 7 days and weighing from 260 to 400 g. were used in all experiments. Animals were maintained on a IO-hour light. l4-hour dark schedule with food and water available ad lib. For 24 hours before sacrifice. animals were fasted with water freely available. All drugs were administered intraperitoneally in less than I ml of vehicle. Debrisoqutn was administered 4 hours before sacrifice. a time interval previously shown to maximire the debrisoquin-produced decrement in plasma HVA (Kendler et al., 1981). Metoclopramide uas administered 60 minutes. sulpiride 90 minutes. and all other neuroleptics 120 minutes before sacrifice. Doses of neuroleptics were chosen to be roughI!, equivalent. from a clinical perspective. to I to 2 mg kg of haloperidol. Molindone. however. because of evidence that at high doses it demonstrates dopamine agonist properties (Juorio. 1980). was administered at a dose of 2.5 mg kg. Debrisoquin sulfate (Hoffmann-La Roche. Nutley. NJ). metoclopramide (Robins Research I.aboratories. Richmond. VA). and molindone (Endo Laboratories. Garden City. NY) Mere dissolved in bacterioatatic water. Sulpiride (Delagrange International. Paris) was dissolved in bacteriostatic water titrated to /‘H 2.X with I 12’ HCI. Fluphenaline hydrochloride injectable
53 (Squibb. Princeton, NJ), loxapine hydrochloride injectable(Lederle Laboratories, Wayne, NJ), chlorpromazine hydrochloride injectable (Smith, Kline & French Laboratories, Philadelphia. PA). and haloperidol injectable (McNeil Laboratories, Fort Washington, PA) were diluted in bacteriostatic water. Blood samples of 5-7 ml were collected by open chest cardiac puncture after brief ether anesthesia into a heparinized syringe. After open chest cardiac puncture, the rats immediately were decapitated and the brains rapidly removed. In some studies, the whole brain was then immediately frozen on dry ice, while in others the striata were bilaterally dissected out and then frozen on dry ice. In the experiment examining the effects of chronic neuroleptic administration, haloperidol, I mg: kg, was administered daily for I. 4. 7, or 2 I days. On the final experimental day, haloperidol was administered 2 hours before sacrifice. The olfactory tubercles were assayed for HVA in this study because the pattern of habituation of HVA to chronic neuroleptic administration in this region is intermediate between that found in the striatum and cortex (Julou et al.. 1977). For all studies. blood samples were prepared and free plasma HVA was assayed by gaschromatographic mass-spectrometric quantitation as previously described (Bacopoulos et al.. 1979). Sufficient plasma was available for nearly all samples to be assayed for free HVA in duplicate, with the mean value for the two samples then being used in the statistical analysis. Free HVA in brain was assayed by high performance liquid chromatography with electrochemical detection using homovanillic alcohol as an internal standard. Statistical analysis was carried out by means of the Students t test, Pearson’s productmoment correlation. and one and two-way analyses of variance (ANOVAs). When needed, a post-hoc test after a significant one-way ANOVA was performed by the Student-NewmanKeuls procedure with ranges set at the 0.05 level. All p values are reported two-tailed; p values greater than 0.05 are not significant (NS).
Results Acute Administration of Neuroleptics. In the first experiment, the effect of metoclopramide on plasma and whole brain HVA was examined with and without pretreatment with debrisoquin, a drug that inhibits peripheral but not central HVA production (Kendler et al., 1981). Debrisoquin produced a significant decrement in plasma HVA (F= II.78; dj’= I, 22;~ q0.002), but no significant effect on whole brain HVA (Fz0.07: c/j'= I,22; NS) (Table 1). Metoclopramide significantly elevated both plasma HVA (F= 9.54; rlf= 1,22;p = 0.005) and whole brain HVA (F= 36.5 I; elf‘= I, 22; p < 0.00I).No significant interaction was seen between the effects of debrisoquin and metoclopramide treatment on either plasma or brain HVA. Following vehicle pretreatment, metoclopramide increased plasma HVA 33%. Table 1. Effect of metoclopramide on plasma and whole brain HVA with and without debrisoquin pretreatment Plasma free HVA (ng/ml) Pretreatment
Treatment
Vehicle
3.6 + 0.5
4.6 i
1.1
Debrisoauin
2.9 2 0.5
3.5 -c 0.5
Whole brain free HVA (ng/g) Pretreatment 226k
71
327 + 132
Treatment 656k317 601 t 331
Debnsoquln 40 mg/kg wasadmlnistered 4 hours before and metoclopramide 120 mg/kgl 1 hour before sacnfwx n = 6 to 7 per group;. Statistical analysis by two-way ANOVA. For plasma HVA, effects of both pretreatment F= 11.76. df= 1. 22; p =0.0021 and treatment are significant lF=9.54; df= 1, 22; p=O.O05j, but their Interaction ~snot fF=O.96; df= 1,22;p=NS1. Forwhole brain HVA,onlythetreatmenteffect issignificant F = 36 51. df = 1. 22;~ < 0 0001 8.Neitherthepretreatmenteffect if =0 07;df= 1.22; NSI northepretreatment X treatment Interaction was significant F = 0 74; df = 1. 22: NS). Results are expressed as means + SD.
54 The acute effects of other neuroleptics on plasma and brain HVA concentrations were examined without debrisoquin pretreatment. With a repeated sampling procedure, fluphenazine produced a significantly greater change in plasma HVA from basal values than seen with vehicle administration (Table 2) (t 2.80, ulf’ = 13, p = 0.0 15). Fluphenazine also significantly increased whole brain HVA (t = 6.04, Q’= 13, p 0.00004). q
q
Table 2. Acute administration
plasma
and whole
brain
HVA responses
to fluphenazine
Treatment
Change in free plasma HVA (ng/ml)
Whole brain free HVA (ng/g)
Vehicle Fluphenazine t idf= 131
0.7 + 0.6 2.6 i 1.6 2.80 0.015
2492 71 932 k 310 6.04 0.0001
P
Animals were etherized and 2.5 cc of blood obtained by closed chest cardiac puncture for basal plasma HVA values. Animals were then administered vehicle or fluphenazlne. 2.5 mg/kg. and 2 hours laterwere sacrificed The change in plasma HVA represents the HVA concentration obtalned at sacrifice minus the basal HVA concentrat!on. Basal concentrations of plasma HVA did not differ in the 2 groups lvehlcle. 2 5 + 0.1 ng/cc; fluphenazine: 2.3 i 0.2 rig/ccc ‘t = 1.33, NS, ‘n = 7-8 per group 8.StatIstical analysis by Student’s t test Results are expressed as means + SD
In the next study, the acute effects of sulpiride, loxapine, and molindone on plasma brain HVA were compared with those found after vehicle administration (Table 3). A one-way ANOVA indicated that both plasma HVA (F = 5. I ; df = 3, 30: I_’= 0.006) and whole brain HVA (F = 123.6; # = 3, 30; p < 0.0001) differed significantly across the four treatment groups. By a Student-Newman-Keuls procedure, loxapine and molindone, but not sulpiride, significantly increased plasma and whole brain HVA. Loxapine produced a 35qo and molindone a 370,;#increase in plasma HVA. Compared to vehicle administration, chlorpromazine produced a highly significant 318% increase in plasma HVA (t = 5.30; (l/‘= 13; p = 0.0001) and a 267%, increase in striatal HVA (t = 7.05; (if= 13. p 0.00001) (Table 3). To investigate further this very large increase in plasma HVA produced by chlorpromazine, the effect of this drug on plasma and whole brain HVA was compared with that of haloperidol (Table 3). Although the increases in whole brain HVA produced by the two neuroleptics were similar. chlorpromazine produced a much greater increase in plasma HVA than did haloperidol. and whole
q
Chronic Administration of Neuroleptics. The effect of up to 21 days of daily haloperidol administration on plasma and brain HVA is seen in Table 4. The greatest increase in HVA in both plasma and olfactory tubercle was produced after a single injection of haloperidol. After repeated administration of the neuroleptic. a modest habituation in the elevation of HVA was seen in both plasma and brain. Over the five time points measured in this study, the correlation between the mean concentration of HVA in plasma and the olfactory tubercle was 0.96 @ < 0.01).
55 Table 3. The acute response of plasma and brain HVA to sulpiride, molindone, chlorpromazine and haloperidol Plasma free HVA
Whole brain free HVA
(w/ml)
(ng44)
4.3 It 0.0
1932
Sulpiride
4.5 * 0.8
265 ? 51
Loxapine
5.8 i 1.1
789 ? 88
Molindone
5.9 ? 1.4
580 -t 99
5.101
123.642
3, 27)
Striatal free HVA
(w/g)
Vehicle
F idf=
loxapine,
13
Chlorpromazine
16.7 t 6.2
3,450 Itr 724
Vehicle
4.0 + 0.6
1,294 + 410
tldf=131
5.302
7.052
Chlorpromazine
13.7 2 5.6
Haloperidol
4.2 ? 0.8
Vehicle
3.6 -t 1.3
/=(df=2.171
605 i 98 657?
111
196t50
19.302
51.802
Loxapine, 5 mg/kg, mollndone, 2.5 mg/kg. chlorpromazine, 25 mg/kg. and haloperidol. 2 mg/kg. were administered intraperitoneally (1.p I 120minutesbeforesacrifice Sulpiride, lOOmg/kg,wasadministeredi.p.90 minutesbeforesacrlficeln=7-8pergroupI. Statisticalanalysiswasbyone-wayANOVAandStudent’sttest. For experiment with vehicle. sulpirtde, loxapine. and molindone. Student-Newman-Keuls procedure divided the plasma WA reponses Into 2 groups: i 1 / vehicle and sulplride and 121 loxapine and molindone. The brain HVA responses were divided Into 3 groups: 11 I vehicle and sulpiride. 121 molindone, and (31 loxaplne. Results are expressed as means +- SD. 1. p
Table 4. Effect of chronic haloperidol the olfactory tubercle Treatment Vehicle
administration
on HVA in plasma and in
Free plasma HVA
Free HVA in olfactory tubercle
(ng/mU
(ngfg)
2.5 ? 0.5
1,011 ? 442
Haloperidol
X 1 day
3.8 + 1.9
2,313 + 992
Haloperidol
X 4 days
3.4 + 0.8
1,895 i- 691
Haloperidol
X 7 days
3.4 2 1.3
1,630 2 683
Haloperidol
X 21 days
3.4 * 0.5
1.7482812
F (df = 4,341 D
9.06
2.82
0.0001
0.04
Haloperldol, 1 ng/kg. was administered daily for prescribed period of timeand 2 hours beforesacrifice group1 StatIstical analysis is by one-way ANOVA. Results are expressed as means i SD
in = 7 per
Discussion Although representing different chemical classes of neuroleptics, metoclopramide. fluphenazine, loxapine, and molindone have acute effects on brain and plasma HVA similar to those previously found with haloperidol (Kendler et al., 198 I). In addition. the rise in plasma HVA produced by metoclopramide after debrisoquin suggests that the effect of this drug on plasma HVA. like that of haloperidol, is not mediated by changes in peripheral HVA production (Kendler et al.. 1981). Debrisoquin inhibits
production of HVA outside of the brain. Thus, if the metoclopramide-induced rise in plasma HVA resulted from increased peripheral HVA production, debrisoquin pretreatment would be expected to eliminate this rise. This, however, was not observed (Table 1). In the dose administered (25 mg: kg), chlorpromazine produced a much greater rise in plasma HVA than seen with any other neuroleptic. Compared to haloperidol, chlorpromazine produced a much greater rise in plasma HVA but a similar rise in whole brain HVA. These results suggest that the large increase in plasma HVA seen after chlorpromazine results from peripheral and not central actions of this drug. Chlorpromazine could increase peripheral production or diminish peripheral clearance of HVA by several mechanisms. The ratio of potency of chlorpromazine at blocking a-adrenergic to dopaminergic receptors is one of the highest found for neuroleptics (Peroutka and Snyder, 1980). Since blockade of a-adrenergic receptors causes hypotension. which could increase peripheral production of HVA or reduce its clearance, the peripheral a-adrenergic blocking action of chlorpromazine is one plausible mechanism to explain its marked effect on plasma HVA concentrations. Sulpiride is among the most potent and specific antagonists of the dopamine vascular receptor (Goldberg et al.. 1978). Administered alone, sulpiride increases arterial resistance in the renal cortex (Chapman et al., 1980). It is therefore possible that sulpiride administration might reduce renal clearance of HVA. However, because of slow and incomplete penetrance into the brain (Nishibe et al.. 1982), sulpiride. when administered at a high dose, produced only a modest increase in whole brain HVA and no substantial increase in plasma HVA. These results suggest that blockade of peripheral vascular dopamine receptors does not substantially diminish clearance of HVA from plasma. This and other studies (Bacopoulos et al.. 1979; Kendler et al.. 1981) suggest that after acute administration of most neuroleptics, the change in plasma HVA concentrations provides a reasonably accurate measure of changes in brain HVA production. However, in clinical settings, neuroleptics are most commonly administered chronically. Chronic neuroleptic administration produces a different response in the brain dopamine system than does acute administration (Julou et al., 1977). Therefore, it was of interest to examine further the response of brain and plasma HVA to chronic neuroleptic administration. Throughout a 21-day course of daily haloperidol administration. the changes in plasma HVA closely paralleled the changes in HVA in the olfactory tubercle, in which the response of HVA to chronic neuroleptic administration is intermediate between that found in the striatum and cortex. In accord with previous findings in the monkey (Bacopoulos et al., 1980), these findings raise the possibility that plasma HVA may continue to reflect changes in brain HVA during chronic neuroleptic administration, particularly in the forebrain dopamine systems where neuroleptic drugs probably produce their antipsychotic effect. The relationship between changes in brain and plasma HVA during chronic neuroleptic administration would be further clarified by a study examining HVA levels in multiple brain areas as well as plasma. Although results from the laboratory should be extrapolated to clinical settings only with great caution, the results of this study have several implications for the use of
57 plasma HVA in clinical research as a reflection of brain dopamine function. First, they suggest that in the dose ranges used, plasma HVA may accurately reflect the acute response of brain dopamine to a wide range of chemically different neuroleptics. Since the only major pharmacologic characteristics shared in common by these various drugs is the ability to block brain dopamine receptors (Peroutka and Snyder, 1980) these results provide further evidence that the rise in plasma HVA after acute neuroleptic administration results from the blockade of brain dopamine receptors. Second, these results indicate that caution should be used in interpreting plasma HVA levels after the commonly used neuroleptic chlorpromazine. Although often used clinically in doses lower than those employed in this study, chlorpromazine apparently can increase plasma HVA by mechanisms other than the blockade of brain dopamine receptors. After the administration of chlorpromazine (and perhaps other neuroleptics which can produce hypotension), plasma HVA may not provide an accurate index of brain dopamine function. Third, this study suggests that the blockade of peripheral dopamine vascular receptors by neuroleptics probably does not substantially alter HVA clearance in the rat and further suggests that this may not be a major problem in interpreting the plasma HVA response to neuroleptics in man. Lastly, in conjunction with previous results of Bacopoulos et al. (1980) in the monkey, this study provides evidence that after the chronic administration of neuroleptic drugs, plasma HVA continues to reflect the effect of the neuroleptic drug on the brain dopamine system. These results raise the possibility that plasma HVA may be used in man as a measure of the response of the brain dopamine system to chronic neuroleptic administration. This work was supported in part by a Schizophrenia Biological Research Center grant from the Veterans Administration and a VA Research Associate Award to Kenneth S. Kendler, M.D. Gregory Austin provided excellent technical assistance.
Acknowledgments.
References Bacopoulos. homovanillic Journal
N.G., Hattox, S.E., and Roth, R.H. 3,4-Dihydroxyphenylacetic acid and acid in rat plasma: Possible indicators of central dopaminergic activity. European
yf’ Pharmacologic,
56,225
( 1979).
Bacopoulos. N.G., Heninger, G.R., and Roth, R.H. Effects of haloperidol and probenecid on plasma and CSF dopamine metabolites in the rhesus monkey (Macacca mulatta). Life Sciences, 23, 1805 (1978).
Bacopoulos, N.G.. Redmond. D.E., Baulu, J.. and Roth. R.H. Chronic haloperidol or fluphenazine: Effects on dopamine metabolism in brain, cerebrospinal fluid and plasma of Cercopirhecus aethiops (vervet monkey). Journal of’ Pharmacology and Experimental Therupeurics. 212, I (1980). Bowers. M.B., Jr., Heninger, G.R.. Sternberg. D., and Meltzer, H.Y. Clinical processes and Communications in Psychopharmacentral dopaminergic activity in psychotic disorders. c,o/og~. 4, 177 ( 1980). Chapman. B.J., Horn, N.M., Munday, K.A., and Robertson, M.J. The actions ofdopamine and of sulpiride on regional blood flows in the rat kidney. Journal of’ Physiology, 298, 437 (1980).
Cutler. N.R., Jeste, D.V., Karoum, serum homovanillic acid concentrations
F., and Wyatt, in schizophrenic
R.J. Low-dose apomorphine reduces patients. L$e Sciences, 30,753 (1982).
58 Goldberg, L.I., Kohli, J.D., Kotake, A.N., and Volkman, P.H. Characteristics ofthevascular dopamine receptor: Comparison with other receptors. Federation Proceedings, 19,2396( 1978). Heninger, G.R., Bacopoulos, N.G., Roth, R.H., Bowers, M.B., Jr., and Sweeney, D. Plasma and CSF dopamine, metabolites in psychiatric patients: Effects of probenecid and haloperidol. In: Usdin, E.. ed. Catecholamines: Basic and Clinical Frontiers. Pergamon Press. Oxford, p. 1887 (1979). Julou, L.. Scatton. B., and Glowinski, J. Acute and chronic treatment with neuroleptics: Similarities and differences in their action on nigrostriatal, mesolimbic. and mesocortical dopaminergic neurons. Advances in Biochemical Psyhopharmacolog_v, 16,617 (1977). Juorio, A.V. Effects of molindone and fluphenazine on the brain concentrations of some phenolic and catecholic amines in the mouse and the rat. British Journal qf’Pharmacolo,gv, 70, 475 (1980).
Kendler, K.S., Heninger, G.R.. and Roth, R.H. Brain contribution to the haloperidoiinduced increase in plasma homovanillic acid. European Journal qf Pharmacology, 71, 321 (1981).
Naber, D.. Pickar. D.. Davis, G.C., Cohen, R.M., Jimerson, D.C., Elchisak. M.A., DeFraites, E.G.. Kalin. N.H.. Risch, S.C.. and Buchsbaum. M.S. Naloxone effects of betaendorphin. cortisol, prolactin. growth hormone, HVA and MHPG in plasma of normal volunteers. P.s.~chopharmaco/og,,, 74, I25 (198 I). Nishibe, Y., Matsuo, Y.. Yoshizaki, T., Eigyo. M.. Shiomi. T., and Hirose, K. Differential effects of sulpiride and metoclopramide on brain: Homovanillic acid levels and shuttle box avoidance after syste,matic and intracerebral administration. Naunjan-Schmiedehergk Archives qf Pharmacology,
321,
190 (1982).
Peroutka, S.J.. and Snyder. S.H. Relationship of neuroleptic drug effects at brain dopamine, serotonin, alpha-adrenergic. and histamine receptors to clinical potency. American Journal q/’ Ps~~chiarrv, 137, 5 I8
( 1980).
51
Elsevier
Acute and Chronic Effects of Neuroleptic Drugs on Plasma and Brain Homovanillic Acid in the Rat Kenneth
S. Kendler
Received Jaman,
and Kenneth
L. Davis
9, 1984; revised version received June 5, 1984; accepred July 24, 1984.
Abstract. In the rat. the acute administration of the neuroleptic drug, haloperidol. produces a parallel increase in brain and plasma concentrations of the dopamine metabolite. homovanillic acid (HVA). The effect of other neuroleptic drugs, which may differ from haloperidol in their central and peripheral actions, on brain and plasma HVA has not been systematically investigated. Therefore, in this report we examine the acute effects of six different neuroleptic drugs. representing most major chemical classes of these drugs, on plasma and brain concentrations of HVA. Metoclopramide, fluphenazine, loxapine, and molindone produce parallel increases in brain and plasma HVA which closely resemble those produced by haloperidol. Compared to these neuroleptics. chlorpromazine produces a much greater increase in plasma HVA but a similar effect on brain HVA. The large chlorpromazine-induced increase in plasma HVA suggests that this drug alters peripheral production or clearance of HVA, perhaps via blockade of peripheral a-receptors. Of the available neuroleptics. sulpiride is one of the most specific and potent at blocking the dopamine vascular receptor. Administration of high doses of sulpiride produces only modest increases in both plasma and brain HVA, suggesting that blockade of peripheral dopamine receptors does not substantially alter peripheral clearance of HVA. After chronic administration of haloperidol for up to 21 days. plasma HVA continued to reflect the brain HVA response to drug administration. Key Words.
Plasma homovanillic
acid, neuroleptics.
In the rat, acute administration of the neuroleptic drug, haloperidol, produces a parallel increase in the concentration of the dopamine metabolite. homovanillic acid (HVA), in brain and plasma (Bacopoulos et al., 1979). This haloperidol-induced increase in plasma HVA appears to result from alterations in brain HVA production and not from changes in the peripheral production or clearance of HVA (Kendler et al., 1981). Haloperidol administration also increases plasma HVA in the monkey (Bacopoulos et al., 1978) and possibly in man (Heninger et al., 1979). Based on these and other studies, plasma HVA is now used in clinical settings as a possible measure of the response of brain dopamine systems to a variety of stimuli including neuroleptic
Kenneth 5 Kendler. M.D.. was t’ormerly Assistant Professor of Psychiatry, Mount Sinai School of Medicine. and is now in the Departments of Psychiatry and Human Genetics. Medical College of Virginia, Rxhmond. VA. Kenneth L. Davis. M.D.. is Professor of Psychiatry and Pharmacology, Mount Sinai School of Medicine, and Chief of Psychiatry. Bronx Veterans Administration Medical Center. (Reprint requests to Dr. K.L. Davis. Psychiatry Service( I 16A). Bronx VA Medical Center. I30 W. Kingsbridge Rd., Bronx. XY 1046X. USA.) 0165-1781
X4 $03.00 % 1984 Elsevier Science Publishers B.V
52 drugs (Bowers et al.. 1980; Naber et al.. 1981; Cutler et al., 1982). The acute effects on plasma HVA of neuroleptics other than haloperidol has not, to our knowledge. been investigated. Neuroleptics differ considerably in their pharmacologic properties both in the brain (Peroutka and Snyder, 1980) and the periphery (Goldberg et al.. 1978). For example. some neuroleptic drugs are considerably more potent than haloperidol at blocking a-adrenergic receptors (Peroutka and Snyder, 1980). The hypotension that often accompanies cu-adrenergic blockade could alter the peripheral production or clearance of HVA. Furthermore. some neuroleptics are more potent than haloperidol at blocking dopamine receptors in the vascular tree. Since administration of at least one such neuroleptic, sulpiride (Chapman et al., 1980). has been shown to alter renal blood flow, it is possible that blockade of renal vascular dopamine receptors could alter HVA clearance. Therefore. the acute effect on plasma HVA of certain neuroleptics might differ considerably from that produced by haloperidol. The chronic effects of neuroleptic drugs on brain dopamine systems differ from their acute effects (Julou et al., 1977). In the monkey, Bacopoulos et al. (1980) demonstrated only a partial correlation between plasma and brain HVA concentrations after chronic administration of fluphenazine and haloperidol. This study was limited because brain samples could be obtained at only one time point (3 weeks) after chronic drug administration. In this report. the acute effects of six different neuroleptic drugs. representing most of the major chemical classes of these compounds. on brain and plasma HVA in the rat were investigated. These drugs include a neuroleptic with a-adrenergic blocking potency, chlorpromazine. and one that is potent at blocking dopamine vascular receptors. sulpiride. Furthermore, the response of plasma and brain HVA to chronic haloperidol administration in the rat was also examined. The results obtained were of potential relevance to the clinical use of plasma HVA in man. However. although similarities exist in the dopamine system of man and rat. sufficient differences have been found to indicate caution in the extrapolation of results from one species to another. Methods Male. Sprague-Dawley rats. habituated to the laboratory environment for at least 7 days and weighing from 260 to 400 g. were used in all experiments. Animals were maintained on a IO-hour light. l4-hour dark schedule with food and water available ad lib. For 24 hours before sacrifice. animals were fasted with water freely available. All drugs were administered intraperitoneally in less than I ml of vehicle. Debrisoqutn was administered 4 hours before sacrifice. a time interval previously shown to maximire the debrisoquin-produced decrement in plasma HVA (Kendler et al., 1981). Metoclopramide uas administered 60 minutes. sulpiride 90 minutes. and all other neuroleptics 120 minutes before sacrifice. Doses of neuroleptics were chosen to be roughI!, equivalent. from a clinical perspective. to I to 2 mg kg of haloperidol. Molindone. however. because of evidence that at high doses it demonstrates dopamine agonist properties (Juorio. 1980). was administered at a dose of 2.5 mg kg. Debrisoquin sulfate (Hoffmann-La Roche. Nutley. NJ). metoclopramide (Robins Research I.aboratories. Richmond. VA). and molindone (Endo Laboratories. Garden City. NY) Mere dissolved in bacterioatatic water. Sulpiride (Delagrange International. Paris) was dissolved in bacteriostatic water titrated to /‘H 2.X with I 12’ HCI. Fluphenaline hydrochloride injectable
53 (Squibb. Princeton, NJ), loxapine hydrochloride injectable(Lederle Laboratories, Wayne, NJ), chlorpromazine hydrochloride injectable (Smith, Kline & French Laboratories, Philadelphia. PA). and haloperidol injectable (McNeil Laboratories, Fort Washington, PA) were diluted in bacteriostatic water. Blood samples of 5-7 ml were collected by open chest cardiac puncture after brief ether anesthesia into a heparinized syringe. After open chest cardiac puncture, the rats immediately were decapitated and the brains rapidly removed. In some studies, the whole brain was then immediately frozen on dry ice, while in others the striata were bilaterally dissected out and then frozen on dry ice. In the experiment examining the effects of chronic neuroleptic administration, haloperidol, I mg: kg, was administered daily for I. 4. 7, or 2 I days. On the final experimental day, haloperidol was administered 2 hours before sacrifice. The olfactory tubercles were assayed for HVA in this study because the pattern of habituation of HVA to chronic neuroleptic administration in this region is intermediate between that found in the striatum and cortex (Julou et al.. 1977). For all studies. blood samples were prepared and free plasma HVA was assayed by gaschromatographic mass-spectrometric quantitation as previously described (Bacopoulos et al.. 1979). Sufficient plasma was available for nearly all samples to be assayed for free HVA in duplicate, with the mean value for the two samples then being used in the statistical analysis. Free HVA in brain was assayed by high performance liquid chromatography with electrochemical detection using homovanillic alcohol as an internal standard. Statistical analysis was carried out by means of the Students t test, Pearson’s productmoment correlation. and one and two-way analyses of variance (ANOVAs). When needed, a post-hoc test after a significant one-way ANOVA was performed by the Student-NewmanKeuls procedure with ranges set at the 0.05 level. All p values are reported two-tailed; p values greater than 0.05 are not significant (NS).
Results Acute Administration of Neuroleptics. In the first experiment, the effect of metoclopramide on plasma and whole brain HVA was examined with and without pretreatment with debrisoquin, a drug that inhibits peripheral but not central HVA production (Kendler et al., 1981). Debrisoquin produced a significant decrement in plasma HVA (F= II.78; dj’= I, 22;~ q0.002), but no significant effect on whole brain HVA (Fz0.07: c/j'= I,22; NS) (Table 1). Metoclopramide significantly elevated both plasma HVA (F= 9.54; rlf= 1,22;p = 0.005) and whole brain HVA (F= 36.5 I; elf‘= I, 22; p < 0.00I).No significant interaction was seen between the effects of debrisoquin and metoclopramide treatment on either plasma or brain HVA. Following vehicle pretreatment, metoclopramide increased plasma HVA 33%. Table 1. Effect of metoclopramide on plasma and whole brain HVA with and without debrisoquin pretreatment Plasma free HVA (ng/ml) Pretreatment
Treatment
Vehicle
3.6 + 0.5
4.6 i
1.1
Debrisoauin
2.9 2 0.5
3.5 -c 0.5
Whole brain free HVA (ng/g) Pretreatment 226k
71
327 + 132
Treatment 656k317 601 t 331
Debnsoquln 40 mg/kg wasadmlnistered 4 hours before and metoclopramide 120 mg/kgl 1 hour before sacnfwx n = 6 to 7 per group;. Statistical analysis by two-way ANOVA. For plasma HVA, effects of both pretreatment F= 11.76. df= 1. 22; p =0.0021 and treatment are significant lF=9.54; df= 1, 22; p=O.O05j, but their Interaction ~snot fF=O.96; df= 1,22;p=NS1. Forwhole brain HVA,onlythetreatmenteffect issignificant F = 36 51. df = 1. 22;~ < 0 0001 8.Neitherthepretreatmenteffect if =0 07;df= 1.22; NSI northepretreatment X treatment Interaction was significant F = 0 74; df = 1. 22: NS). Results are expressed as means + SD.
54 The acute effects of other neuroleptics on plasma and brain HVA concentrations were examined without debrisoquin pretreatment. With a repeated sampling procedure, fluphenazine produced a significantly greater change in plasma HVA from basal values than seen with vehicle administration (Table 2) (t 2.80, ulf’ = 13, p = 0.0 15). Fluphenazine also significantly increased whole brain HVA (t = 6.04, Q’= 13, p 0.00004). q
q
Table 2. Acute administration
plasma
and whole
brain
HVA responses
to fluphenazine
Treatment
Change in free plasma HVA (ng/ml)
Whole brain free HVA (ng/g)
Vehicle Fluphenazine t idf= 131
0.7 + 0.6 2.6 i 1.6 2.80 0.015
2492 71 932 k 310 6.04 0.0001
P
Animals were etherized and 2.5 cc of blood obtained by closed chest cardiac puncture for basal plasma HVA values. Animals were then administered vehicle or fluphenazlne. 2.5 mg/kg. and 2 hours laterwere sacrificed The change in plasma HVA represents the HVA concentration obtalned at sacrifice minus the basal HVA concentrat!on. Basal concentrations of plasma HVA did not differ in the 2 groups lvehlcle. 2 5 + 0.1 ng/cc; fluphenazine: 2.3 i 0.2 rig/ccc ‘t = 1.33, NS, ‘n = 7-8 per group 8.StatIstical analysis by Student’s t test Results are expressed as means + SD
In the next study, the acute effects of sulpiride, loxapine, and molindone on plasma brain HVA were compared with those found after vehicle administration (Table 3). A one-way ANOVA indicated that both plasma HVA (F = 5. I ; df = 3, 30: I_’= 0.006) and whole brain HVA (F = 123.6; # = 3, 30; p < 0.0001) differed significantly across the four treatment groups. By a Student-Newman-Keuls procedure, loxapine and molindone, but not sulpiride, significantly increased plasma and whole brain HVA. Loxapine produced a 35qo and molindone a 370,;#increase in plasma HVA. Compared to vehicle administration, chlorpromazine produced a highly significant 318% increase in plasma HVA (t = 5.30; (l/‘= 13; p = 0.0001) and a 267%, increase in striatal HVA (t = 7.05; (if= 13. p 0.00001) (Table 3). To investigate further this very large increase in plasma HVA produced by chlorpromazine, the effect of this drug on plasma and whole brain HVA was compared with that of haloperidol (Table 3). Although the increases in whole brain HVA produced by the two neuroleptics were similar. chlorpromazine produced a much greater increase in plasma HVA than did haloperidol. and whole
q
Chronic Administration of Neuroleptics. The effect of up to 21 days of daily haloperidol administration on plasma and brain HVA is seen in Table 4. The greatest increase in HVA in both plasma and olfactory tubercle was produced after a single injection of haloperidol. After repeated administration of the neuroleptic. a modest habituation in the elevation of HVA was seen in both plasma and brain. Over the five time points measured in this study, the correlation between the mean concentration of HVA in plasma and the olfactory tubercle was 0.96 @ < 0.01).
55 Table 3. The acute response of plasma and brain HVA to sulpiride, molindone, chlorpromazine and haloperidol Plasma free HVA
Whole brain free HVA
(w/ml)
(ng44)
4.3 It 0.0
1932
Sulpiride
4.5 * 0.8
265 ? 51
Loxapine
5.8 i 1.1
789 ? 88
Molindone
5.9 ? 1.4
580 -t 99
5.101
123.642
3, 27)
Striatal free HVA
(w/g)
Vehicle
F idf=
loxapine,
13
Chlorpromazine
16.7 t 6.2
3,450 Itr 724
Vehicle
4.0 + 0.6
1,294 + 410
tldf=131
5.302
7.052
Chlorpromazine
13.7 2 5.6
Haloperidol
4.2 ? 0.8
Vehicle
3.6 -t 1.3
/=(df=2.171
605 i 98 657?
111
196t50
19.302
51.802
Loxapine, 5 mg/kg, mollndone, 2.5 mg/kg. chlorpromazine, 25 mg/kg. and haloperidol. 2 mg/kg. were administered intraperitoneally (1.p I 120minutesbeforesacrifice Sulpiride, lOOmg/kg,wasadministeredi.p.90 minutesbeforesacrlficeln=7-8pergroupI. Statisticalanalysiswasbyone-wayANOVAandStudent’sttest. For experiment with vehicle. sulpirtde, loxapine. and molindone. Student-Newman-Keuls procedure divided the plasma WA reponses Into 2 groups: i 1 / vehicle and sulplride and 121 loxapine and molindone. The brain HVA responses were divided Into 3 groups: 11 I vehicle and sulpiride. 121 molindone, and (31 loxaplne. Results are expressed as means +- SD. 1. p
Table 4. Effect of chronic haloperidol the olfactory tubercle Treatment Vehicle
administration
on HVA in plasma and in
Free plasma HVA
Free HVA in olfactory tubercle
(ng/mU
(ngfg)
2.5 ? 0.5
1,011 ? 442
Haloperidol
X 1 day
3.8 + 1.9
2,313 + 992
Haloperidol
X 4 days
3.4 + 0.8
1,895 i- 691
Haloperidol
X 7 days
3.4 2 1.3
1,630 2 683
Haloperidol
X 21 days
3.4 * 0.5
1.7482812
F (df = 4,341 D
9.06
2.82
0.0001
0.04
Haloperldol, 1 ng/kg. was administered daily for prescribed period of timeand 2 hours beforesacrifice group1 StatIstical analysis is by one-way ANOVA. Results are expressed as means i SD
in = 7 per
Discussion Although representing different chemical classes of neuroleptics, metoclopramide. fluphenazine, loxapine, and molindone have acute effects on brain and plasma HVA similar to those previously found with haloperidol (Kendler et al., 198 I). In addition. the rise in plasma HVA produced by metoclopramide after debrisoquin suggests that the effect of this drug on plasma HVA. like that of haloperidol, is not mediated by changes in peripheral HVA production (Kendler et al.. 1981). Debrisoquin inhibits
production of HVA outside of the brain. Thus, if the metoclopramide-induced rise in plasma HVA resulted from increased peripheral HVA production, debrisoquin pretreatment would be expected to eliminate this rise. This, however, was not observed (Table 1). In the dose administered (25 mg: kg), chlorpromazine produced a much greater rise in plasma HVA than seen with any other neuroleptic. Compared to haloperidol, chlorpromazine produced a much greater rise in plasma HVA but a similar rise in whole brain HVA. These results suggest that the large increase in plasma HVA seen after chlorpromazine results from peripheral and not central actions of this drug. Chlorpromazine could increase peripheral production or diminish peripheral clearance of HVA by several mechanisms. The ratio of potency of chlorpromazine at blocking a-adrenergic to dopaminergic receptors is one of the highest found for neuroleptics (Peroutka and Snyder, 1980). Since blockade of a-adrenergic receptors causes hypotension. which could increase peripheral production of HVA or reduce its clearance, the peripheral a-adrenergic blocking action of chlorpromazine is one plausible mechanism to explain its marked effect on plasma HVA concentrations. Sulpiride is among the most potent and specific antagonists of the dopamine vascular receptor (Goldberg et al.. 1978). Administered alone, sulpiride increases arterial resistance in the renal cortex (Chapman et al., 1980). It is therefore possible that sulpiride administration might reduce renal clearance of HVA. However, because of slow and incomplete penetrance into the brain (Nishibe et al.. 1982), sulpiride. when administered at a high dose, produced only a modest increase in whole brain HVA and no substantial increase in plasma HVA. These results suggest that blockade of peripheral vascular dopamine receptors does not substantially diminish clearance of HVA from plasma. This and other studies (Bacopoulos et al.. 1979; Kendler et al.. 1981) suggest that after acute administration of most neuroleptics, the change in plasma HVA concentrations provides a reasonably accurate measure of changes in brain HVA production. However, in clinical settings, neuroleptics are most commonly administered chronically. Chronic neuroleptic administration produces a different response in the brain dopamine system than does acute administration (Julou et al., 1977). Therefore, it was of interest to examine further the response of brain and plasma HVA to chronic neuroleptic administration. Throughout a 21-day course of daily haloperidol administration. the changes in plasma HVA closely paralleled the changes in HVA in the olfactory tubercle, in which the response of HVA to chronic neuroleptic administration is intermediate between that found in the striatum and cortex. In accord with previous findings in the monkey (Bacopoulos et al., 1980), these findings raise the possibility that plasma HVA may continue to reflect changes in brain HVA during chronic neuroleptic administration, particularly in the forebrain dopamine systems where neuroleptic drugs probably produce their antipsychotic effect. The relationship between changes in brain and plasma HVA during chronic neuroleptic administration would be further clarified by a study examining HVA levels in multiple brain areas as well as plasma. Although results from the laboratory should be extrapolated to clinical settings only with great caution, the results of this study have several implications for the use of
57 plasma HVA in clinical research as a reflection of brain dopamine function. First, they suggest that in the dose ranges used, plasma HVA may accurately reflect the acute response of brain dopamine to a wide range of chemically different neuroleptics. Since the only major pharmacologic characteristics shared in common by these various drugs is the ability to block brain dopamine receptors (Peroutka and Snyder, 1980) these results provide further evidence that the rise in plasma HVA after acute neuroleptic administration results from the blockade of brain dopamine receptors. Second, these results indicate that caution should be used in interpreting plasma HVA levels after the commonly used neuroleptic chlorpromazine. Although often used clinically in doses lower than those employed in this study, chlorpromazine apparently can increase plasma HVA by mechanisms other than the blockade of brain dopamine receptors. After the administration of chlorpromazine (and perhaps other neuroleptics which can produce hypotension), plasma HVA may not provide an accurate index of brain dopamine function. Third, this study suggests that the blockade of peripheral dopamine vascular receptors by neuroleptics probably does not substantially alter HVA clearance in the rat and further suggests that this may not be a major problem in interpreting the plasma HVA response to neuroleptics in man. Lastly, in conjunction with previous results of Bacopoulos et al. (1980) in the monkey, this study provides evidence that after the chronic administration of neuroleptic drugs, plasma HVA continues to reflect the effect of the neuroleptic drug on the brain dopamine system. These results raise the possibility that plasma HVA may be used in man as a measure of the response of the brain dopamine system to chronic neuroleptic administration. This work was supported in part by a Schizophrenia Biological Research Center grant from the Veterans Administration and a VA Research Associate Award to Kenneth S. Kendler, M.D. Gregory Austin provided excellent technical assistance.
Acknowledgments.
References Bacopoulos. homovanillic Journal
N.G., Hattox, S.E., and Roth, R.H. 3,4-Dihydroxyphenylacetic acid and acid in rat plasma: Possible indicators of central dopaminergic activity. European
yf’ Pharmacologic,
56,225
( 1979).
Bacopoulos. N.G., Heninger, G.R., and Roth, R.H. Effects of haloperidol and probenecid on plasma and CSF dopamine metabolites in the rhesus monkey (Macacca mulatta). Life Sciences, 23, 1805 (1978).
Bacopoulos, N.G.. Redmond. D.E., Baulu, J.. and Roth. R.H. Chronic haloperidol or fluphenazine: Effects on dopamine metabolism in brain, cerebrospinal fluid and plasma of Cercopirhecus aethiops (vervet monkey). Journal of’ Pharmacology and Experimental Therupeurics. 212, I (1980). Bowers. M.B., Jr., Heninger, G.R.. Sternberg. D., and Meltzer, H.Y. Clinical processes and Communications in Psychopharmacentral dopaminergic activity in psychotic disorders. c,o/og~. 4, 177 ( 1980). Chapman. B.J., Horn, N.M., Munday, K.A., and Robertson, M.J. The actions ofdopamine and of sulpiride on regional blood flows in the rat kidney. Journal of’ Physiology, 298, 437 (1980).
Cutler. N.R., Jeste, D.V., Karoum, serum homovanillic acid concentrations
F., and Wyatt, in schizophrenic
R.J. Low-dose apomorphine reduces patients. L$e Sciences, 30,753 (1982).
58 Goldberg, L.I., Kohli, J.D., Kotake, A.N., and Volkman, P.H. Characteristics ofthevascular dopamine receptor: Comparison with other receptors. Federation Proceedings, 19,2396( 1978). Heninger, G.R., Bacopoulos, N.G., Roth, R.H., Bowers, M.B., Jr., and Sweeney, D. Plasma and CSF dopamine, metabolites in psychiatric patients: Effects of probenecid and haloperidol. In: Usdin, E.. ed. Catecholamines: Basic and Clinical Frontiers. Pergamon Press. Oxford, p. 1887 (1979). Julou, L.. Scatton. B., and Glowinski, J. Acute and chronic treatment with neuroleptics: Similarities and differences in their action on nigrostriatal, mesolimbic. and mesocortical dopaminergic neurons. Advances in Biochemical Psyhopharmacolog_v, 16,617 (1977). Juorio, A.V. Effects of molindone and fluphenazine on the brain concentrations of some phenolic and catecholic amines in the mouse and the rat. British Journal qf’Pharmacolo,gv, 70, 475 (1980).
Kendler, K.S., Heninger, G.R.. and Roth, R.H. Brain contribution to the haloperidoiinduced increase in plasma homovanillic acid. European Journal qf Pharmacology, 71, 321 (1981).
Naber, D.. Pickar. D.. Davis, G.C., Cohen, R.M., Jimerson, D.C., Elchisak. M.A., DeFraites, E.G.. Kalin. N.H.. Risch, S.C.. and Buchsbaum. M.S. Naloxone effects of betaendorphin. cortisol, prolactin. growth hormone, HVA and MHPG in plasma of normal volunteers. P.s.~chopharmaco/og,,, 74, I25 (198 I). Nishibe, Y., Matsuo, Y.. Yoshizaki, T., Eigyo. M.. Shiomi. T., and Hirose, K. Differential effects of sulpiride and metoclopramide on brain: Homovanillic acid levels and shuttle box avoidance after syste,matic and intracerebral administration. Naunjan-Schmiedehergk Archives qf Pharmacology,
321,
190 (1982).
Peroutka, S.J.. and Snyder. S.H. Relationship of neuroleptic drug effects at brain dopamine, serotonin, alpha-adrenergic. and histamine receptors to clinical potency. American Journal q/’ Ps~~chiarrv, 137, 5 I8
( 1980).