Research Article Received 25 February 2013,
Revised 14 April 2013,
Accepted 16 April 2013
Published online 5 June 2013 in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/jlcr.3055
19
F/18F exchange synthesis for a novel [18F] S1P3-radiopharmaceutical Johanna Rokka,a Cesare Federico,b Jori Jurttila,a Anniina Snellman,c Merja Haaparanta,c Juha O. Rinne,d and Olof Solina,e* 19 18 F/ F isotope exchange is a useful method to label drug molecules containing 19F-fluorine with 18F without modifying the drug molecule itself. Sphingosine-1-phosphate (S1P) is an important cellular mediator that functions by signaling through cell surface receptors. S1P is involved in several cell responses and may be related to many central nervous system disorders, including neural malfunction in Alzheimer’s disease. In this study, [ 18 F]1-benzyl-N-(3,4-difluorobenzyl)-2-isopropyl-6(2-methoxyethoxy)-1H-indole-3-carboxamide, a novel 18F-labeled positron emission tomography tracer for the S1P3 receptor, was successfully synthesized using the 19F/18F isotope exchange reaction. Parameters of the reaction kinetics were studied, and correlations between the initial 18F-activity, the amount of precursor, radiochemical yield and specific activity (SA) were determined. Contrary to expectations, high initial 18F-activity decreased the radiochemical yield, and only a minor increase of SA occurred. This is most probably due to the complexity of the molecule and the subsequent susceptibility to radiolytic bond disruption. On the basis of the present results, a convenient condition for the 19F/18F exchange reaction is the use of 2 mmol precursor with 20 GBq of 18F-activity. This afforded a radiochemical yield of ~10% with an SA of 0.3 GBq/mmol. Results from this study are of interest for new tracer development where high initial 18F-activity and 19F/18F isotope exchange is used.
Keywords: PET chemistry; sphingosine-1-phosphate; S1P3 receptor; radiolabelling; nucleophilic
Introduction 19
J. Label Compd. Radiopharm 2013, 56 385–391
F-fluorination
(AD).9,10,12,13 Several synthetic compounds have been developed to study the S1PRs, with the best known being FTY720-P, a nonspecific agonist for S1PRs.14 Also, more selective S1P receptor subtype agonists and antagonists have been developed, but the receptor specificity of these compounds have been poor, or application of these compounds has not been reported.9,14–17 A family of 6-substituted indole-3-carboxyclic acid amide compounds is described having S1P receptor agonist or antagonist biological behavior.18 Thus, knowledge of S1PRs’ involvement in health and disease remains limited. Suitable PET tracers will aid in the investigation of the expression and activation of S1PRs and may elucidate the function of these receptors in AD and other CNS disorders.
a Radiopharmaceutical Chemistry Laboratory, Turku PET Centre, University of Turku, Porthaninkatu 3, FI-20500 Turku, Finland b
Siena Biotech S.p.A., Strada del Petriccio e Belriguardo 35, 53100 Siena, Italy
c Preclinical Imaging, Turku PET Centre, University of Turku, Tykistökatu 6, FI-20520 Turku, Finland d Turku PET Centre c/o Turku University Hospital, University of Turku, P.O. Box 52, FI-20521 Turku, Finland e Accelerator Laboratory, Turku PET Centre, Åbo Akademi University, Porthansgatan 3, FI-20500 Turku, Finland
*Correspondence to: Olof Solin, Radiopharmaceutical Chemistry Laboratory, Turku PET Centre, University of Turku, Porthaninkatu 3, FI-20500 Turku, Finland. E-mail: olof.solin@abo.fi
Copyright © 2013 John Wiley & Sons, Ltd.
385
F/18F isotope exchange is a useful method to produce 18Flabeling agents and fluoroaromatic compounds for positron emission tomography (PET). The 19F/18F isotope exchange reaction allows the labeling of drug molecules that contain fluorine in their molecular structure. Several 19F/18F isotope exchange studies examined the effects of the molecular structure of the precursor and have investigated the reaction kinetic parameters to optimize radiochemical yield and specific activity.1–8 Most of these studies have been performed using relatively small amounts of initial activity of 18F-fluorine, and precursor compounds have mostly been relatively simple small molecules.1,2,6,8 Sphingosine-1-phosphate (S1P) is an important cellular mediator. In plasma, S1P is mainly bound to high-density lipoprotein and albumin in nanomolar concentrations and widely distributed throughout the body.9,10 S1P is involved in several cell responses, including cell signaling, neuronal proliferation and cell survival.9–12 Because of the multiple tasks of the S1P, the S1P receptors (S1PRs) are responsible for differentiation of the S1P signaling. S1P modulates through five different receptor subtypes (S1P1–S1P5) belonging to a family of G-protein-coupled receptors.9–13 All S1P receptor subtypes are expressed in the central nervous system (CNS).9–13 The functions of S1P in the CNS are under intense study, and the exact tasks of S1PRs are not fully understood. The expression of S1PRs is possibly related to CNS disorders, and these receptors may be potential drug targets in CNS disorders such as multiple sclerosis and Alzheimer’s disease
18
J. Rokka et al. whereas at a lower temperature (170 C), the reaction mixture did not reach the boiling point of the solution. In entry 7, reaction conditions were held constant, and the end product was purified.
The ADIT project (Design of small molecule therapeutics for the treatment of Alzheimer disease based on the discovery of innovative drug targets, a research project within the European Union’s Sixth Framework Program) was aimed at the development of novel chemical entities endowed with specific neuroprotective activity in AD, up to the preclinical stage.19 During this project, S1P3 was chosen as one of the targets. Preliminary studies within ADIT suggested a possible involvement of S1P3 in regulating neuroinflammation occurring in AD. Our goal in the present study was to synthesize [18F]1-benzylN-(3,4-difluorobenzyl)-2-isopropyl-6-(2-methoxyethoxy)-1H-indole3-carboxamide ([18F]1), a novel 18F-labeled PET tracer for the S1P3 receptor. Radiolabeling was achieved through 19F/18F isotope exchange. During the synthesis development, we also examined the effect of the reaction kinetic parameters on the radiochemical yield and the specific activity of [18F]1.
General procedure for
Chemicals Organic solvents were HPLC grade from Ratherburn Chemicals Ltd (Walkernburn, UK). Water was purified by a Milli-Q Plus Ultra Pure Water system (Millipore, Molsheim, France). Other chemicals were analytical grade from Merck (Darmstadt, Germany). The precursor compound, 1-benzyl-N-(3,4-difluorobenzyl)-2-isopropyl-6-(2-methoxyethoxy)-1H-indole3-carboxamide, was synthesized by AlChemy Fine Chemicals & Research s.r.l. (Baricella, Italy) as HPLC grade.
Purification of the reaction mixture [18F]1-Benzyl-N-(3,4-difluorobenzyl)-2-isopropyl-6-(2-methoxyethoxy)-1Hindole-3-carboxamide ([18F]1) was purified from the reaction mixture using a semi-preparative HPLC method. A Merck Hitachi LaChrom 7000 pump (Merck Hitachi, Darmstadt, Germany) was employed together with a Phenomenex Luna 5 m C18(2) 100 Å (250 10.0 mm) column and gradient system of methanol (A) and water (B). The gradient profile was 0–2 min, 30% A; 2.1 min, 70% A; 10–25 min, 82% A; with a flow rate of 5.0 ml/min. Radioactivity was detected with a NaI(Tl) scintillation detector (200 200 , Bicron Corporation, Newbury, OH, USA) placed on the outflow of the column. The HPLC eluent containing the end product fraction was subsequently removed using solid phase extraction (Sep-PakW light tC18 cartridge, Waters). In this procedure, the fraction isolated by HPLC was first diluted to 20 ml using an ethanol–water solution (5:95 v/v). This solution was applied to the Sep-Pak cartridge. The cartridge was then washed with 20 ml of the ethanol–water solution (5:95 v/v). Finally, the end product, [18F]1, was eluted from the cartridge using 0.5 ml ethanol.
Synthesis F-Fluoride production
18
F-Fluoride was produced using the 18O(p,n)18F nuclear reaction. 18OEnriched water (enrichment grade 98%, 800 ml, Hyox, Rotem Industries Ltd., Israel) was irradiated with 17 MeV protons and a 10 mA beam current for 5–90 min using the MGC-20 cyclotron (Efremov Scientific Research Institute for Electrophysical Apparatuses (NIIEFA), St. Petersburg, Russia) at the Åbo Akademi Accelerator Laboratory. The initial radioactivity of 18 F-fluoride was 3.5–33 GBq.
Reaction conditions for
F/18F-fluorine isotope exchange
The aqueous solution of 18F-fluoride from the cyclotron target was collected into a 2-ml Wheaton V-vial (Wheaton Science Products, Millville, USA) containing 2 mmol potassium bicarbonate and 4 mmol Kryptofix 222. Water was removed using an azeotropic distillation procedure, where acetonitrile (1 ml) was added to the solution and the liquid was evaporated under a stream of helium. The solution was heated at 100 C for 3 min. The distillation procedure was repeated three times. The precursor 1, 1-benzyl-N-(3,4-difluorobenzyl)-2-isopropyl-6-(2methoxyethoxy)-1H-indole-3-carboxamide, was dissolved in 300 ml dimethyl sulfoxide (DMSO). The precursor–DMSO solution was added to the dry [18F]-fluoride–Kryptofix complex. The reaction mixture was then heated at 170 or 190 C (Figure 1). For reaction kinetic studies, aliquots of the reaction mixture were withdrawn at intervals for up to 60 min and then analyzed. Starting with a high initial activity of 18F (Table 1. entry 7), 16 syntheses were performed to produce [18F]1. In these syntheses, reaction mixtures were heated for 20 min, and the product [18F]1 was purified. No samples were taken during these syntheses.
Experimental
18
19
19
F/18F isotope exchange
Analytical procedures
Reaction conditions are listed in Table 1. To define the optimum reaction conditions (Table 1, entries 1–6), we varied the amounts of initial 18Factivity, the unlabeled fluorinated precursor, potassium bicarbonate, and the reaction temperature. At 190 C, the reaction mixture refluxed,
Table 1. Reaction conditions and starting materials for
The radiochemical yield and the specific activity were determined from reaction mixture samples removed at specified intervals and from the end product of the production syntheses. Radiochemical yield was
19
F/18F isotope exchange synthesis of [18F]1
Entry
A0 [GBq]
Precursor [mmol]
KHCO3 [mmol]
K222 [mmol]
Temperature [ C]
1 2 3 4 5 6 7
6.9 0.5 0 0 6.3 0.4 6.7 0.3 3.6–21.6 30.4 2.2
3.9 0.5 2.3 2.0 0.5–8.2 0.5–7.8 2.5 0.3 2.4 0.4
1.0–4.0 2.0 2.0 2.01 0.01 2.00 0.01 2.01 0.01 2.02 0.02
1.9–8.5 5.0 5.0 4.6 0.5 4.5 0.4 4.3 0.3 5.7 1.2
190 170 190 170 190 190 190
386
In each synthesis, target water was removed with acetonitrile and azeotropic distillation, and precursor was dissolved in 300 ml DMSO. A0 denotes initial 18F-activity used in syntheses. Range of investigated parameter is included in the table. Constant amounts are expressed as mean SD. In entry 1, the amount of base is varied; in entries 2 and 3, the speed of chemical decomposition of 1 are studied without adding radioactivity; in entries 4 and 5, the amount of precursor is varied at two different temperatures; in entry 6, the amount of radioactivity is varied using a fixed amount of precursor; and finally, entry 7 contains the parameters used for a set of syntheses (n = 16).
www.jlcr.org
Copyright © 2013 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2013, 56 385–391
J. Rokka et al.
Figure 1. Scheme for
19
F/18F isotope exchange reaction for [18F]1.
determined with a thin layer chromatography (TLC) method, and digital autoradiography. TLC plates were glass plates pre-coated with HPTLC Silica gel 60 RP-18 F254s (art. no. 113724) from Merck (Darmstadt, Germany). TLC plates were developed in a twin-trough chamber (10 10 cm, Camag, Muttenz, Switzerland) using a mixture of 20 mM ammonium acetate (pH 4.3)–methanol (30:70 v/v) as a solvent. Radioactivity on the plate was detected and quantified using photostimulated luminescence digital autoradiography. A TLC plate was placed into an exposure cassette with an imaging plate (Fuji Imaging Plate BAS-MS2025, Fuji Photo Film Co., Ltd., Tokyo, Japan). After exposure, the imaging plate was scanned using the Fuji Analyzer BAS 1800 reader (Fujifilm Co., Ltd., Tokyo, Japan) at 200 mm resolution. Data were analyzed using the Tina 2.1 program (Raytest Isotopenmessgeräte GmbH, Straubenhardt, Germany). The autoradiographic image and its analysis are shown in Figure 2. Radiochemical yield is presented as a percentage of total radioactivity in the sample. Specific activity was measured using an analytical HPLC method. For HPLC analysis, a Merck Hitachi LaChrom 7000 system with Hitachi D-7000 HPLC System Manager software (version 3.0) was used. The UV
PSL/mm2
200
[18F]1
detector of the LaChrom HPLC system was set at 254 nm. Radioactivity was detected with similar NaI(TI) scintillation detector, as described earlier, connected on the outflow of the column. Quality control analysis was carried out using a Phenomenex Gemini NX 3 m C18 110 Å (150 4.60 mm) column and an isocratic system of 20 mM ammonium acetate (pH 4.3)–methanol (30:70 v/v). In the analysis, the concentration of the product was determined. Also, the product peak identified by the radioactive detector was collected, and the radioactivity of this HPLC fraction was measured for specific activity determination. Chromatograms from the reaction mixture samples and end products are shown in Figure 3. The measured specific activity was decay corrected to the end of bombardment. The stability of the precursor was evaluated in studies of reaction solutions to which no radioactivity was added (Table 1, entries 2 and 3). The chemical decomposition of 1 was followed by analyzing the reaction mixture samples with the same analytical HPLC method used for specific activity measurement. The recovery percent of 1 was calculated from these HPLC results. The specific activity and the radiochemical yield were determined using different methods. The TLC method combined with digital autoradiography was used to determine the radiochemical yield. Digital autoradiography was chosen because it has a wide linear range and high sensitivity.20 Specific activity was determined with an analytical HPLC method that enables concentration measurement of 1. The analytical HPLC method was not as sensitive as the TLC method for radioactivity determination (Figure 6B and D); thus, the specific activity was not measurable at all time points.
Statistics
100
Mean values were calculated from the individual measurements and expressed at a precision of one standard deviation (mean SD).
Results 0 0
0.5
1
Rf
J. Label Compd. Radiopharm 2013, 56 385–391
Copyright © 2013 John Wiley & Sons, Ltd.
www.jlcr.org
387
Figure 2. Autoradiographic image of TLC plate showing crude reaction product during reaction. Rf value of [18F]1 was 0.35. The sample was applied on HPTLC Silica gel 60 RP-18 F254s plate, which was developed with a mixture of 20 mM ammonium acetate (pH 4.3)–methanol (30:70 v/v). The radioactivity distribution was detected and analyzed using photostimulated luminescence autoradiography. Radiochemical yield was determined using this method.
Several series of syntheses were made to study the isotopic exchange reaction of 1 (Figure 1, Table 1). The radiochemical yield and specific activity were measured using two different methods (Figures 2 and 3). The optimization of reaction conditions was started with selection of a suitable base for promotion of 19F/18F exchange. Preliminary studies established that 2 mmol potassium bicarbonate (Table 1, entry 1) gave satisfactory results in the 19F/18F isotope exchange reaction
J. Rokka et al. 100
A [ F]1
170 °C 190 °C
0.08
Radioactivity UV absorption
2.5
0.04
Amount of 1 [%]
18
UV absorption [mAFU]
Radioactivity [arb. Units]
5
75
50 2
R = 0.885
25
2
R = 0.955
0 0
0
0 0
5
10
15
20
0.6
Radioactivity UV absorption
0.3
6
UV absorption [mAFU]
Radioactivity [arb. Units]
18
0
0 0
5
10
15
20
Time [min] Figure 3. Radio-HPLC of [18F]1: (A) UV absorption and radioactivity chromatograms of reaction mixture and (B) separated end product. Phenomenex Gemini NX 3 m C18 110 Å column and an isocratic system of 20 mM ammonium acetate (pH 4.3)–methanol (30:70 v/v) was used. Specific activity was measured using this method.
(Figure 4). Further experimentation was thus carried out under these conditions. Exposure of the precursor 1 to heat at both 170 and 190 C caused chemical degradation over time (Table 1, entries 2 and 3; Figure 5). High temperature (Table 1, entries 4 and 5) increased both radiochemical yield (Figure 6A and B) and specific activity (Figure 6C and D). This holds for all experiments except for the yield at the highest concentration of the precursor. Using 8 mmol
Radiochemical yield [%]
20 Amount of KHCO3 1.0 µmol 2.0 µmol 4.0 µmol
10
0 0
20
40
60
Time [min]
388
Figure 4. The effects of exchange reaction time and potassium bicarbonate concentration on radiochemical yield of [18F]1. Whole lines show radiochemical yield at various potassium bicarbonate concentration used in reactions. All other conditions in these reactions were unchanged (Table 1, entry 1).
www.jlcr.org
45
60
Figure 5. Precursor 1 chemical decomposition as a function of time at 170 and 190 C. Precursor was handled in a similar manner as in reactions with radioactivity. However, no radioactivity was added to the reaction vessel (Table 1, entries 2 and 3). Exponential curves were fitted to measured data points.
B [ F]1
30
Time [min]
Time [min]
12
15
Copyright © 2013 John Wiley & Sons, Ltd.
at 170 C resulted in a yield of 50% versus 30% at 190 C. The specific activities were similar in these two experiments. We examined the impact of the amount of precursor supplied in the reaction (Table 1, entries 4 and 5; Figure 6). With higher amounts of precursor, better radiochemical yields were gained, but at the same time, the specific activity of the product was low. Figure 7 correlates the specific activity with the amount of precursor at 190 C after 15 min of reaction. The effects of varied amounts of initial 18F-activities (Table 1, entry 6) on the measured radiochemical yield and specific activity are shown in Figure 8. The reaction containing the highest amount of added radioactivity (21.6 GBq) showed the lowest level of radiochemical yield. At the same time, the specific activity was the highest. In syntheses where product [18F]1 was isolated (Table 1, entry 7), the reaction conditions were constant, and the end product was purified. Total synthesis time was 75 min, and the radiochemical purity exceeded 98% in all syntheses. The average radiochemical yield of these syntheses was 1.0 0.5%, and the specific activity was 0.36 0.19 GBq/mmol (values are decay corrected at the end of bombardment, n = 16). The radiochemical yield and specific activity (Table 1, entry 6) were plotted as a function of initial 18F-activity at the time point of 20 min (Figure 9). As the initial concentration of 18Fradioactivity increased, the radiochemical yield of [18F]1 showed a generally decreasing trend, whereas the specific activity slightly increased as the initial radioactivity increased. Also the mean results from the syntheses in which the product was isolated (n = 16, Table 1, entry 7) are included in Figure 9.
Discussion Previously, direct 19F/18F isotope exchange reactions have been applied to 18F-fluorination of small aromatic molecules. 1,2,5,6 18 F-Labeling of larger molecules with 19F/18F isotope exchange reaction has been reported to be more difficult,4,6 and commonly, these molecules have been produced with at least a two-step synthesis where protecting groups were needed or by coupling a 18F-labeled small molecule with a larger molecule.4,7 In this study, [18F]1 was synthesized using one-step synthesis and 19F/18F isotope exchange reaction (Figure 1). The effects of initial 18F-activity, temperature and amount of precursor on the radiochemical yield and the specific activity
J. Label Compd. Radiopharm 2013, 56 385–391
J. Rokka et al.
A
B
Radiochemical yield [%]
170 °C Amount of precursor
45
0.45 µmol 1.06 µmol 2.11 µmol 8.22 µmol
30
15
Radiochemical yield [%]
30
60
0
190 °C Amount of precursor 0.49 µmol 0.97 µmol 2.13 µmol 7.76 µmol
15
0 0
15
30
45
60
0
15
45
60
D 170 °C
0.3
Amount of precursor 0.45 µmol 1.06 µmol 2.11 µmol 8.22 µmol
0.15
0 0
15
30
45
Specific activity [GBq/µmol]
C Specific activity [GBq/µmol]
30
Time [min]
Time [min]
0.9
190 °C Amount of precursor 0.49 µmol 0.97 µmol 2.13 µmol 7.76 µmol
0.6
0.3
0 0
60
Time [min]
15
30
45
60
Time [min]
Specific activity [GBq/µmol]
Figure 6. (A, B) Radiochemical yield and (C, D) specific activity of [18F]1 as a function of exchange reaction time at 170 and 190 C, respectively. The amount of precursor (different lines in the graphs) and temperature (vertical columns) were varied. Otherwise, reaction conditions were constant (Table 1, entries 4 and 5). The dashed line in panel D denotes the time point used for presenting specific activity as a function of precursor amount in Figure 7.
0.9
0.6
0.3 R2 = 0.857
0 0
4
8
Precursor amount [ mol] Figure 7. Specific activity of [18F]1 as a function of precursor amount at 15-min reaction time and 190 C (see also Figure 6D). The solid line is a fit to the data points.
J. Label Compd. Radiopharm 2013, 56 385–391
Copyright © 2013 John Wiley & Sons, Ltd.
www.jlcr.org
389
for the 19F/18F isotope exchange reaction were studied. [18F]1 has a notably more complex structure than compounds that have been reported before for labeling with direct 19F/18F isotope exchange. Thus, it is not surprising that the yield of [18F]1 is low, presumably because of the complex structure of 1, which is vulnerable to chemical and radiolytic degradation. Preliminary studies established that DMSO is a proper reaction solvent, confirming the results of previous investigations of isotope exchange reactions, especially when high temperature is needed.1,3,6 The amount of the base, KHCO3, affects the radiochemical yield of the reaction (Figure 4). With low amounts of KHCO3, only a minor reaction is observed. With larger amounts of KHCO3, the radiochemical yield increases. To avoid base-catalyzed chemical decomposition of the precursor and [18F]1, 2 mmol of base was used in the subsequent studies. Other bases such as K2CO3 and tetrabutylammonium hydroxide were initially tested, but in these syntheses, the radiochemical yield and the specific activity were not improved (data not shown).
The 19F/18F isotope exchange reaction was performed using relatively harsh conditions. The large substituent attached at the para position of the 1,2-difluorobenzene ring of 1 reduces reactivity of the benzene ring and makes the isotope exchange reaction difficult, an outcome that has been demonstrated earlier with similar compounds.4,6 Using the present reaction conditions, we observed that the chemical decomposition of the precursor and also [18F]1 is rapid (Figure 5). The chemical decomposition of the compound will eventually limit the exchange reaction because the entire amount of precursor is consumed by decomposition. The chemical degradation of 1 is also seen in the radiochemical yield curves of the kinetic reaction studies (Figure 6) where the radiochemical yield drops off at the later time points, especially when lower amounts of precursor are used. This chemical decomposition will, in turn, under otherwise set conditions, limit the extent of radiochemical yield and specific activity. For example, in Figure 6D, after 20 min of reaction time, the amount of 1 is below the detection limit, and the specific activity was thus not measurable. Reaction temperature has an effect on radiochemical yield and specific activity. With a refluxing reaction mixture (190 C), the isotope exchange rate is faster leading to higher radiochemical yield and specific radioactivity than the syntheses carried out under milder conditions (170 C), where the temperature is below the boiling point of the reaction mixture (Figure 6). A similar effect of temperature has also been shown in earlier studies,4–7 although in these studies, less drastic temperatures were used. One exception to this was the synthesis in which we used a very large amount (8 mmol) of precursor at lower temperature. The chemical degradation of 1 is less rapid at 170 C than at 190 C (Figure 5) thus increasing notably the radiochemical yield of this synthesis. On the other hand, at these conditions, the specific activity was low.
J. Rokka et al.
A Initial18 F-activity 3.6 GBq 6.5 GBq 12.3 GBq 21.6 GBq
Radiochemical yield [%]
20
10
0 0
15
30
45
60
Time [min]
Specific activity [GBq/µmol]
B Initial18 F-activity 3.6 GBq 6.5 GBq 12.3 GBq 21.6 GBq
0.4
0.2
0 0
15
30
45
60
Time [min]
Radiochemical yield Specific activity
Radiochemical yield [%]
20
0.60
0.40 10 0.20
0
Specific activity [GBq/µmol]
Figure 8. (A) Radiochemical yield of [18F]1 at 190 C as a function of exchange reaction time. The initial 18F-activity was varied from 3 to 22 GBq. In these studies, the amount of precursor 1 was 2 mmol (Table 1, entry 6). The dashed line denotes the time point used for presenting data in Figure 9. (B) Specific activity of [18F]1 at 190 C as a function of exchange reaction time. The initial 18F-activity was varied from 3 to 22 GBq. The amount of precursor 1 was 2 mmol (Table 1, entry 6). The dashed line denotes the time point used for presenting data in Figure 9.
0.00 0
10
20
18F-radioactivity
30
[GBq]
Figure 9. Radiochemical yield and specific activity at 190 C as a function of initial 18 F-activity at the time point 20-min reaction time. Values at 30 GBq are average values (n = 16; Table 1, entry 7) from the production syntheses of [18F]1.
390
Previous studies have shown that radiochemical yield is proportional and specific activity inversely proportional to the amount of precursor supplied in the reaction.1,2,7 In the present investigation, we observed the same effect (Figures 6 and 7). The correlation between precursor amount and specific activity is not so significant as in previous studies,1,2,7 but on the other
www.jlcr.org
Copyright © 2013 John Wiley & Sons, Ltd.
hand, these studies were performed with compounds of less complex structure than [18F]1. According to expectations based on previously published studies, absolute radiochemical yield and specific activity increase when initial 18F-activity is raised.2,7,8 We did 16 syntheses with high initial 18F-activity (30 GBq), purified the product and found, however, that the end product radiochemical yield is decreased when syntheses are conducted using very high initial 18F-activity. At the same time, the specific activity does not notably increase compared with the syntheses performed with lower initial 18F-activity. To understand the reason for this decrease of the radiochemical yield, a series of reaction kinetic studies was performed by varying the initial 18F-activity while holding other parameters constant (Figure 8). Previously, exchange studies have not addressed the effect of high initial radioactivity.1,2,4,6–8 As Figures 8 and 9 show, high initial 18F-activity reduces the radiochemical yield of [18F]1. Because the only variable in these syntheses is initial 18F-activity, an explanation for this yield decrease is radiolysis. Radiolytic decomposition is caused by the absorption of energy released in radioactive decay either by the labeled molecule itself or by the surrounding (solvent) species that become free radicals. Both of these processes end up attacking the labeled compound.21,22 Radiolysis becomes important especially in solutions where radioactivity concentration and temperature are high. The present reaction conditions caused chemical degradation of 1 (Figure 5), and when a high amount of radioactivity is added to this reaction mixture, radiolysis is evident. Contrary to expectations, initial 18 F-activity had no major impact on specific activity (Figures 8 and 9). The specific activity increased slightly as a function of initial 18F-activity. This is also a consequence of radiolytic and chemical decomposition of [18F]1. However, the specific activity of [18F]1 achieved with high levels of 18F-activity is in the same range as that of molecules of complex structure that have previously been studied.4–6
Conclusion A novel [18F]S1P3 receptor tracer was successfully synthesized using 19F/18F isotope exchange. During the synthesis development, the reaction kinetic parameters were investigated, and correlations between initial 18F-activity, amount of precursor, radiochemical yield and specific activity were observed. High initial 18 F-activity decreases the radiochemical yield and results in a minor increase of specific activity. This is most probably due to the complexity of the molecule and the subsequent susceptibility to radiolytic bond disruption. On the basis of the present results, a convenient condition for the 19F/18F exchange reaction is the use of 2 mmol precursor at a reaction temperature of 190 C starting with 20 GBq of 18F-activity. This afforded a radiochemical yield of ~10% with a specific activity of 0.3 GBq/m mol. The utility of [18F]1 will be assessed in suitable preclinical models. Results from this study are of interest for new tracer development where 19F/18F isotope exchange is used.
Acknowledgement This study received funding from the European Community’s Sixth Framework Program (FP6/2005-2010) under grant agreement no. 511977.
J. Label Compd. Radiopharm 2013, 56 385–391
J. Rokka et al.
Conflict of Interest The authors did not report any conflict of interest.
References [1] F. Cacace, M. Speranza, A. P. Wolf, J. S. Fowler, J. Label. Compd. Radiopharm. 1981, 18, 1721–1730. [2] M. R. Kilbourn, R. Subramanian, J. Label. Compd. Radiopharm. 1990, 28, 1355–1361. [3] J. W. Babich, R. H. Rubin, W. A. Graham, R. A. Wilkingson, J. Vincent, A. J. Fischman, Nucl. Med. Biol. 1996, 23, 995–998. [4] O. Langer, M. Mitterhauser, W. Wadsak, M. Brunner, M. Muller, K. Kletter, M. Muller, J. Label. Compd. Radiopharm. 2003, 46, 715–727. [5] N. N. Ryzhikov, N. A. Gomzina, O. S. Fedorova, D. A. Vasil’ev, A. P. Kostikov, R. N. Krasikova, Radiochemistry 2004, 46(3), 290–294. [6] E. Blom, F. Karimi, B. Långström, J. Label. Compd. Radiopharm. 2009, 52, 504–511. [7] F. Wagner, Ermert J., H. Coenen, J. Nucl. Med. 2009, 50, 1724–1729. [8] N. Malik, W. Voelter, H.-J. Machulla, C. Solbach, J Radioanal Nucl Chem 2010, 287, 287–292. [9] D. Marsolais, H. Rosen, Nat. Rev. Drug Discov. 2009, 8, 298–307. [10] T. Hla, V. Brinkmann, Neurology 2011, 76(suppl 3), S3–S8. [11] S. Milstien, D. Gude, S. Spiegel, Acta Pǽdiatricia 2007, 96, 40–43. [12] L. J. Sim-Silley, P. B. goforth, M. u. Mba, T. L. Macdonald, K. R. Lynch, S. Milstien, S. Spiegel, L. S. Satin, S. P. Welch, D. E. Selley, J. Neurochem. 2009, 110, 1191–1202.
[13] B. Soliven, V. Miron, J. Chun, Neurology 2011, 76(suppl 3), S9–S14. [14] S. Mandala, R. Hajdu, J. Bergstrom, E. Quackenbush, J. Xie, J. Milligan, R. Thornton, G.-J. Shei, D. Card, C. A. Keohane, M. Rosenbach, J. Hale, C. L. Lynch, K. Rupprecht, W. Parsons, H. Rosen, Science 2002, 296, 346–349. [15] A. Huwiler, J. Pfeilschifer, Biochem. Pharmacol. 2008, 75, 1893–1900. [16] D. J. Buzard, S. Han, L. Lopez, A. Kawasaki, J. Moody, L. Thoresen, B. Ullmann, J. Lehmann, I. Calderon, X. Zhu, T. Gharbaoui, D. Sengupta, A. Krishnan, Y. Gao, J. Edwards, J. Barden, M. Morgan, K. Usmani, C. Chen, A. Sadeque, J. Thatte, M. Solomon, L. Fu, K. Whelan, L. Liu, H. Al-Shamma, J. Gatlin, M. Le, C. Xing, S. Espinola, R. M. Jones, Bioorg. Med. Chem. Lett. 2012, 22, 4404–4409. [17] T. Nakamura, M. Asano, Y. Sekiguchi, Y. Mizuno, K. Tamaki, F. Nara, Y. Kawase, Y. Yabe, D. Nakai, E. Kamiyama, Y. Urasaki-Kaneno, T. Shimozato, H. Doi-Komuro, T. Kagari, W. Tomisato, R. Inoue, M. Nagasaki, H. Yuita, K. Oguchi-Oshima, R. Kaneko, T. Nishi, Eur. J. Med. Chem. 2012, 51, 92–98. [18] R. L. Beard, H. Yuan, J. E. Donello, X. Liu, T. Duong, United States Patent Application Publication 2008, US2008/0171772A1. [19] FP6/2005-2010, grant agreement no. 511977, http://www. aditproject.org [12 December 2012]. [20] M. Haaparanta, T. Grönroos, O. Eskola, J. Bergman, O. Solin, J. Chromatogr. A 2006, 1108, 136–139. [21] R. J. Bayly, E. A. Evans, J. Label. Compd. Radiopharm. 1966, 2, 1–34. [22] M. S. Jacobson, H. R. Dankwart, D. W. Mahoney, Appl. Radiat. Isot. 2009, 67, 990–995.
391
J. Label Compd. Radiopharm 2013, 56 385–391
Copyright © 2013 John Wiley & Sons, Ltd.
www.jlcr.org
Revised 14 April 2013,
Accepted 16 April 2013
Published online 5 June 2013 in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/jlcr.3055
19
F/18F exchange synthesis for a novel [18F] S1P3-radiopharmaceutical Johanna Rokka,a Cesare Federico,b Jori Jurttila,a Anniina Snellman,c Merja Haaparanta,c Juha O. Rinne,d and Olof Solina,e* 19 18 F/ F isotope exchange is a useful method to label drug molecules containing 19F-fluorine with 18F without modifying the drug molecule itself. Sphingosine-1-phosphate (S1P) is an important cellular mediator that functions by signaling through cell surface receptors. S1P is involved in several cell responses and may be related to many central nervous system disorders, including neural malfunction in Alzheimer’s disease. In this study, [ 18 F]1-benzyl-N-(3,4-difluorobenzyl)-2-isopropyl-6(2-methoxyethoxy)-1H-indole-3-carboxamide, a novel 18F-labeled positron emission tomography tracer for the S1P3 receptor, was successfully synthesized using the 19F/18F isotope exchange reaction. Parameters of the reaction kinetics were studied, and correlations between the initial 18F-activity, the amount of precursor, radiochemical yield and specific activity (SA) were determined. Contrary to expectations, high initial 18F-activity decreased the radiochemical yield, and only a minor increase of SA occurred. This is most probably due to the complexity of the molecule and the subsequent susceptibility to radiolytic bond disruption. On the basis of the present results, a convenient condition for the 19F/18F exchange reaction is the use of 2 mmol precursor with 20 GBq of 18F-activity. This afforded a radiochemical yield of ~10% with an SA of 0.3 GBq/mmol. Results from this study are of interest for new tracer development where high initial 18F-activity and 19F/18F isotope exchange is used.
Keywords: PET chemistry; sphingosine-1-phosphate; S1P3 receptor; radiolabelling; nucleophilic
Introduction 19
J. Label Compd. Radiopharm 2013, 56 385–391
F-fluorination
(AD).9,10,12,13 Several synthetic compounds have been developed to study the S1PRs, with the best known being FTY720-P, a nonspecific agonist for S1PRs.14 Also, more selective S1P receptor subtype agonists and antagonists have been developed, but the receptor specificity of these compounds have been poor, or application of these compounds has not been reported.9,14–17 A family of 6-substituted indole-3-carboxyclic acid amide compounds is described having S1P receptor agonist or antagonist biological behavior.18 Thus, knowledge of S1PRs’ involvement in health and disease remains limited. Suitable PET tracers will aid in the investigation of the expression and activation of S1PRs and may elucidate the function of these receptors in AD and other CNS disorders.
a Radiopharmaceutical Chemistry Laboratory, Turku PET Centre, University of Turku, Porthaninkatu 3, FI-20500 Turku, Finland b
Siena Biotech S.p.A., Strada del Petriccio e Belriguardo 35, 53100 Siena, Italy
c Preclinical Imaging, Turku PET Centre, University of Turku, Tykistökatu 6, FI-20520 Turku, Finland d Turku PET Centre c/o Turku University Hospital, University of Turku, P.O. Box 52, FI-20521 Turku, Finland e Accelerator Laboratory, Turku PET Centre, Åbo Akademi University, Porthansgatan 3, FI-20500 Turku, Finland
*Correspondence to: Olof Solin, Radiopharmaceutical Chemistry Laboratory, Turku PET Centre, University of Turku, Porthaninkatu 3, FI-20500 Turku, Finland. E-mail: olof.solin@abo.fi
Copyright © 2013 John Wiley & Sons, Ltd.
385
F/18F isotope exchange is a useful method to produce 18Flabeling agents and fluoroaromatic compounds for positron emission tomography (PET). The 19F/18F isotope exchange reaction allows the labeling of drug molecules that contain fluorine in their molecular structure. Several 19F/18F isotope exchange studies examined the effects of the molecular structure of the precursor and have investigated the reaction kinetic parameters to optimize radiochemical yield and specific activity.1–8 Most of these studies have been performed using relatively small amounts of initial activity of 18F-fluorine, and precursor compounds have mostly been relatively simple small molecules.1,2,6,8 Sphingosine-1-phosphate (S1P) is an important cellular mediator. In plasma, S1P is mainly bound to high-density lipoprotein and albumin in nanomolar concentrations and widely distributed throughout the body.9,10 S1P is involved in several cell responses, including cell signaling, neuronal proliferation and cell survival.9–12 Because of the multiple tasks of the S1P, the S1P receptors (S1PRs) are responsible for differentiation of the S1P signaling. S1P modulates through five different receptor subtypes (S1P1–S1P5) belonging to a family of G-protein-coupled receptors.9–13 All S1P receptor subtypes are expressed in the central nervous system (CNS).9–13 The functions of S1P in the CNS are under intense study, and the exact tasks of S1PRs are not fully understood. The expression of S1PRs is possibly related to CNS disorders, and these receptors may be potential drug targets in CNS disorders such as multiple sclerosis and Alzheimer’s disease
18
J. Rokka et al. whereas at a lower temperature (170 C), the reaction mixture did not reach the boiling point of the solution. In entry 7, reaction conditions were held constant, and the end product was purified.
The ADIT project (Design of small molecule therapeutics for the treatment of Alzheimer disease based on the discovery of innovative drug targets, a research project within the European Union’s Sixth Framework Program) was aimed at the development of novel chemical entities endowed with specific neuroprotective activity in AD, up to the preclinical stage.19 During this project, S1P3 was chosen as one of the targets. Preliminary studies within ADIT suggested a possible involvement of S1P3 in regulating neuroinflammation occurring in AD. Our goal in the present study was to synthesize [18F]1-benzylN-(3,4-difluorobenzyl)-2-isopropyl-6-(2-methoxyethoxy)-1H-indole3-carboxamide ([18F]1), a novel 18F-labeled PET tracer for the S1P3 receptor. Radiolabeling was achieved through 19F/18F isotope exchange. During the synthesis development, we also examined the effect of the reaction kinetic parameters on the radiochemical yield and the specific activity of [18F]1.
General procedure for
Chemicals Organic solvents were HPLC grade from Ratherburn Chemicals Ltd (Walkernburn, UK). Water was purified by a Milli-Q Plus Ultra Pure Water system (Millipore, Molsheim, France). Other chemicals were analytical grade from Merck (Darmstadt, Germany). The precursor compound, 1-benzyl-N-(3,4-difluorobenzyl)-2-isopropyl-6-(2-methoxyethoxy)-1H-indole3-carboxamide, was synthesized by AlChemy Fine Chemicals & Research s.r.l. (Baricella, Italy) as HPLC grade.
Purification of the reaction mixture [18F]1-Benzyl-N-(3,4-difluorobenzyl)-2-isopropyl-6-(2-methoxyethoxy)-1Hindole-3-carboxamide ([18F]1) was purified from the reaction mixture using a semi-preparative HPLC method. A Merck Hitachi LaChrom 7000 pump (Merck Hitachi, Darmstadt, Germany) was employed together with a Phenomenex Luna 5 m C18(2) 100 Å (250 10.0 mm) column and gradient system of methanol (A) and water (B). The gradient profile was 0–2 min, 30% A; 2.1 min, 70% A; 10–25 min, 82% A; with a flow rate of 5.0 ml/min. Radioactivity was detected with a NaI(Tl) scintillation detector (200 200 , Bicron Corporation, Newbury, OH, USA) placed on the outflow of the column. The HPLC eluent containing the end product fraction was subsequently removed using solid phase extraction (Sep-PakW light tC18 cartridge, Waters). In this procedure, the fraction isolated by HPLC was first diluted to 20 ml using an ethanol–water solution (5:95 v/v). This solution was applied to the Sep-Pak cartridge. The cartridge was then washed with 20 ml of the ethanol–water solution (5:95 v/v). Finally, the end product, [18F]1, was eluted from the cartridge using 0.5 ml ethanol.
Synthesis F-Fluoride production
18
F-Fluoride was produced using the 18O(p,n)18F nuclear reaction. 18OEnriched water (enrichment grade 98%, 800 ml, Hyox, Rotem Industries Ltd., Israel) was irradiated with 17 MeV protons and a 10 mA beam current for 5–90 min using the MGC-20 cyclotron (Efremov Scientific Research Institute for Electrophysical Apparatuses (NIIEFA), St. Petersburg, Russia) at the Åbo Akademi Accelerator Laboratory. The initial radioactivity of 18 F-fluoride was 3.5–33 GBq.
Reaction conditions for
F/18F-fluorine isotope exchange
The aqueous solution of 18F-fluoride from the cyclotron target was collected into a 2-ml Wheaton V-vial (Wheaton Science Products, Millville, USA) containing 2 mmol potassium bicarbonate and 4 mmol Kryptofix 222. Water was removed using an azeotropic distillation procedure, where acetonitrile (1 ml) was added to the solution and the liquid was evaporated under a stream of helium. The solution was heated at 100 C for 3 min. The distillation procedure was repeated three times. The precursor 1, 1-benzyl-N-(3,4-difluorobenzyl)-2-isopropyl-6-(2methoxyethoxy)-1H-indole-3-carboxamide, was dissolved in 300 ml dimethyl sulfoxide (DMSO). The precursor–DMSO solution was added to the dry [18F]-fluoride–Kryptofix complex. The reaction mixture was then heated at 170 or 190 C (Figure 1). For reaction kinetic studies, aliquots of the reaction mixture were withdrawn at intervals for up to 60 min and then analyzed. Starting with a high initial activity of 18F (Table 1. entry 7), 16 syntheses were performed to produce [18F]1. In these syntheses, reaction mixtures were heated for 20 min, and the product [18F]1 was purified. No samples were taken during these syntheses.
Experimental
18
19
19
F/18F isotope exchange
Analytical procedures
Reaction conditions are listed in Table 1. To define the optimum reaction conditions (Table 1, entries 1–6), we varied the amounts of initial 18Factivity, the unlabeled fluorinated precursor, potassium bicarbonate, and the reaction temperature. At 190 C, the reaction mixture refluxed,
Table 1. Reaction conditions and starting materials for
The radiochemical yield and the specific activity were determined from reaction mixture samples removed at specified intervals and from the end product of the production syntheses. Radiochemical yield was
19
F/18F isotope exchange synthesis of [18F]1
Entry
A0 [GBq]
Precursor [mmol]
KHCO3 [mmol]
K222 [mmol]
Temperature [ C]
1 2 3 4 5 6 7
6.9 0.5 0 0 6.3 0.4 6.7 0.3 3.6–21.6 30.4 2.2
3.9 0.5 2.3 2.0 0.5–8.2 0.5–7.8 2.5 0.3 2.4 0.4
1.0–4.0 2.0 2.0 2.01 0.01 2.00 0.01 2.01 0.01 2.02 0.02
1.9–8.5 5.0 5.0 4.6 0.5 4.5 0.4 4.3 0.3 5.7 1.2
190 170 190 170 190 190 190
386
In each synthesis, target water was removed with acetonitrile and azeotropic distillation, and precursor was dissolved in 300 ml DMSO. A0 denotes initial 18F-activity used in syntheses. Range of investigated parameter is included in the table. Constant amounts are expressed as mean SD. In entry 1, the amount of base is varied; in entries 2 and 3, the speed of chemical decomposition of 1 are studied without adding radioactivity; in entries 4 and 5, the amount of precursor is varied at two different temperatures; in entry 6, the amount of radioactivity is varied using a fixed amount of precursor; and finally, entry 7 contains the parameters used for a set of syntheses (n = 16).
www.jlcr.org
Copyright © 2013 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2013, 56 385–391
J. Rokka et al.
Figure 1. Scheme for
19
F/18F isotope exchange reaction for [18F]1.
determined with a thin layer chromatography (TLC) method, and digital autoradiography. TLC plates were glass plates pre-coated with HPTLC Silica gel 60 RP-18 F254s (art. no. 113724) from Merck (Darmstadt, Germany). TLC plates were developed in a twin-trough chamber (10 10 cm, Camag, Muttenz, Switzerland) using a mixture of 20 mM ammonium acetate (pH 4.3)–methanol (30:70 v/v) as a solvent. Radioactivity on the plate was detected and quantified using photostimulated luminescence digital autoradiography. A TLC plate was placed into an exposure cassette with an imaging plate (Fuji Imaging Plate BAS-MS2025, Fuji Photo Film Co., Ltd., Tokyo, Japan). After exposure, the imaging plate was scanned using the Fuji Analyzer BAS 1800 reader (Fujifilm Co., Ltd., Tokyo, Japan) at 200 mm resolution. Data were analyzed using the Tina 2.1 program (Raytest Isotopenmessgeräte GmbH, Straubenhardt, Germany). The autoradiographic image and its analysis are shown in Figure 2. Radiochemical yield is presented as a percentage of total radioactivity in the sample. Specific activity was measured using an analytical HPLC method. For HPLC analysis, a Merck Hitachi LaChrom 7000 system with Hitachi D-7000 HPLC System Manager software (version 3.0) was used. The UV
PSL/mm2
200
[18F]1
detector of the LaChrom HPLC system was set at 254 nm. Radioactivity was detected with similar NaI(TI) scintillation detector, as described earlier, connected on the outflow of the column. Quality control analysis was carried out using a Phenomenex Gemini NX 3 m C18 110 Å (150 4.60 mm) column and an isocratic system of 20 mM ammonium acetate (pH 4.3)–methanol (30:70 v/v). In the analysis, the concentration of the product was determined. Also, the product peak identified by the radioactive detector was collected, and the radioactivity of this HPLC fraction was measured for specific activity determination. Chromatograms from the reaction mixture samples and end products are shown in Figure 3. The measured specific activity was decay corrected to the end of bombardment. The stability of the precursor was evaluated in studies of reaction solutions to which no radioactivity was added (Table 1, entries 2 and 3). The chemical decomposition of 1 was followed by analyzing the reaction mixture samples with the same analytical HPLC method used for specific activity measurement. The recovery percent of 1 was calculated from these HPLC results. The specific activity and the radiochemical yield were determined using different methods. The TLC method combined with digital autoradiography was used to determine the radiochemical yield. Digital autoradiography was chosen because it has a wide linear range and high sensitivity.20 Specific activity was determined with an analytical HPLC method that enables concentration measurement of 1. The analytical HPLC method was not as sensitive as the TLC method for radioactivity determination (Figure 6B and D); thus, the specific activity was not measurable at all time points.
Statistics
100
Mean values were calculated from the individual measurements and expressed at a precision of one standard deviation (mean SD).
Results 0 0
0.5
1
Rf
J. Label Compd. Radiopharm 2013, 56 385–391
Copyright © 2013 John Wiley & Sons, Ltd.
www.jlcr.org
387
Figure 2. Autoradiographic image of TLC plate showing crude reaction product during reaction. Rf value of [18F]1 was 0.35. The sample was applied on HPTLC Silica gel 60 RP-18 F254s plate, which was developed with a mixture of 20 mM ammonium acetate (pH 4.3)–methanol (30:70 v/v). The radioactivity distribution was detected and analyzed using photostimulated luminescence autoradiography. Radiochemical yield was determined using this method.
Several series of syntheses were made to study the isotopic exchange reaction of 1 (Figure 1, Table 1). The radiochemical yield and specific activity were measured using two different methods (Figures 2 and 3). The optimization of reaction conditions was started with selection of a suitable base for promotion of 19F/18F exchange. Preliminary studies established that 2 mmol potassium bicarbonate (Table 1, entry 1) gave satisfactory results in the 19F/18F isotope exchange reaction
J. Rokka et al. 100
A [ F]1
170 °C 190 °C
0.08
Radioactivity UV absorption
2.5
0.04
Amount of 1 [%]
18
UV absorption [mAFU]
Radioactivity [arb. Units]
5
75
50 2
R = 0.885
25
2
R = 0.955
0 0
0
0 0
5
10
15
20
0.6
Radioactivity UV absorption
0.3
6
UV absorption [mAFU]
Radioactivity [arb. Units]
18
0
0 0
5
10
15
20
Time [min] Figure 3. Radio-HPLC of [18F]1: (A) UV absorption and radioactivity chromatograms of reaction mixture and (B) separated end product. Phenomenex Gemini NX 3 m C18 110 Å column and an isocratic system of 20 mM ammonium acetate (pH 4.3)–methanol (30:70 v/v) was used. Specific activity was measured using this method.
(Figure 4). Further experimentation was thus carried out under these conditions. Exposure of the precursor 1 to heat at both 170 and 190 C caused chemical degradation over time (Table 1, entries 2 and 3; Figure 5). High temperature (Table 1, entries 4 and 5) increased both radiochemical yield (Figure 6A and B) and specific activity (Figure 6C and D). This holds for all experiments except for the yield at the highest concentration of the precursor. Using 8 mmol
Radiochemical yield [%]
20 Amount of KHCO3 1.0 µmol 2.0 µmol 4.0 µmol
10
0 0
20
40
60
Time [min]
388
Figure 4. The effects of exchange reaction time and potassium bicarbonate concentration on radiochemical yield of [18F]1. Whole lines show radiochemical yield at various potassium bicarbonate concentration used in reactions. All other conditions in these reactions were unchanged (Table 1, entry 1).
www.jlcr.org
45
60
Figure 5. Precursor 1 chemical decomposition as a function of time at 170 and 190 C. Precursor was handled in a similar manner as in reactions with radioactivity. However, no radioactivity was added to the reaction vessel (Table 1, entries 2 and 3). Exponential curves were fitted to measured data points.
B [ F]1
30
Time [min]
Time [min]
12
15
Copyright © 2013 John Wiley & Sons, Ltd.
at 170 C resulted in a yield of 50% versus 30% at 190 C. The specific activities were similar in these two experiments. We examined the impact of the amount of precursor supplied in the reaction (Table 1, entries 4 and 5; Figure 6). With higher amounts of precursor, better radiochemical yields were gained, but at the same time, the specific activity of the product was low. Figure 7 correlates the specific activity with the amount of precursor at 190 C after 15 min of reaction. The effects of varied amounts of initial 18F-activities (Table 1, entry 6) on the measured radiochemical yield and specific activity are shown in Figure 8. The reaction containing the highest amount of added radioactivity (21.6 GBq) showed the lowest level of radiochemical yield. At the same time, the specific activity was the highest. In syntheses where product [18F]1 was isolated (Table 1, entry 7), the reaction conditions were constant, and the end product was purified. Total synthesis time was 75 min, and the radiochemical purity exceeded 98% in all syntheses. The average radiochemical yield of these syntheses was 1.0 0.5%, and the specific activity was 0.36 0.19 GBq/mmol (values are decay corrected at the end of bombardment, n = 16). The radiochemical yield and specific activity (Table 1, entry 6) were plotted as a function of initial 18F-activity at the time point of 20 min (Figure 9). As the initial concentration of 18Fradioactivity increased, the radiochemical yield of [18F]1 showed a generally decreasing trend, whereas the specific activity slightly increased as the initial radioactivity increased. Also the mean results from the syntheses in which the product was isolated (n = 16, Table 1, entry 7) are included in Figure 9.
Discussion Previously, direct 19F/18F isotope exchange reactions have been applied to 18F-fluorination of small aromatic molecules. 1,2,5,6 18 F-Labeling of larger molecules with 19F/18F isotope exchange reaction has been reported to be more difficult,4,6 and commonly, these molecules have been produced with at least a two-step synthesis where protecting groups were needed or by coupling a 18F-labeled small molecule with a larger molecule.4,7 In this study, [18F]1 was synthesized using one-step synthesis and 19F/18F isotope exchange reaction (Figure 1). The effects of initial 18F-activity, temperature and amount of precursor on the radiochemical yield and the specific activity
J. Label Compd. Radiopharm 2013, 56 385–391
J. Rokka et al.
A
B
Radiochemical yield [%]
170 °C Amount of precursor
45
0.45 µmol 1.06 µmol 2.11 µmol 8.22 µmol
30
15
Radiochemical yield [%]
30
60
0
190 °C Amount of precursor 0.49 µmol 0.97 µmol 2.13 µmol 7.76 µmol
15
0 0
15
30
45
60
0
15
45
60
D 170 °C
0.3
Amount of precursor 0.45 µmol 1.06 µmol 2.11 µmol 8.22 µmol
0.15
0 0
15
30
45
Specific activity [GBq/µmol]
C Specific activity [GBq/µmol]
30
Time [min]
Time [min]
0.9
190 °C Amount of precursor 0.49 µmol 0.97 µmol 2.13 µmol 7.76 µmol
0.6
0.3
0 0
60
Time [min]
15
30
45
60
Time [min]
Specific activity [GBq/µmol]
Figure 6. (A, B) Radiochemical yield and (C, D) specific activity of [18F]1 as a function of exchange reaction time at 170 and 190 C, respectively. The amount of precursor (different lines in the graphs) and temperature (vertical columns) were varied. Otherwise, reaction conditions were constant (Table 1, entries 4 and 5). The dashed line in panel D denotes the time point used for presenting specific activity as a function of precursor amount in Figure 7.
0.9
0.6
0.3 R2 = 0.857
0 0
4
8
Precursor amount [ mol] Figure 7. Specific activity of [18F]1 as a function of precursor amount at 15-min reaction time and 190 C (see also Figure 6D). The solid line is a fit to the data points.
J. Label Compd. Radiopharm 2013, 56 385–391
Copyright © 2013 John Wiley & Sons, Ltd.
www.jlcr.org
389
for the 19F/18F isotope exchange reaction were studied. [18F]1 has a notably more complex structure than compounds that have been reported before for labeling with direct 19F/18F isotope exchange. Thus, it is not surprising that the yield of [18F]1 is low, presumably because of the complex structure of 1, which is vulnerable to chemical and radiolytic degradation. Preliminary studies established that DMSO is a proper reaction solvent, confirming the results of previous investigations of isotope exchange reactions, especially when high temperature is needed.1,3,6 The amount of the base, KHCO3, affects the radiochemical yield of the reaction (Figure 4). With low amounts of KHCO3, only a minor reaction is observed. With larger amounts of KHCO3, the radiochemical yield increases. To avoid base-catalyzed chemical decomposition of the precursor and [18F]1, 2 mmol of base was used in the subsequent studies. Other bases such as K2CO3 and tetrabutylammonium hydroxide were initially tested, but in these syntheses, the radiochemical yield and the specific activity were not improved (data not shown).
The 19F/18F isotope exchange reaction was performed using relatively harsh conditions. The large substituent attached at the para position of the 1,2-difluorobenzene ring of 1 reduces reactivity of the benzene ring and makes the isotope exchange reaction difficult, an outcome that has been demonstrated earlier with similar compounds.4,6 Using the present reaction conditions, we observed that the chemical decomposition of the precursor and also [18F]1 is rapid (Figure 5). The chemical decomposition of the compound will eventually limit the exchange reaction because the entire amount of precursor is consumed by decomposition. The chemical degradation of 1 is also seen in the radiochemical yield curves of the kinetic reaction studies (Figure 6) where the radiochemical yield drops off at the later time points, especially when lower amounts of precursor are used. This chemical decomposition will, in turn, under otherwise set conditions, limit the extent of radiochemical yield and specific activity. For example, in Figure 6D, after 20 min of reaction time, the amount of 1 is below the detection limit, and the specific activity was thus not measurable. Reaction temperature has an effect on radiochemical yield and specific activity. With a refluxing reaction mixture (190 C), the isotope exchange rate is faster leading to higher radiochemical yield and specific radioactivity than the syntheses carried out under milder conditions (170 C), where the temperature is below the boiling point of the reaction mixture (Figure 6). A similar effect of temperature has also been shown in earlier studies,4–7 although in these studies, less drastic temperatures were used. One exception to this was the synthesis in which we used a very large amount (8 mmol) of precursor at lower temperature. The chemical degradation of 1 is less rapid at 170 C than at 190 C (Figure 5) thus increasing notably the radiochemical yield of this synthesis. On the other hand, at these conditions, the specific activity was low.
J. Rokka et al.
A Initial18 F-activity 3.6 GBq 6.5 GBq 12.3 GBq 21.6 GBq
Radiochemical yield [%]
20
10
0 0
15
30
45
60
Time [min]
Specific activity [GBq/µmol]
B Initial18 F-activity 3.6 GBq 6.5 GBq 12.3 GBq 21.6 GBq
0.4
0.2
0 0
15
30
45
60
Time [min]
Radiochemical yield Specific activity
Radiochemical yield [%]
20
0.60
0.40 10 0.20
0
Specific activity [GBq/µmol]
Figure 8. (A) Radiochemical yield of [18F]1 at 190 C as a function of exchange reaction time. The initial 18F-activity was varied from 3 to 22 GBq. In these studies, the amount of precursor 1 was 2 mmol (Table 1, entry 6). The dashed line denotes the time point used for presenting data in Figure 9. (B) Specific activity of [18F]1 at 190 C as a function of exchange reaction time. The initial 18F-activity was varied from 3 to 22 GBq. The amount of precursor 1 was 2 mmol (Table 1, entry 6). The dashed line denotes the time point used for presenting data in Figure 9.
0.00 0
10
20
18F-radioactivity
30
[GBq]
Figure 9. Radiochemical yield and specific activity at 190 C as a function of initial 18 F-activity at the time point 20-min reaction time. Values at 30 GBq are average values (n = 16; Table 1, entry 7) from the production syntheses of [18F]1.
390
Previous studies have shown that radiochemical yield is proportional and specific activity inversely proportional to the amount of precursor supplied in the reaction.1,2,7 In the present investigation, we observed the same effect (Figures 6 and 7). The correlation between precursor amount and specific activity is not so significant as in previous studies,1,2,7 but on the other
www.jlcr.org
Copyright © 2013 John Wiley & Sons, Ltd.
hand, these studies were performed with compounds of less complex structure than [18F]1. According to expectations based on previously published studies, absolute radiochemical yield and specific activity increase when initial 18F-activity is raised.2,7,8 We did 16 syntheses with high initial 18F-activity (30 GBq), purified the product and found, however, that the end product radiochemical yield is decreased when syntheses are conducted using very high initial 18F-activity. At the same time, the specific activity does not notably increase compared with the syntheses performed with lower initial 18F-activity. To understand the reason for this decrease of the radiochemical yield, a series of reaction kinetic studies was performed by varying the initial 18F-activity while holding other parameters constant (Figure 8). Previously, exchange studies have not addressed the effect of high initial radioactivity.1,2,4,6–8 As Figures 8 and 9 show, high initial 18F-activity reduces the radiochemical yield of [18F]1. Because the only variable in these syntheses is initial 18F-activity, an explanation for this yield decrease is radiolysis. Radiolytic decomposition is caused by the absorption of energy released in radioactive decay either by the labeled molecule itself or by the surrounding (solvent) species that become free radicals. Both of these processes end up attacking the labeled compound.21,22 Radiolysis becomes important especially in solutions where radioactivity concentration and temperature are high. The present reaction conditions caused chemical degradation of 1 (Figure 5), and when a high amount of radioactivity is added to this reaction mixture, radiolysis is evident. Contrary to expectations, initial 18 F-activity had no major impact on specific activity (Figures 8 and 9). The specific activity increased slightly as a function of initial 18F-activity. This is also a consequence of radiolytic and chemical decomposition of [18F]1. However, the specific activity of [18F]1 achieved with high levels of 18F-activity is in the same range as that of molecules of complex structure that have previously been studied.4–6
Conclusion A novel [18F]S1P3 receptor tracer was successfully synthesized using 19F/18F isotope exchange. During the synthesis development, the reaction kinetic parameters were investigated, and correlations between initial 18F-activity, amount of precursor, radiochemical yield and specific activity were observed. High initial 18 F-activity decreases the radiochemical yield and results in a minor increase of specific activity. This is most probably due to the complexity of the molecule and the subsequent susceptibility to radiolytic bond disruption. On the basis of the present results, a convenient condition for the 19F/18F exchange reaction is the use of 2 mmol precursor at a reaction temperature of 190 C starting with 20 GBq of 18F-activity. This afforded a radiochemical yield of ~10% with a specific activity of 0.3 GBq/m mol. The utility of [18F]1 will be assessed in suitable preclinical models. Results from this study are of interest for new tracer development where 19F/18F isotope exchange is used.
Acknowledgement This study received funding from the European Community’s Sixth Framework Program (FP6/2005-2010) under grant agreement no. 511977.
J. Label Compd. Radiopharm 2013, 56 385–391
J. Rokka et al.
Conflict of Interest The authors did not report any conflict of interest.
References [1] F. Cacace, M. Speranza, A. P. Wolf, J. S. Fowler, J. Label. Compd. Radiopharm. 1981, 18, 1721–1730. [2] M. R. Kilbourn, R. Subramanian, J. Label. Compd. Radiopharm. 1990, 28, 1355–1361. [3] J. W. Babich, R. H. Rubin, W. A. Graham, R. A. Wilkingson, J. Vincent, A. J. Fischman, Nucl. Med. Biol. 1996, 23, 995–998. [4] O. Langer, M. Mitterhauser, W. Wadsak, M. Brunner, M. Muller, K. Kletter, M. Muller, J. Label. Compd. Radiopharm. 2003, 46, 715–727. [5] N. N. Ryzhikov, N. A. Gomzina, O. S. Fedorova, D. A. Vasil’ev, A. P. Kostikov, R. N. Krasikova, Radiochemistry 2004, 46(3), 290–294. [6] E. Blom, F. Karimi, B. Långström, J. Label. Compd. Radiopharm. 2009, 52, 504–511. [7] F. Wagner, Ermert J., H. Coenen, J. Nucl. Med. 2009, 50, 1724–1729. [8] N. Malik, W. Voelter, H.-J. Machulla, C. Solbach, J Radioanal Nucl Chem 2010, 287, 287–292. [9] D. Marsolais, H. Rosen, Nat. Rev. Drug Discov. 2009, 8, 298–307. [10] T. Hla, V. Brinkmann, Neurology 2011, 76(suppl 3), S3–S8. [11] S. Milstien, D. Gude, S. Spiegel, Acta Pǽdiatricia 2007, 96, 40–43. [12] L. J. Sim-Silley, P. B. goforth, M. u. Mba, T. L. Macdonald, K. R. Lynch, S. Milstien, S. Spiegel, L. S. Satin, S. P. Welch, D. E. Selley, J. Neurochem. 2009, 110, 1191–1202.
[13] B. Soliven, V. Miron, J. Chun, Neurology 2011, 76(suppl 3), S9–S14. [14] S. Mandala, R. Hajdu, J. Bergstrom, E. Quackenbush, J. Xie, J. Milligan, R. Thornton, G.-J. Shei, D. Card, C. A. Keohane, M. Rosenbach, J. Hale, C. L. Lynch, K. Rupprecht, W. Parsons, H. Rosen, Science 2002, 296, 346–349. [15] A. Huwiler, J. Pfeilschifer, Biochem. Pharmacol. 2008, 75, 1893–1900. [16] D. J. Buzard, S. Han, L. Lopez, A. Kawasaki, J. Moody, L. Thoresen, B. Ullmann, J. Lehmann, I. Calderon, X. Zhu, T. Gharbaoui, D. Sengupta, A. Krishnan, Y. Gao, J. Edwards, J. Barden, M. Morgan, K. Usmani, C. Chen, A. Sadeque, J. Thatte, M. Solomon, L. Fu, K. Whelan, L. Liu, H. Al-Shamma, J. Gatlin, M. Le, C. Xing, S. Espinola, R. M. Jones, Bioorg. Med. Chem. Lett. 2012, 22, 4404–4409. [17] T. Nakamura, M. Asano, Y. Sekiguchi, Y. Mizuno, K. Tamaki, F. Nara, Y. Kawase, Y. Yabe, D. Nakai, E. Kamiyama, Y. Urasaki-Kaneno, T. Shimozato, H. Doi-Komuro, T. Kagari, W. Tomisato, R. Inoue, M. Nagasaki, H. Yuita, K. Oguchi-Oshima, R. Kaneko, T. Nishi, Eur. J. Med. Chem. 2012, 51, 92–98. [18] R. L. Beard, H. Yuan, J. E. Donello, X. Liu, T. Duong, United States Patent Application Publication 2008, US2008/0171772A1. [19] FP6/2005-2010, grant agreement no. 511977, http://www. aditproject.org [12 December 2012]. [20] M. Haaparanta, T. Grönroos, O. Eskola, J. Bergman, O. Solin, J. Chromatogr. A 2006, 1108, 136–139. [21] R. J. Bayly, E. A. Evans, J. Label. Compd. Radiopharm. 1966, 2, 1–34. [22] M. S. Jacobson, H. R. Dankwart, D. W. Mahoney, Appl. Radiat. Isot. 2009, 67, 990–995.
391
J. Label Compd. Radiopharm 2013, 56 385–391
Copyright © 2013 John Wiley & Sons, Ltd.
www.jlcr.org
![Synthesis and evaluation of 4-[18F]fluoropropoxy-3-iodobenzylguanidine ([18F]FPOIBG): A novel 18F-labeled analogue of MIBG.](/assets/img/hold.png)
