Synthesis of 3′-deoxy-3′-fluorothymidine (FLT) 5′-O-glucuronide: A reference standard for imaging studies with [ 18F]FLT

Article abstract of DOI:10.1039/c3md00381g

Reaction of methyl 2,3,4-tri-O-acetyl-1-O-(trichloroacetimidoyl)-α-d- glucopyranuronate with 3′-deoxy-3′-fluorothymidine in the presence of trimethylsilyl trifluoromethanesulfonate gave (2R,3R,4S,5S,6S)-2-(((2R,3S,5R) -3-fluoro-5-(5-methyl-2,4-dioxo-3,4-dihydropyrimidin-1(2H)-yl) tetrahydrofuran-2-yl)methoxy)-6-(methoxycarbonyl)tetrahydro-2H-pyran-3,4, 5-triyl triacetate, which was hydrolysed to 3′-deoxy-3′- fluorothymidine-5′-glucuronide. Analysis of this reference standard by high performance liquid chromatography enabled the identification of [ ¹⁸F]3′-deoxy-3′-fluorothymidine-5′-glucuronide in blood samples from six human patients who had been administered [ ¹⁸F]3′-deoxy-3′-fluorothymidine.

Full text of DOI:10.1039/c3md00381g

MedChemComm
View Article Online
View Journal | View Issue
CONCISE ARTICLE
Synthesis of 30-deoxy-30-fluorothymidine (FLT) 50-
O-glucuronide: a reference standard for imaging
studies with [18F]FLT†
Cite this: Med. Chem. Commun., 2014,
5, 984
Suzannah J. Harnor,a Tommy Rennison,a Martin Galler,b Celine Cano,a
a
b
a
*
*
Roger J. Griffin, David R. Newell and Bernard T. Golding
Reaction of methyl 2,3,4-tri-O-acetyl-1-O-(trichloroacetimidoyl)-a-D-glucopyranuronate with 30-deoxy-
30-fluorothymidine in the presence of trimethylsilyl trifluoromethanesulfonate gave (2R,3R,4S,5S,6S)-
2-(((2R,3S,5R)-3-fluoro-5-(5-methyl-2,4-dioxo-3,4-dihydropyrimidin-1(2H)-yl)tetrahydrofuran-2-yl)-
methoxy)-6-(methoxycarbonyl)tetrahydro-2H-pyran-3,4,5-triyl triacetate, which was hydrolysed to 30-
deoxy-30-fluorothymidine-50-glucuronide. Analysis of this reference standard by high performance liquid
chromatography enabled the identification of [18F]30-deoxy-30-fluorothymidine-50-glucuronide in blood
samples from six human patients who had been administered [18F]30-deoxy-30-fluorothymidine.
Received 13th December 2013
Accepted 25th May 2014
DOI: 10.1039/c3md00381g
www.rsc.org/medchemcomm
b-glucuronidase treatment, 5–15% of plasma 18F was present as
the metabolite at 60 min, and metabolite correction was per-
formed to generate accurate PET image data.1 As part of a
continuing programme to dene the metabolism and distri-
bution of FLT in the body, we have synthesised the 50-glucuro-
nide 2 to provide a denitive metabolic standard.
Introduction
Positron emission tomography (PET) is a key and rapidly
expanding clinical imaging technology, which involves the
systemic administration of a tracer or ligand labelled with a
positron-emitting isotope. Within the body, the positrons
released are annihilated on collision with electrons releasing
two photons of g-radiation, which are detected by placing the
patient in a PET camera. Signal processing allows construction
of tomographic images that are valuable in the diagnosis and
treatment of a wide range of diseases, including cancer and
neurological disorders. The nucleoside analogue 30-deoxy-30-
uorothymidine (FLT, 1), in its 18F-labelled form, [18F]FLT, has
proven to be a valuable biomarker for imaging tumour prolif-
eration by PET. In interpreting PET data, it is important to know
the extent of systemic tracer metabolism, as the metabolites will
contribute to the PET signal, but may not measure the same
process as the parent tracer. Formation of 30-deoxy-30-uoro-
thymidine-50-glucuronide (FLT-50-glucuronide 2, Scheme 1) was
proposed as a major route of metabolism for nucleoside 1,
based on the observation that b-glucuronidase destroyed the
The glucuronidation of drugs and other xenobiotics is a
crucial detoxication process during phase II metabolism.2,3
Thus, glucuronidation is used by humans and animals to assist
the excretion of toxic substances and drugs via the kidneys or in
the bile by the liver. Numerous papers4–9 have described the
synthesis of glucuronides of alcohols and phenols, but the
literature provides limited guidance on which of many pub-
lished procedures is guaranteed to afford success when applied
to a new target. The chemical synthesis of glucuronide 2 has not
been reported. Our objective was to develop an efficient
synthetic route for the preparation of 2, for use as an HPLC
standard in the analysis of plasma samples from patients
undergoing [18F]FLT PET scanning. We report a reliable proce-
dure for the synthesis of glucuronide 2, which should be
applicable to a range of alcohols and phenols. We also report a
potential trap in the use of one published method, which when
applied to FLT, gave primarily an orthoester, isomeric with the
desired protected glucuronide. The application of the synthetic
FLT-50-glucuronide 2 in PET is also described.
metabolite.1 In
5
patients, based on sensitivity to
aNewcastle Cancer Centre, Northern Institute for Cancer Research, School of
Chemistry, Newcastle University, Bedson Building, Newcastle upon Tyne, NE1 7RU,
UK. E-mail: bernard.golding@newcastle.ac.uk; Fax: +44 (0)191 2226929; Tel: +44
(0)191 2226647
bNewcastle Cancer Centre, Northern Institute for Cancer Research, Medical School,
Newcastle University, Paul O'Gorman Building, Framlington Place, Newcastle upon
Tyne, NE2 4HH, UK. E-mail: herbie.newell@ncl.ac.uk; Fax: +44 (0)191 2464301;
Tel: +44 (0)191 2464400
† DRN conceived this study; BTG, SJH and DRN wrote the manuscript; SJH and TR
performed the chemical synthesis supervised by CC, BTG and RJG; MG performed
the HPLC analysis.
Scheme 1 Glucuronidation of 30-deoxy-30-fluorothymidine (FLT).
984 | Med. Chem. Commun., 2014, 5, 984–988
This journal is © The Royal Society of Chemistry 2014

View Article Online
Concise Article
MedChemComm
Results and discussion
A synthesis envisaged for FLT-50-glucuronide 2 used commer-
cially available uorothymidine 1 and the popular glucur-
onidating agent acetobromo-a-D-glucuronic acid methyl ester 3.
Several attempts to synthesise the desired glucuronide were
made using different reaction conditions with 3. Reaction of
ester 3 with Ag2CO3 and quinoline, in the presence of freshly
10
˚
activated 4 A molecular sieves (MS) [Koenigs–Knorr method ]
gave no reaction. A second attempt employed NaH as the base,
Scheme 3 Synthesis of FLT-50-glucuronide. Reagents and conditions:
(i) Ag2CO3, H2O, acetone–water (4 : 1), room temperature, 24 h, 84%;
(ii) CCl3CN, Cs2CO3, CH2Cl2, room temperature, 2 h, 82%; (iii) TMSOTf,
4 A˚ MS, CH2Cl2, ꢁ30 ꢀC–room temperature, 22 h, 23%; (iv) 2 M KOH
(aq.), THF–H2O (4 : 1), 0 ꢀC–room temperature, 2 h, 97%.
˚
as well as Ag2CO3, with the addition of 4 A MS. Due to the poor
solubility of the starting materials, the reaction was carried out
in THF at 0 ꢀC, with slow warming to ambient temperature.
Again, no reaction occurred, which was also the case when NIS
11
˚
and 4 A MS in 1,2-dichloroethane were used.
Partial success was achieved when silver tri-
giving 30-deoxy-30-uorothymidine-50-glucuronide 2 in 97%
yield aer semi-preparative HPLC. The protocol described is
also a reliable method for the synthesis of glucuronides of a
variety of phenolic drugs (B. T. Golding, unpublished results).
To illustrate the utility of the synthetic [19F]FLT-glucuronide,
the material was used to establish HPLC conditions for the
separation of [19F]FLT and [19F]FLT-glucuronide (see Fig. 1,
Chromatogram A). Crucially, the retention times were very
different: 2.0 minutes for [19F]FLT and 1.1 minutes for [19F]FLT-
glucuronide (see below). The method was then applied to the
analysis of levels of [18F]FLT and [18F]FLT-glucuronide in an
extract of blood from a patient given FLT as part of a study to
evaluate the utility of FLT PET as a surrogate response
biomarker in pancreatic cancer. Following the administration
of a single injection dose of 196 MBq of [18F]FLT to a patient, a
blood sample was taken 30 minutes post injection and analysed
using HPLC with radiochemical detection. Chromatogram B
clearly shows [18F]FLT and [18F]FLT-glucuronide in the clinical
sample.
uoromethanesulfonate (AgOTf) was employed,12 along with
2,6-lutidine and 4 A MS in toluene/CH2Cl2.8,13,14 The triate was
˚
ꢀ
added at ꢁ36 C and then the reaction mixture was allowed to
warm to ambient temperature. The desired product was
apparently detected by LC-MS, but the reaction gave several by-
products and no desired product could be isolated. The solvent
was changed from toluene/CH2Cl2 to THF and the reaction was
carried out at low temperature (ꢁ36 ꢀC) with exclusion of
moisture, air and light. The reaction progressed, albeit slowly,
to give 50% of isolated product following column purication.
Although the desired mass was detected by LC-MS, the 1H NMR
spectrum lacked one expected acetate resonance around d 2.0,
which was replaced by a three-proton singlet at d 1.7. With this
observation, and taking into account the detected correct mass,
this product was identied as orthoester 4 (Scheme 2).
In an alternative approach, acetobromo-a-D-glucuronic acid
methyl ester 3 was converted to the hemiacetal 5 [84% yield of a
3 : 1 (a : b) mixture (NMR analysis)] using Ag2CO3 catalysed
hydrolysis.15 Although hemiacetal 5 is itself a useful glycosyl
donor, the compound was reacted with CCl3CN using Cs2CO3 as
base to give the a-trichloroacetimidate 6 (Scheme 3).6,16
[18F]FLT and
[
18F]FLT-glucuronide concentrations were
analysed in the blood plasma of six patients on 10 occasions,
four patients being studied aer two separate [18F]FLT PET
scans. The HPLC retention time for [18F]FLT was within the
range 1.6–2.2 minutes, and the retention time for [18F]FLT-
glucuronide was within the range 1.05–1.12 minutes. The HPLC
retention times for the analysis of non-radioactive FLT and
The trichloroacetimidate was used as an alternative glucur-
ꢀ
onidation reagent for 1, with TMSOTf as catalyst, at ꢁ30 C to
room temperature, to generate the desired protected [19F]FLT-
glucuronide 7 (23% aer semi-preparative HPLC purication)
with no anomer or orthoester formation, although there was
signicant recovered uorothymidine 1. It was found that
unreacted FLT co-eluted with the product upon ‘ash’ column
purication, so it was necessary to perform semi-preparative
HLPC in order to obtain the pure product. Global deprotection
was achieved, as before, using 2 M KOH (aq.) in THF–H2O,
Fig. 1 HPLC analysis of FLT and FLT-glucuronide. (A) Chromatogram
identifying the retention times of FLT and FLT-glucuronide. (B)
Chromatogram showing the detection of [18F]FLT and [18F]FLT-
Scheme 2 Formation of the orthoester. Reagents and conditions: (i) glucuronide in a plasma sample prepared 30 min after an [18F]FLT dose
AgOTf, 2,6-lutidine, 4 A˚ MS, THF, ꢁ36 ꢀC–room temperature, 20 h, of 196 MBq [the glucuronide peak represents 13.6% of the total activity
50%.
recorded for FLT and FLT-glucuronide (FLT accounting for 86.4%)].
This journal is © The Royal Society of Chemistry 2014
Med. Chem. Commun., 2014, 5, 984–988 | 985

View Article Online
MedChemComm
Concise Article
FLT-glucuronide standards were in the range 1.99–2.01 and [petrol–ethyl acetate (100 : 0) / (80 : 20) / (50 : 50)] of the
1.07–1.09 minutes, respectively, and hence the retentions times crude residue afforded methyl 2,3,4-tri-O-acetyl-a,b-glucopyr-
of the FLT and FLT-glucuronide standards and the radioactive anuronate 5 as a white solid (84%), ratio of a to b anomers (by
1
H NMR) ¼ 75 : 25 (a : b); mp 89.2–90.5 C, lit.15 91–92 C; IR
ꢀ
ꢀ
peaks detected in the blood plasma of patients are consistent,
although in the case of [18F]FLT in blood plasma the HPLC (cmꢁ1) 1713, 1734 (C]O), 3317 (O–H); m/z (ES): 357 (M+ + Na);
retention time was more variable.
1H NMR (CDCl3) (a-anomer) d 2.05 (3H, s, OAc), 2.06 (3H, s,
OAc), 2.11 (3H, s, OAc), 3.36 (1H, d, J ¼ 4.0 Hz, OH), 3.77 (3H, s,
CO2CH3), 4.61 (1H, d, J ¼ 10.1 Hz, H-5), 4.94 (1H, dd, J ¼ 3.6 and
10.1 Hz, H-2), 5.21 (1H, t, J ¼ 9.4 Hz, H-4), 5.57 (1H, d, J ¼ 3.8 Hz,
Conclusions
The de novo synthesis of [19F]FLT-glucuronide 2 was successfully H-1) and 5.60 (1H, t, J ¼ 9.7 Hz, H-3); 13C NMR (CDCl3) (a-
achieved in four steps, through glycosylation of FLT 1 with the anomer) d 20.6 (OAc), 20.7 (2 ꢂ OAc), 52.9 (CO2CH3), 68.1 (C–H),
trichloroacetimidate of a suitably protected glucuronic acid. 69.0 (C–H), 69.5 (C–H), 70.7 (C–H), 90.3 (C-1), 168.3 (C]O),
The synthetic FLT-glucuronide was used to dene unequivocally 169.7 (C]O), 170.0 (C]O) and 170.1 (C]O); 1H NMR (500
the HPLC characteristics of the compound and thereby MHz, CDCl3) (b-anomer) d 2.05 (3H, s, OAc), 2.07 (3H, s, OAc),
demonstrate the presence of the metabolite in blood samples 2.11 (3H, s, OAc), 3.78 (3H, s, CO2CH3), 4.15 (1H, d, J ¼ 10.0 Hz,
from patients administered [18F]FLT.
H-5), 4.35 (1H, br d, OH), 4.82 (1H, t, J ¼ 8.1 Hz, H-2), 4.94 (1H,
m, H-1), 5.25 (1H, t, J ¼ 9.5 Hz, H-4) and 5.33 (1H, t, J ¼ 9.4 Hz,
H-3); 13C NMR (125.8 MHz, CDCl3) (b-anomer) d 20.5 (OAc), 20.6
(OAc), 20.6 (OAc), 53.1 (CO2CH3), 69.4 (C–H), 71.4 (C–H), 72.7
(C–H), 73.0 (C–H), 95.6 (C-1), 167.5 (C]O), 169.5 (C]O), 170.2
Experimental
Chemistry materials and methods
Chemicals and solvents were obtained from reputable (C]O) and 170.8 (C]O).
suppliers. Acetobromo-a-D-glucuronic acid methyl ester 3 was
Methyl 2,3,4-tri-O-acetyl-1-O-(trichloroacetimidoyl)-a-D-glu-
purchased from Apollo Scientic (95+%). Solvents were either copyranuronate
6.
Methyl
2,3,4-triacetyl-a,b-gluco-pyr-
dried by standard techniques or purchased as anhydrous. Petrol anuronate 5 (386 mg, 1.15 mmol) was dissolved in anhydrous
was reagent grade (bp range 40–60 ꢀC). All reactions that CH2Cl2 (6 mL) and CCl3CN (1.2 mL, 11.5 mmol) and then
required inert or dry atmosphere were carried out under dry Cs2CO3 (67 mg, 0.207 mmol) were added. The mixture was
nitrogen. Glassware was dried in an oven prior to use. Column stirred at room temperature for 2 h and then passed through a
chromatography was carried out either using 40–60 mm mesh pad of silica, which was washed with petrol–ethyl acetate
silica in glass columns under medium pressure (MPC) or with a (50 : 50). The solvent was removed in vacuo, following which,
Biotage SP4 ash purication system using KP-SilTM silica. Thin MPC [petrol–ethyl acetate (80 : 20) / (50 : 50)] of the crude
layer chromatography (TLC) was performed on 20 mm pre- residue
afforded
methyl
2,3,4-tri-O-acetyl-1-O-(tri-
coated plates of silica gel (Merck, silica gel 60F254); visual- chloroacetimidoyl)-a-D-gluco-pyranuronate 6 (449 mg, 82%) as a
isation was made using ultraviolet light (254 nm) or by staining white solid; mp 108.6–110.2 ꢀC, lit.7 108 ꢀC; IR (cmꢁ1) 1737
with anisaldehyde. Semi-preparative HPLC was performed with (C]O), 2958 (C–H), 3349 (N–H); 1H NMR (CDCl3) d 2.01 (3H, s,
an Agilent 1200 Preparative HPLC machine using a Waters OAc), 2.04 (3H, s, OAc), 2.05 (3H, s, OAc), 3.74 (3H, s, CO2CH3),
XTerra RP18 column (5 mm, 19 ꢂ 150 mm) eluting with 0.1% 4.49 (1H, d, J ¼ 10 Hz, H-5), 5.14 (1H, dd, J ¼ 4.0 Hz and 10.0 Hz,
formic acid in water–acetonitrile gradient. All melting points H-2), 5.27 (1H, t, J ¼ 10.0 Hz, H-4), 5.62 (1H, t, J ¼ 10.0 Hz, H-3),
were measured using a Stuart Scientic SMP3 apparatus or 6.63 (1H, d, J ¼ 3.5 Hz, H-1) and 8.72 (1H, s, NH); 13C NMR
Stuart automatic melting point apparatus (SMP40) and are (CDCl3) d 20.4, 20.5, 20.7 (3 ꢂ OAc), 53.1 (CO2CH3), 68.9 (C–H),
uncorrected. NMR spectra were recorded on a Bruker BioSpin 69.1 (C–H), 69.5 (C–H), 70.5 (C–H), 92.6 (C–H), 160.6 (C]N),
Ultrashield Plus (500 MHz for 1H; 125 MHz for 13C) using 167.2 (C]O), 169.5 (C]O), 169.7 (C]O) and 169.8 (C]O).
deuterated solvent as lock. IR spectra were recorded on an
(2R,3R,4S,5S,6S)-2-(((2R,3S,5R)-3-Fluoro-5-(5-methyl-2,4-dioxo-
Agilent Cary 630 FTIR Spectrometer. UV analysis was performed 3,4-dihydropyrimidin-1(2H)-yl)tetrahydrofuran-2-yl)methoxy)-
using a Hitachi U-2900 spectrophotometer, with the samples 6-(methoxycarbonyl)tetrahydro-2H-pyran-3,4,5-triyl triacetate 7.
dissolved in ethanol. LCMS was carried out on a Waters Acquity 30-Deoxy-30-uorothymidine 1 (113 mg, 0.47 mmol) and methyl
UPLC system with PDA and ELSD operating in positive and 2,3,4-triacetyl-1-O-(trichloroacetimidoyl)-a-D-glucopyranuronate
negative ion electrospray mode, employing an Acquity UPLC 6 (215 mg, 0.47 mmol) were solubilised in anhydrous CH2Cl2
ꢀ
˚
BEH C18, 1.7 mm, 2.1 ꢂ 50 mm column with 0.1% formic acid (2 mL) with 4 A molecular sieves (70 mg) and cooled to ꢁ40 C.
and water–acetonitrile (5–95%) for gradient elution. HRMS were Trimethylsilyl triuoromethanesulfonate (26 mg, 21 mL, 0.116
measured using a Finnigan MAT 95 XP or a Finnigan MAT 900 mmol) was introduced dropwise, and stirring continued for 2 h
ꢀ
XLT.
at ꢁ40 C. The solution was allowed to reach ambient temper-
Methyl 2,3,4-tri-O-acetyl-a,b-glucopyranuronate 5. Methyl ature, and was stirred at room temperature for 18 h. The reac-
2,3,4-triacetyl-a,b-glucopyranuronate 5 was prepared from ace- tion mixture was concentrated directly onto silica and
tobromo-a-D-glucuronic acid methyl ester 3, according to the preliminary purication [dichloromethane–methanol (100 : 0)
literature15 using silver carbonate in acetone–water. Aer / (96 : 4)] of the crude residue was carried out, followed by
passing the reaction mixture through a pad of Celite® the semi-preparative HPLC, to afford title compound 7 (60 mg,
solvent was removed in vacuo to yield a dark orange gum. MPC 23%) as a white solid; Rf ¼ 0.60 (10% CH3OH–CH2Cl2); IR
986 | Med. Chem. Commun., 2014, 5, 984–988
This journal is © The Royal Society of Chemistry 2014

View Article Online
Concise Article
MedChemComm
(cmꢁ1) 1209, 1682, 1748 (C]O), 2956 (C–H); lmax (EtOH)/nm reach ambient temperature over 20 h and then diluted with
265; mp 112.4–113.8 ꢀC; 1H NMR (CDCl3) d 1.95 (3H, d, J ¼ 1.2 CH2Cl2 (3 mL) and passed through Celite® to collect the
Hz, Ar-CH3), 1.98 (3H, s, OAc), 2.00 (3H, s, OAc), 2.03 (3H, s, precipitate. The ltrate was washed with saturated aqueous
OAc), 2.10–2.13 (1H, m, H-2a), 2.48–2.57 (1H, m, H-2b), 3.70 NaHCO3 (3 mL), passed through a phase separator and the
(1H, dd, J ¼ 10.5 and 2.0 Hz, H-5a), 3.74 (3H, s, CO2CH3), 4.07 solvent removed in vacuo. MPC [dichloromethane–methanol
(1H, d, J ¼ 9.5 Hz, H-50), 4.19 (1H, dt, J ¼ 10.6 and 1.9 Hz, H-5b), (100 : 0) / (96 : 4)] of the crude residue afforded the product,
4.32 (1H, dt, J ¼ 27.1 and 2.5 Hz, H-4), 4.51 (1H, d, J ¼ 7.9 Hz, H- which was further triturated with petrol to afford title
10), 4.92 (1H, dd, J ¼ 9.7 and 7.9 Hz, H-20), 5.02 (1H, dd, J ¼ 53.7 compound 4 (167 mg, 50%) as a white solid; Rf ¼ 0.37 (10%
and 4.7 Hz, H-3), 5.12 (1H, dd, J ¼ 9.9 and 9.9 Hz, H-40), 5.24 CH3OH–CH2Cl2); IR (cmꢁ1) 1212, 1369, 1684, 1746 (C]O), 2956
(1H, dd, J ¼ 9.6 and 9.6 Hz, H-30), 6.42 (1H, dd, J ¼ 9.0 and 5.5 (C–H); lmax (EtOH)/nm 265; mp 120.9–123.0 ꢀC; 1H NMR
Hz, H-1), 7.47 (1H, d, J ¼ 1.0 Hz, Ar-H), 8.49 (1H, s, NH); 13C (CDCl3) d 1.70 (3H, s, CH3), 1.87 (3H, s, Ar-CH3), 2.02 and 2.04
NMR (CDCl3) d 12.4 (Ar-CH3), 20.5 (OAc), 20.6 (2 ꢂ OAc), 37.9 (d, (7H, 2 ꢂ s and m, 2 ꢂ OAc, H-2a), 2.54 (1H, m, H-2b), 3.7 (5H, s
J ¼ 21.0 Hz, C-2), 53.0 (CO2CH3), 69.4 (C-40), 69.5 (C-5), 70.9 and m, CO2CH3, H-5a, H-5b*), 4.18 (1H, m, H-20), 4.3–4.4 (2H,
(C-20), 71.4 (C-30), 72.9 (C-50), 83.0 (d, J ¼ 26 Hz, C-4), 84.9 (C-1), 2 ꢂ m, H-4, H-50*), 5.0–5.3 (3H, 3 ꢂ m, H-3, H-30, H-40*), 5.95
94.3 (d, J ¼ 177 Hz, C-3), 100.6 (C-10), 111.6, 135.4 (ArC-H), 150.3, (1H, d, J ¼ 3.5 Hz, H-10), 6.34 (1H, m, H-1), 7.47 (1H, s, NCH),
163.6, 166.6, 169.3, 170.0; 19F NMR (470.7 MHz, CDCl3) d 9.22 (1H, s, NH); 13C NMR (CDCl3) d 12.4, 20.6, 20.7, 23.5, 38.7,
ꢁ173.5; LRMS (ESI-) m/z 559.3 [M ꢁ H]ꢁ; HRMS calcd for 52.7, 62.5, 67.2, 67.5, 70.7, 74.3, 83.4, 85.2, 95.1, 95.5, 111.3,
C
23H28FN2O13 [M ꢁ H]ꢁ 559.1575, found 559.1578; HPLC 96.7% 123.0, 135.3, 150.4, 163.9, 168.7, 168.7, 169.3; 19F NMR (CDCl3) d
purity (elution with 0.1% ammonia/MeCN) and 97.7% purity ꢁ174.03; LRMS (ESI-) m/z 559.4 [M ꢁ H]ꢁ; HRMS calcd for
(elution with 0.1% formic acid/MeCN).
C
C
23H33FN3O13 [M + NH4]+ 578.1992, found 578.1988; calcd for
(2S,3S,4S,5R,6R)-6-(((2R,3S,5R)-3-Fluoro-5-(5-methyl-2,4-dioxo-
23H30FN2O13 [M + H]+ 561.1726, found 561.1723. * Tentative
3,4-dihydropyrimidin-1(2H)-yl)tetrahydrofuran-2-yl)methoxy)- assignments.
3,4,5-trihydroxytetrahydro-2H-pyran-2-carboxylic acid 2. To a
solution of glucuronidated FLT 7 (30 mg, 0.054 mmol) in THF–
Analysis of [18F]FLT and [18F]FLT-glucuronide samples
H2O (4 : 1, 2.5 mL) at 0 ꢀC, was added a 2 M aqueous solution of
KOH (135 mL, 0.27 mmol). The reaction mixture was allowed to The clinical study on which patient samples were collected was
warm to ambient temperature, and stirring continued for 2 h, conducted by the Newcastle upon Tyne Hospitals NHS Foun-
before addition of AcOH to bring the pH to 8, followed by dation Trust (NUTH). The study was approved by the local ethics
removal of the solvent under reduced pressure. Purication of committee and all patients gave fully informed consent
the crude residue by semi-preparative HPLC afforded title according to the Declaration of Helsinki guidelines. The
compound 2 (22 mg, 97%) as a white solid; Rf ¼ 0.60 (10% administration of radioactivity for the PET scans was approved
CH3OH–CH2Cl2); IR (cmꢁ1) 1120, 1664, 2071, 2480, 3374 (O–H); by the Administration of Radioactive Substances Advisory
lmax (EtOH)/nm 262; mp 150.6–152.2 ꢀC; 1H NMR (MeOD) d 1.94 Committee, United Kingdom. Patients received an [18F]FLT
(3H, s, CH3), 2.27–2.44 (1H, m, H-2a), 2.48–2.58 (1H, m, H-2b), dose and blood samples (10 mL) were collected at 5 time points
3.25 (1H, app. t, J ¼ 8.7 Hz, H-20), 3.25 (1H, app. t, J ¼ 9.0 Hz, H- up to 60 min post-injection. Plasma was separated from the
30), 3.43 (1H, app. t, J ¼ 9.6 Hz, H-40), 3.81–3.85 (2H, m, H-50 and blood by centrifuging at 2000 rpm for 15 min (MSE Falcon
H-5a), 4.22 (1H, d, J ¼ 10.5 Hz, H-5b), 4.43–4.47 (2H, m, H-10 and 6/300). Protein was precipitated from the plasma sample (3 mL)
H-4), 5.43 (1H, dd, J ¼ 4.0 and 53.5, H-3), 6.36 (1H, dd, J ¼ 5.5 by the addition of acetonitrile (6 mL). The mixture was centri-
and 9.0 Hz, H-1), 7.84 (1H, s, Ar-H); 13C NMR (MeOD) d 12.7 fuged (5 min at 12 000 rpm, Eppendorf centrifuge 5418). The
(CH3), 39.5 (d, J ¼ 21.0 Hz, C-2), 68.9 (C-5), 70.3 (C-40), 73.1 (C- supernatant was evaporated at 50 ꢀC under nitrogen gas in a
20), 75.0 (C-50), 77.7 (C-30), 85.5 (d, J ¼ 26.0 Hz, C-4), 86.7 (C-1), TurboVap® LV evaporator (Zymark), and resuspended in HPLC
95.8 (d, J ¼ 175.0 Hz, C-3), 104.5 (C-10), 112.0 (C-Ar), 137.7 (Ar-C- mobile phase (150 mL). An aliquot (80 mL) of the resuspended
H); 19F NMR (MeOD) d ꢁ175.5; LRMS (ESI-) m/z 419.3 [M ꢁ H]ꢁ; sample containing [18F]FLT and [18F]FLT-glucuronide was
HRMS calcd for C16H20FN2O10 [M ꢁ H]ꢁ 419.1102, found separated on a Phenomenex Kinetex™ C18 column (100 mm ꢂ
419.1102; HPLC 99.6% purity (elution with 0.1% ammonia/ 4.6 mm, 2.6 mM) using an isocratic HPLC method [0.01 M
MeCN) and 99.7% purity (elution with 0.1% formic acid/MeCN). potassium dihydrogen phosphate–acetonitrile (88 : 12 w/w),
(3aR,5S,6S,7S,7aR)-2-(((2R,3S,5R)-3-Fluoro-5-(5-methyl-2,4- 1 mL minꢁ1], and detected with a Posi-RAM Model 4 Radio-
dioxo-3,4-dihydropyrimidin-1(2H)-yl)tetrahydrofuran-2-yl)- HPLC Detector (Lablogic). This procedure was performed on
methoxy)-5-(methoxycarbonyl)-2-methyltetrahydro-3aH-[1,3]- sample from six patients on 10 occasions, four patients being
dioxolo[4,5-b]pyran-6,7-diyl
diacetate
4.
30-Deoxy-30- studied aer two separate [18F]FLT PET scans.
uorothymidine 1 (150 mg, 0.60 mmol), 2,6-lutidine (78 mL, 0.66
mmol) and silver triuoromethanesulfonate (168 mg, 0.66
Analysis of [19F]FLT and [19F]FLT-glucuronide samples
ꢀ
˚
mmol) were stirred in anhydrous THF (3 mL) at ꢁ36 C with 4 A
molecular sieves (200 mg) for 45 min. Acetobromo-a-D-glucur- An aliquot (10 mL) of mobile phase containing [19F]FLT (208 mM)
onic acid methyl ester 3 (261 mg, 0.66 mmol) in anhydrous THF and [19F]FLT-glucuronide (119 mM) was separated using the
(2.25 mL) was cooled to ꢁ36 ꢀC and then added to the main above HPLC method for [18F]FLT and [18F]FLT-glucuronide,
reaction vessel via syringe. The reaction mixture was allowed to with detection by a Waters 2487 Dual l Absorbance Detector.
This journal is © The Royal Society of Chemistry 2014
Med. Chem. Commun., 2014, 5, 984–988 | 987

View Article Online
MedChemComm
Concise Article
7 B. Fischer, A. Nudelman, M. Ruse, J. Herzig, H. E. Gottlieb
and E. Keinan, J. Org. Chem., 1984, 49, 4988–4993.
8 B. Helferich and K. Weis, Chem. Ber., 1956, 89, 314–321.
9 A. V. Stachulski and X. Meng, Nat. Prod. Rep., 2013, 30, 806–
848.
10 W. Koenigs and E. Knorr, Ber. Dtsch. Chem. Ges., 1901, 34,
957–981.
11 J. R. Harding, C. D. King, J. A. Perrie, D. Sinnott and
A. V. Stachulski, Org. Biomol. Chem., 2005, 3, 1501–1507.
12 S. P. Douglas, D. M. Whiteld and J. J. Krepinsky, J.
Carbohydr. Chem., 1993, 12, 131–136.
Acknowledgements
The authors wish to thank patients, doctors and nurses at the
Newcastle Centre for Cancer Care for providing the blood
samples analysed in this study. In particular, the authors are
grateful to Philip Berry, Karen Haggerty, Ross Maxwell, Noor
Haris and Rachel Pearson for their assistance. The authors
thank the EPSRC UK National Mass Spectrometry Facility
(NMSF), Swansea for mass spectrometric determinations. This
study was supported in part by a Cancer Research UK, EPSRC,
MRC and NIHR Cancer Imaging Programme Grant (SJH, MG,
DRN).
13 H.-K. Biesalski, G. Doepner, H. Kunz, J. Paust and M. John,
Liebigs Ann., 1995, 1995, 717–720.
14 S. Hanessian and J. Banoub, Carbohydr. Res., 1977, 53, C13–
C16.
Notes and references
1 L. M. Kenny, D. M. Vigushin, A. Al-Nahhas, S. Osman,
S. K. Luthra, S. Shousha, R. C. Coombes and
E. O. Aboagye, Cancer Res., 2005, 65, 10104–10112.
2 E. Hodgson, A Textbook of Modern Toxicology, John Wiley &
Sons, Inc., Hoboken, New Jersey, 4th edn, 2010.
3 P. G. Wells, P. I. Mackenzie, J. Roy Chowdhury,
C. Guillemette, P. A. Gregory, Y. Ishii, A. J. Hansen,
F. K. Kessler, P. M. Kim, N. Roy Chowdhury and
J. K. Ritter, Drug Metab. Dispos., 2004, 32, 281–290.
4 R. R. Schmidt and G. Grundler, Synthesis, 1981, 1981, 885–
887.
15 N. Pravdic, D. Keglevic, E. D. Andrews, W. E. Harvey,
R. Bolton, A. F. Casy, J. L. Myers, D. W. Mathiwson,
W. B. Whalley, J. M. Blatchly, R. Taylor, A. B. A. Jansen,
P. J. Stokes, G. W. H. Cheeseman, J. S. Davidson,
J. M. Crabtree, P. N. Gates, E. F. Mooney, J. K. Sutherland,
D. A. Widdowson, S. Mongkolsuk, F. M. Dean, G. W. Kirby,
H. P. Tiwari, D. J. Machin, D. F. C. Morris, E. L. Short,
R. S. Bluck, A. L. Odell, R. W. Olliff, J. Burdon,
W. B. Hollyhead, C. R. Patrick, J. A. Salthouse,
T. C. Waddington, S. M. Gadedar, A. M. Kotsen and
E. Cohen, J. Chem. Soc., 1964, 4633–4668.
5 R. R. Schmidt, H. Gaden and H. Jatzke, Tetrahedron Lett.,
1990, 31, 327–329.
16 F. J. Urban, B. S. Moore and R. Breitenbach, Tetrahedron
Lett., 1990, 31, 4421–4424.
6 A. V. Stachulski and G. V. Jenkins, Nat. Prod. Rep., 1998, 15,
173–186.
988 | Med. Chem. Commun., 2014, 5, 984–988
This journal is © The Royal Society of Chemistry 2014