Syntheses and: In vitro biological evaluation of S1PR1 ligands and PET studies of four F-18 labeled radiotracers in the brain of nonhuman primates

Article abstract of DOI:10.1039/c8ob02609b

A series of seventeen hydroxyl-containing sphingosine 1-phosphate receptor 1 (S1PR1) ligands were designed and synthesized. Their in vitro binding potencies were determined using [³²P]S1P competitive binding assays. Compounds 10a, 17a, 17b, and 24 exhibited high S1PR1 binding potencies with IC₅₀ values ranging from 3.9 to 15.4 nM and also displayed high selectivity for S1PR1 over other S1P receptor subtypes (IC₅₀ > 1000 nM for S1PR2-5). The most potent compounds 10a, 17a, 17b, and 24 were subsequently radiolabeled with F-18 in high yields and purities. MicroPET studies in cynomolgus macaque showed that [¹⁸F]10a, [¹⁸F]17a, and [¹⁸F]17b but not [¹⁸F]24 crossed the blood brain barrier and had high initial brain uptake. Further validation of [¹⁸F]10a, [¹⁸F]17a, and [¹⁸F]17b in preclinical models of neuroinflammation is warranted to identify a suitable PET radioligand to quantify S1PR1 expression in vivo as a metric of an inflammatory response.

Full text of DOI:10.1039/c8ob02609b

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
Syntheses and in vitro biological evaluation of
S1PR1 ligands and PET studies of four F-18 labeled
radiotracers in the brain of nonhuman primates†
Cite this: Org. Biomol. Chem., 2018,
16, 9171
a
Zonghua Luo, a Junbin Han,a Hui Liu, a Adam J. Rosenberg,
Delphine L. Chen,a Robert J. Gropler,a Joel S. Perlmutter a,b and Zhude Tu
*
a
A series of seventeen hydroxyl-containing sphingosine 1-phosphate receptor 1 (S1PR1) ligands were
designed and synthesized. Their in vitro binding potencies were determined using [32P]S1P competitive
binding assays. Compounds 10a, 17a, 17b, and 24 exhibited high S1PR1 binding potencies with IC50
values ranging from 3.9 to 15.4 nM and also displayed high selectivity for S1PR1 over other S1P receptor
subtypes (IC50 > 1000 nM for S1PR2–5). The most potent compounds 10a, 17a, 17b, and 24 were sub-
sequently radiolabeled with F-18 in high yields and purities. MicroPET studies in cynomolgus macaque
showed that [18F]10a, [18F]17a, and [18F]17b but not [18F]24 crossed the blood brain barrier and had high
initial brain uptake. Further validation of [18F]10a, [18F]17a, and [18F]17b in preclinical models of neuro-
inflammation is warranted to identify a suitable PET radioligand to quantify S1PR1 expression in vivo as a
metric of an inflammatory response.
Received 22nd October 2018,
Accepted 15th November 2018
DOI: 10.1039/c8ob02609b
rsc.li/obc
remitting multiple sclerosis (RR-MS). MS is a chronic auto-
immune, inflammatory disease caused by lymphocytic infiltra-
Introduction
Sphingosine 1-phosphate (S1P) is a membrane-derived lyso- tion that leads to a demyelinating neurodegenerative disease
phospholipid that plays a critical regulatory role in inflamma- with no cure. RR-MS is the most common form, characterized
tory diseases.1–3 S1P acts by modulating five highly specific G by intermittent relapses that lead to increasing functional dis-
protein-coupled receptors (GPCRs, S1PR1–5) and has high ability that in large part may be mediated by neuroinflamma-
binding affinities (nanomolar) with these five receptors.4–8 The tory responses in the brain.10 Fingolimod, structurally similar
unique patterns of cellular and temporal expression determine to S1P, functionally modulates S1PR1 on lymphocytes, leading
the specific roles of S1PR1–5 in each organ system. to S1PR1 downregulation and subsequent lymphocyte trapping
Modulation of these receptors is particularly important in the in lymph nodes. This reduces the peripheral blood lymphocyte
central nervous, cardiovascular, and immune systems. Among count and presumably lymphocyte infiltration in the brain.
these five-receptor subtypes, S1PR1 is one of the most abun- The effects of fingolimod on S1PR1 expression in the central
dant receptors in the entire GPCR superfamily. S1PR1 was nervous system (CNS) itself and its contribution to treatment
originally called the endothelial differentiation growth factor efficacy are not well understood. S1PR1 is expressed in
receptor 1 (EDG-1). It is particularly enriched in endothelial neurons and glia in the CNS, including astrocytes, which
cells and vascular smooth muscle cells, and also expressed on modulate inflammatory responses throughout the gray and
immune cells such as lymphocytes.9 The relevance of S1PR1 in white matter; microglia, the resident macrophages in the CNS,
clinical disease has become readily apparent with the U.S. and oligodendrocytes, which produce the myelin needed for
Food and Drug Administration (FDA) approval of the nerve conduction. In the in vitro models of demyelination
S1PR1 modulator fingolimod (FTY720) for treating relapsing- induced by lysophosphatidylcholine (LPS), fingolimod reduced
demyelination and neuronal process extension in the CNS, pre-
sumably through modulation of S1PR1.11–13 It also inhibited
aDepartment of Radiology, Washington University School of Medicine,
510 South Kingshighway Boulevard, St Louis, Missouri, 63110, USA.
E-mail: tuz@mir.wustl.edu; Tel: +1-314-362-8487
bDepartments of Neurology, Neuroscience, Physical Therapy and Occupational
Therapy, Washington University School of Medicine, St Louis, MO 63110, USA
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
c8ob02609b
microglial activation and reduced oligodendrocyte apoptosis.13
Specific deletion of S1PR1 in astrocytes resulted in decreased
experimental autoimmune encephalomyelitis (EAE) pathology
in vivo and a loss of fingolimod efficacy, suggesting that fingo-
limod also acts on S1PR1 in astrocytes.14,15 Fingolimod-treated
MS patients exhibited reduced brain volume loss, lesion
This journal is © The Royal Society of Chemistry 2018
Org. Biomol. Chem., 2018, 16, 9171–9184 | 9171

View Article Online
Paper
Organic & Biomolecular Chemistry
numbers, and severity,14,16–18 confirming the in vitro and clinical investigation.30,31 Our group explored oxadiazole-con-
animal model findings that S1PR1 pathways play important taining analogues and reported two [18F]fluoroethoxy radiotra-
roles in anti-inflammation and neuroprotection. A S1PR1-tar- cers, one having an azetidine-3-carboxylic acid ([18F]5), and
geted positron emission tomography (PET) radiotracer would another tracer having a 1H-1,2,3-triazole-4-methanol ([18F]6).32,33
help us understand the pathophysiology of S1PR1 in the CNS. Although the in vivo study suggested that the mouse liver
Imaging S1PR1 in vivo could be used to develop therapies that uptake of [18F]5 had a strong positive correlation with acute
improve upon the initial treatment successes with fingolimod. liver inflammation induced by LPS injection, this radiotracer
Furthermore, S1PR1 is a therapeutic target of great interest for has limited penetration of the BBB for imaging neuroinflam-
a number of other inflammatory diseases, including inflam- mation. Our preliminary data revealed that [18F]6 exhibited a
matory bowel disease19 and atherosclerotic disease,20,21 as well high rat brain uptake with 0.71%ID per gram at 60 min post
as cancer, where S1PR1 expression promotes treatment resis- injection and no in vivo defluorination. A specific S1PR1
tance22,23 and enhances metastatic potential.23–26 S1PR1 ligand reduced the uptake of [18F]6 binding to S1PR1 in tissue
imaging would enable mechanistic studies and S1PR1-specific sections of the inflamed brain from an LPS-induced neuro-
readouts of targeted therapy in a wide range of inflammatory inflammation mouse model. But follow-up microPET imaging
diseases. Using PET with the promising C-11 radiotracer [11C] studies of [18F]6 in cynomolgus macaque suggested this radio-
TZ3321, we demonstrated increased S1PR1 expression at tracer did not penetrate the nonhuman primate (NHP) brain
inflammatory response sites in three animal models: the rat (seen in ESI†). Species differences in the BBB penetration
EAE model of MS, the ApoE−/− mouse femoral artery wire- ability of PET radiotracers have been previously reported for
injury model of neointimal hyperplasia, and the rat carotid other PET neuroimaging radiotracers.34 Consequently, neither
injury model of vascular inflammation.27–29 The longer half- [18F]5 nor [18F]6 is a suitable PET S1PR1 radiotracer to assess
life of F-18 (T1/2 = 109.8 min) reduces time constraints on the neuroinflammatory responses in humans. To overcome the
tracer production, distribution, and permits longer imaging above-mentioned challenges in the identification of a suitable
sessions for higher target-to-reference ratios than is possible [18F]-labeled S1PR1 radiotracer for neuroinflammation
with C-11. Therefore, the PET radiochemistry community has response in CNS disorders, we focused on further optimization
focused on the development of potent and selective [18F]- of the oxadiazole analogues by replacing the carboxylic acid
labeled S1PR1 radiotracers to quantitatively measure the moiety with hydroxyl group(s) in the terminal of substituted
changes of S1PR1 expression in response to inflammation. side chain. Our previous rat biodistribution of [18F]6 showed
Several [18F]-labeled small molecules have been characterized good brain uptake compared to carboxylic acid containing ana-
for their suitability to be [18F]-labeled PET radiotracers for logues. Herein, we reported the syntheses of new serial of
imaging S1PR1 in vitro and in vivo (Fig. 1). However, either terminal hydroxyl-containing S1PR1 compounds, in vitro
they had inadequate in vivo metabolic stability or did not pene- binding assays to determine their binding potencies and
trate the blood brain barrier (BBB). For instance, three [18F]- selectivity for S1PR1, radiosyntheses of four F-18 radiotracers,
labeled fingolimod derivatives, [18F]3, [18F]4a, [18F]4b reported and microPET studies of these four radiotracers in the
by Haufe’s group showed fast kinetics, low selectivity, or cynomolgus macaque brain to investigate their in vivo kinetics
defluorination in vivo, which prevented their transfer into and ability to penetrate the BBB.
Fig. 1 Structures of fingolimod and S1PR1 radiotracers.
9172 | Org. Biomol. Chem., 2018, 16, 9171–9184
This journal is © The Royal Society of Chemistry 2018

View Article Online
Organic & Biomolecular Chemistry
Paper
Compounds 13a–c were synthesized by employing a similar
procedure described for compounds 10a–c. Treatment of com-
Results and discussion
To identify an [18F]-labeled radiotracer with high potency, high pounds 11a–c with hydroxylamine hydrochloride yielded the
selectivity, and the capability to penetrate the BBB. Initially, intermediates 12a–c, which were subjected to condensation
the (1,2,3-triazol-4-yl)methanol moiety in compound 6 was and then cyclization with compound 9a to yield compounds
replaced by hydroxyl methyl functional group to generate com- 13a–c. Compound 13d was prepared by deprotection of tert-
pound 10a, we then replaced the trifluoromethyl group with a butyloxy-carbonyl (Boc) group using 4.0 M HCl in dioxane
cyano or hydrogen group (Scheme 1). Next, we turned our (Scheme 2).
attention to retaining the trifluoromethyl group and replaced
Compound 10a was utilized as a key intermediate for the
the hydroxyl methyl group of 10a with alkyl alcohols, amino syntheses of analogues 15a–b and 17a–g. Swern oxidation of
alcohols, and ether alcohols. In total, seventeen new analogues the primary alcohol in compound 10a with oxalyl chloride in
were designed, synthesized and screened for in vitro and the presence of triethylamine at −78 °C to yield the intermedi-
in vivo properties. The syntheses of the target compounds are ate aldehyde 14,32 followed by reductive amination using
depicted in Schemes 1–5 following reported procedures with corresponding amines generated the amino alcohol com-
necessary modifications.32,33 For compounds 10a–c, syntheses pounds 15a–b (Scheme 3).
were initiated with 4-(hydroxymethyl)-benzonitrile (7) treated
Chlorination of compound 10a was achieved by reaction
with hydroxylamine hydrochloride in presence of NaHCO3 to with cyanuric chloride in DMF to afford the intermediate
yield amidoxime 8. Condensation and cyclization of the amid- chloride 16, followed by nucleophilic substitution with various
oxime 8 with aromatic carboxylic acids 9a–c in DMF in pres- alcohols to yield compounds 17a–e and 17g. Compound 17f
ence of 2-(1H-benzotriazole-1-yl) 1,1,3,3-tetra-methylaminium was generated by deprotection of Boc group of 17e (Scheme 3).
tetrafluoroborate (TBTU) and a catalytic amount of 1-hydroxy-
benzo-triazole hydrate (HOBt) to generate compounds 10a–c ponding 4-(2-hydroxy-ethoxy)benzonitrile (18a) or 4-((2,3-dihy-
(Scheme 1). droxy-propoxy)methyl)benzonitrile (18b) to generate amidox-
Compounds 20a–b were synthesized using the corres-
Scheme 1 Syntheses of 10a–c. Reagents and conditions: (a) Hydroxylamine hydrochloride, NaHCO3, methanol, reflux; (b)TBTU, HOBt, DIPEA, DMF,
RT–120 °C.
Scheme 2 Syntheses of 13a–d. Reagents and conditions: (a) Hydroxylamine hydrochloride, NaHCO3, methanol, reflux; (b) 9a, TBTU, HOBt, DIPEA,
DMF, RT–120 °C; (c) 4.0 M HCl in dioxane, RT.
This journal is © The Royal Society of Chemistry 2018
Org. Biomol. Chem., 2018, 16, 9171–9184 | 9173

View Article Online
Paper
Organic & Biomolecular Chemistry
Scheme 3 Syntheses of 15a–b and 17a–g. Reagents and conditions: (a) Oxalyl chloride, DMSO, CH2Cl2, triethylamine, −78 °C–RT; (b) NaBH3CN
amines, acetic acid, methanol, RT; (c) cyanuric chloride, DMF, CH2Cl2, RT; (d) NaH, alcohols, THF, reflux; (e) 4.0 M HCl in dioxane, RT.
Scheme 4 Syntheses of 20a–b. Reagents and conditions: (a) Hydroxylamine hydrochloride, NaHCO3, methanol, reflux; (b) 9a, TBTU, HOBt, DIPEA,
DMF, RT–120 °C.
Scheme 5 Syntheses of ( )-24. Reagents and conditions: (a) Hydroxylamine hydrochloride, NaHCO3, methanol, reflux; (b) 9a, TBTU, HOBt, DIPEA,
DMF, RT–120 °C; (c) 4-methylmorpholine N-oxide, OsO4, THF, H2O, RT.
imes 19a or 19b, followed by condensation and cyclization methyl with a cyano, also an electron withdrawing group,
with 9a to yield compounds 20a or ( )-20b (Scheme 4).
resulted in a 2.3-fold decreased potency (10b, IC50 = 15.4 3.8
Finally, compound ( )-24 was prepared from 4-vinylbenzo- nM). Removal of the trifluoromethyl group significantly
nitrile (21) reacted with hydroxylamine hydrochloride to yield reduced the binding potency of compound 10c (IC50 > 1000
the intermediate compound 22, which was subjected to con- nM). Because the CF3 group is a key functional group for
densation followed by cyclization with compound 9a to yield retaining high S1PR1 potency, we retained the trifluoromethyl
compound 23. The terminal olefin in compound 23 was oxi- group in our further exploration of new analogues. In the
dized with 4-methylmorpholine N-oxide and catalytic OsO4 in second series, different amino alcohols were introduced for
THF resulting in target compound ( )-24 (Scheme 5).
compounds ( )-13a, 13b, 13d, 15a, and 15b. We wanted to
We determined the in vitro binding potency of these com- determine if different amino alcohol-containing groups
pounds to S1PR1 as a measure of biological activity by follow- caused significant changes of S1PR1 binding potency. Our
ing our published radioligand assay using freshly synthesized in vitro data showed that all the amino alcohol derivatives
[32P]S1P.32,33,35 The S1PR1 binding data of these new ana- ( )-13a, 13b, 13d, 15a, and 15b were less potent than com-
logues are presented in Table 1. Here we present our obser- pound 10a with IC50 values of 39.9, 38.0, 48.2, 87.2, and 125
vations regarding structure activity relationships. First, among nM, respectively. The above results suggest that the amino
compounds 10a–c, compound 10a with an electron withdraw- alcohols are not favorable to improve the S1PR1 binding
ing trifluoromethyl at the meta-position of the oxadiazole potency.
moiety displayed a high binding potency with an IC50 value of
In the third series, we PEGylated compound 10a using
6.7 nM for S1PR1 receptor.32 Replacement of the trifluoro- different PEG units. PEGylation has been reported as an
9174 | Org. Biomol. Chem., 2018, 16, 9171–9184
This journal is © The Royal Society of Chemistry 2018

View Article Online
Organic & Biomolecular Chemistry
Paper
Table 1 Structures and binding potencies (mean SD) of S1P and newly synthesized compounds against S1PR1a
S1PR1
IC50 SD (nM)
S1PR1
IC50 SD (nM)
Compound
R1
R2
Compound
R1
R2
S1P
—
—
1.4 0.3
17a
CF3
14.0 0.4
10a
10b
10c
( )-13a
CF3
CN
H
6.7 0.7b
15.4 3.8
>1000
17b
17c
17d
17f
CF3
CF3
CF3
CF3
15.4 3.3
106 25
125 22
CF3
39.9 3.6
75.6 19.3
13b
CF3
38.0 6.0
17g
CF3
130 19
13d
15a
CF3
CF3
48.2 9.7
20a
CF3
CF3
56.6 23.8
23.8 5.8
87.2 17.7
( )-20b
15b
CF3
125 27
( )-24
CF3
3.9 0.5
a IC50 values were determined with at least 3 independent experiments, each run was performed in duplicate. b Ref. 32.
effective strategy for improving the solubility, stability, phar- an ethylene glycol directly linked with phenyl group was found
macodynamics, and pharmacokinetics of drug candidates to be the most potent S1PR1 ligand of this study with an IC50
interacting with biologically interesting proteins or value of 3.9 nM. Because compounds 10a, 17a, 17b, and ( )-24
peptides.36–42 The addition of a small number of PEGylaton each had high binding potency for S1PR1 with IC50 < 20 nM,
units (n = 2, 3, 4) is frequently used to improve a compound’s these four compounds were screened to determine their
hydrophilicity which may facilitate crossing the BBB. As one binding potency to other S1P receptor subtypes S1PR2, 3, 4
example, [18F]AV45, the well-known radiotracer used for and 5 to check their binding selectivity. Our in vitro data
imaging β-amyloid (Aβ) aggregates in the AD brain contains shown in Table 2 revealed that all four compounds had no sig-
three PEGylation units.41 Therefore, we next synthesized com- nificant binding toward S1PR2–5 (IC50 > 1000 nM), indicating
pounds 17a–d with 1–4 PEGs and tested their binding potency. that they are selective for S1PR1 over other S1PR subtypes
Our in vitro binding data showed that compound 17a and 17b S1PR2, 3, 4, and 5.
with one and two PEG units had similar binding potencies
Consequently, these four compounds were radiolabeled
with IC50 values of 14.0 and 15.4 nM, respectively. Compound with F-18 to check their potential to be neuroimaging PET
17c (IC50 = 106 nM) with three PEGs, had approximately radiotracers for quantifying S1PR1 expression in the brain.
16-fold lower potency compared to 10a. Compound 17d (IC50
=
Prior to radiolabeling, the corresponding precursors 27a–d
125 nM) with four PEGs also showed further decreased were synthesized by following Scheme 6. Methoxymethyl
binding potency.
(MOM) protected hydroxybenzimidamide 25a–d were prepared
The in vitro S1P binding data of these new PEG-containing
compounds suggested that the addition of one or two PEG
units retained S1PR1 binding potency, but the extension with
more PEG chains (>2 units) decreased S1PR1 binding potency.
In addition, for compounds 17f and 17g, in which a secondary
or tertiary amine was introduced in the PEG chains, S1PR1
binding potencies decreased with IC50 values of 75.6 and 130
nM, respectively. Finally, we modified compound 20a with a
2-hydroxyethoxyl group directly linked to a phenyl group, this
approach also reduced binding potency (IC50 = 56.6 nM) when
compared to compound 17a with a 2-hydroxy ethoxyl linked
with a benzyl group. Further optimization of 17a was con-
ducted by introducing a terminal hydroxyl group using gly-
cerin to yield compound ( )-20b, which showed an IC50 value
of 23.8 nM, which is less potent than the glycol tail compound
17a. The alkyl di-alcohol compound ( )-24, which incorporated
Table 2 Binding potencies (mean
toward S1PR1–5 and Clog P values
SD) of 10a, 17a–b, and ( )-24
IC50 a (nM)
Compd. S1PR1
S1PR2
S1PR3
S1PR4
S1PR5
Clog Pb
S1P
1.4 0.3 3.6
0.4
0.2
151
82
3.1
1.1
0.5
10a
17a
17b
( )-24
6.7 0.7 >1000
14.0 0.4 >1000
15.4 3.3 >1000
3.9 0.5 >1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
4.00
4.02
3.85
3.09
a IC50 values were determined with at least 3 independent experiments,
each run was performed in duplicate; assays for compounds which
showed no activity (IC50 > 1000 nM) were repeated twice. b Clog P
values were calculated by PerkinElmer ChemDraw® 16.0.1.4 (77).
This journal is © The Royal Society of Chemistry 2018
Org. Biomol. Chem., 2018, 16, 9171–9184 | 9175

View Article Online
Paper
Organic & Biomolecular Chemistry
Scheme 6 General syntheses of precursors 27a–d and radiotracers [18F]10a, [18F]17a, [18F]17b, and ( )-[18F]24. Reagents and conditions: (a)
4-Hydroxy-3-(trifluoromethyl)benzoic acid, TBTU, HOBt, DIPEA, DMF, RT–120 °C; (b) ethylene ditosylate, K2CO3, CH3CN, 90 °C; (c) [18F]KF, Kryptofix
222, K2CO3, CH3CN, 110 °C, 15 min, 6 N HCl, 5 min.
by following the general procedure described for 12a–c with performed for each F-18 labeled radiotracer on different days.
MOM protected benzonitriles. The condensation and cycliza- Macaque did not have studies more frequently than every 2
tion of 25a–d and 4-hydroxy-3-(trifluoro-methyl)benzoic acid weeks to allow the subject to recover. Time activity curves
afforded 26a–d, which were subjected to O-alkylation with (TACs) showing brain uptake and washout from our microPET
ethylene ditosylate to afford MOM-protected precursors 27a–d. studies of these four F-18 radiotracers are shown in Fig. 2,
The radiosyntheses of [18F]10a, [18F]17a, [18F]17b, and ( )-[18F] brain TACs indicated that three radiotracers [18F]10a, [18F]17a,
24 were successfully achieved by using nucleophilic reaction and [18F]17b were able to cross the BBB and had high initial
between each tosylate precursor and [18F]KF in acetonitrile brain uptake in the nonhuman primates.
with Kryptofix 222, followed by deprotection of MOM group.
The brain standard uptake value (SUV) for [18F]10a, [18F]17a,
The radioactive products were purified using a semi-prepara- and [18F]17b reached a maximum at 4–6 min after tracer
tive high performance liquid chromatography (HPLC) and injection. [18F]10a had the highest initial brain uptake and
then formulated using 10% of ethanol saline solution. Each quick washout kinetics from the brain while the brain washout
dose of [18F]10a, [18F]17a, [18F]17b, and ( )-[18F]24 was auth- kinetics of [18F]17b was relatively slow. For the most potent
enticated by co-injecting with corresponding cold standard compound ( )-24, with an IC50 value of 3.9 nM, our microPET
compound 10a, 17a, 17b, and ( )-24, respectively. These four studies of ( )-[18F]24 indicated that it did not penetrate the
radiotracers were achieved with radiochemical yields of BBB and its brain accumulation was very low. Lipophilicity of
35–40%, radiochemical purities >98%, and molar activities small molecule drug is a key parameter affecting penetration
>74 GBq µmol−1 (decay corrected to end of synthesis, EOS).
of the BBB. log P value is an index of small molecular lipophili-
To check if each radiotracer had the capability to penetrate city. Most drugs that are active in the central nervous system
the BBB with sufficient accumulation in the animal brain, have log P values that range from 1.5 to 2.7,43 though some
microPET studies of the brain of a cynomolgus macaque were have values beyond this range. The calculated log P (Clog P)
Fig. 2 Representative time-activity curves of microPET studies for [18F]10a, [18F]17a, [18F]17b and ( )-[18F]24 in brains of nonhuman primates. The
curves indicated that three radiotracers [18F]10a, [18F]17a and[18F]17b achieved the max SUV value fast, at 4–6 min post injection. [18F]10a had the
highest initial brain uptake while [18F]17b exhibited relatively slow washout kinetics. ( )-[18F]24 displayed the lowest initial uptake in the brain, indicat-
ing it has no capability in penetrating the BBB of nonhuman primate.
9176 | Org. Biomol. Chem., 2018, 16, 9171–9184
This journal is © The Royal Society of Chemistry 2018

View Article Online
Organic & Biomolecular Chemistry
Paper
values (Table 2) of these four tracers were 4.00, 4.02, 3.85, and internal reference (CDCl3: δ 7.26 ppm; CD3OD: δ 3.31 ppm;
3.09 for [18F]10a, [18F]17a, [18F]17b, and ( )-[18F]24, respect- DMSO-d6: δ 2.50 ppm; acetone-d6: δ 2.05 ppm). Multiplicities
ively. The most potent radiotracer ( )-[18F]24 (IC50 value of 3.9 are indicated by s (singlet), d (doublet), t (triplet), q (quartet),
nM for S1PR1) had a log P value of 3.09 which would predict p (pentet), h (hextet), m (multiplet) and br (broad). High-
better BBB penetration. The unanticipated low nonhuman resolution positive ion mass was acquired by a Bruker MaXis
primate brain uptake of ( )-[18F]24 despite its lipophilicity may 4G Q-TOF mass spectrometer with electrospray ionization
have resulted from the two hydroxyl groups causing different source. The welfare of the animals conformed to the require-
in vivo pharmacological properties, such as binding to ments of National Institutes of Health (NIH). This work was
P-glycoprotein (Pgp).44 Further investigation would be needed conducted at the NHP Facility of Washington University in
to understand the inability of ( )-[18F]24 in penetrating BBB.
St Louis with approval from the IACUC.
Together, our microPET studies in nonhuman primates
(E)-N′-Hydroxy-4-(hydroxymethyl)benzimidamide
(8).
suggested that radiotracers [18F]10a, [18F]17a, and [18F]17b Synthesis of compound 8 followed the published procedure.33
have potential to be promising S1PR1 PET radiotracers for
imaging the neuroinflammatory response in vivo. Further
General procedure for the synthesis of 10a–c
investigations in animal models of neuroinflammatory dis- To a round-bottom flask equipped with a stir bar was added
eases are warranted to characterize their radiopharmaceutical acid 9a–c (1.0 eq.), HOBt (0.2 eq.), TBTU (1.0 eq.), DIPEA
properties.
(3.0 eq.), and DMF (5.0 mL mmol−1). The reaction mixture was
stirred for 0.5 h followed by adding amidoxime 8 (1.0 eq.). The
reaction mixture was stirred for 1 h at room temperature (RT),
then refluxed in a pre-heated 120 °C oil-bath for 4 h and moni-
tored by TLC. After cooling, the reaction mixture was diluted
Conclusions
In conclusion, we developed a series of new S1PR1 ligands and with water and extracted with ethyl acetate. The ethyl acetate
screened in vitro binding potencies for all the newly syn- layer was washed with 1 M HCl, saturated brine, and dried
thesized compounds. Compounds 10a, 17a, 17b, and ( )-24 over anhydrous MgSO4. After filtration and concentration, the
exhibited high potency (IC50 < 20 nM) and high selectivity for residue was purified on a silica gel column to give the product
S1PR1 (IC50 > 1000 nM over S1PR2, 3, 4, and 5). The radio- 10a–c.
syntheses of [18F]10a, [18F]17a, [18F]17b, and ( )-[18F]24 were
(4-(5-(4-(2-Fluoroethoxy)-3-(trifluoromethyl)phenyl)-1,2,4-
accomplished with good yields (35–40%), high radiochemical oxadiazol-3-yl)phenyl)methanol (10a). The characterization is
and chemical purities (>98%), and reasonable molar activities found in our previous publication.32
(>74 GBq µmol−1). MicroPET studies in cynomolgus macaque
2-(2-Fluoroethoxy)-5-(3-(4-(hydroxymethyl)phenyl)-1,2,4-oxa-
indicated that [18F]10a, [18F]17a, and [18F]17b can cross the diazol-5-yl)benzonitrile (10b). Compound 10b was purified by
BBB of NHPs and have good brain uptake while ( )-[18F]24 was flash chromatography, eluted with hexane/ethyl acetate (1/1,
not able to penetrate the BBB. Further in vivo evaluation of v/v) to afford white solid product. Yield: 30%, MP: 150–151 °C.
[18F]10a, [18F]17a, and [18F]17b in animal models of inflamma- 1H NMR (400 MHz, acetone-d6) δ 8.52–8.34 (m, 2H), 8.14–7.95
tory disease mechanisms is needed to check the suitability of (m, 2H), 7.64–7.43 (m, 3H), 5.04–4.85 (m, 2H), 4.78 (d, J =
these radiotracers for assessing the neuroinflammatory 5.0 Hz, 2H), 4.73–4.60 (m, 2H), 4.49 (br, 1H). 13C NMR
response by measuring the S1PR1 expression prior to trans- (101 MHz, acetone-d6) δ 173.8, 168.6, 163.1, 146.3, 134.2,
lation of a lead radiotracer to clinical investigations.
133.5, 127.1, 126.8, 125.1, 117.4, 114.6, 113.7, 102.9, 81.5 (d,
JC–F = 170 Hz), 69.1 (d, JC–F = 18.1 Hz), 63.3. HRMS (ESI) calcd
for C18H14F1N3O3 [M + H+] 340.1092, found 340.1087.
(4-(5-(4-(2-Fluoroethoxy)phenyl)-1,2,4-oxadiazol-3-yl)-phenyl)-
methanol (10c). Compound 10c was purified by flash chrom-
atography, eluted with hexane/ethyl acetate (1/1, v/v) to afford
Experimental section
General information
All reagents and chemicals were obtained from standard com- white solid product. Yield: 45%, MP: 140–142 °C. 1H NMR
mercial sources and used without further purification unless (400 MHz, DMSO-d6) δ 8.21–8.12 (m, 2H), 8.05 (d, J = 8.2 Hz,
otherwise stated. Organic reactions were carried out under 2H), 7.54 (d, J = 8.0 Hz, 2H), 7.24 (d, J = 8.9 Hz, 2H), 5.38 (s,
inert nitrogen and moisture-free with dry solvent. Thin layer 1H), 4.94–4.74 (m, 2H), 4.61 (s, 2H), 4.53–4.34 (m, 2H). 13C
chromatography (TLC) was used to monitor the reaction. Final NMR (101 MHz, DMSO-d6) δ 175.5, 168.5, 162.4, 146.8, 130.4,
organic products were purified by flash column chromato- 127.4, 127.3, 125.0, 116.5, 115.9, 83.2, 81.6, 68.0, 67.8, 62.9.
graphy using 230–400 mesh silica gel purchased from Silicycle. HRMS (ESI) calcd for C17H15FN2O3 [M + H+] 315.1139, found
Melting points were determined on a MEL-TEMP 3.0 apparatus 315.1139.
without correction. All deuterated solvents were purchased
4-(((2,3-Dihydroxypropyl)(methyl)amino)methyl)benzonitrile
from Cambridge Isotope Laboratories. 1H and 13C NMR (11a). To a round-bottom flask equipped with a stir bar was
spectra were recorded on a 400 MHz Varian instrument. added 4-(bromomethyl)benzonitrile (3.9 g, 20.0 mmol),
Chemical shifts were reported in parts per million (ppm) and 3-(methylamino)propane-1,2-diol (2.5 g, 23.0 mmol), K2CO3
were calibrated using a residual undeuterated solvent as an (6.9 g, 50.0 mmol), and acetonitrile (50 mL). The mixture was
This journal is © The Royal Society of Chemistry 2018
Org. Biomol. Chem., 2018, 16, 9171–9184 | 9177

View Article Online
Paper
Organic & Biomolecular Chemistry
stirred overnight at RT. Then, the mixture was concentrated
(E)-N′-Hydroxy-4-(((2-hydroxyethyl)(methyl)amino)methyl)-
under reduced pressure, the residue was purified by flash benzimidamide (12b). Yield: 60%, MP: 162–165 °C. 1H NMR
chromatography, eluted with hexane/ethyl acetate (1/1, v/v) to (400 MHz, DMSO-d6) δ 9.57 (s, 1H), 7.61 (d, J = 7.1 Hz, 2H),
afford oil product. (3.0 g, 68%) 1H NMR (400 MHz, CDCl3) 7.30 (d, J = 7.4 Hz, 2H), 5.76 (s, 2H), 4.39 (s, 1H), 3.50 (s, 4H),
δ 7.64–7.56 (m, 2H), 7.44–7.36 (m, 2H), 3.88–3.79 (m, 1H), 2.42 (t, J = 6.2 Hz, 2H), 2.15 (s, 3H).
3.77–3.66 (m, 2H), 3.60–3.44 (m, 2H), 2.70–2.60 (m, 1H),
2.43–2.34 (m, 1H), 2.25 (s, 3H).
Tert-Butyl (E)-(4-(N′-hydroxycarbamimidoyl)benzyl)(2-hydroxy-
ethyl)carbamate (12c). Yield: 71%, MP: 138–141 °C. 1H NMR
4-(((2-Hydroxyethyl)(methyl)amino)methyl)benzonitrile (11b). (400 MHz, CDCl3) δ 7.66–7.53 (m, 2H), 7.35–7.26 (m, 2H), 5.30
To a round-bottom flask equipped with a stir bar was added (s, 2H), 4.86 (s, 2H), 4.52 (s, 2H), 3.71 (s, 2H), 3.47–3.31 (m,
4-(bromomethyl)benzonitrile (3.0 g, 15.3 mmol), 2-(methyl- 2H), 1.46 (s, 9H).
amino)ethan-1-ol (1.0 g, 13.3 mmol), and acetonitrile (20 mL).
After cooling to 0 °C, triethylamine (2.8 g, 27.7 mmol) was oxadiazol-3-yl)benzyl)(methyl)amino)propane-1,2-diol
( )-3-((4-(5-(4-(2-Fluoroethoxy)-3-(trifluoromethyl)phenyl)-1,2,4-
(13a).
added to the mixture dropwise. The mixture was warmed to RT Compound 13a was synthesized according to the general pro-
and stirred overnight. Then, the mixture was concentrated in cedure as described for compound 10a–c. Yield: 40%, white
vacuum, and the crude residue was purified by flash chromato- solid, MP: 101–103 °C. 1H NMR (400 MHz, MeOD-d4) δ 8.29 (d,
graphy, eluted with hexane/ethyl acetate (3/2, v/v) to afford the J = 7.1 Hz, 2H), 8.01 (d, J = 8.0 Hz, 2H), 7.47 (d, J = 8.1 Hz, 2H),
1
semi-solid product. (2.0 g, 79%) H NMR (400 MHz, CDCl3) δ 7.35 (d, J = 9.4 Hz, 1H), 4.83–4.69 (m, 3H), 4.48–4.37 (m, 3H),
7.63–7.58 (m, 2H), 7.42 (d, J = 7.9 Hz, 2H), 3.68–3.59 (m, 5H), 3.87–3.80 (m, 1H), 3.63 (s, 2H), 3.57–3.46 (m, 2H), 2.53–2.49
2.63–2.57 (m, 2H), 2.23 (s, 3H).
(m, 2H), 2.28 (s, 3H). 13C NMR (101 MHz, MeOD-d4) δ 174.3,
Tert-Butyl (4-cyanobenzyl)(2-hydroxyethyl)carbamate (11c). 168.5, 159.8, 142.1, 133.3, 129.5, 126.9, 126.6 (q, JC–F = 5.4 Hz),
To a round-bottom flask equipped with a stir bar was added 125.4, 122.9 (q, JC–F = 274 Hz), 119.3 (q, JC–F = 31 Hz), 116.4,
4-formylbenzonitrile (6.0 g, 45.8 mmol), ethanolamine (2.8 g, 113.8, 81.2 (d, JC–F = 171 Hz), 68.9, 68.6 (q, JC–F = 20 Hz), 64.9,
45.8 mmol), and methanol (80 mL). The mixture was stirred in 62.0, 60.1, 41.7. HRMS (ESI) calcd for C22H23F4N3O4 [M + H+]
a pre-heated 60 °C oil-bath for 6 h, then cooled to 0 °C fol- 470.1697, found 470.1694.
lowed by adding sodium borohydride (8.7 g, 229 mmol) por-
2-((4-(5-(4-(2-Fluoroethoxy)-3-(trifluoromethyl)phenyl)-1,2,4-
tionwise. The reaction was warmed to room temperature (RT) oxadiazol-3-yl)benzyl)(methyl)amino)ethan-1-ol (13b). Compound
slowly and stirred for an additional 5 h. Then, the reaction 13b was prepared according to the general procedure as
mixture was concentrated under reduced pressure, and the described for compound 10a–c. Yield: 34%, white solid, MP:
1
residue was dissolved in ethyl acetate (200 mL), washed with 108–111 °C. H NMR (400 MHz, CDCl3) δ 8.45 (s, 1H), 8.34 (d,
water (100 mL) and saturated brine (100 mL), and dried over J = 8.7 Hz, 1H), 8.11 (d, J = 8.1 Hz, 2H), 7.45 (d, J = 8.1 Hz, 2H),
anhydrous MgSO4. After filtering and concentration, the 7.17 (d, J = 8.8 Hz, 1H), 4.90–4.74 (m, 2H), 4.47–4.37 (m, 2H),
residual solid was re-dissolved in dichloromethane (100 mL). 3.68–3.62 (m, 4H), 2.64 (t, J = 5.3 Hz, 2H), 2.26 (s, 3H). 13C
To the above solution was added di-tert-butyl dicarbonate NMR (101 MHz, CDCl3) δ 174.2, 168.8, 159.9, 142.1, 133.3,
(6.7 g, 30.6 mmol) and 1 N NaOH (17.0 mL). The mixture was 129.4, 127.8 (q, JC–F = 5.25 Hz), 127.6, 125.6, 122.7 (q, JC–F
=
stirred at RT overnight, then the mixture was washed with 274 Hz), 120.2 (q, JC–F = 32 Hz), 117.1, 113.4, 81.2 (d, JC–F = 172
water and saturated brine, dried over anhydrous MgSO4. After Hz), 68.4 (d, JC–F = 21.2 Hz), 62.0, 58.5, 58.4, 41.6. HRMS (ESI)
filtering and concentration, the crude semi-solid product 11c calcd for C21H21F4N3O3 [M + H+] 440.1592, found 440.1588.
1
was obtained. (8.6 g, 68%) H NMR (400 MHz, CDCl3) δ 7.64
Tert-Butyl (4-(5-(4-(2-fluoroethoxy)-3-(trifluoromethyl)-phenyl)-
(d, J = 7.8 Hz, 2H), 7.38 (d, J = 8.0 Hz, 2H), 4.58 (s, 2H), 3.74 (s, 1,2,4-oxadiazol-3-yl)benzyl)(2-hydroxyethyl)-carbamate (13c).
2H), 3.52–3.29 (m, 3H), 1.41 (s, 9H).
Compound 13c was prepared according to the general pro-
cedure as described for compound 10a–c. Yield: 42%, white
solid, MP: 120–122 °C. 1H NMR (400 MHz, CDCl3) δ 8.47 (s,
General procedure for the synthesis of 12a–c
To a round-bottom flask equipped with a stir bar was added 1H), 8.35 (d, J = 8.8 Hz, 1H), 8.13 (d, J = 8.0 Hz, 2H), 7.38 (d, J =
11a–c (1.0 eq.), hydroxylamine hydrochloride (2.0 eq.), 7.7 Hz, 2H), 7.18 (d, J = 8.8 Hz, 1H), 4.92–4.75 (m, 2H), 4.56 (s,
NaHCO3 (4.0 eq.), and methanol (5.0 mL mmol−1). The reac- 2H), 4.50–4.36 (m, 2H), 3.75 (s, 2H), 3.46 (s, 2H), 2.90 (br, 1H),
tion was refluxed and stirred in a pre-heated 75 °C oil-bath for 1.47 (s, 9H).
6 h. The reaction mixture was cooled to room temperature,
2-((4-(5-(4-(2-Fluoroethoxy)-3-(trifluoromethyl)phenyl)-1,2,4-
and the precipitate was filtered off and washed with methanol. oxadiazol-3-yl)benzyl)amino)ethan-1-ol (13d). To a solution of
The filtrate was concentrated under reduced pressure to give 13c (250 mg, 0.48 mmol) in methanol (5.0 mL) was added 4 M
the product 12a–c and used directly without further HCl in dioxane (5.0 mL). The reaction was stirred at RT for 6 h,
purification.
then the solvent was removed under reduced pressure and the
(E)-4-(((2,3-Dihydroxypropyl)(methyl)amino)methyl)-N′-hydroxy- residue was dissolved in ethyl acetate. The solution was
benzimidamide (12a). Yield: 51%, MP: 160–164 °C. 1H NMR washed with saturated NaHCO3, water, and dried over MgSO4.
(400 MHz, MeOD-d4) δ 7.60 (d, J = 7.8 Hz, 2H), 7.37 (d, J = After filtering and concentration, the final product was
1
7.9 Hz, 2H), 3.85–3.78 (m, 1H), 3.66–3.58 (m, 3H), 3.57–3.43 obtained. (193 mg, 95%) MP: 126–130 °C. H NMR (400 MHz,
(m, 3H), 2.52–2.45 (m, 2H), 2.25 (s, 3H).
MeOD-d4) δ 8.35 (d, J = 4.9 Hz, 2H), 8.06 (d, J = 8.1 Hz, 2H),
9178 | Org. Biomol. Chem., 2018, 16, 9171–9184
This journal is © The Royal Society of Chemistry 2018

View Article Online
Organic & Biomolecular Chemistry
Paper
7.50 (d, J = 8.1 Hz, 2H), 7.39 (d, J = 9.3 Hz, 1H), 4.84–4.68 (m, with water. The crude was extracted with ethyl acetate, washed
2H), 4.51–4.39 (m, 2H), 3.85 (s, 2H), 3.71–3.64 (m, 2H), 2.73 (t, with saturated brine, and dried over MgSO4. After filtering and
J = 5.6 Hz, 2H). 13C NMR (101 MHz, MeOD-d4) δ 174.4, 168.4, concentrated, the residue was purified by flash chromato-
159.8, 142.8, 133.4, 128.7, 127.1, 126.7 (q, JC–F = 5.1 Hz), 125.4, graphy on silica gel column to afford the product.
124.3, 121.6, 116.4, 113.9, 81.2 (d, JC–F = 171 Hz), 68.7 (d, JC–F
=
2-((4-(5-(4-(2-Fluoroethoxy)-3-(trifluoromethyl)phenyl)-1,2,4-
20 Hz), 60.1, 52.5, 50.3. HRMS (ESI) calcd for C20H19F4N3O3 oxadiazol-3-yl)benzyl)oxy)ethan-1-ol (17a). Compound 17a was
[M + H+] 426.1435, found 426.1432.
purified by flash chromatography, eluted with hexane/ethyl
acetate (2/1, v/v) to afford off-white solid. Yield: 35%, MP:
109–110 °C. 1H NMR (400 MHz, CDCl3) δ 8.42 (s, 1H), 8.30 (dd,
General procedure for the synthesis of 15a–b
To a round-bottom flask equipped with a stir bar was added J = 8.6, 1.4 Hz, 1H), 8.11 (d, J = 8.1 Hz, 2H), 7.46 (d, J = 8.0 Hz,
aldehyde 14 32,33 (1.0 eq.), amine (1.5 eq.), methanol (14 mL 2H), 7.13 (d, J = 8.7 Hz, 1H), 4.89–4.71 (m, 2H), 4.62 (s, 2H),
mmol−1), and acetic acid (0.5 mL mmol−1). The reaction 4.46–4.32 (m, 2H), 3.82–3.73 (m, 2H), 3.67–3.59 (m, 2H), 2.19
mixture was stirred for 1 h at which time sodium cyanoborohy- (br, 1H). 13C NMR (101 MHz, CDCl3) δ 174.2, 168.7, 159.6,
dride (1.0 eq.) was added. The reaction mixture was stirred 141.4, 133.3, 127.9, 127.7 (q, JC–F = 5.2 Hz), 127.6, 126.1, 122.7
overnight and diluted with water. The precipitate was filtered (q, JC–F = 271 Hz), 120.2 (q, JC–F = 32 Hz), 117.0, 113.4, 81.3 (d,
off and washed with water to give off-white solid product 15a–b. JC–F = 172 Hz), 72.7, 71.7, 68.4 (d, JC–F = 21 Hz), 61.9. HRMS
2-((4-(5-(4-(2-Fluoroethoxy)-3-(trifluoromethyl)phenyl)-1,2,4- (ESI) calcd for C20H18F4N2O4 [M + H+] 427.1245, found
oxadiazol-3-yl)benzyl)amino)propane-1,3-diol (15a). Yield: 427.1240.
50%, off-white solid, MP: 135–137 °C. 1H NMR (400 MHz,
2-(2-((4-(5-(4-(2-Fluoroethoxy)-3-(trifluoromethyl)phenyl)-1,2,4-
DMSO-d6) δ 8.47–8.41 (m, 1H), 8.34 (s, 1H), 8.04 (d, J = 8.2 Hz, oxadiazol-3-yl)benzyl)oxy)ethoxy)ethan-1-ol (17b). Compound
2H), 7.57 (d, J = 8.0 Hz, 3H), 4.89–4.70 (m, 2H), 4.63–4.50 (m, 17b was purified by flash chromatography, eluted with hexane/
2H), 4.45 (s, 2H), 3.88 (s, 2H), 3.49–3.35 (m, 4H), 2.61–2.53 (m, ethyl acetate (1/1, v/v) to afford an off-white semi-solid. Yield:
1H). 13C NMR (101 MHz, DMSO-d6) δ 174.0, 168.2, 159.5, 39%. 1H NMR (400 MHz, CDCl3) δ 8.43 (s, 1H), 8.32 (d, J =
145.6, 134.1, 128.6, 127.0, 126.6 (q, JC–F = 5.1 Hz), 124.2, 122.4 8.5 Hz, 1H), 7.94 (d, J = 7.8 Hz, 1H), 7.82 (d, J = 10.3 Hz, 1H),
(q, JC–F = 274 Hz) 117.9, 116.0, 115.1, 109.6, 81.7 (d, JC–F = 162 7.71–7.51 (m, 2H), 7.16 (d, J = 8.7 Hz, 1H), 4.91–4.85 (m, 1H),
Hz), 68.8 (q, JC–F = 20 Hz), 61.1, 60.5, 50.5. HRMS (ESI) calcd 4.79–4.74 (m, 1H), 4.70 (s, 2H), 4.47–4.42 (m, 1H), 4.41–4.35
for C21H21F4N3O4 [M + H+] 456.1541, found 456.1537.
(m, 1H), 3.79–3.69 (m, 6H), 3.67–3.60 (m, 2H), 2.41 (br, 1H).
(1-(4-(5-(4-(2-Fluoroethoxy)-3-(trifluoromethyl)phenyl)-1,2,4- 13C NMR (101 MHz, CDCl3) δ 174.4, 168.0, 159.5, 141.8, 133.3,
oxadiazol-3-yl)benzyl)azetidin-3-yl)methanol (15b). Yield: 23%, 128.0, 127.7 (q, JC–F = 5.4 Hz), 127.2, 126.5, 122.8 (q, JC–F = 274
white solid, MP: 114–116 °C. 1H NMR (400 MHz, DMSO-d6) Hz), 119.0 (q, JC–F = 32 Hz), 116.8, 113.4, 81.2 (d, JC–F = 173
δ 8.45 (d, J = 8.8 Hz, 1H), 8.36–8.32 (m, 1H), 8.12 (d, J = 8.2 Hz, Hz), 72.5, 70.3, 70.0, 68.4 (d, JC–F = 21 Hz), 66.4, 61.8. HRMS
2H), 7.65–7.55 (m, 3H), 5.49 (s, 1H), 4.88–4.82 (m, 1H), (ESI) calcd for C22H22F4N2O5 [M + H+] 471.1583, found
4.78–4.71 (m, 1H), 4.64–4.57 (m, 1H), 4.57–4.49 (m, 1H), 471.1589.
3.41–3.27 (m, 8H), 2.34–2.21 (m, 1H).
2-(2-(2-((4-(5-(4-(2-Fluoroethoxy)-3-(trifluoromethyl)phenyl)-
3-(4-(Chloromethyl)phenyl)-5-(4-(2-fluoroethoxy)-3-(trifluoro- 1,2,4-oxadiazol-3-yl)benzyl)oxy)ethoxy)ethoxy)ethan-1-ol (17c).
methyl)phenyl)-1,2,4-oxadiazole (16). To a round-bottom flask Compound 17c was purified by flash chromatography, eluted
was added cyanuric chloride (253 mg, 1.37 mmol) and DMF with hexane/ethyl acetate (1/3, v/v) to afford an off-white semi-
1
(2.5 mL). A solution of 10c (500 mg, 1.31 mmol) in dichloro- solid. Yield: 42%. H NMR (400 MHz, CDCl3) δ 8.45 (br, 1H),
methane (3.5 mL) was added dropwise at RT, and the reaction 8.33 (d, J = 8.7 Hz, 1H), 8.13 (d, J = 8.1 Hz, 2H), 7.49 (d, J =
was continued to stir at RT for overnight. Then, the reaction 8.1 Hz, 2H), 7.16 (d, J = 8.8 Hz, 1H), 4.90–4.75 (m, 2H), 4.64 (s,
mixture was diluted with dichloromethane and water. The di- 2H), 4.49–4.36 (m, 2H), 3.74–3.61 (m, 12H), 2.63 (br, 1H). 13C
chloromethane layer was then separated and concentrated in NMR (101 MHz, CDCl3) δ 174.2, 168.8, 159.6, 141.7, 132.2,
vacuum to afford crude product (400 mg, 73%). The crude 127.9, 127.7 (q, JC–F = 5.2 Hz), 127.6, 125.9, 122.7 (q, JC–F
product was used directly for the next step. 1H NMR (400 MHz, 271 Hz), 120.2 (q, JC–F = 32 Hz), 117.1, 113.4, 81.3 (d, JC–F
=
=
CDCl3) δ = 8.47 (d, J = 2.0 Hz, 1H), 8.35 (dd, J = 8.8 Hz, 2.0 Hz, 171 Hz), 72.7, 72.5, 70.7, 70.6, 70.3, 69.7, 68.4 (d, JC–F = 21 Hz),
1H), 8.17 (d, J = 8.0 Hz, 2H), 7.54 (d, J = 8.4 Hz, 2H), 7.18 (d, J = 61.7. HRMS (ESI) calcd for C24H26F4N2O6 [M + H+] 515.1800,
8.8 Hz, 1H), 4.93–4.76 (m, 2H), 4.65 (s, 2H), 4.48–4.36 (m, 2H).
found 515.1806.
1-(4-(5-(4-(2-Fluoroethoxy)-3-(trifluoromethyl)phenyl)-1,2,4-
oxadiazol-3-yl)phenyl)-2,5,8,11-tetraoxatridecan-13-ol (17d).
General procedure for the synthesis of 17a–e and 17g
To a round-bottom flask equipped with a stir bar was added Compound 17d was purified by flash chromatography, eluted
alcohols (10 eq.) and THF (2 mL mmol−1). After cooling to with hexane/ethyl acetate (1/3, v/v) to afford an off-white semi-
0 °C, sodium hydride (1.8 eq.) was added portionwise, and the solid. Yield: 25%. 1H NMR (400 MHz, CDCl3) δ 8.33 (d, J =
reaction vessel was equipped with a reflux condenser and 5.7 Hz, 1H), 8.25–8.18 (m, 1H), 8.07–8.00 (m, 2H), 7.40 (d, J =
heated to 80 °C and stirred for 30 min. Chloride 16 (1.0 eq.) 7.9 Hz, 2H), 7.06 (d, J = 8.7 Hz, 1H), 4.80–4.64 (m, 3H),
was added to the flask and the mixture was stirred overnight. 4.59–4.52 (m, 2H), 4.36–4.27 (m, 2H), 3.69–3.54 (m, 14H),
The reaction was cooled to room temperature and quenched 3.54–3.48 (m, 2H). 13C NMR (101 MHz, CDCl3) δ 173.9, 168.3,
This journal is © The Royal Society of Chemistry 2018
Org. Biomol. Chem., 2018, 16, 9171–9184 | 9179

View Article Online
Paper
Organic & Biomolecular Chemistry
159.3, 141.8, 132.2, 127.9, 127.7 (q, JC–F = 5.2 Hz), 127.6, 125.9, crude residue was purified by flash chromatography on silica
122.7 (q, JC–F = 271 Hz), 120.2 (q, JC–F = 32 Hz), 117.1, 113.4, gel, eluted with hexane/ethyl acetate (1/1, v/v) to afford an off-
81.3 (d, J = 172.5 Hz), 72.7, 72.5, 70.6, 70.5, 70.4, 70.3, 70.2, white solid product. (3.8 g, 90%) MP: 80–83 °C. 1H NMR
69.7, 68.4 (d, JC–F = 21 Hz), 61.7. HRMS (ESI) calcd for (400 MHz, CDCl3) δ 7.63 (d, J = 8.1 Hz, 2H), 7.43 (d, J = 7.9 Hz,
C26H30F4N2O7 [M + H+] 559.2062, found 559.2066.
2H), 4.61 (s, 2H), 3.97–3.89 (m, 1H), 3.76–3.69 (m, 1H),
Tert-Butyl
(2-((4-(5-(4-(2-fluoroethoxy)-3-(trifluoromethyl)- 3.66–3.56 (m, 3H), 2.75 (s, 1H), 2.29 (s, 1H).
phenyl)-1,2,4-oxadiazol-3-yl)benzyl)oxy)ethyl)(2-hydroxy-ethyl)
(E)-N′-Hydroxy-4-(2-hydroxyethoxy)benzimidamide (19a).
carbamate (17e). Compound 17e was synthesized according to Compound 19a was synthesized according to the general pro-
the general procedure. The crude product was used directly for cedure as described for compound 12a–c. The crude product
the next step reaction without any further purification.
was used directly for the next step reaction without further
1
2-((2-((4-(5-(4-(2-Fluoroethoxy)-3-(trifluoromethyl)phenyl)-1,2,4- purification. H NMR (400 MHz, DMSO-d6) δ 9.41 (s, 1H), 7.56
oxadiazol-3-yl)benzyl)oxy)ethyl)amino)ethan-1-ol (17f). To a (d, J = 8.7 Hz, 2H), 6.89 (d, J = 8.7 Hz, 2H), 5.68 (s, 2H), 4.83 (s,
solution of 17e (284 mg, 0.50 mmol) in methanol (5.0 mL) was 1H), 3.97 (t, J = 4.7 Hz, 2H), 3.68 (d, J = 4.7 Hz, 2H).
added 4 M HCl in dioxane (5.0 mL). The reaction was stirred at
(E)-4-((2,3-Dihydroxypropoxy)methyl)-N′-hydroxybenzimid-
RT for 6 h, then the solvent was concentrated in vacuum and amide (19b). Compound 19b was synthesized according to the
the residue was dissolved in ethyl acetate. The solution was general procedure as described for compound 12a–c. Yield:
washed with saturated NaHCO3, water, and dried over MgSO4. 45%, MP: 142–144 °C. 1H NMR (400 MHz, MeOD-d4) δ 7.61 (d,
After filtering and concentration, the final white solid product J = 8.0 Hz, 2H), 7.39 (d, J = 7.9 Hz, 2H), 4.57 (s, 2H), 3.83–3.75
was obtained. (199 mg, 85%) MP: 126–130 °C. 1H NMR (m, 1H), 3.64–3.45 (m, 4H).
(400 MHz, CDCl3) δ 8.45 (s, 1H), 8.34 (d, J = 8.7 Hz, 1H), 8.14
2-(4-(5-(4-(2-Fluoroethoxy)-3-(trifluoromethyl)phenyl)-1,2,4-
(d, J = 8.1 Hz, 2H), 7.46 (d, J = 8.1 Hz, 2H), 7.17 (d, J = 8.8 Hz, oxadiazol-3-yl)phenoxy)ethan-1-ol (20a). Compound 20a was
1H), 4.91–4.75 (m, 2H), 4.60 (s, 2H), 4.47–4.38 (m, 2H), synthesized according to the general procedure as described
4.35–4.29 (m, 2H), 3.74–3.67 (m, 4H), 3.52 (t, J = 5.0 Hz, 2H). for compound 10a–c. Yield: 37%,
13C NMR (101 MHz, CDCl3) δ 174.3, 168.7, 158.6, 141.3, 133.3, 148–149 °C. 1H NMR (400 MHz, acetone-d6) δ 8.45–8.39 (m,
127.8, 127.7 (q, JC–F = 5.25 Hz), 127.6, 126.1, 122.7 (q, JC–F 2H), 8.08 (d, J = 8.9 Hz, 2H), 7.56 (d, J = 8.7 Hz, 1H), 7.13 (d, J =
a white solid. MP:
=
275 Hz), 120.3 (q, JC–F = 31 Hz), 117.0, 113.4, 86.3 (d, JC–F = 173 8.9 Hz, 2H), 4.95–4.80 (m, 2H), 4.68–4.56 (m, 2H), 4.18 (t, J =
Hz), 72.6, 68.9, 68.4 (d, JC–F = 21.1 Hz), 62.0, 46.0, 44.2. HRMS 4.9 Hz, 2H), 4.05 (br, 1H), 3.92 (d, J = 4.3 Hz, 2H). 13C NMR
(ESI) calcd for C22H23F4N3O4 [M + H+] 470.1697, found (101 MHz, acetone-d6) δ 174.1, 168.4, 161.8, 159.9, 133.7,
470.1694.
128.9, 126.8 (q, JC–F = 5.1 Hz), 123.2 (q, JC–F = 273 Hz), 119.3,
2-((2-((4-(5-(4-(2-Fluoroethoxy)-3-(trifluoromethyl)phenyl)-1,2,4- 119.0, 116.8, 114.9, 114.4, 81.6 (d, JC–F = 170 Hz), 69.9, 68.9 (d,
oxadiazol-3-yl)benzyl)oxy)ethyl)(methyl)amino)ethan-1-ol (17g). JC–F = 20 Hz), 60.3. HRMS (ESI) calcd for C19H16F4N2O4 [M +
Compound 17g was purified by flash chromatography, eluted H+] 413.1119, found 413.1113.
with ethyl acetate/methanol (50/1, v/v) to afford a pale yellow
( )-3-((4-(5-(4-(2-Fluoroethoxy)-3-(trifluoromethyl)phenyl)-
semi-solid. Yield: 26%. 1H NMR (400 MHz, CDCl3) δ 8.40 (d, 1,2,4-oxadiazol-3-yl)benzyl)oxy)propane-1,2-diol (20b). Compound
J = 1.8 Hz, 1H), 8.29 (dd, J = 8.7, 2.0 Hz, 1H), 8.07 (d, J = 8.1 20b was synthesized according to the general procedure as
Hz, 2H), 7.42 (d, J = 8.1 Hz, 2H), 7.11 (d, J = 8.8 Hz, 1H), described for compound 10a–c. Yield: 38%, white solid, MP:
4.88–4.79 (m, 1H), 4.75–4.65 (m, 1H), 4.54 (s, 2H), 4.44–4.25 102–105 °C. 1H NMR (400 MHz, MeOD-d4) δ 8.34–8.28 (m,
(m, 2H), 4.12 (s, 3H), 3.55 (s, 2H), 2.73–2.60 (m, 4H), 2.29 (s, 2H), 8.04 (d, J = 8.1 Hz, 2H), 7.50 (d, J = 8.0 Hz, 2H), 7.35 (d, J =
3H). 13C NMR (101 MHz, CDCl3) δ 174.8, 168.7, 160.2, 159.0, 8.5 Hz, 1H), 4.85–4.69 (m, 2H), 4.61 (s, 2H), 4.48–4.38 (m, 2H),
141.3, 133.3, 128.0, 127.8 (q, JC–F = 5.25 Hz), 127.4, 125.9, 122.8 3.88–3.81 (m, 1H), 3.67–3.51 (m, 4H). 13C NMR (101 MHz,
(q, JC–F = 275 Hz), 120.4 (q, JC–F = 31 Hz), 116.8, 113.4, 72.8, MeOD-d4) δ 174.3, 168.4, 159.8, 142.1, 133.3, 127.6, 126.9,
60.4, 59.9, 57.7, 43.0. HRMS (ESI) calcd for C17H21F4N3O4 [M + 126.6 (q, JC–F = 5.35 Hz), 125.6, 122.9 (q, JC–F = 273 Hz), 119.3
H+] 408.1541, found 408.1536.
(q, JC–F = 31 Hz), 116.3, 113.8, 81.2 (d, JC–F = 170 Hz), 72.3,
4-(2-Hydroxyethoxy)benzonitrile (18a). Compound 18a was 71.6, 70.9, 68.6 (q, JC–F = 20.1 Hz), 63.1. HRMS (ESI) calcd for
synthesized according to the published procedure.45
C21H20F4N2O5 [M + H+] 457.1381, found 457.1385.
4-((2,3-Dihydroxypropoxy)methyl)benzonitrile
(18b).
(E)-N′-Hydroxy-4-vinylbenzimidamide (22). Compound 22
Commercially available DL-1,2-isopropylideneglycerol (3.0 g, was synthesized according to the general procedure as
22.5 mmol) was added to a suspension of NaH (0.6 g, described for compound 12a–c. Yield: 95%. The detailed pro-
25.8 mmol) in THF (30 mL) at 0 °C under nitrogen. The cedure was described in the published literature.46
mixture was stirred for 30 min during which time 4-cyanoben-
5-(4-(2-Fluoroethoxy)-3-(trifluoromethyl)phenyl)-3-(4-vinyl-
zyl bromide (4.0 g, 20.4 mmol) was added dropwise over 5 min phenyl)-1,2,4-oxadiazole (23). Compound 23 was synthesized
followed by warming of the reaction mixture to RT. After 2 h, according to the general procedure as described for compound
the reaction mixture was partitioned between ammonium 10a–c. The crude product was used directly for the next step
chloride solution and ethyl acetate. The aqueous phase was reaction without further purification.
extracted with ethyl acetate and the combined organics dried
( )-1-(4-(5-(4-(2-Fluoroethoxy)-3-(trifluoromethyl)phenyl)-1,2,4-
over anhydrous MgSO4 and concentrated in vacuum. The oxadiazol-3-yl)phenyl)ethane-1,2-diol (24). To a 50 mL of
9180 | Org. Biomol. Chem., 2018, 16, 9171–9184
This journal is © The Royal Society of Chemistry 2018

View Article Online
Organic & Biomolecular Chemistry
Paper
round bottle flask was added 18 (327 mg, 0.86 mmol), and dried over anhydrous MgSO4. After filtration and concen-
4-methylmorpholine N-oxide (122 mg, 1.03 mmol) and THF/ tration under reduced pressure, the crude residue was purified
H2O (3/1, 12.0 mL). The mixture was stirred for 5 min before on a silica gel column to give 26a–d.
adding OsO4 (2.5 wt% in tert-butanol, 545 µL). The reaction
4-(3-(4-((Methoxymethoxy)methyl)phenyl)-1,2,4-oxadiazol-5-
was stirred continuously at RT overnight. Then, the reaction yl)-2-(trifluoromethyl)phenol (26a). Compound 26a was puri-
mixture was diluted with ethyl acetate and water. The ethyl fied by flash chromatography, eluted with hexane/ethyl acetate
acetate layer was separated and concentrated in vacuum to (2/1, v/v) to afford a white solid. Yield: 50%, white solid, MP:
1
afford crude product. Crude product was purified by flash 130–135 °C. H NMR (400 MHz, DMSO-d6) δ = 11.83 (br, 1H),
chromatography, eluted with hexane/ethyl acetate (1/2, v/v) to 8.29–8.24 (m, 2H), 8.07 (d, J = 8.0 Hz, 2H), 7.55 (d, J = 8.0 Hz,
furnish the final product as a white semi-solid. (120 mg, 34%) 2H), 7.27 (d, J = 8.8 Hz, 1H), 4.70 (s, 2H), 4.63 (s, 2H), 3.33 (s,
1H NMR (400 MHz, MeOD-d4) δ 8.39–8.31 (m, 2H), 8.08 (d, J = 3H).
8.2 Hz, 2H), 7.55 (d, J = 8.2 Hz, 2H), 7.38 (d, J = 8.8 Hz, 1H),
4-(3-(4-((2-(Methoxymethoxy)ethoxy)methyl)phenyl)-1,2,4-
4.85–4.70 (m, 3H), 4.50–4.39 (m, 2H), 3.72–3.60 (m, 2H). 13C oxadiazol-5-yl)-2-(trifluoromethyl)phenol (26b). Compound
NMR (101 MHz, MeOD-d4) δ 174.4, 168.5, 159.9, 145.8, 133.3, 26b was purified by flash chromatography, eluted with hexane/
126.9, 126.7 (q, JC–F = 5.2 Hz), 126.6, 125.6, 122.9 (q, JC–F = 273 ethyl acetate (3/2, v/v) to afford a white solid. Yield: 45%, white
Hz), 119.4 (q, JC–F = 32 Hz), 116.4, 113.8, 81.2 (d, JC–F = 171 solid, MP: 120–122 °C. 1H NMR (400 MHz, CDCl3) δ 8.34 (s,
Hz), 74.1, 68.6(d, JC–F = 20.2 Hz), 67.1. HRMS (ESI) calcd for 1H), 8.04–7.93 (m, 3H), 7.53–7.45 (m, 1H), 7.41 (d, J = 8.1 Hz,
C19H16F4N2O4 [M + H+] 413.1119, found 413.1115.
(E)-N′-Hydroxy-4-((methoxymethoxy)methyl)benzimidamide 3.86–3.76 (m, 4H), 3.41 (s, 3H).
2H), 6.99 (d, J = 8.6 Hz, 1H), 4.73 (s, 2H), 4.62 (s, 2H),
(25a). Compound 25a was synthesized according to the
4-(3-(4-(2,4,7,10-Tetraoxaundecan-11-yl)phenyl)-1,2,4-oxadia-
general procedure as described for compound 12a–c. Yield: zol-5-yl)-2-(trifluoromethyl)phenol (26c). Compound 26c was
1
79%, a pale white solid. MP: 158–159 °C. H NMR (400 MHz, purified by flash chromatography, eluted with hexane/ethyl
DMSO-d6) δ 9.64 (s, 1H), 7.70–7.62 (m, 2H), 7.37–7.26 (m, 2H), acetate (3/2, v/v) to afford a white solid. Yield: 60%, MP:
1
5.80 (s, 3H), 4.69–4.61 (m, 2H), 4.53 (s, 2H), 3.30 (s, 2H).
103–105 °C. H NMR (400 MHz, CDCl3) δ 8.26 (d, J = 6.1 Hz,
(E)-N′-Hydroxy-4-((2-(methoxymethoxy)ethoxy)methyl)ben- 1H), 7.90 (d, J = 7.9 Hz, 2H), 7.80 (d, J = 8.6 Hz, 1H), 7.30 (d, J =
zimidamide (25b). Compound 25b was synthesized according 7.9 Hz, 2H), 6.83 (d, J = 8.6 Hz, 1H), 4.61 (s, 2H), 4.54 (s, 2H),
to the general procedure as described for compound 12a–c. 3.85–3.70 (m, 8H), 3.34 (s, 3H), 2.86 (s, 2H).
Yield: 60%, a white solid. MP: 150–151 °C. 1H NMR (400 MHz,
4-(3-(4-(2,4,7,9-Tetraoxadecan-5-yl)phenyl)-1,2,4-oxadiazol-5-
MeOD-d4) δ 7.62 (d, J = 8.1 Hz, 2H), 7.38 (d, J = 8.1 Hz, 2H), yl)-2-(trifluoromethyl)phenol (26d). Compound 26d was puri-
4.63 (s, 2H), 4.57 (s, 2H), 3.74–3.63 (m, 4H), 3.34 (s, 3H). fied by flash chromatography, eluted with hexane/ethyl acetate
(E)-N′-Hydroxy-4-(2,4,7,10-tetraoxaundecan-11-yl)benzimid- (1/1, v/v) to afford a white semi-solid. Yield: 50%. 1H NMR
amide (25c). Compound 25c was synthesized according to the (400 MHz, CDCl3) δ 8.38 (s, 1H), 8.14 (d, J = 8.4 Hz, 1H), 8.08
general procedure as described for compound 12a–c. Yield: (d, J = 8.1 Hz, 2H), 7.48 (d, J = 8.1 Hz, 2H), 7.02 (d, J = 8.6 Hz,
50%, a white solid. MP: 140–142 °C. 1H NMR (400 MHz, 1H), 4.89–4.84 (m, 1H), 4.73–4.69 (m, 1H), 4.68–4.61 (m, 3H),
MeOD-d4) δ 7.65 (d, J = 8.1 Hz, 2H), 7.41 (d, J = 8.1 Hz, 2H), 3.84–3.69 (m, 3H), 3.38 (s, 3H), 3.29 (s, 3H).
4.66 (s, 2H), 4.60 (s, 2H), 3.73–3.65 (m, 8H), 3.37 (s, 3H).
General procedure for the synthesis of 27a–d
(E)-4-(2,4,7,9-Tetraoxadecan-5-yl)-N′-hydroxybenzimidamide
(25d). Compound 25d was synthesized according to the To a round-bottom flask equipped with a stir bar was added
general procedure as described for compound 12a–c. Yield: 26a–d (1.0 eq.), ethylene ditosylate (2.0 eq.), K2CO3 (3.0 eq.),
62%, a yellow solid. MP: 170–172 °C. 1H NMR (400 MHz, and acetonitrile (10 mL mmol−1). The reaction mixture was
CDCl3) δ 8.40 (s, 1H), 8.19 (d, J = 8.8 Hz, 1H), 8.11 (d, J = 8.1 stirred at RT for 1 h, then refluxed in a pre-heated 90 °C oil-
Hz, 2H), 7.50 (d, J = 8.1 Hz, 2H), 7.04 (d, J = 8.6 Hz, 1H), bath overnight and monitored by TLC. The reaction was
4.91–4.87 (m, 1H), 4.75–4.71 (m, 1H), 4.69–4.64 (m, 3H), diluted with ethyl acetate and water, the ethyl acetate layer was
3.84–3.72 (m, 2H), 3.40 (s, 3H), 3.31 (s, 3H).
washed with saturated brine and dried over anhydrous MgSO4.
After filtering and concentrated in vacuum, the crude residue
was purified on a silica gel column to give 27a–d.
General procedure for the synthesis of 26a–d
To a round-bottom flask equipped with a stir bar was added
2-(4-(3-(4-((Methoxymethoxy)methyl)phenyl)-1,2,4-oxa-diazol-
4-hydroxy-3-(trifluoromethyl)benzoic acid (1.0 eq.), HOBt (0.2 5-yl)-2-(trifluoromethyl)phenoxy)ethyl 4-methyl-benzenesulfo-
eq.), TBTU (1.0 eq.), DIPEA (3.0 eq.), and DMF (5.0 mL nate (27a). Compound 27a was purified by flash chromato-
mmol−1). The reaction mixture was stirred for 0.5 h followed graphy, eluted with hexane/ethyl acetate (3/1, v/v) to afford a
1
by adding methoxymethyl (MOM) protected amidoxime 25a–d white solid. Yield: 30%, MP: 125–127 °C. H NMR (400 MHz,
(1.0 eq.). The reaction mixture was stirred for 1 h at room CDCl3) δ 8.44 (s, 1H), 8.32 (d, J = 8.8 Hz, 1H), 8.15 (d, J =
temperature, then refluxed in a pre-heated 120 °C oil-bath for 8.0 Hz, 2H), 7.81 (d, J = 8.0 Hz, 2H), 7.51 (d, J = 8.0 Hz, 2H),
4 h and monitored by TLC. After cooling, the reaction mixture 7.35 (d, J = 8.0 Hz, 2H), 7.09 (d, J = 8.8 Hz, 1H), 4.75 (s, 2H),
was diluted with water and extracted with ethyl acetate. The 4.68 (s, 2H), 4.45–4.35 (m, 4H), 3.44 (s, 3H), 2.45 (s, 3H). 13C
ethyl acetate layer was washed with 1 M HCl, saturated brine, NMR (101 MHz, CDCl3) δ 174.3, 169.0, 159.2, 145.3, 141.6,
This journal is © The Royal Society of Chemistry 2018
Org. Biomol. Chem., 2018, 16, 9171–9184 | 9181

View Article Online
Paper
Organic & Biomolecular Chemistry
133.5, 132.6, 130.1, 128.2, 128.1, 127.8 (q, JC–F = 5.5 Hz), 127.8, in acetonitrile (300 µL). The reaction was placed in a 110 °C
126.1, 122.8 (q, JC–F = 271 Hz), 120.3 (q, JC–F = 32 Hz), 117.5, oil-bath for 15 min. The reaction was removed from the oil-
113.5, 96.1, 68.9, 67.3, 66.8, 55.6, 21.8. HRMS (ESI) calcd for bath, at which time 6 M HCl (150 µL) was added. The reaction
C27H25F3N2O7S [M + H+] 579.1407. Found [M + H+] 579.1415.
mixture was heated in a 110 °C oil-bath for another 5 min.
2-(4-(3-(4-((2-(Methoxymethoxy)ethoxy)methyl)phenyl)-1,2,4- Then, the reaction was removed from the oil-bath and
oxadiazol-5-yl)-2-(trifluoromethyl)phenoxy)ethyl 4-methyl- quenched with 6 M NaOH (150 µL) and diluted with 2.4 mL of
benzenesulfonate (27b). Compound 27b was purified by flash the HPLC mobile phase. The mixture was passed through a
chromatography, eluted with hexane/ethyl acetate (3/1, v/v) to Sep-Pak® Alumina N Cartridge (Part No. WAT020510) and
afford a white solid. Yield: 57%, MP: 115–117 °C. 1H NMR injected onto the semi-preparation HPLC column (Agilent
(400 MHz, CDCl3) δ 8.37 (d, J = 1.4 Hz, 1H), 8.28–8.23 (m, 1H), SB-C18 250 × 9.6 mm, 5 µm, UV = 254 nm, 4.0 mL min−1). The
8.10 (d, J = 8.2 Hz, 2H), 7.78 (d, J = 8.2 Hz, 2H), 7.48 (d, J = 8.1 HPLC fraction was collected into a water bottle with 60 mL of
Hz, 2H), 7.32 (d, J = 8.1 Hz, 2H), 7.05 (d, J = 8.8 Hz, 1H), 4.67 water and then trapped on a Sep-Pak® C-18 Cartridge (Part No.
(s, 2H), 4.64 (s, 2H), 4.40–4.33 (m, 4H), 3.78–3.67 (m, 4H), 3.37 WAT020515). The activity was washed out with 0.6 mL of
(s, 3H), 2.41 (s, 3H). 13C NMR (101 MHz, CDCl3): δ 174.1, ethanol and 5.4 mL of 0.9% saline. After sterile filtration into
168.7, 159.0, 145.2, 141.8, 133.3, 132.3, 129.9, 127.9, 127.8, a glass vial, the radiolabeled product was ready for quality
127.7, 127.5 (q, JC–F = 5.5 Hz), 125.8, 122.6 (q, JC–F = 271 Hz), control (QC) analysis and animal studies. An aliquot of sample
120.0 (q, JC–F = 32 Hz), 117.1, 113.3, 96.5, 72.7, 69.7, 67.3, 66.8, was injected onto an analytical HPLC column (Agilent Eclipse
66.5, 55.2, 21.6. HRMS (ESI) calcd for C29H29F3N2O8S [M + H+] XDB-C18 250 × 4.6 mm, 5 µm) to determine the concentration
623.1669, found 623.1660.
2-(4-(3-(4-(2,4,7,10-Tetraoxaundecan-11-yl)phenyl)-1,2,4-oxa- ing with the non-radiolabeled standard sample solution.
diazol-5-yl)-2-(trifluoromethyl)phenoxy)ethyl 4-methyl-benzene-
Semi-preparation HPLC conditions. [18F]10a, [18F]17a, and
of tracer. Meanwhile, the tracer was authenticated by co-inject-
sulfonate (27c). Compound 27c was purified by flash chrom- [18F]17b: Mobile phase, 50% acetonitrile in 0.1 M ammonium
atography, eluted with hexane/ethyl acetate (3/1, v/v) to afford formate buffer, pH = 4.2; flow rate, 4.0 mL min−1; UV = 254
a white semi-solid. Yield: 38%. 1H NMR (400 MHz, CDCl3) nM; tR = 18–21 min. The radiochemical yields were 38%, 40%,
δ 8.41 (s, 1H), 8.31 (d, J = 8.7 Hz, 1H), 8.13 (d, J = 8.1 Hz, 2H), and 35%, respectively (decay corrected to EOS).
7.80 (d, J = 8.2 Hz, 2H), 7.49 (d, J = 8.1 Hz, 2H), 7.34 (d, J =
8.1 Hz, 2H), 7.08 (d, J = 8.8 Hz, 1H), 4.67 (s, 2H), 4.65 (s, 2H), ammonium formate buffer, pH = 4.2; flow rate, 4.0 mL min−1
( )-[18F]24: Mobile phase, 40% acetonitrile in 0.1
M
;
4.44–4.35 (m, 4H), 3.77–3.67 (m, 8H), 3.37 (s, 3H), 2.44 (s, 3H). UV = 254 nM; tR = 25–27 min. The radiochemical yield was
13C NMR (101 MHz, CDCl3): δ 174.1, 168.8, 159.0, 145.2, 141.9, 38% (decay corrected to EOS).
133.3, 132.3, 129.9, 127.9, 127.9, 127.6 (q, JC–F = 5.5 Hz), 127.5,
QC HPLC conditions. [18F]10a: Mobile phase, 75% aceto-
125.8, 122.6 (q, JC–F = 271 Hz), 120.1 (q, JC–F = 32 Hz), 117.2, nitrile in 0.1 M ammonium formate buffer, pH = 4.2; flow rate,
113.3, 96.5, 72.7, 70.6, 69.8, 67.2, 66.8, 66.5, 55.2, 21.6. HRMS 1.5 mL min−1; UV = 254 nM; tR = 5.5 min.
(ESI) calcd for C31H33F3N2O9S [M + H+] 667.1932, found
667.1925.
[18F]17a: Mobile phase, 70% acetonitrile in 0.1
M
ammonium formate buffer, pH = 4.2; flow rate, 1.0 mL min−1
;
( )-2-(4-(3-(4-(2,4,7,9-Tetraoxadecan-5-yl)phenyl)-1,2,4-oxa- UV = 254 nM; tR = 5.7 min.
diazol-5-yl)-2-(trifluoromethyl)phenoxy)ethyl 4-methyl-ben-
[18F]17b: Mobile phase, 70% acetonitrile in 0.1
M
;
zenesulfonate (27d). Compound 27d was purified by flash ammonium formate buffer, pH = 4.2; flow rate, 1.0 mL min−1
chromatography, eluted with hexane/ethyl acetate (2/1, v/v) to UV = 254 nM; tR = 5.6 min.
afford a white semi-solid. Yield: 44%. 1H NMR (400 MHz,
( )-[18F]24: Mobile phase, 63% acetonitrile in 0.1
M
CDCl3) δ 8.42 (s, 1H), 8.32 (d, J = 8.7 Hz, 1H), 8.15 (d, J = 8.1 ammonium formate buffer, pH = 4.2; flow rate, 1.5 mL min−1
Hz, 2H), 7.81 (d, J = 7.9 Hz, 2H), 7.52 (d, J = 8.2 Hz, 2H), 7.34 UV = 254 nM; tR = 4.6 min.
(d, J = 8.0 Hz, 2H), 7.09 (d, J = 8.7 Hz, 1H), 4.91–4.86 (m, 1H),
;
In vitro S1PR1–5 binding assay
4.75–4.59 (m, 5H), 4.46–4.36 (m, 4H), 3.84–3.70 (m, 2H), 3.38
(s, 3H), 3.29 (s, 3H), 2.44 (s, 3H). 13C NMR (101 MHz, DMSO- The binding potencies of the newly synthesized compounds
d6) δ 174.6, 168.5, 159.5, 145.5, 143.7, 134.5, 132.4, 130.5, were determined by competition against the binding of [32P]
128.3, 128.0, 127.5, 127.1, 127.0 (q, JC–F = 5.5 Hz), 123.1 (q, JC–F S1P to commercial cell membranes expressing recombinant
= 271 Hz) 125.8, 118.8, 116.4, 115.4, 96.2, 94.9, 76.9, 71.1, 68.8, human S1PRs (1, 2, 3, 4, and 5) according to methods reported
67.3, 55.4, 55.1, 21.5. HRMS (ESI) calcd for C30H31F3N2O9S [M previously.32,33,35
+ H+] 653.1775, found 653.1770.
MicroPET studies in cynomolgus macaque
General procedure for the radiosynthesis of [18F]10a, [18F]17a,
Male macaques (9–10 kg) were studied with a microPET Focus
220 scanner (Concorde/CTI/Siemens Microsystems, Knoxville,
[
18F]17b, and ( )-[18F]24
[18F]KF (∼7.4 GBq) aqueous was added to a vial containing TN). Animals were maintained in facilities with 12-hour dark
Kryptofix 222 (∼10 mg), and dried by azeotropic evaporation and light cycles, given access to food, water, and libitum.
with acetonitrile (3 × 1 mL) under N2 flow at 110 °C. To the Animals were also provided a variety of psychologically enrich-
reaction vial was added precursor 27a–d (∼2 mg) as a solution ing tasks to prevent inappropriate deprivation. Animals were
9182 | Org. Biomol. Chem., 2018, 16, 9171–9184
This journal is © The Royal Society of Chemistry 2018

View Article Online
Organic & Biomolecular Chemistry
Paper
scanned under anesthesia (induced with ketamine and glyco-
pyrrolate and maintained with inhalation isoflurane). Core
temperature was kept about 37 °C with a heated water blanket.
8 C. Wang, J. Mao, S. Redfield, Y. Mo, J. M. Lage and
X. Zhou, Exp. Mol. Pathol., 2014, 97, 259–265.
9 M. Maceyka and S. Spiegel, Nature, 2014, 510, 58–67.
The head was secured in a head holder with the brain in the 10 F. D. Lublin and S. C. Reingold, Neurology, 1996, 46, 907–
center of the field of view. Subsequently, a 2 h dynamic emis- 911.
sion scan was acquired after administration of ∼0.35 GBq of 11 G. K. Sheridan and K. K. Dev, Glia, 2012, 60, 382–392.
radiotracer via the venous catheter.
PET scans were collected from 0–120 min with the following
time frames: 3 × 1 min, 4 × 2 min, 3 × 3 min and 20 × 5 min.
12 V. E. Miron, S. K. Ludwin, P. J. Darlington, A. A. Jarjour,
B. Soliven, T. E. Kennedy and J. P. Antel, Am. J. Pathol.,
2010, 176, 2682–2694.
Emission data were corrected for dead time, scatter and 13 S. J. Jackson, G. Giovannoni and D. Baker,
attenuation and then reconstructed to a final resolution of J. Neuroinflammation, 2011, 8, 76.
2.0 mm full-width half maximum in all 3 dimensions at the 14 V. Brinkmann, Br. J. Pharmacol., 2009, 158, 1173–1182.
center of the field of view. PET and MRI images were co-regis- 15 J. W. Choi, S. E. Gardell, D. R. Herr, R. Rivera, C. W. Lee,
tered using automated image registration program (AIR). For
quantitative analyses, three-dimensional ROI (the global brain)
K. Noguchi, S. T. Teo, Y. C. Yung, M. Lu, G. Kennedy and
J. Chun, Proc. Natl. Acad. Sci. U. S. A., 2011, 108, 751–756.
was identified on the MRI and transformed to the recon- 16 S. Trkov, M. Stenovec, M. Kreft, M. Potokar, V. Parpura,
structed PET images to obtain time-activity curves. Activity B. Davletov and R. Zorec, Glia, 2012, 60, 1406–1416.
measures were standardized to body weight and the dose of 17 H. Noda, H. Takeuchi, T. Mizuno and A. Suzumura,
radioactivity injected to yield SUV.
J. Neuroimmunol., 2013, 256, 13–18.
18 E. W. Radue, P. O’Connor, C. H. Polman, R. Hohlfeld,
P. Calabresi, K. Selmaj, N. Mueller-Lenke, C. Agoropoulou,
F. Holdbrook, A. de Vera, L. Zhang-Auberson, G. Francis,
P. Burtin and L. Kappos, Arch. Neurol., 2012, 69, 1259–
1269.
Conflicts of interest
There are no conflicts to declare.
19 T. Karuppuchamy, E. h. Behrens, P. González-Cabrera,
G. Sarkisyan, L. Gima, J. D. Boyer, G. Bamias, P. Jedlicka,
M. Veny, D. Clark, R. Peach, F. Scott, H. Rosen and
J. Rivera-Nieves, Mucosal Immunol., 2016, 10, 162.
Acknowledgements
This work was supported by the USA National Institutes of 20 M. van der Giet, M. Tolle and B. Kleuser, Biol. Chem., 2008,
Health including the National Institute of Neurological 389, 1381–1390.
Disorders and Stroke, and the National Institute on Aging 21 J. E. Hughes, S. Srinivasan, K. R. Lynch, R. L. Proia,
[NS075527 and NS103988], the National Institute of Mental P. Ferdek and C. C. Hedrick, Circ. Res., 2008, 102, 950–958.
Health [MH092797], and National Institute of Biomedical 22 H. Lee, J. Deng, M. Kujawski, C. Yang, Y. Liu, A. Herrmann,
Imaging and Bioengineering [EB025815].
M. Kortylewski, D. Horne, G. Somlo, S. Forman, R. Jove and
This work was also partially supported by USA Department
H. Yu, Nat. Med., 2010, 16, 1421–1428.
of Energy Training Grant titled “Training in Techniques and 23 J. Liang, M. Nagahashi, E. Y. Kim, K. B. Harikumar,
Translation: Novel Nuclear Medicine Imaging Agents for
Oncology and Neurology” [DE-SC0008432]. American
Parkinson Disease Association (APDA), the Greater St Louis
A. Yamada, W. C. Huang, N. C. Hait, J. C. Allegood,
M. M. Price, D. Avni, K. Takabe, T. Kordula, S. Milstien and
S. Spiegel, Cancer Cell, 2013, 23, 107–120.
Chapter of the APDA and the Barnes Jewish Hospital 24 C. Watson, J. S. Long, C. Orange, C. L. Tannahill,
Foundation.
E. Mallon, L. M. McGlynn, S. Pyne, N. J. Pyne and
J. Edwards, Am. J. Pathol., 2010, 177, 2205–2215.
25 N. J. Pyne and S. Pyne, Nat. Rev. Cancer, 2010, 10, 489–503.
26 S. Ponnusamy, S. P. Selvam, S. Mehrotra, T. Kawamori,
A. J. Snider, L. M. Obeid, Y. Shao, R. Sabbadini and
B. Ogretmen, EMBO Mol. Med., 2012, 4, 761–775.
27 H. Liu, H. Jin, X. Yue, Z. Luo, C. Liu, A. J. Rosenberg and
Z. Tu, Mol. Imaging Biol., 2016, 18, 724–732.
Notes and references
1 R. L. Proia and T. Hla, J. Clin. Invest., 2015, 125, 1379–1387.
2 M. Kono and R. L. Proia, Exp. Cell Res., 2015, 333, 178–182.
3 V. A. Blaho and T. Hla, J. Lipid Res., 2014, 55, 1596–1608.
4 M. J. Lee, J. R. Van Brocklyn, S. Thangada, C. H. Liu, 28 H. Jin, H. Yang, H. Liu, Y. Zhang, X. Zhang,
A. R. Hand, R. Menzeleev, S. Spiegel and T. Hla, Science,
1998, 279, 1552–1555.
A. J. Rosenberg, Y. Liu, S. E. Lapi and Z. Tu, J. Nucl.
Cardiol., 2017, 24, 558–570.
5 T. Kohama, A. Olivera, L. Edsall, M. M. Nagiec, R. Dickson 29 H. Liu, H. Jin, X. Yue, J. Han, P. Baum, D. R. Abendschein
and S. Spiegel, J. Biol. Chem., 1998, 273, 23722–23728.
6 T. Hla and T. Maciag, J. Biol. Chem., 1990, 265, 9308–9313.
7 N. Ishizaka, H. Okazaki, K. Kurokawa, M. Kumada and
Y. Takuwa, Biochim. Biophys. Acta, 1994, 1218, 173–180.
and Z. Tu, Mol. Imaging, 2017, 16, 1–7.
30 V. P. Prasad, S. Wagner, P. Keul, S. Hermann, B. Levkau,
M. Schäfers and G. Haufe, Bioorg. Med. Chem., 2014, 22,
5168–5181.
This journal is © The Royal Society of Chemistry 2018
Org. Biomol. Chem., 2018, 16, 9171–9184 | 9183

View Article Online
Paper
Organic & Biomolecular Chemistry
31 R. S. Shaikh, S. S. Schilson, S. Wagner, S. Hermann, 41 H. F. Kung, S. R. Choi, W. Qu, W. Zhang and
P. Keul, B. Levkau, M. Schäfers and G. Haufe, J. Med.
Chem., 2015, 58, 3471–3484.
D. Skovronsky, J. Med. Chem., 2010, 53, 933–
941.
32 A. J. Rosenberg, H. Liu, H. Jin, X. Yue, S. Riley, S. J. Brown 42 Z. Zha, S. R. Choi, K. Ploessl, B. P. Lieberman, W. Qu,
and Z. Tu, J. Med. Chem., 2016, 59, 6201–6220.
33 Z. Luo, A. J. Rosenberg, H. Liu, J. Han and Z. Tu,
Eur. J. Med. Chem., 2018, 150, 796–808.
F. Hefti, M. Mintun, D. Skovronsky and H. F. Kung, J. Med.
Chem., 2011, 54, 8085–8098.
43 H. Pajouhesh and G. R. Lenz, NeuroRx, 2005, 2, 541–
553.
34 V. W. Pike, Trends Pharmacol. Sci., 2009, 30, 431–440.
35 A. J. Rosenberg, H. Liu and Z. Tu, Appl. Radiat. Isot., 2015, 44 D. S. Miller, B. Bauer and A. M. S. Hartz, Pharmacol. Rev.,
102, 5–9. 2008, 60, 196–209.
36 J. M. Harris and R. B. Chess, Nat. Rev. Drug Discovery, 2003, 45 N. Kihara, M. Hashimoto and T. Takata, Org. Lett., 2004, 6,
2, 214.
1693–1696.
37 C. S. Fishburn, J. Pharm. Sci., 2008, 97, 4167–4183.
38 T. Riley and J. Riggs-Sauthier, Pharm. Technol., 2008, 32, 88,
90–92, 94.
39 W. Li, P. Zhan, E. De Clercq, H. Lou and X. Liu, Prog.
Polym. Sci., 2013, 38, 421–444.
40 G. Mattheolabakis, C. C. Wong, Y. Sun, C. A. Amella,
R. Richards, P. P. Constantinides and B. Rigas,
J. Pharmacol. Exp. Ther., 2014, 351, 61–66.
46 J. L. Gilmore, J. E. Sheppeck, S. H. Watterson, L. Haque,
P. Mukhopadhyay, A. J. Tebben, M. A. Galella,
D. R. Shen, M. Yarde, M. E. Cvijic, V. Borowski,
K. Gillooly, T. Taylor, K. W. McIntyre, B. Warrack,
P. C. Levesque, J. P. Li, G. Cornelius, C. D’Arienzo,
A. Marino, P. Balimane, L. Salter-Cid, J. C. Barrish,
W. J. Pitts, P. H. Carter, J. Xie and A. J. Dyckman, J. Med.
Chem., 2016, 59, 6248–6264.
9184 | Org. Biomol. Chem., 2018, 16, 9171–9184
This journal is © The Royal Society of Chemistry 2018