Visualizing dopamine transporter integrity with iodine-123-FP-CIT SPECT in combination with high resolution MRI in the brain of the common marmoset monkey

Page 1

C

Vcm

EGa

b

c

d

e

f

g

h

����

a

ARRA

KMDSMP6

bposs

0h

Journal of Neuroscience Methods 210 (2012) 195– 201

Contents lists available at SciVerse ScienceDirect

Journal of Neuroscience Methods

jou rna l h om epa ge: www.elsev ier .com/ locate / jneumeth

linical Neuroscience

isualizing dopamine transporter integrity with iodine-123-FP-CIT SPECT inombination with high resolution MRI in the brain of the common marmosetonkey

nrique Garea-Rodrígueza,∗, Christina Schlumbohmb, Boldizsár Czéha,c, Jessica Königa,d,unther Helmse, Cornelia Heckmanna, Birgit Meller f,g, Johannes Mellerg, Eberhard Fuchsa,d

Clinical Neurobiology Laboratory, German Primate Center, Göttingen, Kellnerweg 4, D-37077 Göttingen, GermanyEncepharm, Hans-Adolf-Krebs-Weg 9, D-37077 Göttingen, GermanyMolecular Neurobiology, Max-Planck Institute of Psychiatry, Kraepelinstr. 2-10, D-80804 Munich, GermanyCenter for Molecular Physiology of the Brain (CMPB), University of Göttingen, Humboldtallee 23, D-37073 Göttingen, GermanyDepartment of Cognitive Neurology, University Medical Center, Georg-August-University Göttingen, Robert-Koch-Str. 40, D-37075 Göttingen, GermanyDepartment of Nuclear Medicine, Martin-Luther-University Halle, Ernst-Grube-Str. 40, D-06120 Halle, GermanyDepartment of Nuclear Medicine, University Medical Center, Georg-August-University Göttingen, Robert-Koch-Str. 40, D-37075 Göttingen, Germany

i g h l i g h t s

We established an imaging protocol targeting the dopamine transporter in common marmoset monkeys by 123I-FP-CIT SPECT.High resolution imaging was achieved with an upgraded clinical SPECT camera and a 3T MRI system.Combination of 123I-FP-CIT SPECT and high resolution MRI allows for the detection of small dopaminergic brain structures.The presented method is a suitable imaging tool to be used in a marmoset monkey model of Parkinson’s disease.

r t i c l e i n f o

rticle history:eceived 30 April 2012eceived in revised form 14 July 2012ccepted 16 July 2012

eywords:armoset monkeyopamine transporterPECTRI

a b s t r a c t

Considerable progress has been made in small animal single photon emission computed tomogra-phy (SPECT) imaging in the field of Parkinson’s disease. In preclinical research, there is an increasingdemand for in vivo imaging techniques to apply to animal models. Here, we report the first protocol fordopamine transporter (DAT) SPECT in common marmosets using the radioligand 123I-N-�-fluoropropyl-2�-carbomethoxy-3�-{4-iodophenyl}nortropane (123I-FP-CIT). Serial SPECT images were obtained onan upgraded clinical scanner to determine the distribution kinetics of 123I-FP-CIT in the marmoset brain.After intravenous injection of approximately 60 MBq of the radiotracer 123I-FP-CIT, stable and specificstriatal uptake was observed for at least 4 h. Analysis of plasma samples showed rapid disappearanceof the radiotracer from blood plasma within a few minutes after application, with activity declining to

arkinson’s disease-OHDA

4.1% of the administered activity. Structural magnetic resonance imaging (MRI) at 400 �m resolutionprovided the details of the underlying anatomy. In a marmoset model of Parkinson’s disease, which wasgenerated by unilateral injections of 6-hydroxydopamine (6-OHDA) into the nigro-striatal projectionpathway, complete loss of striatal DAT binding in combination with behavioral deficits was observed.The presented study demonstrates that 123I-FP-CIT SPECT is a suitable tool to investigate DAT integrityin preclinical studies on common marmosets.

Abbreviations: 123I-�-CIT, 123I-�-carbomethoxy-3�-(4-iodophenyl)tropane; 123I-IBZenzamide; 123I-FP-CIT, 123I-N-�-fluoropropyl-2�-carbomethoxy-3�-{4-iodophenyl}noorter; DICOM, digital imaging and communications in medicine; FOV, field-of-view; FLIRf gradient echoes; MRI, magnetic resonance imaging; NHP, non-human primate; NIFTI,

ubset expectation maximization; OWM, Old World Monkey; PD, Parkinson’s disease; Rtriatum.∗ Corresponding author. Tel.: +49 551 3851 12; fax: +49 551 3851 307.

E-mail address: [email protected] (E. Garea-Rodríguez).

165-0270/$ – see front matter © 2012 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.jneumeth.2012.07.009

© 2012 Elsevier B.V. All rights reserved.

M, 123I-(S)-(−)-3-iodo-2-hydroxy-6-methoxy-N-[(1-ethyl-2-pyrrolidinyl)methyl]-rtropane; 6-OHDA, 6-hydroxy-dopamine; CB, cerebellum; DAT, dopamine trans-T, FSL linear registration tool; MP-RAGE, magnetization-prepared rapid acquisitionneuroimaging informatics technology initiative; NR, noradrenalin; OSEM, orderedOI, region of interest; SPECT, single photon emission computed tomography; ST,

Page 2

1 euros

1

tadgvtdsDfl(F

cusaaa12

(amTbb(itpwsegm

m2S2Fhm3(

bl1

TpaP

TO

96 E. Garea-Rodríguez et al. / Journal of N

. Introduction

Parkinson’s disease (PD) is a disabling neurodegenera-ive disease characterized by progressive neuronal cell lossnd degeneration of nerve terminals of the nigro-striatalopaminergic system (Dauer and Przedborski, 2003). Sin-le photon emission computed tomography (SPECT) is aaluable technique for clinical diagnosis and for studyinghe pathogenesis of PD, e.g. by quantifying the decrease ofopamine transporters (DATs) in the striatum. As demon-trated in many clinical trials, a reliable marker to imageATs is the �-emitting radioiodinated cocaine analog 123I-N-�-uoropropyl-2�-carbomethoxy-3�-{4-iodophenyl}-nortropane

123I-FP-CIT) (Baldwin et al., 1995; Neumeyer et al., 1994; de lauente-Fernández, 2012).

There is an increasing demand for in vivo imaging appli-ations at the level of preclinical and translational researchsing non-human primates (NHPs), especially for long-termtudies in animal models of Parkinson’s disease. Tradition-lly, the most commonly used NHP species in PD researchre Old World monkeys (OWMs), such as baboons, rhesusnd cynomolgus monkeys, (Baldwin et al., 1995; Booij et al.,997; Prunier et al., 2003; Laruelle et al., 1993; Ma et al.,002).

Despite less frequent usage, the common marmoset monkeyCallithrix jacchus), a New World monkey species, provides a highlyttractive model for preclinical research in PD. The common mar-oset monkey is a small, day-active, NHP that is native to Brazil.

he animals are not threatened or endangered in the wild andreed well in controlled laboratory conditions. They have a smallody size of around 25 cm in length and a weight of 300–450 gMansfield, 2003). Depending on the topic or substance undernvestigation, utilization of marmoset monkeys can prove advan-ageous over other species. This is the case, for example, whenharmacological agents for prospective use in humans cross reactith marmoset – but not with macaque or baboon – binding

ites and/or when the generation of test substances is extremelyxpensive. Due to their much lower body mass relative to OWMs,enerally less than 1/10th of a test substance is needed in the mar-oset compared to OWMs (Mansfield, 2003; Orsi et al., 2011).Since the development of pinhole and multi-pinhole colli-

ation (Beekman and van der Have, 2007; Schramm et al.,003), considerable progress has been made in small animalPECT imaging in the field of Parkinson’s disease (Booij et al.,002; Andringa et al., 2005; Lauwers et al., 2007; Alvarez-ischer et al., 2007). Nevertheless, only Saji and colleaguesave performed a PD-related brain SPECT study using commonarmosets and two radiotracers the 123I-2�-carbomethoxy-

�-(4-iodophenyl)tropane (123I-�-CIT) and 123I-iodobenzamide123I-IBZM) (Saji et al., 2003).

To the best of our knowledge, 123I-FP-CIT SPECT has noteen applied in marmosets. Yet, future preclinical models uti-

izing this technology have a high translational value since23I-FP-CIT SPECT is well established in clinical diagnostics.

herefore, the aim of this study was to establish an imagingrotocol for 123I-FP-CIT SPECT in common marmoset monkeysnd to validate its use in the unilateral 6-OHDA model ofD.

able 1verview of animals used in the present study.

Sex Age (year) Study protocol

Marmoset #1 M 5 Pharmacokinetic study: shortMarmoset #2 M 12 6-OHDA lesioning and behavi

study at age of 12 years for hiMarmoset #3 M 3 Control animal: high resolutio

cience Methods 210 (2012) 195– 201

2. Materials and methods

2.1. Animals

Three adult (3, 5 and 12 years old) male common mar-moset monkeys (Callithrix jacchus) (350–450 g) were used in thisstudy (Table 1). The animals were obtained from the breed-ing colony at the German Primate Center (Göttingen, Germany)and pair-housed in a temperature- (25 ± 1 ◦C) and humidity-controlled (65 ± 5%) facility under a 12-h day/night cycle. Each cage(80 cm × 150 cm × 66 cm, Ebeco GmbH, Castrop-Rauxel, Germany)was furnished with wooden branches and shelves and containeda wooden sleeping box (24 cm × 21 cm × 18 cm). The animals werefed ad libitum with a pelleted marmoset diet (ssniff Spezialdiäten,Soest, Germany). In addition, 20 g mash per animal was served inthe morning and they received 30 g clean-cut fruits or vegetablesmixed with noodles or rice in the afternoon. Water was alwaysavailable. All animal experimentation was carried out in accor-dance with the European Council Directive of November 24 1986(86/EC) including Position 6106/20 of the EU Council of May 262010, and was approved by the Lower Saxony Federal State Officefor Consumer Protection and Food Safety, Germany.

2.2. Pharmacokinetic study

The pharmacokinetic characterization was carried out on twoseparate occasions in a single animal (marmoset #1) under gen-eral anesthesia. Anesthesia was initiated by intramuscular injectionof diazepam (0.3 mg/kg, Diazepam®, Ratiopharm, Germany), alfax-olon (10 mg/kg, Alfaxan®, Vetoquinol, UK) and glycopyrroniumbromide (0.01 mg per animal, Robinul®; Riemser, Germany).Subsequently, the animal was intubated with a polyethylene endo-tracheal tube (inner/outer diameter = 1.3/2.2 mm) and kept underinhalation anesthesia (0.4–1.0% isoflurane in 70:30 N2O O2). Theanimal received an intravenous injection into the right saphenousvein of approximately 60 MBq of 123I-FP-CIT (Datscan®, AmershamBuchler, Braunschweig, Germany). The radioactivity in the injectionsyringe was measured immediately before and after injection withan automated gamma counter (PerkinElmer, Wizard 3, Waltham,MA, USA) to correct the injected dose for physical decay. Repeatedexaminations of one animal yielded 138.1 (test) and 150.7 kBq/gbodyweight (re-test) of 123I-FP-CIT.

Blood samples were obtained from the left saphenous vein 5, 15,30, 60, 120, 180, 210 and 240 min post injection, followed by plasmacentrifugation. Arterialisation of venous blood was achieved byhyperthermia in the animal’s lower extremities, induced by athermostat-controlled infrared radiator. Radioactivity values inblood plasma were corrected for physical decay from t = 0. Thetime course function xt = a1 + a2 × e(−k×t) was fitted to the measuredplasma radioactivity values (mt) in order to calculate the initialslope/rate constant of decay and the steady state radioactivity. Theradioactivity at the time of injection (xt = 0 = a1 + a2) was calculatedfrom the injected dose and the volume of extracellular fluid, which

was assumed to be 25% of the body mass (Swan et al., 1954). It wasestimated by the quotient of injected dose and distribution vol-ume: D [Bq] ÷ (bodyweight [g] × 0.25 [ml/g]). Thus, it was implicitlyassumed that the tracer is initially distributed in the extracellular

acquisition SPECT imaging and blood samplingoral testing was started at age of 3 years. Animal was included into presentedgh resolution SPECT/MRIn SPECT/MRI

Page 3

euros

fla

dCGw

2

cuSwpysIssaa2lltaahSa1

a

2

osS0Hgm(wabiimClovc

2

t8wdo

E. Garea-Rodríguez et al. / Journal of N

uid and that there is a free exchange of 123I-FP-CIT between intra-nd extra-vasal extracellular fluid.

The method of minimizing the weighted sum of squares ofeviations

∑0→t(mt − xt)2 was used. Weight was set to 1/(mt)2.

urve fitting was performed using Sigmaplot 11 (Systat Software,ermany). In parallel with blood collection, serial SPECT imagesere taken to monitor 123I-FP-CIT brain uptake (see below).

.3. Brain SPECT

SPECT imaging was performed using a three-headed gammaamera (Prism 3000XP, Picker International, Cleveland, OH, USA)pgraded with a dedicated small-animal imaging module (HiSPECT,civis, Göttingen, Germany). Each collimator head was equippedith a pyramidal collimator and multi-pinhole aperture plate (6inholes). The inner diameter of each pinhole was 2.5 mm, whichielded a tomographic resolution of 2.1 mm and an average on axisensitivity of 960 cps/MBq. The field-of-view (FoV) was 143 cm3.mages were acquired at 256 × 256 projection matrices with a pixelize of 1.78 mm over 10 gantry steps (30 projection angles) in atep-and-shoot mode. Up to 4 energy windows were acquired. Ancquisition time of 175 s per gantry step was chosen for time-ctivity analysis (marmoset #1) resulting in a total scan time of9.8 min. To obtain high quality images, as done for the 6-OHDA-

esioned (marmoset #2) and control animal (marmoset #3), aonger acquisition time was chosen (300 s per angle and total scanime of 50.2 min). The photo peak energy was set to 159 keV with

window of ±10%. Multi energy window scatter correction waspplied. The anaesthetized animals were placed in a custom-madeolder in prone position and the head was fixed with ear bars.tandard image acquisition (marmosets #2 and #3) was initiatedfter reaching equilibrium state 2 h after injection of 50–60 MBq23I-FP-CIT. In the pharmacokinetic study (marmoset #1), imagecquisition was started immediately after application of the tracer.

.4. SPECT reconstruction and evaluation

SPECT reconstruction was performed using a dedicated iterativerdered subset expectation maximization (OSEM) algorithm con-isting of three iterations and four subsets per iteration (HiSPECT®,civis, Göttingen, Germany). The reconstructed voxel size was.6 mm3. For quantification, brain activity was determined withiSPECT software. For time-course analysis (animal # 1), rectan-ular shaped regions of interest (ROIs) of 74 mm3 volume were setanually for the left (STl) and right striatum (STr) and cerebellum

CB) in reference to the corresponding MRI. 123I-FP-CIT brain uptakeas given as (Bq/�l). Activities of left and right striatum were aver-

ged (ST = (STl + STr)/2). Specific binding in the ST was calculatedy subtracting the mean activity in the CB from the mean activity

n the ST and dividing the result by the mean activity in the CB,.e. (ST-CB)/CB. For the 6-OHDA-lesioned and control animal (ani-

als #2 and #3), ROIs of 370 mm3 volume were chosen in ST andB with standard image acquisition. The binding ratio was calcu-

ated separately for left and right striatum. A ROI with a volumef 95 mm3 was chosen for the midbrain area (substantia nigra andentral tegmental area). The binding ratio of the midbrain area wasalculated analogous to ST.

.5. Magnetic resonance imaging (MRI)

MRI was performed on a 3T clinical MR system (Magne-om TIM TRIO, Siemens Healthcare, Erlangen, Germany) using an

-channel receive coil designed for the human wrist. Marmosetsere scanned in supine position under general anesthesia withiazepam (0.3 mg/kg i.m., Diazepam, Ratiopharm, Germany), alfax-lon (10 mg/kg i.m, Alfaxan®, Vetoquinol, UK) and glycopyrronium

cience Methods 210 (2012) 195– 201 197

bromide (0.01 mg per animal i.m., Robinul®; Riemser, Germany).The total measurement time for structural MRI was about 10 min,thus eliminating the need to intubate the animal.

Structural MRI with T1-w(eighted) and T2-w(eighted) contrastwas performed at 0.4 mm isotropic resolution with non-selectiveexcitation. The three-dimensional (3D) imaging volume coveredthe head and upper chest with the read-out direction along thebody axis. The FoV in the right-left ventral-dorsal direction was51.2 mm.

T1-w contrast was achieved by magnetization-prepared rapidacquisition of gradient echoes (MP-RAGE) using parameters rec-ommended for humans (inversion time TI = 0.9 s, flip angle ̨ = 9◦,bandwidth BW = 200 Hz/pixel, echo time TE = 4.06 ms, repetitiontime TR = 2.25 s,) (Jack et al., 2008), notwithstanding minor putativedifferences in T1 (Bock et al., 2009). The matrix of 320 × 128 × 128was acquired in 4:18 min.

T2-w 3D turbo spin-echo (TSE, not shown) with variable flipangles (2 averages with phase partial Fourier, slice turbo factorof 2, yielding an echo train length of 75, effective TE = 278 ms atBW = 355 Hz/pixel; TR = 2.9 s). The matrix of 256 × 128 × 128 wasacquired in 5:26 min.

2.6. Post-processing of imaging data

The 2D digital imaging and communications in medicine(DICOM) MRI images in axial-to-coronal orientation were con-verted to 3D neuroimaging informatics technology initiative (NIFTI)volumes, changing from the radiological right-left convention toleft-right convention of the SPECT images. The T1-w volume wasaligned to a custom-made T1-w template of the marmoset headand interpolated to 0.25 mm resolution in order to achieve consis-tent angulation of the horizontal plane. This was defined by theintercommisural line, as featured in the stereotaxic atlas of theNational Institute of Neuroscience, Japan (Yuasa et al., 2010). ThisT1-w volume then served as an individual anatomical reference toalign the T2-w and SPECT data, using the FMRIB software library(FSL 4.1, Center for Functional Magnetic Resonance Imaging of theBrain, University of Oxford, UK, www.fmrib.ox.ac.uk/fsl). For spa-tial alignment, the FSL linear registration tool (FLIRT), was used todetermine the 3D rigid body transform (6 parameters) by a mutualinformation criterion. A pixel mask of the brain was defined by thecharacteristic T1-w and T2-w intensities of the brain, followed bymanual editing using FSLview.

2.7. Unilateral 6-hydroxydopamine lesion

The animal (marmoset #2) was given unilateral 6-hydroxydopamine (6-OHDA) injections into the nigrostriatalbundle according to the protocol of Annett et al. (1992). To protectthe noradrenergic and serotonergic pathways from the toxiceffects of 6-OHDA, the marmoset was pre-treated with the nora-drenalin (NA) uptake blocker talsupram (Lundbeck, Copenhagen,Denmark; 20 mg/kg) (Arnt et al., 1985) and the serotonin uptakeblocker citalopram (Lundbeck, 5 mg/kg) (Hyttel, 1982) in sterilesolution subcutaneously 30 min prior to the injections of 6-OHDA.Surgery was carried out under Alfaxalon/Alfadolon anesthesia(Saffan, Schering-Plough Ltd.; 18 mg/kg) and Diazepam (0.05 mgper animal) anesthesia and aseptic conditions. 6-OHDA [8 mg/mlfree base weight 6-OHDA hydrobromide (Sigma–Aldrich, Stein-heim, Germany) dissolved in 0.01% ascorbate-saline] was injectedstereotactically into four sites within the nigrostriatal bundle on

one side of the brain (coordinates: first site: 2 �l at AP+6.5; L+1.2,V+6; second site: 2 �l at AP+7, L+2.2, V+6.5; third site: 2 �l at AP+7,L+2.2, V+7.5; and fourth site: 3 �l at AP+6.5; L+3.2, V+7.5; accord-ing to Stephan et al. (1980). The 6-OHDA was freshly prepared

Page 4

1 euros

amwhattIBhtce

2

r(tahacrrabmsmst1tttdctwadtTAow

Fmhbt

insufficient at concentrations below 30 MBq (<80 kBq/g body-weight, data not shown). No adverse effects of the tracer wereobserved.

98 E. Garea-Rodríguez et al. / Journal of N

nd stored on ice just before use. Each injection of 6-OHDA wasade at a rate of 0.5 �l/min using a 24-gauge injection needle,hich was left in place for a further 4 min after each injectionad been made. Following surgery, the monkey was given annalgesic and kept in a warm incubator until it was well enougho be returned to its home cage. Postoperatively, the animal wasreated with meloxicam (0.2 mg/kg i.m., Metacam®,Boehringerngelheim, Germany) for 3 days and enrofloxacin (5 mg/kg i.m.,aytril®, Bayer, Germany) for 5 days. Before the lesioning, theand preference of the animal was assessed using the staircaseest (see below) and the animal was injected with 6-OHDA on theontra-lateral side of the preferred hand in order to maximize theffect of motor impairments.

.8. Behavioral test: the staircase task

The staircase task is a behavioral test to assess skilled forelimbeaching and was adapted to marmoset monkeys by Eslamboli et al.2003). A single animal (marmoset #2) was tested in the staircaseask. This test requires the animal to collect small baits arranged on

series of five steps either ascending toward the midline (staircaseill) or descending toward the midline (staircase valley). The testpparatus is made of clear Plexiglas fixed to the front of the homeage during testing. Thus, the animal is tested in a familiar envi-onment and handling is not required. The monkey is trained toeach through a vertical slot on the left or right side of the Plexiglaspparatus in order to retrieve small gum arabic pieces (∼5 mm3), sooth hands are needed for the successful completion of the task. Theonkey was presented with 2 trials for each test (staircase hill and

taircase valley). The success rate of the coordinated hand move-ents were scored as follows: grabbing and eating the reward was

cored with 1–5 points, depending on the distance of the stair fromhe vertical slot. The reward on the nearest staircase was worth

point and the most distant was worth 5 points. In many cases,he animal needed several attempts until he could finally obtainhe gum arabic pieces. Thus, if the animal could remove and eathe reward at his first attempt, then the corresponding score wasoubled. The maximum possible scores were 28 and 30 for the stair-ase hill and staircase valley test, respectively. The scores of tworials were averaged. The monkey were trained pre-lesion until heas familiar with the apparatus and could readily retrieve the gum

rabic pieces from all steps. During this period, it also became evi-ent which was the preferred hand for the individual animal, sohe subsequent lesion was made in the contra-lateral hemisphere.he animal was trained for the task for 6 months before lesioning.fter the lesion, he was tested twice a month for two years and then

nce a month for 3 years. Importantly, the animal was not food- orater-deprived prior to the test.

ig. 1. Pharmacokinetics of 123I-FP-CIT brain uptake. Representative image of ani-al #1 (left) taken at an early time point (50 min post i.v. injection of tracer) shows

igh fraction of unspecific binding. Two hours post-injection, unbound tracer haseen washed out and intense 123I-FP-CIT binding is visible in the striatum (1) andhe lacrimal glands (2) (right). Acquisition time: 29.8 min.

cience Methods 210 (2012) 195– 201

3. Results

3.1. Dose determination

The mean activity applied in the present study was 56.9 ± 6.3MBq [I-123]-FP-CIT (corresponding to 130.7 ± 19.1 kBq/g body-weight). In our experience, a suitable concentration rangesbetween 50 and 70 MBq (100–170 kBq/g bodyweight, data notshown). Concentrations above 70 MBq (170 kBq/g bodyweight)were not tested. The uptake of 123I-FP-CIT in the brain was

Fig. 2. (A) Time-activity curves for regional 123I-FP-CIT brain uptake. Repeated mea-surements were performed on two separate occasions (1st scan: circles, 2nd scan:squares) in a single animal (animal #1) with an acquisition time of 29.8 min perscan. (B) Striatal to cerebellar binding ratios increase with time. (C) Radioactivity vs.time curves for parent plasma radioactivity.

Page 5

E. Garea-Rodríguez et al. / Journal of Neuroscience Methods 210 (2012) 195– 201 199

F MRI.c triatalr

3

tith5iaowr

wias

3

t

Fwm

ig. 3. 123I-FP-CIT SPECT images of the common marmoset superimposed on T1-wontrol animal (animal #3) (upper row). Unilateral 6-OHDA lesioning of the nigro-sow, lesion is indicated by arrows). Acquisition time of SPECT images: 50.2 min.

.2. Pharmacokinetic study

Serial SPECT images were obtained to describe the distribu-ion kinetics of 123I-FP-CIT in the marmoset brain. Shortly after.v. injection of 123I-FP-CIT, the whole brain was flooded with theracer. The unbound tracer washed out over time, revealing aigh and stable uptake of 123I-FP-CIT in the striatum after about0 min (Figs. 1 and 2A). The calculated striatal to cerebellar bind-

ng ratios increased with time and reached highest values at 4 hfter injection of the tracer (Fig. 2B). Considering the stable plateauf striatal 123I-FP-CIT uptake, the optimum acquisition time-pointas determined to be between 2 and 4 h after application of the

adioligand.Analysis of blood samples showed that 123I was eliminated

ithin a few minutes following application. Thirty minutes afternjection, the plasma activity declined on average to 4.1% of thedministered activity and remained stable until the end of the mea-urement after 4 h (Fig. 2C).

.3. DAT scans of control and 6-OHDA-lesioned animals

High uptake of 123I-FP-CIT was observed in both striata inhe control animal, perfectly matching with the corresponding

ig. 4. Sagittal view of the control animal (animal #3) of the co-aligned volumes of SPECith MRI allows for detection of deep brain dopaminergic nuclei, such as the substantia niucosa (4) and the thyroid gland (5). Caudatoputamen (1).

The high tracer uptake in the striatum matches with anatomical structures in the forebrain bundle leads to complete loss of striatal DAT activity (animal #2) (lower

anatomical structures as revealed by the overlay onto MRI (Fig. 3).The highest 123I-FP-CIT uptake was observed in the more ros-tral parts of striatum, where the signals from caudate nucleusand putamen (referred to as caudatoputamen) could not be sep-arated without the anatomical information of the underlying MRI.The DAT signal was less intense in the caudal part of the stria-tum. Calculated binding ratios revealed values of 2.94 for the leftand 2.70 for the right caudatoputamen (Fig. 5B). The high sensi-tivity of SPECT in combination with the high resolution of MRIeven enabled the detection of small dopaminergic midbrain struc-tures (Fig. 4). Quantification of the signals in the substantia nigratogether with the ventral tegmental area yielded a binding ratio of1.91.

The suitability of DAT imaging was also tested in the uni-lateral 6-OHDA lesion model of PD as a proof of principle.Injections of 6-OHDA into the nigro-striatal projection path-way resulted in complete loss of striatal DAT binding (Fig. 3),which was accompanied by overt behavioral deficits (Fig. 6).Quantification of 123I-FP-CIT uptake revealed binding ratios of

4.73 for the untreated contra lateral side and 0.09 for thelesioned side (Fig. 5A). DAT scans were taken eight years afterlesion, proving long-lasting chronic effects of 6-OHDA in thismodel.

T (color overlay) and MRI (greyscale). High resolution 123I-FP-CIT SPECT co-alignedgra (2). Outside of the brain, 123I-FP-CIT accumulates in the lacrimal gland (3), nasal

Page 6

200 E. Garea-Rodríguez et al. / Journal of Neuros

Fig. 5. Striatal to cerebellar binding ratios of a unilaterally 6-OHDA-lesioned animal(animal #2) (A) and a healthy control (animal #3) (B).

Fig. 6. Long-term behavioral assessment of a 6-OHDA-lesioned monkey (animal#2). After lesioning, the animal failed to score in the staircase hill (A) and staircasevalley test (B) when using the affected left hand (filled circles). Motor impairmentremained chronic for at least 5 years.

cience Methods 210 (2012) 195– 201

3.4. Behavioral analysis

Before surgery, the animal showed similar scores for the left(26.72 ± 1.54) and right hand (26.06 ± 1.94) in the staircase hill task(Fig. 6A). Following 6-OHDA lesioning, the animal’s left hand finemotor skills were severely impaired and the animal could not com-plete the task with this hand (1st year score: 1.1 ± 1.64). In contrast,when using the right unaffected hand, the average annual score(1st year: 25.08 ± 3.54) was similar to the pre-lesion values. Finemotor skill impairments of the left hand remained stable over atime period of 5 years (2nd year: 0.13 ± 0.39, 3rd year: 0.11 ± 0.40,4th year: 0.14 ± 0.45, 5th year: 0.23 ± 0.83). As expected, fine motorskills of the right hand were not affected by lesion and remained inthe control range over the years (average annual scores: 2nd year:24.6 ± 2.73, 3rd year: 25.18 ± 4.79, 4th year: 25.73 ± 1.55, 5th year:25.88 ± 2.32).

Pre-lesion scores were similar for the left (29.11 ± 1.45) andright hand (27.0 ± 3.79) in the stair case valley task (Fig. 6B). Aftersurgery, a strong decrease in the average annual score was observedfor the left affected hand (1st year: 1.56 ± 2.94). The score remainedlow over the following 3 years (2nd year: 0.03 ± 0.11, 3rd year:0.0, 4th year: 1.91 ± 3.20), but slightly increased in the fifth year(5th year: 7.0 ± 5.47). In contrast, fine motor skills of the righthand remained unaffected throughout the entire experiment andno change in the scores of this hand was observed (1st year:25.32 ± 3.78, 2nd year: 24.63 ± 5.32, 3rd year: 27.18 ± 2.82, 4thyear: 26.73 ± 4.1, 5th year: 25.88 ± 3.83).

4. Discussion

The aim of this study was to establish DAT imaging in the com-mon marmoset using 123I-FP-CIT SPECT. Here, we used clinicalequipment specially adapted for high resolution scanning of mar-moset monkeys.

Pharmacodynamic time course analysis showed high and sta-ble striatal binding of 123I-FP-CIT for at least 4 h, indicating highspecificity of 123I-FP-CIT for DAT (Laruelle et al., 1993; Lundkvistet al., 1995). Similar patterns of brain uptake were reported forhumans (Kuikka et al., 1995), Old World monkeys (Baldwin et al.,1995; Lundkvist et al., 1995; Booij et al., 1997) and rodents (Booijet al., 1997, 2002; Alvarez-Fischer et al., 2007), suggesting uni-versal applicability of 123I-FP-CIT SPECT for DAT-related diseases.The individual differences in binding ratio between the unaffectedcontra lateral side of the 6-OHDA-lesioned animal and the con-trol animal are within range of previously reported studies (Booijet al., 2002; Andringa et al., 2005). KD binding experiments werenot performed in the present study. However, the specificity of123I-FP-CIT for DAT and its suitability for PD diagnostics has beenproven in other studies (Seibyl et al., 1998; Booij et al., 1998; de laFuente-Fernández, 2012).

Measurements of 123I-FP-CIT activity in plasma of common mar-mosets demonstrated that the elimination of the radioligand fromplasma share the same kinetics as in humans (Booij et al., 1998) andis similar to the elimination kinetics of 11C-�-CIT-FP in cynomolgusmonkeys (Lundkvist et al., 1995). Lipophilic radioiodinated plasmametabolites may marginally interfere with SPECT quantification(Lundkvist et al., 1995; Bergström et al., 1996; Heinz et al., 1997).

Although the DAT scans did not feature sufficient anatomicaldetail to outline nuclear territories of the basal ganglia, the globalintensity pattern could be robustly aligned to high-resolutionstructural MRI scans. Thus, ROIs of single deep brain nuclei canbe defined for specific analysis and monitoring of concomitant

volumetric changes. Our co-registration approach employed theintercommisural line as stereotaxic reference to ensure compa-rability between scans of the same animal as well as consistentdisplay of different individuals.

Page 7

euros

pcsicoc1

(2Dc1

noeedtlt

5

chtSim

A

K(ppiE

R

A

A

A

A

A

B

B

B

E. Garea-Rodríguez et al. / Journal of N

Unilateral 6-OHDA injections into the nigro-striatal projectionathway led to complete loss of striatal DAT binding. This resultorresponds well to the behavioral impairments of fine motorkills which were clearly detectable over 5 years following lesion-ng. Thus, the in vivo imaging and behavioral data exemplify thehronicity of the lesion protocol. These results are supported byther behavioral and histological data obtained after 6-OHDA appli-ations in marmosets (Annett et al., 1994; Eslamboli et al., 2003).23I-FP-CIT SPECT was evaluated in various rodent models of PDLauwers et al., 2007; Andringa et al., 2005; Alvarez-Fischer et al.,007; Booij et al., 2002). Until now, the only study that appliedAT SPECT to common marmosets was a reported by Saji andolleagues, who evaluated 123I-�-CIT in the 1-methyl-4-phenyl-,2,3,6-tetrahydropyridine (MPTP) model of PD (Saji et al., 2003).

In future studies, 123I-FP-CIT SPECT may be an addition or alter-ative to behavioral assessments in non-human primate modelsf PD. Long-term monitoring in vivo by SPECT imaging would bespecially advantageous in viral vector-mediated alpha-synucleinxpressing approaches, in which parkinsonian symptoms slowlyevelop (Kirik et al., 2003; Eslamboli et al., 2007). Especially, futureherapeutic approaches in marmoset models of PD will benefit fromongitudinal studies of individual disease courses, thus increasinghe relevance and impact for translational research.

. Conclusion

The presented work provides the first 123I-FP-CIT SPECT proto-ol for application in common marmoset monkeys. The upgradeduman scanner used in the present study, allows for future long-erm and translational approaches. High resolution 123I-FP-CITPECT and MR imaging appears to be a suitable combination fornvestigating DAT integrity and individual morphology in mar-

oset models of PD.

cknowledgements

The authors wish to thank S. Leineweber, O. Rautz, L.M.rautheim and J. Wülker for expert animal care, Dr. U. Engeland

Scivis, Göttingen), J. Schmiereck and J. Krenzek for technical sup-ort, and N. Yee for critical reading of the manuscript. The study wasartially supported by the DFG Research Center of Molecular Phys-

ology of the Brain (CMPB) and GE Healthcare (Munich, Germany).GR was funded by EU ERA-Net NEURON.

eferences

lvarez-Fischer D, Blessmann G, Trosowski C, Behe M, Schurrat T, Hartmann A, et al.Quantitative [(123)I]FP-CIT pinhole SPECT imaging predicts striatal dopaminelevels, but not number of nigral neurons in different mouse models of Parkin-son’s disease. Neuroimage 2007;38:5–12.

ndringa G, Drukarch B, Bol JG, de Bruin K, Sorman K, Habraken JB, et al. PinholeSPECT imaging of dopamine transporters correlates with dopamine transporterimmunohistochemical analysis in the MPTP mouse model of Parkinson’s disease.Neuroimage 2005;26:1150–8.

nnett LE, Martel FL, Rogers DC, Ridley RM, Baker HF, Dunnett SB. Behaviouralassessment of the effects of embryonic nigral grafts in marmosets with unilateral6-OHDA lesions of the nigrostriatal pathway. Exp Neurol 1994;125:228–46.

nnett LE, Rogers DC, Hernandez TD, Dunnett SB. Behavioural analysis of unilateralmonoamine depletion in the marmoset. Brain 1992;115(Pt. 3):825–56.

rnt J, Christensen AV, Hyttel J. Pharmacology in vivo of the phenylindan deriva-tive, Lu 19-005, a new potent inhibitor of dopamine, noradrenalin and5-hydroxytryptamine uptake in rat brain. Naunyn Schmiedebergs Arch Phar-macol 1985;329:101–7.

aldwin RM, Zea-Ponce Y, al-Tikriti MS, Zoghbi SS, Seibyl JP, Charney DS, et al.Regional brain uptake and pharmacokinetics of [123I]N-omega-fluoroalkyl-2beta-carboxy-3 beta-(4-iodophenyl)nortropane esters in baboons. Nucl MedBiol 1995;22:211–9.

eekman F, van der Have F. The pinhole: gateway to ultra-high-resolutionthree-dimensional radionuclide imaging. Eur J Nucl Med Mol Imaging2007;34:151–61.

ergström K, Halldin C, Lundkvist, Swahn CG, Akerman KK, Kuikka JT, et al. Charac-terization of C-11 or I-123 labelled �-CIT-FP and �-CIT-FE metabolism measured

cience Methods 210 (2012) 195– 201 201

in monkey and human plasma. Identification of two labelled metabolites withHPLC. Hum Psychopharmacol 1996;11:483–90.

Bock NA, Kocharyan A, Liu JV, Silva AC. Visualizing the entire cortical myelinationpattern in marmosets with magnetic resonance imaging. J Neurosci Methods2009;185:15–22.

Booij J, Andringa G, Rijks LJ, Vermeulen RJ, De Bruin K, Boer GJ, et al. [123I]FP-CITbinds to the dopamine transporter as assessed by biodistribution studies in ratsand SPECT studies in MPTP-lesioned monkeys. Synapse 1997;27:183–90.

Booij J, Busemann Sokole E, Stabin MG, Janssen AG, de Bruin K, van Royen EA. Humanbiodistribution and dosimetry of [123I]FP-CIT: a potent radioligand for imagingof dopamine transporters. Eur J Nucl Med 1998;25:24–30.

Booij J, de Bruin K, Habraken JB, Voorn P. Imaging of dopamine transporters in ratsusing high-resolution pinhole single-photon emission tomography. Eur J NuclMed Mol Imaging 2002;29:1221–4.

Dauer W, Przedborski S. Parkinson’s disease: mechanisms and models. Neuron2003;39:889–909.

de la Fuente-Fernández R. Role of DaTSCAN and clinical diagnosis in Parkinsondisease. Neurology 2012;78:696–701.

Eslamboli A, Baker HF, Ridley RM, Annett LE. Sensorimotor deficits in a unilateralintrastriatal 6-OHDA partial lesion model of Parkinson’s disease in marmosetmonkeys. Exp Neurol 2003;183:418–29.

Eslamboli A, Romero-Ramos M, Burger C, Bjorklund T, Muzyczka N, Mandel RJ,et al. Long-term consequences of human alpha-synuclein overexpression in theprimate ventral midbrain. Brain 2007;130:799–815.

Heinz A, Jones DW, Gorey JG, Knable MB, Lee KS, Saunders RC, et al. Analysis of themetabolites of [I-123] beta-CIT in plasma of human and nonhuman primates.Synapse 1997;25:306–8.

Hyttel J. Citalopram – pharmacological profile of a specific serotonin uptakeinhibitor with antidepressant activity. Prog Neuropsychopharmacol Biol Psy-chiatry 1982;6:277–95.

Jack Jr CR, Bernstein MA, Fox NC, Thompson P, Alexander G, Harvey D, et al. Thealzheimer’s disease neuroimaging initiative (ADNI): MRI methods. J Magn ResonImaging 2008;27:685–91.

Kirik D, Annett LE, Burger C, Muzyczka N, Mandel RJ, Bjorklund A. Nigrostriatal alpha-synucleinopathy induced by viral vector-mediated overexpression of humanalpha-synuclein: a new primate model of Parkinson’s disease. Proc Natl AcadSci U S A 2003;100:2884–9.

Kuikka JT, Bergstrom KA, Ahonen A, Hiltunen J, Haukka J, Lansimies E,et al. Comparison of iodine-123 labelled 2 beta-carbomethoxy-3 beta-(4-iodophenyl)tropane and 2 beta-carbomethoxy-3 beta-(4-iodophenyl)-N-(3-fluoropropyl)nortropane for imaging of the dopamine transporter in the livinghuman brain. Eur J Nucl Med 1995;22:356–60.

Laruelle M, Baldwin RM, Malison RT, Zea-Ponce Y, Zoghbi SS, al-Tikriti MS,et al. SPECT imaging of dopamine and serotonin transporters with [123I]beta-CIT: pharmacological characterization of brain uptake in nonhuman primates.Synapse 1993;13:295–309.

Lauwers E, Beque D, Van Laere K, Nuyts J, Bormans G, Mortelmans L, et al.Non-invasive imaging of neuropathology in a rat model of alpha-synuclein over-expression. Neurobiol Aging 2007;28:248–57.

Lundkvist C, Halldin C, Swahn CG, Hall H, Karlsson P, Nakashima Y, et al. [O-methyl-11C]beta-CIT-FP, a potential radioligand for quantitation of the dopaminetransporter: preparation, autoradiography, metabolite studies, and positronemission tomography examinations. Nucl Med Biol 1995;22:905–13.

Ma KH, Huang WS, Chen CH, Lin SZ, Wey SP, Ting G, et al. Dual SPECT of dopamine sys-tem using [99mTc]TRODAT-1 and [123I]IBZM in normal and 6-OHDA-lesionedformosan rock monkeys. Nucl Med Biol 2002;29:561–7.

Mansfield K. Marmoset models commonly used in biomedical research. Comp Med2003;53:383–92.

Neumeyer JL, Wang S, Gao Y, Milius RA, Kula NS, Campbell A, et al. N-omega-fluoroalkyl analogs of (1R)-2 beta-carbomethoxy-3 beta-(4-iodophenyl)-tropane (beta-CIT): radiotracers for positron emission tomography and singlephoton emission computed tomography imaging of dopamine transporters. JMed Chem 1994;37:1558–61.

Orsi A, Rees D, Andreini I, Venturella S, Cinelli S, Oberto G. Overview of the mar-moset as a model in nonclinical development of pharmaceutical products. RegulToxicol Pharmacol 2011;59:19–27.

Prunier C, Bezard E, Montharu J, Mantzarides M, Besnard JC, Baulieu JL, et al.Presymptomatic diagnosis of experimental Parkinsonism with 123I-PE2I SPECT.Neuroimage 2003;19:810–6.

Saji H, Iida Y, Kawashima H, Ogawa M, Kitamura Y, Mukai T, et al. In vivoimaging of brain dopaminergic neurotransmission system in small animalswith high-resolution single photon emission computed tomography. Anal Sci2003;19:67–71.

Schramm NU, Ebel G, Engeland U, Schurrat T, Béhé M, Behr TM. High-resolution SPECT using multipinhole collimation. IEEE Trans Nucl Sci 2003;50:315–20.

Seibyl JP, Marek K, Sheff K, Zoghbi S, Baldwin RM, Charney DS, et al. Iodine-123-beta-CIT and iodine-123-FPCIT SPECT measurement of dopamine transportersin healthy subjects and Parkinson’s patients. J Nucl Med 1998;39:1500–8.

Stephan H, Baron G, Schwerdtfeger WK. The brain of the common marmoset(Callathrix jacchus). A stereotaxic atlas. Berlin: Springer-Verlag; 1980.

Swan RC, Madisso H, Pitts RF. Measurement of extracellular fluid volume in nephrec-tomized dogs. J Clin Invest 1954;33:1447–56.

Yuasa S, Nakamura K, Kohsaka S. Stereotaxic atlas of the marmoset brain: withimmunohistochemical architecture and MR images. National Institute of Neu-roscience (JP): Tokyo; 2010.