- Email: [email protected]
Uptake and utilization of [beta-11C] 5-hydroxytryptophan in human brain studied by positron emission tomography
215
Psychiatry Research: Neuroimaging, 451215-225 Elsevier
Uptake and Utilization of [Beta-W]5-Hydroxytryptophan in Human Brain Studied by Positron Emission Tomography Lars Reibring, Hans Agren, Per Hartvig, Joakim Tedroff, Peter Bjurling, Tor Kihlberg, and Bengt Lflngstrijm
Hans Lundqvist,
Received May 8, 1992; revised version received September 2, 1992; accepted November I I, 1992. Abstract. The immediate precursor in the serotonin synthetic route, 5hydroxytryptophan (5HTP), labeled with 142 in the j3 position, has become available for studies using positron emission tomography (PET) to examine serotonin formation in human brain. Normalized uptake and intracerebral utilization of tracer amounts of [P-11C]S-HTP were studied twice in six healthy male volunteers, three of them before and after pharmacological pretreatments. The kinetic model defines regional utilization as the relative regional radioactivity accumulation rate. Repeat studies showed good reproducibility. Pretreatments with benserazide, p-chlorophenylalanine (PCPA), and unlabeled 5-HTP all significantly increased uptake of [P-liC]S-HTP. The utilization rates in both striatal and frontal cortex were higher than those in the surrounding brain, indicating that PET studies using [P-llC]S-HTP as a ligand quantitate selective processes in the utilization of 5-HTP. We tentatively interpret uptake and utilization as a measure of brain serotonin turnover, the selectivity of which was shown by pharmacological interventions in vivo. Key Words. Serotonin,
frontal cortex, striatum,
benserazide, p-chlorophenyl-
alanine. The monoamine hypotheses of the pathogenesis of several psychiatric disorders demand qualitative and quantitative in vivo studies of transmitter synthesis, transmitter turnover, and receptor function. Positron emission tomography (PET), which permits the use of various r*C tracers, has emerged as a focus for studies on all aspects of neurotransmission. The immediate precursor in the serotonin synthetic route, Shydroxytryptophan (SHTP), labeled with rrC in the p position, is now available for direct studies on precursor uptake mechanisms and for attempts to estimate presynaptic serotonin synthesis (Agren et al., 1991; Hartvig et al., 1992). An alternative approach to the study of brain uptake mechanisms is through the measurement of selective accumulation in brain of [rlC]-methyltryptophan (Diksic et al., 1991) but this substance is not transformed to serotonin.
Lam Reibring, M.D., is Clinical Researcher, and Hans digren, M.D., Ph.D., is Associate Professor, Department of Psychiatry, University Hospital, Uppsala. Per Hartvig, Pharm.D., Ph.D., is Professor, University Hospital Pharmacy. Joakim Tedroff, M.D., Ph.D., is Assistant Professor, Department of Neurology, University Hospital. Hans Lundqvist, Ph.D., Peter Bjurling, Ph.D., and Tor Kihlberg, Ph.D., are Researchers, and Bengt LAngstrom, Ph.D., is Professor, PET Center, Uppsala University. (Reprint requests to Dr. L. Reibring, Dept. of Psychiatry, University Hospital, S-751 85 Uppsala, Sweden.) 0165-1781/92/$05.00
@ 1992 Elsevier Scientific
Publishers
Ireland
Ltd.
In the present report, results are presented from PET studies that examined brain utilization of 5-HTP in healthy volunteers before and after the administration ot pharmacological agents chosen to alter certain aspects of monoamine metabolism-~ the peripheral r-aromatic acid decarboxylase (AADC) inhibitor benserazide, the central inhibitor of 5-HTP synthesis y-chlorophenylalanine (PCPA), and various doses of unlabeled 5-HTP. In addition, three pairs of identical experiments on uptake and utilization of 5-HTP were carried out to test the reproducibility of findings.
Methods Six healthy volunteers took part in the study. The subjects were mentally healthy men (physicians and medical students) between 28 and 43 years of age (mean = 37 years, SD = 6). Table 1 presents pertinent demographic characteristics of the subject and experimental details. In one triple experiment, the radioactive dose of 5-HTP was given to subject D simultaneously with intravenous doses of 0, 1, and 10 mg, respectively, of unlabeled 5-HTP. We also studied S-HTP brain utilization after benserazide, 50 mg given orally to subject E 2 hours before a second scan, and after PCPA, 1 g given in two oral doses to subject F within 24 hours before a second scan. Another three volunteers (subjects A, B, and C) were scanned twice to study reproducibility; the imaging sessions being separated by 2 hours, 2 weeks, and 7 weeks, respectively. Subjects.
Table 1. Description of 6 male healthy volunteers investigated with (p11C]5hydroxyttyptophan (5-HTP) Age ID
Pretreatment
Al
Baseline
A2
None
Bl
Baseline
82
None
Cl
Baseline
C2
None
Dl
Baseline 5-HTP, 10 mg, i.v.
El
Baseline
E2
Benserazide, p.o.
Fl
Baseline
F2
p-Chlorophenylalanine,
Date (month-
(W
(MW
43
78
151 142
21:02
06-22
37
108
70
l&18
OS-02
40
11:37
05-16
39
76
164
1:30
02-02
28
70
80
43
72
30 p.o. 37k6
Time of day
(yr)
1 mg, i.v.
SHTP,
D3
f SD
Radioactivity
190
D2
Mean
Weight
42
day) 08-22
13:16
03-22
lo:58
09-21
100
12:50
09-21
108
14:51
09-21 04-l 1
30
17:37
151
21:51
02-03
132
18:17
01-29
52 80113
(hr) 19~18
15:31
_____
05-04
123zk48
Radiochemistry. The radionuclide “C was obtained from the Tandem accelerator at The Svedberg Laboratory of Uppsala University and used in the synthesis of [P-*rC]S-HTP, as described elsewhere (Bjurling et al., 1989). The “C was introduced in the p position in the chain (Fig. l), thus yielding a serotonin molecule labeled with rrC after in vivo decarboxylation by AADC (Lovenberg et al., 1962). After analysis to ascertain chemical and radiochemical purity, a 0.1 mol/l phosphate buffer solution at pH 7.4 of [P-rrC]S-HTP was
217
Fig. 1. Radioactivityin arterial blood and ratio of brain tissues and blood in a healthy volunteer 3oc 25
-‘Arterial
blood
--o-
Ref. tlsslJe/A. blood
--e
Caudate n./A. blood
, 1.5
Time (min) Equilibrium is not reached during the imaging time span. The ~-11C)5-Hydroxytryptophan molecule is illustrated with its radioactive atom.
filtered through a 0.22 ,um filter and given intravenously was 30-164 MBq, corresponding to 9-50 pg of 5-HTP.
to the patient. The radioactive dose
Blood Samples. To serve as reference values, arterial samples were collected from subject D. Radioactivity was measured in whole blood in a well counter connected to the main computer (VAX) that controlled the PET camera and adjusted to time of blood drawing.
The PET scan was performed with the patient lying with his head fixed in a Scanditronix PC 384-3B tomograph. Imaging started immediately after intravenous administration of the “C-labeled 5-HTP and lasted for 40-60 minutes. This procedure delivers about 18 consecutive exposures of three horizontal brain slices. With guidance from a computed tomographic (CT) scan, the top-most level 3 was set to include the basal ganglia, parallel to and approximately 50 mm above the canthomeatal line. Level 2 was 13.5 mm down from level 3, and level 1 was a further 13.5 mm down in an attempt to reach cerebellar regions. Regions of interest (ROIs) were determined from a CT scan-derived outline map. The following ROIs were selected for further analysis: whole brain at level 2 (total area of brain slice); dorsoluteralprefrontal cortex (DLPFC; corresponding to Brodmann areas 9 and 10) at levels 2 and 3; medial PFC (MPFC; corresponding to Brodmann area 32) at levels 2 and 3; cuudute nucleus at level 3; and lentlform nucleus (including putamen and globus pallidus) at level 3. Measurements on the right and left sides were averaged. PET Procedure.
Measured radioactivities per cm3 were corrected for physical decay to the time of administration and given as uptake after correction for the administered dose per gram of body weight. Uptake data from ROIs were transferred to a Macintoshru computer for mathematical treatment by various procedures in programs such as ExcelTM, KaleidaGraphTM, StatView IIru, and SuperAnovarM.
Calculations.
Analysis of Uptake Data. Focusing on intracerebral
radioactivity, we excluded exposures before 5 minutes of run time because the initial blood radioactivity peak may conceal intracerebral radiation (Fig. 1). Averaged regional uptake values from 5 minutes to the end of
21X
the experiment (Table 3).
were termed
“averaged
uptake”
and were compared
by analysis
of covariance
Analysis of Change of Uptake over Time. Change of radioactivity over time was observed in certain brain regions. Several approaches are applicable for the quantification of this temporal change. Abbreviations are explained in Table 2.
Table 2. Abbreviations is the concentration
Uptake Averaged
uptake
of radioactivity in reference tissue or target tissue at time t.
is calculated
from exposures
Concentration
of radioactivity
CP
after 5 minutes from injection
of radioactivity.
t
in free plasma at time
t
CT
Total concentration
of radioactivity
in target tissue at time
CR
Total concentration
of radioactivity
in reference
CE
Total concentration
of extracellular
radioactivity
CET
Concentration
of extracellular
radioactivity
in target tissue at time
CER
Concentration
of extracellular
radioactivity
in reference
C/T
Concentration
of intracellular
tissue at time in reference
t
tissue and target
t
tissue at time
of intracellular
t
radioactivity
in target tissue at time
radioactivity
in reference
t
C/R
Concentration
K
Radioactivity
kl k2 k,
Fractional
rate constants
for transfer
of radioactivity
in target tissue
'1 4 13
Fractional
rate constants
for transfer
of radioactivity
in reference
E
Blood volume
t
Time, where
“accumulation
fraction
rate coefficient”
t
tissue at time tissue at time
t
in target tissue tissue
in tissue
t = 0 at time of injection
Note. All radioactivity measures above are normalized as for body weight, injected radioactivity, and physical decay of isotope to time of injection.
Table 3. Positron emission tomographic measures of averaged uptake and interest (utilization measures X 0.0001)
Volunteer Premedlcatlon Al
Bswllne
Whole brain Level 2
Dwsolateral prefrontal cortex Level 2
auptake
a.uptake
Medial Level
Level 3
K
k3-I3
a.uptaks
K
k3-I3
a.uptake
K
0.653
0.836
-7
-6
0.692
-7
-6
0.933
6
A2 None
0.863
0.880
-11
-12
0.878
8
9
0.950
-6
29
29
0.792
6
7
0.667
-1
12
12
0.935
6
6
0.885
9
1
1
1.075
5
4
1.013
11
Bl Baseline
0.699
0.7521
82 None
0.897
0.9581
Cl Baseline
1.0441
1.059
C2 None
0.752
Dl Baseline
0.6Oi
D2 5-HTP. 1 mg, I.“. 03 SHTP,
10 mg, I.V.
El Baseline
0.842
1
0.756 0.776
1
0.885
0.879
0.905
0.933
0.852
E2 Benserazide, 50 mg. p.o.
1.620
Fl Baseline
0.83i
F2 p-Chlorophenylalanine,
0.922.
3
1.504
0.909
-2
-2
0.764
-43
-39
0.660
I: 1 ! -9
I:
0.821 2
I
-9
2
0.877
9
0
-1
0.906
-9
20
20
0.954
3
? -3 2
I:
2
1-17
1
-3
1
2 Z-15
I 1
1.484
-22
0.802
-1
0.905
0.925
1 9,P.0. Mean of baselines
0.903
0.893
1 23
-2
-1
11 21
I
8
I
0.738
37
0.939
-37
0.978
13
-9
1.023
7
1
0.661
-45
-25
1
1.674 3
11
-2
0.831
30
341
321
1.068
39
7
6
0.910
-6
1 1
Note. Baselines denoted by bold characters. Significant differences calculated by analyses of covariance:
1. p < 0.05.
219 Approach A. An “accumulation rate coefficient” K is easily estimated for each ROI by plotting the ratio of uptake in target tissue Cr divided by uptake in a large neutral reference area Ca (whole brain at level 2), as a function of ordinary time t. If the blood-biased exposures before 5 minutes are excluded, the plots are approximately linear. K is the slope of the obtained regression line. Slope differences were tested by analysis of covariance (Table 3). Approach B. An alternative way to interpret this linear increase in radioactivity would be to apply a three-compartment model containing plasma C,, free plus nonspecifically bound ligand concentrations in extracellular space C, and intracellular space C,. An easily delineated region that is free from 5-HTP utilization and can be used as reference tissue is not readily avaifable. Thus, a model with reference tissue including tracer utilization I, has to be accepted (Fig. 2). The illustrated symmetric model gives
CT=&.Cp+(l
- &)(CET+
(1)
CIT)
or, after manipuIation (2)
C~-t’C~=(1-&e)(C~+C~T)
E is the relative blood volume fraction in tissue. The ratio of Eq. (2) and the corresponding equation for reference tissue gives the following formulation referring to Fig. 2.
CT -
ecp
CR - ecp
=
CET
CET + EE3 So’Cmdt
+ CIT
cER+t&T
=
CER
+ t lo’ c.sRdt
(3)
=
utilization of radioactiveShydroxytryptophan (5HTP) in various regionsof prefrontal cortex 2
Level 3
k3-la
a.uptake
K
Caudate nucleus k,-4
Lentiform nucleus
a.uptake
K
ka-13
a.uptake
K
ks-G
6
0.697
33
37
0.906
62
66
0.983
33
34
-6
0.845
-18
-18
0.915
70
72
1.013
21
22
-2
0.943
SO
49
0.607
46
46
0.621
0
7
1.047
-32
-31
0.594
71
29
3
10
0.993
47
43
1.126
44
43
36
0.732
41
41
0.763
73
57
-35 12
0.627 0.862
36 29
31 27
0.696 0.928
50 37
0.861
2
2
0.963
81
0.806
33
31
1.020
93
101
0.960
22
25
1.280
46
54
1.293
6 -46 12
1 1 I
1.625
1
1 I
1
24 90 ] 2 40
53
55
46
53
28
0.602
21
19
0.768
115
106
0.962
92
66
35
0.914
31
26
0.960
82
75
1.021
95
86
-7
0.895
21
19
0.885
75
72
0.877
51
48
2.p<0.01.3.p<0.001.
220 If the time expression ~C’,dt, C6a = (C,&r/C,, and the exposures during the inmal 5 minutes are ignored, Eq. (3) is approximately reduced to CT -=CR
Assuming k, <
ICI
j-0’ e-_(~z+~d(-)&
lt
j-0’e-(h+b)(t-S)&
(
\ So’CERdt \
’ + (” - 13) CR
)
(4)
I3 << k,, k2 = I2 and CKK= CR further reduces Eq. (4) to
(5) where F is a constant. Eq. (5) has been applied to [lrC]~-DOPA utilization calculation (Hartvig et al., 1991). As for S-HTP, none of these assumptions are justified. However, if the extracellular space is considered to be a homogeneous compartment all over the brain, then Eq. (4) is alternatively reduced to CT CR
=
1+
(k3 -
13)-
ra’ C&t
(6)
Cl2
jC,dt cannot be measured by PET, and for this reason the integral is changed tolC,dt. Finally, Cr/ CR is plotted as ordinate against /CRdt/ CR, yielding k, - 1, as the slope of the linear regression line. The slope reflects the rate of specific utilization of serotonin precursor, assuming specific utilization in reference tissue to be at a zero level. The slope differences were tested by analysis of covariance (Table 3).
Fig. 2. Compartmental tissue
model including specific utilization
Reference tissue
in reference
Target tissue
CT and CR are measurable by positron emission tomography
Results The averaged uptake values from baselines showed no obvious intracerebral differences, indicating a homogeneous blood-brain barrier (BBB) transport on the basis of [iiC]S-HTP and PET. The uptake ratio plots were clearly rectilinear 5 minutes after injection, with evenly distributed residuals (Fig. 3). Curve fitting based on exponential or other functions did not change the distribution of residuals. Striatal utilization was highest, while cortical utilization was at reference tissue level, except for the medial PFC at level 3. Although there were 57 comparisons of uptake and utilization measures from the same ROIs in the dual experiments on the three
221
Fig. 3. Consecutiveuptake values from dorsolateral prefrontal cortex (PFC) at level 2 and lentiformnucleus 1.6 'A
.
_
DorsolateralPFC level2
1.4 -
1.4 1.2 - -
*:
Lentifonnnucleus * *.
. **
.
.
Data points were smoothed by a 30% weighted curve fit (without assumption of any m~hematicai model). From 5 minutes after injection of radioactive ligand, the smoothed fitted lines are clearly r~tilinear, with an even dist~bution of residuals
volunteers who did not receive additional pharmacological challenges, only two significant differences could be detected, which suggests good reproducibility. The correlation between K and k,-Z, was extremely high and strongly significant (r2 = 0.99), illustrating that JCRdz/ CR is almost equal to true time in the case of 5-HTP. Very weak correlations were observed between k3-I3 or K and averaged uptake values (r* = 0.003 and 0.002, respectively), suggesting that k3-Z3 or K and averaged uptake are orthogonal variables and independent of each other. A dose~ependent increase in averaged uptake of labeled S-HTP was seen with the injection of increasing amounts of unlabeled 5-HTP. Similar alterations followed pretreatment with PCPA. Pretreatment with 50 mg of benserazide caused a general increase in averaged uptake. Table 3 shows the effects of the pharmacological interventions.
Discussion The physiological formation of serotonin is a complex process. The use of PET to study this process will involve several aspects, i.e., tryptophan transport across the BBB, uptake of t~ptophan into serotonergic nerve terminals, hydroxylation to 5-HTP by the rate-l~iting tryptophan hydroxylase, and finally decarboxylation to serotonin by AADC. After intravenous injection, [rrC]S-HTP crosses through the BBB by a simple diffusion mechanism and is subsequently decarboxylated to w-rrC]S-hydroxytryptamine. To attempt to determine which one of these mechanisms is relevant for an understanding of PET data with this ligand, we used pharmacological interventions such as inhibition of peripheral serotonin synthesis, inhibition of 5-HTP synthesis within the brain, pretreatment with unlabeled 5-HTP, and repeated study without pharmacological interventions. The extraction from blood is probably low, since a rat study showed that only 7.4% (relative to radio-
222 active water) of arterially injected l>l_-[r4C]5-HTP entered the brain during the first pass (Oldendorf, 1971) indicating that small differences in regional cerebral blood flow are negligible. The corresponding uptake index for [‘4C] L-DOPA was 20%. The accumulation of 5-HTP in the striatum (where monoaminergic neurons do exist) suggests that our utilization rate constants reflect decarboxylation. A recent 1992) showed that the centrally active PET study on monkeys (Hartvig et al.. decarboxylase inhibitor NSD 10 15 almost completely inhibits striatal 5-HTP utilization, and that administration of 5-HTP labeled with “C in the carboxy position flattens the striatal utilization slope. This carboxy label is split off during decarboxylation to form [r’C]carbon dioxide, delivering unlabeled serotonin. These findings suggest that measures of v-riC]5-HTP accumulation reflect AADC activity. Interregional differences in the brain can be explained by heterogeneous distribution of AADC, most likely indicating the density of serotonin and dopamine neurons. Saavedra (1976) showed in a study of rats that decarboxylation of exogenously administrated 5-HTP occurs predominantly in serotonin-containing neurons. Another rat study (Sims et al., 1973) showed nonparallel cerebral and cerebellar distribution of L-DOPA decarboxylase and 5-HTP decarboxylase. However, the definitive proof that intravenously administrated radioactive 5-HTP is utilized in serotonin neurons only, and thus could be considered to mark physiological serotonin synthesis. awaits animal studies of synaptic serotonin synthesis and release by microdialysis. Compartmental models typically depend on blood radioactivity as the input function (Mintun et al., 1984; Wong et al., 1984; Perlmutter et al., 1986; Eckernh et al., 1987; Logan et al., 1987; Diksic et al., 1991). However, efficient use of blood data was precluded by the length of time needed to reach steady state between blood and brain (Fig. 1) (more than 40 minutes, which is double the 20.3-minute half-life of I’C) and the late emergence of presumably active metabolites. Use of an intracranial reference tissue instead of blood samples offers an alternative in attempts to avoid fuzzy and poorly detectable signals from peripheral 5-HTP metabolism. This approach has previously been used to good advantage in studies of L-DOPA uptake and utilization for cerebellum (Brooks et al., 1990; Tedroff et al., 1990) and a whole horizontal substriatal brain slice according to Eq. 5 (Hartvig et al., 1991). In the present study, intraneuronal utilization of [rtC]S-HTP (K and k,-1,). which we tentatively interpret as measures of specific serotonin synthesis, was estimated using whole brain at a level below the basal ganglia as reference. Whole brain delineated at this level contains a large fraction of white matter. The white matter has low enzymatic activity (Lloyd and Hornykiewicz, 1972), making this large region a better reference than cerebellum or blood. It is relevant to note that in PET research on brain metabolism, the calculation of relative regional metabolic rates of deoxyglucose consumption is commonly based on whole brain slices or slices from the ipsilateral hemisphere for the purpose of reducing the influence of experimental factors (Wiesel, 1992). With regard to opiate receptors, the influence of specific utilization in biasing reference tissue has been estimated and found to be inconsequential (Cunningham et al.. 1991). However, applying the L-DOPA approximation of Eq. 4 to the 5-HTP
223 case (Hartvig et al., 1991) yields some negative k3 rates, and a specific ut~ization of 5-HTP in the reference tissue cannot be refuted. This situation necessitates introducing a compartmental model that compensates for specific utilization in reference tissue. By the introduction of 1, , negative rate values become interpretable. A graphical estimation of k, according to Eq. 5 requires dividing the slope by the intercept F. The use of F adds a degree of freedom and precludes analyses of covariance. Furthermore, F is not a constant in the case of 5-HTP since the statement k,<
depressive episode, then the locus minoris in depression would be located between the capillary lumen and the site of hydroxylation. If this is true, a PCPA-induced depression could be prevented by pretreatment with unlabeled 5-HTP, but not with tryptophan. Current PET methodology has a number of limitations. The size of our striatal ROIs allows only fair resolution. Minor head movements during imaging might mimic utilization in small ROIs. Moreover, the PET camera measures dynamic regional intracranial radioactivity--not 5-HTP density or serotonin synthesis, and certainly not radioactivity within serotonergic neurons. Nevertheless, the pattern of utilization reported here follows the known distribution of serotonergic neurons, and the pattern fades after decarboxylase inhibition (Hartvig et al., 1992). In conclusion, the present PET studies of 5-HTP in human brain, which are based on a pharmacologic~ challenge strategy, appear to offer a means of gaining information about processes relevant to the in vivo utilization of S-HTP for serotonin synthesis. Similar challenge studies in patients suffering from psychiatric disorders might indicate whether responses are altered during illness, or whether the disorder itself mimics any known pharmacological perturbation. This project was supported by the Swedish Medical Research Council (grants 6355,8461, and 8645), the S~erstr~m-K~nig Foundation, and the Fredrik and Ingrid Thuring Foundation.
Acknowledgment.
References Agren, H.; Reibring, L.; Hartvig, P.; Tedroff, J.; Bjurling, P.; Homfeldt, K.; Andersson, Y.; Lundqvist, H.; and LHngstrom, B. Low brain uptake of L~‘iC]5-hydroxyt~tophan in major depression: A positron emission tomography study on patients and healthy volunteers. Acta Psychiatrica Scandinavica, 83M9-455, 1991. Bjurling, P.; Watanabe, Y.; Tokushige, M.; Oda, T.; and LILngstriim, B. Syntheses of /3-l’C-labelled L-tryptophan and 5-hydroxy-t,-tryptophan using a multi-enzymatic reaction route. Journal of the Chemical Society Perkin Transactions, 1:1331-1334, 1989. Brooks, D.J.; Salmon, E.P.; Mathias, C.J.; Quinn, N.; Leenders, K.L.; Bannister, R.; Marsden, C.D.; and Frackowiak, R.S.J. The relationship between locomotor disability, autonomic dysfunction, and the integrity of the striatal dopaminergic system in patients with multiple system atrophy, pure autonomic failure, and Parkinson’s disease, studied with PET. Brain, 113:1539-1552, 1990. Cunningham, V.J.; Hume, S.P.; Price, G.R.; Ahier, R.G.; Cremer, J.E.;and Jones, A.K.P. Compartmental analysis of diprenorphine binding to opiate receptors in the rat in vivo and its comparison with equilibrium data in vitro. Journal o~Cerebru1 Blood Flow and ~etabo~~srn. 1 l:I-9, 1991. Diksic, M.; Nagahiro, S.; Chaly, T.; Sourkes, T.L.; Yamamoto, Y.L.; and Feindel, W. Serotonin synthesis rate measured in living dog brain by positron emission tomography. Journal of Neurochemistry, 56: 153-162, 199 1. EckernHs, S.-A.; Aquilonius, S.-M.; Hartvig, P.; Hiigglund, J.; Lundqvist, H.; Nligren, K.; and LBngstriim, B. Positron emission tomography (PET) in the study of dopamine receptors in the primate brain: Evaluation of a kinetic model using I’C-N-methylspiperone. Acta Neuro~ogica Scandinavica, 75: 168-178, 1987. Hartvig, P.; Agren, H.; Reibring, L.; Tedroff, J.; Bjurling, P.; Kihlberg, T.; and LAngstrom, B. Brain kinetics of L-v-1lC]DOPA in humans studied by positron emission tomography. Journal qf Neural Transmission, 86:25-4 1. 199I.
225 Hartvig, P.; Lindner, K.J.; Tedroff, J.; Andersson, Y.; Bjurling, P.; and Langstrom, B. Brain kinetics of IlC-labelled.L-tr~tophan and 5-hydroxy-L-tr~tophan in the rhesus monkey: A study using positron emission tomography. Journal of Neural Transmission, 88: I-10, 1992. Lloyd, K.G., and Hornykiewicz, 0. Occurrence and distribution of aromatic L-amino acid (I.-DOPA) decarboxylase in the human brain. Journal of Neurochemistry, 19:1549-1559, 1972.
Logan, J.; Wolf, A.P.; Shiue, C.-Y.; and Fowler, J.S. Kinetic modeling of receptor-ligand binding applied to positron emission tomographic studies with neuroleptic tracers. Journal of Neurochemistry,
48:73-83, 1987.
Lovenberg, W.; Weissbach, H.; and Udenfriend, Journal of biological C~misrry,
S. Aromatic L-amino acid decarboxylase.
237~89-93, 1962.
Melamed, E.; Hefti, F.; Pettibone, D.J.; Liebman, J.; and Wurtman, R.J. Aromatic Lamino acid decarboxylase in rat corpus striatum: Implication for actions of r-DOPA in parkinsonism. Neurology, 31:651-655, 1981. Mintun, M.A.; Raichle, M.E.; Kilbourn, M.R.; Wooten, G.F.; and Welch, M.J. A quantitative model for the in vivo assessment of drug binding sites with positron emission tomography. Annals of Neurology, 15:217-227, 1984. Oldendorf, W.H. Brain uptake of radiolabeled amino acids, amines, and hexoses after arterial injection. American Journal of Phy.~iology, 221: 1629-1639, 197 I. Perlmutter, J.S.: Larson, K.B.; Raichie, M.E.; Markham, J.; Mintun, M.A.; Kilbourn, M.R.; and Welch, M.J. Strategies for in vivo measurement of receptor binding using positron emission tomography. Journal of Cerebral Blood Flow and Metabolism, 6: 154-169, 1986. Saavedra, J.M. S-hydroxy-L-tryptophan decarboxylase activity: Microassay and distribution in discrete rat brain nuclei. Journal of Neurochemistty, 26:585-589, 1976. Shopsin, B.; Friedman, E.; and Gershon, S. Parachlorophenylalanine reversal of tranylcypromine effects in depressed patients. Archives of General Psychiatry, 33:81 l-819, 1976.
Sims, K.L.; Davis, G.A.; and Bloom, FE. Activities of 3,4~ihydroxy-L-phenylalanine and 5-hydroxy-L-tryptophan decarboxylases in rat brain: Assay characteristics and distribution. Journal of Neurochemistry, 20:449-464,
1973.
Tedroff, J.; Aquilonius, S.-M.; Laihinen, A.; Rinne, A.; Hartvig, P.; Andersson, J.; Lundqvist, H.; Haaparanta, M.; Solin, 0.; Antoni, G.; Gee, A.G.; Ulin, J.; and Langstrom, B. Striatal kinetics of [“Cl-(f)-nomifensine and 6-[i*F]fluoro-L-DOPA in Parkinson’s disease measured with positron emission tomography. Acta Neurologica Scandinavica, 82:24-30, 1990.
Wiesel, F.-A. Glucose metabolism in psychiatric disorders: How can we facilitate comparisons among studies? In: Agren, H.; Martinot, J.-L.; and Wiesel, F.-A., eds. Studies on Brain Metabolism in Psychiatric Patients: Can Standards Be Drawn? Wien: Springer-Verlag, 1992. pp. I-18. Wong, D.F.; Wagner, H.N., Jr.; Dannals, R.F.; Links, J.M.; Frost, J.J.; Ravert, H.T.; Wilson, A.A.; Rosenbaum, A.E.; Gjedde, A.; Douglass, K. H.; Petronis, J.D.; Folstein, M.F.; Toung, J.K.T.; Burns, H.D.; and Kuhar, M.J. Effects of age on dopamine and serotonin receptors measured by positron tomography in the living human brain. Science, 226:13931395, 1984.
Psychiatry Research: Neuroimaging, 451215-225 Elsevier
Uptake and Utilization of [Beta-W]5-Hydroxytryptophan in Human Brain Studied by Positron Emission Tomography Lars Reibring, Hans Agren, Per Hartvig, Joakim Tedroff, Peter Bjurling, Tor Kihlberg, and Bengt Lflngstrijm
Hans Lundqvist,
Received May 8, 1992; revised version received September 2, 1992; accepted November I I, 1992. Abstract. The immediate precursor in the serotonin synthetic route, 5hydroxytryptophan (5HTP), labeled with 142 in the j3 position, has become available for studies using positron emission tomography (PET) to examine serotonin formation in human brain. Normalized uptake and intracerebral utilization of tracer amounts of [P-11C]S-HTP were studied twice in six healthy male volunteers, three of them before and after pharmacological pretreatments. The kinetic model defines regional utilization as the relative regional radioactivity accumulation rate. Repeat studies showed good reproducibility. Pretreatments with benserazide, p-chlorophenylalanine (PCPA), and unlabeled 5-HTP all significantly increased uptake of [P-liC]S-HTP. The utilization rates in both striatal and frontal cortex were higher than those in the surrounding brain, indicating that PET studies using [P-llC]S-HTP as a ligand quantitate selective processes in the utilization of 5-HTP. We tentatively interpret uptake and utilization as a measure of brain serotonin turnover, the selectivity of which was shown by pharmacological interventions in vivo. Key Words. Serotonin,
frontal cortex, striatum,
benserazide, p-chlorophenyl-
alanine. The monoamine hypotheses of the pathogenesis of several psychiatric disorders demand qualitative and quantitative in vivo studies of transmitter synthesis, transmitter turnover, and receptor function. Positron emission tomography (PET), which permits the use of various r*C tracers, has emerged as a focus for studies on all aspects of neurotransmission. The immediate precursor in the serotonin synthetic route, Shydroxytryptophan (SHTP), labeled with rrC in the p position, is now available for direct studies on precursor uptake mechanisms and for attempts to estimate presynaptic serotonin synthesis (Agren et al., 1991; Hartvig et al., 1992). An alternative approach to the study of brain uptake mechanisms is through the measurement of selective accumulation in brain of [rlC]-methyltryptophan (Diksic et al., 1991) but this substance is not transformed to serotonin.
Lam Reibring, M.D., is Clinical Researcher, and Hans digren, M.D., Ph.D., is Associate Professor, Department of Psychiatry, University Hospital, Uppsala. Per Hartvig, Pharm.D., Ph.D., is Professor, University Hospital Pharmacy. Joakim Tedroff, M.D., Ph.D., is Assistant Professor, Department of Neurology, University Hospital. Hans Lundqvist, Ph.D., Peter Bjurling, Ph.D., and Tor Kihlberg, Ph.D., are Researchers, and Bengt LAngstrom, Ph.D., is Professor, PET Center, Uppsala University. (Reprint requests to Dr. L. Reibring, Dept. of Psychiatry, University Hospital, S-751 85 Uppsala, Sweden.) 0165-1781/92/$05.00
@ 1992 Elsevier Scientific
Publishers
Ireland
Ltd.
In the present report, results are presented from PET studies that examined brain utilization of 5-HTP in healthy volunteers before and after the administration ot pharmacological agents chosen to alter certain aspects of monoamine metabolism-~ the peripheral r-aromatic acid decarboxylase (AADC) inhibitor benserazide, the central inhibitor of 5-HTP synthesis y-chlorophenylalanine (PCPA), and various doses of unlabeled 5-HTP. In addition, three pairs of identical experiments on uptake and utilization of 5-HTP were carried out to test the reproducibility of findings.
Methods Six healthy volunteers took part in the study. The subjects were mentally healthy men (physicians and medical students) between 28 and 43 years of age (mean = 37 years, SD = 6). Table 1 presents pertinent demographic characteristics of the subject and experimental details. In one triple experiment, the radioactive dose of 5-HTP was given to subject D simultaneously with intravenous doses of 0, 1, and 10 mg, respectively, of unlabeled 5-HTP. We also studied S-HTP brain utilization after benserazide, 50 mg given orally to subject E 2 hours before a second scan, and after PCPA, 1 g given in two oral doses to subject F within 24 hours before a second scan. Another three volunteers (subjects A, B, and C) were scanned twice to study reproducibility; the imaging sessions being separated by 2 hours, 2 weeks, and 7 weeks, respectively. Subjects.
Table 1. Description of 6 male healthy volunteers investigated with (p11C]5hydroxyttyptophan (5-HTP) Age ID
Pretreatment
Al
Baseline
A2
None
Bl
Baseline
82
None
Cl
Baseline
C2
None
Dl
Baseline 5-HTP, 10 mg, i.v.
El
Baseline
E2
Benserazide, p.o.
Fl
Baseline
F2
p-Chlorophenylalanine,
Date (month-
(W
(MW
43
78
151 142
21:02
06-22
37
108
70
l&18
OS-02
40
11:37
05-16
39
76
164
1:30
02-02
28
70
80
43
72
30 p.o. 37k6
Time of day
(yr)
1 mg, i.v.
SHTP,
D3
f SD
Radioactivity
190
D2
Mean
Weight
42
day) 08-22
13:16
03-22
lo:58
09-21
100
12:50
09-21
108
14:51
09-21 04-l 1
30
17:37
151
21:51
02-03
132
18:17
01-29
52 80113
(hr) 19~18
15:31
_____
05-04
123zk48
Radiochemistry. The radionuclide “C was obtained from the Tandem accelerator at The Svedberg Laboratory of Uppsala University and used in the synthesis of [P-*rC]S-HTP, as described elsewhere (Bjurling et al., 1989). The “C was introduced in the p position in the chain (Fig. l), thus yielding a serotonin molecule labeled with rrC after in vivo decarboxylation by AADC (Lovenberg et al., 1962). After analysis to ascertain chemical and radiochemical purity, a 0.1 mol/l phosphate buffer solution at pH 7.4 of [P-rrC]S-HTP was
217
Fig. 1. Radioactivityin arterial blood and ratio of brain tissues and blood in a healthy volunteer 3oc 25
-‘Arterial
blood
--o-
Ref. tlsslJe/A. blood
--e
Caudate n./A. blood
, 1.5
Time (min) Equilibrium is not reached during the imaging time span. The ~-11C)5-Hydroxytryptophan molecule is illustrated with its radioactive atom.
filtered through a 0.22 ,um filter and given intravenously was 30-164 MBq, corresponding to 9-50 pg of 5-HTP.
to the patient. The radioactive dose
Blood Samples. To serve as reference values, arterial samples were collected from subject D. Radioactivity was measured in whole blood in a well counter connected to the main computer (VAX) that controlled the PET camera and adjusted to time of blood drawing.
The PET scan was performed with the patient lying with his head fixed in a Scanditronix PC 384-3B tomograph. Imaging started immediately after intravenous administration of the “C-labeled 5-HTP and lasted for 40-60 minutes. This procedure delivers about 18 consecutive exposures of three horizontal brain slices. With guidance from a computed tomographic (CT) scan, the top-most level 3 was set to include the basal ganglia, parallel to and approximately 50 mm above the canthomeatal line. Level 2 was 13.5 mm down from level 3, and level 1 was a further 13.5 mm down in an attempt to reach cerebellar regions. Regions of interest (ROIs) were determined from a CT scan-derived outline map. The following ROIs were selected for further analysis: whole brain at level 2 (total area of brain slice); dorsoluteralprefrontal cortex (DLPFC; corresponding to Brodmann areas 9 and 10) at levels 2 and 3; medial PFC (MPFC; corresponding to Brodmann area 32) at levels 2 and 3; cuudute nucleus at level 3; and lentlform nucleus (including putamen and globus pallidus) at level 3. Measurements on the right and left sides were averaged. PET Procedure.
Measured radioactivities per cm3 were corrected for physical decay to the time of administration and given as uptake after correction for the administered dose per gram of body weight. Uptake data from ROIs were transferred to a Macintoshru computer for mathematical treatment by various procedures in programs such as ExcelTM, KaleidaGraphTM, StatView IIru, and SuperAnovarM.
Calculations.
Analysis of Uptake Data. Focusing on intracerebral
radioactivity, we excluded exposures before 5 minutes of run time because the initial blood radioactivity peak may conceal intracerebral radiation (Fig. 1). Averaged regional uptake values from 5 minutes to the end of
21X
the experiment (Table 3).
were termed
“averaged
uptake”
and were compared
by analysis
of covariance
Analysis of Change of Uptake over Time. Change of radioactivity over time was observed in certain brain regions. Several approaches are applicable for the quantification of this temporal change. Abbreviations are explained in Table 2.
Table 2. Abbreviations is the concentration
Uptake Averaged
uptake
of radioactivity in reference tissue or target tissue at time t.
is calculated
from exposures
Concentration
of radioactivity
CP
after 5 minutes from injection
of radioactivity.
t
in free plasma at time
t
CT
Total concentration
of radioactivity
in target tissue at time
CR
Total concentration
of radioactivity
in reference
CE
Total concentration
of extracellular
radioactivity
CET
Concentration
of extracellular
radioactivity
in target tissue at time
CER
Concentration
of extracellular
radioactivity
in reference
C/T
Concentration
of intracellular
tissue at time in reference
t
tissue and target
t
tissue at time
of intracellular
t
radioactivity
in target tissue at time
radioactivity
in reference
t
C/R
Concentration
K
Radioactivity
kl k2 k,
Fractional
rate constants
for transfer
of radioactivity
in target tissue
'1 4 13
Fractional
rate constants
for transfer
of radioactivity
in reference
E
Blood volume
t
Time, where
“accumulation
fraction
rate coefficient”
t
tissue at time tissue at time
t
in target tissue tissue
in tissue
t = 0 at time of injection
Note. All radioactivity measures above are normalized as for body weight, injected radioactivity, and physical decay of isotope to time of injection.
Table 3. Positron emission tomographic measures of averaged uptake and interest (utilization measures X 0.0001)
Volunteer Premedlcatlon Al
Bswllne
Whole brain Level 2
Dwsolateral prefrontal cortex Level 2
auptake
a.uptake
Medial Level
Level 3
K
k3-I3
a.uptaks
K
k3-I3
a.uptake
K
0.653
0.836
-7
-6
0.692
-7
-6
0.933
6
A2 None
0.863
0.880
-11
-12
0.878
8
9
0.950
-6
29
29
0.792
6
7
0.667
-1
12
12
0.935
6
6
0.885
9
1
1
1.075
5
4
1.013
11
Bl Baseline
0.699
0.7521
82 None
0.897
0.9581
Cl Baseline
1.0441
1.059
C2 None
0.752
Dl Baseline
0.6Oi
D2 5-HTP. 1 mg, I.“. 03 SHTP,
10 mg, I.V.
El Baseline
0.842
1
0.756 0.776
1
0.885
0.879
0.905
0.933
0.852
E2 Benserazide, 50 mg. p.o.
1.620
Fl Baseline
0.83i
F2 p-Chlorophenylalanine,
0.922.
3
1.504
0.909
-2
-2
0.764
-43
-39
0.660
I: 1 ! -9
I:
0.821 2
I
-9
2
0.877
9
0
-1
0.906
-9
20
20
0.954
3
? -3 2
I:
2
1-17
1
-3
1
2 Z-15
I 1
1.484
-22
0.802
-1
0.905
0.925
1 9,P.0. Mean of baselines
0.903
0.893
1 23
-2
-1
11 21
I
8
I
0.738
37
0.939
-37
0.978
13
-9
1.023
7
1
0.661
-45
-25
1
1.674 3
11
-2
0.831
30
341
321
1.068
39
7
6
0.910
-6
1 1
Note. Baselines denoted by bold characters. Significant differences calculated by analyses of covariance:
1. p < 0.05.
219 Approach A. An “accumulation rate coefficient” K is easily estimated for each ROI by plotting the ratio of uptake in target tissue Cr divided by uptake in a large neutral reference area Ca (whole brain at level 2), as a function of ordinary time t. If the blood-biased exposures before 5 minutes are excluded, the plots are approximately linear. K is the slope of the obtained regression line. Slope differences were tested by analysis of covariance (Table 3). Approach B. An alternative way to interpret this linear increase in radioactivity would be to apply a three-compartment model containing plasma C,, free plus nonspecifically bound ligand concentrations in extracellular space C, and intracellular space C,. An easily delineated region that is free from 5-HTP utilization and can be used as reference tissue is not readily avaifable. Thus, a model with reference tissue including tracer utilization I, has to be accepted (Fig. 2). The illustrated symmetric model gives
CT=&.Cp+(l
- &)(CET+
(1)
CIT)
or, after manipuIation (2)
C~-t’C~=(1-&e)(C~+C~T)
E is the relative blood volume fraction in tissue. The ratio of Eq. (2) and the corresponding equation for reference tissue gives the following formulation referring to Fig. 2.
CT -
ecp
CR - ecp
=
CET
CET + EE3 So’Cmdt
+ CIT
cER+t&T
=
CER
+ t lo’ c.sRdt
(3)
=
utilization of radioactiveShydroxytryptophan (5HTP) in various regionsof prefrontal cortex 2
Level 3
k3-la
a.uptake
K
Caudate nucleus k,-4
Lentiform nucleus
a.uptake
K
ka-13
a.uptake
K
ks-G
6
0.697
33
37
0.906
62
66
0.983
33
34
-6
0.845
-18
-18
0.915
70
72
1.013
21
22
-2
0.943
SO
49
0.607
46
46
0.621
0
7
1.047
-32
-31
0.594
71
29
3
10
0.993
47
43
1.126
44
43
36
0.732
41
41
0.763
73
57
-35 12
0.627 0.862
36 29
31 27
0.696 0.928
50 37
0.861
2
2
0.963
81
0.806
33
31
1.020
93
101
0.960
22
25
1.280
46
54
1.293
6 -46 12
1 1 I
1.625
1
1 I
1
24 90 ] 2 40
53
55
46
53
28
0.602
21
19
0.768
115
106
0.962
92
66
35
0.914
31
26
0.960
82
75
1.021
95
86
-7
0.895
21
19
0.885
75
72
0.877
51
48
2.p<0.01.3.p<0.001.
220 If the time expression ~C’,dt, C6a = (C,&r/C,, and the exposures during the inmal 5 minutes are ignored, Eq. (3) is approximately reduced to CT -=CR
Assuming k, <
ICI
j-0’ e-_(~z+~d(-)&
lt
j-0’e-(h+b)(t-S)&
(
\ So’CERdt \
’ + (” - 13) CR
)
(4)
I3 << k,, k2 = I2 and CKK= CR further reduces Eq. (4) to
(5) where F is a constant. Eq. (5) has been applied to [lrC]~-DOPA utilization calculation (Hartvig et al., 1991). As for S-HTP, none of these assumptions are justified. However, if the extracellular space is considered to be a homogeneous compartment all over the brain, then Eq. (4) is alternatively reduced to CT CR
=
1+
(k3 -
13)-
ra’ C&t
(6)
Cl2
jC,dt cannot be measured by PET, and for this reason the integral is changed tolC,dt. Finally, Cr/ CR is plotted as ordinate against /CRdt/ CR, yielding k, - 1, as the slope of the linear regression line. The slope reflects the rate of specific utilization of serotonin precursor, assuming specific utilization in reference tissue to be at a zero level. The slope differences were tested by analysis of covariance (Table 3).
Fig. 2. Compartmental tissue
model including specific utilization
Reference tissue
in reference
Target tissue
CT and CR are measurable by positron emission tomography
Results The averaged uptake values from baselines showed no obvious intracerebral differences, indicating a homogeneous blood-brain barrier (BBB) transport on the basis of [iiC]S-HTP and PET. The uptake ratio plots were clearly rectilinear 5 minutes after injection, with evenly distributed residuals (Fig. 3). Curve fitting based on exponential or other functions did not change the distribution of residuals. Striatal utilization was highest, while cortical utilization was at reference tissue level, except for the medial PFC at level 3. Although there were 57 comparisons of uptake and utilization measures from the same ROIs in the dual experiments on the three
221
Fig. 3. Consecutiveuptake values from dorsolateral prefrontal cortex (PFC) at level 2 and lentiformnucleus 1.6 'A
.
_
DorsolateralPFC level2
1.4 -
1.4 1.2 - -
*:
Lentifonnnucleus * *.
. **
.
.
Data points were smoothed by a 30% weighted curve fit (without assumption of any m~hematicai model). From 5 minutes after injection of radioactive ligand, the smoothed fitted lines are clearly r~tilinear, with an even dist~bution of residuals
volunteers who did not receive additional pharmacological challenges, only two significant differences could be detected, which suggests good reproducibility. The correlation between K and k,-Z, was extremely high and strongly significant (r2 = 0.99), illustrating that JCRdz/ CR is almost equal to true time in the case of 5-HTP. Very weak correlations were observed between k3-I3 or K and averaged uptake values (r* = 0.003 and 0.002, respectively), suggesting that k3-Z3 or K and averaged uptake are orthogonal variables and independent of each other. A dose~ependent increase in averaged uptake of labeled S-HTP was seen with the injection of increasing amounts of unlabeled 5-HTP. Similar alterations followed pretreatment with PCPA. Pretreatment with 50 mg of benserazide caused a general increase in averaged uptake. Table 3 shows the effects of the pharmacological interventions.
Discussion The physiological formation of serotonin is a complex process. The use of PET to study this process will involve several aspects, i.e., tryptophan transport across the BBB, uptake of t~ptophan into serotonergic nerve terminals, hydroxylation to 5-HTP by the rate-l~iting tryptophan hydroxylase, and finally decarboxylation to serotonin by AADC. After intravenous injection, [rrC]S-HTP crosses through the BBB by a simple diffusion mechanism and is subsequently decarboxylated to w-rrC]S-hydroxytryptamine. To attempt to determine which one of these mechanisms is relevant for an understanding of PET data with this ligand, we used pharmacological interventions such as inhibition of peripheral serotonin synthesis, inhibition of 5-HTP synthesis within the brain, pretreatment with unlabeled 5-HTP, and repeated study without pharmacological interventions. The extraction from blood is probably low, since a rat study showed that only 7.4% (relative to radio-
222 active water) of arterially injected l>l_-[r4C]5-HTP entered the brain during the first pass (Oldendorf, 1971) indicating that small differences in regional cerebral blood flow are negligible. The corresponding uptake index for [‘4C] L-DOPA was 20%. The accumulation of 5-HTP in the striatum (where monoaminergic neurons do exist) suggests that our utilization rate constants reflect decarboxylation. A recent 1992) showed that the centrally active PET study on monkeys (Hartvig et al.. decarboxylase inhibitor NSD 10 15 almost completely inhibits striatal 5-HTP utilization, and that administration of 5-HTP labeled with “C in the carboxy position flattens the striatal utilization slope. This carboxy label is split off during decarboxylation to form [r’C]carbon dioxide, delivering unlabeled serotonin. These findings suggest that measures of v-riC]5-HTP accumulation reflect AADC activity. Interregional differences in the brain can be explained by heterogeneous distribution of AADC, most likely indicating the density of serotonin and dopamine neurons. Saavedra (1976) showed in a study of rats that decarboxylation of exogenously administrated 5-HTP occurs predominantly in serotonin-containing neurons. Another rat study (Sims et al., 1973) showed nonparallel cerebral and cerebellar distribution of L-DOPA decarboxylase and 5-HTP decarboxylase. However, the definitive proof that intravenously administrated radioactive 5-HTP is utilized in serotonin neurons only, and thus could be considered to mark physiological serotonin synthesis. awaits animal studies of synaptic serotonin synthesis and release by microdialysis. Compartmental models typically depend on blood radioactivity as the input function (Mintun et al., 1984; Wong et al., 1984; Perlmutter et al., 1986; Eckernh et al., 1987; Logan et al., 1987; Diksic et al., 1991). However, efficient use of blood data was precluded by the length of time needed to reach steady state between blood and brain (Fig. 1) (more than 40 minutes, which is double the 20.3-minute half-life of I’C) and the late emergence of presumably active metabolites. Use of an intracranial reference tissue instead of blood samples offers an alternative in attempts to avoid fuzzy and poorly detectable signals from peripheral 5-HTP metabolism. This approach has previously been used to good advantage in studies of L-DOPA uptake and utilization for cerebellum (Brooks et al., 1990; Tedroff et al., 1990) and a whole horizontal substriatal brain slice according to Eq. 5 (Hartvig et al., 1991). In the present study, intraneuronal utilization of [rtC]S-HTP (K and k,-1,). which we tentatively interpret as measures of specific serotonin synthesis, was estimated using whole brain at a level below the basal ganglia as reference. Whole brain delineated at this level contains a large fraction of white matter. The white matter has low enzymatic activity (Lloyd and Hornykiewicz, 1972), making this large region a better reference than cerebellum or blood. It is relevant to note that in PET research on brain metabolism, the calculation of relative regional metabolic rates of deoxyglucose consumption is commonly based on whole brain slices or slices from the ipsilateral hemisphere for the purpose of reducing the influence of experimental factors (Wiesel, 1992). With regard to opiate receptors, the influence of specific utilization in biasing reference tissue has been estimated and found to be inconsequential (Cunningham et al.. 1991). However, applying the L-DOPA approximation of Eq. 4 to the 5-HTP
223 case (Hartvig et al., 1991) yields some negative k3 rates, and a specific ut~ization of 5-HTP in the reference tissue cannot be refuted. This situation necessitates introducing a compartmental model that compensates for specific utilization in reference tissue. By the introduction of 1, , negative rate values become interpretable. A graphical estimation of k, according to Eq. 5 requires dividing the slope by the intercept F. The use of F adds a degree of freedom and precludes analyses of covariance. Furthermore, F is not a constant in the case of 5-HTP since the statement k,<
depressive episode, then the locus minoris in depression would be located between the capillary lumen and the site of hydroxylation. If this is true, a PCPA-induced depression could be prevented by pretreatment with unlabeled 5-HTP, but not with tryptophan. Current PET methodology has a number of limitations. The size of our striatal ROIs allows only fair resolution. Minor head movements during imaging might mimic utilization in small ROIs. Moreover, the PET camera measures dynamic regional intracranial radioactivity--not 5-HTP density or serotonin synthesis, and certainly not radioactivity within serotonergic neurons. Nevertheless, the pattern of utilization reported here follows the known distribution of serotonergic neurons, and the pattern fades after decarboxylase inhibition (Hartvig et al., 1992). In conclusion, the present PET studies of 5-HTP in human brain, which are based on a pharmacologic~ challenge strategy, appear to offer a means of gaining information about processes relevant to the in vivo utilization of S-HTP for serotonin synthesis. Similar challenge studies in patients suffering from psychiatric disorders might indicate whether responses are altered during illness, or whether the disorder itself mimics any known pharmacological perturbation. This project was supported by the Swedish Medical Research Council (grants 6355,8461, and 8645), the S~erstr~m-K~nig Foundation, and the Fredrik and Ingrid Thuring Foundation.
Acknowledgment.
References Agren, H.; Reibring, L.; Hartvig, P.; Tedroff, J.; Bjurling, P.; Homfeldt, K.; Andersson, Y.; Lundqvist, H.; and LHngstrom, B. Low brain uptake of L~‘iC]5-hydroxyt~tophan in major depression: A positron emission tomography study on patients and healthy volunteers. Acta Psychiatrica Scandinavica, 83M9-455, 1991. Bjurling, P.; Watanabe, Y.; Tokushige, M.; Oda, T.; and LILngstriim, B. Syntheses of /3-l’C-labelled L-tryptophan and 5-hydroxy-t,-tryptophan using a multi-enzymatic reaction route. Journal of the Chemical Society Perkin Transactions, 1:1331-1334, 1989. Brooks, D.J.; Salmon, E.P.; Mathias, C.J.; Quinn, N.; Leenders, K.L.; Bannister, R.; Marsden, C.D.; and Frackowiak, R.S.J. The relationship between locomotor disability, autonomic dysfunction, and the integrity of the striatal dopaminergic system in patients with multiple system atrophy, pure autonomic failure, and Parkinson’s disease, studied with PET. Brain, 113:1539-1552, 1990. Cunningham, V.J.; Hume, S.P.; Price, G.R.; Ahier, R.G.; Cremer, J.E.;and Jones, A.K.P. Compartmental analysis of diprenorphine binding to opiate receptors in the rat in vivo and its comparison with equilibrium data in vitro. Journal o~Cerebru1 Blood Flow and ~etabo~~srn. 1 l:I-9, 1991. Diksic, M.; Nagahiro, S.; Chaly, T.; Sourkes, T.L.; Yamamoto, Y.L.; and Feindel, W. Serotonin synthesis rate measured in living dog brain by positron emission tomography. Journal of Neurochemistry, 56: 153-162, 199 1. EckernHs, S.-A.; Aquilonius, S.-M.; Hartvig, P.; Hiigglund, J.; Lundqvist, H.; Nligren, K.; and LBngstriim, B. Positron emission tomography (PET) in the study of dopamine receptors in the primate brain: Evaluation of a kinetic model using I’C-N-methylspiperone. Acta Neuro~ogica Scandinavica, 75: 168-178, 1987. Hartvig, P.; Agren, H.; Reibring, L.; Tedroff, J.; Bjurling, P.; Kihlberg, T.; and LAngstrom, B. Brain kinetics of L-v-1lC]DOPA in humans studied by positron emission tomography. Journal qf Neural Transmission, 86:25-4 1. 199I.
225 Hartvig, P.; Lindner, K.J.; Tedroff, J.; Andersson, Y.; Bjurling, P.; and Langstrom, B. Brain kinetics of IlC-labelled.L-tr~tophan and 5-hydroxy-L-tr~tophan in the rhesus monkey: A study using positron emission tomography. Journal of Neural Transmission, 88: I-10, 1992. Lloyd, K.G., and Hornykiewicz, 0. Occurrence and distribution of aromatic L-amino acid (I.-DOPA) decarboxylase in the human brain. Journal of Neurochemistry, 19:1549-1559, 1972.
Logan, J.; Wolf, A.P.; Shiue, C.-Y.; and Fowler, J.S. Kinetic modeling of receptor-ligand binding applied to positron emission tomographic studies with neuroleptic tracers. Journal of Neurochemistry,
48:73-83, 1987.
Lovenberg, W.; Weissbach, H.; and Udenfriend, Journal of biological C~misrry,
S. Aromatic L-amino acid decarboxylase.
237~89-93, 1962.
Melamed, E.; Hefti, F.; Pettibone, D.J.; Liebman, J.; and Wurtman, R.J. Aromatic Lamino acid decarboxylase in rat corpus striatum: Implication for actions of r-DOPA in parkinsonism. Neurology, 31:651-655, 1981. Mintun, M.A.; Raichle, M.E.; Kilbourn, M.R.; Wooten, G.F.; and Welch, M.J. A quantitative model for the in vivo assessment of drug binding sites with positron emission tomography. Annals of Neurology, 15:217-227, 1984. Oldendorf, W.H. Brain uptake of radiolabeled amino acids, amines, and hexoses after arterial injection. American Journal of Phy.~iology, 221: 1629-1639, 197 I. Perlmutter, J.S.: Larson, K.B.; Raichie, M.E.; Markham, J.; Mintun, M.A.; Kilbourn, M.R.; and Welch, M.J. Strategies for in vivo measurement of receptor binding using positron emission tomography. Journal of Cerebral Blood Flow and Metabolism, 6: 154-169, 1986. Saavedra, J.M. S-hydroxy-L-tryptophan decarboxylase activity: Microassay and distribution in discrete rat brain nuclei. Journal of Neurochemistty, 26:585-589, 1976. Shopsin, B.; Friedman, E.; and Gershon, S. Parachlorophenylalanine reversal of tranylcypromine effects in depressed patients. Archives of General Psychiatry, 33:81 l-819, 1976.
Sims, K.L.; Davis, G.A.; and Bloom, FE. Activities of 3,4~ihydroxy-L-phenylalanine and 5-hydroxy-L-tryptophan decarboxylases in rat brain: Assay characteristics and distribution. Journal of Neurochemistry, 20:449-464,
1973.
Tedroff, J.; Aquilonius, S.-M.; Laihinen, A.; Rinne, A.; Hartvig, P.; Andersson, J.; Lundqvist, H.; Haaparanta, M.; Solin, 0.; Antoni, G.; Gee, A.G.; Ulin, J.; and Langstrom, B. Striatal kinetics of [“Cl-(f)-nomifensine and 6-[i*F]fluoro-L-DOPA in Parkinson’s disease measured with positron emission tomography. Acta Neurologica Scandinavica, 82:24-30, 1990.
Wiesel, F.-A. Glucose metabolism in psychiatric disorders: How can we facilitate comparisons among studies? In: Agren, H.; Martinot, J.-L.; and Wiesel, F.-A., eds. Studies on Brain Metabolism in Psychiatric Patients: Can Standards Be Drawn? Wien: Springer-Verlag, 1992. pp. I-18. Wong, D.F.; Wagner, H.N., Jr.; Dannals, R.F.; Links, J.M.; Frost, J.J.; Ravert, H.T.; Wilson, A.A.; Rosenbaum, A.E.; Gjedde, A.; Douglass, K. H.; Petronis, J.D.; Folstein, M.F.; Toung, J.K.T.; Burns, H.D.; and Kuhar, M.J. Effects of age on dopamine and serotonin receptors measured by positron tomography in the living human brain. Science, 226:13931395, 1984.