- Email: [email protected]
[11C]NS8880, a promising PET radiotracer targeting the norepinephrine transporter
Nuclear Medicine and Biology 41 (2014) 758–764
Contents lists available at ScienceDirect
Nuclear Medicine and Biology journal homepage: www.elsevier.com/locate/nucmedbio
[ 11C]NS8880, a promising PET radiotracer targeting the norepinephrine transporter☆ Karina H. Vase a,⁎, Dan Peters b, Elsebet Ø. Nielsen c, Aage K.O. Alstrup a, Dirk Bender a a b c
PET Center, Aarhus University Hospital, DK-8000 Aarhus C, Denmark DanPET AB, Rosenstigen 7, SE-216 19 Malmö, Sweden NeuroSearch A/S, Pederstrupsvej 93, DK-2750 Ballerup, Denmark
a r t i c l e
i n f o
Article history: Received 16 September 2013 Received in revised form 5 June 2014 Accepted 17 June 2014 Keywords: Norepinephrine transporter PET NS8880 MeNER Preclinical evaluation Radiosynthesis
a b s t r a c t Introduction: Positron emission tomography (PET) imaging of the norepinephrine transporter (NET) is still hindered by the availability of useful PET imaging probes. The present study describes the radiosynthesis and pre-clinical evaluation of a new compound, exo-3-(6-methoxypyridin-2-yloxy)-8-H-8-azabicyclo[3.2.1]octane (NS8880), targeting NET. NS8880 has an in vitro binding profile comparable to desipramine and is structurally not related to reboxetine. Methods: Labeling of NS8880 with [11C] was achieved by a non-conventional technique: substitution of pyridinyl fluorine with [11C]methanolate in a Boc-protected precursor. The isolated [11C]NS8880 was evaluated preclinically both in a pig model (PET scanning) and in a rat model (μPET scanning) and compared to (S,S)-[11C]-Omethylreboxetine ([11C]MeNER). Results: The radiolabeling technique yielded [11C]NS8880 in low (b10%) but still useful yields with high purity. The PET in vivo evaluation in pig and rat revealed a rapid brain uptake of [11C]NS8880 and fast obtaining of equilibrium. Highest binding was observed in thalamic and hypothalamic regions. Pretreatment with desipramine efficiently reduced binding of [11C]NS8880. Conclusion: Based on the pre-clinical results obtained so far [11C]NS8880 displays promising properties for PET imaging of NET. © 2014 Elsevier Inc. All rights reserved.
1. Introduction The norepinephrine transporter (NET) is the membrane glycoprotein responsible for reuptake of norepinephrine from the synapse into the presynaptic nerve terminals and is thereby essential in regulation of norepinephrine (NE) related actions at cellular levels [1]. NET is considered to play important roles in both physiological and pathological processes in the brain and is associated with several neuropsychiatric disorders, including attention deficit hyperactivity disorder (ADHD), substance abuse, neurodegenerative disorders, schizophrenia, and depression [2–4]. Several potent NET inhibitors have been radiolabelled, but high nonspecific binding and relatively low specific signal limit their use as NET imaging agents [5,6]. Currently, the most promising and most widely used positron emission tomography (PET) radiotracers for imaging NET are derivatives of reboxetine: (S,S)-[11C]-O-methylreboxetine ((S,S)-[11C] MeNER) [7–9], and its [18 F]fluoromethyl-analog [18 F]-FMeNER-D2 [10]. They demonstrate specific binding to NET in vivo, but still with a relative low specific-to-non-specific binding ratio, furthermore their uptake
kinetics in human are rather slow [11,12]. Hence, more suitable PET radioligands are still of interest. A general challenge for NET imaging is the widespread distribution of NET in the brain which complicates the identification of a suitable reference region. In human the small region of locus coeruleus has the highest NET density with about 500 fmol/mg protein [13]. Other NET-rich regions are thalamus and hypothalamus [14,15]. Additionally, the regional density of NET is low relative to other monoamine transporters like dopamine and serotonin transporters, which lower the contrast between NET-poor and NET-rich regions. This fact emphasizes the necessity of potential radioligands to display high affinity and excellent selectivity for binding to NET. In the present study, we radiolabelled [ 11C]NS8880 as a potential NET ligand and preliminary evaluated it in vivo in a rat and pig model. NS8880 (exo-3-(6-methoxypyridin-2-yloxy)-8-H-8-azabicyclo[3.2.1] octane) is a novel compound with high affinity and selectivity towards NET (IC50: NET: 5 nM; SERT: 260 nM; DAT: 2000 nM), which structurally is not related to reboxetine, see Fig. 1. For comparison, binding in pig brain of the more widely used [ 11C]MeNER was studied as well. 2. Materials and methods
☆ Results presented in 2009 in part at the EANM Annual Congress in Barcelona. ⁎ Corresponding author at: Department of Nuclear Medicine and PET Center, Aarhus University Hospital, Noerrebrogade 44, DK-8000 Aarhus C, Denmark. Tel.: +45 7846 3034; fax: +45 7846 3020. E-mail address: [email protected] (K.H. Vase). http://dx.doi.org/10.1016/j.nucmedbio.2014.06.004 0969-8051/© 2014 Elsevier Inc. All rights reserved.
2.1. Chemicals All chemicals were purchased from Sigma-Aldrich Ltd. and were used as received without further purification. Aqueous NaOH (3 M)
K.H. Vase et al. / Nuclear Medicine and Biology 41 (2014) 758–764
759
Fig. 1. Structures of NS8880 (1), (S,S)-MeNER (2), (S,S)-FMeNER-D2, and reboxetine (4).
and aqueous phosphate buffer (70 mM) were manufactured in pharmaceutical grade at the pharmacy of Aarhus University Hospital. Solvents used for HPLC were of HPLC grade. Non-radioactive NS8880 and the Boc-protected precursor KIB14261 were synthesized according to previously described procedures [16]. (S,S)-O-desmethylreboxetine was kindly donated by Alan Wilson, Toronto, CA. 2.2. In vitro determination of IC50 Studies of serotonin, dopamine and noradrenaline uptake were carried out in vitro using brain tissue obtained from male Wistar rats (150–200 g). Samples of cerebral cortices were used for serotonin uptake, samples of corpus striata were used for dopamine uptake, and samples of hippocampi were used for noradrenaline uptake. All samples were homogenized in ice-cold 0.32 M sucrose containing 1 mM pargyline (Sigma Chemicals). Preparation of synaptosomes and incubation of samples for ten minutes in the presence of [ 3H] serotonin (1 nM), [ 3H]dopamine (1 nM) or [ 3H]noradrenaline (1 nM) were carried out according to conventional procedures [17,18]. All tritiated radiopharmaceuticals were from GE Healthcare (Little Chalfont, UK). Samples were co-incubated with NS8880 at concentrations ranging from 0.01 to 30 μM for noradrenaline and dopamine uptake, and 0.0001 to 1 μM for serotonin uptake. After completion of the incubations, samples were poured directly onto Whatman GF/C glass fibre filters under reduced pressure. The filters were washed three times with five ml of ice-cold 0.9% (w/v) NaCl solution, and the concentration of radioactivity retained on the filters was determined by conventional liquid scintillation counting. Specific uptake was calculated as the difference between total uptake and uptake measured in the presence of a selective uptake inhibitor (citalopram, Lundbeck Pharmaceuticals; benztropine, RBI; or desipramine, Sigma Chemicals for inhibition of serotonin, dopamine or noradrenaline uptake respectively) at a final concentration of 1 μM. 2.3. Radiochemistry [ 11C]Carbon dioxide was produced at the PET center with a PETtrace cyclotron (GEMS, Uppsala, Sweden) using 16.4 MeV protons in the 14 N(p,α) 11C reaction on nitrogen gas. [ 11C]Methyl iodide was
prepared from [ 11C]carbon dioxide by reduction to [ 11C]methane followed by gas phase methylation employing a GE Healthcare MeI Box. The labeling procedure including HPLC purification, rotary evaporation, and labeled product formulation, was achieved using a fully automated system (Synthia, Uppsala University PET center, Uppsala, Sweden). The radiosynthesis of [ 11C]NS8880 was accomplished in a 2-step procedure by radiolabelling of the Boc-protected precursor with [ 11C] methanolate prepared in situ and a subsequent deprotection step as shown in Scheme 1. [ 11C]methyl iodide was trapped in a DMSO solution (300 μl) containing NaOH (10 μl, 3 M) and precursor KIB14261 (1.0 mg). After heating for 5 min at 130 °C, the crude Boc-protected product was purified by preparative HPLC using a Phenomenex Luna C18 (10 μm, 250 x 10 mm) column with 70% acetonitrile/30% aqueous NaH2PO4 (70 mM) as eluent (10 ml/min, λ = 230 nm). A typical preparative HPLC chromatogram is shown in Fig. 2. The fraction containing the intermediate (retention time, 910 min) was collected and subsequently evaporated to remove the mobile phase. The residue was treated with hydrochloric acid (2 ml, 1.0 M) for 7 min at 90 °C to remove the Boc protecting group. Finally, the resultant product solution was neutralized by the addition of NaOH (3 M) and used without further formulation. The radiochemical purity of the final product was determined by reversed phase HPLC using a Phenomenex Luna CN 5 μm column (250 x 4.6 mm) with acetonitrile and 70 mM NaH2PO4 (30:70) as mobile phase at flow rate 1 ml/min. [ 11C]NS8880 was identified by co-injection with unlabeled NS8880. [ 11C]MeNER was synthesized by the common N-methylation of the phenolic precursor using [ 11C]methyl iodide according to previously described procedures [7,8].
2.4. PET imaging Animal experiments were performed in accordance with the Danish Animal Experimentation Act with a license granted by the Danish Animal Experimentation Inspectorate. The in vivo brain uptake of [ 11C]NS8880 was studied in a single male Wistar rat and a single female domestic pig (Danish Landrace Yorkshire). Both animals had free access to tap water. The pig was deprived of food 16 hours prior to intramuscular premedication with midazolam and S-ketamine. In both animals, the anaesthesia was induced and maintained with
Scheme 1. Radiosynthesis of [11C]NS8880.
760
K.H. Vase et al. / Nuclear Medicine and Biology 41 (2014) 758–764
Fig. 2. Preparative HPLC chromatogram from the purification of Boc-protected [11C]NS8880.
1.5-2.0% isoflurane in O2/N2O. Catheters were surgically inserted into the femoral vein and artery for both animals. The rat was placed in a MicroPET scanner (Concorde Microsystems Inc., Knoxville, TN) and the pig in a Siemens ECAT Exact HR scanner (Siemens, Knoxville, TN). During imaging the anaesthesia levels were monitored by corneal reflexes, and furthermore, the pig heart rate, blood pressure and blood gases were monitored, and any aberrations were corrected. After short attenuation scans, both animals were intravenously injected with a bolus of [ 11C]NS8880 (rat: 26 MBq, and pig: 80 MBq) and baseline scanned for 90 minutes. Subsequently, pre-treatment experiments were performed in which the selective NET inhibitor desipramine (1 mg/kg) was injected intravenously in both animals. After ten minutes a second 90 minutes PET scan was performed after bolus injections (rat: 16 MBq and pig: 60 MBq) with a new production of [ 11C]NS8880. Arterial blood samples were repeatedly collected from both animals during the baseline and the second scan to establish time activity curves. In pig a total of 30 arterial blood samples were collected at intervals increasing from 5 seconds to 10 minutes. In rat the number of blood samples was restricted to 3 samples used for metabolite analysis. Hypothermia was avoided by warming the rat and pig, and changes in body temperature were automatically controlled by self-regulating heating systems. Furthermore, in another pig study we were testing the uptake of [ 11C]MeNER for comparison to [11C]NS8880 uptake.
This animal was prepared for PET scanning similar to the first pig. After a bolus injection of 388 MBq [ 11C]MeNER, it was PET scanned in 90 minutes (baseline). Subsequently, pre-treatment experiments were performed by bolus injection with the blocker desipramine (1 mg/kg) IV, and a second 90 minutes PET scan was performed with a new production of 404 MBq [ 11C]MeNER. At the end of the experiments all three animals were euthanized by an overdose of pentobarbitone. For the pig studies, the summed emission image of [ 11C]NS8880 and [ 11C]MeNER in the baseline condition was manually co-registered to the statistical MR atlas of porcine brain [19]. The calculated transformation matrix was used to resample the dynamic emission sequences into the common stereotaxic space. For direct comparison of the two tracers, voxel-vise parametric maps of the distribution volume, Vd (ml g -1), were calculated using the metabolite-corrected arterial input and the linearization of Zhou et al. [20] in the baseline condition and after the desipramine challenge. The acquired 3-dimensional μPET emission data were reconstructed to temporally framed sinograms by use of Fourier rebinning and an ordered-subsets expectation maximization reconstruction algorithm without attenuation correction. Image visualization was performed with ASIPro software (Concorde Microsystems Inc.). Regions of interest (ROIs) were manually assigned according to a rat brain atlas [21]. 2.5. Metabolite analysis
Table 1 IC50 values [nM] for NS8880. MeNER, reboxetine, and desipramine are included for comparison [6]. Compound
DAT
SERT
NET
NS8880⁎ MeNER⁎ Reboxetine⁎ Desipramine⁎⁎
2000 N10 000 N10 000 3190 ± 40
260 310 1070 17.6 ± 0.7
5.0 2.48 8.2 0.83 ± 0.05
⁎ SD values not available. ⁎⁎ For desipramine, the affinities are expressed by Kd values.
Arterial plasma analysis of [11C]NS8880 metabolism was performed in pig and rat for each scan. Blood samples were collected from the pig at 6, 10, 20, 30, 40, 60 and 80 minutes after injection of [11C]NS8880 and at 5, 10, and 20 minutes from the rat. The blood samples were centrifuged (1 min × 13,000 rpm) and 500 μl of plasma was mixed with 500 μl acetonitrile to precipitate plasma proteins. After centrifugation (5 min × 13,000 rpm) the supernatant was analyzed and fractionated by HPLC (Agilent Zorbax SB CN 5 μm, 250 × 9.4 mm, methanol:ammonium formate (0.1 M, pH 4.1) 35:65, 6 ml/min). Detection consisted of serial ultraviolet detection (λ = 230 nm) and gamma detection.
A
10000
thalamus hypothalamus frontal cortex occipital cortex striatum cerebellum
8000 6000 4000 2000 0 0
20
40
60
80
Radioactivity concentration (Bq/cc)
Radioactivity concentration (Bq/cc)
K.H. Vase et al. / Nuclear Medicine and Biology 41 (2014) 758–764
B
300000
thalamus frontal cortex striatum cerebellum
250000 200000 150000 100000 50000 0 0
20
C Thalamus
8000 6000 4000 2000
[11C]NS8880 Desipramine + [11C]NS8880
0 0
20
40
60
150000
100000
50000 [11C]NS8880 Desipramine + [11C]NS8880
0 0
20
Radioactivity concentration (Bq/cc)
Radioactivity concentration (Bq/cc)
8000 6000 4000
[11C]NS8880 Desipramine + [11C]NS8880
0 20
40
60
40
60
80
Time after injection (min)
Striatum
0
80
Thalamus
80
E
2000
60
D
200000
Time after injection (min) 10000
40
Time after injection (min) Radioactivity concentration (Bq/cc)
Radioactivity concentration (Bq/cc)
Time after injection (min)
10000
761
80
F
200000
Striatum 150000
100000
50000 [11C]NS8880 Desipramine + [11C]NS8880
0 0
Time after injection (min)
20
40
60
80
Time after injection (min)
Fig. 3. Time-activity curves after administration of [11C]NS8880 in pig (A) and rat (B). Time-activity curves after administration of [11C]NS8880 at baseline conditions and after pretreatment with desipramine (1 mg/kg) for selective ROIs in pig brain (C), (E), and in rat brain (D), (F).
11
C radioactivity concentrations were measured in a well counter (Packard Biosciences). 3. Results
(from EOB) up to 180 MBq of [ 11C]NS8880 was isolated with a radiochemical purity N99% and a specific radioactivity in the range of 20–70 GBq/μmol. [ 11C]MeNER was isolated with a radiochemical purity N98% and a specific radioactivity of around 50 GBq/μmol.
3.1. In vitro determination of IC50 IC50 values determined in rat synaptosomes revealed a comparable selective binding to norepinephrine transporters as summarized in Table 1. The IC50 values are comparable to those of reboxetine and MeNER, though with less selectivity than reboxetine. 3.2. Radiochemistry The applied radiosynthesis yielded [ 11C]NS8880 in low but still useful radiochemical yields, 5-8% decay corrected yield based on produced [ 11C]methyl iodide. Within 45 minutes total synthesis time
3.3. PET imaging The PET study in pig and the μPET studies in rat both revealed for [ 11C]NS8880 a fast brain uptake with highest accumulation in thalamus, hypothalamus, and frontal cortex. Interestingly, considerable uptake was found as well in striatum. Based on time activity curves equilibrium conditions were reached already after 15 minutes in pig and 10 minutes in rat. The administration of desipramine resulted in a reduction of around 20% of the area under the curve of the time activity curves in pig and around 25% in rat, Fig. 3.
762
K.H. Vase et al. / Nuclear Medicine and Biology 41 (2014) 758–764
Fig. 4. Maps of distribution volume, Vd (ml g−1) of [11C]NS8880 in brain of living pig (A) at baseline conditions and (B) after treatment with desipramine (1 mg/kg). From left to right: coronal, sagittal, and transverse images. The distribution volumes were calculated by the method of Zhou relative to the arterial input function.
Parametric mapping of the distribution volume (Vd) obtained in pig confirmed highest Vd values in thalamus, frontal cortex, and striatum. Pretreatment with desipramine reduced the Vd generally by roughly 40%; see Fig. 4 and Table 2. PET imaging with [ 11C]MeNER revealed the expected pattern with highest Vd in thalamic regions and lowest binding in occipital cortex. Binding is reduced around 25% by desipramine treatment, see Fig. 5 and Table 2. Based on an estimate from the time activity curves, equilibrium conditions were first attained within the end of the 90 minutes scan period, cf. Fig. 6. Upon desipramine block we found an initial higher brain up-take of [ 11C]MeNER, reflecting somewhat more available radiotracer due to blocking of peripheral binding sites. 3.4. Metabolite analysis [ 11C]NS8880 showed a rapid metabolism both in rat and in pig. In pig less than 20% of parent radiotracer was detected in plasma 20 minutes after tracer injection and only 5% after 30 minutes. In rat, 35% of unchanged [ 11C]NS8880 was observed after 20 minutes. 4. Discussion The radiolabeling method used in the present study successfully produced [ 11C]NS8880 in low but still useful yields for in vivo evaluation. The straightforward direct methylation of O-desmethyl NS8880 using [ 11C]methyl iodide or [ 11C]methyl triflate could not be applied as this hydroxy starting material is not stable. We therefore Table 2 Vd values of [11C]NS8880 and [11C]MeNER obtained in pig brain. Region
Striatum Thalamus Frontal cortex Occipital cortex Cerebellum
[11C]NS8880
[11C]MeNER
Baseline
Desipramine block
Baseline
Desipramine block
3.6 4.1 3.4 3.2 3.3
2.2 2.5 1.9 1.9 1.8
3.1 4.2 3.1 3.1 3.4
2.7 3.2 2.6 2.5 2.6
± ± ± ± ±
0.3 0.3 0.5 0.5 0.4
± ± ± ± ±
0.2 0.2 0.3 0.3 0.4
± ± ± ± ±
0.3 0.3 0.5 0.5 0.4
± ± ± ± ±
0.2 0.1 0.3 0.4 0.3
considered the aromatic substitution of a halogen by [ 11C]methanolate prepared in situ as labeling approach. Pyridines with halogens in 2-position are activated aromatic systems and the halogens are comparable easy to substitute within aromatic nucleophilic substitution reactions. However, these reactions normally require excessive amounts of nucleophile, high temperatures and long reaction times, which explain our use of rather harsh reaction conditions. In the first experiments a 2-chloro analog was applied, but no product formation was observed (data not shown), probably due to too slow reaction kinetics. We therefore decided to use the more reactive 2-fluoro derivative as starting material (see Scheme 1). To our knowledge we could hereby for the first time demonstrate the application of a 2-pyridinyl fluorine substrate and [ 11C]methanolate in PET radiochemistry. The amount of isolated [ 11C]NS8880 was comparable low, which probably is due to the combination of two parameters; firstly the loss of gaseous [ 11C]methyl iodide from the reaction mixture owing to the harsh reaction conditions and secondly a slow reaction kinetics. We decided to generate the [ 11C]methanolate starting from [ 11C]methyl iodide and sodium hydroxide in situ. It might be argued that the isolated radiochemical yields could be improved simply by first generating [ 11C]methanolate under milder reaction conditions and applying the starting material and the harsh reaction conditions afterwards. However, we did not prosecute this approach as our C-11 chemistry set-up would have required a manual addition of the precursor, which is in terms of radiation protection unfavorable. The assumption that higher radiochemical yields could be obtained upon usage of pre-prepared [ 11C]methanolate is supported by preparative radio-HPLC which revealed up to 30% of the radiolabeled Bocprotected intermediate in the reaction mixture (Fig. 2), whereas the radiochemical yield of [ 11C]NS8880 based on the average [ 11C]methyl iodide production is only around 5%. It might still be argued, that the actual labeling route proceeds via substitution of fluoride by hydroxide, followed by [ 11C]methylation of this phenolic compound. However, these 2-hydroxy pyridins are not stable compounds and would have resulted in a decomposition of the intermediate material. This decomposition was not observed.
K.H. Vase et al. / Nuclear Medicine and Biology 41 (2014) 758–764
763
Fig. 5. Maps of distribution volume, Vd (ml g-1) of [11C]MeNER in brain of living pig (A) at baseline conditions and (B) after treatment with desipramine (1 mg/kg). From left to right: coronal, sagittal, and transverse images. The distribution volumes were calculated by the method of Zhou relative to the arterial input function.
A
70000
thalamus hypothalamus frontal cortex occipital cortex striatum cerebellum
60000 30000 20000 10000 0 0
20
40
60
Time after injection (min)
80
[ 11C]NS8880 is more sensitive towards desipramine block compared to [ 11C]MeNER, around 40% vs 25% in all regions. This might reflect a somewhat lower degree of non-specific binding of [ 11C]NS8880. It is a general challenge for NET imaging to identify a suitable reference region because of the widespread and relatively uniform distribution of NET throughout the brain and several regions have been evaluated as potential reference regions [22,23]. In the present study this also constitute a problem and so far a suitable reference region could not be identified due to considerable uptake in all regions. Like other groups we found a lower uptake in occipital cortex compared to cerebellum [23,24]. A rough estimate of the binding potential is around 0.3 obtained by the ratio between Vd in thalamus and occipital cortex minus 1. Binding potential might be higher, if a more suitable reference region could be identified. Based on the limited data, Vd in thalamic regions is slightly higher for [ 11C]MeNER compared to [ 11C]NS8880, resulting in a slightly higher estimate for the binding potential (0.4 vs 0.3). It should be noted, that the effective binding potential for [ 11C]MeNER in humans is likewise only around 0.4-0.5 [11,25]. A disadvantage using [ 11C]MeNER in human studies is
Radioactivity concentration (Bq/cc)
Radioactivity concentration (Bq/cc)
In pig and rat [ 11C]NS8880 quickly metabolized with only around 20-30% parent compound after 20 minutes. Interestingly we found a slightly faster metabolism in pig compared to rat. This is in contrast to our normal observation with other radiotracers with faster metabolism in small animals. [ 11C]NS8880 enters rapidly the brain and equilibrium is reached remarkably fast, both in pig and rat. Highest binding is observed in regions with high NET concentration, i.e. thalamic and hypothalamic regions. The maps of Vd reveal a binding pattern comparable to those of desipramine as demonstrated by the blocking experiment using desipramine. Considerable uptake was observed in striatum and the striatal binding seems to be displaceable by desipramine. In another experiment in rat, uptake of [ 11C]NS8880 in all regions was not affected by blocking with citalopram, thus the striatal uptake is likely not due to the binding to serotonin transporters (data not shown). In pig brain the binding pattern of [ 11C]NS8880 is close to that of 11 [ C]MeNER (see Figs. 4 and 5) with highest uptake in thalamic and hypothalamic regions. However, we found a considerable higher Vd for [ 11C]NS8880 in striatum compared to [ 11C]MeNER. The binding of
B
70000
thalamus hypothalamus frontal cortex occipital cortex striatum cerebellum
60000 30000 20000 10000 0 0
20
40
60
80
Time after injection (min)
Fig. 6. Time-activity curves for selective ROIs in pig brain after administration of [11C]MeNER (A) at baseline conditions and (B) after pre-treatment with desipramine (1 mg/kg).
764
K.H. Vase et al. / Nuclear Medicine and Biology 41 (2014) 758–764
the slow kinetics of this compound; peak equilibrium binding is not reached during a 90-minute PET procedure [11] and as a consequence data acquisition times of 120 minutes have been applied in recent studies [25]. In pig, we found similarly a slow kinetics for [ 11C]MeNER. [ 11C]NS8880 shows here at least in pig and rat a considerable faster kinetics, which is advantageous. Of course it is unknown if this fast kinetics would be observed in humans as well. 5. Conclusion Our initial data indicate that [ 11C]NS8880 is usable for imaging NET and due to its faster kinetics might in some respect be superior to [ 11C]MeNER, even so the binding potential is still low. Acknowledgement We acknowledge the technical assistance of Mette Simonsen, Kim Vang, Jørgen Bach Pedersen, Gitte Friberg, Kirsten Braad Iskov, and Tove Thomsen. Support to DanPET AB from CCJobs, Interreg, Sweden-Denmark, and COGNITO, NRU, Copenhagen is gratefully acknowledged. References [1] Blakely RD, De Felice LJ, Hartzell HC. Molecular physiology of norepinephrine and serotonin transporters. J Exp Biol 1994;196:263–81. [2] Biederman J, Spencer T. Attention-deficit/hyperactivity disorder (ADHD) as a norepinephrine disorder. Biol Psychiatry 1999;46:1234–42. [3] Zhou J. Norepinephrine transporter inhibitors and their therapeutic potential. Drugs Future 2004;29:1235–44. [4] Gulyás B, Brockschnieder D, Nag S, Pavlova E, Kása P, Beliczai Z, et al. The norepinephrine transporter (NET) radioligand (S, S)-[18 F]FMeNER-D2 shows significant decreases in NET density in the human brain in Alzheimer's disease: A post-mortem autoradiographic study. Neurochem Int 2010;56:789–98. [5] Ding Y-S, Fowler J. New-generation radiotracers for nAChR and NET. Nucl Med Biol 2005;32:707–18. [6] Ding Y-S, Lin K-S, Logan J. PET imaging of norepinephrine transporters. Curr Pharm Des 2006;12:3831–45. [7] Schou M, Halldin C, Sovago J, Pike VW, Gulyas B, Mozley PD, et al. Specific in vivo binding to the norepinephrine transporter demonstrated with the PET radioligand, (S, S)-[11C]MeNER. Nucl Med Biol 2003;30:707–14. [8] Wilson AA, Johnson DP, Mozley D, Hussey D, Ginovart N, Nobrega J, et al. Synthesis and in vivo evaluation of novel radiotracers for the in vivo imaging of the norepinephrine transporter. Nucl Med Biol 2003;30:85–92.
[9] Ding YS, Lin KS, Garza V, Carter P, Alexoff D, Logan J, et al. Evaluation of a new norepinephrine transporter PET ligand in baboons, both in brain and peripheral organs. Synapse 2003;50:345–52. [10] Schou M, Halldin C, Sóvágó J, Pike VW, Hall H, Gulyás B, et al. PET evaluation of novel radiofluorinated reboxetine analogs as norepinephrine transporter probes in the monkey brain. Synapse 2004;53:57–67. [11] Logan J, Wang G-J, Telang F, Fowler JS, Alexoff D, Zabroski J, et al. Imaging the norepinephrine transporter in humans with (S, S)-[11C]O-methyl reboxetine and PET: problems and progress. Nucl Med Biol 2007;34:667–79. [12] Takano A, Varrone A, Gulyás B, Karlsson P, Tauscher J, Halldin C. Mapping of the norepinephrine transporter in the human brain using PET with (S, S)-[18 F] FMeNER-D2. NeuroImage 2008;42:474–82. [13] Klimek V, Stockmeier C, Overholser J, Meltzer HY, Kalka S, Dilley G, et al. Reduced levels of norepinephrine transporters in the locus coeruleus in major depression. J Neurosci 1997;17:8451–8. [14] Gross-Isseroff R, Israeli M, Biegon A. Autoradiographic analysis of [3H] desmethylimipramine binding in the human brain postmortem. Brain Res 1988;456:120–6. [15] Schou M, Halldin C, Pike VW, Mozley PD, Dobson D, Innis RB, et al. Post-mortem human brain autoradiography of the norepinephrine transporter using (S, S)[18 F]FMeNER-D2. Eur Neuropsychopharmacol 2005;15:517–20. [16] Peters D, Eriksen BL, Nielsen EØ, Scheel-Krüger J, Olsen GM. Novel 8-aza-bicyclo[3. 2.1]octane derivatives and their use as monoamine neurotransmitter re-uptake inhibitors. WO 2004/113334 A1. [17] Low WC, Whitehorm D, Hendley ED. Genetically related rats with differences in hippocampal uptake of norepinephrine and maze performance. Brain Res Bull 1984;12:703–9. [18] Bolden-Watson C, Richelson E. Blockade by newly-developed antidepressants of biogenic amine uptake into rat brain synaptosomes. Life Sci 1993;52:1023–9. [19] Watanabe H, Andersen F, Simonsen CZ, Evans SM, Gjedde A, Cumming P. MR-based statistical atlas of the Göttingen Minipig brain. NeuroImage 2001;14:1089–96. [20] Zhou Y, Huang S-C, Bergsneider M, Wong DF. Improved parametric image generation using spatial-temporal analysis of dynamic PET studies. NeuroImage 2002;15:697–707. [21] Paxinos G, Watson C. The rat brain in stereotaxic coordinates. 3rd ed. San Diego: Academic press; 1997. [22] Logan J, Ding Y-S, Lin K-S, Pareto D, Fowler J, Biegon A. Modeling and analysis of PET studies with norepinephrine transporter ligands: the search for a reference region. Nucl Med Biol 2005;32:531–42. [23] Gallezot J-D, Weinzimmer D, Nabulsi N, Lin S-F, Fowles K, Sandiego C, et al. Evaluation of [ 11C]MRB for assessment of occupancy of norepinephrine transporter: Studies with atomoxetine in non-human primates. NeuroImage 2011;56:268–79. [24] Hannestad J, Gallezot J-D, Planeta-Wilson B, Lin S-F, Williams WA, van Dyck CH, et al. Clinically relevant doses of methylphenidate significantly occupy norepinephrine transporters in human in vivo. Biol Psychiatry 2010;68:854–60. [25] Ding Y-S, Singhal T, Planeta-Wilson B, Gallezot J-D, Nabulsi N, Labaree D, et al. PET imaging of the effects of age and cocaine on the norepinephrine transporter in the human brain using (S, S)-[11C]O-methylreboxetine and HRRT. Synapse 2010;64:30–8.
Contents lists available at ScienceDirect
Nuclear Medicine and Biology journal homepage: www.elsevier.com/locate/nucmedbio
[ 11C]NS8880, a promising PET radiotracer targeting the norepinephrine transporter☆ Karina H. Vase a,⁎, Dan Peters b, Elsebet Ø. Nielsen c, Aage K.O. Alstrup a, Dirk Bender a a b c
PET Center, Aarhus University Hospital, DK-8000 Aarhus C, Denmark DanPET AB, Rosenstigen 7, SE-216 19 Malmö, Sweden NeuroSearch A/S, Pederstrupsvej 93, DK-2750 Ballerup, Denmark
a r t i c l e
i n f o
Article history: Received 16 September 2013 Received in revised form 5 June 2014 Accepted 17 June 2014 Keywords: Norepinephrine transporter PET NS8880 MeNER Preclinical evaluation Radiosynthesis
a b s t r a c t Introduction: Positron emission tomography (PET) imaging of the norepinephrine transporter (NET) is still hindered by the availability of useful PET imaging probes. The present study describes the radiosynthesis and pre-clinical evaluation of a new compound, exo-3-(6-methoxypyridin-2-yloxy)-8-H-8-azabicyclo[3.2.1]octane (NS8880), targeting NET. NS8880 has an in vitro binding profile comparable to desipramine and is structurally not related to reboxetine. Methods: Labeling of NS8880 with [11C] was achieved by a non-conventional technique: substitution of pyridinyl fluorine with [11C]methanolate in a Boc-protected precursor. The isolated [11C]NS8880 was evaluated preclinically both in a pig model (PET scanning) and in a rat model (μPET scanning) and compared to (S,S)-[11C]-Omethylreboxetine ([11C]MeNER). Results: The radiolabeling technique yielded [11C]NS8880 in low (b10%) but still useful yields with high purity. The PET in vivo evaluation in pig and rat revealed a rapid brain uptake of [11C]NS8880 and fast obtaining of equilibrium. Highest binding was observed in thalamic and hypothalamic regions. Pretreatment with desipramine efficiently reduced binding of [11C]NS8880. Conclusion: Based on the pre-clinical results obtained so far [11C]NS8880 displays promising properties for PET imaging of NET. © 2014 Elsevier Inc. All rights reserved.
1. Introduction The norepinephrine transporter (NET) is the membrane glycoprotein responsible for reuptake of norepinephrine from the synapse into the presynaptic nerve terminals and is thereby essential in regulation of norepinephrine (NE) related actions at cellular levels [1]. NET is considered to play important roles in both physiological and pathological processes in the brain and is associated with several neuropsychiatric disorders, including attention deficit hyperactivity disorder (ADHD), substance abuse, neurodegenerative disorders, schizophrenia, and depression [2–4]. Several potent NET inhibitors have been radiolabelled, but high nonspecific binding and relatively low specific signal limit their use as NET imaging agents [5,6]. Currently, the most promising and most widely used positron emission tomography (PET) radiotracers for imaging NET are derivatives of reboxetine: (S,S)-[11C]-O-methylreboxetine ((S,S)-[11C] MeNER) [7–9], and its [18 F]fluoromethyl-analog [18 F]-FMeNER-D2 [10]. They demonstrate specific binding to NET in vivo, but still with a relative low specific-to-non-specific binding ratio, furthermore their uptake
kinetics in human are rather slow [11,12]. Hence, more suitable PET radioligands are still of interest. A general challenge for NET imaging is the widespread distribution of NET in the brain which complicates the identification of a suitable reference region. In human the small region of locus coeruleus has the highest NET density with about 500 fmol/mg protein [13]. Other NET-rich regions are thalamus and hypothalamus [14,15]. Additionally, the regional density of NET is low relative to other monoamine transporters like dopamine and serotonin transporters, which lower the contrast between NET-poor and NET-rich regions. This fact emphasizes the necessity of potential radioligands to display high affinity and excellent selectivity for binding to NET. In the present study, we radiolabelled [ 11C]NS8880 as a potential NET ligand and preliminary evaluated it in vivo in a rat and pig model. NS8880 (exo-3-(6-methoxypyridin-2-yloxy)-8-H-8-azabicyclo[3.2.1] octane) is a novel compound with high affinity and selectivity towards NET (IC50: NET: 5 nM; SERT: 260 nM; DAT: 2000 nM), which structurally is not related to reboxetine, see Fig. 1. For comparison, binding in pig brain of the more widely used [ 11C]MeNER was studied as well. 2. Materials and methods
☆ Results presented in 2009 in part at the EANM Annual Congress in Barcelona. ⁎ Corresponding author at: Department of Nuclear Medicine and PET Center, Aarhus University Hospital, Noerrebrogade 44, DK-8000 Aarhus C, Denmark. Tel.: +45 7846 3034; fax: +45 7846 3020. E-mail address: [email protected] (K.H. Vase). http://dx.doi.org/10.1016/j.nucmedbio.2014.06.004 0969-8051/© 2014 Elsevier Inc. All rights reserved.
2.1. Chemicals All chemicals were purchased from Sigma-Aldrich Ltd. and were used as received without further purification. Aqueous NaOH (3 M)
K.H. Vase et al. / Nuclear Medicine and Biology 41 (2014) 758–764
759
Fig. 1. Structures of NS8880 (1), (S,S)-MeNER (2), (S,S)-FMeNER-D2, and reboxetine (4).
and aqueous phosphate buffer (70 mM) were manufactured in pharmaceutical grade at the pharmacy of Aarhus University Hospital. Solvents used for HPLC were of HPLC grade. Non-radioactive NS8880 and the Boc-protected precursor KIB14261 were synthesized according to previously described procedures [16]. (S,S)-O-desmethylreboxetine was kindly donated by Alan Wilson, Toronto, CA. 2.2. In vitro determination of IC50 Studies of serotonin, dopamine and noradrenaline uptake were carried out in vitro using brain tissue obtained from male Wistar rats (150–200 g). Samples of cerebral cortices were used for serotonin uptake, samples of corpus striata were used for dopamine uptake, and samples of hippocampi were used for noradrenaline uptake. All samples were homogenized in ice-cold 0.32 M sucrose containing 1 mM pargyline (Sigma Chemicals). Preparation of synaptosomes and incubation of samples for ten minutes in the presence of [ 3H] serotonin (1 nM), [ 3H]dopamine (1 nM) or [ 3H]noradrenaline (1 nM) were carried out according to conventional procedures [17,18]. All tritiated radiopharmaceuticals were from GE Healthcare (Little Chalfont, UK). Samples were co-incubated with NS8880 at concentrations ranging from 0.01 to 30 μM for noradrenaline and dopamine uptake, and 0.0001 to 1 μM for serotonin uptake. After completion of the incubations, samples were poured directly onto Whatman GF/C glass fibre filters under reduced pressure. The filters were washed three times with five ml of ice-cold 0.9% (w/v) NaCl solution, and the concentration of radioactivity retained on the filters was determined by conventional liquid scintillation counting. Specific uptake was calculated as the difference between total uptake and uptake measured in the presence of a selective uptake inhibitor (citalopram, Lundbeck Pharmaceuticals; benztropine, RBI; or desipramine, Sigma Chemicals for inhibition of serotonin, dopamine or noradrenaline uptake respectively) at a final concentration of 1 μM. 2.3. Radiochemistry [ 11C]Carbon dioxide was produced at the PET center with a PETtrace cyclotron (GEMS, Uppsala, Sweden) using 16.4 MeV protons in the 14 N(p,α) 11C reaction on nitrogen gas. [ 11C]Methyl iodide was
prepared from [ 11C]carbon dioxide by reduction to [ 11C]methane followed by gas phase methylation employing a GE Healthcare MeI Box. The labeling procedure including HPLC purification, rotary evaporation, and labeled product formulation, was achieved using a fully automated system (Synthia, Uppsala University PET center, Uppsala, Sweden). The radiosynthesis of [ 11C]NS8880 was accomplished in a 2-step procedure by radiolabelling of the Boc-protected precursor with [ 11C] methanolate prepared in situ and a subsequent deprotection step as shown in Scheme 1. [ 11C]methyl iodide was trapped in a DMSO solution (300 μl) containing NaOH (10 μl, 3 M) and precursor KIB14261 (1.0 mg). After heating for 5 min at 130 °C, the crude Boc-protected product was purified by preparative HPLC using a Phenomenex Luna C18 (10 μm, 250 x 10 mm) column with 70% acetonitrile/30% aqueous NaH2PO4 (70 mM) as eluent (10 ml/min, λ = 230 nm). A typical preparative HPLC chromatogram is shown in Fig. 2. The fraction containing the intermediate (retention time, 910 min) was collected and subsequently evaporated to remove the mobile phase. The residue was treated with hydrochloric acid (2 ml, 1.0 M) for 7 min at 90 °C to remove the Boc protecting group. Finally, the resultant product solution was neutralized by the addition of NaOH (3 M) and used without further formulation. The radiochemical purity of the final product was determined by reversed phase HPLC using a Phenomenex Luna CN 5 μm column (250 x 4.6 mm) with acetonitrile and 70 mM NaH2PO4 (30:70) as mobile phase at flow rate 1 ml/min. [ 11C]NS8880 was identified by co-injection with unlabeled NS8880. [ 11C]MeNER was synthesized by the common N-methylation of the phenolic precursor using [ 11C]methyl iodide according to previously described procedures [7,8].
2.4. PET imaging Animal experiments were performed in accordance with the Danish Animal Experimentation Act with a license granted by the Danish Animal Experimentation Inspectorate. The in vivo brain uptake of [ 11C]NS8880 was studied in a single male Wistar rat and a single female domestic pig (Danish Landrace Yorkshire). Both animals had free access to tap water. The pig was deprived of food 16 hours prior to intramuscular premedication with midazolam and S-ketamine. In both animals, the anaesthesia was induced and maintained with
Scheme 1. Radiosynthesis of [11C]NS8880.
760
K.H. Vase et al. / Nuclear Medicine and Biology 41 (2014) 758–764
Fig. 2. Preparative HPLC chromatogram from the purification of Boc-protected [11C]NS8880.
1.5-2.0% isoflurane in O2/N2O. Catheters were surgically inserted into the femoral vein and artery for both animals. The rat was placed in a MicroPET scanner (Concorde Microsystems Inc., Knoxville, TN) and the pig in a Siemens ECAT Exact HR scanner (Siemens, Knoxville, TN). During imaging the anaesthesia levels were monitored by corneal reflexes, and furthermore, the pig heart rate, blood pressure and blood gases were monitored, and any aberrations were corrected. After short attenuation scans, both animals were intravenously injected with a bolus of [ 11C]NS8880 (rat: 26 MBq, and pig: 80 MBq) and baseline scanned for 90 minutes. Subsequently, pre-treatment experiments were performed in which the selective NET inhibitor desipramine (1 mg/kg) was injected intravenously in both animals. After ten minutes a second 90 minutes PET scan was performed after bolus injections (rat: 16 MBq and pig: 60 MBq) with a new production of [ 11C]NS8880. Arterial blood samples were repeatedly collected from both animals during the baseline and the second scan to establish time activity curves. In pig a total of 30 arterial blood samples were collected at intervals increasing from 5 seconds to 10 minutes. In rat the number of blood samples was restricted to 3 samples used for metabolite analysis. Hypothermia was avoided by warming the rat and pig, and changes in body temperature were automatically controlled by self-regulating heating systems. Furthermore, in another pig study we were testing the uptake of [ 11C]MeNER for comparison to [11C]NS8880 uptake.
This animal was prepared for PET scanning similar to the first pig. After a bolus injection of 388 MBq [ 11C]MeNER, it was PET scanned in 90 minutes (baseline). Subsequently, pre-treatment experiments were performed by bolus injection with the blocker desipramine (1 mg/kg) IV, and a second 90 minutes PET scan was performed with a new production of 404 MBq [ 11C]MeNER. At the end of the experiments all three animals were euthanized by an overdose of pentobarbitone. For the pig studies, the summed emission image of [ 11C]NS8880 and [ 11C]MeNER in the baseline condition was manually co-registered to the statistical MR atlas of porcine brain [19]. The calculated transformation matrix was used to resample the dynamic emission sequences into the common stereotaxic space. For direct comparison of the two tracers, voxel-vise parametric maps of the distribution volume, Vd (ml g -1), were calculated using the metabolite-corrected arterial input and the linearization of Zhou et al. [20] in the baseline condition and after the desipramine challenge. The acquired 3-dimensional μPET emission data were reconstructed to temporally framed sinograms by use of Fourier rebinning and an ordered-subsets expectation maximization reconstruction algorithm without attenuation correction. Image visualization was performed with ASIPro software (Concorde Microsystems Inc.). Regions of interest (ROIs) were manually assigned according to a rat brain atlas [21]. 2.5. Metabolite analysis
Table 1 IC50 values [nM] for NS8880. MeNER, reboxetine, and desipramine are included for comparison [6]. Compound
DAT
SERT
NET
NS8880⁎ MeNER⁎ Reboxetine⁎ Desipramine⁎⁎
2000 N10 000 N10 000 3190 ± 40
260 310 1070 17.6 ± 0.7
5.0 2.48 8.2 0.83 ± 0.05
⁎ SD values not available. ⁎⁎ For desipramine, the affinities are expressed by Kd values.
Arterial plasma analysis of [11C]NS8880 metabolism was performed in pig and rat for each scan. Blood samples were collected from the pig at 6, 10, 20, 30, 40, 60 and 80 minutes after injection of [11C]NS8880 and at 5, 10, and 20 minutes from the rat. The blood samples were centrifuged (1 min × 13,000 rpm) and 500 μl of plasma was mixed with 500 μl acetonitrile to precipitate plasma proteins. After centrifugation (5 min × 13,000 rpm) the supernatant was analyzed and fractionated by HPLC (Agilent Zorbax SB CN 5 μm, 250 × 9.4 mm, methanol:ammonium formate (0.1 M, pH 4.1) 35:65, 6 ml/min). Detection consisted of serial ultraviolet detection (λ = 230 nm) and gamma detection.
A
10000
thalamus hypothalamus frontal cortex occipital cortex striatum cerebellum
8000 6000 4000 2000 0 0
20
40
60
80
Radioactivity concentration (Bq/cc)
Radioactivity concentration (Bq/cc)
K.H. Vase et al. / Nuclear Medicine and Biology 41 (2014) 758–764
B
300000
thalamus frontal cortex striatum cerebellum
250000 200000 150000 100000 50000 0 0
20
C Thalamus
8000 6000 4000 2000
[11C]NS8880 Desipramine + [11C]NS8880
0 0
20
40
60
150000
100000
50000 [11C]NS8880 Desipramine + [11C]NS8880
0 0
20
Radioactivity concentration (Bq/cc)
Radioactivity concentration (Bq/cc)
8000 6000 4000
[11C]NS8880 Desipramine + [11C]NS8880
0 20
40
60
40
60
80
Time after injection (min)
Striatum
0
80
Thalamus
80
E
2000
60
D
200000
Time after injection (min) 10000
40
Time after injection (min) Radioactivity concentration (Bq/cc)
Radioactivity concentration (Bq/cc)
Time after injection (min)
10000
761
80
F
200000
Striatum 150000
100000
50000 [11C]NS8880 Desipramine + [11C]NS8880
0 0
Time after injection (min)
20
40
60
80
Time after injection (min)
Fig. 3. Time-activity curves after administration of [11C]NS8880 in pig (A) and rat (B). Time-activity curves after administration of [11C]NS8880 at baseline conditions and after pretreatment with desipramine (1 mg/kg) for selective ROIs in pig brain (C), (E), and in rat brain (D), (F).
11
C radioactivity concentrations were measured in a well counter (Packard Biosciences). 3. Results
(from EOB) up to 180 MBq of [ 11C]NS8880 was isolated with a radiochemical purity N99% and a specific radioactivity in the range of 20–70 GBq/μmol. [ 11C]MeNER was isolated with a radiochemical purity N98% and a specific radioactivity of around 50 GBq/μmol.
3.1. In vitro determination of IC50 IC50 values determined in rat synaptosomes revealed a comparable selective binding to norepinephrine transporters as summarized in Table 1. The IC50 values are comparable to those of reboxetine and MeNER, though with less selectivity than reboxetine. 3.2. Radiochemistry The applied radiosynthesis yielded [ 11C]NS8880 in low but still useful radiochemical yields, 5-8% decay corrected yield based on produced [ 11C]methyl iodide. Within 45 minutes total synthesis time
3.3. PET imaging The PET study in pig and the μPET studies in rat both revealed for [ 11C]NS8880 a fast brain uptake with highest accumulation in thalamus, hypothalamus, and frontal cortex. Interestingly, considerable uptake was found as well in striatum. Based on time activity curves equilibrium conditions were reached already after 15 minutes in pig and 10 minutes in rat. The administration of desipramine resulted in a reduction of around 20% of the area under the curve of the time activity curves in pig and around 25% in rat, Fig. 3.
762
K.H. Vase et al. / Nuclear Medicine and Biology 41 (2014) 758–764
Fig. 4. Maps of distribution volume, Vd (ml g−1) of [11C]NS8880 in brain of living pig (A) at baseline conditions and (B) after treatment with desipramine (1 mg/kg). From left to right: coronal, sagittal, and transverse images. The distribution volumes were calculated by the method of Zhou relative to the arterial input function.
Parametric mapping of the distribution volume (Vd) obtained in pig confirmed highest Vd values in thalamus, frontal cortex, and striatum. Pretreatment with desipramine reduced the Vd generally by roughly 40%; see Fig. 4 and Table 2. PET imaging with [ 11C]MeNER revealed the expected pattern with highest Vd in thalamic regions and lowest binding in occipital cortex. Binding is reduced around 25% by desipramine treatment, see Fig. 5 and Table 2. Based on an estimate from the time activity curves, equilibrium conditions were first attained within the end of the 90 minutes scan period, cf. Fig. 6. Upon desipramine block we found an initial higher brain up-take of [ 11C]MeNER, reflecting somewhat more available radiotracer due to blocking of peripheral binding sites. 3.4. Metabolite analysis [ 11C]NS8880 showed a rapid metabolism both in rat and in pig. In pig less than 20% of parent radiotracer was detected in plasma 20 minutes after tracer injection and only 5% after 30 minutes. In rat, 35% of unchanged [ 11C]NS8880 was observed after 20 minutes. 4. Discussion The radiolabeling method used in the present study successfully produced [ 11C]NS8880 in low but still useful yields for in vivo evaluation. The straightforward direct methylation of O-desmethyl NS8880 using [ 11C]methyl iodide or [ 11C]methyl triflate could not be applied as this hydroxy starting material is not stable. We therefore Table 2 Vd values of [11C]NS8880 and [11C]MeNER obtained in pig brain. Region
Striatum Thalamus Frontal cortex Occipital cortex Cerebellum
[11C]NS8880
[11C]MeNER
Baseline
Desipramine block
Baseline
Desipramine block
3.6 4.1 3.4 3.2 3.3
2.2 2.5 1.9 1.9 1.8
3.1 4.2 3.1 3.1 3.4
2.7 3.2 2.6 2.5 2.6
± ± ± ± ±
0.3 0.3 0.5 0.5 0.4
± ± ± ± ±
0.2 0.2 0.3 0.3 0.4
± ± ± ± ±
0.3 0.3 0.5 0.5 0.4
± ± ± ± ±
0.2 0.1 0.3 0.4 0.3
considered the aromatic substitution of a halogen by [ 11C]methanolate prepared in situ as labeling approach. Pyridines with halogens in 2-position are activated aromatic systems and the halogens are comparable easy to substitute within aromatic nucleophilic substitution reactions. However, these reactions normally require excessive amounts of nucleophile, high temperatures and long reaction times, which explain our use of rather harsh reaction conditions. In the first experiments a 2-chloro analog was applied, but no product formation was observed (data not shown), probably due to too slow reaction kinetics. We therefore decided to use the more reactive 2-fluoro derivative as starting material (see Scheme 1). To our knowledge we could hereby for the first time demonstrate the application of a 2-pyridinyl fluorine substrate and [ 11C]methanolate in PET radiochemistry. The amount of isolated [ 11C]NS8880 was comparable low, which probably is due to the combination of two parameters; firstly the loss of gaseous [ 11C]methyl iodide from the reaction mixture owing to the harsh reaction conditions and secondly a slow reaction kinetics. We decided to generate the [ 11C]methanolate starting from [ 11C]methyl iodide and sodium hydroxide in situ. It might be argued that the isolated radiochemical yields could be improved simply by first generating [ 11C]methanolate under milder reaction conditions and applying the starting material and the harsh reaction conditions afterwards. However, we did not prosecute this approach as our C-11 chemistry set-up would have required a manual addition of the precursor, which is in terms of radiation protection unfavorable. The assumption that higher radiochemical yields could be obtained upon usage of pre-prepared [ 11C]methanolate is supported by preparative radio-HPLC which revealed up to 30% of the radiolabeled Bocprotected intermediate in the reaction mixture (Fig. 2), whereas the radiochemical yield of [ 11C]NS8880 based on the average [ 11C]methyl iodide production is only around 5%. It might still be argued, that the actual labeling route proceeds via substitution of fluoride by hydroxide, followed by [ 11C]methylation of this phenolic compound. However, these 2-hydroxy pyridins are not stable compounds and would have resulted in a decomposition of the intermediate material. This decomposition was not observed.
K.H. Vase et al. / Nuclear Medicine and Biology 41 (2014) 758–764
763
Fig. 5. Maps of distribution volume, Vd (ml g-1) of [11C]MeNER in brain of living pig (A) at baseline conditions and (B) after treatment with desipramine (1 mg/kg). From left to right: coronal, sagittal, and transverse images. The distribution volumes were calculated by the method of Zhou relative to the arterial input function.
A
70000
thalamus hypothalamus frontal cortex occipital cortex striatum cerebellum
60000 30000 20000 10000 0 0
20
40
60
Time after injection (min)
80
[ 11C]NS8880 is more sensitive towards desipramine block compared to [ 11C]MeNER, around 40% vs 25% in all regions. This might reflect a somewhat lower degree of non-specific binding of [ 11C]NS8880. It is a general challenge for NET imaging to identify a suitable reference region because of the widespread and relatively uniform distribution of NET throughout the brain and several regions have been evaluated as potential reference regions [22,23]. In the present study this also constitute a problem and so far a suitable reference region could not be identified due to considerable uptake in all regions. Like other groups we found a lower uptake in occipital cortex compared to cerebellum [23,24]. A rough estimate of the binding potential is around 0.3 obtained by the ratio between Vd in thalamus and occipital cortex minus 1. Binding potential might be higher, if a more suitable reference region could be identified. Based on the limited data, Vd in thalamic regions is slightly higher for [ 11C]MeNER compared to [ 11C]NS8880, resulting in a slightly higher estimate for the binding potential (0.4 vs 0.3). It should be noted, that the effective binding potential for [ 11C]MeNER in humans is likewise only around 0.4-0.5 [11,25]. A disadvantage using [ 11C]MeNER in human studies is
Radioactivity concentration (Bq/cc)
Radioactivity concentration (Bq/cc)
In pig and rat [ 11C]NS8880 quickly metabolized with only around 20-30% parent compound after 20 minutes. Interestingly we found a slightly faster metabolism in pig compared to rat. This is in contrast to our normal observation with other radiotracers with faster metabolism in small animals. [ 11C]NS8880 enters rapidly the brain and equilibrium is reached remarkably fast, both in pig and rat. Highest binding is observed in regions with high NET concentration, i.e. thalamic and hypothalamic regions. The maps of Vd reveal a binding pattern comparable to those of desipramine as demonstrated by the blocking experiment using desipramine. Considerable uptake was observed in striatum and the striatal binding seems to be displaceable by desipramine. In another experiment in rat, uptake of [ 11C]NS8880 in all regions was not affected by blocking with citalopram, thus the striatal uptake is likely not due to the binding to serotonin transporters (data not shown). In pig brain the binding pattern of [ 11C]NS8880 is close to that of 11 [ C]MeNER (see Figs. 4 and 5) with highest uptake in thalamic and hypothalamic regions. However, we found a considerable higher Vd for [ 11C]NS8880 in striatum compared to [ 11C]MeNER. The binding of
B
70000
thalamus hypothalamus frontal cortex occipital cortex striatum cerebellum
60000 30000 20000 10000 0 0
20
40
60
80
Time after injection (min)
Fig. 6. Time-activity curves for selective ROIs in pig brain after administration of [11C]MeNER (A) at baseline conditions and (B) after pre-treatment with desipramine (1 mg/kg).
764
K.H. Vase et al. / Nuclear Medicine and Biology 41 (2014) 758–764
the slow kinetics of this compound; peak equilibrium binding is not reached during a 90-minute PET procedure [11] and as a consequence data acquisition times of 120 minutes have been applied in recent studies [25]. In pig, we found similarly a slow kinetics for [ 11C]MeNER. [ 11C]NS8880 shows here at least in pig and rat a considerable faster kinetics, which is advantageous. Of course it is unknown if this fast kinetics would be observed in humans as well. 5. Conclusion Our initial data indicate that [ 11C]NS8880 is usable for imaging NET and due to its faster kinetics might in some respect be superior to [ 11C]MeNER, even so the binding potential is still low. Acknowledgement We acknowledge the technical assistance of Mette Simonsen, Kim Vang, Jørgen Bach Pedersen, Gitte Friberg, Kirsten Braad Iskov, and Tove Thomsen. Support to DanPET AB from CCJobs, Interreg, Sweden-Denmark, and COGNITO, NRU, Copenhagen is gratefully acknowledged. References [1] Blakely RD, De Felice LJ, Hartzell HC. Molecular physiology of norepinephrine and serotonin transporters. J Exp Biol 1994;196:263–81. [2] Biederman J, Spencer T. Attention-deficit/hyperactivity disorder (ADHD) as a norepinephrine disorder. Biol Psychiatry 1999;46:1234–42. [3] Zhou J. Norepinephrine transporter inhibitors and their therapeutic potential. Drugs Future 2004;29:1235–44. [4] Gulyás B, Brockschnieder D, Nag S, Pavlova E, Kása P, Beliczai Z, et al. The norepinephrine transporter (NET) radioligand (S, S)-[18 F]FMeNER-D2 shows significant decreases in NET density in the human brain in Alzheimer's disease: A post-mortem autoradiographic study. Neurochem Int 2010;56:789–98. [5] Ding Y-S, Fowler J. New-generation radiotracers for nAChR and NET. Nucl Med Biol 2005;32:707–18. [6] Ding Y-S, Lin K-S, Logan J. PET imaging of norepinephrine transporters. Curr Pharm Des 2006;12:3831–45. [7] Schou M, Halldin C, Sovago J, Pike VW, Gulyas B, Mozley PD, et al. Specific in vivo binding to the norepinephrine transporter demonstrated with the PET radioligand, (S, S)-[11C]MeNER. Nucl Med Biol 2003;30:707–14. [8] Wilson AA, Johnson DP, Mozley D, Hussey D, Ginovart N, Nobrega J, et al. Synthesis and in vivo evaluation of novel radiotracers for the in vivo imaging of the norepinephrine transporter. Nucl Med Biol 2003;30:85–92.
[9] Ding YS, Lin KS, Garza V, Carter P, Alexoff D, Logan J, et al. Evaluation of a new norepinephrine transporter PET ligand in baboons, both in brain and peripheral organs. Synapse 2003;50:345–52. [10] Schou M, Halldin C, Sóvágó J, Pike VW, Hall H, Gulyás B, et al. PET evaluation of novel radiofluorinated reboxetine analogs as norepinephrine transporter probes in the monkey brain. Synapse 2004;53:57–67. [11] Logan J, Wang G-J, Telang F, Fowler JS, Alexoff D, Zabroski J, et al. Imaging the norepinephrine transporter in humans with (S, S)-[11C]O-methyl reboxetine and PET: problems and progress. Nucl Med Biol 2007;34:667–79. [12] Takano A, Varrone A, Gulyás B, Karlsson P, Tauscher J, Halldin C. Mapping of the norepinephrine transporter in the human brain using PET with (S, S)-[18 F] FMeNER-D2. NeuroImage 2008;42:474–82. [13] Klimek V, Stockmeier C, Overholser J, Meltzer HY, Kalka S, Dilley G, et al. Reduced levels of norepinephrine transporters in the locus coeruleus in major depression. J Neurosci 1997;17:8451–8. [14] Gross-Isseroff R, Israeli M, Biegon A. Autoradiographic analysis of [3H] desmethylimipramine binding in the human brain postmortem. Brain Res 1988;456:120–6. [15] Schou M, Halldin C, Pike VW, Mozley PD, Dobson D, Innis RB, et al. Post-mortem human brain autoradiography of the norepinephrine transporter using (S, S)[18 F]FMeNER-D2. Eur Neuropsychopharmacol 2005;15:517–20. [16] Peters D, Eriksen BL, Nielsen EØ, Scheel-Krüger J, Olsen GM. Novel 8-aza-bicyclo[3. 2.1]octane derivatives and their use as monoamine neurotransmitter re-uptake inhibitors. WO 2004/113334 A1. [17] Low WC, Whitehorm D, Hendley ED. Genetically related rats with differences in hippocampal uptake of norepinephrine and maze performance. Brain Res Bull 1984;12:703–9. [18] Bolden-Watson C, Richelson E. Blockade by newly-developed antidepressants of biogenic amine uptake into rat brain synaptosomes. Life Sci 1993;52:1023–9. [19] Watanabe H, Andersen F, Simonsen CZ, Evans SM, Gjedde A, Cumming P. MR-based statistical atlas of the Göttingen Minipig brain. NeuroImage 2001;14:1089–96. [20] Zhou Y, Huang S-C, Bergsneider M, Wong DF. Improved parametric image generation using spatial-temporal analysis of dynamic PET studies. NeuroImage 2002;15:697–707. [21] Paxinos G, Watson C. The rat brain in stereotaxic coordinates. 3rd ed. San Diego: Academic press; 1997. [22] Logan J, Ding Y-S, Lin K-S, Pareto D, Fowler J, Biegon A. Modeling and analysis of PET studies with norepinephrine transporter ligands: the search for a reference region. Nucl Med Biol 2005;32:531–42. [23] Gallezot J-D, Weinzimmer D, Nabulsi N, Lin S-F, Fowles K, Sandiego C, et al. Evaluation of [ 11C]MRB for assessment of occupancy of norepinephrine transporter: Studies with atomoxetine in non-human primates. NeuroImage 2011;56:268–79. [24] Hannestad J, Gallezot J-D, Planeta-Wilson B, Lin S-F, Williams WA, van Dyck CH, et al. Clinically relevant doses of methylphenidate significantly occupy norepinephrine transporters in human in vivo. Biol Psychiatry 2010;68:854–60. [25] Ding Y-S, Singhal T, Planeta-Wilson B, Gallezot J-D, Nabulsi N, Labaree D, et al. PET imaging of the effects of age and cocaine on the norepinephrine transporter in the human brain using (S, S)-[11C]O-methylreboxetine and HRRT. Synapse 2010;64:30–8.