SYNAPSE 5:213-223 (1990)

Mapping Muscarinic Receptors in Human and Baboon Brain Using

[ N-' ' C-Methyl]-Benz tropine STEPHEN L. DEWEY, ROBERT R. MACGREGOR, JONATHAN D. BRODIE, BERNARD BENDRIEM,

PAYTON T. KING, NORA D. VOLKOW, DAVID J. SCHLYER, JOANNA S. FOWLER, ALFRED P. WOLF, S. JOHN GATLEY, AND ROBERT HITZEMANN

Department of Chemistry, Brookhaven National Laboratory, Upton, New York 11973 (S.L.D., R.R.M., B.B., P.T.K., N.D.V., D.J.S., J.S.F., A.P.W., S.J.G.); Department of Psychiatry, New York University Medical

Center, New York, New York 10016 (J.D.B.); Department of Psychiatry, State University of New York, Stony Brook, New York 11974 (R.H.); Psychiatry Service, Veterans' Administration Medical Center, Northport, New

York 11 768 (R.H.)

KEY WORDS Positron emission tomography, Muscarinic, Cholinergic, Benztropine, Brain, Radiopharmaceuticals

ABSTRACT The muscarinic cholinergic system has been mapped in vivo in human and baboon brain using [N-llC-methyl]-benztropine and high resolution positron emis- sion tomography (PET). [N-llC-methyl]-benztropine uptake was observed in frontal, parietal, occipital, and temporal cortices as well as in subcortical structures including the corpus striatum and thalamus. Uptake continued to increase in baboon and human brain in all areas over an 80 minute experimental period with the exception of the cerebellum where the accumulation of radioactivity began to decrease by 25 minutes postinjection. The ratio of incorporation of [N-llC-methyll-benztropine between corpus striatuml cerebellum was 1.53 and 1.46 in humans and baboons, respectively, at 60 minutes. Blocking studies in baboons using the muscarinic cholinergic antagonists scopolamine and benztropine and the muscarinic cholinergic agonist pilocarpine combined with blocking studies in humans using benztropine indicate that the binding of this compound is specific for the muscarinic cholinergic system. Pretreatment with the potent dopamine reuptake blocker nomifensine produced no effect on the incorporation of radioactivity in any baboon brain region examined. Analysis of labelled plasma metabolites indicates that in humans, the rate of metabolism of [N-llC-methyll-benztropine is slow (83.0% un- changed at 30 minutes postinjection) differing quite dramatically from the rate of metabolism observed in baboons (43.4% unchanged at 30 minutes postinjection). These data combined with postmortem studies in humans and primates demonstrate that [N-llC-methyll-benztropine is a suitable muscarinic cholinergic ligand for use in humans and baboons with PET

INTRODUCTION Positron emission tomography (PET) has been widely

utilized for investigating the neuroanatomical distribu- tion of radiolabelled neurotransmitter-specific receptor ligands (Wagner, 1986). Defects in neurotransmitter systems are responsible for many debilitating diseases of the central nervous s stem (CNS) (Davies and Verth, 1977; Meltzer and Stah[ 1976). The muscarinic system, for example, plays an essential role in many memory and cognitive functions (Drachman and Leavitt, 1974; Dubois et al., 1985; Sitaram et al., 1978). Alterations in choline acetyltransferase activity as well as changes in cell number in the nucleus basalis, the primary source of cholinergic activity in the CNS, have been found at autopsy in the brains of patients suffering from demen- tia of the Alzheimer's t pe (Mountjo ,1986; Price et al.,

ity may underlie the negative symptomatolo associ- ated with schizophrenic illness (Tandon anrGreden,

1982; Whitehouse et a 9 ., 1982). Cho?kerpc hyperactiv-

0 1990 WILEY-LISS, INC.

1989). Cholinergic systems have been implicated in other CNS disorders as well (for review see McGeer and McGeer, 1984). In light of these data, it is im ortant to

humans with PET to examine this system in vivo in order to understand more fully the consequences of pathophysiologic and neuroanatomical changes that occur in these devastating diseases. [llCI-Dexetimide and [llCI-levetimide have been synthesized and used in animals for this purpose (Dannals et al., 1988). Re- cently, Mulholland and colleagues (1989) synthesized [llCl-tropanyl benzilate (TRB) and [ llCl-N-methyl pip- eridyl benzilate (NMPB) and performed PET studies in baboons with the former.

develop specific muscarinic probes that can bp e used in

Received September 26,1989; accepted October 10,1989

Address reprint requests to Stephen L. Dewey, Ph.D., Chemistry Department, Brookhaven National Laboratory, Upton, NY 11973.

214 S.L. DEWEY ET AL.

Neurotransmitter interactions provide for appropri- ate responses by the CNS to changes in the internal and external environment. The muscarinic system interacts both functionally and anatomically with other neuro- transmitter systems, i.e., the dopaminergic system (Bartholini et al., 1973; Elhert et al., 1981; McGeer et al., 1974; Roth and Bunney, 1976; Xu et al., 1989). Recently, we demonstrated that central cholinergic blockade significantly reduced the uptake of [ “FI-N- methylspiroperidol, a potent dopamine D, receptor ligand (Arnett et al., 1986) in the corpus striatum of adult male volunteers (Dewey et al., 1988). In disease states where primary neurotransmitter alterations ex- ist, as in schizophrenia, Alzheimer’s disease, and Hun- tington’s chorea, it is likely that alterations in other functionally linked neurotransmitter systems are in- volved. Understanding these secondary or transneu- ronal changes may be of clinical significance in the proper therapeutic management of these CNS disor- ders.

Benztropine (Cogentino) is a clinically prescribed synthetic anticholinergic agent which rapidly alleviates extrapyramidal side effects commonly associated with neuroleptic induced antipsychotic therapy (Coyle and Snyder, 1969). In the present study, we synthesized and injected [N-ilC-methyl]-benztropine into adult female baboons (Papio anubis) and adult male human volun- teers. We examined the neuroanatomical distribution, the time courses of radioactivity in the brain and la- belled metabolites in plasma, and the effects of specific muscarinic ligands and of a dopamine reuptake blocker on the distribution of [N-llC-methyl]-benztropine in the brain with high resolution PET. Abstracts of this work have appeared previously (Dewey et al., 1988,1989a,b).

MATERIALS AND METHODS [N-llC-methyll-benztropine synthesis

Norbenztropine was prepared from commercially available benztropine mesylate (Aldrich Chemical Co.) by conversion to its free base and demethylatlon by the reaction with phosgene followed by hydrolysis (Ban- holzer et al., 1975). [N-llC-methyl]-benztropine was

re ared by the reaction of carbon-11 (20.4 minute RalElife) labelled methyl iodide (Langstrom and Lund vist, 1976) with norbenztropine (1.0 mg) dis-

su1foxide:dimethylformamide (1:4) at 135°C (Fig. 6). The labelled benztropine was purified by semi-prepara- tive HPLC (silica gel, Ultrasphere-Si, 10 X 250 mm) using a solvent mixture consisting of acetonitrile:O.Ol M (NH4),HP04 (60:40) and a flow rate of 5 mumin. The retention time of norbenztropine in this system was 10-11 minutes and the retention time of benztropine was 15-16 minutes. The solvent was removed from the fraction containing the labelled benztropine and the residue was dissolved in 3.0 ml isotonic saline (USP) for injection. The total synthesis time was 40 minutes. The radiochemical purity was determined to be >98% by thin layer chromatography (ch1oroform:ethanol (5:l); Rf = 0.27). The specific activit determined by HPLC

solve 8 in 300 p1 of acetonitrile and 200 ~1 of dimethyl-

was typically 750 mCi/p,mol at t K e end of bombardment. Assay of [N-llC-methyl]-benztropine in plasma Plasma (0.2 ml) sampled at 1, 5, 10, 30, and 60

minutes after injection of [N-llC-methyll-benztropine was added to 1.0 cc of methanol, and the mixture was

sonicated and centrifuged. The supernatant was ana- lyzed by HPLC using the solvent system described above and a 4.6 x 250 mm ultrasphere-Si column and a flow rate of 2.0 mumin with ultraviolet detection and radioactivity assay of fractions. A sample of unlabelled benztropine was added to the supernatant and the fraction of labelled benztropme in each sample was taken as the amount of radioactivity co-elutin with the

activities eluting from the column. Mouse studies

Female mice (BNL strain, 25-31 g, n = 121, housed on a regular 12:12 li ht/dark cycle, received a tail vein injection of [N-G methyll-bt3nztropine (0.1 cc, 8-11 uCi). Animals were killed at 1, 5, and 30 minutes postinjection by cervical dislocation. Selected tissues were dissected, blotted free of excess blood, weighed, and measured for radioactivitl. content using a Packard automatic gamma counter. The data obtained at 5 minutes were used for calculating the dosimetry of [N-l’C-methyll-benztropine using the Medical Internal Radiation Dose (MIRD) Committee’s methods and a computer program “Dose-A Fortran Program for the Calculation of Radiation Dose from Radiopharmaceuti- cals” (Australian Atomic Enerby Commission).

HPLC analysis of soluble radioactivity in mouse brain

In a separate group of experiments, two female mice (BNL strain, 27-30 g) received a tail vein injection of [N-llC-methyll-benztropine (0.1 cc, 200-400 $3). At 30 minutes postinjection animals were killed by cervical dislocation. The brains were rapidly removed and ho- mogenized in 2 ml of cold methanol using a Polytron tissue disru ter. The homogenate was centrifuged for 4

rinsed with 3 ml of methanol and the mixture was centrifuged. The pellet and combined supernatant frac- tions were counted. An aliquot of the supernatant was diluted with a small amount of unlabelled benztropine and then injected onto a normal-phase analytical HPLC column. Fractions (30 second intervals) were collected in small tubes and counted in the Packard gamma counter.

Baboon studies Adult female baboons were initially anesthetized

with an intramuscular injection of ketamine hydrochlo- ride (10 mgkg), intubated, and transported to the PET facility. Animals were maintained on gas anesthesia (oxygen, nitrous oxide, and isofluorane) throughout the study. Vital signs were monitored and recorded. Cathe- ters were placed in an antecubital vein for isotope injection, and a femoral artery for blood sampling. Continuous arterial blood sampling was performed by an automated device (Ole Dich. Denmark) for the first 2 minutes in order to obtain multiple samples throughout the time during which the blood activity peaks. Manual sampling was used for the remainder of the study as described previously (Fowler et al., 1990). Plasma was counted for total radioactivity, and selected plasma samples taken at predetermined intervals were ana- lyzed for the presence of labelled metabolites as de- scribed above. Baboons were positioned in the tomo- graph using a horizontal and vertical laser alignment

unlabelled benztropine relative to the sum o P all other

minutes an B the supernatant decanted. The pellet was

IN-"C-METHYL]-BENZTROPINE BINDING IN BRAIN 215

system. Each animal had its own individual head- holder (Kearfott et al., 1984). In addition, an infared motion monitoring device (Tri-Tronics, Smarteye Model SD) attached directly to the head-holder and aimed at the tip of the baboons' nose was used in order to alert the investigators of any head movement (>1.0 mm). Scan- ning was performed with eyes closed for 84 minutes in a Computer Technology Imaging (CTI) positron tomo- graph (model 931-08/12; 15 slice, 6.5 mm slice thick- ness, full width at half maximum, (FWHM) with an in plane resolution of 6.0 x 6.0 mm (FWHM). The follow- ing scanning protocol was used: 4 scans of 30 seconds followed by 4 scans of 1 minute, 4 scans of 2 minutes, 5 scans of 10 minutes, and 1 scan of 20 minutes. In all studies, two injections of [N-llC-methyll-benztropine (6.07-17.82 mCi, 28.0-81.3 pg) were performed at least 2 hours apart on the same day in order to assess either the variability of the PET measurement (n = 3) or the effect of pharmacological intervention on the incorpora- tion of radioactivity in the brain (n = 7). In all studies, baboons remained in the PET gantry between isotope injections. Animals recovered for a t least 3 weeks prior to their second study.

Human studies Normal adult male volunteers (19-81 years) gave

informed consent and received a series of two intrave- nous injections of [N-llC-methyll-benztropine (12.00- 13.93 mCi) on the same day. Vital signs were monitored and recorded. Scanning was performed with eyes opened for 80 minutes in the CTI 931 positron tomo- graph using the following scanning protocol: 5 scans of 1 minute followed by 3 scans of 5 minutes and 3 scans of 20 minutes. Human subjects were positioned in the tomo- graph with individual head-holders (Kearfott et al., 1984) using a vertical and horizontal laser alignment system. In order better to visualize the corpus striatum and the temporal lobe, the gantry angle was adjusted to -25" from the cantho-meatal line. Subjects remained in the gantry throughout each individual scanning period but were removed between isotope injections. Radial arterial blood sampling was performed as previously described (see baboon studies), and selected samples were analyzed for the presence of unchanged tracer as described above. In all studies, two injections of [N- llC-methyll-benztropine were performed at least 2 hours apart on the same day in order to assess either the variability of the PET measurement (n = 2) or the effect of unlabelled benztropine on the incorporation of radio- activity in the brain. In four studies, volunteers received an i.v. injection of benztropine (2.0 cc, 2.0 mg) 60 min- utes prior to the second isotope injection.

In both human and baboon studies, prior to the first isotope injection, a transmission scan was performed in order to correct for the attenuation of the annihilation photons. The tomogra h was cross calibrated with the

in the plasma samples. well-counter used for t Yl e measurement of radioactivity

Data analysis Region of interest (ROI) selection was performed

directly on the transaxial PET images for human and baboon studies using computer software supplied with the tomograph. ROI's were drawn on slices from the last scan as it consistently provided the greatest definition of anatomical structure. These regions were directly

copied onto each temporally correspondin slice. By

if any movement of the head had occurred during the scanning period and subsequently made adjustments in only ROI position and not size or shape. This was more important in human studies where the subject was removed from the gantry after the first radiotracer injection and then repositioned for the second ra- diotracer injection.

In all studies, the regional brain uptake was normal- ized to the amount of tracer in the blood by calculating an incorporation quotient (IQ, Patlak, 1981). The I& normalizes the uptake in a tissue at a particular time (TI to the concentration of the tracer in the plasma from the time of injection to the same time (T); i.e.,

examining ROI placement on each slice we f etermined

IQ = C,(T)/J I$ C,(t)dt

where C,(T) is the concentration of tracer in tissue at time T and

is the integrated concentration of plasma from the time of injection to time T. The values for C, were obtained from counting the total radioactivity in the plasma and correcting these values for the amount of unchanged labelled benztropine obtained by HPLC analysis of se- lected plasma samples. These values for both baboons and humans are presented in Tables 11-IV.

RESULTS Mouse studies

Table V shows the anatomical distribution of [N- llC-methyll-benztropine in mice at 1,5, and 30 minutes postinjection. Radioactivity accumulated in several re- gions, with the highest amount being in the lungs, heart, and kidneys. Radioactivity concentration contin- ued to increase in the brain reaching a value of 3.07 (% injected dose/g) at 30 minutes.

Studies where radioactivity was extracted from mouse brain indicated that of the total radioactivity in the homogenate, 16% was associated with the pellet, and 84% was extracted in methanol. HPLC assay of the extracted radioactivity showed the presence of three radioactive peaks, one eluting at 11 minutes, corre- sponded to unchanged tracer, one eluting at 2.5 minutes (unidentified), and one eluting at 3.5 minutes (uniden- tified). Thus, of the total carbon-11 in the mouse brain at 30 minutes, 16% was associated with the pellet, 68% was unchanged [N-llC-methyl]-benztropine, and 16% was unidentified labelled metabolites. Radioactivity re- covery from the HPLC was >98%.

Control baboon studies The regional distribution of carbon-11 can be seen in

different transaxial levels of the brain (Fig. 2). With the exception of the cerebellum and the thalamus, radioac- tivity accumulated from the time of injection through- out the 84 minutes of scanning in the cortex and corpus striatum (Fig. 1). The greatest incorporation of radioac- tivity, and consequently the highest I&, can be seen in the corpus striatum, bilaterally, with the smallest incor- poration being in the cerebellum (Table 11). The ratio of the average incorporation in the corpus striatum over the average incorporation in the cerebellum (CSr/Cb)

216 S.L. DEWEY ET AL.

was 1.46 at 60 minutes. Radioactivity appeared to be homogeneously distributed throughout the cortex. There was no incorporation of radioactivity in any white matter structure. In those studies where animals re- ceived two injections of [N-llC-methyll-benztropine without an pharmacological intervention, the test/ retest varia ility in the region of hi hest incorporation, the corpus striatum, was less t an 19.0% (range 0.0-18.8; table 11). However, the testhetest variability in the same animal on different days ranged from 1.6% (studies 26A and 34A) to 52.6% (studies 43A and 46A).

Analysis of carbon-11 labelled metabolites revealed that at 30 minutes postinjection, 43.4% of the total radioactivity remained as unchanged tracer (Table I). This is in marked contrast to the rate of metabolism seen in human plasma (Table I).

Blocking studies in baboons Pretreatment with scopolamine (0.1 mgkg) or benz-

tropine (0.1 mgkg) 60 minutes prior to administration of [N-l'C-methyl]-benztropine reduced the I& in all regions. This reduction, however, was not uniform across brain regions. Reductions in the corpus striatum, cortex, and cerebellum avera ed 35.8, 32.9, and 25.3%,

from 1.46 to 1.32 at 60 minutes. There was no preferen- tial effect of scopolamine or benztropine on reducing radioactivity incorporation in any brain structure. In- creasing the blocking dose of benztropine from 0.1 mgl kg to 0.2 mgkg resulted in a greater decrease in incor- poration of radioactivity in all brain repons (Table 11). The administration of these unlabelled compounds in- creased heart and respiratory rate immediately after injection which returned to baseline values 40 minutes prior to injection of the isotope. These drugs did not

roduce an effect on the rate of metabolism of the abelled benztropine.

Pretreatment with the cholinergw agonist pilocarpine (0.1 mgkg, Sigma Chemical Co.) 45 minutes prior to isotope injection reduced the incorporation of radioac- tivity in all regions including the cerebellum. However, this reduction was not greater that the variability of the measurement (<20%). The CSriCb ratio was 1.43 at 60 minutes. Administration of pilocarpine roduced

were transient and returned to normal resting values well before administration of the isotope. Pilocarpine did not alter the rate of metabolism of [N-llC-methyll-

a

respectively (Table 11). The 8 Sr/Cb ratio was reduced

Y

changes in both cardiac and respiratory rates E ut these

benztropine. Pretreatment with nomifensine (2.0 mgkg, 60 min-

utes prior to isotope injection) did not produce changes in the I& in any of the regions examined (Table 11). Nomifensine did not alter the rate of metabolism of the labelled compound or produce any change in cardiac or respiratory rate.

Control human studies The regional distribution of carbon-11 can be seen in

different transaxial levels extending from a rostra1 level above the centrum semiovale to a more caudal midcere- bellar level (Fig. 4). Incorporation of radioactivity in the brain began to occur almost immediately following ad- ministration of the isotope and continued throughout the length of the study in all cortical and subcortical regions with the exception of the cerebellum which began to clear of radioactivity within 25 minutes postin-

jection(Fig. 3). The CSr/Cb ratio at 60 minutes was 1.53. The greatest incorporation of radioactivity observed in the PET images and measured by the I&, was seen in the corpus striatum, bilaterally, and occipital cortex (Tables 111, IV). Punctate incorporation of radioactivity was observed in frontal, parietal, temporal, and occipi- tal cortices. The occipital cortex contained the greatest amount of radioactivity of all the cortical regions (Table 111). Radioactivity was also seen in the pituitary gland along the midline. There appeared to be no incorpora- tion of radioactivity in any white matter structure. Two subjects who received sequential injections 2 hours apart without any pharmacological intervention had a testhetest variability in the corpus striatum and cere- bellum of 6.4% or less (Table IV).

Analysis of carbon-1 1 labelled plasma metabolites revealed that at 30 minutes postinjection, 83.0% of the radioactivity remained as unchanged tracer (Table I).

Blocking studies in humans Four subjects received an intravenous injection of

benztropine (2.0 mg) 60 minutes prior to the second isotope injection. Administration of unlabelled benz- tro ine reduced the I& in the corpus striatum, cortex, an B thalamus by an average of 20.3% (range 19.7-23.7), 18.4% (range 12.2-22.51, and 15.4% (range 11.3-19.61, respectively, in all but one subject (study 3, Table IV). The smallest effect of unlabelled benztropine was ob- served in the cerebellum with an average change of 9.1%, which is greater than the testhetest variability. The CSr/Cb ratio was reduced to 1.36 at 60 minutes. Unlabelled benztropine did not alter the rate of [N- llC-methyll-benztropine metabolism.

DISCUSSION On the basis of pharmacologic studies, muscarinic

receptors in the CNS have been classified into two subtypes-M, and M, receptors (Hammer and Gia- chetti, 1982; Hammer et al., 1980). Studies using selec- tive muscarinic agents have demonstrated that MI re- ceptors are located redominantly in the neocortex and

gastrointestinal tract are richer in M, receptors (Watson et al., 1986). Whether these subtypes can be classified on the basis of their functional properties remains unclear. It is, therefore, difficult to assess whether a specific M, or an M2 agent should be utilized with PET in order to investigate the neuroanatomical consequences of muscarinic receptor-linked CNS disor- ders.

The specificity of the [N-"C-methyl]-benztropine for the muscarinic sites is supported by several lines of evidence. First, the neuroanatomical distribution of radioactivity, in this study, closely corresponds to the distribution of muscarinic receptors as demonstrated in postmortem autoradiographic studies in human brain

the neostriatum, w R ereas thcb brain stem, heart, and

TABLE I . Percent unchanged [N-"C-ozethyl/-benztropine in human and hahoon uilasma'

Time (minutes) 1 .n 5.0 : 0.0 30.0 60.0

Human 96.1 + 2.8 94.5 * 3.5 94.0 k 3.4 83.0 i 7.1 - Baboon 96.2 k 2.7 79.9 !C 8.9 68.2 k 10.9 43.4 + 9.4 33.5 & 5.0

'Values represent the mean i standard deviikion.

TAB

LE II

. Inc

orpo

ratio

n qu

otie

nt (

min

-') f

or [N

-"C

-met

hyl]

-ben

ztro

pine

at

60 m

inut

es p

ostin

ject

ion

in se

lect

ed b

rain

reg

ions

in

babo

ons'

Stud

y B

aboo

n st

riat

um

chan

ge!

Tha

lam

us

chan

ge'

Cor

tex

chan

ge'

Cer

ebel

lum

ch

ange

Z

Inte

rven

tion

23A

Pe

ace

0.17

6 0.

171

0.16

1 0.

142

Con

trol

23

B

0.14

3 -1

8.8

0.14

7 -1

4.0

0.13

4 -1

6.7

0.12

6 -1

1.3

Con

trol

45

A

Peac

e 0.

198

0.16

6 0.

165

0.13

9 C

ontr

ol

45B

0.

188

-5.1

0.

172

3.6

0.16

1 -2

.4

0.13

2 -5

.0

Con

trol

46

A

Ora

l 0.

092

0.09

2 0.

084

0.06

4 C

ontr

ol

46B

0.

090

-2.2

0.

086

-6.5

0.

084

0.0

0.06

3 -1

.6

Con

trol

29

A

Peac

e 0.

227

0.18

4 0.

218

0.15

4 C

ontr

ol

29B

0.

217

-4.4

0.

192

4.3

0.24

8 13

.7

0.16

6 7.

8 N

omif

ensi

ne3

44A

O

ral

0.12

5 0.

109

0.11

2 0.

090

Con

trol

-3

6.7

Ben

ztro

pine

4 44

B

0.07

5 -4

0.0

0.08

1 -2

5.6

0.06

5 -4

1.9

0.05

7 43

A

Ora

l 0.

194

0.16

2 0.

182

0.14

2 C

ontr

ol

43B

0.

133

-31.

4 0.

129

-20.

3 0.

126

-30.

8 0.

104

-26.

8 B

enzt

ropi

ne?

26A

E

lly

0.13

0 0.

117

0.11

7 0.

059

Con

trol

0.

044

-25.

4 Sc

opol

amin

e6

26B

0.

065

-50.

0 0.

079

-32.

4 0.

070

-40.

2 34

A

Elly

0.

132

0.10

5 0.

117

0.09

9 C

ontr

ol

34B

0.

103

-21.

9 0.

099

-5.7

0.

095

-18.

8 0.

087

-12.

1 Sc

opol

amin

e6

31A

O

ral

0.13

2 0.

110

0.10

8 0.

079

Con

trol

31

B

0.10

9 -1

7.4

0.09

6 -1

2.7

0.10

1 -6

.5

0.07

0 -1

1.4

Pilo

carp

ine7

35

A

Ora

l 0.

158

0.14

9 0.

134

0.11

4 C

ontr

ol

35B

0.

147

-6.9

0.

142

-4.6

0.

130

-3.0

0.

114

0.0

Pilo

carp

ine7

'A a

nd B

refe

r to

the

fir

st a

nd s

econ

d PE

T sc

ans,

resp

ectiv

ely,

on

the

sam

e ba

boon

on

the

sam

e da

y.

'Cal

cula

ted

as 1

00(r

un A

- ru

nB)/

runA

. '2

.0 r

nglk

g i.v

. at 6

0 rn

in b

efor

e in

ject

ion

of t

race

r.

40.Z

rng

lkg

i.v. a

t 60

min

bef

ore

inje

ctio

n of

tra

cer.

'0.

1 m

g/kg

i.v

. at

60

rnin

bef

ore

inje

ctio

n of

tra

cer.

60

.1 m

g/kg

i.v.

at 6

0 m

in b

efor

e in

ject

ion

of t

race

r.

70.1

mg/

kg i.

v. a

t 45

min

bef

ore

inje

ctio

n of

tra

cer.

Cor

pus

Perc

ent

Perc

ent

Perc

ent

Perc

ent

TA

BL

E II

I. I

ncor

pora

tion

quot

ient

imin

-'1 f

or IN

-"C

-mrt

hyll

-ben

ztro

pine

at 6

0 m

inut

es p

osti

njec

tion

in c

ortic

al r

egio

ns o

f hu

man

bra

in'

Stu

dy

Fro

ntal

%

cha

nge

Par

ieta

l (X

I cha

nge

Tem

pora

l '%I

cha

nge

Occ

ipita

l '4

chan

ge

1A

0.13

5 0.

135

0.12

3 0.

143

1R

0.12

3 -8

.9

0.13

4 -0

.7

0.12

5 1.

6 0.

148

3.5

2A

0.10

6 0.

093

0.09

2 0.

129

2B

0.10

5 -0

.1

0.09

3 0.

0 0.

096

4.3

0.13

7 6.

2 3A

0.

093

0.08

8 0.

088

0.12

2 3 B

0.

105

12.9

0.

088

0.0

0.09

0 2.

3 0.

113

-7.4

4A

0.

105

0.10

2 0.

108

0.12

7 4R

0.

086

-18.

1 0.

08'2

-19.

6 0.

086

-20.

4 0.

102

-19.

7 5A

0.

091

0.08

7 0.

091

0.12

0 5B

0.

072

-20.

9 0.

071

-18.

4 0.

077

-15.

4 0.

093

-22.

5 6A

0.

092

0.09

3 0.

081

0.11

5 6R

0.

075

-18.

5 0.

075

- 19

.4

0.06

8 -1

6.0

0.10

1 -1

2.2

'A a

nd B

refe

r to

the

firs

t and

seco

nd P

E'T

scan

s,re

spec

tive

ly, o

n th

e sa

me i

ndiv

idua

l on

the

sam

e day

. On

stud

ies

3-6

the

B s

can

was

per

form

ed 6

0 m

inut

es f

ollo

win

g a

2.0

mg

i.v.

dos

e of

be

nztr

opiu

e. P

erce

nt c

hang

e w

as c

alcu

late

d as

100

(run

A

run

B)/

run

A.

TA

BL

E IV

. In

coru

orat

ion

ouot

ient

(min

-'i f

or IN

-"C

-met

hvll-

benz

troD

ine

at 6

0 m

inu

tes

uost

inie

ctco

n in

sub

cort

ical

reg

ions

and

cer

ebel

lum

of h

uman

bra

in'

Cor

pus

9% ch

ange

C

ereb

ellu

m

Stu

dy

stri

atum

'E

cha

nge

Tha

lam

us

% c

hang

e

1A

1B

2A

2B

3A

3R

4A

6B

6A

0.15

3 0.

162

0.14

1 0.

139

0.11

4 0.

119

0.13

1 u.

iX

0.11

9

5.9

-1.4

4.4

-a,,/

0.12

0 0.

145

0.11

8 0.

111

0.11

7 0.

105

0.11

5 0.

102

0.09

1

20.8

-5.9

-10.

3

-11.

3

0.11

0 0.

117

0.08

4 0.

081

0.08

3 0.

093

0.08

5 0.

078

0.06

9

6.4

-3.6

12.0

-8.2

~~

~ 5R

0.09

8 -1

7.6

0.07

7 -1

5.4

0.06

4 -7

.2

6A

0.11

7 0.

107

0.08

4 6B

0.

094

-19.

7 0.

086

-19.

6 0.

074

-1 1

.9

'A a

nd B

refe

r to

the

firs

t and

seco

nd P

ET

sca

ns, r

espe

ctiv

ely,

on

the

sam

e in

divi

dual

on

the

sam

e da

y. O

n st

udie

s 3-

6 th

eB sc

an w

as p

erfo

rmed

60

min

utes

foll

owin

g R

2.0m

g i.v

. dos

e of

' be

nztr

opin

e. P

erce

nt c

hang

e w

as c

alcu

late

d as

100

(run

A -

. ru

n B

)fru

n A.

[N-"C-METHYLJ-BENZTROPINE BINDING IN BRAIN 219

0.051

TABLE V. Anatomical distribution of [N-"C-methyll-benztropine i n adult female mice at I , 5, and 30 minutes postinjection (% injected

dose/grami'

Time (minutes) i n 5 0 3n.o

Brain Blood Heart Lung Liver Spleen Kidneys Small intestine Ovaries

1.26 f 0.06 3.11 k 0.55 18.1 & 0.73 30.1 * 3.0 4.46 f 0.80 5.55 f 0.88 14.0 & 1.1 4.08 * 0.55 5.75 * 1.2

1.81 rt 0.17 1.50 f 0.12 11.1 f 1.1 35.2 i 6.3 7.87 k 1.6 9.29 + 2.1 23.0 i 4.8 6.22 i 1.1 11.8 i 3.0

3.07 5z 0.14 1.47 5 0.11 4.09 + 0.53 22.8 i 7.6 14.7 & 1.8 12.0 i 1.4 23.2 i 1.8 12.8 + 0.62 10.0 i 2.5

'Values represent the mean k standard deviation (n = 4)

0.04 -1

Striatum n Thalamus o F Cortex A Cerebellum

. I I , . I . I . I . I . I ' I ' I

10 20 30 40 50 60 70 80 90

Time post Injection (min.)

Fig. 1. Time-activity curves in the baboon following an i.v. injection of [N-l'C-methyl]-benztropine. Note: radioactivity continues to accu- mulate in the corpus striatum, bilaterally, and the cortex while tha- lamic and cerebellar activity decreases. Data taken from study 23-4, Table 11.

Fig. 2. Transaxial PET images (9 levels) of the baboon brain a t 60 minutes postinjection. The most rostra1 level is in the upper left and the most caudal is in the lower right. The greatest incorporation is seen in the corpus striatum, while the lowest is in the cerebellum, posteriorly. Images taken from study 23A.

220 S.L. DEWEY ET AL.

0 006-

h 0 0005- 0

6 0004- > n - - ' Oo3-* M p) 0002- v

2

3 0.007

0.006 Striutum I.

~

~

Temporal 2-

Cm-i<ululstZ ** Parietal Thalamus * ff Frontal

,

Cerebellum

0 001

0 000 0 10 20 30 40 50 60 70

Time post Injection (min.)

Fig. 3. Time-activity curves in the human brain following an i.v. injection of IN-"C-methyl]-benztropine. Radioactivity continues to accumulate in the corpus striatum, thalamus, and occipital cortex, while cerebellar activity begins to decrease by 25 minutes post injec- tion. Data taken from study 5A, Tables 111 and lV.

4 O o o l i 0000/~ I . , . 1 , I

0 100 200 300 400 500 600 700

Receptor Density (fmole/mg protein)

Fig. 5. Comparison of regional uptake of [N-"C-methyl]- benztropine in study 5A (% injected dosekc) with the known muscar- inic receptor distribution measured with [3Hl-QNB in human postmor- tem material. Postmortem data taken from Lin et a]. (1986).

Fig. 4. Transaxial PET images (15 levels) of the human brain a t 60 minutes postinjection. The most rostra1 level is in the upper left and the most caudal in the lower right. Punctate incorporation of radioactivity can be seen in all cortical regions. The greatest incorporation appears in the corpus striatum, bilaterally, and occipital cortex, posteriorly. Data taken from study 5A.

[N-llC-METHYL,l-BENZTROPINE BINDING IN BRAIN

1. LiAlH4

22 1

2 . HI 1 1 ~ 0 2 - ~ I C H ~ I

Fig. 6. Synthetic scheme of 1N-"C-methyl]-benztropine.

using [3Hl-quinuclidinyl benzilate (QNB) (Lin et al., 1986) (Fig. 5). The greatest incorporation of radioactiv- ity occurs in the corpus striatum in human and baboons, which is consistent with the known regional distribu- tion of rece tor density as demonstrated in humans by

of [N-llC-methyl]-benztrogine in human subjects paral- lels the distribution of [ 31]-QNB as measured with single photon emission computed tomography (SPECT) in living human brain (Drayer et al., 1982; Weinberger et al., 1989).

Second, the uptake of radioactivity is reduced in humans by benztropine and in baboons by scopolamine and benztropine. It is interesting to note, however, that in one human volunteer (study 3, Tables I11 and IV) [N-llC-methyll-benztropine uptake was not reduced fol- lowing a 2.0 mg i.v. dose of benztropine. This is consis- tent with observations that patients clinically respond differently to similar doses of anticholinergic agents. This may reflect individual variations in either musca- rinic rece tor number (Lin et al., 1986) or mechanisms

brain barrier. Decreases in [N-llC-methyll-benztropine uptake due to regional changes in cerebral blood flow following administration of scopolamine, benztropine, or pilocarpine cannot be ruled out, as electron micro- scopic evidence suggests that the muscarinic system may play a regulatory role in local cerebral blood flow (Armstrong, 1986). Physostigmine increases, while sco- polamine administration decreases, cortical blood flow (Dam and London, 1984; Helen and London, 1984; Weinberger et al., 1979). Honer and colleagues (19881, using the 133Xe inhalation technique for assessing re- gional cerebral blood flow (rCBF), demonstrated in humans that doses of 7.3 and 6.1 Fgkg produced signif- icant cognitive impairment, while rCBF changes in frontal cortex occurred only at the higher dose. It is unlikely, however, that changes in rCBF are in them- selves responsible for the reduction in incorporation of [N-llC-methyll-benztropine, as the time course of changes in rCBF due to anticholiner 'c agents demon- strated that by 60 minutes rCBF ha f returned to nor- ma1 pretreatment values (Honer et al., 1988).

Benztropine inhibits dopamine reuptake in the cor- pus striatum in vitro (Coyle and Snyder, 1969). Previ- ously we showed that [N-l C-methyl]-cocaine binding in the corpus striatum was inhibited by nomifensine, dem- onstrating that the dopamine reuptake site can be

Lin and col P eagues (1986). In addition, the distribution

responsib P e for transport of benztropine across the blood

imaged with a suitable labelled ligand (Fowler et al., 1990). However, pretreatment with nomifensine prior to the injection of [N-llC-methyll-benztropine did not produce changes in the IQ in the dopamine rich corpus striatum (Table 11). Therefore, it appears that the bind- ing of [N-llC-methyl]-benztropine to the dopamine re- uptake site does not make a significant contribution to the PET images of the baboon brain. Together these studies demonstrate that [N-llC-methyll-benztropine binding is primarily to the muscarinic receptor.

The relative affinites of benztropine for the muscar- inic subtypes suggests that the specific binding is to both M1 and M2 receptors (Snyder and Yamamura, 1977). Recently, Lin and his coworkers (19861, using L3H1- irenzepine and r3H]-QNB, demonstrated that M1 baboon brain, the highest amount of radioactivity was seen in the M1 rich neostriatum and neocortex. Our mouse data suggests that M2 receptors also bind [N- llC-methyll-benztropine, as there was a higher incorpo- ration of radioactivity in the heart and gastrointestinal tract than in the brain (Tablev). As the [N-l1C-methyl1- benztropine uptake was observed in regions with vary- ing ratios of M1 to M, receptors, it appears that the binding we observed is to both M1 and M2 subtypes. However, this needs to be demonstrated by specifically blockin M1 and M2 subtypes and measuring the effect

The metabolism of [N-llC-methyll-benztropine in hu- man plasma is very slow in sharp contrast to that in baboons (Table I). This is consistent with clinical obser- vations that a single daily dose of benztropine is suffi- cient in preventing extrapyramidal symptoms. In this respect, [N-W-rneth 11 benztropine is superior to the

metabolized in human plasma (Erhenkaufer, personal comm.; Frey et al., 1985).

These results demonstrate that [N-llC-methyl]-benz- tropine is a suitable ligand for use with PET for investi- gations of the neuroanatomical distribution of the mus- carinic cholinergic system. Further studies designed to address questions of [N-llC-methyll-benztropine bind- ing in the periphery as well as investigations probing interactions between the muscarinic and other function- ally linked neurotransmitter systems in schizophrenia are in progress. Age-related alterations in muscarinic receptor density are also being examined with [N-W- methyll-benztropine. In addition, work is in progress to

and K sites exist in the human CNS. In human and

on upta a e of labelled benztropine.

muscarinic ligand, [ 17 Cl-scopolamine - which is rapidly

222 S.L. DEWEY ET AL.

synthesize "F-labelled benztropine. The longer half-life of 18F relative t o llC (110 minutes vs. 20.4 minutes) will increase the effective imaging time to several hours allowing for clearance of radioactivity from areas low in muscarinic receptor number such as the cerebellum. This will provide optimal delineation of receptor rich areas.

ACKNOWLEDGMENTS This work was carried out at Brookhaven National

Laboratory under contract DE-AC0276CH00016 with the U S . Department of Energy and supported by its Office of Health and Environmental Research. We also acknowledge support from the National Institutes of Health (NS-15638). We thank Dr. Henry I. Yamamura for his helpful discussion concerning the in vitro binding affinities of benztropine. The authors gratefully ac- knowledge Karin Karlstrom, Colleen Shea, Rosangela Casati, David Alexoff, and Betty Jellett, for their help in metabolite analysis and radiotracer preparation. The authors are also grateful t o Diane Clarke for her help in developing the analytical system for the labeled tracer. We would also like to thank Renee Moadel for PET operations; Don Warner, Clarence Barrett, and Robert Carciello for cyclotron operations; and Noelwah Netusil, R.N., Mary Franchi, R.N., and Ted Johnson for expert patient care.

REFERENCES Armstrong, D.M. (1986) Ultrastructural characterization of choline

acetyltransferase-containing neurons in the basal forebrain of rat: Evidence for a cholinergic innervation of intracerebral blood vessels. J . Comp. Neurol., 2503-92.

Arnett, C.D.. Wolf, A.P.. Shiue, C.-Y.. Fowler. J.S.. MacGregor. R.R.. Christman, D.R., and Smith, M. (1986) Improved delinueation of human dopamine receptors using ['*FI-N-methylspiroperidol and PET. J. Nucl. Med., 27:187%1882.

Banholzer, R., Heusner, A., and Schulz, W. (1975) Entmethylierung von Tropanderivaten mit Phosgen. Liebigs. Ann. Chem., 2227-2231.

Bartholini, G., Stadler, H., and Lloyd, K.G. (1973) Cholinergic-dopa- minergic interactions in the extrapyramidal system. Adv. Neurol., n 0r.n 0,. J:LJJ-Z41.

Coyle, J.T., and Snyder, S.H. (1969) Antiparkinsonian drugs: Inhibi- tion of dopamine uptake in the corpus striatum as a possible mecha- nism of action. Science, 166:809-811.

Dam, M., and London, E. (1984) Glucose utilization in the Papez circuit: Effects of oxotremorine and scopolamine. Brain Res.,

Dannals, R.F., Langstrom, B., Ravert, H.T., Wilson, A.A., and Wagner, H.N. (1988) Synthesis of radiotracers for studvinn muscarinic cholin-

295: 137-144.

ergic receptors in the living human brain u"sing positron emission tomography: ['lCldexetimide and [l'Cllevetimide. Appl. Radiat. Isot.. 39:291-295. ~ ~~~

Davies, P., and Verth, A.H. (1977) Regional distribution of the musca- rinic acetylcholine receptor in normal and Alzheimer's-type demen- tia brain. Brain Res., 138:385-392.

Dewey, S.L., MacGregor, R.R., Bendriem, B., Fowler, J.S., King, P.T., Schlyer, D.J., Wolf, A.P., Volkow, N.D., Christman, D.R., Brodie, J.D., and Hitzemann, R. (1989a) [l'CI-benztropine as a possible muscarinic cholinergic receptor ligand for use in PET. J. Nucl. Med., 30:741.

Dewey, S.L., MacGregor, R.R., Brodie, J.D., Volkow, N.D., Fowler, J.S., Schlyer, D., Wolf, A.P., and Bendriem, B. (1989b) I"CI-benztropine as a possible muscarinic receptor ligand in human brain for PET. SOC. Neurosci. Abstr., 15:96.5.

Dewey, S.L., Wolf, A.P., Fowler, J.S., Brodie, J., Shiue, C.-Y., Alavi, A,, Hiesiger, E., Schlyer, D., Volkow, N., Raulli, R., and Christman, D. (1988) The effects of central cholinergic blockade on [lRF1-N-methyl- spiroperidol binding in human brain using PET. XVI C.I.N.P. Con- gress.

Drachman, D.A., and Leavitt, J. (1974) Human memory and the cholinergic system: A relationship to aging? Arch. Neurol., 30: 1 13-121.

Drayer, B., Jaszczak, R., Coleman, E., Storni, A., Greer, K., Petry, N., Lischko, M., and Flanagan, S. (1982) Muscarinic cholinergic receptor binding: In vivo dejjction using sin& photon emission computed tomography and ra ioiodinated quinuclidinyl benzilate. J. Comput. Assist. Tomogr., 6:536-543.

Dubois, B., Maya, W., Agid, Y., Le Moal, M., and Simon, H. (1985) Profound disturbance of spontaneoua and learned behaviors follow- ing lesions of the nucleus basalis magnocellularis in the rat. Brain Res., 338:249-258.

Elhert, F.J., Roeske, W.R., and Yamamura, H.I. (1981) Striatal musca- rinic receptors: Regulation by dopaminergic agonists. Life Sci., 28:2441-2448 ~~ ~ ~ . _

Fowler, J.S., Volkow, N.D., Wolf, A.P., Dewey, S.L., Schlyer, D.J., MacGregor, R.R., Hitzemann, R., L(lgan, J., Bendriem, B., Gatley, S.J., and Christman, D. (1990) Mapping cocaine binding sites in human and baboon brain in vivo. Synapse, in Press.

Frey, K.A., Hichwa, R.D., Ehrenkaufer, R.L.E., and Agranoff, B.W. (1985) Tracer kinetic modeling of niuscarinic cholinergic receptor binding. Proc. Natl. Acad. Sci. U.S.A., 82:6771-6775.

Hammer, R., Berrie, C.P., Birdsall, N.J.M., Burgen, A.S.V., and Hulme, E.C. (1980) Pirenzepine distinguishes between different subclasses of muscarinic receptors. hature, 283:90-92.

Hammer, R., and Giachetti, A. (1982) Muscarinic receptor subtypes: M, and M, biochemical and functional characterization. Life Sci.,

Helen, P., and London, E.D. (1984) Muscimol-scopolamine interactions in the rat brain: A study with 2-deoxy-D-ll-'4Clglucose. J. Neurosci.,

Honer, W.G., Prohovnik, I., Smith, G., and Lucas, L.R. (1988) Scopola- mine reduces frontal cortex perfusion. J. Cereb. Blood Flow Metab., 8:635441.

Kearfott, K.J., Rottenberg, D.A., and Knowles, R.J.R. (1984) A new headholder for PET, CT, and NMR imaqng. J . Comput. Assist, Tomogr., 8: 12 17-1 220.

Langstrom, B., and Lundqvist, H. (1'376) The preparation of W- methyl iodide and its use in the synthesis of l'C-methyl-L- methionine. Int. J. Appl. Radiat. Isot., 27:357-363.

Lin, S.-C., Olson, K.C., Okazaki, H., and Richelson, E. (1986) Studies on muscarinic binding sites in human brain identified with ['Hlpirenzepine. J. Neurochem., 46:2 74-279.

McGeer, P.L., Grewaal, D.S., and McGeer, E.G. (1974) Influence of noncholinergic drugs on rat striatal acetylcholine levels. Brain Res., 50:211-217.

McGeer, P.L., and McGeer, E.G. (19,341 Cholinergic systems and cholinergic pathology. In: Handbook of Neurochemistry. A. Lajtha, ed. Plenum Press, New York, pp. 379-410.

Meltzer, H.Y., and Stahl, S.M. (1976) The dopamine hypothesis of schizophrenia: A review. Schizophr. EM. , 2:19.

Mountjoy, C.Q. (1986) Correlations between neuropatholoqcal and neurochemical changes. Br. Med. Bul I., 42:81-85.

Mulholland, G.K., Otto, C.A., Jewett, D M., Kilbourn, M.R., Sherman, P.S., Koeppe, R.A., Frey, K.A., and Kuhl, D.E. (1989) Radiosynthesis and comparisons in the biodistributicn of carbon-11 labeled musca- rinic antagonists: ( t 1-2-tropanyl benzilate and N-methyl-4- piperidyl benzilate. J. Label Compds. Radiopharm., 26:202-203.

Patlak, C.S. (1981) Derivation of equations for the steady-state reac- tion velocity of a substance based on the use of a second substance. J . Cereb. Blood Flow Metab., 1:129-131

Price, D.L., Whitehouse, P.J., and Strubie, R. G. (1982) Basal forebrain cholinergic systems in Alzheimer's disease and related dementias. Neurosci. Commun., 1:84-92.

Roth, R.H., and Bunney, B.S. (1976) Ir teractions of cholinergic neu- rons with other chemically defined neiironal systems in the CNS. In: Biology of Cholinergic Function. I. Hanin and A. Goldberg, eds. Raven Press, New York, pp. 379-395.

Sitaram, N., Weingartner, H., and Gillin, J. (1978) Human senile learning: Enhancement with anecoline and impairment with scopo- lamine correlated with performance on placebo. Science, 201: 274-276.

Snyder, S., and Yamamura, H.I. (19'17) Antidepressants and the muscarinic acetylcholine receptor. Arch. Gen. Psychiatry, 34: 236-239.

Tandon, R., and Greden, J.F. (1989) Cholinergic hyperactivity and negative schizophrenic symptoms. Arch. Gen. Psychiatry, 46: 745-753.

Wagner, H.N. (1986) Quantitative imaging of neuroreceptors in the living human brain. Semin. Nucl. Metl., 16:51-62.

Watson, M., Roeske, W.R., Vickroy, T.W., Smith, T.L., Akiyama, K., Guyla, K., Duckles, S.P., Serra, M., Adem, A., Nordberg, A., Gehlert, D.R., Wamsley, J.K., and Yamamura, H.I. 11986) Biochemical and

31~2991-2998.

6~1405-1413.

[N-"C-METHYL]-BENZTROPINE BINDING IN BRAIN 223

functional basis of putative muscarinic receptor subtypes and its implications. Trends Pharmacol. Sci., 7:4&55.

Weinberger, D.R., Gibson, R.E., Coppola, R., Jones, D.W., Braun, A.R., Mann, U., Berman, K.F., Sunderland, T., Chase, T.N., and Reba, R.C. (1989) Distribution of muscarinic receptors in patients with dementia:Acontrolled~tudyof~~~I QNB and SPECT. J. Cereb. Blood Flow Metab., 9:S537.

Weinberger, J., Greenberg, J.H., Waldman, M.T.G., Sylvestro, A,, and

Reivich, M. (1979) The effect of scopolamine on local glucose metab- olism in elderly people. Brain Res., 177:337-345.

Whitehouse, P.J., Price, D.L., and Struble, R.G. (1982) Alzheimer's disease and senile dementia: Loss of neurons in the basal forebrain. Science, 215:1237-1239.

Xu, M., Mizobe, F., Yamamoto, T., and Kato, T. (1989) Differential effects of M ~ and M muscarinic drugs on striatal dopamine release and metabohsm in fZrkely moving rats. Brain Res., 495:232-243.