Applied Radiation and Isotopes 99 (2015) 133–137
Contents lists available at ScienceDirect
Applied Radiation and Isotopes journal homepage: www.elsevier.com/locate/apradiso
Automated radiosynthesis of [18F]ciprofloxacin Johann Stanek a,b, Severin Mairinger a, Thomas Wanek a, Claudia Kuntner a, Markus Müller b, Oliver Langer a,b,n a b
Health and Environment Department, AIT Austrian Institute of Technology GmbH, A-2444 Seibersdorf, Austria Department of Clinical Pharmacology, Medical University of Vienna, Währiger-Gürtel 18-20, A-1090 Vienna, Austria
H I G H L I G H T S
G R A P H I C A L
A B S T R A C T
Automated synthesis of [18F]ciprofloxacin in a TRACERlabTM FXFDG (GE Healthcare) synthesis module was developed. Dependence of radiochemical yield on reactor type was observed. 3-mL V-shaped borosilicate glass reactor gave higher radiochemical yield as compared with standard 15-mL glassy carbon reactor. V-shaped borosilicate glass reactor might also give higher radiochemical yield for other [18F]radiotracers than [18F]ciprofloxacin.
art ic l e i nf o
a b s t r a c t
Article history: Received 19 November 2014 Received in revised form 18 February 2015 Accepted 26 February 2015 Available online 27 February 2015
We transferred the previously published manual synthesis of [18F]ciprofloxacin (decay-corrected RCY: 5.5 71.0%) to an automated synthesis module (TRACERlabTM FXFDG, GE Healthcare) and observed a strong decrease in RCY (0.4 70.4%). When replacing the standard 15-mL glassy carbon reactor of the synthesis module with a 3-mL V-shaped borosilicate glass reactor a considerable improvement in RCY was observed. [18F]Ciprofloxacin was obtained in a RCY of 2.7 71.4% (n ¼23) with a specific activity at EOS of 1.47 0.5 GBq/mmol in a synthesis time of 160 min. & 2015 Elsevier Ltd. All rights reserved.
Keywords: [18F]ciprofloxacin Automated synthesis TRACERlabTM FXFDG Radiochemical yield Reactor type 18 19 F/ F exchange
1. Introduction Ciprofloxacin (1-cyclopropyl-6-fluoro-1,4-dihydro-4-oxo-7-(1piperazinyl)-quinolone-3-carboxylic acid) is a still widely prescribed fluoroquinolone antibiotic which displays a broad spectrum of activity against Gram-negative and Gram-positive bacteria.
n Corresponding author at: Health and Environment Department, AIT Austrian Institute of Technology GmbH, A-2444 Seibersdorf, Austria. Fax: þ43 50550 2136. E-mail address: [email protected] (O. Langer).
http://dx.doi.org/10.1016/j.apradiso.2015.02.024 0969-8043/& 2015 Elsevier Ltd. All rights reserved.
Ciprofloxacin has been described as a substrate of several different active membrane transporter proteins belonging to the solute carrier (SLC) or adenosine triphosphate-binding cassette (ABC) families, such as organic anion transporter 3 (OAT3), breast cancer resistance protein (BCRP), multidrug resistance protein 4 (MRP4) and multidrug and toxin extrusion protein 1 (MATE1) (Mulgaonkar et al., 2012). These transporters mediate hepatobiliary and renal excretion of their substrates and drug–drug interactions involving inhibition of one or several of these transporters could lead to altered tissue distribution of ciprofloxacin, which may have an impact
134
J. Stanek et al. / Applied Radiation and Isotopes 99 (2015) 133–137
2.2. Automated synthesis of [18F]ciprofloxacin
Fig. 1. Reaction scheme for the synthesis of [18F]ciprofloxacin.
on its efficacy and safety (Giacomini et al., 2010). Ciprofloxacin contains fluorine in its structure which can be replaced with the positron-emitting radionuclide fluorine-18 (18F, half-life 109.8 min) (Langer et al., 2003a,b). We have previously shown that positron emission tomography (PET) imaging with [18F]ciprofloxacin can be used to non-invasively study the tissue distribution of ciprofloxacin in humans (Langer et al., 2005). Studying the tissue distribution of [18F]ciprofloxacin in genetically modified mice, which lack one or several membrane transporter proteins, may help to identify the relevant transporters, which are likely to affect ciprofloxacin tissue distribution in humans. In previous studies, ciprofloxacin has been prepared by a manual two-step one-pot synthesis method comprising a nucleophilic 18F/19F exchange reaction of 7-chloro-1-cyclopropyl-6fluoro-1,4-dihydro-4-oxoquinoline-3-carboxylic acid followed by reaction of the 18F-labeled carboxylic acid derivative with piperazine (Fig. 1) (Langer et al., 2003a,b). Aim of this work was to develop an automated synthesis of [18F]ciprofloxacin in a commercially available automated synthesis module (TRACERlabTM FXFDG, General Electric Healthcare, Uppsala, Sweden) in order to reduce the radiation exposure associated with manual radiosynthesis.
2. Materials and methods 2.1. General All chemicals were purchased from Sigma-Aldrich Handels GmbH (Vienna, Austria) and used without further purification. Aqueous (aq.) [18F]fluoride was produced using a PETtrace cyclotron (GE Healthcare) via the 18O(p,n)18F nuclear reaction by irradiation of a 2.6 mL water target containing 95.9% enriched [18O]water (ABX-advanced biochemical compounds, Radeberg, Germany) with a 16.5 MeV proton beam. Radiochemical purity and specific activity of [18F]ciprofloxacin were determined with analytical high-performance liquid chromatography (HPLC) using an Agilent 1200 system (Agilent Technologies Österreich GmbH, Vienna, Austria) consisting of a quaternary pump, an auto-sampler and a column oven. Ultraviolet (UV) absorption was detected with an Agilent 1200 diode array detector at a wavelength of 280 nm in series with a Raytest “Gabi Star” detector (raytest Isotopenmessgeräte GmbH, Straubenhardt, Germany) for radioactivity detection. A PRP-1 column (4.1 250 mm2, 5 mm, Hamilton Bonaduz AG, Bonaduz, Switzerland), heated to 40 °C, was isocratically eluted with a 85/15 (v/v) mixture of aq. 0.01 M phosphoric acid (H3PO4) and ethanol at a flow rate of 1 mL/min. Osmolality (mosmol/kg) of formulated [18F]ciprofloxacin solution was measured using a Wescor Vapro 5520 Pressure Osmometer (Wescor Inc., Logan, USA). The pH value was determined with a pH-Meter Inolab pH720 (WTW Wissenschaftlich-Technische Werkstätten GmbH, Weilheim, Germany).
Radiosynthesis of [18F]ciprofloxacin was performed as a twostep one-pot reaction (Fig. 1) in a custom-modified dual-reactor TRACERlabTM FXFDG synthesis module (GE Healthcare). This synthesis module originally contained two independent synthesis units in a single housing and was originally designed to operate two consecutive [18F]FDG syntheses without reloading and opening the hot cell. Our synthesis module was custom-modified by the manufacturer to enable two-pot [18F]radiosyntheses (Fig. 2). We used the synthesis module either in standard configuration with two large-volume glassy carbon reactors (15 mL) (configuration A) or in a modified configuration in which one large-volume glassy carbon reactor (reactor 1) had been replaced with a small-volume V-shaped borosilicate glass reactor (3 mL) (configuration B, Fig. 3). In configuration B also the heating block of the reactor was replaced so that the bore size fitted to the smaller diameter of the reactor (Fig. 3, left picture). In order to facilitate reflux of the reaction solution in configuration B the 3-mL V-shaped borosilicate glass reactor was mounted in a way that 1.5 cm remained outside the heating block. In both configurations the two reaction steps of the [18F]ciprofloxacin synthesis were performed in reactor 1 (Figs. 2 and 3). All reagent solutions were prepared prior to start of synthesis and placed in the synthesis module storage vessels. Amounts of reagents and solvent volumes were different for configurations A and B as indicated below. After delivery of the irradiated [18O]water to the synthesis module, [18F]fluoride was trapped on an anion exchange cartridge (PS-HCO3, 45 mg, Macherey-Nagel, Düren, Germany), which had been pre-activated with ethanol (3 mL) and water (5 mL). [18F]Fluoride was eluted into the synthesis reactor by rinsing the cartridge with a mixture of kryptofix 2.2.2 (4,7,13,16,21,24-hexaoxa-1,10-diazabicyclo[8.8.8] hexacosane, configuration A: 16 mg, 42.5 mmol; configuration B: 12 mg, 31.9 mmol) in acetonitrile (0.9 mL) and potassium carbonate (3.5 mg, 25.3 μmol) in water (0.1 mL) (added via valve 1, V1, see Fig. 2). After adding acetonitrile (0.5 mL) to the synthesis reactor (via V2), the solvent was evaporated azeotropically under vacuum, first for 3 min at 60 °C and then for further 5 min at 120 °C to remove the remaining water. To the dried K[18F]F-K222 complex, radiolabelling precursor 7-chloro1-cyclopropyl-6-fluoro-1,4-dihydro-4-oxoquinoline-3-carboxylic acid (configuration A: 12.0 mg, 42.6 mmol; configuration B: 6.0 mg, 21.3 mmol) dissolved in dimethyl sulfoxide (DMSO, configuration A: 0.8 mL; configuration B: 0.3 mL) was added (V6) and the reaction mixture was heated at 175 °C for 40 min. The reaction mixture was then cooled to 40 °C and a mixture of trimethylborate (configuration A: 100 mL, 897 mmol; configuration B: 20 mL, 179 mmol) and glacial acetic acid (configuration A: 100 mL, 1747 mmol; configuration B: 20 mL, 349 mmol) in DMSO (configuration A: 0.3 mL; configuration B: 0.1 mL) was added (V5) and the reaction mixture was stirred at 40 °C for 2 min. Then a solution of piperazine (configuration A: 57 mg, 661.7 mmol; configuration B: 20 mg, 232.2 mmol) in DMSO (configuration A: 0.7 mL; configuration B: 0.35 mL) was added (V4) and the mixture was reacted at 175 °C for 40 min. After cooling to 40 °C, a mixture of aq. 0.01 M H3PO4 and ethanol (85/15, v/v, configuration A: 2.5 mL; configuration B: 1 mL) was added (V3) and the crude reaction mixture was injected into the built-in semipreparative HPLC system. A Hamilton PRP-1 column (10 250 mm2, 10 mm) equipped with a PRP-1 guard column (10 40 mm2) was eluted at a flow rate of 3 mL/min for the first 12 min with a mixture of aq. 0.01 M H3PO4 and ethanol (97/3, v/v) followed by an increase of the ethanol percentage to 15%. The HPLC eluate was monitored in series for radioactivity and UV absorption at a wavelength of 280 nm. [18F]Ciprofloxacin, which eluted with a retention time of 25–30 min in a volume of 15– 20 mL (Fig. 4), was directly passed over a strong cation exchange
J. Stanek et al. / Applied Radiation and Isotopes 99 (2015) 133–137
135
Fig. 2. Scheme of the automated synthesis module for the synthesis of [18F]ciprofloxacin. The module is a dual-reactor TRACERlabTM FXFDG synthesis module, which has been custom-modified to enable two-pot radiosyntheses.
Fig. 3. Pictures of dual-reactor synthesis module in modified configuration with one V-shaped borosilicate glass reactor (reactor 1) and one 15-mL glassy carbon reactor (reactor 2). The one-pot radiosynthesis was performed in reactor 1. Note that the heating block of reactor 1 was replaced so that the bore size fitted to the smaller diameter of the reactor.
cartridge (Isolute 100 mg SCX, International Sorbent Technology Ltd., Hengoed, UK), which had been pre-activated with ethanol (5 mL) and mobile phase for semipreparative HPLC (5 mL). The cartridge was washed with water (5 mL, V8) and [18F]ciprofloxacin was then eluted from the cartridge with 0.1 M aq. sodium hydroxide solution (2.5 mL, V7). The final product was formulated for intravenous injection into rodents by adding sodium dihydrogenphosphate dihydrate (45 mg, 288 mmol) and aq. sodium chloride solution (10%, w/v, 50 mL). 2.3. Manual synthesis of [18F]ciprofloxacin The manual synthesis of [18F]ciprofloxacin was performed as published previously (Langer et al., 2003a,b) using a 3-mL Wheaton V-vial (clear type I borosilicate glass, Wheaton Science Products, Millville, USA) equipped with a polytetrafluoroethylene/silicone septum and a magnetic stirrer. Amounts of reagents and
solvent volumes were identical with the automated synthesis (configuration B). Product purification by HPLC and solid-phase extraction was performed as described for the automated synthesis.
3. Results and discussion [18F]Ciprofloxacin was synthesized manually in moderate yield via a previously described two-step one-pot reaction (Fig. 1) (Langer et al., 2003a,b). The first step comprised a nucleophilic 18 19 F/ F exchange reaction of 7-chloro-1-cyclopropyl-6-fluoro-1,4dihydro-4-oxoquinoline-3-carboxylic acid. Since the 6-position of the carboxylic acid was only moderately activated for nucleophilic substitution high temperature (175 °C) and long reaction times (40 min) were required. As expected the 18F/19F exchange reaction yielded a final product with considerably lower specific activity as
136
J. Stanek et al. / Applied Radiation and Isotopes 99 (2015) 133–137
Fig. 4. Representative semipreparative HPLC chromatogram for the purification of [18F]ciprofloxacin (upper channel¼ UV signal and lower channel ¼radioactivity signal). In this HPLC system radiolabelled intermediate 7-chloro-1-cyclopropyl-6-[18F]fluoro-1,4-dihydro-4-oxoquinoline-3-carboxylic acid remained on the column.
compared with no-carrier-added [18F]fluorinations (Table 1). However, as [18F]ciprofloxacin will be used to study the tissue distribution of ciprofloxacin at therapeutic doses and as ciprofloxacin is a marketed drug with a good safety profile, high specific activity of [18F]ciprofloxacin is not mandatory. In the second step of the synthesis, an in-situ generated boron complex of 7-chloro-1-cyclopropyl-6-[18F]fluoro-1,4-dihydro-4-oxoquinoline3-carboxylic acid (Langer et al., 2003b) was reacted with piperazine to yield [18F]ciprofloxacin. Crude product was purified by semipreparative HPLC (Fig. 4) and solid-phase extraction employing a strong cation exchange cartridge. In our hands, this previously described manual synthesis procedure afforded [18F]ciprofloxacin in a total synthesis time of 130 min in a mean (7 standard deviation) decay-corrected radiochemical yield (RCY) based on starting [18F]fluoride of 5.5 71.0% (n ¼4) with a specific activity at end of synthesis of 0.3 70.2 GBq/mmol (Table 1), which was in good agreement with previously published values (Langer et al., 2003a). We then attempted to transfer the manual synthesis of [18F]ciprofloxacin to a custom-modified dual-reactor synthesis module (TRACERlabTM FXFDG, GE Healthcare) (Figs. 2 and 3), which is similar to the TRACERlabTM FXFN synthesis module, which has been shown to be a versatile synthesis platform for the production of a range of different 18F-labeled radiopharmaceuticals (Shao et al., 2011). Unexpectedly we observed a marked decrease in RCY to only 0.4 70.4% (n ¼5) when employing this automated synthesis module. The most likely explanation for the decrease in RCY as compared with the manual synthesis, in which a 3-mL Wheaton
V-vial was used, was the large-volume glassy carbon reactor (15 mL) of the synthesis module. This reactor had a round bottom (Fig. 3) and the surface area of the reaction solution was approximately two times larger than for the Wheaton V-vial. Owing to a high reaction temperature (175 °C) and long reaction times (40 min), solvent (DMSO) may have evaporated during the reaction and entered the tubing, where it may have condensed and remained as liquid. As a result part of the reaction mixture may have not been in solution but rather adhered to the reactor wall as a solid, which may not have reacted thereby resulting in a drop in RCY. In contrast, in the manual synthesis the entire reaction mixture remained in solution in the Wheaton V-vial in a closed system without any solvent loss. In the automated synthesis, a decrease in reaction temperature resulted in a decrease in incorporation yields while an increase in solvent volume was not possible as injection of 42 mL of DMSO into the semipreparative HPLC system caused [18F]ciprofloxacin to elute with the void volume. Removal of DMSO prior to semipreparative HPLC purification was not practical due to the high boiling point of DMSO (189 °C). Other possible explanations for the lower RCY in the automated synthesis as compared with the manual synthesis may be slightly different concentrations of reagents and lack of stirring in the automated synthesis or different adsorption of reagents and/or product on the walls of the Wheaton V-vial and the glassy carbon reactor. Replacement of one 15-mL glassy carbon reactor in the standard configuration of the synthesis module with a 3-mL V-shaped borosilicate glass reactor, in which the one-pot reaction was
Table 1 Overview of three different methods to synthesize [18F]ciprofloxacin. Method
n
Start activity (GBq)
Product (GBq)a
RCY (%)b
Radiochemical purity (%)a
Specific activity (GBq/mmol)a
Manual Automated (configuration A) Automated (configuration B)
4 5 23
61.6 7 19.9 91.8 7 100.0 134.6 7 21.3
1.2 7 0.6 0.17 0.1 1.3 7 0.6
5.5 71.0 0.4 70.4 2.7 71.4
99.17 0.8 93.3 7 5.5 99.0 7 0.9
0.3 7 0.2 0.17 0.1 1.4 7 0.5
All values are given as mean 7 standard deviation. a b
At end of synthesis. Decay-corrected to end of bombardment and based on starting [18F]fluoride.
J. Stanek et al. / Applied Radiation and Isotopes 99 (2015) 133–137
performed (Fig. 3), allowed for reducing the DMSO volume and enabled to reflux the reaction mixture without significant loss of solvent. This led to a considerable improvement in RCY affording [18F]ciprofloxacin in a mean RCY of 2.7 71.4% (n¼ 23) with a specific activity of 1.4 70.5 GBq/mmol at end of synthesis in a total synthesis time of 160 min (Table 1). To the best of our knowledge, such a dependence of RCY on the reactor type has not been reported before for [18F]fluorinations. As the custom modification of the TRACERlabTM FXFDG synthesis module is not relevant to the synthesis of [18F]ciprofloxacin we expect that our findings will be also applicable to other TRACERlabTM synthesis modules equipped with large-volume glassy carbon reactors, such as the TRACERlabTM FXFN (Shao et al., 2011). Our automated synthesis procedure reliably afforded final [18F]ciprofloxacin amounts, readily formulated for intravenous injection into rodents, of 1.370.6 GBq, which was sufficient to perform four consecutive small-animal PET scans. Osmolality of the formulated product solution was 276 732 mosmol/kg and pH was 6.5 70.3. Radiochemical purity as assessed by analytical HPLC was Z 99%, which stayed Z98%, when the product solution was stored at room temperature for 136 741 min.
4. Conclusion When attempting to transfer a manual labeling method for the synthesis of [18F]ciprofloxacin to an automated synthesis module we observed a strong decrease in RCY as compared to manual labeling. We could show that replacement of one standard largevolume glassy carbon reactor (15 mL) in the automated synthesis module by a small-volume V-shaped borosilicate glass reactor (3 mL), in which the two-step one-pot reaction was performed, led to increased RCYs, affording [18F]ciprofloxacin in acceptable and reproducible amounts and high radiochemical purity for smallanimal PET experiments. We speculate that use of a small-volume V-shaped borosilicate glass reactor might also give higher RCYs for other [18F]radiotracers than [18F]ciprofloxacin, in particular those
137
for which high reaction temperatures and prolonged heating are required. Alternatively, microfluidic synthesis approaches may be of advantage to synthesize such radiotracers (Pascali et al., 2013).
Acknowledgments The research leading to these results has received funding from the Austrian Science Fund (FWF) project “Transmembrane Transporters in Health and Disease” (Grant F 3513-B20, awarded to M. Müller). The authors thank the staff of Seibersdorf Laboratories GmbH for technical assistance.
References Giacomini, K.M., Huang, S.M., Tweedie, D.J., Benet, L.Z., Brouwer, K.L., Chu, X., Dahlin, A., Evers, R., Fischer, V., Hillgren, K.M., Hoffmaster, K.A., Ishikawa, T., Keppler, D., Kim, R.B., Lee, C.A., Niemi, M., Polli, J.W., Sugiyama, Y., Swaan, P.W., Ware, J.A., Wright, S.H., Yee, S.W., Zamek-Gliszczynski, M.J., Zhang, L., 2010. Membrane transporters in drug development. Nat. Rev. Drug Discov. 9, 215–236. Langer, O., Karch, R., Müller, U., Dobrozemsky, G., Abrahim, A., Zeitlinger, M., Lackner, E., Joukhadar, C., Dudczak, R., Kletter, K., Müller, M., Brunner, M., 2005. Combined PET and microdialysis for in vivo assessment of intracellular drug pharmacokinetics in humans. J. Nucl. Med. 46, 1835–1841. Langer, O., Mitterhauser, M., Brunner, M., Zeitlinger, M., Wadsak, W., Mayer, B.X., Kletter, K., Müller, M., 2003a. Synthesis of fluorine-18-labeled ciprofloxacin for PET studies in humans. Nucl. Med. Biol. 30, 285–291. Langer, O., Mitterhauser, M., Wadsak, W., Brunner, M., Müller, U., Kletter, K., Müller, M., 2003b. A general method for the fluorine-18 labelling of fluoroquinolone antibiotics. J. Label. Compd. Radiopharm. 46, 715–727. Mulgaonkar, A., Venitz, J., Sweet, D.H., 2012. Fluoroquinolone disposition: identification of the contribution of renal secretory and reabsorptive drug transporters. Expert Opin. Drug Metab. Toxicol. 8, 553–569. Pascali, G., Watts, P., Salvadori, P.A., 2013. Microfluidics in radiopharmaceutical chemistry. Nucl. Med. Biol. 40, 776–787. Shao, X., Hoareau, R., Hockley, B.G., Tluczek, L.J.M., Henderson, B.D., Padgett, H.C., Scott, P.J.H., 2011. Highlighting the versatility of the tracerlab synthesis modules. Part 1: fully automated production of [18F]labelled radiopharmaceuticals using a Tracerlab FXFN. J. Label. Compd. Radiopharm. 54, 292–307.
Contents lists available at ScienceDirect
Applied Radiation and Isotopes journal homepage: www.elsevier.com/locate/apradiso
Automated radiosynthesis of [18F]ciprofloxacin Johann Stanek a,b, Severin Mairinger a, Thomas Wanek a, Claudia Kuntner a, Markus Müller b, Oliver Langer a,b,n a b
Health and Environment Department, AIT Austrian Institute of Technology GmbH, A-2444 Seibersdorf, Austria Department of Clinical Pharmacology, Medical University of Vienna, Währiger-Gürtel 18-20, A-1090 Vienna, Austria
H I G H L I G H T S
G R A P H I C A L
A B S T R A C T
Automated synthesis of [18F]ciprofloxacin in a TRACERlabTM FXFDG (GE Healthcare) synthesis module was developed. Dependence of radiochemical yield on reactor type was observed. 3-mL V-shaped borosilicate glass reactor gave higher radiochemical yield as compared with standard 15-mL glassy carbon reactor. V-shaped borosilicate glass reactor might also give higher radiochemical yield for other [18F]radiotracers than [18F]ciprofloxacin.
art ic l e i nf o
a b s t r a c t
Article history: Received 19 November 2014 Received in revised form 18 February 2015 Accepted 26 February 2015 Available online 27 February 2015
We transferred the previously published manual synthesis of [18F]ciprofloxacin (decay-corrected RCY: 5.5 71.0%) to an automated synthesis module (TRACERlabTM FXFDG, GE Healthcare) and observed a strong decrease in RCY (0.4 70.4%). When replacing the standard 15-mL glassy carbon reactor of the synthesis module with a 3-mL V-shaped borosilicate glass reactor a considerable improvement in RCY was observed. [18F]Ciprofloxacin was obtained in a RCY of 2.7 71.4% (n ¼23) with a specific activity at EOS of 1.47 0.5 GBq/mmol in a synthesis time of 160 min. & 2015 Elsevier Ltd. All rights reserved.
Keywords: [18F]ciprofloxacin Automated synthesis TRACERlabTM FXFDG Radiochemical yield Reactor type 18 19 F/ F exchange
1. Introduction Ciprofloxacin (1-cyclopropyl-6-fluoro-1,4-dihydro-4-oxo-7-(1piperazinyl)-quinolone-3-carboxylic acid) is a still widely prescribed fluoroquinolone antibiotic which displays a broad spectrum of activity against Gram-negative and Gram-positive bacteria.
n Corresponding author at: Health and Environment Department, AIT Austrian Institute of Technology GmbH, A-2444 Seibersdorf, Austria. Fax: þ43 50550 2136. E-mail address: [email protected] (O. Langer).
http://dx.doi.org/10.1016/j.apradiso.2015.02.024 0969-8043/& 2015 Elsevier Ltd. All rights reserved.
Ciprofloxacin has been described as a substrate of several different active membrane transporter proteins belonging to the solute carrier (SLC) or adenosine triphosphate-binding cassette (ABC) families, such as organic anion transporter 3 (OAT3), breast cancer resistance protein (BCRP), multidrug resistance protein 4 (MRP4) and multidrug and toxin extrusion protein 1 (MATE1) (Mulgaonkar et al., 2012). These transporters mediate hepatobiliary and renal excretion of their substrates and drug–drug interactions involving inhibition of one or several of these transporters could lead to altered tissue distribution of ciprofloxacin, which may have an impact
134
J. Stanek et al. / Applied Radiation and Isotopes 99 (2015) 133–137
2.2. Automated synthesis of [18F]ciprofloxacin
Fig. 1. Reaction scheme for the synthesis of [18F]ciprofloxacin.
on its efficacy and safety (Giacomini et al., 2010). Ciprofloxacin contains fluorine in its structure which can be replaced with the positron-emitting radionuclide fluorine-18 (18F, half-life 109.8 min) (Langer et al., 2003a,b). We have previously shown that positron emission tomography (PET) imaging with [18F]ciprofloxacin can be used to non-invasively study the tissue distribution of ciprofloxacin in humans (Langer et al., 2005). Studying the tissue distribution of [18F]ciprofloxacin in genetically modified mice, which lack one or several membrane transporter proteins, may help to identify the relevant transporters, which are likely to affect ciprofloxacin tissue distribution in humans. In previous studies, ciprofloxacin has been prepared by a manual two-step one-pot synthesis method comprising a nucleophilic 18F/19F exchange reaction of 7-chloro-1-cyclopropyl-6fluoro-1,4-dihydro-4-oxoquinoline-3-carboxylic acid followed by reaction of the 18F-labeled carboxylic acid derivative with piperazine (Fig. 1) (Langer et al., 2003a,b). Aim of this work was to develop an automated synthesis of [18F]ciprofloxacin in a commercially available automated synthesis module (TRACERlabTM FXFDG, General Electric Healthcare, Uppsala, Sweden) in order to reduce the radiation exposure associated with manual radiosynthesis.
2. Materials and methods 2.1. General All chemicals were purchased from Sigma-Aldrich Handels GmbH (Vienna, Austria) and used without further purification. Aqueous (aq.) [18F]fluoride was produced using a PETtrace cyclotron (GE Healthcare) via the 18O(p,n)18F nuclear reaction by irradiation of a 2.6 mL water target containing 95.9% enriched [18O]water (ABX-advanced biochemical compounds, Radeberg, Germany) with a 16.5 MeV proton beam. Radiochemical purity and specific activity of [18F]ciprofloxacin were determined with analytical high-performance liquid chromatography (HPLC) using an Agilent 1200 system (Agilent Technologies Österreich GmbH, Vienna, Austria) consisting of a quaternary pump, an auto-sampler and a column oven. Ultraviolet (UV) absorption was detected with an Agilent 1200 diode array detector at a wavelength of 280 nm in series with a Raytest “Gabi Star” detector (raytest Isotopenmessgeräte GmbH, Straubenhardt, Germany) for radioactivity detection. A PRP-1 column (4.1 250 mm2, 5 mm, Hamilton Bonaduz AG, Bonaduz, Switzerland), heated to 40 °C, was isocratically eluted with a 85/15 (v/v) mixture of aq. 0.01 M phosphoric acid (H3PO4) and ethanol at a flow rate of 1 mL/min. Osmolality (mosmol/kg) of formulated [18F]ciprofloxacin solution was measured using a Wescor Vapro 5520 Pressure Osmometer (Wescor Inc., Logan, USA). The pH value was determined with a pH-Meter Inolab pH720 (WTW Wissenschaftlich-Technische Werkstätten GmbH, Weilheim, Germany).
Radiosynthesis of [18F]ciprofloxacin was performed as a twostep one-pot reaction (Fig. 1) in a custom-modified dual-reactor TRACERlabTM FXFDG synthesis module (GE Healthcare). This synthesis module originally contained two independent synthesis units in a single housing and was originally designed to operate two consecutive [18F]FDG syntheses without reloading and opening the hot cell. Our synthesis module was custom-modified by the manufacturer to enable two-pot [18F]radiosyntheses (Fig. 2). We used the synthesis module either in standard configuration with two large-volume glassy carbon reactors (15 mL) (configuration A) or in a modified configuration in which one large-volume glassy carbon reactor (reactor 1) had been replaced with a small-volume V-shaped borosilicate glass reactor (3 mL) (configuration B, Fig. 3). In configuration B also the heating block of the reactor was replaced so that the bore size fitted to the smaller diameter of the reactor (Fig. 3, left picture). In order to facilitate reflux of the reaction solution in configuration B the 3-mL V-shaped borosilicate glass reactor was mounted in a way that 1.5 cm remained outside the heating block. In both configurations the two reaction steps of the [18F]ciprofloxacin synthesis were performed in reactor 1 (Figs. 2 and 3). All reagent solutions were prepared prior to start of synthesis and placed in the synthesis module storage vessels. Amounts of reagents and solvent volumes were different for configurations A and B as indicated below. After delivery of the irradiated [18O]water to the synthesis module, [18F]fluoride was trapped on an anion exchange cartridge (PS-HCO3, 45 mg, Macherey-Nagel, Düren, Germany), which had been pre-activated with ethanol (3 mL) and water (5 mL). [18F]Fluoride was eluted into the synthesis reactor by rinsing the cartridge with a mixture of kryptofix 2.2.2 (4,7,13,16,21,24-hexaoxa-1,10-diazabicyclo[8.8.8] hexacosane, configuration A: 16 mg, 42.5 mmol; configuration B: 12 mg, 31.9 mmol) in acetonitrile (0.9 mL) and potassium carbonate (3.5 mg, 25.3 μmol) in water (0.1 mL) (added via valve 1, V1, see Fig. 2). After adding acetonitrile (0.5 mL) to the synthesis reactor (via V2), the solvent was evaporated azeotropically under vacuum, first for 3 min at 60 °C and then for further 5 min at 120 °C to remove the remaining water. To the dried K[18F]F-K222 complex, radiolabelling precursor 7-chloro1-cyclopropyl-6-fluoro-1,4-dihydro-4-oxoquinoline-3-carboxylic acid (configuration A: 12.0 mg, 42.6 mmol; configuration B: 6.0 mg, 21.3 mmol) dissolved in dimethyl sulfoxide (DMSO, configuration A: 0.8 mL; configuration B: 0.3 mL) was added (V6) and the reaction mixture was heated at 175 °C for 40 min. The reaction mixture was then cooled to 40 °C and a mixture of trimethylborate (configuration A: 100 mL, 897 mmol; configuration B: 20 mL, 179 mmol) and glacial acetic acid (configuration A: 100 mL, 1747 mmol; configuration B: 20 mL, 349 mmol) in DMSO (configuration A: 0.3 mL; configuration B: 0.1 mL) was added (V5) and the reaction mixture was stirred at 40 °C for 2 min. Then a solution of piperazine (configuration A: 57 mg, 661.7 mmol; configuration B: 20 mg, 232.2 mmol) in DMSO (configuration A: 0.7 mL; configuration B: 0.35 mL) was added (V4) and the mixture was reacted at 175 °C for 40 min. After cooling to 40 °C, a mixture of aq. 0.01 M H3PO4 and ethanol (85/15, v/v, configuration A: 2.5 mL; configuration B: 1 mL) was added (V3) and the crude reaction mixture was injected into the built-in semipreparative HPLC system. A Hamilton PRP-1 column (10 250 mm2, 10 mm) equipped with a PRP-1 guard column (10 40 mm2) was eluted at a flow rate of 3 mL/min for the first 12 min with a mixture of aq. 0.01 M H3PO4 and ethanol (97/3, v/v) followed by an increase of the ethanol percentage to 15%. The HPLC eluate was monitored in series for radioactivity and UV absorption at a wavelength of 280 nm. [18F]Ciprofloxacin, which eluted with a retention time of 25–30 min in a volume of 15– 20 mL (Fig. 4), was directly passed over a strong cation exchange
J. Stanek et al. / Applied Radiation and Isotopes 99 (2015) 133–137
135
Fig. 2. Scheme of the automated synthesis module for the synthesis of [18F]ciprofloxacin. The module is a dual-reactor TRACERlabTM FXFDG synthesis module, which has been custom-modified to enable two-pot radiosyntheses.
Fig. 3. Pictures of dual-reactor synthesis module in modified configuration with one V-shaped borosilicate glass reactor (reactor 1) and one 15-mL glassy carbon reactor (reactor 2). The one-pot radiosynthesis was performed in reactor 1. Note that the heating block of reactor 1 was replaced so that the bore size fitted to the smaller diameter of the reactor.
cartridge (Isolute 100 mg SCX, International Sorbent Technology Ltd., Hengoed, UK), which had been pre-activated with ethanol (5 mL) and mobile phase for semipreparative HPLC (5 mL). The cartridge was washed with water (5 mL, V8) and [18F]ciprofloxacin was then eluted from the cartridge with 0.1 M aq. sodium hydroxide solution (2.5 mL, V7). The final product was formulated for intravenous injection into rodents by adding sodium dihydrogenphosphate dihydrate (45 mg, 288 mmol) and aq. sodium chloride solution (10%, w/v, 50 mL). 2.3. Manual synthesis of [18F]ciprofloxacin The manual synthesis of [18F]ciprofloxacin was performed as published previously (Langer et al., 2003a,b) using a 3-mL Wheaton V-vial (clear type I borosilicate glass, Wheaton Science Products, Millville, USA) equipped with a polytetrafluoroethylene/silicone septum and a magnetic stirrer. Amounts of reagents and
solvent volumes were identical with the automated synthesis (configuration B). Product purification by HPLC and solid-phase extraction was performed as described for the automated synthesis.
3. Results and discussion [18F]Ciprofloxacin was synthesized manually in moderate yield via a previously described two-step one-pot reaction (Fig. 1) (Langer et al., 2003a,b). The first step comprised a nucleophilic 18 19 F/ F exchange reaction of 7-chloro-1-cyclopropyl-6-fluoro-1,4dihydro-4-oxoquinoline-3-carboxylic acid. Since the 6-position of the carboxylic acid was only moderately activated for nucleophilic substitution high temperature (175 °C) and long reaction times (40 min) were required. As expected the 18F/19F exchange reaction yielded a final product with considerably lower specific activity as
136
J. Stanek et al. / Applied Radiation and Isotopes 99 (2015) 133–137
Fig. 4. Representative semipreparative HPLC chromatogram for the purification of [18F]ciprofloxacin (upper channel¼ UV signal and lower channel ¼radioactivity signal). In this HPLC system radiolabelled intermediate 7-chloro-1-cyclopropyl-6-[18F]fluoro-1,4-dihydro-4-oxoquinoline-3-carboxylic acid remained on the column.
compared with no-carrier-added [18F]fluorinations (Table 1). However, as [18F]ciprofloxacin will be used to study the tissue distribution of ciprofloxacin at therapeutic doses and as ciprofloxacin is a marketed drug with a good safety profile, high specific activity of [18F]ciprofloxacin is not mandatory. In the second step of the synthesis, an in-situ generated boron complex of 7-chloro-1-cyclopropyl-6-[18F]fluoro-1,4-dihydro-4-oxoquinoline3-carboxylic acid (Langer et al., 2003b) was reacted with piperazine to yield [18F]ciprofloxacin. Crude product was purified by semipreparative HPLC (Fig. 4) and solid-phase extraction employing a strong cation exchange cartridge. In our hands, this previously described manual synthesis procedure afforded [18F]ciprofloxacin in a total synthesis time of 130 min in a mean (7 standard deviation) decay-corrected radiochemical yield (RCY) based on starting [18F]fluoride of 5.5 71.0% (n ¼4) with a specific activity at end of synthesis of 0.3 70.2 GBq/mmol (Table 1), which was in good agreement with previously published values (Langer et al., 2003a). We then attempted to transfer the manual synthesis of [18F]ciprofloxacin to a custom-modified dual-reactor synthesis module (TRACERlabTM FXFDG, GE Healthcare) (Figs. 2 and 3), which is similar to the TRACERlabTM FXFN synthesis module, which has been shown to be a versatile synthesis platform for the production of a range of different 18F-labeled radiopharmaceuticals (Shao et al., 2011). Unexpectedly we observed a marked decrease in RCY to only 0.4 70.4% (n ¼5) when employing this automated synthesis module. The most likely explanation for the decrease in RCY as compared with the manual synthesis, in which a 3-mL Wheaton
V-vial was used, was the large-volume glassy carbon reactor (15 mL) of the synthesis module. This reactor had a round bottom (Fig. 3) and the surface area of the reaction solution was approximately two times larger than for the Wheaton V-vial. Owing to a high reaction temperature (175 °C) and long reaction times (40 min), solvent (DMSO) may have evaporated during the reaction and entered the tubing, where it may have condensed and remained as liquid. As a result part of the reaction mixture may have not been in solution but rather adhered to the reactor wall as a solid, which may not have reacted thereby resulting in a drop in RCY. In contrast, in the manual synthesis the entire reaction mixture remained in solution in the Wheaton V-vial in a closed system without any solvent loss. In the automated synthesis, a decrease in reaction temperature resulted in a decrease in incorporation yields while an increase in solvent volume was not possible as injection of 42 mL of DMSO into the semipreparative HPLC system caused [18F]ciprofloxacin to elute with the void volume. Removal of DMSO prior to semipreparative HPLC purification was not practical due to the high boiling point of DMSO (189 °C). Other possible explanations for the lower RCY in the automated synthesis as compared with the manual synthesis may be slightly different concentrations of reagents and lack of stirring in the automated synthesis or different adsorption of reagents and/or product on the walls of the Wheaton V-vial and the glassy carbon reactor. Replacement of one 15-mL glassy carbon reactor in the standard configuration of the synthesis module with a 3-mL V-shaped borosilicate glass reactor, in which the one-pot reaction was
Table 1 Overview of three different methods to synthesize [18F]ciprofloxacin. Method
n
Start activity (GBq)
Product (GBq)a
RCY (%)b
Radiochemical purity (%)a
Specific activity (GBq/mmol)a
Manual Automated (configuration A) Automated (configuration B)
4 5 23
61.6 7 19.9 91.8 7 100.0 134.6 7 21.3
1.2 7 0.6 0.17 0.1 1.3 7 0.6
5.5 71.0 0.4 70.4 2.7 71.4
99.17 0.8 93.3 7 5.5 99.0 7 0.9
0.3 7 0.2 0.17 0.1 1.4 7 0.5
All values are given as mean 7 standard deviation. a b
At end of synthesis. Decay-corrected to end of bombardment and based on starting [18F]fluoride.
J. Stanek et al. / Applied Radiation and Isotopes 99 (2015) 133–137
performed (Fig. 3), allowed for reducing the DMSO volume and enabled to reflux the reaction mixture without significant loss of solvent. This led to a considerable improvement in RCY affording [18F]ciprofloxacin in a mean RCY of 2.7 71.4% (n¼ 23) with a specific activity of 1.4 70.5 GBq/mmol at end of synthesis in a total synthesis time of 160 min (Table 1). To the best of our knowledge, such a dependence of RCY on the reactor type has not been reported before for [18F]fluorinations. As the custom modification of the TRACERlabTM FXFDG synthesis module is not relevant to the synthesis of [18F]ciprofloxacin we expect that our findings will be also applicable to other TRACERlabTM synthesis modules equipped with large-volume glassy carbon reactors, such as the TRACERlabTM FXFN (Shao et al., 2011). Our automated synthesis procedure reliably afforded final [18F]ciprofloxacin amounts, readily formulated for intravenous injection into rodents, of 1.370.6 GBq, which was sufficient to perform four consecutive small-animal PET scans. Osmolality of the formulated product solution was 276 732 mosmol/kg and pH was 6.5 70.3. Radiochemical purity as assessed by analytical HPLC was Z 99%, which stayed Z98%, when the product solution was stored at room temperature for 136 741 min.
4. Conclusion When attempting to transfer a manual labeling method for the synthesis of [18F]ciprofloxacin to an automated synthesis module we observed a strong decrease in RCY as compared to manual labeling. We could show that replacement of one standard largevolume glassy carbon reactor (15 mL) in the automated synthesis module by a small-volume V-shaped borosilicate glass reactor (3 mL), in which the two-step one-pot reaction was performed, led to increased RCYs, affording [18F]ciprofloxacin in acceptable and reproducible amounts and high radiochemical purity for smallanimal PET experiments. We speculate that use of a small-volume V-shaped borosilicate glass reactor might also give higher RCYs for other [18F]radiotracers than [18F]ciprofloxacin, in particular those
137
for which high reaction temperatures and prolonged heating are required. Alternatively, microfluidic synthesis approaches may be of advantage to synthesize such radiotracers (Pascali et al., 2013).
Acknowledgments The research leading to these results has received funding from the Austrian Science Fund (FWF) project “Transmembrane Transporters in Health and Disease” (Grant F 3513-B20, awarded to M. Müller). The authors thank the staff of Seibersdorf Laboratories GmbH for technical assistance.
References Giacomini, K.M., Huang, S.M., Tweedie, D.J., Benet, L.Z., Brouwer, K.L., Chu, X., Dahlin, A., Evers, R., Fischer, V., Hillgren, K.M., Hoffmaster, K.A., Ishikawa, T., Keppler, D., Kim, R.B., Lee, C.A., Niemi, M., Polli, J.W., Sugiyama, Y., Swaan, P.W., Ware, J.A., Wright, S.H., Yee, S.W., Zamek-Gliszczynski, M.J., Zhang, L., 2010. Membrane transporters in drug development. Nat. Rev. Drug Discov. 9, 215–236. Langer, O., Karch, R., Müller, U., Dobrozemsky, G., Abrahim, A., Zeitlinger, M., Lackner, E., Joukhadar, C., Dudczak, R., Kletter, K., Müller, M., Brunner, M., 2005. Combined PET and microdialysis for in vivo assessment of intracellular drug pharmacokinetics in humans. J. Nucl. Med. 46, 1835–1841. Langer, O., Mitterhauser, M., Brunner, M., Zeitlinger, M., Wadsak, W., Mayer, B.X., Kletter, K., Müller, M., 2003a. Synthesis of fluorine-18-labeled ciprofloxacin for PET studies in humans. Nucl. Med. Biol. 30, 285–291. Langer, O., Mitterhauser, M., Wadsak, W., Brunner, M., Müller, U., Kletter, K., Müller, M., 2003b. A general method for the fluorine-18 labelling of fluoroquinolone antibiotics. J. Label. Compd. Radiopharm. 46, 715–727. Mulgaonkar, A., Venitz, J., Sweet, D.H., 2012. Fluoroquinolone disposition: identification of the contribution of renal secretory and reabsorptive drug transporters. Expert Opin. Drug Metab. Toxicol. 8, 553–569. Pascali, G., Watts, P., Salvadori, P.A., 2013. Microfluidics in radiopharmaceutical chemistry. Nucl. Med. Biol. 40, 776–787. Shao, X., Hoareau, R., Hockley, B.G., Tluczek, L.J.M., Henderson, B.D., Padgett, H.C., Scott, P.J.H., 2011. Highlighting the versatility of the tracerlab synthesis modules. Part 1: fully automated production of [18F]labelled radiopharmaceuticals using a Tracerlab FXFN. J. Label. Compd. Radiopharm. 54, 292–307.
