243

Psychiatry Research, 34243-257 Elsevier

Serum Amino Acids, Central Monoamines, and Hormones in Drug-Naive, Drug-Free, and Neuroleptic-Treated Schizophrenic Patients and Healthy Subjects Marie Luise Rao, Gisela Gross, Bernd Strebel, and Joachim Klosterk6tter

Peter Brfiunig, Gerd Huber,

Received February 22, 1990; revised version received August 6.1990; accepted August 19.1990. Abstract. Basal serum amino acids (including central monoamine precursors), central monoamines, and hormones were studied in schizophrenic patients (drugnaive, n = 20; drug-withdrawn for 3 or more days, n = 67; neuroleptic-treated, n = 23) and healthy subjects (n = 90) to answer the following questions: (1) Do neuroleptic-withdrawn and neuroleptic-naive patients differ on these serum measures? (2) What are the effects of neuroleptic treatment on these measures? (3) On which variables do drug-free and neuroleptic-treated patients differ? Because serum amino acid, central monoamine, and hormone levels were similar in drugnaive and drug-withdrawn patients, data from these groups (“drug-free”) were combined and compared to those of healthy subjects and neuroleptic-treated patients. Asparagine, citrulline, phenylalanine, and cysteine were higher, while tyrosine, tryptophan, and the ratio of tryptophan to competing amino acids were significantly lower in drug-free schizophrenic patients than in healthy subjects. Dopamine was increased, and melatonin and thyroid hormones were decreased in drug-free schizophrenic patients compared to healthy subjects. Norepinephrine, epinephrine, and prolactin were higher in neuroleptic-treated men compared to drug-free male patients or healthy men. These results are consistent with the hypothesis of dopaminergic overactivity in schizophrenia, which might be caused by altered amino acid precursor availability and could be related to the decrease in melatonin and reduction in thyroid hormone levels.

Key Words. Schizophrenia, drug-naive, central monoamines, hormones.

neuroleptic

drug response,

amino acids,

Dysregulation of dopaminergic pathways has been postulated as a pathophysiological basis for schizophrenia; this hypothesis has been extended to the involvement of other central monoamines such as norepinephrine and serotonin (see van Kammen and Antelman, 1984; Meltzer, 1987; Bleich et al., 1988). The metabolism of central monoamines is related to the availability of their amino acid precursors, i.e., serum This article is part of the Dissertation of Bernd Strebel, Hohe Medizinische Fakultti der Universitl Bonn. The results were presented at the Annual Meeting of the American College of Neuropsychopharmacology, December 1989, Maui, HI. Marie Luise Rao, Ph.D., is Assistant Professor of Biochemical Endocrinology and Head of the Neurochemistry Laboratories; Gisela Gross, M.D., is Associate Professor of Psychiatry; Bernd Strebel is a Postgraduate in Medicine; Gerd Huher, M.D., is Professor Emeritus of Psychiatry; and Joachim Klosterk8tter, M.D., is Assistant Professor, Department of Psychiatry, Rheinische Friedrich-WilhelmsUniversitt, Bonn, Germany; Peter Braunig, M.D., is Senior Psychiatrist at the Westftilische Zentrum fiir Psychiatric, Universitiitsklinik Bochum, Germany. (Reprint requests to Dr. M.L. Rao. Universitlsnervenklinik, Psychiatric, Sigmund-Freud-Str. 25, 5300 Bonn 1, Germany.) 0165-1781/90/$.03.50

@ 1990 Elsevier Scientific

Publishers

Ireland

Ltd.

244 concentrations of tyrosine and tryptophan (Fernstrom and Wurtman, 1972; FernStrom and Faller, 1978) and to the activity of the endocrine system, where central monoamines govern hormonal release into the peripheral circulation (Roth and Grunfeld, 1985). Serum concentrations of amino acid precursors, monoamines, and hormones may be altered in schizophrenia, but research results are inconsistent. This inconsistency could be due to the heterogeneity of the schizophrenias, to previous neuroleptic treatment, to the time of sampling, or to the methods used. Many reports focus on one biochemical variable rather than groups of biochemically related compounds. We hypothesize that the measurement of a group of related substances in the peripheral circulation might provide important information about the link between central monoamine synthesis and action. In this context, we addressed the following questions: (1) Do neuroleptic-withdrawn and neurolepticnaive patients differ in serum levels of amino acids, central monoamines, and hormones? (2) Which variables differ between drug-free schizophrenic patients and controls? (3) How does neuroleptic treatment affect the sequence of events from the amino acid precursor to central monoamine synthesis and hormonal output? In an attempt to answer these questions, we determined basal levels of serum amino acids, monoamines, and hormones in schizophrenic patients (drug-naive, drug-withdrawn, and neuroleptic-treated) and healthy control subjects.

Methods Subjects included 90 healthy controls and 110 schizophrenic inpatients from the Psychiatric Clinic of the University of Bonn (Table 1). The control subjects (medical/technical personnel and students) spent the day of study in the hospital. A questionnaire covering health history, personal data, and current activities was completed to ensure the exclusion of subjects who smoke, drank, or exercised excessively, who were overweight, who were not eating the standard diet required for participation, or who had first-degree relatives with neurological and psychiatric diseases. Patients and control subjects received a standard hospital diet, which consisted of 80 f 5 g protein, 100 f 5 g fat, and 220 f 20 g carbohydrate per day; 50% of the caloric intake was at noon. Psychiatric patients and healthy subjects did not show clinical signs of endocrinopathies or other diseases, and their weight and height were in the normal range. Diagnostic assessments and clinical ratings were performed independently by three psychiatrists who were unaware of the laboratory findings. Patients were diagnosed according to the criteria of K. Schneider (1980); all patients met Research Diagnostic Criteria for schizophrenia or schizoaffective psychosis (Spitzer et al., 1978). Patients were initially divided into three neuroleptic drug categories: (1) drug-naive, (2) drug-withdrawn (patients who underwent a washout of 3 or more days), and (3) neuroleptic-treated (patients medicated according to clinical requirements for at least 5 days) (see Table 1). Subjects.

Table 1. Characteristics

of schizophrenic

patients and healthy subjects Me

Men

Subiects

Women

Mean

SD

Patients Drug-naive Neuroleptic-withdrawn 3 or more days

11

9

33

11

36

31

35

12

11

12

34

13

49

41

251

for

Neuroleptic-treated Healthy subiects 1. p < 0.05 vs. the other 3 groups.

5

245

Serum neuroleptic bioactivity was analyzed in all participants, except in the first and second patients (Table 2). Healthy subjects and drug-naive or drug-withdrawn patients were included in the study when their serum neuroleptic activity was below the limit of detection (< 0.3 ng/ml). Patients on depot neuroleptics were not included. Except for chloral hydrate, no psychopharmacological agents were administered. Approximately one-fourth of the females took oral contraceptives. Initially 165 patients and 96 healthy subjects entered the study. Fiveteen patients were excluded because blood was taken later than 8 a.m., 40 because followup revealed diagnoses other than schizophrenia or because serum neuroleptic determinations in “drug-free” patients showed detectable neuroleptic activity. Out of 96 healthy subjects, six were excluded: some were on a vegetarian diet, others engaged in vigorous exercise, and one had neuroleptic levels comparable to steady-state levels of patients. Circadian profiles of patients and subjects were also monitored for cosinor analysis (Rao et al., in preparation).

Table 2. Serum neuroleptic

activities

determined

by radioreceptor Neuroleptic

Drug Benperidol Haloperidol

Number

Mean dose (mg/day)

Range (nW

assay activity Mean (nW

1

16

11

11

13

20

2.5-61

20

14

14

Trifluperidol

1

5

Promethazine

4

50

2-20

11

Thioridazine

2

125

ND

ND

Chlorwothixene

1

30

42

42

Note. ND = not determined

Procedures. Blood was drawn from bed-rested patients and healthy subjects by venipuncture after an overnight fast between 7 and 8 a.m.; blood (2 ml) was withdrawn into tubes containing EDTA for the determination of serotonin. The remaining blood clotted for 30 min and was centrifuged for 10 min at 3,000g. Samples were divided and stored separately to permit the determination of each parameter without repeated freezing and thawing. For the analysis of amino acids, serum was diluted 1: 1 with 5% sulfosalicylic acid, swirled for 30 set, and centrifuged for 2 min at 12,OOOg. The supernatant was frozen at -28 “C. One part of the serum was stabilized by the addition of EGTA (final concentration 6.1 mmol/ 1) and gluthathione (4.8 mmol/l) and frozen at -82 ‘C for the catecholamine determination; serum was also stored at -82 “C for the melatonin assay and at -28 “C for the determination of pituitary hormones and cortisol. From internal quality control procedures which were run with each assay, it was inferred that the amino acids, central monoamines, and hormones were stable for more than 12 months under the storage conditions used; samples were usually analyzed within 6 months. Amino acids (Table 3) were determined by ion exchange chromatography on a DTC-27 10 resin (LC 5000, Biotronik, Maintal, Germany) using a modification of the method of Moore and Stein (195 1). The chromatograms were evaluated with a Shimadzu integrator. The intraassay and interassay coefficients of variation for tryptophan and other amino acids were 4% and 8%, respectively. The lower limit of detection, as inferred from standard curves for five key amino acids (tyrosine, phenylalanine, tryptophan, taurine, and glutamate), was 3 pmol/ 1 (for methodological details and quality control procedures, see Rao and Fels, 1987). Serotonin was determined by sensitive high-performance liquid chromatography with electrochemical detection (Rao and Fels, 1987); the intra-assay and interassay coefficients of variation were 3.6% and 570, respectively. The lower limit of sensitivity was 0.05 pmol/l. The catecholamines were analyzed by a modification (Rao et al., 1984) of the radioenzymatic assay of Peuler and Johnson (1977). Pretreatment of rats for the isolation of catechol-0 methyltransferase according to Brown and Jenner (198 1) improved the sensitivity of the assay to

246 65, 59, and 54 pmol/l for dopamine, norepinephrine, and epinephrine, respectively, when 0.05 ml serum was analyzed. The respective intra-assay coefficients of variation for dopamine, norepinephrine, and epinephrine were 5%, 4%, and 3% (n = 10). The respective interassay coefficients of variation were 13%, 8%, and 10%. The influence of stability and posture of patients on serum catecholamines was described by Rao and Mager (1987). Hormonal levels were analyzed by radioimmunoassay (RIA) with commercially available kits. Serum melatonin was determined by RIA using antibodies obtained from Dr. G, M. Brown (McMaster University, Hamilton, Ontario). The intra-assay and interassay coefficients of variation were 10% and 1I%, respectively. When increasing amounts of exogenous melatonin were added to the serum samples, the regression line passed through the origin after subtraction of endogenous immunoreactive melatonin; the lower limit of sensitivity of the assay was 0.1 nmoljl. The thyroid stimulating hormone (TSH) assay (Henning-Berlin, Germany) was calibrated with the World Health Organization (WHO) standard 68/38; the sensitivity was 0.19 mUi 1;the intra-assay and interassay coefficients of variation were 2% and 8%, respectively. The human growth hormone (HGH) assay (Sorin Biomedica) was calibrated with the WHO standard 66/ 2 17. The lower limit of sensitivity was 0.01 nmol/ I; the intra-assay and interassay coefficients of variation were 6% and 8%, respectively. The intra-assay and interassay coefficients of variation for cortisol (Sorin Biomedica) were 4%and 8%, respectively. Cross-reactivity with progesterone, testosterone, and dihydrotestosterone was < 0.6%. Serum neuroleptic activity was assayed by a modification (Rao, 1986) of the radioreceptor assay of Creese and Snyder (1977). Serum neuroleptic levels were expressed in neuroleptic units. One neuroleptic unit was equivalent to the displacement of receptor-bound JH-spiroperidol brought about by adding 0.1 ml serum containing 1 nmol/l haloperidol. The lower limit of sensitivity was 1 nmol haloperidol equivalent/l, corresponding to 0.37 ng/ml. The intra-assay and interassay coefficients of variation were 5% and S%, respectively. Statistics. Assays were carried out in duplicate. Data were grouped when no difference in serum levels between men and women was observed with the Mann-Whitney U test. Linear regression analysis was performed to test for correlation. Serum concentrations were tested for normal distribution with the Kolmogoroff-Smirnoff test. Group differences were tested with a multivariate analysis of variance (MANOVA) and the Bonferroni test; we applied the latter to obtain information regarding the significance of mean differences of the different variables in the multiple groups (e.g., healthy subjects vs. drug-naive patients, vs. drug-withdrawn patients, vs. neuroleptic-treated patients). The values given are means * 1 SD; the level of difference was 5%. Results Serum Levels of Amino Acids, Biogenic Amines, and Hormones. Except for aspartate, glutamate, proline, cysteine, methionine, lysine, histidine, 3-methylhistidine, arginine, taurine, epinephrine, dopamine, melatonin, prolactin, and growth hormone, all other variables were normally distributed in healthy subjects. Serum levels in men and women differed for cysteine, methionine, lysine, serine, glutamine, proline, valine, isoleucine, leucine, lysine, norepinephrine, epinephrine, prolactin, and growth hormone, and these were analyzed for men and women separately (Tables 3-5). Age was similar in the schizophrenic subgroups @ > 0.05); however, all schizophrenic subgroups had a higher mean age than that for the healthy subjects @ < 0.05). In healthy subjects, no correlation was observed between age and serum concentrations of the biochemical variables. Drug-naive vs. drug-withdrawn patients. Drug-naive patients and drugwithdrawn patients had serum neuroleptic activity levels below the limit of detection. Because serum variables did not differ between the drug-naive and drug-withdrawn

241 groups (Tables 3-5) data for the two groups were combined to explore whether serum levels in the “drug-free” (drug-free + drug withdrawn) group differed from those in the neuroleptic-treated and healthy control groups. Drug-free patients vs. controls. Tyrosine, phenylalanine, tryptophan, arginine, cysteine, glutamine, and the ratio of tryptophan to competing amino acids differed in drug-free patients as compared to values for healthy subjects (Table 3). Serum dopamine levels were higher in drug-free patients than in controls (Table 4). Serum melatonin and triiodothyronine levels were lower in patients than in controls @ < 0.05); there was also a trend for lowered TSH and thyroxine in patients compared to controls @ < 0.10; Table 5). Effect of neuroleptic treatment. Neuroleptic treatment was associated with a change in asparagine, citrulline, and 3-methylhistidine-amino acids that are not involved in the central monoamine pathways (Table 3). Norepinephrine and epinephrine levels were higher in neuroleptic-treated male patients than in drug-free male patients (Table 4). Serum prolactin levels were increased in medicated schizophrenic men compared to the drug-free cohort (Table 5). In women the levels were also higher in the medicated than in the drug-free group, but this increase was not statistically significant. We retested the female patients at 10 a.m. and found serum prolactin levels to be higher in the medicated than in the drug-free cohort (0.88 + 0.36 vs. 0.17 + 0.02 nmol/l). Serum neuroleptic activity correlated with serum prolactin concentrations in men (r = 0.57, n = 11, p < 0.05) and women (I = 0.77, n = 12, p < 0.001).

Discussion Serum amino acid, biogenic monoamine, and hormone concentrations were similar in drug-naive and drug-withdrawn schizophrenic patients. We are not aware of any other study offering comparable data. The schizophrenic patients were older than the healthy subjects. In the Bonn area, it was not possible to recruit a similar proportion of subjects over 40-50 years who were healthy, refrained from medication and substance abuse, and were willing to spend 30 hours in the hospital. Nevertheless, the patient-control comparisons appear appropriate since age dependency of the biochemical variables was not demonstrated in the controls. In addition, several investigators have pointed out that between 18 and 60 years, age does not influence serum levels of hormones (Wilke, 1983; Yamada et al., 1984). Recent discussions in clinical sciences have suggested that fluctuations in circulating substances should not necessarily be ascribed to the aging process, but that a thorough investigation should be made instead; this statement applies to the most thoroughly studied variables such as blood lipids and thyroid hormones (Gambert and Tsitouras, 1985; Expert Panel, 1988). We observed serum phenylalanine to be higher, and tryptophan and tyrosine (the large neutral amino acid precursors for central monoamine synthesis) to be lower in schizophrenic patients than in healthy subjects; the same applied to the ratio between tryptophan and its competing amino acids, suggesting lowered amino acid precursor availability-that is, lowered availability of tryptophan and possibly tyrosine in the central nervous system. Bjerkenstedt et al. (1985) observed that the large neutral

81 90 90 88 90 90

150+ 31 585 12 71t 60 261k 61 383k113

Threonine

Asparagine

Glutamate

Glycine

Alanine

66 90 85

26k 16 85+ 26 1032 16

1-Methylhistidine

Tryptophan

Histidine

342 31?

Methionine Men Women

Serine Men Women

62i 10 57f 7

Cysteine Men Women

1282 23 141k 28

7 7

107k 18

Arginine

49 41

49 41

49 41

90

90

90

88k 12

Ornithine

7

90

74t 12

Phenylalanine

132

90

82f 18

Tyrosine

3-Methylhistidine

88

9

36k

Citrulline

Aspartate

5

90

112f 44

Taurine ilk

n

Control subjects

Amino acid (ctmol/l) 20 11 20

105i 40 5

147rt 29

20 11 201 201 20 10 201 20 20

372-t 97 32i- 11 73* 12 82k 26 86k 21 202 13 67k 19 99k 16 7

7 6 134t 27 1342 23

32t 3Ok

75k 19 67F 4

1162 19

11 9

11 9

111 91

201

20

2602 66

ilk

19 20

562 16 792 42

9k

n

Drug-free Drugnaive 6

9

7 9 133k 30 155k 31

32+ 27k

73f 20 7Ok 11

1162 26

12k

1012 16

70+ 25

272 15

9Ok 21

82k 17

74i 19

38k 12

37Ok 96

267? 66

91? 53

54+ 13

1452 30

122

11Ok 38

Offdrug

patients

Patients

34 31

36 31

36 31

66

61

67

67

33

67

67

67

48

67

67

62

61

66

40

67

n 4

35

33

8 6 1392 20 147? 24

36i 28k

73k 25 75+ 81

1132 17

62

102i 16

64-t 141

39+ 24

86k 13

86+ 181

74i 17

48i 183

354f 87

246f 49

76k

64k 182

159f 25

lOi

105k 25

Ondrug

Table 3. Serum amino acid levels of controls and schizophrenics

n

11 12

19 12

11 12

23

21

23

22

4

23

23

23

10

23

23

21

20

23

11

23

34 acids

0.082kO.016 0.083ko.ol4

11 9

11 9

46 40

861

46 40

211+32 189+ 25

158-t 25 131+ 24

87k 16 73i- 15

256k44 232+ 44

244k 70 215k 35

558+126 569k124

52 34

208+ 34 185k 36

0.125kO.024 0.140~0.019

0.066~0.011

-

1502 55 123k 18

92+ 19 74% 13

248+ 224+

274k104 221i 64

548+ 55 499+_119

0.077+0.006 0.086~0.017

36 31

36 31

36 31

36 31

27 20

36 31

211

10 11

11 12

11 12

11 12

11 12

10 6

10 11

1. p < 0.05 vs. controls.

2. p < 0.05 vs drug-free

patients.

3. p < 0.05 vs. controls

8 drug-free

patients.

Note. Means (k 1 SD1 aregfvenseparatelyfordrug-nafveand drug-withdrawn patients. Dataforthedrug-naive and drug-withdrawn patients were combined in the multtvariate analysis of variance, sop values correspond to the entire group of drug-free patients.

0.130f0.040 0.13ako.034

49 41

212+ 57 1882 19

152k 31 135+ a

11 9

11 9

6

111 9

0.077kO.025

ratio

49 41

49 41

83i 15 75k 9

2521- 44 235k 18

25Ok 36 211k 53

561-tl38 525k 52

Phenylalaninekompeting amino acids ratio 0.126+0.025 49 Men 0.131+0.035 41 Women

0.090f0.009 0.089+0.018

amino

lag+ 35

202k

157k 25 136? 26

49 41

49 41

47 34

49 41

Tryptophankompeting amino acids ratio 0.092+0.025 90 Men &women

Men Women

Tyrosinekompeting

Lysine Men Women

Leucine Men Women

al? 14 71? 17

2Olk 40 233+ 36

Valine Men Women

lsoleucine Men Women

267k 92 22Ok 77

68Ok157 605k al

Men Women

Proline

Men Women

Glutamine

250 amino acids valine, isoleucine, leucine, and phenylalanine were higher in schizophrenic patients than in controls. In the context of phenylalanine as a precursor of tyrosine, it has been shown in experimental animals that increasing phenylalanine levels decrease the central tyrosine/phenylalanine ratio and inhibit the rate-limiting step in central monoamine synthesis, the tyrosine-hydroxylase step, thus decreasing tyrosine’s turnover for central monoamine synthesis (Maher, 1988). In the present study, however, an increase in the serum tyrosine/phenylalanine ratio was not observed, thus still leaving the influence of amino acid availability on dopamine synthesis in schizophrenia open for discussion. Roth et al. (1988) speculated that low amino acid precursor availability might be related to an enhanced rate of activity of midbrain dopaminergic neurons since it could promote up-regulation of dopaminergic receptors and increased firing, as may occur during a psychotic phase when the release of dopamine is believed to be increased. It should be noted, however, that whether central dopamine receptors are actually increased in schizophrenia is still the subject of debate since in vivo results (i.e., dopamine receptor estimations by positron emission tomography in schizophrenic patients) are equivocal in that no change (Farde et al., 1987) and an increase in dopamine receptors (Wong et al., 1986) have been observed; the conflicting results may have been partially due to methodological differences (Andreasen et al., 1988). Table 4. Serum catecholamine and blood serotonin concentrations healthy control subjects and schizophrenic patients

in

Patients Drug-free patients Drug-

Control Variable Dopamine jnmol/l) Norepinephrine Men Women Epinephrine Men Women

n

Offdrug

n

0.24t0.22

201

0.31kO.25

65

0.25kO.14

20

48 41

3.6 i2.0 4.6 f2.1

11 91

3.6 k2.1 4.3 k2.1

35 30

5.7 k2.73 4.7 f3.1

10

48 41

0.29+0.14 0.32f0.21

11 9

0.32+0.2 0.30t0.19

35 30

0.57kO.532 0.35kO.27

10 10

67

1.06+0.30

10

0.99+0.48

35

0.91+0.32

subjects

n

0.20f0.11

89

(nmol/l) 3.4 21.2 3.0 21.0

(nmol/l) 0.36kO.24 0.23kO.09

naive

Ondrug

n

10

Serotonin (~molll)

0.89f0.25

6

Note. The concentrations are means k 1 SD. Means are given separately for drug-naive patients and for drugwithdrawn (off-drug) patients. For the multivariate analysis of variance, the data for these 2 groups were combined. Therefore, the p values correspond to the entire group of drug-free patients. 1. p < 0.05 vs. healthy control subjects. 2. p 4 0.05 vs. drug-free patients. 3. p < 0.05 vs. healthy control subjects and drug-free

patients.

Changes in the serum phenylalanine/competing amino acid ratio were observed in schizophrenic patients and positively associated with tardive dyskinesia (Richardson et al., 1989); a link was made in this context to phenylketonuria since mentally retarded phenylketonurics are particularly vulnerable to tardive dyskinesia (Richard-

251 Table 5. Serum hormone

levels in controls

and schizophrenics Patients

Drug-free patients Hormone

n

Drugnaive

n

Offdrug

n

Ondrug

n

0.34kO.2

79

0.26kO.21

181

0.21f0.15

59

0.25+0.25

21

0.58+0.19

90

0.56f0.20

20

0.55kO.19

67

0.50f0.04

23

0.30f0.16 0.69f0.45

49 40

0.33+0.13 0.52f0.23

11 9

0.30+0.21 0.64kO.58

36 31

0.84t0.883 0.79kO.62

11 12

1.90+1.2

90

1.32f0.95

202

1.59k1.16

67

1.6221.67

22

108+27

90

95+20

202

96+_21

67

98522

22

2.1550.42

90

1.78t0.38

201

1.88kO.47

67

1.76kO.422

22

49 40

0.025+0.01 0.079f0.12

11 9

0.036f0.028 0.117+0.184

36 31

Control subjects

Melatonin (nmol/l) Cortisol i~mol/l) Prolactin(nmol/l) Men Women Thyrotropin (mu/I) Thyroxine (nmol/l) Triiodothyronine (nmol/l) Growth hormone Men Women

(nmol/l) 0.039+0.06 0.137+0.17

0.052+0.06 0.102kO.21

11 12

Note. The contentrations are means k 1 SD. Means are given separately for drug-naive patients and fordrugwithdrawn (off-drug) patients. For the multivariate analysis of variance, the data for these 2 groups were combined. Therefore, the p values correspond to the entire group of drug-free patients. 1. p < 0.05 vs. healthy control subjects. 2. p c 0.1 vs. healthy control subjects. 3. p < 0.1 vs. healthy control subjects; p < 0.05 with respect to drug-free subjects

son et al., 1986). In the present study, no difference was found in the phenylalanine/competing amino acid ratio among the different schizophrenic subgroups. Unfortunately, the patients were not screened specifically for tardive dyskinesia, so it was not possible to analyze this group separately. None of the serum amino acids that function as excitatory (glutamate and aspartate) or inhibitory (glycine and taurine) neurotransmitters were altered in drug-free or neuroleptic-treated patients compared to healthy subjects, as has also been observed by Gershon and Shader (1969). In neuroleptic-treated schizophrenics, Macciardi et al. (1990) observed an increase in glycine, glutamate, and serine. They speculated that these elevated amino acids might disrupt N-methyl-Baspartate (NMDA) receptors. In the present study and in that of Perry and Hansen (1985), amino acids which participate in transmethylation processes such as serine and glycine, whose disturbed metabolism might be related to a faulty transmitter metabolite production, were normal. This might be due to a different patient population since Bruinvels et al. (1980) observed a decrease in plasma serine only in patients with altered sensory perception. Plasma levels of dopamine and its metabolite, homovanillic acid, have been assessed in humans and in the rat to reflect central dopaminergic activity (Bacopoulos et al., 1979; Rao et al., 1984; Bowers and Swigar, 1987; Davidson and Davis, 1988;

252 Davila et al., 1988). Serum dopamine concentrations are generally low, making measurement problematical. Therefore, serum dopamine concentrations were analyzed with a radioenzymatic assay, for which rats were pretreated for the preparation of catechol-Gmethyltransferase to lower the blank and increase the sensitivity. Basal dopamine concentrations were higher in drug-free schizophrenic patients than in healthy controls. Serum dopamine levels of the neuroleptic-treated patients were in between those of healthy controls and drug-free patients. Previous studies have shown that long-term neuroleptic treatment elicited a decrease in dopamine turnover, as reflected in a lowering of the dopamine metabolite homovanillic acid in plasma (Pickar et al., 1984, 1986; Reynolds, 1989), but increases have also been noted (Harris et al., 1984). The latter finding might be related to the overall severity of the illness, which was correlated with plasma homovanillic acid levels (Davidson and Davis, 1988). Hyperarousal and disturbances in attention and information processing in schizophrenic patients have been attributed to increased noradrenergic activity (Hornykiewicz, 1982; van Kammen and Gelernter, 1987); both increased CSF (Jeste et al., 1984; Barbeito et al., 1984; Kemali et al., 1985) and plasma norepinephrine levels (Dajas et al., 1983, van Kammen and Antelman, 1984) have been observed in schizophrenic patients; these observations are supported by the present finding of higher serum norepinephrine and a tendency for higher epinephrine concentrations in the drug-free schizophrenic women when compared to healthy subjects. Both catecholamines were also higher in the neuroleptic-treated than in the drug-free schizophrenic men, as has also been observed by Naber et al. (1980b). A point made by van Kammen et al. (1989) in this context may also be applicable-namely, that clinically stable patients are the ones most likely to be able to undergo drug-withdrawal periods and the lower serum norepinephrine of these patients compared to neuroleptic-treated patients might reflect their clinical stability as well as the effect of neuroleptic withdrawal. There are reports casting doubt upon the relevance of serum norepinephrine levels to psychopathology (Kemali et al., 1982); however, others point to a significant correlation between serum measures of norepinephrine and measures of psychopathology (Cooper et al., 1985) and also to a significant correlation between serum and CSF norepinephrine (Ziegler et al., 1977), which seems important for the interpretation of these results. Blood serotonin concentrations were somewhat, albeit not significantly, elevated in untreated schizophrenic patients; however, the standard deviation was twice as high in patients as in controls. Previous studies have shown large intraindividual fluctuations in blood serotonin that could be traced to the level of psychopathology in the individual patients (Rao and Braunig, 1989). Elevated blood serotonin levels have been associated with auditory hallucinations, lack of insight, and disorganization; lowered serotonin levels, with depressive mood and suicidality (for reviews, see Asberg and Nordstrom, 1988; Bleich et al., 1988). The present study did not include suicidal patients. The finding by DeLisi et al. (1981) of an increase in blood serotonin in schizophrenic patients might be due to the inclusion of a different patient population compared to the one studied here. Concomitant with the observed increase in serum dopamine, one would expect a decrease in prolactin, since the pituitary lies outside the blood-brain barrier and since

253 the administration of dopamine decreases serum prolactin and TSH (Heinen et al., 198 1). Although serum dopamine levels were increased in the drug-free schizophrenic patients compared to healthy subjects, prolactin concentrations did not differ between the groups, as has also been found in most other studies (for review, see Meltzer, 1984). However, there was a trend for lowered TSH that was paralleled by lower thyroxine and triiodothyronine levels. This is in keeping with the observation by McLarty et al. (1978) that the single most common biochemical abnormality in psychiatric patients was a subnormal thyroid hormone level. Basal cortisol and growth hormone levels were normal in the schizophrenic patients studied here, as has been shown before (Meltzer, 1984). It is well known that prolactin increases after neuroleptic therapy (Brown and Laughren, 1981); however, prolonged neuroleptic administration leads to “tolerance” to the prolactin-elevating effect, with 50% of long-term neuroleptic-treated patients showing prolactin in the normal range (Naber et al., 1980a; &man and Axelsson, 1980; Davis et al., 1984). Thus, it is not surprising that no significant prolactin increase was found in neuroleptic-treated women at 8 a.m., although a significant increase was found in men. The increase for women did not become significant until 10 a.m., probably reflecting a slight phase delay in prolactin’s acrophase that was more pronounced in women than in men. There was a good correlation between serum neuroleptic units and serum prolactin concentrations in men and women. Although challenge tests with t_-dopa, insulin, or apomorphine have shown neuroleptic-treated patients to react differently with respect to growth hormone compared to drug-free schizophrenic patients (Davis et al., 1984) basal serum growth hormone concentrations were similar in medicated and unmedicated schizophrenic patients (Saldanha et al., 1972), as was also seen in the present study. Neuroleptics had no influence on most amino acids, on basal cortisol, or on thyroid hormones, as has been noted before (Wode-Helgodt et al., 1977; Naber et al., 1980~). The schizophrenic patients studied here had lower melatonin levels. This “low melatonin syndrome” might be biochemically related to the elevated dopamine, since low melatonin disinhibits hypothalamic dopamine release (Zisapel et al., 1985) and thus might contribute to the “hyperdopaminergia” in schizophrenia. Fanget et al. (1989) noted that melatonin levels in schizophrenic patients tend to be lower than those in healthy subjects, but their patients were neuroleptic-treated. The major findings of this study are as follows: (1) Neuroleptic-naive schizophrenic patients did not differ from neuroleptic-withdrawn patients with respect to blood levels of the biochemical variables studied. (2) Drug-free schizophrenic patients showed changes in the availability of amino acid precursors for monoamine synthesis, which might be involved in dopamine receptor up-regulation. (3) Levels of serum dopamine were increased in drug-free schizophrenic patients as compared to controls; the decreased levels of melatonin noted in these patients could also have promoted a disinhibition of dopamine release. (4) This serum dopamine increase exerted an inhibitory effect on the pituitary-thyroid axis in lowering TSH and thyroid hormone levels. (5) Neuroleptic-treated patients had elevated levels of asparagine, citrulline, and 3-methylhistidine-compounds that do not appear to be directly related to amino acid availability for monoamine synthesis; in addition, these patients showed the well-known elevation of prolactin and an increase in norepinephrine and epinephrine.

254 Acknowledgments. The authors thank Dr. G.M. Brown, Department of Neurosciences, McMaster University, Hamilton, Ontario, Canada, for providing the melatonin antiserum and Dr. W.A.Brown, VA Center, Providence, RI, for expert advice during the preparation of this article. The project was supported in part by the Deutsche Forschungsgemeinschaft and the Ministry of North-Rhein-Westfalia, Germany.

References Andreasen, N.C.; Carson R.; Diksic, M.; Evans, A.; Farde, L.; Gjedde, A.; Hakim, A.; Lal, S.; Nair, N.; Sedvall, G.; Tune, L.; and Wong, D. Workshop on schizophrenia, PET, and dopamine D2 receptors in the human neostriatum. Schizophrenia Bulletin, 14:471-484, 1988. Asberg, M., and Nordstrom, P. Biological correlates of suicidal behavior. In: Moller, H.-J.; Schmidtke, A.; and Welz, R., eds. Current Issues of Suicidology. Berlin: Springer, 1988. pp. 223-241. Bacopoulos, N.G.; Hattox, S.E.; and Roth, R.H. 3,4_Dihydroxyphenylacetic acid and homovanillic acid in rat plasma: Possible indicators of central dopaminergic activity. European Journal

of Pharmacology,

56:225-236,

1979.

Barbeito, L.; Lista, A.; Silbeira, R.; and Dajas, F. Resting urinary catecholamine excretion in schizophrenics: Methodology and interpretation of results. Biological Psychiatry, 19:14191425,

1984.

Bjerkenstedt, L.; Edman, G.; Hagenfeldt, L.; Sedvall, G.; and Wiesel, F.-A. Plasma amino acids in relation to cerebrospinal fluid monoamine metabolites in schizophrenic patients and healthy controls. British Journal of Psychiatry, 147:276-282, 1985. Bleich, A.; Brown, S.L.; Kahn, R.; and van Praag, H.M. The role of serotonin in schizophrenia. Schizophrenia Bulletin, 14:297-315, 1988. Bowers, M.B., Jr., and Swigar, M.E. Acute psychosis and plasmacatecholamine metabolites. Archives

of General

Psychiatry,

44: 190, 1987.

Brown, M.L., and Jenner, D.A. Novel double-isotope technique for enzymatic assay of catecholamines, permitting high precision, sensitivity and plasma sampling capacity. Clinical Science, 61:591-598, 1981. Brown, W.A., and Laughren, T.P. Low serum prolactin and early relapse following neuroleptic withdrawal. American Journal of Psychiatry, 138:237-239, 198 1. Bruinvels, J.; Pepplinkhuizen, L.; van Tuijl, H.R.; Moleman, P.; and Blom, W. Role of serine, glycine, and the tetrahydrofolic acid cycle in schizoaffective psychosis: A hypothesis relating porphyrin biosynthesis and transmethylation. In: Usdin, E.; Sourkes, T.L.; and Youdim, M.B.H., eds. Enzymes and Neurotransmitters in Mental Disease. New York: John Wiley & Sons, Ltd., 1980. pp. 139-153. Cooper S.J.; Kelly, J.G.; and King, D.J. Adrenergic receptors in depression: Effects of electroconvulsive therapy. British Journal of Psychiatry, 147:23-29, 1985. Creese, I.E., and Snyder, S.H. A simple and sensitive radioreceptor assay for antischizophrenic drugs in blood. Nature, 270:180-182, 1977. Dajas, F.; Barbeito, L.; Martinez-Pesquera, G.; Lista, A.; Puppo, D.; and Puppo-Touriz, H. Plasma noradrenaline and clinical psychopathology in schizophrenia. Neuropsychobiology, 10:70-74,

1983.

Davidson, M., and Davis, K.L., A comparison of plasma homovanillic acid concentrations in schizophrenic patients and normal controls. Archives of General Psychiatry, 45:561-563, 1988. Davila, R.; Manero, E.; Zumarraga, M.; Andia I.; Schweitzer, J.W.; and Friedhoff, A.J. Plasma homovanillic acid as a predictor of response to neuroleptics. Archives of General Psychiatry,

451564-567,

1988.

Davis, J.M.; Vogel, C.; Gibbons, R.; Pavkovic, 1.; and Zhang, M. Pharmacoendocrinology of schizophrenia. In: Brown, G.M.; Koslow, S.H.; and Reichlin, S.R., eds. Neuroendocrinology and Psychiatric Disorder. New York: Raven Press, 1984. pp. 29-53. DeLisi, L.E.; Neckers, L.M.; Weinberger, D.R.; and Wyatt, R.J. Increased whole blood serotonin concentrations in chronic schizophrenic patients. Archives of General Psychiatry, 38:647-650,

1981.

255 Expert Panel. Report of the National Cholesterol Education Program Expert Panel on detection, evaluation and treatment of high blood cholesterol in adults. Archives of Internal Medicine, 148:36-69, 1988. Fanget, F.; Claustrat, B.; Dalery, J.; Brun, J.; Terra, J.-L.; Marie-Cardine, M.; and Guyotat, J. Nocturnal plasma melatonin levels in schizophrenic patients. Biological Psychiatry. 25:499501, 1989. Farde, L.; Wiesel, F.-A.; Hall, H.; Halldin, C.; Stone-Elander, S.; and Sedvall, G. No D2 receptor increase in a PET study of schizophrenia. Archives of General Psychiatry, 44:671-672, 1987. Fernstrom, F.K., and Faller, D.V. Neutral amino acids in the brain: Changes in response to food ingestion. Journal of Neurochemistry, 30: 153 l-1538, 1978. Fernstrom, F.K., and Wurtman, R.J. Brain serotonin content: Physiological regulation by plasma neutral amino acids. Science, 178:414-416, 1972. Gambert, S.R., and Tsitouras, P.D. Effect of age on thyroid hormone physiology and function. Geriatric Bioscience, 33:360-365, 1985. Gershon, E.S., and Shader, R.I. Screening for aminoacidurias in psychiatric inpatients. Archives of General Psychiatry, 2 1:82-88, 1969. Harris, P.Q.; Brown, S.J.; Friedman, J.; and Bacopoulos, N.G. Plasma drug and homovanillic acid levels in psychotic patients receiving neuroleptics. Biological Psychiatry, 19:849860, 1984. Heinen, E.; Herrmann, J.; Konigshausen, T.; and Kriiskemper, H.L. Secondary hypothyroidism in severe non-thyroidal illness. Hormone and Metabolic Research, 13:284-288, 1981. Hornykiewicz, 0. Brain catecholamines in schizophrenia-A good case for noradrenaline. Nature, 299:484-486, 1982. Jeste, D.V.; Doongaji, D.R.; and Linnoila, M. Elevated cerebrospinal fluid noradrenaline in tardive dyskinesia. British Journal of Psychiatry, 144:177-180, 1984. Kemali D.; DelVecchio, M.; and Maj, M. Increased noradrenaline levels in CSF and plasma of schizophrenic patients. Biological Psychiatry, 17:71 I-717, 1982. Kemali, D.; Maj, M.; Iorio, G.; Marciano, F.; Nolfe, G.; Galderisi, S.; and Salvati, A. Relationship between CSF noradrenaline levels, C-EEG indicators of activation and psychosis ratings in drug-free schizophrenic patients. Acta Psychiatrica Scandinavica, 7 1: 19-24, 1985. Macciardi, F.; Lucca A.; Catalano, M.; Marino C.; Zanardi, R.; and Smeraldi, E. Amino acid patterns in schizophrenia: Some new findings. Psychiatry Research, 32:63-70, 1990. Maher, T.J. Modification of synthesis, release and function of catecholaminergic systems by phenylalanine. In: Huether, G., ed. Amino Acid Availability and Brain Function in Health and Disease. (NATO ASI Series H: Cell Biology) Vol. 20. Berlin: Springer-Verlag, 1988. pp. 20 l-206. McLarty, D.G.; Ratcliffe, W.A.; Ratcliffe, J.G.; Shimmins, J.G.; and Goldberg, A. A study of thyroid function in psychiatric in-patients. British Journal of Psychiatry, 133:21 l-218, 1978. Meltzer, H.Y. Neuroendocrine abnormalities in schizophrenia: Prolactin, growth hormone, and gonadotropins. In: Brown, G.M.; Koslow, S.H.; and Reichlin, S.R., eds. Neuroendocrinology and Psychiatric Disorder. New York: Raven Press, 1984. pp. l-28. Meltzer, H.Y. Biological studies in schizophrenia. Schizophrenia Bulletin, 13:77-l 11, 1987. Moore, S., and Stein W.H. Chromatography of amino acids on sulfonated polystyrene resins. Journal of Biological Chemistry, 192:663-68 1, 195 1. Naber D.; Ackenheil, M.; Laakman, G.; Fischer, H.; and von Werder, K. Basal and stimulated levels of prolactin, TSH and LH in serum of chronic schizophrenic patients, long-term treated with neuroleptics. Pharmakopsychiatry, 13:325-330, 1980a. Naber, D.; Finkbeiner, C.; Fischer, B.; Zander, K.J.; and Ackenheil, M. Effect of long-term neuroleptic treatment on prolactin and norepinephrine levels in serum of chronic schizophrenics: Relation to psychopathology and extrapyramidal symptoms. Neuropsychobiology. 6:181-189, 1980b. Naber, D.; Steinbock, H.; and Greil, W. Effects of short- and long-term neuroleptic treatment on thyroid function. Progress in Neuro-Psychopharmacology, 4: 199-206, 1980~.

256 &man, R., and Axelsson, R. Prolactin response to neuroleptics: Clinical and theoretical implications. Journal of Neural Transmission, Suppl. 17, 1980. Perry, T.L., and Hansen, S. Interconversion of serine and glycine is normal in psychotic patients. Psychiatry Research, 15: 109-l 13, 1985. Peuler, J.D., and Johnson, G.A. Simultaneous single isotope radioenzymatic assay of plasma norepinephrine, epinephrine and dopamine. Lzfe Sciences, 21:625-636, 1977. Pickar, D.; Labarca, R.; Doran, A.R.; Wolkowitz, O.M.; Roy, A.; Breier, A.; Linnoila, M.; and Paul, SM. Longitudinal measurement of plasma homovanillic acid levels in schizophrenic patients. Archives of General Psychiatry, 43:669-676, 1986. Pickar, D.; Labarca, R.; Linnoila, M.; Roy, A.; Hommer, D.; Everett, D.; and Paul, S.M. Neuroleptic-induced decrease in plasma homovanillic acid and antipsychotic activity in schizophrenic patients. Science, 225~954-957, 1984. Rao, M.L. Modification of the radioreceptor assay technique for estimation of serum neuroleptic drug levels leads to improved precision and sensitivity. Psychopharmacology, 90:548-553, 1986. Rao, M.L., and Braunig, P. Peripheral serotonin and catecholamine levels and suicidal behavior. In: Gershon, S., and Lerer, B., eds. New Directions in Affective Disorder. Berlin: Springer, 1989. pp. 320-325. Rao, M.L., and Fels, K. Beeinflussen Tryptophan und Serotonin beim Menschen die Melatonin-Ausschtittung und damit die Funktion des “Regulators der Regulatoren” (Zirbeldrtise). In: Stille, G.; Wagner, W.; and Herrman, W.M., eds. Fortschritte in der Pharmakotherapie. Basel: Karger, 1987. pp. 87-99. Rao, M.L.; Gross, G.; and Huber G. Altered interrelationship of dopamine, prolactin, thyrotropin and thyroid hormone in schizophrenic patients. European Archives of Psychiatric and Neurological Sciences, 234:8- 12, 1984. Rao, M.L., and Mager, T. Influence of the pineal gland on pituitary function in humans. Psychoneuroendocrinofogy, 12: 141-147, 1987. Reynolds, G.R. Beyond the dopamine hypothesis: The neurochemical pathology of schizophrenia. British Journal of Psychiatry, 155:305-316, 1989. Richardson, M.A.; Haughland, G.; Pass, R.; and Craig, T.J. The prevalence of tardive dyskinesia in a mentally retarded population. Psychopharmacology Bulletin, 22~243-249, 1986. Richardson, M.A.; Suckow, R.; Whittaker, R.; Boggiano, W.; Sziraki, I.; Kushner, H.; and Perumal, A. The plasma phenylalanine/ large neutral amino acid ratio: A risk factor for tardive dyskinesia. Psychopharmacology Bulletin, 25147-51, 1989. Roth, J., and Grunfeld, C. Mechanism of action of peptide hormones and catecholamines. In: Wilson, J.D., and Foster, D.W. Williams, eds. Textbook of Endocrinology. 7th ed. Philadelphia: W.B. Saunders, 1985. pp. 76-122. Roth, R.H.; See-Ying, T.; Bradberry, C.W.; Karasic, D.H.; and Deutch, A.Y. Precursor availability and function of dopaminergic neurons. In: Huether, G., ed. Amino AcidAvailability and Brain Function in Health and Disease. (NATO AS1 Series H: Cell Biology) Vol. 20. Berlin: Springer-Verlag, 1988. pp. 19 l-200. Saldanha, V.F.; Havard, C.W.H.; Bird, R.; and Gardner, R. The effect of chlorpromazine on pituitary function. Clinical Endocrinology, 1: 173-180, 1972. Schneider, K. Khnische Psychopathologic. 12th ed. Stuttgart: Thieme Verlag, 1980. Spitzer, R.L.; Endicott, J.; and Robins, E. Research Diagnostic Criteria: Rationale and reliability. Archives of General Psychiatry, 35:773-782, 1978. van Kammen, D.P., and Antelman, S. Impaired noradrenergic transmission in schizophrenia? Ltfe Sciences, 34: 1403-1413, 1984. van Kammen, D.P., and Gelernter, J. Biochemical instability in schizophrenia: I. The norepinephrine system. In: Meltzer, H.Y., ed. Psychopharmacology: The Third Generation of Progress. New York: Raven Press, 1987. pp. 745-751.

257 van Kammen, D.P.; Peters, J.; van Kammen, W.B.; Nugent, A.; Goetz, K.L.; Yao, J.; and Linnoila, M. CSF norepinephrine in schizophrenia is elevated prior to relapse after haloperidol withdrawal. Biological Psychiatry, 26:176-188, 1989. Wilke T.J. Influence of age and sex on the concentration of free thyroxine in serum and on the free thyroxine/total thyroxine ratio. Clinical Chemistry, 29:1428-1430, 1983. Wode-Helgodt, B.; Eneroth, P.; Fyro, B.; Gullberg, B.; and Sedvall, G. Effect of chlorpromazine treatment on prolactin levels in cerebrospinal fluid and plasma of psychotic patients. Acta Psychiatrica Scandinavica, 56:280-293, 1977. Wong, D.R.; Wagner, H.N.; Tune, L.E.; Dannals, R.F.; Pearlson, G.D.; Links, J.M.; Tamminga, C.A.; Broussolle, E.P.; Raver, H.T.; Wilson, A.A.; Toung, J.K.T.; Malat, J.; Williams, J.A.; O’Tuama, L.A.; Snyder, S.H.; Kuhar, M.J.; and Gjedde, A. Positron emission tomography reveals elevated D, dopamine receptors in drug-naive schizophrenics. Science, 234:1558-1563, 1986. Yamada, T.; Naka, M.; Komiya, 1.; Ichikawa, M.; Aizawa, T.; Hashizume, K.; Takasu, N.; and Watanabe, T. Age-related alterations of pituitary-thyroid function in normal female subjects and in female patients with simple goitre. Acta Endocrinologica, 107:346-351, 1984. Ziegler, M.G.; Lake, C.R..; Wood, J.H.; Brooks, B.R.; and Ebert, M.H. Relationship between norepinephrine in blood and cerebrospinal fluid in the presence of a blood cerebrospinal fluid barrier for norepinephrine. Journal of Neurochemistry, 28:677-679, 1977. Zisapel, N.; Egozi, Y.; and Laudon, M. Circadian variations in the inhibition of dopamine :-elease from adult and newborn rat hypothalamus by melatonin. Neuroendocrinology, 40: 102108, 1985.