Synthesis of 2-[18F]Fluoro-2-deoxyisosorbide 5-mononitrate and Assessment of Its in vivo Biodistribution as Determined by Dynamic Positron Emission Tomography (PET)

Article abstract of DOI:10.1002/cmdc.201500275

Herein we disclose the synthesis of 2-fluoro-2-deoxyisosorbide 5-mononitrate (2F-IS-5MN), a fluorinated analogue of the commonly prescribed vasodilator isosorbide 5-mononitrate (IS-5MN). X-ray structural data for IS-5MN and its C2-epimeric congener IM-5MN are presented together with structural data for 2F-IS-5MN. Radioisotope labeling of 2F-IS-5MN has, for the first time, allowed observation of the in vivo biodistribution of this organic nitrate by means of dynamic positron emission tomography (PET) in wild-type mice. Visualization breakthrough: Herein we disclose the synthesis of a fluorinated analogue of the commonly prescribed vasodilator isosorbide 5-mononitrate (IS-5MN). Radioisotope labeling of 2F- IS-5MN has, for the first time, allowed observation of the in vivo biodistribution of this organic nitrate by means of dynamic positron emission tomography (PET) in wild-type mice.

Full text of DOI:10.1002/cmdc.201500275

DOI: 10.1002/cmdc.201500275
Full Papers
Synthesis of 2-[18F]Fluoro-2-deoxyisosorbide 5-
mononitrate and Assessment of Its in vivo Biodistribution
as Determined by Dynamic Positron Emission Tomography
(PET)
Nico Santschi,[a, b] Stefan Wagner,[c] Constantin Daniliuc,[a] Sven Hermann,[b, d]
Michael Schäfers,*[b, c, d] and Ryan Gilmour*[a, b]
Herein we disclose the synthesis of 2-fluoro-2-deoxyisosorbide
5-mononitrate (2F-IS-5MN), a fluorinated analogue of the com-
monly prescribed vasodilator isosorbide 5-mononitrate (IS-
5MN). X-ray structural data for IS-5MN and its C2-epimeric
congener IM-5MN are presented together with structural data
for 2F-IS-5MN. Radioisotope labeling of 2F-IS-5MN has, for the
first time, allowed observation of the in vivo biodistribution of
this organic nitrate by means of dynamic positron emission
tomography (PET) in wild-type mice.
Introduction
Since the late 1800s, organic nitrates have found widespread
application as potent vasodilators for the treatment of ische-
mic heart disease.[1–4] More than a century later, this condition
remains a major threat to human life, with the World Health
Organization having estimated mortality rates at around seven
million cases in 2012.[5] The success of organic nitrate prodrugs
relies on a rapid metabolic de-nitration reaction with concomi-
tant release of the cellular signaling agent nitric oxide (NO);
this ultimately leads to vascular smooth-muscle cell relaxa-
tion.[6] Amongst the most commonly studied organic nitrates
to date, glyceryl trinitrate (GTN) and isosorbide dinitrate (ISDN,
1) feature most prominently (Figure 1). Interestingly, however,
whereas GTN and ISDN exhibit short in vivo half-lives (t1/2) of
<10 and 30–50 min, respectively,[7] the primary ISDN metabo-
lite isosorbide 5-mononitrate (IS-5MN, (S)-2) does not display
a first-pass effect. Accordingly, it is longer lived (t1/2 =4–5 h),
and, as a direct consequence, is almost fully bioavailable with
Figure 1. Structures of glyceryl trinitrate (GTN), isosorbide dinitrate (ISDN, 1)
and its primary metabolite isosorbide 5-mononitrate (IS-5MN, (S)-2).
an enhanced pharmacokinetic profile relative to ISDN (bioavail-
ability: 30–60%).[8,9] As such, several synthetic strategies to IS-
5MN have been reported,[10–12] and the drug has been commer-
cialized. More recently, the conjugation of aspirin to IS-5MN
has been considered to potentially decrease aspirin-induced
damage to the gastrointestinal tract.[13] In view of the continu-
ing threat posed by ischemic heart disease and the current in-
terest surrounding the therapeutic use of the IS-5MN scaffold,
an in vivo dynamic positron emission tomography (PET) study
was initiated based on a suitably 18F-radiolabeled probe. Cog-
nizant of the fact that only the R-ONO2 motif is required for
NO generation, the C2 hydroxy group should not be vital to
the drug’s vasodilatory function. Hence, the C2 position was
selected as the locus for the radioactive reporter moiety. To
the best of our knowledge, this study constitutes the first real-
time, whole-body biodistribution pattern of an organic nitrate
by means of PET in vivo. Furthermore, this study provided an
opportunity to collect crystallographic structural data for IS-
5MN, which are conspicuously absent despite the interest in
this class of materials.[14]
[a] Dr. N. Santschi, Dr. C. Daniliuc, Prof. Dr. R. Gilmour
Institut für Organische Chemie, Westfälische Wilhelms-Universität Münster
Corrensstraße 40, 48149 Münster (Germany)
E-mail: ryan.gilmour@uni-muenster.de
Homepage: www.uni-muenster.de/Chemie.oc/gilmour/en/index.html
[b] Dr. N. Santschi, Dr. S. Hermann, Prof. Dr. M. Schäfers, Prof. Dr. R. Gilmour
Excellence Cluster EXC 1003 “Cells in Motion”
Westfälische Wilhelms-Universität Münster, 48149 Münster (Germany)
[c] Dr. S. Wagner, Prof. Dr. M. Schäfers
Klinik für Nuklearmedizin, Westfälische Wilhelms-Universität Münster
Albert-Schweitzer-Campus 1, 48149 Münster (Germany)
[d] Dr. S. Hermann, Prof. Dr. M. Schäfers
European Institute of Molecular Imaging
Westfälische Wilhelms-Universität Münster
Waldeyerstraße 15, 48149 Münster (Germany)
E-mail: schafmi@uni-muenster.de
Results and Discussion
Chemistry
The synthesis of IS-5MN ((S)-2) began with the selective protec-
tion of the sterically more accessible exo-hydroxy group of iso-
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201500275.
ChemMedChem 2015, 10, 1724 – 1732
1724
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Scheme 1. Syntheses of IS-5MN ((S)-2) and its diastereomer IM-5MN ((R)-2).
Reagents and conditions: a) DCC, BzOH, DMAP, CH2Cl2, 08C!RT; b) KOH,
EtOH, 508C. Descriptors: (S/R) refer to the absolute stereochemistry at C2.
sorbide (S)-3 at C2, thereby providing product (S)-4 in 58%
yield on a 5-gram scale (Scheme 1).[15] The subsequent nitration
of (S)-4 in a mixture of nitric acid (10 equiv) and acetic anhy-
dride (10 equiv) was performed on a 1-gram scale and afforded
compound (S)-5 as pale-yellow needles (81% yield) after re-
crystallization.[16] Next, cleavage of the protecting group was
achieved by treatment with potassium hydroxide (0.3 equiv) in
ethanol at 508C for 20 min.[17] The desired product was purified
by recrystallization from hot chloroform and n-hexane to fur-
nish IS-5MN ((S)-2) in 40% yield over three steps. By extension
of the same strategy, congener (R)-2 was accessed by starting
from commercially available isomannide (R)-3. Whereas the ni-
tration and deprotection protocols performed comparably
with 83 and 73% yields of (R)-5 and (R)-2, respectively, the ini-
tial mono-protection provided product (R)-4 in only 34% yield.
For all isosorbide (S)-3- or isomannide (R)-3-derived com-
pounds, an identical numbering scheme as for (S)-2 was adopt-
ed, displaying the exo-ONO2 group at C5.
Scheme 2. Synthesis of fluorinated analogues (S)-6 (2F-IS-5MN) and (S)-[18F]6.
Reagents and conditions: a) DAST, NEt3, CH2Cl2, 08C!RT; b) [18F]KF, K2.2.2
,
K2CO3, CH3CN, 110 8C, 20 min; c) TsCl, py, CH2Cl2, 08C!RT.
Gratifyingly, crystals suitable for X-ray diffraction were ob-
tained for compounds (S)-2, (R)-2, and (S)-6 through vapor dif-
fusion of solutions in methanol, chloroform, and diethyl ether,
respectively, at À208C in an atmosphere of pentane. Selected
bond lengths and dihedral angles are summarized in Table 1,
and graphical representations are shown in Figure 2.
All systems include cis-fused five-membered rings that
adopt an envelope conformation, with the overall molecular
structure reminiscent of the energetically most favorable
chair–boat conformation of cyclooctane.[19] The envelope-
shaped substructures are further supported by the dihedral
angles around the C3ÀC4 bond with the methylene C1 and C6
groups in the envelope endo configuration. Consequently, the
C5-nitro groups are oriented in an axial fashion with similar
bond lengths (~1.45 Š). Furthermore, the C2ÀX bonds de-
crease with the substitution X=ONO2, OH, F, with respective
values of 1.459(2), 1.448(3), 1.431(10), and 1.407(2) Š.
1
Inspection and comparison of the H NMR spectra of com-
pounds (S)-2 and (R)-2 allow the stereochemistry at C2 to be
assigned based on coupling constant analysis. Whereas in IS-
5MN (S)-2 no coupling of the bridgehead hydrogen C3-H to
C2-H was observed, the inversion of configuration in (R)-2 at
C2 was indicated by the emergence of a triplet at 4.47 ppm
(3JHH =5.3 Hz).
To provide the 19F-analogue required for characterization
purposes, (R)-2 was exposed to the deoxofluorination reagent
diethylaminosulfur trifluoride (DAST, 1.2 equiv) in dichlorome-
thane at 08C. Whilst this initial attempt furnished the expected
product in poor yield (<20%), the stoichiometric addition of
a base notably increased the reaction efficiency. Performing
the reaction in the presence of triethylamine (1.2 equiv) fur-
nished the fluorinated product (S)-6 (2F-IS-5MN) in 46–73%
yield. The resultant pale-yellow oil solidified at À208C. As ex-
pected, the reaction proceeded with inversion of configuration
Table 1. Summary of selected bond lengths and (dihedral) angles in 1,
(S)-2, (R)-2, and (S)-6.
ISDN
(1)[a]
IS-5MN
((S)-2)[b,f]
IM-5MN
((R)-2)[c,f]
2F-IS-5MN
((S)-6)[f]
Bond [Š]
C2ÀX[d]
1.459(2)
1.450(2)
1.545(2)
1.428(3)
1.448(3)
1.545(3)
1.431(10)
1.454(11)
1.532(12)
1.407(2)
1.451(2)
1.543(3)
C5ÀONO2
C3ÀC4
Angle [8]
O1-C4-C3-C2
O2-C3-C4-C5
H-C2-C3-H
H-C5-C4-H
LSP[e]
0.35
À5.38
À10.18
À92.80
À14.59
62.38
À9.13
À13.09
33.26
À12.20
63.38
À5.97
À10.38
À92.75
À13.78
62.68
À7.85
3
at C2, as was corroborated by the characteristic JHH coupling
À94.87
À13.69
60.85
constant between C2-H and C3-H of the core. Rigorous struc-
tural proof was attained by X-ray crystallography (see below).
The precursor for the preparation of (S)-[18F]6 required for PET
studies was accessed by mono-tosylation of isomannide (R)-3
to afford (R)-7 in 39% yield,[18] followed by the standard nitra-
tion conditions delineated above. Recrystallization from diethyl
ether and dichloromethane at À208C yielded (R)-8 in 86%
yield (Scheme 2).[16] Coupling constant analysis was again con-
sistent with the endo-tosylate (t, 3JHH =5.1 Hz, C3-H).
[a] CSD refcode: PIMKOL. [b] Asymmetric unit contains two molecules.
[c] Asymmetric unit contains three molecules. [d] ISDN (1, X=ONO2); IS-
5MN ((S)-2, X=OH); IM-5MN ((R)-2, X=OH); 2F-IS-5MN ((S)-6, X=F).
[e] Angle between least-square planes defined by O1-C1-C2-C3-C4 and
O2-C3-C4-C5-C6. [f] CCDC 1400345 ((S)-2), 1400347 ((R)-2), and 1400346
((S)-6) contain the supplementary crystallographic data for this paper.
These data are provided free of charge by The Cambridge Crystallograph-
ic Data Centre.
ChemMedChem 2015, 10, 1724 – 1732
www.chemmedchem.org
1725
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Biology
Compound (S)-[18F]6 possess
moderate
lipophilicity,
with
a log D (exp.) value of 0.73Æ
0.03. Relative to the calculated
logD value (clog D) of 0.24,[21]
the actual solubility of (S)-[18F]6
in 1-octanol in the biphasic 1-oc-
tanol/PBS system is increased by
a factor of 3. Next, an in vitro
stability study was carried out
Figure 2. ORTEP representations of IS-5MN ((S)-2), IM-5MN ((R)-2), and 2F-IS-5MN ((S)-6). Thermal ellipsoids are set
to 50% probability.
3
Because JHH couplings for the H-C2-C3-H system were
absent in (S)-2 and (S)-6, dihedral angles of ~908 were expect-
ed based on the Karplus equation. Measurement of the torsion
angles in the solid state provided values of À92.80 and
À92.758 for (S)-2 and (S)-6, respectively. For (R)-2, in which the
configuration at C2 is inverted, the H-C2-C3-H angle widens to
33.268. In all cases, the H-C5-C4-H angles are similar and range
from À14.598 for (S)-2 to À12.208 for (R)-2.
using human blood serum. During incubation for up to 90 min
at 378C, (S)-[18F]6 was found to possess high serum stability. As
shown in Figure 3, only the parent compound (S)-[18F]6 was
detected by radio-HPLC, and significant radiometabolites or
decomposition products were not observed.
Consequently, biodistribution studies with labeled (S)-[18F]6
were conducted in adult wild-type mice (C57/Bl6) in vivo (n=
10) and ex vivo (n=12). The PET imaging setup covers the
whole body of a mouse in a single field of view. PET acquisition
was started, and after intravenous (i.v.) bolus injection of (S)-
[18F]6 (400 kBqgÀ1 body weight in 100 mL saline) the distribu-
tion of radioactivity was continuously measured for 90 min. Re-
constructed PET images (Figure 4) confirm the efficient and
rapid tissue availability of (S)-[18F]6 throughout the whole body
In addition, the angles between the least-square planes de-
fined by the two five-membered rings are similar to that in
ISDN (60.858), with 62.38, 63.38, and 62.688 for (S)-2, (R)-2, and
(S)-6, respectively. All congeners form heavily hydrogen
bonded structures. Notably, (S)-2 and (S)-6 display similar inter-
actions with NO2ÀHC5NO2 distances of 2.695/2.919 Š and
2.603/2.779 Š, respectively. The motifs found for (R)-2 involve
NO2ÀHC2OH (2.702 Š), NO2ÀHC4 (2.634 Š), and NO2À
HC6 (2.856 Š) contacts. However, all these interac-
tions are significantly longer than values reported for
the prototypical NO2ÀHR fragment (2.30 Š).[20] Finally,
the fluorine substituent does not engage in hydro-
gen bonds, in contrast to the C2OH substituents of
(S)-2 and (R)-2.
The radiosyntheses of (S)-[18F]6 were realized in
one step via nucleophilic substitution (SN2) of the to-
sylate leaving group of the precursor compound (R)-
8 with [18F]fluoride. The target compound (S)-[18F]6
was obtained in radiochemical yields (rcy) of 14.3Æ
4.4% [n=5, decay-corrected (d.c.) based on cyclo-
tron-derived [18F]fluoride ions] 101Æ12 min from the
end of radionuclide production. The target com-
pound was isolated with specific activities (As) of 2–
16 GBqmmolÀ1 and with high radiochemical purities
(ꢀ99%). Due to the volatility of the target com-
pound (S)-[18F]6, it was not possible to evaporate the
aqueous HPLC product fraction at increased tempera-
tures in vacuo and then to formulate the compound
in the final injection solution without losing the main
part of the product activity. This drawback was cir-
cumvented by extraction of the aqueous HPLC prod-
uct fraction with dichloromethane and evaporation
of this (S)-[18F]6-containing organic layer at ambient
temperature (or lower) to dryness in vacuo.
Figure 3. Typical quality control radio-HPLC chromatograms of a) a batch of (S)-[18F]6,
and of the in vitro stability of (S)-[18F]6 (tR =10.44 min) after incubation in human blood
serum at 378C after b) 10 and c) 90 min.
ChemMedChem 2015, 10, 1724 – 1732
www.chemmedchem.org
1726
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
vivo data are consistent with the
initial assessment based on dy-
namic PET measurements. The
highest concentration of radio-
activity was found in the kid-
neys, closely followed by fat
tissue. Moreover,
a
twofold
excess of radioactivity resided in
fat tissue as compared with
blood.
Conclusions
Herein we disclose a synthetic
strategy to access a novel IS-
5MN derivative, namely, 2-deoxy-
2-fluoroisosorbide 5-mononitrate
(S)-6, over four steps by starting
from commercially available
Figure 4. Representative PET/CT study of the in vivo distribution of (S)-[18F]6 at selected timeframes post-injection
(p.i.). Whereas within the first minutes p.i. the distribution of radioactivity is determined mainly by tissue perfu-
sion, later timeframes (5–90 min p.i.) present a diffusion-like distribution pattern accompanied by elimination of
radioactivity via the kidneys into the urinary bladder.
within seconds. In parallel, radioactivity is eliminated by the
kidneys and accumulates in the urinary bladder.
starting material. In addition, (S)-2 and its epimer (R)-2 were
prepared, and their solid-state structures were compared with
that of (S)-6. To the best of our knowledge, this constitutes the
first X-ray structural analysis of the commonly prescribed vaso-
dilator IS-5MN. Subsequent radiolabeling starting from a suita-
ble precursor yielded (S)-[18F]6 with an acceptable rcy of 14.3Æ
4.4% and excellent radiochemical purities (ꢀ99%), thereby fa-
cilitating the real-time in vivo biodistribution analysis of an
organic nitrate by dynamic PET.
PET imaging data were analyzed quantitatively for selected
regions of interest (Figure 5). Time curves of regional radio-
activity concentrations expressed as percentage of injected
dose per volume (% IDcmÀ3) reveal a rapid and efficient distri-
bution of (S)-[18F]6 in the organism with a fast washout from
blood (Figure 5a,b). In fact, blood levels reach a peak at 18 s
post-injection (p.i.) (27% IDcmÀ3) with a subsequent rapid de-
crease (50% decline from peak activity within 16 s). (S)-[18F]6 is
taken up by the liver rapidly, with a peak at ~1 min p.i.
(7.4% IDcmÀ3). Thereafter, a slow and steady decrease of radio-
activity in the liver to 4.2% IDcmÀ3 at 90 min p.i. was observed.
No accumulation of radioactivity in the gall bladder or intes-
tine was observed, thus excluding the hepatobiliary route as
the major elimination pathway of (S)-[18F]6. The kidneys pres-
ent with an uptake of radioactivity to a level of 5.7% IDcmÀ3
at 2.5 min p.i., which decreases to 5.1% IDcmÀ3 at 20 min p.i.,
and finally increases again to 5.7% IDcmÀ3 at the end of the
imaging study. Concurrently, radioactivity in the urinary blad-
der increases continuously from the first seconds until 90 min
p.i., which indicates elimination of radioactivity via the kidneys.
Further regional analysis of brain, lung, heart, and spleen (Fig-
ure 5c,d) show an initial peak of radioactivity due to tissue per-
fusion followed by a slow washout to final concentrations of
2.9, 2.7, 3.2, and 3.3% IDcmÀ3 at 90 min, respectively. In con-
trast, muscle and fat tissue present with a slow accumulation
of activity over time, reaching a maximum at 35 min p.i. with
3.2% IDcmÀ3 in the muscle and 5.2% IDcmÀ3 in the fat tissue
(Figure 5e,f). With this, the radioactivity concentration in the
fat tissue exceeds levels in the blood from 12 min p.i. by up to
54% (55 min p.i).
The biodistribution of nitrates after oral or intravenous injec-
tions in rats has been studied predominantly by using 14C-la-
beled ISDN or nitroglycerin, or by analysis of nitrate/metabolite
concentrations by a capillary gas chromatographic electron-
capture analysis of blood samples and harvested tissues over
time, often with contradicting results.[22,23] Having established
a route to the novel radiotracer (S)-[18F]6, a full, temporally re-
solved whole-body biodistribution study based on high-resolu-
tion PET imaging was successfully performed. With a dedicated
large field-of-view small-animal PET camera,[24] it was possible
to perform whole-body imaging of wild-type mice dynamically
upon i.v. tail vein injection.
In line with previous findings in rats, a rapid clearance of i.v.
injected (S)-[18F]6 from blood was observed. In parallel, virtually
all tissues including brain, lungs, heart, and muscle showed
a fast and efficient uptake of the tracer in mice. Whereas most
of these tissues were cleared of radioactivity in response to
a decline in blood, skeletal muscle and fat tissues showed a dis-
tinctively different pattern. In these tissues radioactivity accu-
mulates until ~35 min p.i., with only a minor, non-blood count
decrease-correlated washout thereafter. Whereas muscle con-
centrations stay below blood levels, higher concentrations of
radioactivity (up to 50%) remain present in fat tissues after
90 min relative to blood. To the best of our knowledge, this is
the first description of this intriguing nitrate uptake by fat
tissue. This effect may lead to an improved understanding of
the biodistribution and availability of organic nitrates.
PET/CT studies were followed by organ explantation and
radioactivity counting/weighing ex vivo to overcome potential
limitations of in vivo quantification such as limited sensitivity,
spillover, and partial volume effects. Furthermore, the ex vivo
approach allows precise analysis of the radioactivity concentra-
tion in vessel walls. Results are shown in Figure 6. These ex
ChemMedChem 2015, 10, 1724 – 1732
www.chemmedchem.org
1727
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Experimental Section
Chemistry. All chemicals were pur-
chased as reagent grade and used
without further purification. Ana-
lytical thin-layer chromatography
(TLC) was performed on aluminum
plates pre-coated with SiO2-60 F254
(Merck) and visualized with UV
light (l=254 nm) or with KMnO4
or CAM solution. Column chroma-
tography was performed on SiO2-
60 (230–400 mesh ASTM, Fluka).
Analytical methods. 1H, 13C, and
19F NMR spectra were recorded at
ambient temperature on Bruker
AV300, AV400, or Agilent DD2 600
instruments by the NMR service of
the Organic Chemistry Institute,
WWU Münster. Chemical shifts are
reported in ppm and are refer-
enced to residual solvent signals.
Melting points were measured on
a Büchi B-545 melting-point appa-
ratus. IR spectra were recorded on
a PerkinElmer 100 FTIR spectrome-
ter. Optical rotations were mea-
sured on a JASCO P2000 polarime-
ter. High-resolution mass spec-
trometry (HRMS) data were mea-
sured by the MS service of the Or-
ganic Chemistry Institute, WWU
Münster.
(3S,3aR,6R,6aR)-6-Hydroxyhexa-
hydrofuro[3,2-b]furan-3-yl benzo-
ate ((S)-4). Prepared according to
a published procedure.[15] Isosor-
bide (5 g, 34.2 mmol, 1 equiv) and
benzoic acid (4.6 g, 37.7 mmol,
1.1 equiv) were dissolved in CH2Cl2
(100 mL) and cooled to 08C (ice
Figure 5. Time–activity concentration curves (n=10 mice) confirm elimination of radioactivity by the kidneys
[a),b)] but not by the hepatobiliary system. Although most tissues present with a time course determined pre-
dominantly by tissue perfusion [c),d)], radioactivity in muscle and fat tissues accumulates over time [e),f)]. Panels
b), d), and f) are magnifications of curves shown in a), c), and e), respectively.
bath).
N,N’-Dicyclohexylcarbodii-
mide (DCC; 7.77 g, 37.7 mmol,
1.1 equiv) and 4-(dimethylamino)-
pyridine (DMAP; 44 mg, 0.4 mmol,
0.01 equiv) were added, and stirring was continued for
8.5 h at 08C. The mixture was then allowed to warm
slowly to ambient temperature overnight (13.5 h). The
suspension was filtered, and the residual white solid was
washed with CH2Cl2. Chromatographic purification (SiO2,
cHex/EtOAc 2:1!0:1) afforded a colorless oil that was
taken up in acetone to precipitate residual dicyclo-
hexylurea. After filtration and removal of the solvent the
product was obtained as a crystalline solid (5.01 g,
20.0 mmol, 58%). Spectral data are in agreement with
published values.[15] Rf =0.25 in cHex/EtOAc 2:1; mp:
71.3–71.98C. ½aŠ25 =67.88 (c=0.10, MeOH); 1H NMR
D
(400 MHz, CDCl3): d=8.06–8.00 (m, 2H, OBzortho), 7.61–
7.56 (m, 1H, OBzpara), 7.48–7.42 (m, 2H, OBzmeta), 5.48 (d,
J=3.4 Hz, 1H, H-C3OBz), 4.72 (t, J=4.8 Hz, 1H, H-C6a),
4.64 (d, J=4.4 Hz, 1H, H-C3a), 4.35 (q, J=5.8 Hz, 1H, H-
C6OH), 4.19 (d, J=10.7 Hz, 1H, H-C2), 4.12 (dd, J=10.7,
3.5 Hz, 1H, H-C2), 3.93 (dd, J=9.5, 6.0 Hz, 1H, H-C5), 3.62
Figure 6. Ex vivo distribution of radioactivity 2 h after i.v. injection of (S)-[18F]6 in adult
mice (n=12). Error bars represent standard deviations.
ChemMedChem 2015, 10, 1724 – 1732
www.chemmedchem.org
1728
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
(dd, J=9.5, 6.0 Hz, 1H, H-C5), 2.66 ppm (s, 1H, C6O-H); 13C NMR
(101 MHz, CDCl3): d=165.6 (OBzmeso), 133.6 (OBzpara), 129.9 (2C, OB-
zortho), 129.5 (OBzipso), 128.6 (2C, OBzmeta), 85.8 (C3a), 82.2 (C6a), 79.0
(C3OBz), 73.8 (C2), 73.7 (C5), 72.5 ppm (C6OH); HRMS (ESI) calcd
m/z: 273.0733, found m/z: 273.0739 [M+Na]+; IR (ATR): n˜ =3430
(w), 2986 (w), 2936 (w), 2886 (w), 2872 (w), 1980 (w), 1722 (s), 1599
(w), 1489 (w), 1455 (w), 1401 (w), 1373 (w), 1342 (w), 1318 (w),
1296 (m), 1260 (s), 1197 (w), 1180 (w), 1109 (s), 1083 (s), 1073 (s),
1043 (s), 1027 (s), 1008 (s), 974 (m), 941 (m), 924 (m), 896 (m), 879
(m), 853 (m), 819 (m), 805 (m), 750 (m), 705 (s), 681 (s), 673 cmÀ1
(s).
H); 13C NMR (101 MHz, CDCl3): d=88.9 (C6a), 81.5 (C3ONO2),
81.3 (C3a), 75.9 (C5), 75.8 (C6OH), 69.4 ppm (C2); HRMS (ESI) calcd
m/z: 214.0322, found m/z: 214.0328 [M+Na]+; IR (ATR): n˜ =3194
(w), 2991 (w), 2949 (w), 2914 (w), 2902 (w), 1649 (s), 1633 (s), 1455
(w), 1363 (w), 1348 (w), 1335 (w), 1279 (s), 1253 (w), 1236 (w), 1221
(w), 1117 (m), 1105 (m), 1090 (s), 1059 (m), 1001 (m), 985 (m),
965 (s), 947 (w), 910 (m), 869 (m), 850 (s), 769 (m), 746 (s), 671 cmÀ1
(w).
(3R,3aR,6R,6aR)-6-Hydroxyhexahydrofuro[3,2-b]furan-3-yl ben-
zoate ((R)-4). Isomannide (2 g, 13.7 mmol, 1 equiv) and benzoic
acid (1.84 g, 15.1 mmol, 1.1 equiv) were dissolved in CH2Cl2 (40 mL)
and cooled to 08C (ice bath). DCC (3.1 g, 15.0 mmol, 1.1 equiv) and
DMAP (20 mg, 0.2 mmol, 0.01 equiv) were added, then the mixture
was allowed to warm to ambient temperature and was stirred for
two days. The suspension was cooled to À208C, filtered, and the
residual white solid was washed with CH2Cl2. Subsequently, the fil-
trate was concentrated under reduced pressure. Chromatographic
purification (SiO2, cHex/EtOAc 2:1!0:1) afforded the pure product
as a white solid (1.15 g, 4.6 mmol, 34%). Rf =0.36 in cHex/EtOAc
(3S,3aR,6R,6aS)-6-(Nitrooxy)hexahydrofuro[3,2-b]furan-3-yl ben-
zoate ((S)-5). A solution of Ac2O (1.83 mL, 19.4 mmol, 10 equiv) in
CH2Cl2 (30 mL) was cooled to 08C (ice bath), then HNO3 (99%,
0.85 mL, 20.4 mmol, 10 equiv) was added dropwise. After complete
addition, stirring at 08C was continued for an additional 10 min.
Subsequently, (S)-4 (0.5 g, 2.00 mmol, 1 equiv) was added in one
portion, and the mixture was stirred for 1 h at 08C, then poured
onto ice-cold aqueous NaHCO3 (satd.). The organic layer was sepa-
rated, and the aqueous phase was extracted once with CH2Cl2. The
combined organic layers were dried over MgSO4 and concentrated
under reduced pressure. Chromatographic purification (SiO2, cHex/
EtOAc 1:1) afforded a yellow oil that was crystallized from Et2O/
pentane at À208C to afford the product as pale-yellow needles
(0.48 g, 1.6 mmol, 81%). Rf =0.70 in cHex/EtOAc 1:1; mp: 70.5–
1:1; mp: 88.5–89.48C. ½aŠ25 =114.18 (c=1.0, MeOH); 1H NMR
D
(400 MHz, CDCl3): d=8.11–8.02 (m, 2H, OBzortho), 7.63–7.53 (m, 1H,
OBzpara), 7.44 (t, J=7.8 Hz, 2H, OBzmeta), 5.39 (q, J=5.9 Hz, 1H, H-
C3OBz), 4.83 (t, J=5.3 Hz, 1H, H-C3a), 4.52 (t, J=5.3 Hz, 1H, H-
C6a), 4.31 (q, J=6.4 Hz, 1H, H-C6OH), 4.20 (dd, J=9.6, 6.1 Hz, 1H,
H-C2), 4.03 (dd, J=9.6, 5.9 Hz, 1H, H-C2), 3.96 (dd, J=9.1, 6.3 Hz,
1H, H-C5), 3.60 (dd, J=9.1, 7.3 Hz, 1H, H-C5), 2.71 ppm (s, 1H,
C6OH); 13C NMR (101 MHz, CDCl3): d=166.0 (Bzmeso), 133.4 (Bzpara),
129.9 (2C, OBzortho), 129.5 (Bzipso), 128.6 (2C, OBzmeta), 81.7 (C6a), 80.8
(C3a), 74.6 (C3OBz), 73.9 (C5), 72.3 (C6OH), 71.3 ppm (C2); HRMS
(ESI) calcd m/z: 273.0733, found m/z: 273.0738 [M+Na]+; IR (ATR):
n˜ =3461 (w), 2954 (w), 2876 (w), 1715 (s), 1626 (w), 1601 (w), 1583
(w), 1492 (w), 1476 (w), 1452 (w), 1406 (m), 1367 (w), 1348 (w),
1314 (m), 1276 (s), 1263 (s), 1240 (m), 1176 (m), 1125 (s), 1073 (s),
1047 (s), 1024 (m), 1011 (m), 1001 (m), 969 (m), 939 (w), 890 (m),
877 (m), 841 (m), 806 (m), 733 (m), 713 (s), 688 (m), 672 cmÀ1 (m).
71.08C. ½aŠ25 =108.48 (c=1.02, MeOH); 1H NMR (600 MHz, CDCl3):
D
d=8.02 (d, J=7.6 Hz, 2H, OBzortho), 7.58 (t, J=7.4 Hz, 1H, OBzpara),
7.45 (t, J=7.7 Hz, 2H, OBzmeta), 5.48 (d, J=3.2 Hz, 1H, H-C3OBz),
5.39 (td, J=5.5, 2.8 Hz, 1H, H-C6ONO2), 5.07 (t, J=5.2 Hz, 1H, H-
C6a), 4.64 (d, J=4.9 Hz, 1H, H-C3a), 4.18 (d, J=10.8 Hz, 1H, H-C2),
4.11 (dd, J=10.8, 3.2 Hz, 1H, H-C2), 4.07 (dd, J=11.3, 3.0 Hz, 1H, H-
C5), 3.94 ppm (dd, J=11.2, 5.6 Hz, 1H, H-C5); 13C NMR (151 MHz,
CDCl3): d=165.6 (OBzmeso), 133.6 (OBzpara), 129.9 (2C, OBzortho), 129.4
(OBzipso), 128.6 (2C, OBzmeta), 86.9 (C3a), 81.7 (C6a), 81.4 (C6ONO2),
78.0 (C3OBz), 73.8 (C2), 69.4 ppm (C5); HRMS (ESI) calcd m/z:
318.0584, found m/z: 318.0576 [M+Na]+; IR (ATR): n˜ =2990 (w),
2917 (w), 1719 (m), 1627 (m), 1599 (w), 1474 (w), 1455 (w), 1351
(w), 1315 (w), 1300 (w), 1283 (m), 1264 (s), 1240 (m), 1207 (w), 1176
(w), 1111 (m), 1097 (s), 1071 (m), 1059 (m), 1023 (m), 1001 (m), 967
(m), 914 (m), 884 (m), 868 (m), 854 (s), 810 (m), 769 (m), 749 (m),
713 (s), 685 (m), 673 cmÀ1 (m).
(3R,3aR,6R,6aS)-6-(Nitrooxy)hexahydrofuro[3,2-b]furan-3-yl ben-
zoate ((R)-5). A solution of Ac2O (3.8 mL, 40.3 mmol, 10 equiv) in
CH2Cl2 (60 mL) was cooled to 08C (ice bath), then HNO3 (99%,
1.7 mL, 40.7 mmol, 10 equiv) was added dropwise. After complete
addition, stirring at 08C was continued for an additional 10 min.
Subsequently, (R)-4 (1.0 g, 4.00 mmol, 1 equiv) was added in one
portion, and the mixture was stirred for 30 min at 08C, then
poured onto ice-cold aqueous NaHCO3 (satd.). The organic layer
was separated, and the aqueous layer was extracted once with
CH2Cl2. The combined organic layers were dried over MgSO4 and
concentrated under reduced pressure. The crude material was crys-
tallized from Et2O/nHex to afford the pure product as a white solid
(3R,3aS,6S,6aR)-6-Hydroxyhexahydrofuro[3,2-b]furan-3-yl nitrate
((S)-2). (S)-5 (1.7 g, 5.8 mmol, 1 equiv) was dissolved in EtOH
(40 mL) and warmed to 508C, then KOH (0.1 g, 1.8 mmol,
0.3 equiv) was added and the solution incubated at 508C for
20 min. After cooling to ambient temperature, the mixture was
acidified with aqueous HCl (1.2m, 1.8 mL) and concentrated under
reduced pressure. The residual was dissolved in aqueous NaHCO3
(satd.) and CH2Cl2, and then the organic layer was separated. The
aqueous layer was extracted four times with CH2Cl2 and the com-
bined organic layers were dried over MgSO4 and concentrated
under reduced pressure. The crude material was washed with
nHex and then recrystallized from hot CHCl3/nHex to afford the
pure product as white, felted crystals (0.93 g, 4.9 mmol, 85%). Suit-
able crystals for X-ray diffraction were obtained through slow evap-
oration of a MeOH solution at À208C under an atmosphere of
(0.98 g, 3.3 mmol, 83%); mp: 113.6–113.98C. ½aŠ25 =262.58 (c=1.0,
D
MeOH); 1H NMR (400 MHz, CDCl3): d=8.12–8.04 (m, 2H, OBzortho),
7.63–7.56 (m, 1H, OBzpara), 7.50–7.42 (m, 2H, OBzmeta), 5.37–5.27 (m,
2H, H-C3OBz and H-C6ONO2), 4.92 (t, J=5.4 Hz, 1H, H-C6a), 4.86 (t,
J=5.4 Hz, 1H, H-C3a), 4.22 (dd, J=9.3, 6.5 Hz, 1H, H-C2), 4.06–
3.98 ppm (m, 3H, H-C2 and HH-C5); 13C NMR (101 MHz, CDCl3): d=
166.0 (Bzmeso), 133.6 (Bzpara), 130.0 (2C, OBzortho), 129.3 (Bzipso), 128.6
(2C, OBzmeta), 81.14 (C6a), 81.05 (C3a), 81.02 (C6ONO2), 73.5
(C3OBz), 70.8 (C2), 69.6 ppm (C5); HRMS (ESI) calcd m/z: 318.0584,
found m/z: 318.0589 [M+Na]+; IR (ATR): n˜ =2987 (w), 2942 (w),
2883 (w), 1716 (s), 1628 (s), 1603 (m), 1586 (w), 1493 (w), 1470 (w),
1452 (m), 1386 (w), 1329 (m), 1316 (m), 1272 (s), 1240 (m), 1182
(w), 1096 (s), 1072 (m), 1033 (s), 998 (m), 964 (m), 909 (w), 878 (s),
854 (s), 823 (m), 764 (m), 751 (m), 710 (s), 690 cmÀ1 (s).
pentane; mp: 89.5–89.88C.[26] ½aŠ25 =160.38 (c=1.4, EtOH);[10]
D
1H NMR (400 MHz, CDCl3): d=5.35 (td, J=5.5, 2.6 Hz, 1H, H-
C3ONO2), 5.00 (t, J=5.2 Hz, 1H, H-C3a), 4.41 (d, J=4.7 Hz, 1H, H-
C6a), 4.36 (d, J=2.6 Hz, 1H, H-C6OH), 4.00 (dd, J=11.3, 2.7 Hz, 1H,
H-C2), 3.95 (d, J=10.2 Hz, 1H, H-C5), 3.91 (dd, J=10.4, 2.7 Hz, 1H,
H-C5), 3.87 (dd, J=11.3, 5.5 Hz, 1H, H-C2), 1.88 ppm (s, 1H, C6O-
ChemMedChem 2015, 10, 1724 – 1732
www.chemmedchem.org
1729
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
(3R,3aS,6R,6aR)-6-Hydroxyhexahydrofuro[3,2-b]furan-3-yl nitrate
((R)-2). (R)-5 (1 g, 3.4 mmol, 1 equiv) was dissolved in EtOH (60 mL)
and warmed to 508C, then KOH (50 mg, 0.9 mmol, 0.26 equiv) was
added and the solution incubated at 508C for 30 min. The mixture
was concentrated under reduced pressure to 5 mL and acidified
with aqueous HCl (1.2m, 2 mL), then all solvents were removed in
vacuo. The residual was dissolved in aqueous NaHCO3 (satd.) and
CH2Cl2. The aqueous layer was extracted twice with CH2Cl2. The
combined organic layers were dried over MgSO4 and concentrated
under reduced pressure. The crude material was washed with
nHex and then recrystallized from CHCl3/nHex to afford the pure
product as a white solid (0.47 g, 2.5 mmol, 73%). Suitable crystals
for X-ray diffraction were obtained through slow evaporation of
a CHCl3 solution at À208C under an atmosphere of pentane; mp:
layer was diluted with CH2Cl2, then separated and the aqueous
phase was extracted once with CH2Cl2. The combined organic
layers were dried over MgSO4. The crude material was crystallized
from Et2O/CH2Cl2 at À208C to afford the pure product as fine
white needles (0.99 g, 2.9 mmol, 86%); mp: 121.1–121.58C.[26]
½aŠ25 =114.18 (c=1.01, MeOH); H NMR (400 MHz, CDCl3): d=7.87–
1
D
7.79 (m, 2H, OTsortho), 7.40–7.33 (m, 2H, OTsmeta), 5.30 (td, J=5.8,
4.2 Hz, 1H, H-C6ONO2), 4.82 (m, 2H, H-C3OTs and H-C6a), 4.51 (t,
J=5.1 Hz, 1H, H-C3a), 4.03 (dd, J=11.0, 4.2 Hz, 1H, H-C5), 3.97 (dd,
J=11.0, 5.9 Hz, 1H, H-C5), 3.94 (dd, J=9.0, 6.8 Hz, 1H, H-C2), 3.72
(t, J=8.9 Hz, 1H, H-C2), 2.45 ppm (s, 3H, CH3 in OTs); 13C NMR
(101 MHz, CDCl3): d=145.6 (OTsipso), 133.0 (OTspara), 130.1 (2C,
OTsmeta), 128.1 (2C, OTsortho), 81.1 (2C, C6ONO2 and C6a), 80.5 (C3a),
77.0 (C3OTs), 69.8 (C5), 69.3 (C2), 21.8 ppm (CH3 in OTs); HRMS (ESI)
calcd m/z: 368.0411, found m/z: 368.0406 [M+Na]+; IR (ATR): n˜ =
2999 (w), 2885 (w), 1635 (s), 1597 (w), 1496 (w), 1475 (w), 1454 (w),
1396 (w), 1354 (s), 1319 (w), 1308 (w), 1280 (s), 1249 (w), 1192 (m),
1177 (s), 1138 (w), 1114 (m), 1103 (m), 1077 (m), 1051 (m), 1034 (s),
1003 (s), 941 (m), 880 (s), 854 (s), 831 (s), 810 (s), 796 (s), 767 (m),
753 (m), 703 (w), 671 cmÀ1 (s).
70.9–71.38C.[26] ½aŠ25 =227.38 (c=1.0, MeOH); 1H NMR (400 MHz,
D
CDCl3): d=5.37 (td, J=5.5, 4.1 Hz, 1H, H-C3ONO2), 4.87 (t, J=
5.4 Hz, 1H, H-C3a), 4.47 (t, J=5.3 Hz, 1H, H-C6a), 4.30 (qd, J=8.6,
6.0 Hz, 1H, H-C6OH), 4.13–4.00 (m, 3H, HH-C2 and H-C5), 3.55 (t,
J=8.7 Hz, 1H, H-C5), 2.42 ppm (d, J=9.4 Hz, 1H, C6O-H); 13C NMR
(101 MHz, CDCl3): d=82.0 (C6a), 81.7 (C3ONO2), 81.3 (C3a), 73.3
(C5), 72.3 (C6OH), 69.8 ppm (C2); HRMS (ESI) calcd m/z: 214.0322,
found m/z: 214.0321 [M+Na]+; IR (ATR): n˜ =3282 (w), 2962 (w),
1936 (w), 2891 (w), 2865 (w), 1634 (s), 1618 (s), 1480 (w), 1450 (w),
1375 (w), 1347 (w), 1327 (w), 1311 (w), 1288 (s), 1274 (s), 1259 (m),
1237 (m), 1208 (w), 1123 (s), 1088 (s), 1042 (s), 998 (m), 965 (s), 911
(m), 876 (s), 854 (s), 763 (s), 752 (s), 683 cmÀ1 (m).
(3R,3aS,6S,6aS)-6-Fluorohexahydrofuro[3,2-b]furan-3-yl nitrate
((S)-6). To a solution of (R)-8 (0.1 g, 0.52 mmol, 1 equiv) and NEt3
(0.09 mL, 0.65 mmol, 1.3 equiv) in CH2Cl2 (2 mL) at 08C (ice bath)
was added diethylaminosulfur trifluoride (DAST; 0.08 mL,
0.61 mmol, 1.2 equiv) dropwise. The reaction was allowed to warm
to ambient temperature and was stirred for two days. The brown
mixture was poured onto aqueous NaHCO3 (satd.), and the aque-
ous layer was extracted twice with CH2Cl2. The combined organic
layers were dried over MgSO4 and concentrated under reduced
pressure. Chromatographic purification (SiO2, cHex/EtOAc 3:1) af-
forded the pure product as a pale-yellow solid (74 mg, 0.38 mmol,
73%). Suitable crystals for X-ray diffraction were obtained through
slow evaporation of a Et2O solution at À208C under an atmos-
phere of pentane. Rf =0.75 in cHex/EtOAc 1:1; mp: 57.1–58.28C.
(3R,3aS,6R,6aR)-6-Hydroxyhexahydrofuro[3,2-b]furan-3-yl
4-
methylbenzenesulfonate ((R)-7). Prepared according to a pub-
lished procedure.[18] A solution of isomannide (2 g, 13.7 mmol,
1 equiv) in pyridine (2.2 mL, 27.3 mmol, 2 equiv) and CH2Cl2 (6 mL)
was cooled to 08C (ice bath), then 4-toluenesulfonyl chloride (TsCl;
3.04 g, 15.9 mmol, 1.2 equiv) was added, the mixture was allowed
to warm to ambient temperature and was stirred for 21.5 h. Then,
the solution was diluted with CH2Cl2, washed with aqueous HCl
(0.5m) and the organic layer was dried over MgSO4. After concen-
trating under reduced pressure, chromatographic purification (SiO2,
cHex/EtOAc 2:1) afforded the pure product as a white solid (1.62 g,
5.4 mmol, 39%). Spectral data are in agreement with those previ-
ously published.[18] Rf =0.18 in cHex/EtOAc 1:1; mp: 105.9–107.78C.
½aŠ25 =156.98 (c=1.03, MeOH); 1H NMR (300 MHz, CDCl3): d=5.38
D
(td, J=5.5, 2.8 Hz, 1H, H-C3ONO2), 5.10 (dd, 1JHF =50.1J=2.5 Hz,
1H, H-C6F), 5.05 (t, J=5.3 Hz, 1H, H-C3a), 4.62 (dd, 3JHF =11.1J=
3
5.1 Hz, 1H, H-C6a), 4.20 (ddd, JHF =22.0J=11.5, 1.2 Hz, 1H, H-C5),
4.01 (ddm, J=11.3, 2.8 Hz, 1H,H-C2), 3.92 (ddd, 3JHF =41.1J=11.4,
2.6 Hz, 1H, H-C5), 3.91 ppm (ddd, J=11.3, 5.3, 0.9 Hz, 1H, H-C2);
½aŠ25 =86.08 (c=1.00, MeOH); 1H NMR (400 MHz, CDCl3): d=7.84–
D
1
7.74 (m, 2H, OTsortho), 7.38–7.32 (m, 2H, OTsmeta), 4.89 (ddd, J=7.5,
6.5, 5.1 Hz, 1H, H-C3OTs), 4.47 (t, J=4.9 Hz, 1H, H-C3a), 4.41 (t, J=
5.0 Hz, 1H, H-C6a), 4.27 (td, J=7.0, 5.5 Hz, 1H, H-C6OH), 4.00 (dd,
J=9.4, 6.6 Hz, 1H, H-C2), 3.94 (dd, J=9.3, 6.4 Hz, 1H, H-C5), 3.77
(dd, J=9.3, 7.5 Hz, 1H, H-C2), 3.53 (dd, J=9.3, 7.2 Hz, 1H, H-C5),
2.44 (s, 3H, CH3 in OTs), 2.39 ppm (s, 1H, C6O-H); 13C NMR
(101 MHz, CDCl3): d=145.4 (OTsipso), 133.1 (OTspara), 130.0 (2C,
OTsmeta), 128.1 (2C, OTsortho), 81.5 (C6a), 80.1 (C3a), 78.5 (C3OTs), 74.1
(C5), 72.4 (C6OH), 70.1 (C2), 21.8 ppm (CH3 in OTs); HRMS (ESI)
calcd m/z: 323.0560, found m/z: 323.0558 [M+Na]+; IR (ATR): n˜ =
3523 (w), 2933 (w), 2867 (w), 1596 (w), 1493 (w), 1455 (w), 1387
(m), 1357 (s), 1305 (m), 1277 (w), 1240 (w), 1188 (m), 1171 (s), 1120
(m), 1098 (m), 1045 (s), 1033 (s), 1017 (s), 973 (m), 921 (m), 883 (m),
840 (s), 816 (s), 795 (s), 751 (m), 707 (m), 666 cmÀ1 (s).
13C NMR (101 MHz, CDCl3): d=95.0 (d, JCF =180.4 Hz, C6F), 86.3 (d,
2JCF =31.5 Hz, C6a), 81.6 (C3a), 81.3 (C3ONO2), 73.9 (d, 2JCF =21.9 Hz,
C5), 69.4 ppm (C2); 19F NMR (282 MHz, CDCl3): d=À190.30 ppm
(dddd, 1JHF =51.5 3JHF =41.2, 22.0, 11.1 Hz); IR (ATR): n˜ =2936 (w),
2880 (w), 1621 (s), 1460 (w), 1380 (w), 1353 (m), 1328 (w), 1312 (w),
1278 (s), 1235 (m), 1207 (w), 1113 (m), 1091 (s), 1074 (s), 1011 (m),
980 (m), 964 (s), 948 (m), 909 (m), 871 (s), 853 (s), 778 (s), 764 (s),
750 (s), 676 cmÀ1 (m).
General methods for radiochemistry. Radiosynthesis was carried
out on a modified PET tracer radiosynthesizer (TRACERLab FxFDG
,
GE Healthcare). The recorded data were processed by TRACERLab
Fx software (GE Healthcare). Separation and purification of the ra-
diosynthesized compounds were performed on the following semi-
preparative radio-HPLC system A: K-500 and K-501 pump, K-2000
UV detector (Herbert Knauer GmbH), NaI(TI) Scintibloc 51 SP51 g-
detector (Crismatec), and an ACE 5 AQ column (250 mm”10 mm).
Method A started with a linear gradient of 10!80% CH3CN in H2O
(0.1% TFA) over 30 min, holding for 15 min and followed by
a linear gradient of 80!10% CH3CN in H2O (0.1% TFA) over 5 min,
with detection at l=220 nm and a flow rate of 5.5 mLminÀ1. Radi-
ochemical purities and specific activities were determined using
the analytical radio-HPLC system B: Two Smartline 1000 pumps
(3R,3aS,6R,6aS)-6-(Nitrooxy)hexahydrofuro[3,2-b]furan-3-yl
4-
methylbenzenesulfonate ((R)-8). A solution of Ac2O (3.14 mL,
33.3 mmol, 10 equiv) in CH2Cl2 (50 mL) was cooled to 08C (ice
bath), then HNO3 (99%, 1.4 mL, 33.6 mmol, 10 equiv) was added
dropwise. After complete addition, stirring at 08C was continued
for additional 10 min. Subsequently, (R)-7 (1.0 g, 3.3 mmol, 1 equiv)
was added in one portion, and the mixture was stirred for 1 h at
08C, then poured onto ice-cold aqueous NaHCO3. The organic
ChemMedChem 2015, 10, 1724 – 1732
www.chemmedchem.org
1730
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
and a Smartline UV detector 2500 (Herbert Knauer GmbH), a Gabi-
Star g-detector (Raytest Isotopenmessgeräte GmbH), and a Nucleo-
sil 100-5 C18 column (250 mm”4 mm). Method B started with
a linear gradient of 10!90% CH3CN in H2O (0.1% TFA) over
15 min, followed by a linear gradient of 90!10% CH3CN in H2O
(0.1% TFA) over 3 min, with detection at l=220 nm and a flow
rate of 1.0 mLminÀ1. The recorded data of both HPLC systems
were processed by the GINA Star software package (Raytest
Isotopenmessgeräte GmbH). No-carrier-added aqueous [18F]fluoride
was produced on an RDS 111e cyclotron (CTI-Siemens) by irradia-
tion of a 2.8 mL water target using 10 MeV proton beams on
97.0% enriched [18O]H2O by the 18O(p,n)18F nuclear reaction.
incubated at 378C. Samples of 20 mL each were taken after periods
of 10, 30, 60, and 90 min and quenched in 100 mL ice-cold DNA-
grade CH3CN followed by centrifugation for ꢀ5 min (2000 g). The
supernatant was analyzed by analytical radio-HPLC system B
(method B, (S)-[18F]6: tR =10.44 min).
Animals. Adult male C57/BL6 mice (n=12, age 15–20 weeks, body
weight 23–27 g) were anesthetized by isoflurane/O2, and one later-
al tail vein was cannulated using a 27-G needle connected to
15 cm polyethylene catheter tubing. (S)-[18F]6 (400 kBqgÀ1 body
weight) was injected as a bolus (100 mL compound flushed with
100 mL saline, syringe pump speed: 300 mLminÀ1) via the tail vein,
and subsequent PET scanning was performed. All experiments in
this study were performed in accordance with German Law on the
Care and Use of Laboratory Animals and were approved by the
local authorizing agency of North Rhine–Westphalia.
Radiosynthesis of (S)-[18F]6. In a computer-controlled TRACERLab
FxFDG synthesizer, the batch of aqueous [18F]fluoride ions (3594–
5077 MBq) from the cyclotron target was passed through an
anion-exchange resin (Sep-Pak Light Waters Accell Plus QMA car-
tridge, preconditioned with 5 mL 1m K2CO3 and 10 mL H2O for in-
jection). [18F]Fluoride ions were eluted from the resin with a mixture
of 40 mL 1m K2CO3, 200 mL H2O for injection, and 800 mL DNA-
grade CH3CN containing 20 mg (53 mmol) Kryptofix2.2.2 (K2.2.2) in
the reactor. Subsequently, the aqueous [18F]K(K2.2.2)F solution was
carefully evaporated to dryness in vacuo. Precursor (R)-8 (20 mg,
58 mmol) in 0.5 mL DNA-grade CH3CN was added, and the mixture
was heated at 110 8C for 20 min. Then the solution was cooled to
508C, 0.5 mL DNA-grade CH3CN and 1.0 mL H2O were added, and
the mixture was passed through a Waters Sep-Pak Alumina N Plus
Light cartridge (preconditioned with 10 mL H2O for injection). The
raw product solution was purified by gradient radio-HPLC system A
(method A). To the product fraction of (S)-[18F]6 (tR =10.21 min),
5.0 mL H2O for injection were added and the mixture was extract-
ed with 5.0 mL CH2Cl2. The organic layer was dried with Na2SO4,
and the Na2SO4 was washed with an additional 2.0 mL CH2Cl2. The
collected organic layers were evaporated to dryness in vacuo at
ambient temperature. The residue was re-dissolved in 1 mL H2O/
EtOH (9:1 v/v) for injection. Product compound (S)-[18F]6 was ob-
tained in a radiochemical yield (rcy) of 14.3Æ4.4% (decay-correct-
ed based on cyclotron-derived [18F]fluoride ions (d.c.), n=5) in
101Æ12 min from the end of radionuclide production. The target
compound was isolated in radiochemical purities of ꢀ99% and
with specific activities of 2–16 GBqmmolÀ1. Radiochemical purities
and specific radioactivities of (S)-[18F]6 (tR =10.44 min) were deter-
mined by analytical radio-HPLC system B (method B).
Small animal PET scanning. PET experiments were carried out
using a sub-millimeter high-resolution (0.7 mm full width at half
maximum) small-animal scanner (32 module quadHIDAC, Oxford
Positron Systems Ltd., Oxford, UK) with uniform spatial resolution
(<1 mm) over a large cylindrical field (165 mm diameter, 280 mm
axial length).[24] List-mode data were acquired for 90 min and re-
constructed into dynamic timeframes using an iterative reconstruc-
tion algorithm. Subsequently, the scanning bed was transferred to
the computed tomography (CT) scanner (Inveon, Siemens Medical
Solutions, USA) and a CT acquisition with a spatial resolution of
80 mm was performed for each mouse. Reconstructed image data
sets were co-registered based on extrinsic markers attached to the
multimodal scanning bed and the image analysis software (Inveon
Research Workplace 3.0, Siemens Medical Solutions, USA). Three-di-
mensional volumes of interest (VOIs) were defined over the respec-
tive organs in CT data sets, transferred to the co-registered PET
data and analyzed quantitatively. Regional uptake was calculated
as percentage of injected dose by dividing counts per milliliter in
the VOI by total counts in the mouse multiplied by 100
(% IDmLÀ1).
Ex vivo tissue counting. Image studies were complemented by
precise preparation of tissue samples after euthanasia of the mice
2 h post-injection of the radioactive compound. Tissue samples
were dabbed dry and weighed on a microbalance (Mettler-Toledo
M3, Mettler-Toledo GmbH, Germany). Then, radioactivity of each
sample was measured (counts per minute; cpm) using an automat-
ic g counter system (2480 Wizard2, PerkinElmer, USA). The results
were normalized to the weight of the tissue sample and to the in-
jected dose and expressed as percentage of injected dose per
gram (% IDgÀ1).
Determination of the partition coefficient (logD (exp.)). The lipo-
philicity of radioligand (S)-[18F]6 was assessed by determination of
the water–octanol partition coefficient following a published pro-
cedure.[25] In brief, ~70 kBq of (S)-[18F]6 were mixed with equal
amounts (0.5 mL) of PBS (pH 7.4) and 1-octanol, and the resulting
biphasic system was mixed vigorously for 3 min at ambient tem-
perature. The tubes were centrifuged (2000 g, 5 min). Then 400 mL
of the 1-octanol layer was separated, and 400 mL PBS were added.
The mixture was mixed vigorously for 10 min at ambient tempera-
ture followed by centrifugation of the tubes for 5 min (2000 g).
Two samples of 100 mL of each layer were counted in a g counter
(Wallac Wizard, PerkinElmer Life Sciences). The partition coefficient
was determined by calculating the ratio cpmoctanol/cpmPBS and ex-
pressed as logD (exp.) [log(cpmoctanol/cpmPBS)]. Three independent
experiments were performed in duplicate, and data were provided
as mean values Æ standard deviation.
Acknowledgements
We gratefully acknowledge generous financial support from
WWU Münster, the Swiss National Science Foundation (SNF)
(P2EZP2-148757, N.S.), the Deutsche Forschungsgemeinschaft
(DFG) (SFB 656 C06, Z05 and SFB 858), the Interdisciplinary
Centre of Clinical Research (IZKF Münster, core unit PIX), the Clus-
ter-of-Excellence DFG EXC 1003 “Cells in Motion”, and the Europe-
an Research Council (ERC-2013-StG Starter Grant project number
336376-ChMiFluorS).
Stability in human serum. The serum stability of radioligand (S)-
[18F]6 was evaluated by incubation in human serum at 378C for up
to 90 min. An aliquot of formulated (S)-[18F]6 (20 mL, 4.6 MBq) was
added to a sample of human serum (200 mL), and the mixture was
Keywords: biodistribution · dynamic PET · fluorine · organic
nitrates · radiosynthesis
ChemMedChem 2015, 10, 1724 – 1732
www.chemmedchem.org
1731
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
[1] W. Murrell, Lancet 1879, 113, 225–227.
[2] W. Murrell, Lancet 1879, 113, 80–81.
[3] W. Murrell, Lancet 1879, 113, 113–115.
[4] W. Murrell, Lancet 1879, 113, 151–152.
[5] World Health Organization, Fact Sheet #310, The top 10 causes of death:
www.who.int/mediacentre/factsheets/fs310/en/ (accessed May 20,
2015).
[16] A. Ali, I. K. Sebhat, C. L. Franklin, K. M. Rupprecht, R. K. Baker, R. P. Nar-
gund, L. Yan, P. Huo, D. Shen, N. Almirante, S. Biondi, M. Ferrario, A. Nic-
otra (Cicletanine Derivatives), US Pat. No. US 20100029678 A1, 2012.
[17] K. Sandrock, G. Cordes (Sanol Schwarz-Monheim, GmbH), Pat No.
CA 1132993 A1, 1982.
[18] S. Kumar, U. Ramachandran, Tetrahedron 2005, 61, 4141–4148.
[19] K. B. Wiberg, J. Org. Chem. 2003, 68, 9322–9329.
[20] F. H. Allen, C. A. Baalham, J. P. M. Lommerse, P. R. Raithby, E. Sparr, Acta
Crystallogr. Sect. B 1997, 53, 1017–1024.
[21] Calculated logD value (clogD) was determined by ACD/ChemSketch
version 6.00, ACD/Labs; [logD=logP at physiological pH (7.4)].
[22] H. L. Fung, S. C. Sutton, A. Kamiya, J. Pharmacol. Exp. Ther. 1984, 228,
334–341.
[23] T. B. Tzeng, H. L. Fung, J. Pharmacol. Exp. Ther. 1992, 261, 692–700.
[24] K. P. Schäfers, A. J. Reader, M. Kriens, C. Knoess, O. Schober, M. Schäfers,
J. Nucl. Med. 2005, 46, 996–1004.
[6] K. E. Torfgård, J. Ahlner, Cardiovasc. Drugs Ther. 1994, 8, 701–717.
[7] M. G. Bogaert, Cardiovasc. Drugs Ther. 1994, 8, 693–699.
[8] U. Abshagen, G. Betzien, R. Endele, B. Kaufmann, Eur. J. Clin. Pharmacol.
1981, 20, 269–275.
[9] U. W. Abshagen, Am. J. Cardiol. 1992, 70, G61–G66.
[10] K. S. Ravikumar, S. Chandrasekaran, Synthesis 1994, 1032–1034.
[11] C. Brown, R. W. Marston, P. F. Quigley, S. M. Roberts, J. Chem. Soc. Perkin
Trans. 1 2000, 1809–1810.
[12] R. Seemayer, N. Bar, M. P. Schneider, Tetrahedron: Asymmetry 1992, 3,
1123–1126.
[25] O. Prante, C. Hocke, S. Lçber, H. Hübner, P. Gmeiner, T. Kuwert, Nuklear-
medizin 2006, 45, 41–48.
[26] L. D. Hayward, D. J. Livingstone, M. Jackson, V. M. Csizmadia, Can. J.
Chem. 1967, 45, 2191–2194.
[13] J. F. Gilmer, L. M. Moriarty, D. F. McCafferty, J. M. Clancy, Eur. J. Pharm.
Sci. 2001, 14, 221–227.
[14] J. A. Kanters, A. Schouten, G. J. Sterk, M. H. de Jong, J. Mol. Struct. 1993,
298, 113–120.
Received: June 25, 2015
Published online on August 12, 2015
ˇ
ˇ
´
´
[15] z. Cekovic, Z. Tokic, Synthesis 1989, 610–612.
ChemMedChem 2015, 10, 1724 – 1732
www.chemmedchem.org
1732
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim