Radiosynthesis and in vivo evaluation of [11C]-labelled pyrrole-2-carboxamide derivates as novel radioligands for PET imaging of monoamine oxidase A

Available online at www.sciencedirect.com

Nuclear Medicine and Biology 37 (2010) 459 – 467 www.elsevier.com/locate/nucmedbio

Radiosynthesis and in vivo evaluation of [11 C]-labelled pyrrole-2-carboxamide derivates as novel radioligands for PET imaging of monoamine oxidase A Sylvie De Bruynea , Giuseppe La Reginab , Steven Staelensc , Leonie Wyffelsa , Steven Deleyec , Romano Silvestrib , Filip De Vosa,⁎ a

b

Laboratory for Radiopharmacy, Ghent University, 9000 Ghent, Belgium Istituto Pasteur, Fondazione Cenci Bolognetti, Dipartimento di Chimica e Tecnologie del Farmaco, Sapienza Università di Roma, I-00185 Rome, Italy c IBITECH-Medisip, Ghent University-IBBT, 9000 Ghent, Belgium Received 21 August 2009; received in revised form 17 September 2009; accepted 28 September 2009

Abstract Introduction: Since MAO-A is an enzyme involved in the metabolism of neurotransmitters, fluctuations in MAO-A functionality are associated with psychiatric and neurological disorders as well as with tobacco addiction and behaviour. This study reports the radiolabelling of two [11C]-labelled pyrrole-2-carboxamide derivates, RS 2315 and RS 2360, along with the characterization of their in vivo properties. Methods: The radiolabelling of [11C]-RS 2315 and [11C]-RS 2360 was accomplished by alkylation of their amide precursors with [11C] CH3I. Biodistribution, blocking and metabolite studies of both tracers were performed in NMRI mice. Finally, a PET study in SpragueDawley rats was performed for [11C]-RS 2360. Results: Both tracers were obtained in a radiochemical yield of approximately 30% with radiochemical purity of N98%. Biodistribution studies showed high brain uptake followed by rapid brain clearance for both radiotracers. In the brain, [11C]-RS 2360 was more stable than [11C]-RS 2315. Blocking studies in mice could not demonstrate specificity of [11C]-RS 2315 towards MAO-A or MAO-B. The blocking and imaging study with [11C]-RS 2360 on the other hand indicated specific binding in MAO-A at the earliest time points. Conclusions: [11C]-RS 2315 displayed a high nonspecific binding and is therefore not suitable for visualization of MAO-A in vivo. [11C]-RS 2360 on the other hand has potential for mapping MAO-A since specific binding is demonstrated. © 2010 Published by Elsevier Inc. Keywords: MAO; PET; [11C]-Pyrrole-2-carboxamide derivates; Brain

1. Introduction Monoamine oxidase (MAO) is a flavin-containing enzyme [1] that occurs in the outer mitochondrial membrane [2] of neuronal and nonneuronal cells in the brain and peripheral organs [3]. MAO catalyzes the oxidative deamination of amines from both endogenous and exogenous sources. There exist two different isoforms of MAO that present a 70% degree of homology in their amino acid sequence. These two isoforms, termed MAO-A and MAO-B,

⁎ Corresponding author. Tel.: +32 9 264 8066; fax: +32 9 264 8071. E-mail address: [email protected] (F. De Vos). 0969-8051/$ – see front matter © 2010 Published by Elsevier Inc. doi:10.1016/j.nucmedbio.2009.09.005

are different gene products with molecular weights of about 59.7 and 58.8 kDa, respectively [4,5]. MAO-A and MAO-B differ in substrate specificity and in their sensitivity to inhibitors [6]. MAO-A preferentially deaminates noradrenaline and serotonin and is selectively inhibited by clorgyline, whereas MAO-B breaks down phenethylamine and benzylamine and is inhibited by L-deprenyl. Both forms oxidize dopamine and tyramine [7,8]. The relative ratios of MAO-A and MAO-B are different between organs and species. In rat brain for example, MAO-A predominates, whereas in human brain MAO-B appears in the highest concentration [9–11]. Because of the ability of MAO to catabolise neurotransmitters, MAO is involved in psychiatric and neurological disorders such as depression [12] and Parkinson's disease

460

S. De Bruyne et al. / Nuclear Medicine and Biology 37 (2010) 459–467

[13] as well as in tobacco addiction [14,15] and behaviour [3,16–18]. A very efficient noninvasive method to study enzymes in vivo is visualization with PET. There are only a few PET tracers for MAO-A designed and they all have their own drawbacks. The first developed radiotracer, [11C] clorgyline, displayed an unexplained species difference [19,20]. In contrast to results in humans, [11C]clorgyline was not retained in baboon brain. Another MAO-A tracer, [11C]befloxatone, is synthesized with [11C]phosgene, which is rather rarely available [21]. The drawback of [11C]harmine is the extensive metabolization in plasma with only 10% intact product at 10 min postinjection (pi) [22]. The search for new ligands with optimal properties is justified by the lack of an ideal ligand and by the observation that fluctuations in MAO-A functionality are associated with human diseases and tobacco addiction. A series of new pyrrole derivates has been synthesized and evaluated for their MAO-A and MAO-B inhibitory activity and selectivity [23–25]. Out of this new class of MAO inhibitors we selected the most potent inhibitors for MAO-A: N-(phenethyl)-N-methyl-1H-pyrrole-2-carboxamide (RS 2315) and (R)-N-(α-cyclohexylethyl)-N-methyl1H-pyrrole-2-carboxamide (RS 2360) (Fig. 1). Their Ki values are 7 and 1.7 nM for MAO-A and 12 and 30 nM for MAO-B, for RS 2315 and RS 2360, respectively [23–25]. The present study reports the radiolabelling and in vivo evaluation of 11C-labelled RS 2315 and RS 2360. 2. Materials and methods 2.1. General All chemicals were purchased from commercial sources (Sigma-Aldrich or Acros Organics, Belgium) and were used without further purification. The starting compounds N(phenethyl)-1H-pyrrole-2-carboxamide (RS 2115) and (R)N-(α-cyclohexylethyl)-1H-pyrrole-2-carboxamide (RS 2226) as well as the cold reference compounds RS 2315 and RS 2360 for this study were kindly provided by Istituto Pasteur, Fondazione Cenci Bolognetti, Dipartimento di Chimica e Tecnologie del Farmaco, Sapienza Università di Roma, Rome, Italy. The synthesis and in vitro evaluation of these compounds are reported elsewhere [23–25].

Unless otherwise mentioned, the HPLC system used consisted of a Waters 515 HPLC pump, a Waters 2487 UV detector (Waters, Milford, MA, USA) set at 262 nm, a Ludlum model 2200 scalar ratemeter equipped with a Geiger Müller tube (Ludlum Measurements, Inc., Sweetwater, TX, USA) and a Shimadzu C-RSA chromatopac data analyser. Absorption units at full scale were set at 0.0001. The columns, mobile phases and flow rates are indicated in the text below. All radioactivity counting was performed with a Cobra gamma-ray spectrometer equipped with five 1×1-in. NaI(Tl) crystals (Cobra Autogamma, Packard, Canberra, Australia). The values are corrected for background radiation and physical decay during counting. All animal studies were approved by the local ethics committee of Ghent University (ECD 08/35). Mice and rats were obtained from Bioservices (Uden, The Netherlands) and were allowed free access to food and water. 2.2. Production of [11C]-iodomethane [11C]CH4 was produced in a Cyclone 18 twin cyclotron (IBA, Louvain-La-Neuve, Belgium) via the 14N(p,α)11C reaction induced by irradiation of N2 gas containing 5% H2 with proton beam (18 MeV, 14 μA) for 20 min. This target gas is pressurized, and transfer of the obtained [11C]CH4 from the cyclotron target to the homemade synthesis module housed in a hot cell was performed by expansion of the target gas. The transferred [11C]CH4 was trapped on a loop filled with Porapak N (divinylbenzene/vinyl pyrolidone polymer), cooled in liquid argon. Once all [11C]CH4 was released from the target, the loop was flushed with helium to remove N2/H2 and allowed to warm to room temperature. The [11C]CH4 was swept off the loop with a helium flow and mixed with I2 vapours at 50°C, after which the mixture was passed through an oven heated to 600°C to yield [11C]CH3I. After the reaction, the remaining iodine and produced HI were trapped in ascarite, while [11C]CH3I was collected on a Porapak N trap at room temperature. The unreacted [11C]CH4 was further circulated until no more [11C]CH3I was produced. A helium stream released [11C]CH3I from the Porapak N trap by heating the trap to 120°C.

Fig. 1. Chemical structure of RS 2315 and RS 2360.

S. De Bruyne et al. / Nuclear Medicine and Biology 37 (2010) 459–467

2.3. Radiosynthesis 11

2.3.1. [ C]-RS 2315 A solution of 3 μmol of precursor RS 2115 in DMSO/ DMF (189 μl/60 μl) and 1 μl Bu4NOH (aqueous solution of 1.5 M) was added to a reaction vessel, and the mixture was cooled in an ice bath. A stream of helium containing the alkylating agent [11C]CH3I was bubbled through the reaction mixture until a maximum of 11C activity was trapped. The reaction mixture was heated in an oil bath at 65°C for 10 min and subsequently diluted with 200 μl HPLC eluent and purified with semipreparative C18 HPLC (Econosphere, 10 μm, 10×250 mm; Grace Davison Discovery Sciences, Lokeren, Belgium) using a smartline UV detector 2500 (Knauer, Berlin, Germany) set at 254 nm. Radiodetection occurred by a solar-blind PIN photodiode. Elution was carried out at 6 ml/min flow rate with 45% (v/v) CH3CN in 0.02 M sodium acetate buffer (pH 4.5) as mobile phase. The fraction containing [11C]-RS 2315 (retention time=10 min) was collected into a vessel with 40 ml of sterile water and loaded on a C18 Sep-Pak (Alltech Maxi-Clean SPE Prevail C18, previously activated with 1 ml ethanol and 5 ml sterile water). After the cartridge had been washed with 5 ml sterile water, the desired product, [11C]-RS 2315, was eluted with 1 ml ethanol. For biodistribution studies, the ethanol fraction was diluted with 10 ml saline. For metabolite analysis, ethanol was evaporated to dryness and the residue was redissolved in an adequate amount of ethanol/physiological saline (8/92 v/v). 2.3.2. [11C]-RS 2360 The same procedure as described for [11C]-RS 2315, with minor modifications, was applied to synthesize [11C]-RS 2360. [11C]-RS 2360 was prepared by methylation with [11C]CH3I of the normethylderivate RS 2226 (3 μmol) in DMF/DMSO (60 μl/188 μl) for 7 min at 70°C using 1.5 μl Bu4NOH (aqueous solution of 1.5 M). Purification was accomplished using CH3CN/H2O/HCOOH (50/50/0.1 v/v/v) as mobile phase at a flow rate of 6 ml/min. The collected HPLC fraction containing [ 11 C]-RS 2360 (retention time=8.5 min) was diluted with 40 ml 0.01 M Dulbecco's phosphate buffered saline (pH 7.4) before loading onto the Sep-Pak (Alltech Maxi-Clean SPE Prevail C18, previously activated with 1 ml EtOH and 5 ml sterile water). After the cartridge was washed with 5 ml sterile water, the desired product, [11C]-RS 2360, was eluted with 1 ml EtOH. For biodistribution and blocking studies the EtOH fraction was diluted with 10 ml physiological saline. For metabolite and imaging studies, EtOH was evaporated to dryness and the residue was redissolved in an adequate amount of EtOH/ physiological saline (8/92 v/v). 2.4. Quality control Quality control consisted of the determination of radiochemical purity and specific activity, calculated by analytical HPLC assay using a Gracesmart C18 column (5

461

μm, 4.6×250 mm; Grace Davison Discovery Sciences, Lokeren, Belgium) at a flow rate of 1 ml/min. The eluents were the same as described for the purification. Co-injection on analytical HPLC of the final end product and cold reference compound was performed to confirm the identity of the obtained radiotracer. Specific activities, decay corrected back to the end of purification, were obtained from the tracer activity determined by Capintec readings and from the mass of the tracer which was calculated by the use of a calibration curve of unlabelled reference compound (0.02–1 μM). 2.5. Log D7.4 measurement Measurements of partition coefficient (n=3) for [11C]-RS 2315 and [11C]-RS 2360 were performed using a modified literature procedure [26]. An aliquot (10–20 μl) of the 11C tracer was added to a test tube containing 3 ml n-octanol and 3 ml phosphate buffer (0.01 M, pH 7.4). The mixture was vigorously shaken by hand for 1 min, vortexed for 2 min and centrifuged for 3 min at 3000×g. 0.5 ml of both phases was taken and placed in separate vials, taking care to avoid cross contamination between the phases. The remaining aqueous phase was discarded, and 2.5 ml fresh buffer (0.01 M, pH 7.4) was added to the initial test tube. The procedure was repeated three more times to give four vials of each phase for measuring radioactivity. The partition coefficient was calculated as: radioactivity in n-octanol (cpm)/radioactivity in phosphate buffer (cpm). 2.6. Biodistribution studies The biodistribution of [11C]-RS 2315 and [11C]-RS 2360 was studied in male Naval Medical Research Institute (NMRI) mice (approximately 25 g and 6 weeks old). Mice (n=3 at each time point) were injected intravenously with approximately 200 μl of an EtOH:physiological saline mixture (10:90 v/v) containing 7.5–11 MBq (0.2–0.3 mCi) of [11C]-RS 2315 or [11C]-RS 2360. At 1, 10, 30 and 60 min pi, mice were euthanized and dissected. Blood, urine and organs were weighed and counted for radioactivity. All organs were rinsed with water prior to weighing and counting. For calculation of the injected dose, five aliquots of the injection solution were weighed and counted for activity. Results are decay corrected and expressed as percentage of injected dose per gram of tissue±standard deviation (%ID/g±S.D.). 2.7. Blocking studies Blockade studies were carried out with preinjection of 100 μl clorgyline hydrochloride (MAO-A inhibitor) or R(−)-deprenyl hydrochloride (MAO-B inhibitor) intravenously both in a dose of 10 mg/kg. Approximately 9 MBq (0.24 mCi) of 11C tracer (100 μl) was injected 30 min later via the tail vein. The mice were sacrificed and dissected at 1, 10 and 30 min after 11C tracer injection. Blood, urine and organs were weighed and counted for radioactivity. Results are decay corrected and expressed as %ID/g±S.D.

462

S. De Bruyne et al. / Nuclear Medicine and Biology 37 (2010) 459–467

Statistical analysis was performed using one-sided, unpaired Student's t test. Only P values b.05 are considered significant. 2.8. Metabolite analysis The in vivo metabolic stability of [11C]-RS 2315 and [11C]RS 2360 in plasma and brain was studied in male NMRI mice. Twenty-two to 37 MBq (0.59–1 mCi) of [11C]-RS 2315 or [11C]-RS 2360 dissolved in 200 μl of an EtOH:physological (8:92 v/v) mixture was injected through the tail vein, and the mice (n=3 for each time point) were sacrificed at 1, 10 and 30 min pi. Blood and whole brain were isolated. Blood was collected into a vacutest tube containing 3.6 mg K3EDTA and centrifuged at 4000×g for 6 min to separate plasma. The supernatant plasma (200 μl) was mixed with 0.8 ml CH3CN, whereas the brain was homogenized in 1 ml CH3CN. Both samples were vortexed for 0.5 min and centrifuged at 3000×g for 3 min. Pellet and supernatant were separated and counted for radioactivity. An aliquot (500 μl) of the supernatant obtained from the plasma and brain homogenates was subjected to HPLC analysis using the same conditions as used for purification. The HPLC eluate was collected in 0.5min fractions and their radioactivity was measured. To determine the recovery capabilities of [11C]-RS 2315 and [11C]-RS 2360 as well as the stability of the radiotracer during the workup, control experiments (n=3) were done using mouse plasma and brain spiked with approximately 3 MBq (81 μCi) of authentic [11C]-RS 2315 or [11C]-RS 2360. Sample workup was done as described above. Results are expressed as percentages of the total activity±S.D. 2.9. Imaging study Additional in vivo tests with [11 C]-RS 2360 were performed using microPET imaging technology. SpragueDawley rats (n=4 for each treatment group) were injected with physiological saline, clorgyline hydrochloride (10 mg/ kg) or (R)-(−)-deprenyl hydrochloride (10 mg/kg) 30 min prior to their microPET scan. The latter acquisitions are performed using Gamma Medica Ideas' LabPET 8, a stateof-the-art microPET device consisting of 2×2×10-mm3 LYSO/LGSO scintillators in an 8-pixel, quad-APD detector module arrangement. This system is capable of delivering 1 mm spatial resolution in rodents at a sensitivity of 4%, thereby covering a field-of-view of 10 cm transaxially by 8 cm axially. All animals were injected with 35–40 MBq (0.95–1.08 mCi) of [11C]-RS 2360 on the camera bed at the start of a dynamic acquisition and anesthetized throughout the microPET scan by inhalation of 1.5% isoflurane. Frames of 4×0.5, 3×1 and 1×5 min were accordingly sequentially recorded. The resulting data were reconstructed using 30 iterations of the maximum likelihood expectation maximization algorithm in 160×160×63 voxels of 0.5×0.5×1.175 mm voxel size. An a posteriori three-dimensional Gaussian filter of 2×2×2 mm kernel width was applied to all frames. Statistical

analysis was performed using one-sided, unpaired Student's t test. Only P values b.05 are considered significant. Images were analysed with PMOD whereby volumes of interest (VOIs) were drawn on the brain region (Fig. 2A) and on the background region (Fig. 2B) of each animal. Since the selection of the background region is critical, we delineated the background VOI on the last frame of 5-min acquisition. In this way, noise is eliminated and a robust reference value is obtained to normalize the brain uptake for injected activity in all other frames on an individual animal basis. We selected the background region so that no specific muscle activity was present.

3. Results 3.1. Radiosynthesis The synthesis of radioligands [11C]-RS 2315 and [11C]RS 2360 is shown in Fig. 3. Their respective amide precursors were labelled with [11C]CH3I prepared from [11C]CH4, in the presence of tetrabutylammonium hydroxide as a base through N-[11C]methylation and isolated by semipreparative HPLC. [11C]-RS 2315 elutes with a retention time of 10 min. Extraction efficiency of the Sep-Pak was calculated to be 82±8% (n=6). Based on the amount of [11C]CH3I added to the reaction vial, the radiochemical yield was 28±4% (n=6). The total synthesis, from the end of 11CH3I delivery to the reaction vessel to delivery for in vivo studies, was completed in 35 min. The overall synthesis, purification and formulation time to obtain [11C]-RS 2360 was 30 min. In a typical experiment, target tracer [11C]-RS 2360 was provided in a radiochemical yield of 30±6% (n=6), decay corrected to the start of the reaction, based on the amount of [11C]CH3I added. Almost all activity was released from the Sep-Pak (93±4%, n=6). 3.2. Quality control, specific activity and stability The identity of both tracers was confirmed by co-elution with authentic reference compound after co-injection on HPLC. Calculation of specific activity resulted in a range from 25 to 92 GBq/μmol (0.68–2.49 Ci/μmol) for [11C]-RS 2315 (n=6) and from 41 to 106 GBq/μmol (1.11–2.87 Ci/ μmol) for [11C]-RS 2360 (n=6). HPLC-based radiochemical purity of the tracer recovered at the end of the experiment was N98% for both radiotracers. The HPLC-based radiochemical purity of both [11C]-RS 2315 and [11C]-RS 2360 remained higher than 98% during the time span of the experiments. 3.3. In vitro lipophilicity We determined log D7.4 as an indicator for lipophilicity and blood–brain permeability. Octanol–buffer partition coefficient measurements gave a log D7.4 of 1.65±0.19 for [11C]-RS 2315, slightly higher than the log D7.4 of [11C]-RS 2360 (1.48±0.04). Both values are suitable for brain penetration.

S. De Bruyne et al. / Nuclear Medicine and Biology 37 (2010) 459–467

463

Fig. 2. Example VOI for the brain region (A) and for the background region (B). (A) MicroPET image of a 30-s time frame at 1.5 min pi of [11C]-RS 2360. (B) MicroPET image of a 5-min time frame at 10 min pi of [11C]-RS 2360.

Fig. 3. Radiosynthesis of [11C]-RS 2315 and [11C]-RS 2360.

464

S. De Bruyne et al. / Nuclear Medicine and Biology 37 (2010) 459–467

3.4. Biodistribution studies

Table 2 Tissue distribution of [11C]-RS 2360 in male NMRI mice

Following intravenous injection of [11C]-RS 2315 into NMRI mice, the time course of radioactivity was determined in several tissues (Table 1). [11C]-RS 2315 entered the brain quickly and was rapidly cleared from the brain (from 4.75±1.62 %ID/g at 1 min pi to 0.24±0.07 %ID/g at 60 min pi). Except at the earliest time point, the radioactivity in blood was higher than the brain uptake. High initial uptake in heart, lungs, pancreas and kidneys (5.52±0.41, 5.96±1.11, 5.19±1.63 and 6.81±3.23 %ID/g, respectively, at 1 min pi) was observed, but, except for the kidney (8.34±1.09 %ID/g at 60 min pi), the studied organs did not show retention of [11C]-RS 2315. In the liver, the activity was the highest at 10 min pi with 9.09±0.97 %ID/g and decreased over time. The activity in small intestines had the highest increase between 10 min pi (3.39±1.57 %ID/g) and 30 min pi (9.68±2.69 %ID/g), indicating hepatobiliary clearance. Urinary clearance was also observed (data not shown). Table 2 summarizes the tissue time-course distribution of [11C]-RS 2360 in NMRI mice. [11C]-RS 2360 displayed a rapid and high brain uptake of 7.08±0.95 %ID/g at 1 min pi indicating a good penetration of the radiotracer into the brain. The activity in the blood pool showed a slow clearance (2.35±1.21 %ID/g at 1 min pi and 1.00±0.27 %ID/g at 60 min pi, respectively). The lungs, heart and pancreas showed high initial uptake (23.62±4.22, 11.51±0.93 and 3.93±1.78 %ID/g at 1 min pi) followed by a fast washout (6.01±2.28, 1.75±0.16 and 1.66±0.37 %ID/g at 30 min pi). As for [11C]-RS 2315, the uptake in the liver was also gradually increased for 10 min after injection (7.45±0.35 %ID/g) and then decreased (4.24±0.99 %ID/g at 60 min pi). Clearance of radioactivity occurred by both the hepatobiliary and urinary pathway. 3.5. Blocking studies In vivo selectivity and specificity of both radiotracers were examined by preinjection of clorgyline hydrochloride (MAO-A inhibitor) or R-(−)-deprenyl hydrochloride (MAOTable 1 Tissue distribution of [11C]-RS 2315 in male NMRI mice %ID/g±S.D.

Blood Brain Heart Lungs Stomach Spleen Liver Kidneys Small intestine Large intestine Bladder Pancreas

1 min

10 min

30 min

60 min

2.95±0.59 4.75±1.62 5.52±0.41 5.96±1.11 1.91±0.23 2.05±0.64 3.63±0.56 6.81±3.23 2.35±0.48 1.05±0.42 2.28±0.25 5.19±1.63

2.69±0.10 1.11±0.16 2.53±0.25 2.72±0.29 2.28±0.90 1.73±0.25 9.09±0.97 7.12±1.24 3.39±1.58 0.98±0.14 8.75±1.40 2.19±0.24

1.62±0.28 0.39±0.04 1.35±0.11 1.80±0.10 1.55±0.40 1.02±0.11 5.41±1.19 9.69±3.48 9.68±2.69 0.94±0.14 16.49±5.83 1.74±0.17

0.67±0.13 0.24±0.07 0.75±0.17 0.96±0.20 1.29±0.23 0.87±0.27 2.87±0.74 8.34±1.09 14.25±1.21 1.64±1.24 16.57±12.11 1.09±0.27

Mice were intravenously injected with 7.5–11 MBq (0.2–0.3 mCi) [11C]RS 2315. Values are averaged (n=3) and decay corrected.

%ID/g±S.D.

Blood Brain Heart Lungs Stomach Spleen Liver Kidneys Small intestine Large intestine Bladder Pancreas

1 min

10 min

30 min

60 min

2.35±1.21 7.08±0.95 11.51±0.93 23.62±4.22 2.28±1.21 1.53±0.66 2.62±1.59 6.56±2.06 1.58±0.62 1.07±0.47 3.14±1.21 3.93±1.78

1.62±0.31 1.68±0.32 3.42±0.51 16.88±6.58 1.84±0.46 1.73±0.39 7.45±0.35 6.33±2.13 1.56±0.40 0.97±0.13 4.52±0.53 3.24±1.58

1.41±0.12 0.51±0.06 1.75±0.16 6.01±2.28 1.19±0.06 1.41±0.21 5.17±0.79 4.82±0.19 2.77±0.72 1.54±0.36 8.04±3.22 1.66±0.37

1.00±0.27 0.26±0.10 1.20±0.31 3.39±1.00 1.40±0.37 1.15±0.26 4.24±0.99 4.91±1.66 5.68±1.36 2.86±0.38 25.07±29.69 1.81±1.06

Mice were intravenously injected with 7.5–11 MBq (0.2–0.3 mCi) [11C]-RS 2360. Values are averaged (n=3) and decay corrected.

B inhibitor). Uptake of radioactivity in the brain (Fig. 4) and in peripheral organs was measured at 1, 10 and 30 min pi. [11C]-RS 2315 brain uptake could not be decreased by administrating clorgyline or R-(−)-deprenyl. Surprisingly, a significant higher brain uptake is observed at 30 min pi after both MAO-A and MAO-B inhibition. The effect on the brain/ blood ratio, however, is not significant. A rise in brain uptake also occurred at 1 min pi, but since the blood activity increased as well, the brain/blood ratio was not affected. Peripherally, an increased liver uptake is observed at 1 min pi both after clorgyline (from 3.63±0.56 to 7.15±3.02 %ID/g) and after R(−)-deprenyl (from 3.63±0.56 to 7.81±0.73 %ID/g) pretreatment. At 10 and 30 min pi, a rise in intestinal activity [from 4.36±1.68 to 7.84±3.21 %ID/g after clorgyline pretreatment and to 5.74±0.73 %ID/g after R-(−)-deprenyl pretreatment at 10 min pi] was seen, suggesting an enhanced hepatobiliary clearance after administration of clorgyline or R-(−)-deprenyl. This effect, however, failed to reach statistical significance possibly caused by the large standard error. At 1 min pi, pretreatment with clorgyline resulted in a significant decrease in brain uptake of [11C]-RS 2360 (P=.02). (R)-(−)-Deprenyl administration also lowered brain uptake, but this decrease was not significant. This effect is in accordance with what we would expect from the in vitro properties (Ki=1.7 nM for MAO-A and Ki=30 nM for MAOB). At 10 and 30 min pi, however, no effect of clorgyline or R(−)-deprenyl preadministration could be observed. No clear effect of blocking MAO-A or MAO-B could be distinguished in the peripheral organs (data not shown). 3.6. Metabolite analysis in mice Control experiments with spiked plasma and brain revealed extraction efficiencies of 96% for plasma and 93±1% for brain samples for [11C]-RS 2315 and 93±6% for plasma and 85±6% for brain samples for [11C]-RS 2360, respectively. Plasma and brain obtained from mice at 1, 10

S. De Bruyne et al. / Nuclear Medicine and Biology 37 (2010) 459–467

465

Fig. 5. Metabolite chromatogram of [11C]-RS 2315 at 10 min (A) and 30 min (B) pi.

Fig. 4. Brain uptake of [11C]-RS 2315 (A) and [11C]-RS 2360 (B) in control mice and in mice pretreated with clorgyline or (R)-(−)-deprenyl. Values are the mean of three experiments, *Pb.05, Student's t test.

can be concluded when compared to injection of physiological solution (P=.006). An inhibition of 16% by R-(−)deprenyl is noticeable, albeit not significant (P=.07). These observations indicate that [11C]-RS 2360 is more selective towards MAO-A than to MAO-B, which is in accordance

and 30 min pi were analysed by RP-HPLC to investigate the metabolism pattern of [11C]-RS 2315 (Fig. 5) and [11C]-RS 2360 (Fig. 6). RP-HPLC analysis of radioactivity in mouse plasma showed that [11C]-RS 2315 was rapidly metabolized with only 21±2% parent compound remaining at 10 min pi. All detected metabolites were more polar than the original radiotracer. The percentage of radioactivity in mouse brain extracts derived from unmetabolized [11C]-RS 2315 was N99% at 1 min pi, 62±3% at 10 min pi and 53±7% at 30 min pi, respectively. The relative percentage of intact [11C]-RS 2360 in plasma as a function of time after injection was 29±10% at 10 min pi and 21±5% at 30 min pi. HPLC analysis of the brain samples showed 81±4% and 66±4% of the radioactivity existed as unchanged [11C]-RS 2360 at 10 and 30 min pi, respectively. No degradation products were found in either plasma or the brain at 1 min pi. 3.7. PET Scans The results shown in Fig. 7 illustrate that, for [11C]-RS 2360, an overall significant inhibition by clorgyline of 30%

Fig. 6. Metabolite chromatogram of [11C]-RS 2360 at 10 min (A) and 30 min (B) pi.

466

S. De Bruyne et al. / Nuclear Medicine and Biology 37 (2010) 459–467

Fig. 7. Results for the PET imaging experiments. Rats were injected with 35–40 MBq (0.95 - 1.08 mCi) [11C]-RS 2360 and anesthetized throughout the μSPECT scan by inhalation of 1.5% isoflurane.

with the in vitro-determined Ki values. When we investigate the influence of clorgyline pretreatment at each time frame separately, a significant inhibition is observed at the first five time frames but not at the last three time frames. R-(−)Deprenyl administration caused no significant inhibition of brain uptake at any of the separate time frames. These observations are in accordance with the blocking study. 4. Discussion We have developed two [11C]-labelled pyrrole-2-carboxamide derivates as potential radioligands for MAO-A imaging. The radiosynthesis of both radiotracers was achieved by nucleophilic substitution of the desmethyl precursor with [11C]CH3I in the presence of tetrabutylammonium hydroxide. After HPLC purification, the radiochemical yields (approximately 30%) as well as the radiochemical purity (N98%) were comparable for both [11C]-RS 2315 and [11C]-RS 2360. Specific activity ranged from 25 to 106 GBq/ μmol (0.68–2.87 Ci/μmol). These values are comparable with those reported for [11C]clorgyline (18–87 GBq/μmol), [11 C]harmine (18–93 GBq/μmol) and [11 C]befloxatone (18.5–74 GBq/μmol) [20–22]. The log D7.4 determination presented values that are suitable for brain penetration. The tissue distribution profile of both radiotracers is similar. Upon intravenous administration, high initial levels of activity were observed in mouse brain, indicating an excellent passage through the blood–brain barrier of both [11C]-RS 2315 and [11C]-RS 2360. The brain uptake of both radiotracers showed that they entered the brain rapidly with a maximum at 1 min pi. The brain activity diminished very quickly, and at 10 min pi for [11C]-RS 2315 and at 30 min pi for [11C]-RS 2360

blood activity exceeded brain activity levels. Metabolization studies revealed that at 1 min pi almost no degradation occurred either in plasma or in the brain. At 10 and 30 min pi, [11 C]-RS 2315 and [11 C]-RS 2360 were extensively metabolized in plasma and remained more stable in the brain. Neither in plasma nor in brain extracts was a more lipophilic metabolite derived from any of the two radiotracers detected. It seems quite likely that the hydrophilic degradation fragment with a retention time of 2.5 min was methanol, formaldehyde or formic acid, which was formed after demethylation of the radiotracer. At all time points examined, the extent of metabolism was greater for [11C]-RS 2315 than for [11C]-RS 2360. Compared to [11C]harmine, which in plasma is present only as a 10% intact product at 10 min pi, the metabolism pattern was in both cases superior [22]. The blocking study failed to prove the specificity of [11C]RS 2315, indicating only nonspecific binding in the brain. [11C]-RS 2360 showed specificity towards MAO-A at 1 min pi. At 10 and 30 min pi, however, no effect of clorgyline pretreatment could be observed. Administration of R-(−)deprenyl had no significant effect on the brain uptake of [11C]-RS 2360. The PET imaging study with [11C]-RS 2360 displayed a significant inhibition at the first five time frames but not at the last three time frames. R-(−)-Deprenyl administration caused no significant inhibition of brain uptake at any of the separate time frames. These observations are in accordance with the blocking study. The absence of selectivity at the later time points might be explained by the appearance of metabolites in the brain. Another possible reason could be that [11C]-RS 2360 has a fast reversible binding to the enzyme which is demonstrated by the fast clearance pattern of [11C]-RS 2360. When we investigate the brain ratios of [11C]-RS 2360 over the complete duration of the PET scans, an overall significant inhibition by clorgyline can be concluded, indicating specific binding in MAO-A. Administration of R-(−)-deprenyl caused a decreased brain uptake, although this effect was not significant. These observations indicate that [11C]-RS 2360 is more selective towards MAO-A than to MAO-B, which is in accordance with the in vitro-determined Ki values.

5. Conclusion The present study shows that both radiotracers were efficiently labelled with the positron emitter carbon-11. In vivo, both [11C]-RS 2315 and [11C]-RS 2360 penetrate rapidly in the brain, followed by an efficient washout. Furthermore, [11C]-RS 2315 displays a high nonspecific binding and is therefore not suitable for visualization of MAO-A in vivo. [11C]-RS 2360 on the other hand might have potential for mapping MAO-A when PET scans are performed directly after injection of the tracer since specific binding is demonstrated at the earliest time points after injection. Further studies are necessary to evaluate the potential role of [11C]-RS 2360 in MAO-A imaging in humans.

S. De Bruyne et al. / Nuclear Medicine and Biology 37 (2010) 459–467

Acknowledgments LRG and SR thank the Italian MUIR (Ministero dell'Istruzione, Università e Ricerca) for financial support. References [1] Kearney EB, Salach JI, Walker WH, Seng RL, Kenney W, Zeszotek E, et al. The covalently-bound flavin of hepatic monoamine oxidase. Eur J Biochem 1971;24:321–7. [2] Schnaitman C, Erwin VG, Greenawalt WJ. The submitochondrial localization of monoamineoxidase an enzymatic marker for the outer membrane of rat liver mitochondria. J Cell Biol 1967;32:719–35. [3] Youdim MBH, Edmondson D, Tipton KF. The therapeutic potential of monoamine oxidase inhibitors. Nature 2006;7:295–309. [4] Lan NC, Chen C, Shih JC. Expression of functional human monoamine oxidase A and B cDNAs in mammalian cells. J Neurochem 1989;52:1652–4. [5] Bach AW, Lan NC, Johnson DL, Abell CW, Bembenek ME, et al. cDAN cloning of human liver monoamine oxidase A and B: molecular basis of differences in enzymatic properties. Proc Natl Acad Sci U S A 1988;85:4934–8. [6] Dostert PL, Strolin-Benedetti M, Tipton KF. Interactions of monoamine oxidase with substrates and inhibitors. Med Res Rev 1989;9:45–89. [7] Schoepp DD, Azzaro AJ. Specificity of endogenous substrates for types A and B monoamine oxidase in rat striatum. J Neurochem 1981;36:2025–31. [8] Shih JC, Chen K, Ridd MJ. Monoamine oxidase: from genes to behaviour. Annu Rev Neurosci 1999;22:197–217. [9] Westlund KN, Denney RM, Kochersperger LM, Rose RM, Abell CW. Distinct monoamine oxidase A and B populations in primate brain. Science 1985;230:181–3. [10] Saura J, Richards G, Mahy N. Differential age-related changes of MAO-A and MAO-B in mouse brain and peripheral organs. Neurobiol Aging 1994;15:399–408. [11] Rodriguez MJ, Saura J, Billett EE, Finch CC, Mahy N. Cellullar localization of monoamine oxidase A and B in human tissues outside of the central nervous system. Cell Tissue Res 2001;304:215–20. [12] Meyer JH, Ginovart N, Boovariwala A, Sagrati S, Hussey D, Garcia A, et al. Elevated monoamine oxidase A levels in the brain. Arch Gen Psychiatry 2006;63:1209–16.

467

[13] Heafield MTE, Williams AC. Parkinson's disease: clinical and therapeutics effects. Curr Opin Neurol Neurosurg 1992;5:288–94. [14] Fowler JS, Volkow ND, Wang GJ, Pappas N, Logan J, Shea C, et al. Brain monoamine oxidase A inhibition in cigarette smokers. Proc Natl Acad Sci U S A 1996;93:14065–9. [15] Berlin I, Said S, Spreux-Varoquaux O, Olivares R, Launay JM, Puech AJ. Monoamine oxidase A and B activities in heavy smokers. Biol Psychiatry 1995;38:756–61. [16] Shih JC, Thompson RF. Monoamine oxidase in neuropsychiatry and behaviour. Am J Human Genet 1999;65:593–8. [17] Alia-Klein N, Goldstein RZ, Kriplani A, Logan J, Tomasi D, Williams B, et al. Brain monoamine oxidase A activity predicts trait aggression. J Neurosci 2008;28:5099–104. [18] Brunner HG, Nelen M, Breakefield XO, Ropers HH, van Oost BA. Abnormal behaviour associated with a point mutation in the structural gene for monoamine oxidase A. Science 1993;262: 578–80. [19] Fowler JS, Ding YS, Logan J, MacGregor RR, Shea C, et al. Species differences in [11C]clorgyline binding in brain. Nucl Med Biology 2008;28:779–85. [20] Bergström M, Westerberg G, Kihlberg T, Långström B. Synthesis of some 11C-labelled MAO-A inhibitors and their in vivo uptake kinetics in rhesus monkey brain. Nucl Med Biol 1997;24:381–8. [21] Dollé F, Valette H, Bramoulle Y, Guenther I, Fuseau C, Coulon C, et al. Synthesis and in vivo imaging properties of [11C]befloxatone: a novel highly potent positron emission tomography ligand for monoamine oxidase-A. Bioorg Med Chem 2003;13:1771–5. [22] Bergström M, Westerberg G, Långström B. 11C-Harmine as a tracer for monoamine oxidase A (MAO-A): in vitro and in vivo studies. Nucl Med Biol 1997;24:287–93. [23] Silvestri R, La Regina G, De Martino G, Artico M, Befani O, Palumbo M, et al. Simple, potent, and selective pyrrole inhibitors of monoamine oxidase types A and B. J Med 2003;46:917–20. [24] La Regina G, Silvestri R, Artico M, Lavecchia A, Novellino E, Befani E, et al. New pyrrole inhibitors of monoamine oxidase: synthesis, biological evaluation, and structural determinants of MAO-A and MAO-B selectivity. J Med Chem 2007;50:922–31. [25] La Regina G, Silvestri R, Gatti V, Lavecchia A, Novellino E, Befani O, et al. Synthesis, structure–activity relationships and molecular modeling studies of new indole inhibitors of monoamine oxidases A and B. Bioorg Med Chem 2008;22:9729–40. [26] Wilson AA, Garcia A, DaSilva JN, Houle S. An admonition when measuring the lipophilicity of radiotracers using counting techniques. Appl Rad Isot 2001;54:203–8.