A potent and selective C-11 labeled PET tracer for imaging sphingosine-1-phosphate receptor 2 in the CNS demonstrates sexually dimorphic expression

Article abstract of DOI:10.1039/c5ob00951k

Sphingosine-1-phosphate receptor 2 (S1PR2) plays an essential role in regulating blood-brain barrier (BBB) function during demyelinating central nervous system (CNS) disease. Increased expression of S1PR2 occurs in disease-susceptible CNS regions of female versus male SJL mice and in female multiple sclerosis (MS) patients. Here we reported a novel sensitive and noninvasive method to quantitatively assess S1PR2 expression using a C-11 labeled positron emission tomography (PET) radioligand [¹¹C]5a for in vivo imaging of S1PR2. Compound 5a exhibited promising binding potency with IC₅₀ value of 9.52 ± 0.70 nM for S1PR2 and high selectivity over S1PR1 and S1PR3 (both IC₅₀ > 1000 nM). [¹¹C]5a was synthesized in ～40 min with radiochemistry yield of 20 ± 5% (decayed to the end of bombardment (EOB), n > 10), specific activity of 222-370 GBq μmol^-1 (decayed to EOB). The biodistribution study in female SJL mice showed the cerebellar uptake of radioactivity at 30 min of post-injection of [¹¹C]5a was increased by Cyclosporin A (CsA) pretreatment (from 0.84 ± 0.04 ID% per g to 2.21 ± 0.21 ID% per g, n = 4, p < 0.01). MicroPET data revealed that naive female SJL mice exhibited higher cerebellar uptake compared with males following CsA pretreatment (standardized uptake values (SUV) 0.58 ± 0.16 vs. 0.48 ± 0.12 at 30 min of post-injection, n = 4, p < 0.05), which was consistent with the autoradiographic results. This data suggested that [¹¹C]5a had the capability in assessing the sexual dimorphism of S1PR2 expression in the cerebellum of the SJL mice. The development of radioligands for S1PR2 to identify a clinical suitable S1PR2 PET radiotracer, may greatly contribute to investigating sex differences in S1PR2 expression that contribute to MS subtype and disease progression and it will be very useful for detecting MS in early state and differentiating MS with other patients with neuroinflammatory diseases, and monitoring the efficacy of treating diseases using S1PR2 antagonism.

Full text of DOI:10.1039/c5ob00951k

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
A potent and selective C-11 labeled PET tracer for
imaging sphingosine-1-phosphate receptor 2 in
the CNS demonstrates sexually dimorphic
expression†
Cite this: Org. Biomol. Chem., 2015,
13, 7928
Xuyi Yue,a Hongjun Jin,a Hui Liu,a Adam J. Rosenberg,a Robyn S. Klein*b and
Zhude Tu*a
Sphingosine-1-phosphate receptor 2 (S1PR2) plays an essential role in regulating blood–brain barrier
(BBB) function during demyelinating central nervous system (CNS) disease. Increased expression of S1PR2
occurs in disease-susceptible CNS regions of female versus male SJL mice and in female multiple sclero-
sis (MS) patients. Here we reported a novel sensitive and noninvasive method to quantitatively assess
S1PR2 expression using a C-11 labeled positron emission tomography (PET) radioligand [11C]5a for in vivo
imaging of S1PR2. Compound 5a exhibited promising binding potency with IC50 value of 9.52 0.70 nM
for S1PR2 and high selectivity over S1PR1 and S1PR3 (both IC50 > 1000 nM). [11C]5a was synthesized in
∼40 min with radiochemistry yield of 20
5% (decayed to the end of bombardment (EOB), n > 10),
specific activity of 222–370 GBq µmol−1 (decayed to EOB). The biodistribution study in female SJL mice
showed the cerebellar uptake of radioactivity at 30 min of post-injection of [11C]5a was increased by
Cyclosporin A (CsA) pretreatment (from 0.84 0.04 ID% per g to 2.21 0.21 ID% per g, n = 4, p < 0.01).
MicroPET data revealed that naive female SJL mice exhibited higher cerebellar uptake compared with
males following CsA pretreatment (standardized uptake values (SUV) 0.58 0.16 vs. 0.48 0.12 at 30 min
of post-injection, n = 4, p < 0.05), which was consistent with the autoradiographic results. This data
suggested that [11C]5a had the capability in assessing the sexual dimorphism of S1PR2 expression in the
cerebellum of the SJL mice. The development of radioligands for S1PR2 to identify a clinical suitable
S1PR2 PET radiotracer, may greatly contribute to investigating sex differences in S1PR2 expression that
contribute to MS subtype and disease progression and it will be very useful for detecting MS in early state
and differentiating MS with other patients with neuroinflammatory diseases, and monitoring the efficacy
of treating diseases using S1PR2 antagonism.
Received 12th May 2015,
Accepted 9th June 2015
DOI: 10.1039/c5ob00951k
www.rsc.org/obc
lation.3 The life expectancy of patients with MS is significantly
reduced and the quality of life is adversely affected in terms of
1. Introduction
Multiple sclerosis (MS) is a neuroinflammatory demyelinating mobility and social functioning. MS also has a strong sex bias,
disease of the central nervous system (CNS) which leads to with the female to male ratio currently approaching 4 : 1.4–6
focal plaques of primary demyelination in conjunction with Relapsing-remitting MS (RRMS), the most common form of
BBB disruption in the gray and white matter of the brain and the disease in women, is a condition in which recurrent epi-
spinal cord.1,2 MS affects 2.5 million people worldwide with a sodes of new neurological dysfunction (relapses) are separated
prevalence of approximately 0.1% in the Caucasian popu- by periods of clinical stability (remission). Most RRMS patients
(>50%) will develop secondary progressive MS (SPMS), with
increased and sustained disability within 10–20 years after
diagnosis.7,8 The mechanisms of that contribute to pro-
aDepartment of Radiology, Washington University School of Medicine, St. Louis,
gression of MS are incompletely understood.
MO 63110, USA. E-mail: tuz@mir.wustl.edu; Fax: +1-314-362-8555;
Tel: +1-314-362-8487
Sphingolipids, main components of nervous tissue, have
been linked to MS several decades ago. Nowadays, especially
sphingosine-1-phosphate (S1P) is in the focus of pathophysio-
logical research and therapy development. S1P is a membrane-
derived lysophospholipid that binds with five subtype
bDepartments of Medicine, Anatomy & Neurobiology, Pathology & Immunology,
Washington University School of Medicine, St. Louis, MO 63131, USA.
E-mail: rklein@dom.wustl.edu; Fax: +1-314-362-9320; Tel: +1-314-286-2140
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
c5ob00951k
7928 | Org. Biomol. Chem., 2015, 13, 7928–7939
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
G-protein coupled receptors S1PR1–S1PR5. The activation of methylpyrazole with ethyl isobutyrylacetate using acetic acid as
S1PRs modulates a spectrum of key biological events including the solvent afforded compound 1.19 The reaction yield was low
activities in the CNS and immune response.9 Sphingolipid-like when propionic acid was employed as the solvent. Moreover,
immunomodulator fingolimod (FTY-720, Gilenya) is the first it’s very challenging to remove the acylated side product from
oral disease-modifying therapy approved by the US Food and the reaction of 5-amino-1,3-dimethylpyrazole with solvent pro-
Drug Administration (FDA) for RRMS.10 Phosphorylated Fingo- pionic acid due to its close polarity as the product. However,
limod sequesters lymphocytes within the lymph nodes and use of acetic acid as the solvent allowed the separation of the
prevents their trafficking to the CNS, via binding to S1PR sub- product from the acylated side product. Bromination of
types S1PR1, S1PR3, S1PR4, and S1PR5, but not S1PR2.11 1 afforded compound 2, followed by reaction with hydrazine to
Encouraged by the success of Fingolimod, much effort has produce the key intermediate 3. The 2-chloropyridine
been devoted to developing selective compounds targeting moiety was synthesized from commercially available 2-chloro-
S1PR subtypes, especially S1PR1 or S1PR2.12
6-methoxyisonicotinic acid. Treatment of 2-chloro-6-methoxyi-
In a recent study, we found increased expression of S1PR2 sonicotinic acid with diphenylphosphoryl azide afforded acyl
in disease-susceptible CNS regions of female MS patients, and azide 4. Reflux of compound 4 in toluene in situ produced the
suggest cell intrinsic sex differences in the S1PR2 expression isocyanate. A solution of hydrazine 3 in tetrahydrofuran (THF)
on CNS vasculature contribute to the development of disease was subsequently added to the above solution to give the first
cycles in females compared with males.13 Quantification of target compound 5a in moderate yield (Scheme 1).
S1PR2 levels would greatly contribute to studies of examining
Next, we focused on the modification of the methoxy group
sex differences in MS, and the identification of patients that in compound 5a to explore new analogues targeting on S1PR2.
may benefit from treatment using S1PR2 antagonists, which The syntheses of the target compounds 5b–5f are depicted
are less likely to cause lymphopenia.14 Combined with a suit- in Scheme 2. Reaction of methyl 2-chloroisonicotinate with
able radiotracer, positron emission tomography (PET), a silver(II) fluoride, followed a reported procedure20 to introduce
reliable well established noninvasive imaging method, is able a fluorine atom on the pyridine ring to generate compound 6.
to offer a unique sensitive way to quantitatively assess neuro- Saponification of compound 6 gave carboxylic acid 7 in quanti-
transmitter receptors, enzymes, and transporters in vivo.15 tative yield. Conversion of the acid to azide provided acyl azide
Currently PET is widely used in clinical diagnosis. Identifying 8, which was further reacted with hydrazine 3 to give target
a suitable PET tracer that can specifically assess the expression compound 5b. By replacing one of the chlorine atoms in 2,6-
of S1PR2 in vivo, will ultimately lead to better understanding dichloroisonicotinic acid with methoxyethoxy functional group
of the function of S1PR2 in the neuropathogenesis of MS.
generated intermediate 9, which was readily converted to
The inbred SJL mouse strain has been used as a model of target compound 5c using a similar procedure for preparing
the sexual dimorphism observed in MS, as SJL females are compound 5a. Alternatively, hydrolysis of 2,6-dichloroisonicoti-
more susceptible to experimental autoimmune encephalo- nic acid with aqueous sodium hydroxide solution at 130 °C,
myelitis (EAE) than males; and exhibit a relapsing-remitting followed by precipitation of the product by acidification using
disease pattern similar to that observed in MS patients.16,17 6 M HCl, produced the compound 11 in high yield. Protection
Moreover, sex difference has also been found in the CNS of 11 with chloromethyl methyl ether afforded two regio-
expression of S1PR2 in SJL mice, especially in the cerebel- isomers 12 and 13 in a 5/1 ratio, which were readily separated.
lum,13 which provides a good target for the validation of new- The esters were saponified to afford acids 14 and 16 in high
synthesized S1PR2 radioligands.
yields. Both compounds were converted to the acyl azide and
Herein we report the design and synthesis of a series of reacted with hydrazine 3 to provide target compounds 5d and
S1PR2 ligands containing similar core structures to the well- 5e. Attempts to remove the methoxymethoxyl (MOM) protect-
known S1PR2 selective antagonist JTE-013.18 In vitro competi- ing group in compound 5e using hydrogen chloride/dioxane
tive cell membrane binding assays are conducted to determine led to a complex resulted from the poor solubility of com-
the binding affinities of the newly synthesized analogues pound 5e in the solvent. To resolve the difficulty, deprotection
towards S1PR1, S1PR2, and S1PR3. Radiosynthesis of a S1PR2 of the MOM group of 5e was accomplished by using trifluoroa-
radioligand [11C]5a, in vivo evaluation of [11C]5a via auto- cetic acid in dichloromethane that smoothly produced the
radiography, biodistribution, and microPET studies on SJL C-11 radiolabeling precursor 18 in moderate yield. One of the
mice are accomplished. Our studies suggest that [11C]5a chlorine atoms in 2,6-dichloroisonicotinic acid was converted
demonstrates sexual dimorphism of S1PR2 expression in the into a fluoroethoxy group to afford fluorine containing inter-
cerebellum of SJL mice.
mediate 19. Following the above mentioned procedure, com-
pound 19 reacted with compound 3 to give the target
compound 5f.
2. Results
2.2. In vitro competitive binding assay
2.1. Chemistry
The in vitro competitive binding assays against [32P]-S1P for
The synthesis of S1PR2 ligands starts with the construction of the new synthesized target compounds 5a–5f were conducted
key hydrazine intermediate 3. Condensation of 5-amino-1,3-di- following our published protocol.21 Results showed that com-
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem., 2015, 13, 7928–7939 | 7929

View Article Online
Paper
Organic & Biomolecular Chemistry
Scheme 1 Synthesis of the target compound 5a. Reagents and conditions: (a) 5-amino-1,3-dimethylpyrazole, acetic acid, 140 °C, overnight, 18%;
(b) POBr3, anisole, reflux, 54%; (c) NH2NH2, EtOH, reflux, 92%; (d) diphenylphosphoryl azide, Et3N, EtOAc, 57%; (e) (i) toluene, reflux; (ii) THF, 3, 50 °C,
32% over two steps.
pounds 5a, 5e, and 5f exhibited promising binding potency 2.4. Biodistribution study
with IC50 value of 9.52 0.70 nM, 8.09 0.91 nM, 8.12 0.62
nM, respectively, while compounds 5b (IC50 = 135 21 nM)
Biodistribution studies were conducted to assess the tracer
uptake in the brain and peripheral organs/tissues. The uptake
and 5c (IC50 = 234
34 nM) only had moderate binding
of [11C]5a in female SJL mice at 30 min post injection showed
that the heart, lung, kidney and liver had the highest uptake,
which was consistent with S1PR2 distribution in mice.22 The
cerebellum and total brain displayed moderate tracer accumu-
lation with the ID% per gram values of 0.84 0.04 and 0.78
0.04 respectively. Pretreatment of mice with Cyclosporin A
(CsA), which improves blood brain barrier permeability via
modulation of P-glycoprotein (P-gp) with inhibition of the ATP
binding cassette efflux transporters,23,24 the cerebellar uptake
(ID% per g) within female mice was notably increased to
2.21 0.21 (n = 4, p < 0.01) upon CsA treatment. The uptake of
[11C]5a in lung, thymus, kidney and liver showed similar trend
in the CsA treatment group shown in Table 2.
potency towards S1P2 receptor. No binding potency was
observed for compound 5d towards S1PR2. More importantly,
compound 5a was seven-fold more potent than the well-known
S1PR2 antagonist – JTE-013 (IC50 = 68.47 7.45 nM, Fig. 1)
and also showed good selectivity towards S1P1 and S1P3 recep-
tors (IC50 > 1000 nM) (Table 1). Compounds 5a, 5e, 5f showed
similar calculated Log D7.4 values as JTE-013 except the calcu-
lated Log D7.4 for compound 5d was 1.01, which may cause it
to lose binding potency for S1PR2.
2.3. Radiochemistry
With the promising in vitro competitive binding potency for
several ligands having IC50 < 10 nM, we elected to radiolabel
5a using [11C]methyl iodide to make [11C]5a for further in vivo
validation. Various labeling conditions in terms of 11C-methyl-
2.5. Ex vivo autoradiography and microPET/computerized
tomography (CT) study
ating agents, reaction temperature, base, mobile phase were To further validate the application of [11C]5a for imaging
explored, the optimized reaction conditions of this reaction S1PR2, ex vivo autoradiography studies and microPET/CT
were: 1.0–1.3 mg precursor, 200 µL DMF, 3 µL of 5 M aqueous scans in male and female SJL mice with CsA pre-treatment
KOH solution, 85 °C, 5 min (Scheme 3). Under optimized were also performed. The results from both autoradiography
semi-preparative HPLC conditions (Agilent Zorbax SB-C18 and PET/CT studies consistently demonstrated higher accumu-
column, 250 × 9.2 mm, mobile phase 45% acetonitrile in lation of [11C]5a in the cerebellum of female SJL mice than
0.1 M ammonium formate, pH 6.5, flow rate 4.0 mL min−1
,
that observed in males (indicated by white dashed circles in
detection wavelength 254 nm), the retention time was Fig. 2A). Furthermore, quantitative microPET analysis showed
15–17 min for [11C]5a, 4–5 min for precursor 18, and 5–6 min that the average cerebellar uptake of [11C]5a in female mice
for a radioactive vice-product. It took about 40 min starting was statistically higher (n = 5, p = 0.021) than that in male
from the release of [11C]methyl iodide to formation of the final mice (standardized uptake values (SUV) 0.58 0.16 vs. 0.48
injection dose of [11C]5a with radiochemistry yield of 20 5% 0.12, Fig. 2B). The autoradiography and microPET/CT scans
(decayed to the end of bombardment (EOB), n > 10), specific clearly demonstrated a sex difference in the cerebellar tracer
activity of 222–370 GBq µmol−1 (EOB).
uptake of [11C]5a in SJL mice and was consistent with our pre-
7930 | Org. Biomol. Chem., 2015, 13, 7928–7939
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
Scheme 2 Synthesis of target compounds 5b–5f. Reagents and conditions: (a) silver(II) fluoride, acetonitrile; (b) LiOH, THF/H2O (1/2, v/v), then 1 M
HCl; (c) diphenylphosphoryl azide, Et3N, 1,4-dioxane; (d) (i) toluene, reflux; (ii) THF, hydrazine 3, 50 °C; (e) 2-methoxyethanol, tBuOK, THF; (f) (i) 2 M
aqueous NaOH, 130 °C; (ii) 6 M HCl; (g) MOMCl, DIPEA, CH2Cl2; (h) TFA, CH2Cl2, 0 °C to rt, 2 h; (i) (i) 2-fluoroethanol, tBuOK, THF; (ii) diphenylphos-
phoryl azide, Et3N, 1,4-dioxane.
vious report of S1PR2 expression in the cerebellum of SJL levels will facilitate better understanding of the neuropatholo-
mice,13 which support the target selectivity and specificity of genesis of MS, especially with regard to the role of S1PR2 in
this newly developed S1PR2 radiotracer.
neuroinflammation. Here we reported the design and syn-
thesis of six S1PR2 ligands containing similar core structures
as the well-known S1PR2 selective antagonist JTE-013. The syn-
thesis took 6–9 steps. The synthesized ligands 5a–5f were
tested for binding potency towards S1PR1, S1PR2, and S1PR3
by in vitro competitive cell membrane binding assay. Three
compounds showed good binding potency (IC50 < 10 nM) for
S1PR2, and high selectivity over S1PR1 and S1PR3. Based on
the in vitro data, we conclude that minor optimization of the
3. Discussion
Our previous study revealed that S1PR2 plays an essential role
in regulating blood–brain barrier (BBB) function during
demyelinating CNS disease.13 Development of a reliable and
noninvasive imaging method that detects S1PR2 expression
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem., 2015, 13, 7928–7939 | 7931

View Article Online
Paper
Organic & Biomolecular Chemistry
The radiolabeling for the [11C]5a was challenging due to
two competitive O-11C-methylation and N-11C-methylation
product formulated in the reaction. To increase the desired
O-11C-methylation radiolabeling product, different conditions
in terms of reaction temperature, 11C-methylating agents
including [11C]methyl iodide and [11C]methyl triflate, base and
mobile phase were tested (see Table S1†). Although cesium car-
bonate was reported to be an optimized base for the O-11C-
methylation of similar pyridin-2-ol substrates,25,26 only mode-
rate yield was achieved in our reaction when using cesium car-
bonate as the base. In our system, we observed that potassium
hydroxide combined with [11C]methyl iodide gave the highest
O-11C-methylation to N-11C-methylation ratio compared to
other conditions including [11C]MeOTf/KOH, [11C]MeI/Cs2CO3,
[11C]MeI/CsOH. Using pH 6.5 buffered aqueous ammonium
formate solution improved resolution and radiochemical
purity over pH 4.5 buffered aqueous ammonium formate.
Under the optimized semi-preparative HPLC conditions, [11C]
5a with retention time at 15–17 min was separated well from
the precursor 18 (tR = 4–5 min) and radiolabeled by-product
(tR = 5–6 min). [11C]5a was synthesized in about 40 min starting
from the release of [11C]methyl iodide to formation of the final
injection dose with radiochemistry yield of 20 5% (decayed
to EOB, n > 10), specific activity of 222–370 GBq µmol−1 (EOB).
Biodistribution study of [11C]5a indicated a higher cerebel-
lar uptake of the tracer in female SJL mice with CsA pre-treat-
Fig. 1 Competitive binding curves of compound 5a and JTE-013 for
S1PR2. A CHO cell membrane containing recombinant human S1PR2
was used in a [32P]S1P competitive binding assay to measure the binding
affinity for compound 5a (red line, fitted IC50 = 9.52
JTE-013 (blue dashed line, fitted IC50 = 68.47 7.45 nM).
0.70 nM) and
functional group may led to significant change on the binding
affinities, which will be useful for our future exploring the
these structures reported in this manuscript.
Table 1 Binding affinities (IC50 values, nM) of new synthesized compounds towards S1P2, S1P2, S1P3 receptors
IC50 (nM)
a
Compound
c Log D7.4
S1PR2
S1PR1
S1PR3
JTE-013
5a
5b
5c
5d
5e
5f
4.37
4.43
3.36
4.10
1.01
3.83
4.66
68.47 7.45
9.52 0.70
135 21
234 34
>1000
8.09 0.91
8.12 0.62
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
a Calculated value at pH = 7.4 by ACD/I-Lab ver. 7.0 (Advanced Chemistry Development, Inc., Canada).
7932 | Org. Biomol. Chem., 2015, 13, 7928–7939
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
Scheme 3 Radiolabeling of the precursor 18 for [11C]5a and postulated by-product.
ment than that in control the group. It is notable that CsA
increased the cerebellar uptake of [11C]5a in female SJL mice
by ∼2.6-fold, which suggest the brain uptake towards [11C]5a is
modestly modulated by P-gp27–29 and thus is limited. Although
species difference is common in P-gp transport of radio-
ligands30 and [11C]5a may have relatively high uptake in monkey
or human brains. On the other hand, the low brain uptake of
[11C]5a might also be partially attributed to the relatively low
S1PR2 expression level in the brain, since the localization of
S1PR2 in SJL mice brain is mainly on endothelial cells.13 Struc-
tural optimization of 5a in the pyrazole-pyridine moiety to
avoid P-gp efflux transport and enhance BBB penetration,
while retaining binding potency is ongoing. Among the syn-
thesized ligands, compound 5e with a fluoroethoxy group also
displayed high binding potency (IC50 = 8.09 0.91 nM). Here
Table 2 Biodistribution of [11C]5a at 30 min post-injection in control
versus CsA pre-treated female SJL mice
ID% per g
Female control
Female with CsA
Blood
Heart
Lung
Muscle
Fat
Thymus
Spleen
Kidney
Liver
Cerebellum
Total brain
3.52 0.29
3.56 0.56
5.71 0.47
1.81 0.18
3.30 0.54
2.81 0.16
2.95 0.24
7.96 1.29
13.31 1.34
0.84 0.04
0.78 0.04
5.55 0.61
5.89 0.62
13.79 2.56
2.65 0.30
3.13 1.70
4.60 0.22
3.99 0.39
9.43 0.96
15.91 1.06
2.21 0.21
1.96 0.18
Fig. 2 (A) Autoradiography and PET/CT images of [11C]5a in the brains of female and male SJL mice (n = 5) upon CsA pre-treatment; (B) quantifi-
cation of cerebellar uptake of [11C]5a in female and male SJL mice upon CsA treatment from microPET. All the histograms were average values from
5 independent experiments, error bars were the standard derivations.
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem., 2015, 13, 7928–7939 | 7933

View Article Online
Paper
Organic & Biomolecular Chemistry
we focused on our first investigation of [11C]5a as a S1PR2 PET out on Merck 60 F254 silica gel glass plates, and visualization
ligand. Exploration of 5f and other F-18 labeled ligands is was aided by UV.
undergoing within our group.
4-Isopropyl-1,3-dimethyl-1H-pyrazolo[3,4-b]pyridin-6-ol (1).33
While the pathogenesis of MS involves both genetic suscep- A mixture of 5-amino-1,3-dimethylpyrazole (5.4 g, 49 mmol)
tibility in terms of immune function and activation,31 and and ethyl isobutyrylacetate (10 g, 63 mmol) in acetic acid
environmental causes such as Epstein-Barr virus infections,32 (100 mL) was heated at 140 °C in a sealed tube for 20 h, the
studies focused on sex differences are needed for new insights mixture was concentrated to remove the solvent and subjected
into hormonal and genetic sex-related factors that may impact to silica gel chromatography using CH2Cl2/MeOH (20/1) as the
on disease susceptibility and to identify new treatment targets. eluent to afford compound 1 (1.8 g, 18% yield) as a light
1
In addition, defined biomarkers that improve diagnostic evalu- yellow solid. H NMR (400 MHz, CDCl3) δ 6.14 (s, 1H), 3.96 (s,
ation and treatment follow-up in a sex-specific fashion are 3H), 3.32–3.22 (m, 1H), 2.49 (s, 3H), 1.27 (d, J = 6.4 Hz, 6H).
urgently needed. The identification of sex differences in S1PR2
6-Bromo-4-isopropyl-1,3-dimethyl-1H-pyrazolo[3,4-b]pyridine
expression and signaling in the neuropathogenesis of CNS (2).33 To a solution of 1 (1.0 g, 4.9 mmol) in anisole (10 mL)
autoimmunity13 is a critical step towards gender-based medi- was added phosphorus oxybromide (2.0 g, 7.0 mmol). The
cine for the treatment of MS. Targeting S1PR2 may also treat reaction was heated at 130 °C overnight. Water was added to
MS without impacting normal immune function. Detection of quench the reaction which was then extracted with EtOAc. The
S1PR2 via a novel, sensitive and noninvasive imaging PET combined organic layers were washed with saturated aqueous
modality will improve the assessment of the receptor in vivo. NaHCO3, and saturated aqueous NaCl. The organic phase was
The current developed tracer meets the above requirement and concentrated in vacuo, and the residue was subjected to silica
is able to detect the sexual dimorphism of S1PR2 expression gel chromatography using CH2Cl2/MeOH (30/1) as the eluent
within autoimmune susceptible SJL mice in terms of bio- to afford compound 2 (0.70 g, 54% yield) as a light yellow
1
distribution, autoradiography, microPET/CT scans.
solid. H NMR (400 MHz, CDCl3) δ 6.99 (s, 1H), 3.95 (s, 3H),
3.53–3.43 (m, 1H), 2.59 (s, 3H), 1.28 (d, J = 6.8 Hz, 6H).
6-Hydrazinyl-4-isopropyl-1,3-dimethyl-1H-pyrazolo[3,4-b]pyr-
idine (3).33 To a solution of 2 (0.45 g, 1.7 mmol) in ethanol
(1.0 mL) was added anhydrous hydrazine (2.8 g, 88 mmol).
4. Conclusions
We designed and synthesized a spectrum of S1PR2 ligands, The reaction was heated in a sealed tube at 120 °C overnight.
chose a ligand that has high binding potency and high selecti- The mixture was concentrated in vacuo, and the residue was
vity for S1P receptor 2 to be labeled with [11C]methyl iodide. subjected to silica gel chromatography using CH2Cl2/MeOH
Utilizing the radioligand [11C]5a, we successively demonstrated (20/1) as the eluent to afford compound 3 (0.34 g, 92% yield)
the sexual dimorphism of S1PR2 expression in SJL mice with as a yellow solid. 1H NMR (400 MHz, CDCl3) δ 6.14 (s, 1H),
CsA pre-treatment. Further exploring the lead structures to 4.00 (br, 2H), 3.84 (s, 3H), 3.40–3.30 (m, 1H), 2.50 (s, 3H), 1.21
identify a promising radioligands for S1PR2 with ability to (d, J = 7.2 Hz, 6H).
penetrate the blood brain barrier, may greatly contribute to the
2-Chloro-6-methoxyisonicotinoyl azide (4). To a solution of
diagnosis and treatment of MS and other neuroinflammatory 2-chloro-6-methoxyisonicotinic acid (0.28 g, 1.5 mmol) in ethyl
diseases.
acetate (3 mL) was added Et3N (0.38 g, 3.8 mmol). The solution
was cooled to 0 °C and diphenylphosphoryl azide (0.45 g,
1.7 mmol) was added dropwise. The reaction was warmed to
room temperature upon addition and was stirred overnight.
The mixture was concentrated in vacuo, and the residue was
subjected to silica gel chromatography using hexane/EtOAc
5. Materials and methods
5.1. Chemistry
General. All reagents and chemicals were purchased from (30/1) as the eluent to afford compound 4 (0.18 g, 57% yield)
commercial suppliers and used without further purification as a white solid. Mp 67–69 °C; 1H NMR (400 MHz, Acetone-d6)
unless otherwise stated. Melting points were determined on a δ = 7.21 (s, 1H), 6.99 (s, 1H), 3.82 (s, 3H); 13C NMR
MEL-TEMP 3.0 apparatus and are uncorrected. 1H NMR and (100.7 MHz, Acetone-d6) δ 169.8, 164.5, 149.3, 142.9, 114.5,
13C NMR spectra were recorded at 300 or 400 MHz on a Varian 108.9, 54.2; HRMS (ESI) calcd for C7H6ClN4O2 213.0179,
Mercury-VX spectrometer using CDCl3, Acetone-d6, CD3OD, [M + H]+, found 213.0174.
DMSO-d6 as solvent. All chemical shift values were reported in
N-(2-Chloro-6-methoxypyridin-4-yl)-2-(4-isopropyl-1,3-dimethyl-
parts per million (ppm) (δ). The following abbreviations were 1H-pyrazolo[3,4-b]pyridin-6-yl)hydrazine-1-carboxamide (5a).
used to describe peak patterns wherever appropriate: br = Under nitrogen, a solution of 4 (40 mg, 0.19 mmol) in toluene
broad, s = singlet, d = doublet, t = triplet, q = quartet, m = (0.5 mL) was refluxed for 3 hours to in situ produce the iso-
multiplet. HRMS analyses were conducted in Washington cyanate. The reaction mixture was allowed to cool to 50 °C and
University Resource for Biomedical and Bio-organic Mass a solution of 3 (31 mg, 0.14 mmol) in anhydrous THF (1 mL)
Spectrometry. Preparative chromatography was performed on was added dropwise. The reaction was then stirred at 50 °C
Chemglass chromatography column using 230–400 mesh overnight. The mixture was concentrated in vacuo, and the
silica gel purchase from Silicycle. Analytical TLC was carried residue was subjected to silica gel chromatography using
7934 | Org. Biomol. Chem., 2015, 13, 7928–7939
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
CH2Cl2/MeOH (20/1) as the eluent to afford compound 5a residue was subjected to silica gel chromatography using
1
(18 mg, 32% yield) as a light yellow solid. Mp 243–245 °C; H CH2Cl2/MeOH (30/1) as the eluent to afford compound 5b
NMR (400 MHz, CD3OD) δ 7.09 (s, 1H), 6.82 (s, 1H), 6.40 (s, (8 mg, 17% yield) as a light yellow solid. Mp 166–168 °C;
1H), 3.74 (s, 3H), 3.73 (s, 3H), 3.42–3.37 (m, 1H), 2.45 (s, 3H), 1H NMR (400 MHz, CDCl3) δ 8.16 (br, 1H), 7.26 (s, 1H), 7.14 (s,
1.22 (d, J = 6.4 Hz, 6H); 13C NMR (100.7 MHz, CD3OD) δ 166.6, 1H), 6.61 (br, 1H), 6.44 (br, 1H), 6.40 (s, 1H), 3.85 (s, 3H),
164.7, 160.0, 156.2, 150.6, 148.5, 145.9, 140.0, 110.0, 106.2, 3.50–3.43 (m, 1H), 2.55 (s, 3H), 1.26 (d, J = 6.8 Hz, 6H);
99.1, 96.1, 53.1, 31.8, 29.4, 21.7, 13.4; HRMS (ESI) calcd for 13C NMR (100.7 MHz, CDCl3) δ 164.4, 163.0, 156.3 (d, J =
C18H22ClN7NaO2 426.1421, [M + Na]+, found 426.1403.
225.1 Hz), 150.8, 150.6, 149.2, 140.2, 110.5, 110.3, 97.5, 96.8,
Methyl 2-chloro-6-fluoroisonicotinate (6). To a solution of 96.4, 33.3, 29.8, 22.9, 15.3; HRMS (ESI) calcd for
methyl 2-chloroisonicotinate (1.0 g, 6.0 mmol) in anhydrous C17H19ClFN7NaO 414.1221, [M + Na]+, found 414.1207.
acetonitrile (50 mL) was added silver(II) fluoride (2.6 g,
2-Chloro-6-(2-methoxyethoxy)isonicotinic acid (9). To a solu-
18 mmol) under nitrogen. The reaction was stirred in a sealed tion of 2,6-dichloroisonicotinic acid (1.1 g, 5.7 mmol) in an-
tube at 40 °C overnight. The suspension was filtered through hydrous THF (5 mL) was added a solution of potassium tert-
Celite and the solution was concentrated by rotary evaporation. butoxide (1.6 g, 14.3 mmol) in THF (14 mL) dropwise at 0 °C.
The residue was subjected to silica gel chromatography using 30 min later, 2-methoxyethanol was added dropwise and the
hexane/EtOAc (40/1) as the eluent to afford compound 6 reaction was continued at room temperature overnight. 1 N
(0.25 g, 0.58 g of the starting material was recovered, 50% yield aqueous hydrochloric acid was added to adjust the solution
based on recovered starting material) as a clear liquid. This pH to 4. The mixture was extracted with EtOAc, the combined
compound is a known compound and commercially available. organic layers were dried over sodium sulfate. The concen-
1H NMR (400 MHz, Acetone-d6) δ 7.65–7.63 (m, 1H), 7.37–7.36 trated residue was subjected to silica gel chromatography
(m, 1H), 3.85 (s, 3H).
using CH2Cl2/MeOH (4/1) as the eluent to afford compound 9
2-Chloro-6-fluoroisonicotinic acid (7). To a solution of 6 as a white solid (1.6 g, 98% yield). Mp 138–140 °C; 1H NMR
(0.24 g, 1.3 mmol) in THF (2 mL) was added aqueous lithium (400 MHz, CD3OD) δ 7.36 (s, 1H), 7.17 (s, 1H), 4.41 (t, J =
hydroxide solution (40 mg in 4 mL water, 1.6 mmol) dropwise 5.2 Hz, 2H), 3.71 (t, J = 4.4 Hz, 2H), 3.38 (s, 3H); 13C NMR
at 0 °C. The reaction was stirred at room temperature over- (100.7 MHz, CD3OD) δ 164.9, 163.8, 148.7, 143.8, 115.5, 109.4,
night. 2 M HCl(aq) was added to neutralize the reaction. The 70.3, 65.8, 57.7; HRMS (ESI) calcd for C9H11ClNO4 232.0377,
mixture was extracted with EtOAc, the combined organic layers [M + H]+, found 232.0368.
were dried over sodium sulfate. The concentrated residue was
2-Chloro-6-(2-methoxyethoxy)isonicotinoyl azide (10). To a
subjected to silica gel chromatography using CH2Cl2/MeOH solution of 9 (0.15 g, 0.65 mmol) in anhydrous 1,4-dioxane
(3/1) as the eluent to afford compound 7 (0.22 g, 99% yield) as (2 mL) was added Et3N (0.10 g, 0.97 mmol). The solution
a white soild. Mp 118–120 °C; 1H NMR (400 MHz, CD3OD) was cooled to 0 °C and diphenylphosphoryl azide (0.27 g,
δ 7.52 (s, 1H), 7.20 (d, J = 1.6 Hz, 1H); 13C NMR (100.7 MHz, 0.97 mmol) was added dropwise. The reaction was warmed to
CD3OD) δ 164.7, 162.7 (d, J = 245.3 Hz), 148.8 (d, J = 14.2 Hz), room temperature upon completion of the addition and was
148.2, 121.3 (d, J = 12.2 Hz), 107.8 (d, J = 44.2 Hz); HRMS (ESI) stirred overnight. The mixture was concentrated in vacuo, and the
calcd for C6H4ClFNO2 175.9915, [M + H]+, found 175.9905.
residue was subjected to silica gel chromatography using hexane/
2-Chloro-6-fluoroisonicotinoyl azide (8). To a solution of 7 EtOAc (20/1) as the eluent to afford compound 10 (200 mg, 84%
1
(0.10 g, 0.57 mmol) in anhydrous 1,4-dioxane (2 mL) was yield) as a light yellow liquid. H NMR (400 MHz, Acetone-d6)
added Et3N (0.12 g, 1.1 mmol). The solution was cooled to δ 7.25 (s, 1H), 7.05 (s, 1H), 4.33 (t, J = 4.4 Hz, 2H), 3.58 (t, J =
0 °C and diphenylphosphoryl azide (0.31 g, 1.1 mmol) was 4.8 Hz, 2H), 3.22 (s, 3H); 13C NMR (100.7 MHz, Acetone-d6)
added dropwise. The reaction was warmed to room tempera- δ 169.9, 164.2, 149.1, 143.2, 114.6, 109.1, 70.1, 66.5, 57.9; HRMS
ture upon addition and was stirred overnight. The mixture was (ESI) calcd for C9H10ClN4O3 257.0441, [M + H]+, found 257.0436.
concentrated in vacuo, and the residue was subjected to silica
N-(2-Chloro-6-(2-methoxyethoxy)pyridin-4-yl)-2-(4-isopropyl-
gel chromatography using hexane/EtOAc (50/1) as the eluent to 1,3-dimethyl-1H-pyrazolo[3,4-b]pyridin-6-yl)hydrazine-1-car-
afford compound 8 (35 mg, 31% yield) as a yellow solid. Mp boxamide (5c). A solution of 10 (126 mg, 0.49 mmol) in toluene
1
39–40 °C; H NMR (400 MHz, Acetone-d6) δ 7.71 (s, 1H), 7.44 (1.2 mL) was refluxed at 120 °C for 3 hours under nitrogen to
(d, J = 2.0 Hz, 1H); 13C NMR (100.7 MHz, Acetone-d6) δ 169.0, in situ produce the isocyanate. To the above mixture was added
162.8 (d, J = 244.9 Hz), 149.3, 145.8, 120.9, 107.8 (d, J = dropwise a solution of 3 (140 mg, 0.64 mmol) in anhydrous
43.6 Hz).
THF (1.2 mL) at 50 °C. The reaction continued to stir at 50 °C
N-(2-Chloro-6-fluoropyridin-4-yl)-2-(4-isopropyl-1,3-dimethyl- for overnight. The mixture was concentrated in vacuo, and the
1H-pyrazolo[3,4-b]pyridin-6-yl)hydrazine-1-carboxamide (5b). residue was subjected to silica gel chromatography using
Under nitrogen, a solution of 8 (30 mg, 0.15 mmol) in toluene CH2Cl2/MeOH (30/1) as the eluent to afford compound 5c
1
(0.8 mL) was refluxed for 3 hours to in situ produce the iso- (78 mg, 35% yield) as a light yellow solid. Mp 210–212 °C; H
cyanate. The reaction mixture was allowed to cool to 50 °C and NMR (400 MHz, CDCl3) δ 8.22 (br, 1H), 7.45 (br, 1H), 7.07 (br,
a solution of 3 (30 mg, 0.12 mmol) in anhydrous THF (0.8 mL) 1H), 6.99 (s, 1H), 6.84 (s, 1H), 6.36 (s, 1H), 4.36 (t, J = 4.0 Hz,
was added dropwise. The reaction was then stirred at 50 °C 2H), 3.80 (s, 3H), 3.65 (t, J = 4.4 Hz, 2H), 3.41–3.38 (m, 1H),
overnight. The mixture was concentrated in vacuo, and the 3.35 (s, 3H), 2.50 (s, 3H), 1.21 (d, J = 6.8 Hz, 6H); 13C NMR
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem., 2015, 13, 7928–7939 | 7935

View Article Online
Paper
Organic & Biomolecular Chemistry
(100.7 MHz, CDCl3) δ 163.8, 158.4, 156.7, 156.2, 150.5, 149.1, 75.7, 56.4; HRMS (ESI) calcd for C8H8ClNNaO4 240.0040, [M +
148.7, 140.0, 109.6, 106.9, 97.7, 97.3, 70.8, 65.6, 58.9, 33.2, Na]+, found 240.0034.
29.6, 22.8, 15.1; HRMS (ESI) calcd for
470.1683, [M + Na]+, found 470.1674.
C
20H26ClN7NaO3
6-Chloro-1-(methoxymethyl)-2-oxo-1,2-dihydropyridine-4-carb-
onyl azide (15). To a solution of 14 (0.41 g, 1.9 mmol) in an-
6-Chloro-2-oxo-1,2-dihydropyridine-4-carboxylic acid (11). hydrous 1,4-dioxane (16 mL) was added Et3N (0.28 g,
2,6-Dichloroisonicotinic acid (2.5 g, 13 mmol) was added to 2.8 mmol). The solution was cooled to 0 °C and diphenylphos-
2 N aqueous NaOH solution (32 mL). The solution was phoryl azide (0.77 g, 2.8 mmol) was added dropwise. The reac-
refluxed at 130 °C for overnight. 6 N HCl was added to the tion was warmed to room temperature upon completion of the
solution to precipitate the product when the solution was addition and stirred overnight. The mixture was concentrated
cooled down to room temperature. The suspension was filtered in vacuo, and the residue was subjected to silica gel chromato-
and the white solid was collected, dried in vacuo to afford the graphy using hexane/EtOAc (6/1) as the eluent to afford com-
product 11 (2.2 g, 99% yield) as a white solid. Mp 186–188 °C; pound 15 (0.28 g, 62% yield) as a yellow solid. Mp 99–100 °C;
1H NMR (400 MHz, DMSO-d6) δ 12.82 (br, 1H), 7.18 (s, 1H), 1H NMR (400 MHz, CDCl3) δ 7.06 (s, 1H), 6.69 (s, 1H), 5.55 (s,
7.02 (s, 1H); 13C NMR (100.7 MHz, DMSO-d6) δ 165.2, 164.8, 2H), 3.39 (s, 3H); 13C NMR (100.7 MHz, Acetone-d6) δ 170.0,
148.3, 144.7, 114.3, 108.9; HRMS (ESI) calcd for C6H5ClNO3 162.1, 140.7, 138.9, 120.4, 103.5, 75.6, 56.7; HRMS (ESI) calcd
173.9958, [M + H]+, found 173.9946.
Methoxymethyl 6-chloro-1-(methoxymethyl)-2-oxo-1,2-di-
for C16H15Cl2N8O6 485.0492, [2M + H]+, found 485.0490.
N-(6-Chloro-1-(methoxymethyl)-2-oxo-1,2-dihydropyridin-4-yl)-
hydropyridine-4-carboxylate (12)/methoxymethyl 2-chloro-6- 2-(4-isopropyl-1,3-dimethyl-1H-pyrazolo[3,4-b]pyridin-6-yl)hydra-
(methoxymethoxy)isonicotinate (13). To a solution of com- zine-1-carboxamide (5d). A solution of 15 (75 mg, 0.31 mmol)
pound 11 (0.87 g, 5 mmol) in CH2Cl2 (100 mL) was added in toluene (2 mL) was refluxed for 3 hours under nitrogen to
i-Pr2NEt (3.2 g, 25 mmol). The solution was cooled to 0 °C and in situ produce the isocyanate. To the above mixture was added
chloromethyl methyl ether (1.6 g, 20 mmol) was added drop- dropwise a solution of 3 (53 mg, 0.24 mmol) in anhydrous
wise. The reaction was warmed to room temperature after com- THF (2 mL) at 60 °C. The reaction continued to stir at 60 °C
pletion of the addition and stirred overnight. The solution was overnight. The mixture was concentrated in vacuo, and the
quenched with water and extracted with CH2Cl2. The com- residue was subjected to silica gel chromatography using
bined organic phase was washed with saturated aqueous CH2Cl2/MeOH (40/1) as the eluent to afford compound 5d
NaHCO3, saturated sodium chloride solution, and dried with (92 mg, 88% yield) as a light yellow solid. Mp 172–174 °C;
anhydrous sodium sulfate. The concentrated residue was sub- 1H NMR (400 MHz, CD3OD) δ 6.77 (s, 1H), 6.71 (s, 1H), 6.43 (s,
jected to silica gel chromatography using CH2Cl2/MeOH (40/1) 1H), 5.43 (s, 2H), 3.79 (s, 3H), 3.45–3.38 (m, 1H), 3.32 (s, 3H),
as the eluent to afford compound 12 (polar, 876 mg, 67% 2.49 (s, 3H), 1.25 (d, J = 6.8 Hz, 6H); 13C NMR (100.7 MHz,
yield) as a clear liquid and 13 (less polar, 180 mg, 14% yield) CD3OD) δ 164.6, 159.7, 155.6, 150.8, 150.0, 139.8, 137.6, 107.9,
as a clear liquid.
102.1, 100.2, 98.6, 98.5, 74.9, 56.2, 31.9, 29.4, 21.9, 13.7; HRMS
Compound 12: 1H NMR (400 MHz, CDCl3) δ 7.11 (s, 1H), (ESI) calcd for C19H24ClN7NaO3 456.1527, [M + Na]+, found
6.73 (s, 1H), 5.56 (s, 2H), 5.39 (s, 2H), 3.48 (s, 3H), 3.39 (s, 3H); 456.1514.
13C NMR (100.7 MHz, CDCl3) δ 163.12, 163.05, 140.4, 138.4,
2-Chloro-6-(methoxymethoxy)isonicotinic acid (16). To
121.3, 105.5, 92.1, 75.8, 58.1, 57.8; HRMS (ESI) calcd for solution of 13 (0.48 g, 1.8 mmol) in THF (2.4 mL) was added
C10H12ClNNaO5 284.0302, [M + Na]+, found 284.0300.
aqueous lithium hydroxide solution (57 mg in 4.8 mL water,
a
Compound 13: 1H NMR (400 MHz, CDCl3) δ 7.40 (d, J = 2.4 mmol) dropwise at 0 °C. The reaction was stirred at room
0.8 Hz, 1H), 7.20 (d, J = 0.8 Hz, 1H), 5.43 (s, 2H), 5.40 (s, 2H), temperature overnight. 2 N HCl(aq) was added to neutralize the
3.47 (s, 3H), 3.44 (s, 3H); 13C NMR (100.7 MHz, CDCl3) δ 163.1, reaction at 0 °C. The mixture was extracted with EtOAc, the
162.8, 149.2, 142.5, 116.7, 109.8, 92.8, 91.9, 58.0, 57.3; HRMS combined organic phase was dried over sodium sulfate. The
(ESI) calcd for C10H12ClNNaO5 284.0302, [M + Na]+, found concentrated residue was subjected to silica gel chromato-
284.0289.
graphy using CH2Cl2/MeOH (4/1) as the eluent to afford com-
6-Chloro-1-(methoxymethyl)-2-oxo-1,2-dihydropyridine-4-carb- pound 16 (0.31 g, 77% yield) as a light yellow solid. Mp
oxylic acid (14). To a solution of 12 (1.4 g, 5.3 mmol) in THF 150–152 °C; 1H NMR (400 MHz, Acetone-d6) δ 7.33 (s, 1H),
(5 mL) was added an aqueous solution of lithium hydroxide 7.12 (s, 1H), 5.36 (s, 2H), 3.35 (s, 3H); 13C NMR (100.7 MHz,
(0.17 g in 10 mL water, 7.0 mmol) dropwise at 0 °C. The reac- acetone-d6) δ 164.0, 163.0, 148.7, 144.1, 116.6, 109.8, 92.7,
tion was stirred at room temperature for overnight. 2 N HCl 56.4; HRMS (ESI) calcd for C8H8ClNNaO4 240.0040, [M + Na]+,
solution was added to acidify the reaction. The mixture was found 240.0036.
extracted with EtOAc, the combined organic phase was
2-Chloro-6-(methoxymethoxy)isonicotinoyl azide (17). To a
dried over sodium sulfate. The concentrated residue was sub- solution of 16 (0.24 g, 1.1 mmol) in anhydrous 1,4-dioxane
jected to silica gel chromatography using CH2Cl2/MeOH (4/1) (5 mL) was added Et3N (0.14 g, 1.4 mmol). The solution was
as the eluent to afford compound 14 (0.82 g, 71% yield) as a cooled to
0 °C and diphenylphosphoryl azide (0.39 g,
white solid. Mp 161–163 °C; 1H NMR (400 MHz, CD3OD) 1.4 mmol) was added dropwise. The reaction was warmed to
δ 6.94 (s, 1H), 6.74 (s, 1H), 5.52 (s, 2H), 3.30 (s, 3H); 13C NMR room temperature and stirred overnight. The mixture was con-
(100.7 MHz, CD3OD) δ 164.5, 163.9, 142.3, 138.3, 119.7, 106.2, centrated in vacuo, and the residue was subjected to silica gel
7936 | Org. Biomol. Chem., 2015, 13, 7928–7939
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
chromatography using hexane/EtOAc (50/1) as the eluent to in vacuo, and the residue was subjected to silica gel chromato-
1
afford compound 17 (0.11 g, 41% yield) as a yellow liquid. H graphy using hexane/EtOAc (20/1) as the eluent to afford com-
NMR (400 MHz, Acetone-d6) δ 7.35 (s, 1H), 7.14 (s, 1H), 5.41 pound 19 (103 mg, 8% yield for two steps) as a yellow solid.
1
(s, 2H), 3.38 (s, 3H); 13C NMR (100.7 MHz, Acetone-d6) δ 181.1, Mp 47–48 °C; H NMR (400 MHz, Acetone-d6) δ 7.26 (s, 1H),
169.8, 163.2, 149.0, 115.7, 109.2, 92.9, 56.4; the acyl azide was 7.07 (s, 1H), 4.66 (dt, J = 47.6 Hz, 4.0 Hz, 2H), 4.47 (dt, J =
converted to free amine during mass analysis. HRMS (ESI) 29.2 Hz, 4.0 Hz, 2H); 13C NMR (100.7 MHz, Acetone-d6)
calcd for C7H9ClN2NaO2 211.0250, [M + Na]+, found 211.0243.
δ 169.8, 163.8, 149.0, 143.4, 115.1, 109.2, 81.4 (d, J = 209.1 Hz),
N-(2-Chloro-6-(methoxymethoxy)pyridin-4-yl)-2-(4-isopropyl- 66.5 (d, J = 24.3 Hz).
1,3-dimethyl-1H-pyrazolo[3,4-b]pyridin-6-yl)hydrazine-1-carbox-
N-(2-Chloro-6-(2-fluoroethoxy)pyridin-4-yl)-2-(4-isopropyl-1,3-
amide (5e). A solution of 17 (108 mg, 0.45 mmol) in toluene dimethyl-1H-pyrazolo[3,4-b]pyridin-6-yl)hydrazine-1-carboxamide
(2.2 mL) was refluxed at 120 °C for 3 hours under nitrogen to (5f). A solution of 19 (64 mg, 0.26 mmol) in toluene (1.5 mL)
in situ produce the isocyanate. To the above mixture was added was refluxed for 3 hours under nitrogen to in situ produce the
dropwise a solution of 3 (76 mg, 0.35 mmol) in anhydrous isocyanate. To the above mixture was added dropwise a solu-
THF (2.5 mL) at 50 °C. The reaction continued to stir at 50 °C tion of 3 (44 mg, 0.2 mmol) in anhydrous THF (2 mL) at 50 °C.
for overnight. The mixture was concentrated in vacuo, and the The reaction continued to stir at 50 °C for overnight. The
residue was subjected to silica gel chromatography using mixture was concentrated in vacuo, and the residue was sub-
CH2Cl2/MeOH (40/1) as the eluent to afford compound 5e jected to silica gel chromatography using CH2Cl2/MeOH as the
(130 mg, 86% yield) as a pale white solid. Mp 208–209 °C; eluent to afford compound 5f (54 mg, 62% yield) as a pale
1H NMR (400 MHz, CD3OD) δ 7.25 (s, 1H), 7.00 (s, 1H), 6.48 solid. Mp 240–242 °C; 1H NMR (400 MHz, CD3OD) δ 7.20 (s,
(br, 1H), 5.37 (s, 2H), 3.81 (s, 3H), 3.53–3.46 (m, 1H), 3.44 (s, 1H), 6.93 (s, 1H), 6.47 (s, 1H), 4.64 (dt, J = 47.6 Hz, 3.6 Hz, 2H),
3H), 2.54 (s, 3H), 1.31 (d, J = 6.8 Hz, 6H); 13C NMR (100.7 MHz, 4.42 (dt, J = 25.2 Hz, 4.0 Hz, 2H), 3.81 (s, 3H), 3.52–3.42 (m,
CD3OD) δ 163.0, 159.9, 155.7, 151.1, 150.9, 148.5, 139.9, 107.8, 1H), 2.52 (s, 3H), 1.30 (d, J = 6.8 Hz, 6H); 13C NMR (100.7 MHz,
107.0, 98.6, 98.5, 96.9, 92.0, 55.8, 31.7, 29.3, 21.8, 13.6; HRMS CD3OD) δ 163.7, 159.9, 155.7, 150.9, 150.8, 148.3, 139.9, 119.7,
(ESI) calcd for C19H24ClN7NaO3 456.1527, [M + Na]+, found 106.7, 98.6, 98.5, 96.5, 81.4 (d, J = 209.1 Hz), 65.5 (d, J =
456.1515.
24.3 Hz), 31.7, 29.4, 21.8, 13.6; HRMS (ESI) calcd for
N-(2-Chloro-6-hydroxypyridin-4-yl)-2-(4-isopropyl-1,3-dimethyl- C19H23ClFN7NaO2 458.1483, [M + Na]+, found 458.1492.
1H-pyrazolo[3,4-b]pyridin-6-yl)hydrazine-1-carboxamide (18).
5.2. Radiochemistry
Compound 5e (54 mg, 0.12 mmol) was dissolved in CH2Cl2
(2 mL). Trifluoroacetic acid (1 mL) was added dropwise at 0 °C. Production of [11C]CH3I followed the reported method.34
The solution was stirred at room temperature for 2 hours. Satu- Briefly, [11C]CH3I was produced on-site using a GE PETtrace MeI
rated aqueous NaHCO3 solution was added to quench the reac- Microlab through consecutive reaction starting from [11C]CO2.
tion. The mixture was extracted with CH2Cl2. The combined Up to 1400 mCi of [11C]CO2 was produced from the JSW
organic phase was dried over sodium sulfate and concentrated BC-16/8 cyclotron by irradiating a gas target of 0.5% O2 in N2
through rotary evaporation. The residue was subjected to silica for 15–30 min with a 40 μA beam of 16 MeV protons in the
gel chromatography using CH2Cl2/MeOH (10/1) as the eluent Barnard Cyclotron Facility of Washington University School of
to afford the precursor 18 (21 mg, 43% yield) for radiolabeling. Medicine. [11C]CH3I was produced by reduction of [11C]CO2
Mp 226–228 °C; 1H NMR (400 MHz, CD3OD) δ 6.90 (s, 1H), using a nickel catalyst (Shimalite-Ni (reduced), Shimadzu,
6.71 (s, 1H), 6.40 (s, 1H), 3.74 (s, 3H), 3.46–3.37 (m, 1H), 2.46 Japan P.N.221-27719) in the presence of hydrogen gas at
(s, 3H), 1.24 (d, J = 6.8 Hz, 6H); 13C NMR (100.7 MHz, CD3OD) 360 °C, followed by iodination of the resulting [11C]CH4 with
δ 164.7, 159.9, 158.0, 155.7, 151.4, 151.0, 140.8, 139.9, 107.8, iodine at 690 °C. Approximately 12.2 min following the end-of-
103.3, 98.6, 98.5, 31.7, 29.4, 21.8, 13.6; HRMS (ESI) calcd for bombardment (EOB), 800–1000 mCi of [11C]CH3I were deli-
C17H21ClN7O2 390.1445, [M + H]+, found 390.1441.
2-Chloro-6-(2-fluoroethoxy)isonicotinoyl azide (19). To
vered via gas tubing to the reaction vial in the hot cell.
a
[11C]CH3I was bubbled for a period of 2–3 min into a solu-
solution of 2,6-dichloroisonicotinic acid (0.96 g, 5 mmol) in tion of precursor (1.0–1.3 mg) in DMF (0.2 mL) containing
anhydrous THF (5 mL) was added a solution of potassium tert- 3.0 µL of potassium hydroxide (5.0 M) at room temperature.
butoxide (1.4 g, 12.5 mmol) in THF (12 mL) dropwise at 0 °C. When the trap of radioactivity was completed, the sealed reac-
5 min later, 2-fluoroethanol (480 mg, 7.5 mmol) was added tion vessel was heated at 85 °C for 5 min, shaking was applied
dropwise and the reaction was continued at room temperature to the vial using a long clamp during the reaction. The oil bath
for overnight. 1 N aqueous hydrochloric acid was added to was removed and 1.7 mL of the HPLC mobile phase (45%
adjust the solution pH to 4.0. The mixture was extracted with acetonitrile in 0.1 M ammonium formate, v/v, pH 6.5) was
EtOAc, the combined organic phase was dried with sodium added into to the reaction vessel. The mixture was loaded onto
sulfate, and concentrated in vacuo to give the crude ether. The a reversed phase HPLC system to purify the mixture (Agilent
ether was then dissolved in anhydrous 1,4-dioxane (10 mL). Zorbax SB-C18 column, 250 × 9.2 mm, mobile phase 45%
Et3N (0.18 mL, 1.3 mmol), diphenylphosphoryl azide (0.36 g, acetonitrile in 0.1 M ammonium formate, pH 6.5, flow rate
1.3 mmol) was added successively. The reaction was stirred at 4.0 mL min−1, detection wavelength 254 nm). Under these con-
room temperature overnight. The mixture was concentrated ditions, the product with retention time at 15–17 min was col-
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem., 2015, 13, 7928–7939 | 7937

View Article Online
Paper
Organic & Biomolecular Chemistry
lected into a vial that contained 50 mL Milli-Q water. Then the Fuji Bio-Imaging Analyzer FLA-7000 (Fuji Photo Film Co.,
collected fraction was passed through a C-18 Plus Sep-Pak® Tokyo, Japan). Photo-stimulated luminescence (PSL) from the
cartridge to concentrate the target component in the Sep-Pak cerebellum was quantified using Multi Gauge v3.0 software
cartridge. The Sep-Pak cartridge was rinsed using 20 mL of (Fuji Photo Film Co., Tokyo, Japan). Data were background-
sterile water. Finally, the tracer trapped on the Sep-Pak® was corrected and expressed as photo-stimulated luminescence
eluted with 0.3 mL of ethanol, following with 2.7 mL 0.9% signals per square millimeter (PSL per mm2).
sodium chloride solution, passing through
a
0.22 µm
Brain microPET imaging was performed using two Siemens
(Whatman Puradisc 13 mm syringe filter) sterile filter into a microPET scanners (Siemens Preclincial Solutions, Knoxville,
sterile pyrogen-free glass vial for delivery. For quality control, TN, USA) – microPET-Focus-F220 and a microPET-Inveon Multi-
an aliquot of sample was assayed by an analytical HPLC Modality scanners. Imaging studies were done using 19–25 g
system (Agilent Zorbax SB-C18 column, 250 × 4.6 mm, mobile SJL mice pretreated with Cyclosporin A. Animals were anesthe-
phase 55% acetonitrile in 0.1 M ammonium formate, pH 4.5, tized using 2% isoflurane/oxygen and a tail vein catheter
flow rate 1.2 mL min−1, detection wavelength 254 nm). The placed in the lateral tail vein. Gas anesthesia was maintained
sample was authenticated by co-injecting with the corres- at <1.5% isoflurane during the imaging session; each mice was
ponding nonradiolabeled standard solution. The retention time positioned on the scanner bed at least 30 min prior to the
was 5.5 min with radiochemical purity was >99%. The radio- tracer injection. Body temperature was maintained using a
synthesis typically took 40 min starting from the release of warming lamp. A dose of 25 mg kg−1 CsA in saline adminis-
[11C]methyl iodide, with yield 20 5% (n > 10, decay corrected tered i.v. 30 min prior to radiotracer injection was used for
to the end of synthesis), specific activity 222–370 GBq µmol−1 modulation of the BBB efflux transporters. Both CsA and the
(decayed to EOB).
radiotracer (400–900 μCi of [11C]5a) were administered using
an i.v. catheter placed in the lateral tail vein. The imaging ses-
sions were carried out as 1 h dynamic scan using the Micro-
5.3. In vitro competitive binding assay
In vitro competitive binding assay against [32P]S1P assay was PET Focus 220 and Inveon scanners (Siemens Medical
conducted according to our published protocol.21
Solutions USA). Acquired list mode data were histogrammed
into a 3D set of sinograms and binned to the following time
frames: 1 × 3 s, 6 × 2 s, 9 × 5 s, 6 × 10 s, 4 × 30 s, 2 × 60 s, 2 ×
5.4. Biodistribution, autoradiography, PET/CT study
All animal experiments were conducted in compliance with 2 min, 10 × 5 min. Sinogram data was then processed using
the Guidelines for the Care and Use of Research Animals filter back projection algorithm with attenuation and scatter
under protocols approved by Washington University’s Animal corrections. Regions of interest (the cerebellum) were manually
Studies Committee. For mice biodistribution studies, 64–89 drawn with the software ASIPro (Acquisition Sinogram Image
µCi of [11C]5a was injected via the tail vein (i.v.) into mature Processing, Siemens Medical Solutions, Malvern, PA, USA) using
female SJL mice (n = 4 per study group, mice weight 18–24 g) IDL’s Virtual Machine (ITT Visual Information Solutions,
under 2–3% isoflurane/oxygen anesthesia. Eight mice were Boulder, CO, USA) to calculate the average cerebellar uptake.
divided into two groups: the control group without CsA pre-
treatment and the CsA pre-treatment group (25 mg kg−1 of
CsA, i.v. 30 min prior to the tracer injection). At 30 min post-
injection (p.i.), mice were anesthetized again and euthanized.
Acknowledgements
The whole brain was quickly harvested, the blood was removed This work was financially supported by the USA DOE-Training
by blotting, and the cerebellum was separated, the remainder Grant: #DESC0008432 and the Washington University, Mal-
of the brain was collected to determine the total brain uptake. linckrodt Institute of Radiology (MIR) Pilot Grant: #14-017, the
Peripheral organs and tissues, including blood, heart, lung, USA National Institute of Mental Health (NIMH) of the
muscle, fat, thymus, spleen, kidney, and liver, were also col- National Institutes of Health (no. MH092797) and National
lected; all samples were weighed and counted in an automated Institute of Neurological Disorders and Stroke (NINDS, no.
well counter with a standard dilution of the injectate. Counts NS075527, NS061025, P01 NS059560), and National Multiple
were decay-corrected and the %ID per g was calculated. A two- Sclerosis Society grants PP2043, RG 4632.
tailed paired Student’s t test was used to calculate the standard
derivations. A p value less than 0.05 was considered statisti-
cally significant.
References
For the ex vivo autoradiography study, SJL mice were
injected with 400–900 µCi of [11C]5a and euthanized 30 min
p.i. as described above. The whole brain was quickly removed
and snap-frozen. Horizontal sections (100 μm) were cut with a
Microm cryotome and mounted on Superfrost Plus glass slides
(Fisher Scientific, Pittsburgh, PA). Frozen slides were directly
exposed to film in an imaging cassette for 4 hours at −80 °C in
the dark. The distribution of radioactivity was visualized by a
1 D. H. Mahad, B. D. Trapp and H. Lassmann, The Lancet.
Neurology, 2015, 14, 183–193.
2 R. M. Ransohoff, D. A. Hafler and C. F. Lucchinetti, Nat.
Rev. Neurol., 2015, 11, 134–142.
3 McAlpine’s Multiple Sclerosis. Section 1: The story of multiple
sclerosis, Elsevier, Churchill, Livingstone, Philadelphia, 2006.
4 A. Compston and A. Coles, Lancet, 2008, 372, 1502–1517.
7938 | Org. Biomol. Chem., 2015, 13, 7928–7939
This journal is © The Royal Society of Chemistry 2015

View Article Online
Organic & Biomolecular Chemistry
Paper
5 A. E. Handel, L. Jarvis, R. McLaughlin, A. Fries, G. C. Ebers 21 A. J. Rosenberg, H. Liu and Z. Tu, Appl. Radiat. Isot., 2015,
and S. V. Ramagopalan, PLoS One, 2011, 6, e14606. 102, 5–9.
6 S. M. Orton, B. M. Herrera, I. M. Yee, W. Valdar, 22 M. Adada, D. Canals, Y. A. Hannun and L. M. Obeid, FEBS
S. V. Ramagopalan, A. D. Sadovnick and G. C. Ebers, The
Lancet. Neurology, 2006, 5, 932–936.
7 B. G. Weinshenker, B. Bass, G. P. Rice, J. Noseworthy,
J., 2013, 280, 6354–6366.
23 Z. Tu, S. Li, J. Xu, W. Chu, L. A. Jones, R. R. Luedtke and
R. H. Mach, Nucl. Med. Biol., 2011, 38, 725–739.
W. Carriere, J. Baskerville and G. C. Ebers, Brain: J. Neurol., 24 W. Loscher and H. Potschka, NeuroRx: J. Am. Soc. Exp.
1989, 112(Pt 6), 1419–1428. Neurotherapeutics, 2005, 2, 86–98.
8 H. Tremlett, Z. Yinshan and V. Devonshire, Mult. Scler., 25 M. D. Moran, A. A. Wilson, C. S. Elmore, J. Parkes, A. Ng,
2008, 14, 314–324.
9 D. Marsolais and H. Rosen, Nat. Rev. Drug Discovery, 2009,
8, 297–307.
O. Sadovski, A. Graff, Z. J. Daskalakis, S. Houle,
M. J. Chapdelaine and N. Vasdev, Bioorg. Med. Chem., 2012,
20, 4482–4488.
10 L. Kappos, E. W. Radue, P. O’Connor, C. Polman, 26 C. N. Wang, C. K. Moseley, S. M. Carlin, C. M. Wilson,
R. Hohlfeld, P. Calabresi, K. Selmaj, C. Agoropoulou,
M. Leyk, L. Zhang-Auberson and P. Burtin, N. Engl. J. Med.,
2010, 362, 387–401.
R. Neelamegam and J. M. Hooker, Bioorg. Med. Chem. Lett.,
2013, 23, 3389–3392.
27 K. Ishiwata, K. Kawamura, K. Yanai and N. H. Hendrikse,
J. Nucl. Med., 2007, 48, 81–87.
28 E. F. J. De Vries, J. Doorduin, N. A. R. Vellinga, A. Van
Waarde, R. A. Dierckx and H. C. Klein, Nucl. Med. Biol.,
2008, 35, 459–466.
11 D. Pelletier and D. A. Hafler, N. Engl. J. Med., 2012, 366,
339–347.
12 E. Roberts, M. Guerrero, M. Urbano and H. Rosen, Expert
Opin. Ther. Pat., 2013, 23, 817–841.
13 L. Cruz-Orengo, B. P. Daniels, D. Dorsey, S. A. Basak, 29 K. A. Kurdziel, D. O. Kiesewetter, R. E. Carson,
J. G. Grajales-Reyes, E. E. McCandless, L. Piccio,
R. E. Schmidt, A. H. Cross, S. D. Crosby and R. S. Klein,
J. Clin. Invest., 2014, 124, 2571–2584.
W. C. Eckelman and P. Herscovitch, J. Nucl. Med., 2003, 44,
1330–1339.
30 S. Syvanen, O. Lindhe, M. Palner, B. R. Kornum,
O. Rahman, B. Langstrom, G. M. Knudsen and
M. Hammarlund-Udenaes, Drug Metab. Dispos., 2009, 37,
635–643.
14 G. T. Kunkel, M. Maceyka, S. Milstien and S. Spiegel, Nat.
Rev. Drug Discovery, 2013, 12, 688–702.
15 M. Inglese and M. Petracca, Prion, 2013, 7, 47–54.
16 T. L. Papenfuss, C. J. Rogers, I. Gienapp, M. Yurrita, 31 P. A. Gourraud, H. F. Harbo, S. L. Hauser and
M. McClain, N. Damico, J. Valo, F. Song and
S. E. Baranzini, Immunol. Rev., 2012, 248, 87–
C. C. Whitacre, J. Neuroimmunol., 2004, 150, 59–69.
103.
17 T. M. Rivers, D. H. Sprunt and G. P. Berry, J. Exp. Med., 32 A. Ascherio and K. L. Munger, Annals Neurol., 2007, 61,
1933, 58, 39–53. 288–299.
18 C. Li, X. X. Chi, W. Xie, J. A. Strong, J. M. Zhang and 33 R. E. Swenson, PCT Int. Appl, 2013 WO 2013148460 A1
G. D. Nicol, J. Neurophysiol., 2012, 108, 1473–1483. 20131003.
19 N. L. Nam, I. I. Grandberg and V. I. Sorokin, Chem. Hetero- 34 Z. Tu, C. S. Dence, D. E. Ponde, L. Jones, K. T. Wheeler,
cycl. Compd., 2003, 937–942.
20 P. S. Fier and J. F. Hartwig, Science, 2013, 342, 956–960.
M. J. Welch and R. H. Mach, Nucl. Med. Biol., 2005, 32,
423–430.
This journal is © The Royal Society of Chemistry 2015
Org. Biomol. Chem., 2015, 13, 7928–7939 | 7939