www.elsevier.com/locate/jns

Journal of the Neurological Scienc

[11C]Vinpocetine: a prospective peripheral benzodiazepine receptor

ligand for primate PET studies

Balazs Gulyasa,b,*, Christer Halldina, Adam Vasc, Richard B. Banatid,e, Evgeny Shchukina,

Sjoerd Finnemaa, Jari Tarkainena, Karoly Tihanyic, Geza Szilagyif, Lars Fardea

aKarolinska Institute, Psychiatry Section, Department of Clinical Neuroscience, Karolinska Hospital, S-17176 Stockholm, SwedenbKarolinska Institute, Department of Neuroscience, S-171 77 Stockholm, Sweden

cChemical Works of Gedeon Richter Ltd., Budapest, H-1103 HungarydMRC Clinical Sciences Centre, Hammersmith Hospital (Cyclotron Unit) and Department of Neuropathology,

Division of Neuroscience and Psychological Medicine Faculty of Medicine, Imperial College, London W6 8RF, United KingdomeRamaciotti Centre for Brain Imaging (Brain-Mind Research Institute) and School of Medical Radiation Sciences, University of Sydney,

East Street PO Box 170, Lidcombe NSW 1825, AustraliafNational Stroke Center, National Institute of Neurology and Psychiatry, H-1021 Budapest, Hungary

Available online 16 December 2004

Abstract

Vinpocetine, a synthetic derivative of the Vinca minor alkaloid vincamine, is a widely used drug in neurological practice. We tested the

hypothesis that vinpocetine binds to peripheral benzodiazepine binding sites (PBBS) and is therefore a potential ligand of PBBS. Positron

emission tomography (PET) measurements in two cynomolgous monkeys showed that pretreatment with vinpocetine markedly reduced the

brain uptake of [11C]PK11195, a known PBBS radioligand. On the other hand, whereas pretreatment with PK11195 increased the brain

uptake of [11C]vinpocetine due to the blockade of PBBS in the periphery, it significantly reduced the binding potential (BP) values of

[11C]vinpocetine in the whole brain and in individual brain structures to PK11195. These findings indicate that, whereas the two ligands have

different affinities to PBBS, vinpocetine is a potent ligand of PBBS, which in turn suggests that the pharmacological activity of vinpocetine

may involve the regulation of glial functions.

D 2004 Elsevier B.V. All rights reserved.

Keywords: Vinpocetine; PK11195; Peripheral benzodiazepine binding site (PBBS); Monkey brain; Glia; Positron emission tomography (PET)

1. Introduction

Vinpocetine (ethyl-apovincaminate) is a synthetic com-

pound, structurally related to the Vinca minor alkaloid

vincamine. The drug (CavintonR, Gedeon Richter, Buda-

pest, Hungary) has since 1978 been widely used as a

neuroprotective agent for the prevention and treatment of

various cerebrovascular diseases [1].

Whereas the drug’s clinical efficacy has been docu-

mented by several trials, the mechanisms of action have

0022-510X/$ - see front matter D 2004 Elsevier B.V. All rights reserved.

doi:10.1016/j.jns.2004.11.032

* Corresponding author. Karolinska Institute, Psychiatry Section,

Department of Clinical Neuroscience, S-171 76 Stockholm, Sweden. Tel.:

+46 8 5248 7858; fax: +46 8 308 572.

E-mail address: balazs.gulyas@neuro.ki.se (B. Gulyas).

been a subject for physiological, pharmacological and

neuroimaging studies over the past decades. It has been

demonstrated that vinpocetine is a potent inhibitor of the

voltage-dependent Na+ channels and the Ca2+/calmodu-

line-dependent phosphodiesterase 1 [1]. Using positron

emission tomography (PET), we have shown in both

monkeys and man that [11C]vinpocetine rapidly enters the

brain after intravenous injection, the maximal uptake

being approximately 4% of the total injected radioactivity

[2,3]. The distribution pattern of vinpocetine in the brain

is heterogenous, with the highest uptake in the thalamus,

basal ganglia and visual cortex. A similar distribution,

with absence of any evident binding to transmitter

receptors or transporters, has been demonstrated ex vivo

by [11C]vinpocetine autoradiography on cryosectioned

es 229–230 (2005) 219–223

B. Gulyas et al. / Journal of the Neurological Sciences 229–230 (2005) 219–223220

whole hemisphere human brain slices [4]. Brain imaging

has shown that vinpocetine administered in pharmacolog-

ical doses increases the regional cerebral glucose uptake

and glucose metabolism in the intact brain tissue in

peristroke regions [5,6].

A 2-week-long treatment increased the regional

cerebral blood flow especially in the thalamus, basal

ganglia and visual cortex of the nonaffected hemisphere

[7].

This presence of a region-specific pattern of [11C]vin-

pocetine binding in brain that corresponds to physio-

logical changes in cerebral blood flow and metabolism in

the same brain regions supports that vinpocetine binds to

biologically active specific sites that may mediate the

known direct clinical effects of the drug.

Recent in vitro studies confirm the receptogram of

vinpocetine and show that the drug has the highest

binding affinity to peripheral benzodiazepine binding sites

(PBBS) in the rat heart (IC-50=0.2 AM) (Table 1). This

affinity is higher than the affinity of other previously

reported sites of vinpocetine binding in the brain (cf. e.g.,

Refs. [2,8]).

Based on the relatively high affinity for PBBS, we

hypothesised that in vivo binding of vinpocetine competes

with the binding of the isoquinoline PK11195, a reference

ligand for the PBBS [9]. To explore this hypothesis, we

measured the brain uptake and distribution of [11C]vinpo-

cetine and [11C]PK11195 in two cynomolgous monkeys at

baseline conditions and during pretreatment with the

unlabelled ligands.

2. Methods

2.1. Animals and ethical committee approval

Two female cynomolgous monkeys (weight: 4.0 and 7.1

kg; age: 4 and 8 years, respectively) were used in

consecutive PET measurements. The study was covered

Table 1

In vitro receptorial effects of vinpocetine (PanLabs, SpectrumScreenRresults, best 10 values selected)

Receptor Tissue IC-50

(AM)

Benzodiazepine,

peripheral

Rat heart 0.2

Adrenergic a2B Rat kidney cortex 0.9

Adrenergic a2A Human recombinant 1.9

Na+ channel, site 2 Rat brain 1.9

Ca++ channel, L-type,

benzothiazepin

Rat cerebral cortex 2.1

Adrenergic a1A Rat submaxillary gland 2.9

Ca++ channel, L-type,

phenylalkylamine

Rat brain 4.1

Dopamine D4.2 Human recombinant 7.9

Adenosine A1 Rat brain 8.3

by permission of the Animal Research Ethical Committee of

the Northern Stockholm Region (1998/14).

2.2. Radiochemistry

2.2.1. Synthesis of [11C]vinpocetine

The synthesis was performed according to a procedure

previously described [2]. Vinpocetine (Batch No.: K6B0110)

and the precursor for the radiochemical synthesis, apovinca-

minic acid (Batch No.: H-2190), were obtained from Gedeon

Richter. The specific radioactivity obtained at time of

injection of [11C]vinpocetine was higher than 2.8 GBq/Amol,

mol, corresponding to a total mass injected of less than 20 Ag.The radiochemical purity was better than 99%.

2.2.2. Synthesis of [11C]PK11195

The synthesis was performed according to a procedure

previously described [10]. The specific radioactivity obtained

at time of injection of [11C]PK11195was higher than 50GBq/

Amol, corresponding to a total mass injected of less than 2 Ag.The radiochemical purity was better than 99%.

2.3. The PET system

The PET system used was ECAT EXACT HR47 (Sie-

mens), with a spatial resolution (in-plane) of about 3.8 mm

full width at half maximum (FWHM). The system was used

in the three-dimensional mode, and the reconstructed volume

was displayed as 47 horizontal sections with a center-to-

center distance of 3.125 mm. The axial field-of-view was 15

cm. Radioactivity in the brain was measured continuously

according to a preprogrammed sequence. Each PETmeasure-

ment was made according to a 63-min-long data acquisition

protocol consisting of 15 time frames (3�1 min, 4�3 min,

8�6 min). Attenuation correction was obtained from a 10-

min transmission scan using a 68Ge radiation source.

2.4. Experimental procedures

The monkeys were anaesthetised with repeated intra-

muscular injections of a combination (50–50%; 0.8–1.0 ml/

h) of ketamine (KetalarR, Parke-Davis; 50 g/ml) and

xylazine (RompunR vet., Bayer; 20 mg/ml). The monkeys

were positioned in the PET system with the help of a head

fixation device that maintained the same head position

during and between the measurements. The image planes

were parallel to that defined by the canto-meatal line. In

each monkey, a cannula was fixed in a sural vein for

intravenous administration of the tracer.

The first monkey received two injections [66 and 66 MBq

(1.78 mCi)] of [11C]PK11195. After the first injection, a

baseline PET measurement was made. Eight minutes prior to

the second injection, 3 mg/kg of unlabelled vinpocetine was

injected intravenously (pretreatment measurement). The

second monkey received two injections [63 and 66 MBq

(1.70 and 1.78 mCi)] of [11C]vinpocetine. After the first

Fig. 1. Global radioactivity uptake vs. time curves in the brain. Radio-

activity levels are expressed as percentage of total injected radioactivity. (A)

[11C]PK11195: baseline and pretreatment conditions. (B) [11C]vinpocetine:

baseline and pretreatment conditions.

Fig. 2. (A) MRI brain scan of one of the monkeys used in the experiments. This an

PET images of themonkey brain in Talairach stereotactic convention, z=2mm. (B)M

volume of interest (VOI) used as the reference region for determining binding po

condition. (D) The brain distribution [11C]PK11195 in the pretreatment condition fo

[11C]vinpocetine in baseline condition. (F) The brain distribution of [11C]vinpocetine

(For interpretation of the references to colour in this figure legend, the reader is ref

B. Gulyas et al. / Journal of the Neurological Sciences 229–230 (2005) 219–223 221

injection, a baseline PET measurement was made. In the

pretreatment measurement, 9 min prior to the injection, 1 mg/

kg of unlabelled PK11195 was injected intravenously.

2.5. Image analysis and calculations

The individual brain images were stereotactically

standardized using the monkey version of the compu-

terised brain atlas system of the Karolinska Institutet

[11]. Global and regional radioactivity values were

measured in whole brain and selected volumes of interest

(VOIs). Global and regional uptake and distribution

measurements were made on the basis of the summation

images representing an average for the period between 9

and 63 min after tracer administration. The tissue uptake

of labelled vinpocetine and PK11195 and the regional

and whole brain binding potential (BP) values were

determined by using the linear graphical analysis accord-

ing to Logan et al. [12] using the cerebellar cortex as

reference region (cf. Fig. 2) and corresponding to the

linear part of the plot covering the last 39–63 min of

measurement.

3. Results

3.1. Global brain radioactivity and cerebral kinetics of

[11C]PK11195 and [11C]vinpocetine

After iv injection of [11C]PK11195, 0.77% of total

injected activity was found in brain between 9 and 63

d the following pictures show stereotactically standardized individual MR or

RI scan, horizontal slice, z=�10mm. In yellow: the contours of the cerebellar

tential (BP) values. (C) The brain distribution of [11C]PK11195 in baseline

llowing pretreatment with 3 mg/kg vinpocetine. (E) The brain distribution of

in the pretreatment condition following pretreatment with 1mg/kg PK11195.

erred to the web version of this article.)

Fig. 3. (A) Radioactivity uptake curves in selected brain regions. Regional

radioactivity levels are expressed in nCi/ml. (A) [11C]PK11195: baseline

and pretreatment conditions. (B) [11C]vinpocetine: baseline and pretreat-

ment conditions.

B. Gulyas et al. / Journal of the Neurological Sciences 229–230 (2005) 219–223222

min after injection. The maximum fraction of radio-

activity in brain was 1.17% at 1.5 min after radioligand

injection. In the pretreatment condition, only 0.60% of

total injected radioactivity was found in the brain

(compared to the baseline condition, �22% change).

The maximum fraction of radioactivity in brain under

pretreatment condition was 1.00% at 1.5 min after tracer

injection (Fig. 1A).

After iv injection of [11C]vinpocetine, 3.84% of total

injected activity was found in brain between 9 and 63 min

after injection, and the maximum amount of radioactivity

in the brain was 5.90% at 7.5 min after tracer injection.

After pretreatment with unlabelled PK11195, the average

amount of radioactivity in brain was 5.96% of the total

injected radioactivity (compared to the baseline condition:

36% increase). The maximum amount of radioactivity in

the brain (6.32%) was at 10.5 min after tracer injection

(Fig. 1B).

Table 2

Binding potential (BP) values in selected brain VOIs to [11C]vinpocetine (baselin

Cerebrum Mesencephalon Thalamus Caudate

nucleus

P

Baseline 0.60 0.33 0.99 0.77

Pretreatment 0.36 0.13 0.71 0.50

Change in % �40 �61 �28 �35 �Reference region: cerebellar grey matter, cf. Fig. 2B.

3.2. Regional radioactivity distribution in brain of

[11C]PK11195 and [11C]vinpocetine

In Fig. 2, average PET summation images, obtained after

the injection of [11C]PK11195 and [11C]vinpocetine, are

shown, displaying both baseline and pretreatment conditions.

The changes of regional radioactivity values in selected brain

volumes of interest (VOIs) are shown in Fig. 3.

3.3. Binding potential values before and after pretreatment

The binding potential (BP) was calculated using a linear

graphical approach. The BP for [11C]PK11195 was 0.07 and

0.06 at baseline and pretreatment conditions, respectively.

Due to the low brain uptake of the ligand, no reasonable

binding potential (BP) analysis could have been performed

for individual structures in the brain. On the other hand, the

BP values in the parotis, an organ with high PBBS density,

were 1.95 and 1.70 (left and right, respectively) at baseline

and 1.50 and 1.45 after pretreatment with vinpocetine

(�19% change).

The BP values in the case of [11C]vinpocetine are shown in

Table 2. As shown in the table, there was amarked decrease in

global BP values in the cerebrum (�40%), as well as in all

selected brain regions, following pretreatment with PK11195.

4. Discussion

In line with earlier observations [13], the brain uptake of

PK11195 was low, and only 0.77% of the total injected

radioactivity of [11C]PK11195 entered the brain. This

amount was markedly reduced following pretreatment with

vinpocetine. This indicates that vinpocetine may bind to

binding sites that are occupied by PK11195. However, full

examination of regional BP values of [11C]PK11195 is

limited due to the low cerebral concentration of the ligand

and the low density of the PBBS in the normal brain as

demonstrated by autoradiography [14].

In line with earlier observations [2], [11C]vinpocetine

passed the blood–brain barrier and entered the brain readily.

Of the total injected radioactivity, 3.84% was in the brain,

and this fraction markedly increased to 5.94% following

pretreatment with the known PBBS ligand PK11195. This

behaviour is characteristic for pretreatment conditions of

those receptors which have a relatively large number in the

periphery, and the blockade of the peripheral receptors by a

e and pretreatment with 1 mg/kg PK11195)

utamen Occipital

cortex

Temporal

cortex

Parietal

cortex

Frontal

cortex

Cingulate

cortex

0.97 0.50 0.50 0.71 0.60 0.69

0.68 0.34 0.26 0.48 0.33 0.40

30 �32 �48 �32 �45 �42

B. Gulyas et al. / Journal of the Neurological Sciences 229–230 (2005) 219–223 223

known ligand, administered in pharmacological doses (in the

present case by PK11195), results in an increased amount of

a radioligand, selective for the same receptor type and

administered in tracer doses, in the brain.

The two pretreatment conditions resulted in fundamen-

tally different patterns. Compared to the baseline con-

dition following the injection of [11C]PK11195,

pretreatment with vinpocetine resulted in a lower global

brain radioactivity uptake (0.77% and 0.60% of the total

injected radioactivity, respectively). Compared to the

baseline condition following the injection of [11C]vinpo-

cetine, in the case of pretreatment with PK11195, the

global brain radioactivity uptake was markedly higher

(3.84% and 5.96%, respectively). The measured radio-

activity uptake in the brain is the result of mainly two

factors: (i) the relative relationship between both ligands

with regard to their respective ratios between the affinity

of ligand binding in the periphery as compared to that in

the brain and (ii) the relative differences in the extraction

of both ligands. For the cold ligand to displace the hot

ligand in the peripheral binding compartment more than

in the brain and thus result in relatively higher uptake

values in the brain (as was the case of pretreatment with

PK11195 prior to the injection of [11C]vinpocetine), the

affinity of the hot ligand in the periphery has to be less

than in the brain. This affinity difference, together with

the marked difference in the brain uptake values of the

two ligands, may explain the above observation. As at

this time no data are available regarding the two ligands’

relative affinity values, further experiments are needed to

clarify this issue.

Despite the fact that approximately five times more

vinpocetine enters the brain than PK11195, pretreatment

with the known PBBS ligand PK11195 reduces markedly

the BP values for vinpocetine in the brain, supporting the

fact that the two ligands bind to the same receptor site (and

that PK11195 does enter the brain albeit much less than

vinpocetine).

These observations taken together, the present experi-

ments support the hypothesis that vinpocetine is a ligand

with relatively high affinity for PBBS. [11C]vinpocetine can

be used as a potential PET marker of PBBS and

consequently glia cells in the primate brain. Furthermore,

the present observations may support the view that, if

indeed the PBBS, which is absent in neurons, is a major

binding site for vinpocetine, its therapeutic action is likely to

involve the modulation of glial cells [15,16].

Acknowledgement

The authors express their gratitude to Ms. Kjerstin Lind

and Mr. Julio Gabriel for their technical assistance during

the experiments.

References

[1] Bfnfczk P, Gulyas B, Adam. Vizi V, Nemes A, Karpati E, Kiss B, et al.

Role of sodium channel inhibition in neuroprotection: effect of

vinpocetine. Brain Res Bull 2000;53:245–54.

[2] Gulyas B, Halldin C, Karlsson P, Chou Y-H, Swahn C-G, Bfnfczk P,

et al. Brain uptake and plasma metabolism of [11C]vinpocetine: a

preliminary PET study in a Cynomolgous monkey. J Neuroimaging

1999;9:217–22.

[3] Gulyas B, Halldin C, Sandell J, Swahn C-G, Bfnfck P, Kiss B, et al.

PET studies on the uptake and regional distribution of [11C]vinpoce-

tine in human subjects. Acta Neurol Scand 2002;106:325–32.

[4] Hall H, Varn7s K, Sandell J, Halldin C, Farde L, Vas A., et al.

Autoradiographic evaluation of [11C]vinpocetine binding in the

human postmortem brain. Acta Biol Hung 2002;53:59–66.

[5] Szakall S, Boros I, Balkay L, Emri M, Fekete I, Kerenyi L, et al. The

cerebral effects of a single-dose intravenous vinpocetine in chronic

stroke patients: a PET study. J Neuroimaging 1998;8:97–204.

[6] Gulyas B, Csiba L, Kerenyi L, Galuska L, Tron L. PET studies

on chronic stroke patients before and after a single-dose

intravenous Cavinton infusion. In: Gulyas B, Mqller-G7rtnerHW, editors. Positron emission tomography: a critical assessment

of recent trends. Dordrecht7 Kluwer Academic Publisher; 1998.

p. 291–306.

[7] Vas A., Gulyas B, Szabo Z, Bfnfczk P, Csiba L, Kiss B, et al. Clinicaland non-clinical investigations using positron emission tomography,

near infrared spectroscopy and transcranial Doppler studies on the

neuroprotective drug vinpocetine: a summary of evidences. J Neurol

Sci 2002;203–204:259–62.

[8] Gulyas B, Sovago J, Sandell J, Halldin C, Cselenyi ZM, Vas A., et al.

Drug distribution in man: a positron emission tomography study after

oral administration of the labelled neuroprotective drug vinpocetine.

Eur J Nucl Med 2002;29:1031–8.

[9] Benavides J, Cornu P, Dennis T, Dubois A, Hauw JJ, MacKenzie ET,

et al. Imaging of human brain lesions with an omega 3 site

radioligand. Ann Neurol 1988;24(6):708–12.

[10] Cremer JE, Hume SP, Cullen BM, Myers R, Manjil LG, Turton DR,

et al. The distribution of radioactivity in brains of rats given [N-

methyl-11C]PK 11195 in vivo after induction of a acortical ischaemic

lesion. Nucl Med Biol 1992;19:159–66.

[11] Roland PE, Graufelds CJ, W7hlin J, Ingelman L, Andersson M,

Ledberg A, et al. Human brain atlas: for high-resolution functional and

anatomical mapping. Hum Brain Mapp 1994;1:173–84.

[12] Logan J, Fowler JS, Volkow ND, Wang GJ, Ding YS, Alexoff DL.

Distribution volume ratios without blood sampling from graphical

analysis of PET data. J Cereb Blood Flow 1996;16:834–40.

[13] Petit-Taboue MC, Baron JC, Barre L, Travere JM, Speckel D,

Camsonne R, et al. Brain kinetics and specific binding of [11C]PK

11195 to omega 3 sites in baboons: positron emission tomography

study. Eur J Pharmacol 1991;200(2–3):347–51.

[14] Banati RB, Newcombe J, Gunn RN, Cagnin A, Turkheimer F,

Heppner F, et al. The peripheral benzodiazepine binding site in the

brain in multiple sclerosis: quantitative in vivo imaging of microglia

as a measure of disease activity. Brain 2000;123:2321–37.

[15] Banati RB, Gehrmann J, Schubert P, Kreutzberg GW. Cytotoxicity of

microglia. Glia 1993;7(1):111–8.

[16] Banati RB. Visualising microglial activation in vivo. Glia 2002;40(2):

206–17.